Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Compartmentalization and B cell antigen receptor signaling Jackson, Teresa Lynn 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2006-130075.pdf [ 27.89MB ]
Metadata
JSON: 831-1.0092779.json
JSON-LD: 831-1.0092779-ld.json
RDF/XML (Pretty): 831-1.0092779-rdf.xml
RDF/JSON: 831-1.0092779-rdf.json
Turtle: 831-1.0092779-turtle.txt
N-Triples: 831-1.0092779-rdf-ntriples.txt
Original Record: 831-1.0092779-source.json
Full Text
831-1.0092779-fulltext.txt
Citation
831-1.0092779.ris

Full Text

C O M P A R T M E N T A L I Z A T I O N A N D B C E L L A N T I G E N RECEPTOR SIGNALING  by TERESA L Y N N J A C K S O N B.Sc, The University of British Columbia, 1997  A THESIS SUBMITTED IN P A R T I A L F U L F I L L M E N T OF T H E REQUIRENENTS FOR T H E D E G R E E OF DOCTOR OF PHILOSOPHY in  THE F A C U L T Y OF G R A D U A T E STUDIES  (Zoology)  THE UNIVERSITY OF BRITISH C O L U M B I A December 2005  © Teresa Lynn Jackson, 2005  ABSTRACT This thesis focuses on two aspects of compartmentalization with respect to B C R signaling.  In the first section,  compartmentalization of the B C R to lipid rafts is considered and in the second section the  subsequent  compartmentalization of B L N K and PLCy to the B C R is considered. Recent studies have suggested that the B C R translocates into lipid rafts in certain B cell stages. Yet these reports have been ambiguous and the mechanisms regulating such translocation have remained elusive. In this thesis it is demonstrated that the B C R can translocate into lipid rafts following B C R cross-linking in the immature B cell lines, WEHI 231 and CH31 (Jackson et al., 2005). Additionally, it is shown that the Iga/(3 heterodimer, in the absence of the mlgM subunit, can translocate into lipid rafts in the immature B cell line, WEHI 303.1.5 (Jackson et al., 2005).  Previous studies have likewise  demonstrated that the mlgM subunit, in the absence of Iga/p, can translocate into lipid rafts (Cheng et al., 2001). Together, these findings may be used to help define a structural feature, common to both subunits, involved in mediating lipid raft association.  The PLCy pathway is an integral part of the B C R signaling network. Loss-of-function studies have indicated that the B C R is coupled to PLCy via Syk, B T K and B L N K .  In this thesis, a non-lymphoid reconstitution system was  used to determine if these components are sufficient to couple the B C R to PLCy. From this it was determined that co-expression of the B C R , Syk and B L N K is sufficient to reconstitute BCR-induced PLCy activation in the system. However, this activation is hypothesized to represent only a partial reconstitution of the pathway as neither B L N K nor PLCy are recruited to the plasma membrane upon B C R cross-linking and as PLCy phosphorylation appears very limited.  It was hypothesized that this might be due to the absence of B T K ; however, further expression of B T K  within the system inhibited rather than enhanced PLCy activation. Subsequent investigations determined that B T K is constitutively activated within this system and as such, may be inappropriately affecting the pathway. Additionally, it was hypothesized that the limited reconstitution may be a consequence of the inability to reconstitute B C R induced B L N K and PLCy membrane recruitment.  Thus, B L N K and PLCy were constitutively targeted to the  plasma membrane within the system. From this, it was determined that membrane-targeting of PLCy is sufficient to reconstitute BCR-induced, Syk-dependent PLCy activation.  In contrast, membrane-targeting of B L N K is not  sufficient to reconstitute BCR-induced PLCy membrane recruitment or to enhance BCR-induced PLCy activation within this system.  This suggests that there may be an additional defect in the system that is preventing the  formation of a functional B C R / B L N K / P L C y signaling complex. Moreover, these findings suggest that there may be a deficit in our current understanding of the BCR/PLCy pathway.  In summary, these findings highlight the importance of compartmentalization in B C R signaling both with respect to compartmentalization of the B C R to lipid rafts and the subsequent compartmentalization of B L N K and PLCy to the BCR.  ii  T A B L E of C O N T E N T S  Abstract  ii  Table of Contents  iii  List of Tables  x  List of Figures  xi  List of Abbreviations  .  . "  xvii  Acknowledgments  xxi  Chapter 1: Introduction  1  1.1  The Biological Problem  1  1.2  A n Overview of the Human Immune System  1  1.3  B C R Structure  4  1.4  A Brief Overview of B Cell Development  9  1.5  Compartmentalization and Signal Transduction Pathways  12  1.5.1 1.5.2  The Role of Adapter Proteins in Signal Transduction Pathways The Role of Lipid Rafts in B C R Signaling  14 15  1.6  A n Overview of B C R Signaling 1.6.1 Initiation of B C R Signaling 1.6.2 The PI3K Pathway 1.6.3 The PLCy Pathway 1.6.4 The Ras/MAPK Pathway .  19 19 22 25 28  1.7  The Enigmas of B C R Signaling  31  1.8  The BCR/PLCy Pathway in Detail . 1.8.1 PLCy Structure and Function 1.8.2 BCR-Induced Tyrosine Phosphorylation of PLCy 1.8.3 BCR-Induced Membrane Recruitment of PLCy  31 31  iii  33 36  1.9 Thesis Goals  42  1.10 Thesis Summary  45  Chapter 2: Materials and Methods 2.1  2.2  2.3  Reagents 2.1.1 2.1.2 2.1.3  48  Antibodies Plasmids Plasmids Created for This Thesis  Molecular 2.2.1 2.2.2 2.2.3 2.2.4 2.2.5 2.2.6 2.2.7  biology methods Restriction endonuclease digests Alkaline phosphatase reaction Phenol/Chloroform extraction Agarose gel electrophoresis Gel purification of D N A D N A ligation reactions Transformation of competent Escherichia coli bacteria 2.2.8 Polymerase chain reactions 2.2.9 Qiagen-mediated preparation of D N A 2.2.10 Cesium chloride-mediated preparation of D N A  48 48 49 49 60 60 60 60 61 61 61 61 62 62 62  Tissue culture 2.3.1 Tissue culture cell lines 2.3.2 Maintenance of tissue culture cell lines 2.3.3 Calcium phosphate-mediated transfections of AtT20-derived cell lines 2.3.4 Drug selection of transfected cells and isolation of individual clones  64  2.4  Stimulation and lysis of cell lines  65  2.5  SDS-PAGE and immunoblot analysis  66  2.6  Immunoprecipitation studies  67  2.7  Membrane enrichment of cell lines  67  2.8  Cytoskeletal-based lipid raft preparation  68  2.9  Inositiol phosphate assay  69  2.10  Population-based calcium flux assay  69  2.11  Single cell-based calcium assay  70 iv  63 63 63 64  2.12  Production of anti-BLNK polyclonal antibodies  71  2.13  Surface biotinylation of cells  72  2.14  Summary of solutions  72  2.14.1 2.14.2 2.14.3 2.14.4 2.14.5 2.14.6 2.14.7 2.14.8 2.14.9 2.14.10 2.14.11 2.14.12 2.14.13 2.14.14 2.14.15 2.14.16 2.14.17 2.14.18 2.14.19 2.14.20 2.14.21 2.14.22 2.14.23 2.14.24 2.14.25 2.14.26 2.14.27 2.14.28 2.14.29 2.14.30 2.14.31 2.14.32 2.14.33  Tris/Boric Acid/EDTA (TBE) 6x D N A Sample Buffer Lauria-Bertani Broth (LB Broth) Lauria-Bertani Agar (LB Agar) M 9 Growth Media Sucrose Solution Bacteria Lysis Buffer I Triton Lytic M i x Bacteria Lysis Buffer II Bead Wash Buffer Tris/EDTA (TE) Complete D M E M Low Serum D M E M Complete RPMI-1640 Low Serum RPMI-1640 Trypsin Solution 2x HEPES-Buffered Saline (2x HBS) l x HEPES-Buffered Saline (lx HBS) Modified HBS Calcium Free Modified HBS TritonX-100 Lysis Buffer D M Lysis Buffer NP40 Lysis Buffer Non-Detergent Lysis Buffer Low Salt Cytoskeletal Stabilization Buffer High Salt Cytoskeletal Stabilization Buffer l x Running Sample Buffer ( l x RSB) 5x Running Sample Buffer (5x RSB) Running Buffer Transfer Buffer Tris Buffer Saline (TBS) Stripping Tris Buffered Saline (TBS) Tris Buffered Saline with Tween20 (TBST)  Chapter 3: The Solo Iga/p Heterodimer Can Localize to Lipid Rafts  . .  72 73 73 73 73 73 73 73 74 74 74 74 74 75 75 75 75 75 76 76 76 76 77 77 77 78 78 78 78 78 79 79 79  80  3.1  Introduction  80  3.2  Results  82  3.2.1 3.2.2  The B C R Translocated into Lipid Rafts in the Immature B Cell Lines, WEHI 231 and CH31 A Portion of Solo Iga/p Localizes to Lipid Rafts v  82  in the Mutant Immature B Cell Line, WEHI 303.1.5 3.3  Discussion  88 95  Chapter 4: Co-Expression of the B C R , Syk, and B L N K Is Sufficient to Reconstitute BCR-Induced PLCy Activation in AtT20-Derived Cell Lines  99  4.1  Introduction  99  4.2  Results 4.2.1  101  4.2.2  4.2.3  4.2.4  4.2.5  4.2.6  4.3  Expression of the B C R , Syk, B T K and B L N K within AtT20-Derived Cell Lines Co-Expression of the B C R , B L N K and/or B T K Is Not Sufficient To Reconstitute BCR-Induced Erk Phosphorylation in AtT20-Derived Cell Lines Co-Expression of the B C R and Syk Is Sufficient to Reconstitute BCR-Induced, P L C Independent Erk Phosphorylation in AtT20Derived Cell Lines Co-Expression of the B C R , Syk and B L N K Is Sufficient to Reconstitute BCR-Induced, P L C Dependent Erk Phosphorylation in AtT20Derived Cell Lines Co-Expression of the BCR, Syk and B T K is Not Sufficient to Reconstitute BCR-Induced, P L C Dependent Erk Phosphorylation in AtT20Derived Cell Lines Co-Expression of the BCR, Syk, B T K and B L N K Is Not Sufficient to Reconstitute B C R Induced, PLC-Dependent Erk Phosphorylation in AtT20-Derived Cell Lines  Discussion 4.3.1 Syk is Necessary and Sufficient to Reconstitute BCR-Induced, PLC-Independent Erk Phosphorylation in the AtT20 System 4.3.2 The Effects of Syk, B T K and/or B L N K Expression on BCR-Induced Erk Phosphorylation in the AtT20 System  Chapter 5: Phosphorylation, Protein Association and Compartmentalization Status of B L N K , B T K and PLCy in the AtT20 System 5.1  Introduction  101  104  107  110  113  116 119  119  120  125 125  vi  5.2  Results 5.2.1  5.2.2  5.2.3  5.2.4 5.2.5  5.2.6  5.2.7  5.2.8  5.2.9  5.3  127 Co-Expression of the B C R and B L N K Is Not Sufficient to Reconstitute BCR-induced PLCy phosphorylation in AtT20-Derived Cell Lines Co-Expression of the B C R with Syk and/or B T K is Sufficient to At Least Partially Reconstitute BCR-induced PLCy Phosphorylation in the AtT20-Derived System Co-Expression of the B C R and B L N K with Syk and/or B T K Does Not Significantly Enhance BCR-induced PLCyPhosphorylation B T K Is Constitutively Phosphorylated in AtT20-Derived Cell Lines B T K Phosphorylation Is At Least Partially Dependent on Fyn Activity in AtT20-Derived Cell Lines Co-Expression of the B C R , B L N K and B T K Is Not Sufficient to Reconstitute BCR-induced B L N K Phosphorylation in AtT20-Derived Cell Lines Co-Expression of the B C R , Syk, and B L N K Is Sufficient to Reconstitute BCR-induced B L N K Phosphorylation in AtT20- Derived Cell Lines Protein Association Studies are Inconclusive in Lymphoid and AtT20-Derived Cell Lines (refer to Appendix II) BCR-induced Membrane Recruitment of Syk, B L N K , B T K and PLCy Is Not Reconstituted in the AtT20-Derived Cell Lines  Discussion 5.3.1 Recalling the Proposed Model of the BCR/PLCy Pathway 5.3.2 BCR-induced PLCy phosphorylation in the AtT20 System 5.3.3 BCR-induced Syk and B T K Phosphorylation in the AtT20 System : 5.3.4 BCR-induced B L N K Phosphorylation in the AtT20 System 5.3.5 BCR-induced Compartmentalization of Syk, B T K , B L N K and PLCy in the AtT20 System  Chapter 6: Co-Expression of the B C R , Syk and Acylated-PLCy2 Is Sufficient to Reconstitute BCR-induced PLCy Activation in AtT20Derived Cell Lines 6.1  Introduction  127  129  132 135  140  143  147  151  156 159 159 160 163 165 165  167 169  vii  6.2  Results 6.2.1  6.2.2  6.2.3  6.2.4  6.2.5 . 6.2.6  6.2.7  6.2.8  6.3  Discussion 6.3.1  6.3.2  6.3.3  170 Expression of Membrane-Targeted B L N K (TmBLNK), Acylated B L N K ( A c B L N K ) and Acylated PLCy2 (AcPLCy2) in AtT20-Derived Cell Lines Co-Expression of the B C R , Syk and T m B L N K or A c B L N K is Sufficient to Reconstitute B C R Induced, PLC-Dependent Erk Phosphorylation in AtT20-Derived Cell Lines Co-Expression of the B C R , Syk and T m B L N K is Sufficient to Reconstitute BCR-Induced Phosphorylation of T m B L N K in AtT20Derived Cell Lines Co-Expression of the B C R , Syk, and T m B L N K Appears to Enhance BCR-Induced P L C y l Phosphorylation in AtT20-Derived Cell Lines Co-Expression of the B C R , Syk and T m B L N K or A c B L N K Is Not Sufficient to Reconstitute Inducible Membrane Recruitment of P L C y l in AtT20-Derived Cell Lines Co-Expression of the B C R , Syk and AcPLCy2 Appears to Enhance BCR-Induced, P L C Dependent Erk Phosphorylation in AtT20Derived Cell Lines Co-Expression of the B C R , Syk and AcPLCy2 Is Sufficient to Reconstitute BCR-Induced AcPLCy2 Phosphorylation in AtT20-Derived Cell Lines Co-Expression of PLCy2, With the B C R , B L N K and Syk, Does Not Appear to Enhance BCR-Induced Erk Phosphorylation in AtT20Derived Cell Lines  170  176  182  187  191  193  199  201 204  Targeting Human B L N K and Human PLCy to the Plasma Membrane in AtT20-Derived Cell Lines Co-Expression of the B C R , Syk and Membrane Targeted B L N K Is Sufficient to Reconstitute BCR-Induced PLCy Activation in AtT20Derived Cell Lines Co-Expression of the B C R , Syk and AcPLCy2 Is Sufficient to Reconstitute BCR-Induced PLCy Activation in AtT20-Derived Cell Line  Chapter 7: Discussion  204  206  209  210 viii  7.1  Introduction  210  7.2  Review of Proposed Model and Initial Hypotheses  210  7.3  Review of Key Findings  212  7.4  Discussion of Findings and Future Considerations  212  7.4.1 7.4.2 7.4.3 7.5  Findings Regarding the BCR-Induced P L C Independent Erk Pathway Initial Findings Regarding the BCR/PLCy Pathway Subsequent Findings Regarding the BCR/PLCy Pathway  212 215 222  Outstanding Questions, Possible Explanations and Future Work 7.5.1 Mis-Compartmentalization of B L N K Within the AtT20 System 7.5.2 Mis-Compartmentalization of PLCy Within the AtT20 System .  229  7.6  Compartmentalization of the B C R to Lipid Rafts  230  7.7  Final Words  231  225 225  References  233  Appendix I: Summary of Cell Lines  249  Appendix II: Confirmation of Specificity of the Various Antibodies Utilized  253  Appendix III: Summary of Immunoprecipitation Studies  254  Appendix IV: Summary of Inositol Phosphate Studies  259  Appendix V : Summary of Population-Based Calcium Flux Assays  265  Appendix VI: Summary of Single-Cell Calcium Flux Assays  270  ix  L I S T of T A B L E S  Table 2.1.  Summary of Primers Used To Create the Various Plasmids  Table 4.1.  Table 5.1.  50  The Effects of Syk, B T K and/or B L N K Expression on BCR-Induced Erk Phosphorylation in the AtT20 System  121  Summary of Key Findings Regarding BCR-Induced P L C y l Phosphorylation and Activation in the AtT20 System  161  Table 7.1.  Summary of the Key Findings of This Thesis  213  Table A L L  Summary of Protein Expression in AtT20-Derived Cell  Table A3.1.  Table A3.2.  Table A3.3.  Table A3.4.  Table A3.5.  Table A3.6.  Table A6.1.  Lines  252  Summary of B L N K / P L C y 1 Immunoprecipitation Studies in AtT20-derived Cell Lines  254  Summary of BLNK/PLCy2 Immunoprecipitation Studies in AtT20-derived Cell Lines  256  Summary of Membrane-Targeted B L N K / P L C y 1 CoAssociation Studies in AtT20-derived Cell Lines . . .  257  Summary of B T K / B L N K Co-Association Studies in AtT20-derived Cell Lines  257  Summary of BLNK/Ig-a Co-Association Studies in AtT20-derived and Lymphoid Cell Lines  258  Summary of B L N K / P L C y Co-Association Studies in Lymphoid Cell lines  258  Classification Criteria For an Individual Cell Based on Its Calcium Flux Response to an Agonist  271  x  L I S T of F I G U R E S Figure 1.1.  Schematic Representation of the B C R  8  Figure 1.2.  Cell Types Derived from the Haematopoietic Stem Cell  10  Figure 1.3.  Summary of Key B Cell Developmental Stages and Indication of Some of The Cell Properties Associated With the Respective Stages  12  Figure 1.4.  Schematic Representation of Key Protein and Lipid Interaction Domains  13  Figure 1.5  Schematic Representation of Lipid Rafts  18  Figure 1.6.  Summary of the Initial Events of B C R Signaling  21  Figure 1.7.  Summary of the BCR/PI3K Pathway  23  Figure 1.8.  Summary of the BCR/PLCy Pathway  27  Figure 1.9.  The Ras/MAPK Pathway  30  Figure 1.10.  General Structure of PLCy 1 and PLCy2  33  Figure 1.11.  General Structure of Syk and B T K  Figure 1.12.  General Structure of h B L N K  38  Figure 2.1.  Schematic representation of how the RSVpLpA-mychuman-BLNK (pp70) expression vector was developed  53  Schematic representation of how the RSVpLpA-humanPLCy2 expression vector (-8.5 kB) was developed  54  Schematic representation of how the RSVpLpA-TM expression vector (~ 5.2 kB) was developed  55  Schematic representation of how the RSVpLpA-TMhuman-BLNK expression vector (~ 6.8 kB) was created .  56  Schematic representation of how the RSVpLpA-Ac expression vector (~ 4.56 kB) was developed  57  Schematic representation of how the RSVpLpA-Achuman-BLNK expression vector (~ 6.16 kB) was developed  58  Figure 2.2.  Figure 2.3.  Figure 2.4.  Figure 2.5.  Figure 2.6.  Figure 2.7.  Schematic representation  of how the xi  .36  RSVpLpA-Ac-  human-PLCy2 expression vector (~ 8.56 kB) was developed  59  Characterization of Iga, IgP and mlgM (as determined by u chain) expression in the experimental B cell lines  85  The B C R translocates into a detergent-insoluble, saltextractable lipid raft fraction in the WEHI 231 immature B cell line following B C R cross-linking  86  The B C R translocates into a detergent-insoluble, saltextractable lipid raft fraction in the CH31 immature B cell line Following B C R cross-linking  87  The B C R translocates into a detergent-insoluble, saltextractable lipid raft fraction in the WEHI 231 immature B cell line following Igp cross-linking  91  A portion of the solo Iga/IgP heterodimer localizes to detergent-insoluble, salt-extractable lipid raft fraction in the mutant WEHI 303.1.5 cell line (mlgM negative) .  92  A portion of the solo Iga/IgP heterodimer localizes to the detergent-insoluble, salt-extractable lipid raft fraction in the K40B-1 pro-B-like cell line (mlgM negative)  93  A portion of the Iga/IgP heterodimer localizes to the detergent-insoluble, salt-extractable lipid raft fraction in the K40B-2 pre-B-like cell line (mlgM positive)  94  Figure 4.1.  Basic Overview of the BCR/PLCy Signaling Pathway .  102  Figure 4.2.  Characterization of a, u, Syk, B L N K , B T K and P L C y l Expression in Transfected AtT20 Cell Lines  103  Basic Overview of BCR-Induced Erk Phosphorylation Via the PLCy-Independent and PLCy-Dependent Signaling Pathways  105  Co-Expression of the BCR, B L N K and/or B T K Is Not Sufficient to Reconstitute BCR-Induced Erk Phosphorylation in AtT20-Derived Cell Lines  106  Co-Expression of the B C R and Syk Is Sufficient to Reconstitute PLC-Independent, BCR-Induced Erk Phosphorylation in AtT20-Derived Cell Lines  108  Figure 3.1.  Figure 3.2.  Figure 3.3.  Figure 3.4.  Figure 3.5.  Figure 3.6.  Figure 3.7.  Figure 4.3.  Figure 4.4.  Figure 4.5a.  Figure 4.5b.  Comparison of Mean Pixel Intensity of the Phosphorylated Erk "Bands" in the BCR/Syk Cell Line Xll  in Non-Inhibited and PLC-Inhibited Samples  109  Figure 4.6. Co-Expression of B L N K , Along with Syk and the BCR, Appears to Slightly Inhibit BCR-Induced Erk Phosphorylation in AtT20-Derived Cell Lines Figure 4.6c.  Figure 4.7.  Ill  Comparison of Mean Pixel Intensity of the Phosphorylated Erk "Bands" in the B C R / S y k / B L N K Cell Line in Non-Inhibited and PLC-Inhibited Samples  112  Co-Expression of B T K , Along with Syk and the B C R , Appears to Slightly Inhibit BCR-Induced Erk Phosphorylation in AtT20-Derived Cell Lines . . . . . .  Figure 4.7c.  Figure 4.8.  Figure 4.8c.  Figure 5.1.  Figure 5.2.  Figure 5.3.  Figure 5.4.  Figure 5.5.  Figure 5.6.  114  Comparison of Mean Pixel Intensity of the Phosphorylated Erk "Bands" in the BCR/Syk/BTK Cell Line in Non-Inhibited and PLC-Inhibited Samples . .  115  Co-Expression of the BCR, Syk, B L N K and B T K Is Not Sufficient to Reconstitute BCR-Induced, PLC-Dependent Erk Phosphorylation in AtT20-Derived Cell Lines  117  Comparison of Mean Pixel Intensity of the Phosphorylated Erk "Bands" in the B C R / S y k / B T K / B L N K Cell Line in Non-Inhibited and PLC-Inhibited Samples  119  Review of Proposed Model of BCR-Induced PLCy Activation  126  Co-Expression of the B C R and B L N K Is Not Sufficient to Reconstitute BCR-Induced P L C y l Phosphorylation in AtT20-Derived Cell Lines  128  Co-Expression of the B C R , along with Syk or B T K , Is Sufficient to Reconstitute BCR-Induced PLCyl Phosphorylation in AtT20-Derived Cell Lines  . . . . . . 1 3 1  Co-Expression of the B C R and Syk Is Sufficient to Reconstitute BCR-Induced P L C y l Phosphorylation in AtT20-Derived Cell Lines  134  Co-Expression of the B C R and B T K is Sufficient to Reconstitute BCR-Induced B T K Phosphorylation in AtT20-Derived Cell Lines  136  The phospho-BTK (Tyr223) specific antibody is specific or phosphorylated B T K  139  xiii  Figure 5.7.  Figure 5.8.  Figure 5.9.  Figure 5.10.  Figure 5.11.  Figure 5.12.  Figure 5.13.  Figure 5.14.  Figure 5.15.  Figure 6.1.  Figure 6.2.  Figure 6.3.  Figure 6.4.  Constitutive Phosphorylation of B T K is, At Least, Partly Dependent on Fyn Activity in AtT20-Derived Cell Lines  142  Co-Expression of the B C R and B L N K is Not Sufficient to Reconstitute BCR-Induced B L N K Phosphorylation in AtT20-Derived Cell Lines. In Contrast, Co-Expression of the B C R , Syk and B L N K is Sufficient to Reconstitute BCR-Induced B L N K Phosphorylation in AtT20-Derived Cell Lines  145  Co-Expression of the B C R , B T K and B L N K is Not Sufficient to Reconstitute BCR-Induced B L N K Phosphorylation in AtT20-Derived Cell Lines  146  Co-Expression of the BCR, Syk and B L N K is Sufficient to Reconstitute BCR-Induced B L N K Phosphorylation in AtT20-Derived Cell Lines  149  Co-Expression of the B C R , Syk, B L N K and B T K is Sufficient to Reconstitute BCR-Induced B L N K Phosphorylation in AtT20-Derived Cell Lines  150  B L N K Does Not Appear to Co-Immunoprecipitate with Either P L C y l or PLCy2 in the Daudi B Cell Line  153  P L C y l Does Not Appear to Co-Immunoprecipitate with B L N K in the Daudi B Cell Line  154  PLCy2 Does Not Appear to Effectively CoImmunoprecipitate with B L N K in the Daudi B Cell Line  155  BCR-Induced Membrane Recruitment of Syk, B L N K , B T K and/or PLCy is not Apparent in AtT20 Derived Cell Lines  158  Strategies for Constitutiyely Targeting human B L N K and human PLCy2 to the Plasma Membrane Within AtT20Derived Cell Lines  169  Characterization of TmBLNK, A c B L N K , AcPLCy2 and P L C y l Expression in Transfected AtT20 Cell Lines  171  Both Molecular Weight Forms of T m B L N K (-114 kd and -105 kd) Appear To Constitutively Associate with the Membrane Fraction in AtT20-Derived Cell Lines  174  The Heavier Form of T m B L N K (-114 kD) Appears to Expressed on the Cell Surface in AtT20-Derived Cell xiv  Figure 6.5.  Figure 6.6.  Figure 6.6b.  Figure 6.6c.  Figure 6.7.  Figure 6.8.  Figure 6.9.  Figure 6.10.  Figure 6.11.  Figure 6.12.  Figure 6.13.  Lines  175  Co-Expression of the B C R , Syk and T m B L N K or A c B L N K May Inhibit Rather than Enhance B C R Induced Erk Phosphorylation in AtT20-Derived Cell Lines  178  Co-Expression of the B C R , Syk and T m B L N K or A c B L N K is Sufficient to Reconstitute BCR-Induced, PLC-Dependent Erk Phosphorylation in AtT20-Derived Cell Lines  179  Comparison of Mean Pixel Intensity of the Phosphorylated Erk "Bands" in the BCR/Syk/TmBLNK Cell Line in Non-Inhibited and PLC-Inhibited Samples  180  Comparison of Mean Pixel Intensity of the Phosphorylated Erk "Bands" in the B C R / S y k / A c B L N K Cell Line in Non-Inhibited and PLC-Inhibited Samples  181  Co-Expression of the B C R , Syk and T m B L N K is Sufficient to Reconstitute BCR-Induced T m B L N K Phosphorylation in AtT20-Derived Cell Lines  184  Co-Expression of the B C R , Syk and T m B L N K is Sufficient to Reconstitute BCR-Induced T m B L N K Phosphorylation in AtT20-Derived Cell Lines  185  Co-Expression of the B C R , Syk and A c B L N K is Sufficient to Reconstitute BCR-Induced A c B L N K Phosphorylation in AtT20-Derived Cell Lines  186  Co-Expression of the B C R , Syk and T m B L N K Appears to Enhance BCR-Induced P L C y l Phosphorylation in AtT20-Derived Cell Lines  188  Co-Expression of the B C R , Syk and T m B L N K Appears to Enhance BCR-Induced P L C y l Phosphorylation in AtT20-Derived Cell Lines  189  Co-Expression of A c B L N K along with the B C R and Syk, Does Not Appear to Significantly Enhance B C R Induced P L C y l Phosphorylation in AtT20-Derived Cell Lines  190  Co-expression of T m B L N K or A c B L N K , Along with the B C R and Syk, Does not Appear Sufficient to Reconstitute Inducible Membrane Recruitment of P L C y l xv  in AtT20-Derived Cell Lines  192  Co-Expression of the B C R , Syk, and AcPLCy2 Significantly Enhances BCR-Induced Erk Phosphorylation in AtT20-Derived Cell Lines  195  Co-Expression of the B C R , Syk and AcPLCy2 is Sufficient to Reconstitute BCR-Induced, PLC-Dependent Erk Phosphorylation in AfT20-Derived Cell Lines  196  Figure 6.15b. Comparison of Mean Pixel Intensity of the Phosphorylated Erk "Bands" in the BCR/Syk/AcPLCy2 Cell Line in Non-Inhibited and PLC-Inhibited Samples  197  Figure 6.15c. Comparison of Mean Pixel Intensity of the Phosphorylated Erk "Bands" in the BCR/Syk//AcBLNKAcPLCy2 Cell Line in NonInhibited and PLC-Inhibited Samples  198  Figure 6.14.  Figure 6.15.  Figure 6.16.  Figure 6.17.  Figure 6.18.  Figure A L L  Figure A l . 2 .  Figure A2.1.  Figure A5.1.  Figure A6.1.  Co-Expression of the AcPLCy2, Along with the B C R and Syk, Appears to Sufficient to Reconstitute B C R Induced AcPLCy2 Phosphorylation in AtT20-Derived Cell Lines  200  Characterization of Syk, B L N K , and PLCy2 Expression in Transfected AtT20 Cell Lines  202  Co-Expression of AcPLCy2, Along with the B C R and Syk, Does Not Appear to Enhance BCR-Induced Erk Phosphorylation in AtT20-Derived Cell Lines  203  Summary Diagram of Lymphoid Cell Lines Utilized in This Thesis  249  Summary Diagram of Key AtT20-Derived Cell Lines Utilized in This Thesis  250  Confirmation of Specificity of the Erk, PLCy 1, PLCy2, B L N K , B T K , Syk, heavy chain, X light chain, P and a antibodies  253  Calcium Flux in the Daudi and SRI Cell Line As Determined by Fura-2 Based Flourometric Ratio Analysis  269  Serotonin-Induced Calcium Flux in SRI Cells  273  xvi  L I S T of A B B R E V I A T I O N S Ac AM ASK1 AtT20 APC B cell BASH BCA BCAP BCR Blk BLNK BME BSA BTK CaCb CD 16 CLP CO2 CSB CsCl C-terminal D D DH DAG DIGs D M lysis buffer DMEM DMSO DNA DOC D-PBS DRMs DTT E E. coli EC EDTA Erk g GDP GEMs GPI GST  Acylation (used to refer to fusion proteins that are targeted to the membrane via fusion with the Lyn acylation sequence) acetoxymethyl ester apoptosis-signaling regulating kinase mouse pituitary gland tumor cell line antigen presenting cell B lymphocyte B lymphocyte adapter protein containing a Src homology 2 domain bicinchoninic acid B cell adapter for PI3K B cell antigen receptor B lymphocyte kinase B cell linker protein 2-P-mercaptoethanol bovine serum albumin Bruton's tyrosine kinase calcium chloride cluster of differentiation 16 (also known as FcyRIIIa) common lymphoid progenitor carbon dioxide cytoskeletal stabilization buffer cesium chloride carboxyl terminal aspartic acid diversity segment diversity segment of the heavy chain diacylglycerol detergent-insoluble glycolipid-enriched membranes n-Dodecyl-P-d-maltoside lysis buffer Dulbecco's modified Eagle Medium dimethylsulfoxide deoxyribonucleic acid deoxycholate Dulbecco's phosphate buffered saline detergent-resistant membranes dithiothreitol glutamic acid Escherichia coli extracellular (ethylenedinitrilo)tetraacetic acid extracellular regulated kinase gram guanine diphosphate glycolipid-enriched membranes glycophosphatidylinositol glutathione S-transferase xvii  GSK-3p GTP hBLNK HBS HC1 HM79-16  HRP HSC I  Ig Iga/p IkB IkK  IP3 IT A M J JH JL  K40B-1 K40B-2 KC1 L L LAB LAT LB LiCl LTR MAPK MAPKKK MgS0 mg MHC I M H C II mlgA mlgD mlgE mlgG mlgM 4  ml mM MPC mPLCy Mg uM  glycogen synthase 3P guanine triphosphate human B cell linker protein HEPES-buffered saline hydrochloric acid monoclonal antibody specific for the extracellular domain of murine IgP that is purified from the HM76-19 hamster hybridoma cell line horseradish peroxidase haematopoietic stem cell isoleucine immunoglobulin immunoglobulin alpha/beta heterodimer, signaling subunit of the BCR inhibitor of N F - k B Ik kinase inositol tris phosphate immunoreceptor tyrosine-based activation motif joining segment joining segment of heavy chain joining segment of light chain pro-B like cell line pre-B like cell line potassium chloride litre leucine linker of activated B cell linker of activated T cell Luria-Bertani lithium chloride long terminal repeat mitogen activated protein kinase M A P K kinase kinase magnesium sulfate milligram major histocompatibility complex class I major histocompatibility complex class II membrane-bound immunoglobulin A membrane-bound immunoglobulin D membrane-bound immunoglobulin E membrane-bound immunoglobulin G membrane-bound immunoglobulin M , antigen-binding subunit of the B C R millilitre millimolar myeloid progenitor cell membrane-targeted rat Phospholipase C gamma microgram micromolar xviii  NaCl Na2HP04 Na3VC>4 NAPS Unit NF A T NP40 lysis buffer N-terminal OD PAMP pBS PDGRF PDK1 PDK2 PCR PH PI3K PI-4-P PI-3-4-P2 PIP PI-3-4-5-P3 PIP PLC PLCyl PLCy2 PRR PTK PTP RasGRP RSV Rpm RPMI RSB S. cerevisiae SAPK SDS SDS-PAGE SFK SH2 SH3 SHP-2 Slp-65 SOS Syk T cell TBE TBS TBST TCR TE C  2  3  sodium chloride sodium hydrogen phosphate sodium orthovanadate Nucleic Acid Protein Service Unit at U B C nuclear factor of activated T cells (cytosolic component) Nonidet P40 lysis buffer amino terminal optical density pathogen-associated molecular patterns refers to the Bluescript plasmid platelet-derived growth factor receptor phosphoinositide-dependent kinase 1 phosphoinositide-dependent kinase 2 polymerase chain reaction pleckstrin homology domain phosphatidyly-inositol 3 kinase phosphatidyl inositol-4-phosphate phosphatidyl inositol-3-4-bisphosphate phosphatidyl inositol bisphosphate phosphatidyl inositol-3-4-5-trisphosphate phosphatidyl inositol trisphosphate phospholipase C phospholipase C gamma 1 phospholipase C gamma 2 pattern recognition receptor protein tyrosine kinase protein tyrosine phosphatase Ras guanyl nucleotide-releasing protein Rous sarcoma virus revolutions per minute Roswell Park Memorial Institute running sample buffer Saccharomyces cerevisiae stress-activated protein kinase sodium dodecyl sulfate sodium dodecylsulfate-polyacrylamide gel electrophoresis Src family kinase Src homology 2 domain Src homology 3 domain Src homology 2 domain-containing protein tyrosine phosphatase-2 Src homology 2 domain-containing leukocyte protein of 65 kD Son of Sevenless Spleen tyrosine kinase T lymphocyte Tris-buffered E D T A tris-buffered saline tris-buffered saline with Tween 20 detergent T cell receptor tris/EDTA solution xix  TLR TM  Toll-like receptor transmembrane (used to refer to fusion proteins that are targeted to the plasma membrane via fusion to the T cell receptor extracellular domain and the C D 16 transmembrane domain) Triton X-100 University of British Columbia variability segment variability segment of heavy chain variability segment of light chain Walter and Eliza Hall Institute immature B cell line mlgM-deficient immature B cell tyrosine monoclonal phosphotyrosine specific antibody r  TX-100 UBC V V  H  V  L  WEHI WEHI 231 WEHI 303.1.5 Y 4G10  xx  ACKNOWLEDGEMENTS Brevity has its place. But it is not here; for it is far too rare that we have or take the opportunity to thank the people in our lives that have helped us along our ways.  I consider myself fortunate in knowing that I could not possibly  acknowledge all those who have helped me to reach this point; I only hope that those who are not directly acknowledged here still know how much I have appreciated their support, advice and friendships.  From the Matsuuchi Lab I thank the directed study students, Gabe Woolham, Aneez Mohamed and Eric Zhou for their assistance. I thank Lorna Santos and Jared Lopes for our many conversations, scientific and otherwise. I thank May Dang-Lawson, without whom the lab would cease to function. I thank Janis Dylke, whose calm, patience and quick intelligence is always a source of inspiration. I thank Steve Macthaler and Emily McWalter for so kindly sharing their place, for their support and for their many efforts quantifying my data. I especially thank Steve for the many  00  favours and for making me laugh when all else fails.  I thank Colm Condon for encouraging me and  introducing me to Dr. Linda Matsuuchi. A n d I thank Linda Matsuuchi for all her assistance, patience and welldelivered lessons.  Linda, I especially thank you for accepting me as me and understanding and supporting my  passion for rugby and for teaching. I also thank you for teaching Biology 441 where I began to understand what a good teacher really was. I can only hope that I can translate those lessons to my students.  From other labs, I thank Mike Gold and his group for all their assistance. their assistance with my calcium flux assays.  I thank Peter Knight and John Church for  John, I especially thank you for being so generous with your  equipment, time and expertise. I thank my committee, Dr. Vanessa Aiild, Dr. Hugh Brock, Dr. Ljerka Kunst and Dr. Nelly Pante, for their many hours, for their advice and for always being available at the last minute.  From my life outside the lab, I thank my rugby team, the U B C Thunderbirds. Words fail me - and you all know how rare that is! Nonetheless, thank you for ensuring that I lived life instead of just studying it. I thank my family. I thank my grandparents Bob and Lorna Ritchie and Jean Hutchin (Jackson) who have always believed in and supported me. I thank my siblings, Dale, Dylan, Dave and Stefania who always make sure I have both feet planted firmly on the ground while I stand on their shoulders! I thank the Corkins for taking me into their family and home. I thank my Mom who always taught me to dream big, saying "Don't just drive the tow truck, own the whole company!" I thank my Dad who taught me the work ethic, perseverance and many other skills necessary to achieve my dreams. Dad, thank you for being you and for always being there.  And finally, I thank my husband James Corkin, without whom this thesis truly would not exist. James, thank you for typing in all my references! Beyond that, I can not possible thank you for all that you do for me. With you, I am my best.  '  -  .  xxi  CHAPTER 1  Introduction  1.1 The Biological Problem  Every moment of every day our bodies are invaded by pathogens such as bacteria, viruses, fungi and parasites. And every day our immune system staves off such invasions. Unfortunately, at times our immune system fails, leaving us vulnerable to illness, disease and even death. Encouragingly, humanity has gained a vast knowledge of how our immune system functions and fails to function such that we can medically manipulate and reinforce our immune system to help our bodies in their continual defense against disease and death. Nonetheless, many aspects of immune function remain to be elucidated and many illnesses, immunodeficiencies and autoimmune diseases remain to be eradicated.  1.2 An Overview of the Human Immune System  Our immune system includes two cooperative branches of defense namely, the innate response and the adaptive response. The innate response is an immediate yet less-specific response that is primarily mediated by macrophages, neutrophils and dendritic cells.  These cells circulate  throughout our peripheral tissues where they encounter pathogens. Upon recognizing a pathogen these cells will either secrete products to degrade the pathogen or they will engulf the pathogen and degrade it intracellularly. In contrast, the adaptive response is a delayed yet more-specific response that is primarily mediated by B and T lymphocytes (B and T cells, respectively). These cells circulate throughout our blood, lymph and lymph nodes as naive cells where they encounter antigen presenting cells (APCs; e.g. macrophages and dendritic cells of the innate response and B cells of the adaptive response). Upon specifically recognizing an A P C , and upon receiving the appropriate co-stimulatory signals, B and T cells differentiate into effector cells and memory cells. Effector cells then help the innate system to specifically control the immediate infection while memory cells provide long-lasting and specific protection against re-infection. Thus, the innate and adaptive responses cooperate to provide both an immediate and a long term defense against pathogens.  1  A critical aspect of immune system function is the ability to recognize self from non-self (i.e., pathogen). Innate cells achieve this via germ-line encoded pattern recognition receptors (PRRs) (reviewed by Aderam and Ulevitch, 2000; Takeda and Akira, 2001; Janeway, Jr. and Medzhitov, 2002; Sieling and Modlin, 2002). PRRs bind conserved pathogen-associated molecular patterns (PAMPs) that are uniquely yet widely expressed by pathogens (reviewed by Aderam and Ulevitch, 2000; Takeda and Akira, 2001; Janeway, Jr. and Medzhitov, 2002; Sieling and Modlin, 2002). Hence, PRR-PAMP interactions enable individual innate cells to distinguish self from non-self and to respond to a broad spectrum of pathogens.  The best characterized PRRs include members of the Toll-like receptor (TLR) family.  TLRs  were first identified for their role in Drosophila development and later for their role in Drosophila immune defense (Lemaitre et al., 1996; reviewed in Medzhitov, 2001, Takeda and Akira, 2001). Subsequently, at least ten TLRs have been identified in the mammalian immune system (reviewed in Janeway Jr. and Medzhitov, 2002). TLRs are most commonly expressed on phagocytic cells where their engagement leads the cell to engulf and degrade the pathogen and to secrete chemokines that further enhance the innate response. However, on dendritic cells, T L R engagement leads the cell to engulf the pathogen and to migrate to a lymph node where the cell then presents the pathogen to naive T cells.  If a naive T cell specifically recognizes the  presented pathogen it will differentiate into effector and memory T cells. Thus, TLRs play a pivotal role both in mediating an immediate innate response and in activating a long-term adaptive response.  Adaptive cells distinguish self from non-self via cell-surface antigen receptors that bind to immunogenic peptides (antigens) that are typically derived from pathogens.  As an entire  population, B and T cells express a large repertoire of antigen receptors that are capable of recognizing and responding to vast array of pathogens. However, any given cell is capable of recognizing and responding to only one specific pathogen as it expresses multiple copies of only one unique, single-specificity antigen receptor. The specificity and variability of these receptors is achieved through the process of random somatic recombination coupled with variable pairing of different polypeptide chains to form the binding site of the receptor.  However, receptors  capable of recognizing self-peptides are sometimes generated as the recombination event is random. Fortunately, cells expressing such receptors are typically eliminated before maturation through the process of clonal selection and deletion (refer to Chapter 1.4). .  2  Following  maturation, naive B and T cells circulate throughout our blood and peripheral tissues where they may encounter their cognate antigen. If this encounter is accompanied by the appropriate costimulatory signals the cell will undergo clonal expansion where it will proliferate and differentiate into progeny effector and memory cells. These progeny express receptors of the same specificity of the parental naive cell such that they are able to specifically respond to initiating pathogen. Thus, the adaptive response is comprised of a population of B and T cells that can distinguish self from non-self to provide a specific and tailored defense against pathogens.  Antigen receptor engagement leads to a variety of immune responses depending on the identity and developmental stage of the cell, on the identity of the antigen receptor-antigen interaction and on the context of signaling. In the case of T cells, the T cell antigen receptor (TCR) can only recognize antigens that are presented on the surface of APCs. recognize antigens  In particular, TCRs can only  that are associated with specialized host molecules termed  major  histocompatibity complexes (MHCs). Cells infected with intracellular pathogens tend to present the resulting antigens on M H C class I molecules ( M H C I) whereas cells that have engulfed extracellular pathogens (e.g., macrophages and dendritic cells) tend to present the resulting antigens on M H C class II molecules (MHC II). Depending on the antigen:MHC class presented different T cell responses will be initiated. Thus, M H C molecules help to direct the T cell response to ensure that the appropriate response is mounted against the eliciting pathogen.  Many T cell responses are initiated when a naive T cell interacts with a mature dendritic cell of the innate response.  Mature dendritic cells present antigen:MHCs on their cell surface along  with co-stimulatory molecules that drive naive T cell differentiation. Dendritic cells presenting antigen:MHC I tend to drive naive T cells to differentiate into cytotoxic T cells that are armed to recognize and destroy cells that present antigens from intracellular pathogens.  In contrast,  dendritic cells presenting antigen:MHC II tend to drive naive T cells to differentiate into helper T cells that are armed to recognize cells that present antigens from extracellular pathogens. Upon such recognition, helper T cells may release cytokines that induce macrophage recruitment, differentiation and activation such that the macrophage is better able to destroy the pathogen. Alternatively, upon recognition of cognate antigen-presenting B cells, helper T cells release cytokines that induce the B cell to differentiate into an effector cell. As well, the naive T cell proliferates and differentiates into memory T cells that provide a specific and more rapid 3  response upon re-infection with the same pathogen. Thus, T cell responses can provide longterm, specific adaptive immunity as well as enhance the innate response..  In the case of B cells, the B C R can recognize soluble antigens. When the B C R of a naive B cell binds a soluble antigen it takes up the antigen via receptor-mediated endocytosis. The B cell then processes the antigen and presents that antigen on its cell surface in the context of M H C II molecules. If the antigen-presenting B cell then interacts with a helper T cell that expresses a cognate TCR both cells will become activated (reviewed in Bernard et al., 2005). Initially, the B cell will activate the helper T cell to release cytokines. In turn, these cytokines will activate the B cell to proliferate and differentiate into antibody-secreting plasma cells.  The resulting  antibodies circulate throughout the blood where they can then specifically neutralize the initiating pathogen by either inhibiting it from interacting with host cells or by targeting it for destruction by various components of the immune response. And similar to T cells, the naive B cell will also proliferate and differentiate into long-living memory B cells that provide a specific and more rapid response to the pathogen upon re-infection with the same pathogen.  Ultimately, the innate response and the adaptive response function coordinately to defend the body against pathogens.  Initially innate cells provide a more general and immediate defense  against pathogens.  This defense is then reinforced when innate cells recruit and activate the  adaptive response.  The adaptive response then aids the innate response by specifically  eliminating the initiating pathogen while also establishing a long-term defense  against  subsequent re-infection. Thereby, our immune system protects us from disease.  1.3 B C R Structure  The experiments of this thesis utilized the murine B C R and as such it is the murine B C R that is detailed below. Nonetheless, the overall structure of the human B C R and the murine B C R are comparable.  The B C R is expressed on the cell surface in various forms throughout B cell  development. The earliest form of the B C R to be expressed is the pro-BCR which is expressed on pro-B cells (refer to Chapter 1.4). The pro-BCR is multi-peptide complex that is comprised of an Iga/p disulfide-linked heterodimer that is expressed on the cell surface in association with the chaperone protein, calnexin (Koyama et al., 1997; Nagata et al., 1997). 4  As well, this  complex may include a surrogate light chain (VpreB and X5) that is expressed on the cell surface in association with the chaperone protein, gpl30 (Karasuyama et al, 1993; Shinjo et al., 1994).  Iga and IgP are encoded for by the mbl and B29 gene, respectively (Hermanson et al., 1988; Hombach et al, 1988; Sakaguchi et al, 1988; Campbell et al, 1991; Matsuo et al, 1991). Iga and IgP are both glycoproteins that possess an Ig-like extracellular domain followed by a highly charged spacer region. Within the spacer region are cysteine residues that form a disulfide bridge between Iga and IgP to form the Iga/p heterodimer.  The spacer region is followed by a  hydrophobic, 22 amino acid transmembrane region that anchors Iga/p to the plasma membrane. This transmembrane region is then followed by a cytoplasmic domain, which in murine cells contains 61 amino acids and 48 amino acids for Iga and IgP, respectively. Combined, these domains result in an approximately 34 kilodalton (kD) Iga protein and an approximately 39 kD IgP protein.  The Iga/p heterodimer is a constant component of all functional permutations of the B C R and is required for receptor signaling. The ability of the heterodimer to transduce signals is partiallydependent upon the Immunoreceptor Tyrosine-based Activation Motif (ITAM) that is contained within the cytoplasmic domain of both Iga and IgP (Reth, 1989). The I T A M is a conserved motif defined by the consensus sequence D/E(X) D/E(X) Y(X)2L/I(X) L/I (where D = aspartic acid, E 7  2  7  = glutamic acid, Y = tyrosine, L = leucine, I = isoleucine and X = any amino acid) that is typically associated with immunoreceptor signal transduction (Reth, 1989). This motif becomes tyrosine phosphorylated and enables Iga/p to associate with a variety of signaling proteins that can specifically recognize and bind to the phosphorylated I T A M via Src homology 2 (SH2) domains.  The surrogate light chain is comprised of two non-covalently associated Ig-like light chain proteins, V p  r e B  and X5, that are encoded for by the X5 gene (Kudo and Melchers, 1987; Tsubata  and Reth, 1990; Misener et al, 1991; Minegishi et al, 1999). Together, these two proteins are believed to be expressed on the cell surface in association with the chaperone protein, gpl30 (Karasuyama et al, 1993; Shinjo et al, 1994) where they associate with the Iga/p/calnexin complex to form the pro-BCR. While the function of the surrogate light chain is unclear it may have a role both in facilitating cell surface expression of the pro-BCR and in mediating some form of ligand binding and/or receptor aggregation such that pro-BCR signaling can be initiated. 5  The function of the pro-BCR itself remains unclear. However, the receptor may be involved in providing signals that help drive the developmental progression from pro-B cell to pre-B cell (Nagata ef a/., 1997).  The developmental successor to the pro-BCR is the pre-BCR which is expressed on pre-B cells (refer to Chapter 1.4). The pre-BCR (Fig. 1.1) is comprised of two non-covalently associated subunits namely, the Iga/p heterodimer and the membrane bound IgM subunit (mlgM) (reviewed in Martensson and Ceredig, 2000). The Iga/p heterodimer is essentially as described above except that it is now expressed on the cell surface in association with the mlgM subunit as opposed to calnexin (Tsubata and Reth, 1990; Minegishi et al., 1999). The mlgM subunit of the pre-BCR is a 67-78 kD, disulfide linked tetramer composed of a pair of surrogate light chain/u heavy chain heterodimers (i.e., mlgM =  [u]2[V eB/^5]2). pr  The surrogate light chain is essentially  as described above except that it is now expressed on the cell surface in association with the u heavy chain as opposed the gpl30 (Kudo and Melchers, 1987; Tsubata and Reth, 1990). Meanwhile, the murine u. heavy chain is encoded for by the heavy chain locus which undergoes random somatic recombination of the variable (V), diversity (D) and joining (J) segments during the transition from the pro-B to the pre-B cell stage. The structure of the u heavy chain includes an extracellular region comprised of a variable domain (VH) followed by four constant domains (CHI-4),  a transmembrane domain of 22 amino acids and a brief intracellular domain of 3 amino  acids. Once the pre-BCR is expressed on the cell surface the cell undergoes clonal selection whereby cells expressing inappropriately reactive receptors are eliminated (reviewed in Martensson and Ceredig, 2000).  Cells surviving clonal selection continue to develop,  progressing from the pre-B cell stage to the immature B cell stage.  The developmental successor to the pre-BCR is the B C R which is expressed on the surface of both immature and mature B cells (refer to Chapter 1.4).  The B C R (Fig. 1.1) is initially  comprised of two non-covalently associated functional subunits, the signal-transducing Iga/p heterodimer and the antigen-binding membrane Ig (mlg) subunit (Schamel and Reth, 2000). The Iga/p subunit of the B C R is essentially as described above for the pre-BCR. In contrast, the mlg subunit now contains rearranged light chains (either k or A,) in place of the surrogate light chain. The murine K and X light chains are encoded for by the light chain locus which undergoes random somatic recombination of the V and J segments during the transition from the pre-B cell stage to the immature B cell stage. The structure of the 25-28 kD light chain includes a variable 6  domain (VL) similar to the heavy chain, followed by a single constant domain (CL). Unlike, the heavy chain, the light chain does not contain a transmembrane or an intracellular domain but rather is expressed on the cell surface by way of its disulfide linkage to the u heavy chain. As well, the B C R mlg may contain one of five isotypes of the heavy chain including a, 8, s, y, and p (mlgA, mlgD, mlgE, mlgG and mlgM, respectively). Together, the variable domains of the light and heavy chain form an antigen binding region and in turn, two disulfide linked light chain/heavy chain complexes for a mlg complex. The mlg subunit must then associate with the Iga/p heterodimer to form a functional B C R that is expressed on the cell surface.  7  Figure 1.1. Schematic Representation of the BCR. The B C R is composed of an antigen-binding, membranebound subunit (mlg) that is non-covalently associated with a signaling, membrane-bound subunit (Iga/p). The mlg is composed of two disulfide linked heavy chains that are each disulfide linked to a light chain. The Iga/p subunit is composed of disulfide-linked Iga and IgP accessory chains, each of which contains an I T A M within its cytoplasmic tail. V and V represent the variable domains of the heavy and light chain, respectively. Ci_ represent the constant domains of the heavy and light chains. Ig-like indicates the Ig-like domains of Iga and Igp. The insets in the top right hand corner indicate the B C R as it is represented throughout this thesis. H  L  4  8  1.4 A Brief Overview of B Cell Development  From haematopoietic stem cell to antibody-secreting plasma cell, the B cell passes through multiple developmental checkpoints.  These checkpoints help to ensure that a self-tolerant,  diverse and maximally responsive B cell population is achieved.  A l l haematopoietic cells,  including the B cell, derive from the haematopoietic stem cell (HSC) within the bone marrow (Fig. 1.2). The HSC initially gives rise to a myeloid progenitor cell and a common lymphoid progenitor cell.  The myeloid progenitor subsequently gives rise to polymorphonuclear  leukocytes, erythrocytes and platelets while the common lymphoid progenitor gives rise to B and T cells.  The earliest defined B cell stage is that of the pro-B cell (Fig 1.3). The pro-B cell is typically characterized by cell-surface expression of the pro-BCR (described above) and is marked by the first V D J recombination event where the D and J segments of the heavy chain locus are recombined (i.e., DHJH rearrangement). Signaling studies have suggested that the pro-BCR may be involved in driving the pro-B cell to further differentiate into a pre-B cell (Nagata et al., 1997). However, a pro-BCR ligand and/or a mechanism for this process has yet to be identified. Regardless, the next defined B cell stage is that of the pre-B cell.  The pre-B cell is typically characterized by the cell surface expression of the pre-BCR (detailed above) and is marked by the completion of VHDRJH recombination of the u heavy chain locus. Following VHDHJH recombination the resultant u heavy chain is expressed on the cell surface in association with the surrogate light chain and the Iga/p heterodimer as the pre-BCR. Loss-offunction studies indicate that the pre-BCR is then involved in signaling the pre-B cell to cease u heavy chain rearrangements and to initiate VLJL recombination (Miyazaki et al., 1999; reviewed in Benschop and Cambier, 1999 and Martensson and Ceredig, 2000).  This cessation of  rearrangement is termed allelic exclusion and is necessary to ultimately ensure that each B cell expresses multiple copies of only one single-specificity antigen receptor. It is also important to note that the pre-B cell proliferates before initiating VLJL recombination and progressing to the immature B cell stage. This round of proliferation is important as it ultimately results in progeny cells, each of which contains an identically rearranged heavy chain associated  9  Haematopoietic Stem Cell ( H S C )  Myeloid Progenitor Cell  Polymorphonuclear Leukocytes  Platelets  Common Lymphoid Progenitor  B Cells  T Cells  Erythrocytes  Memory B Cells  Plasma B Cells  Some Cell Types Derived from the Haematopoietic Stem Cell. The Haematopoietic Stem Cell (HSC) differentiates to give rise to either a Myeloid Progenitor Cell (MPC) or a Common Lymphoid Progenitor (CLP) cell. The MPC goes on to differentiate into myeloid lineages including polymorphonuclear leukocytes, platelets and erythrocytes. The CLP goes on to differentiate into lymphoid lineages including B and T cells. B cells can further differentiate into Memory B Cells or into antibody secreting Plasma B Cells. Figure 1.2.  10  with a uniquely rearranged light chain; a situation which effectively increases the overall diversity of the receptor repertoire as both the heavy chain and light chain combine to form a specific antigen-binding site.  The immature B cell stage is characterized by the cell surface expression of the B C R (described above) where the rearranged light chain has replaced the surrogate light chain to produce a mature receptor. At this point the immature B cell circulates within the bone marrow where it undergoes the process of clonal selection whereby most self-reactive immature B cells are eliminated (reviewed in Rolink et al, 2001). Immature B cells that survive clonal selection then migrate from the bone marrow to the blood and lymphoid organs where they progress to the mature B cell stage.  The mature B cell, like the immature BCR, is characterized by the cell surface expression of the BCR. However, the mature B cell typically undergoes isotype switching such that the B C R is comprised predominantly of mlgD as opposed to the mlgM prevalent in immature B cells (reviewed in Benschop and Cambier, 1999). The mature B cell then circulates throughout the lymphoid system where it may encounter its cognate antigen. Upon such an encounter B C R signaling is initiated. Such signaling, i f accompanied by the appropriate co-stimulatory signals, culminates in the mature B cell clonally expanding and terminally differentiating into memory B cells and antibody-secreting plasma cells that ultimately provide a specific and long-term immune defense against the initiating pathogen.  11  antibodies  Developmental Stage  Pro-B Cell  Pre-B Cell  BCR Expression  Iga/(3 with calnexin  Heavy Chain  Immature BCell  Mature BCell  Plasma Cell  Iga/p, mlgM with surrogate light chain  Iga/p, mlgM  Iga/p mlgM & mlgD  germline  VDJ rearranged  VDJ rearranged  VDJ rearranged  soluble form (secreted antibodies) VDJ rearranged, soluble form  Light Chain  germline  germline  V J rearranged  V J rearranged  V J rearranged  Surface Markers  B220 , CD19 c-Kit ,  B220 , CD19 c-Kit",  B220 , CD19 c-Kit",  B220 , CD19 c-Kit",  X5\  X5 ,  mlgM", mlgD",  mIgM , mlgD",  +  +  +  +  +  +  +  +  +  X5\  X5\  +  mIgM , mIgD ,  mIgM , mlgD",  +  +  +  +  Figure. 1.3. Summary of Key B Cell Developmental Stages and Indication of Some of The Cell Properties Associated With the Respective Stages.  1.5 Compartmentalization and Signal Transduction Pathways  Before a discussion of B C R signaling can be initiated compartmentalization must first be considered.  Compartmentalization refers to the regulated localization of proteins to specific  cellular compartments  be they organelles, the cytosol, cellular membranes,  microdomains, or macromolecular complexes.  membrane  Such compartmentalization is essential to  maintain the integrity of all cellular signaling pathways.  Important to this process of  compartmentalization, and therefore important to any in-depth discussion of cellular signaling, are several highly conserved modular protein domains that mediate specific protein and lipid interactions. These domains include, but are not limited to, the Src homology 2 (SH2) domain, the Src homology 3 (SH3) domain, the pleckstrin homology (PH) domain, proline-rich motifs 12  and tyrosine-phosphorylation motifs (Fig. 1.4) (reviewed in Pawson and Nash, 2000; Castagnoli et al, 2004).  phosphoinositides l i p i d bilayer  SH3 domain  P-rich domain  SH2 domain  Y(P) motif  PH domain  Figure 1.4. Schematic Representation of Key Protein and Lipid Interaction Domains. Proteins often contain one or more conserved protein or lipid interaction domains including; SH3 domains that associate with proline rich (P-rich) domains, SH2 domains that associate with tyrosine phosphorylation motifs [Y(P)] and P H domains that associate with membrane-bound phosphoinositides.  SH2 domains are modules of approximately 100 amino acids that specifically bind to phosphorylated tyrosine motifs (reviewed in Pawson and Nash, 2000; Castagnoli et al, 2004). Different SH2 domains recognize different phosphorylated tyrosine motifs due to a secondary binding cleft within the domain that specifically recognizes the first six residues C-terminal to the phosphorylated tyrosine (reviewed in Pawson and Nash, 2000; Castagnoli et al, 2004). SH2 domains not only facilitate specific protein-protein interactions, they also enable these reactions to be regulated due to the fact that their binding site (i.e., the tyrosine) can be rapidly modified. More precisely, a protein's tyrosine phosphorylation status, and therefore its ability to associate with SH2 domain-containing proteins, can be regulated and modified by the coordinate action of protein tyrosine kinases (PTKs) which phosphorylate tyrosine residues and protein tyrosine phosphatases (PTPs) which dephosphorylate tyrosine residues.  SH3 domains are conserved modules that bind to proline rich motifs with the consensus sequence proline-x-x-proline where x represents any amino acid (reviewed in Pawson and Nash, 2000; Castagnoli et al, 2004). Similar to SH2 domains, different SH3 domains can specifically bind to different proline-rich motifs (reviewed in Pawson and Nash, 2000; Castagnoli et al, 2004). Similar to SH2 domains, SH3 domain-mediated interactions can be regulated through phosphorylation (Broome and Hunter, 1996; MacCarthy-Morrough et al, 1999). However, in 13  this case phosphorylation occurs on a residue within the SH3 domain to alter the domain's binding potential. Alternatively, phosphorylation of the SH3 domain-containing protein and/or its cognate proline-rich containing protein can cause the protein to undergo a conformational change such that their interaction domains' are subsequently revealed or concealed.  Dependent  on  their  particular  structure,  PH  domains  can  specifically  bind  to  phosphatidylinositol-4,5-bisphosphate (PI-4,5-P), phosphatidylinositol-3,4-bisphosphate (PI-3,4P), or phosphatidylinositol-3,4,5-trisphosphate (PI-3,4,5-P) (reviewed in Fruman et al., 1999). Because these phosphoinositides are membrane-bound, P H domains can serve to recruit their respective proteins to the membrane, presumably to juxtapose them with their activators and/or substrates. Importantly, membrane-recruitment of various P H domain-containing proteins can be regulated by regulating the membrane concentration of their cognate phosphoinositide. This can be  achieved  through  the  coordinate  action  of  phosphatidylinositol  kinases  and  phosphatidylinositol phosphatases. Proline-rich motifs and tyrosine phosphorylation motifs bind to SH3 and SH2 domains, respectively, as detailed above. Thus, described are several protein and lipid interaction domains that are crucial to cellular signaling in general and to B C R signaling in particular.  1.5.1  The Role of Adapter Proteins in Signal Transduction Pathways  As mentioned previously, compartmentalization is essential to maintain the integrity of all cellular signaling pathways. Such compartmentalization is achieved, in part, by adapter proteins. Adapter proteins are non-enzymatic proteins that contain one or more of the above-mentioned conserved protein-protein or protein-lipid interaction domains (reviewed in Leo and Schraven, 2001). These domains enable adapter proteins to function as molecular scaffolds upon which macromolecular signaling complexes can be specifically assembled and regulated (reviewed in Tomlinson et al., 2003; Jordon et al., 2000; Pawson and Nash, 2000). The function of adapter proteins is well-exemplified by the Ste5 adapter protein of Saccharomyces cerevisiae (S. cerevisiae) (reviewed in Schaeffer and Weber, 1999; Kelly and Chan, 2000). In S. cerevisiae the M A P K pathway can be activated by numerous environmental cues, and depending on these cues, can elicit numerous cellular outcomes (reviewed in Schaefer and Weber, 1999). The specificity of the M A P K pathway is determined in part by the Ste5 14  adapter protein which, upon pheromone stimulation, specifically assembles S t e l l (a M A P K kinase kinase) into a Ste20/Stell/Ste7/Fus3 M A P K signaling module that regulates the mating response (reviewed in Schaefer and Weber, 1999). Alternatively, Stel 1 can be assembled into a Shol/Stel l/Pbs2/Hogl M A P K signaling module that regulates the high osmolarity response (reviewed in Schaefer and Weber, 1999).  Thus, Ste5 establishes signaling specificity by  constraining promiscuous components of the M A P K pathway (i.e., Stel 1) into a specific M A P K signaling module and thereby limiting inappropriate cross-talk. Furthermore, the formation of a Ste5/MAPK signaling module is thought to enhance the kinetics of the M A P K cascade by providing the reactions with an entropic advantage (reviewed in Schaefer and Weber, 1999; Reinhardt, 2004).  Thus, the Ste5 adapter illustrates how adapter proteins can contribute to  protein compartmentalization and in doing so regulate both signal sensitivity and specificity.  Adapter proteins are critical components of B C R signaling. In fact, several severe phenotypes have been documented in knock-out mice that lack these adapters (Torres et al., 1996; Gong and Nussenzweig, 1996, Pappu et al, 1999; Hayashi et al, 2000; X u et al, 2000). However, these adapters are too numerous to be reviewed within this thesis and thus, the interested reader is directed to a series of excellent reviews on the topic including those by Tomlinson and colleagues (2000), by Leo and Schraven (2001), by Jordan and colleagues (2003) and by Reinhardt (2004). Rather, this thesis will focus on two key adapters that are involved in the BCR/PLCy pathway namely the Iga/(3 heterodimer and the B cell linker protein (BLNK).  1.5.2  The Role of Lipid Rafts in B C R Signaling  Proteins can be sequestered in lipid rafts. Lipid rafts are detergent-insoluble, sphingolipid- and cholesterol-rich membrane microdomains. Lipid rafts are also referred to as detergent-insoluble glycolipid-enriched membranes (DIGs), detergent-resistant membranes (DRMs) or glycolipidenriched membranes (GEMs) (reviewed in Simons and Ikonen, 1997; Brown and London, 2000; Simons and Toomre, 2000; Maxfield, 2002; Dykstra et al., 2003).  Rafts form when  sphingolipids and cholesterol associate into a liquid-ordered phase.  In the absence of  cholesterol, sphingolipids associate to form a relatively tightly packed gel-like bilayer. This tight association is facilitated by the sphingolipids' saturated acyl tails which have a relatively straight three-dimensional structure yet, is limited by the sphingolipids' hydrophobic head groups which are relatively bulky and result in voids existing between the acyl chains of adjacent 15  sphingolipids. Cholesterol is able to fill these voids to form a liquid-ordered phase. This liquid order phase floats as lipid rafts (-70-300 nm in size) within the liquid-disordered phase of the phospholipid bilayer (Figure 1.5).  Lipid rafts are evolutionarily conserved, appearing in Saccharomyces cerevisiae (yeast), Caenorhabitis elegans (worms), Drosophila melangaster (fruit flies) as well as in mammals (reviewed in Langlet et al, 2000).  This conservation, coupled with experimental evidence,  suggests that lipid rafts play a pivotal role in many cellular processes including in protein sorting and trafficking (reviewed in Brown and London, 1998; Gruenberg 2001), in cell migration (Manes et al, 1999; Manes et al, 2001), in cell adhesion (reviewed in Pande, 2000), in cellpolarity (reviewed in Gomez-Mouton et al, 2001) and in signal transduction (reviewed in Pierce et al, 2002, Manes et al, 2001, Simons and Toomre, 2000). Moreover, lipid rafts function in chemokine receptor signaling (reviewed in Manes et al, 2001), growth receptor signaling (reviewed in Simons and Toomre, 2000) and in immunoreceptor signaling (reviewed in Dykstra et al, 2003; Pierce, 2002; Cherukuri et al, 2001).  Central to their role in receptor signaling, lipid rafts segregate proteins within the lateral plane of the plasma membrane (i.e., compartmentalize proteins). Lipid rafts specifically include and/or exclude proteins from within their domain based on biochemical interactions between the lipid raft and the proteins. While the natures of these interactions have yet to be fully elucidated, some generalizations are becoming apparent (reviewed in Dykstra et al, 2003; Simons and Ikonen; 1997). For example glycophosphatidylinositol (GPI)-anchored proteins preferentially associate with lipid rafts (Danielsen, 1995; Fra et al, 1994; Brown and Rose, 1992) as do dually acylated proteins (e.g., Lyn and Blk of the Src family kinases) (Kosugi et al., 2001; McCabe and Berthiaume, 2001; McCabe and Berthiaume, 1999; van't Hof and Resh, 1997). A variety of cell surface receptors associate inducibly with lipid rafts following receptor engagement (reviewed in Pierce, 2002; Cheng et al, 2001; Manes et al, 2001; Simons and Toomre, 2000). Thus, lipid rafts may aid in the spatiotemporal regulation of signaling by serving as platforms upon which macromolecular signaling complexes can be specifically assembled and regulated.  Importantly, disruption of lipid rafts impairs BCR-induced PLCy signaling (Aman and Ravichandran, 2000; Guo et al, 2000). Therefore, lipid rafts may play an important role in the BCR/PLCy pathway. Chapter three of this thesis investigates the mechanisms regulating B C R 16  lipid raft translocation. At the onset of this thesis it was largely unclear how this event was mediated. Nonetheless, previous studies have demonstrated that the antigen-binding subunit (mlgM) of the BCR is capable of translocating into lipid rafts in the absence of Iga/p heterodimer (Cheng et al,  2001).  This suggested that the mlgM subunit may alone be  responsible for directing BCR lipid raft translocation upon BCR cross-linking.  Yet, the  possibility remained that the Iga/p heterodimer contributes to this process. Thus, I investigated whether the Iga/p heterodimer alone is able to translocate into the lipid rafts following crosslinking. These studies helped to define the molecular mechanisms underlying BCR lipid raft translocation and further defined the roles of the mlgM and the Iga/p heterodimer in BCR signaling.  17  extracellular  cytoplasmic  Legend  ^  phospholipid (unsaturated tail) phospholipid (saturated tail) sphingolipid  ft cholesterol  Figure 1.5 Schematic Representation of Lipid Rafts. Lipid rafts are sphingolipid- and cholesterol-rich membrane microdomains that are proposed to float within the phospholipid bilayer (reviewed in Simons and Ikonen, 1997; Brown and London, 2000; Simons and Toomre, 2000; Maxfield, 2002; Dykstra et al., 2003). Based on their biochemical structure lipid rafts are proposed to preferentially include or exclude various proteins within their domains and thus, represent a specific cellular compartment. The inset in the top right hand corner indicates lipid rafts as they are represented throughout this thesis.  18  1.6  An Overview of BCR Signaling  B cells play an integral role in the immune system's defense against invasive microorganisms. Requisite to the B cell's protective role is its ability to signal through cell-surface BCRs. Defects in initiating or transducing B C R signals can lead to immunodeficiency diseases (in the case of abrogated signaling) or to autoimmune diseases and lymphomas (in the case of aberrant signaling). Thus, B C R signaling has been extensively researched in an effort to understand and eradicate these diseases.  While the models of B C R signaling are ever-evolving, several key pathways have been largely identified and deciphered (reviewed in Kurosaki, 1999; Kurosaki, 2000; Kurosaki et al, 2000; Niiro and Clark, 2002; Gold, 2002; Pierce, 2002). B C R signaling involves three key pathways: the phosphatidyl-inositol 3 kinases (PI3K) pathway, the Ras/Mitogen Activated Protein Kinase (Ras/MAPK) pathway and the phospholipase C gamma (PLCy) pathway. These pathways are detailed below as being linear and independent. This is done to provide the reader with some initial clarity.  However, such an approach does little to illustrate the intricacies of B C R  signaling that are so essential to regulating B cell fate. Thus, the reader is asked to bear in mind that these pathways are truly interconnected, forming a B C R signaling network that ultimately exists influentially and receptively within the cell's complex signaling milieu.  1.6.1  Initiation of BCR Signaling  The B C R exists as pre-formed oligomeric complexes on the surface of resting mature B cells (Schamel and Reth, 2000; reviewed in Reth et al, 2000; Matsuuchi and Gold, 2001).  These  oligomers may weakly associate with downstream signaling components to provide a low level, antigen-dependent signal that is required for mature B cell survival (Lam et al, 1997; Schamel and Reth, 2000; reviewed in Reth et al, 2000; Matsuuchi and Gold, 2001).  Upon antigen-  binding, the oligomers coalesce into larger complexes that subsequently translocate into lipid rafts (Cheng et al, 1999; Aman and Ravichandran, 2000; Petrie et al, 2000; Weintraub et al, 2000; Dillon et al, 2000; Schamel and Reth, 2000; reviewed in Reth et al, 2000; Matsuuchi and Gold, 2001; Pierce, 2002). Such translocation co-localizes the B C R oligomers with proximal signaling components while isolating them from negative regulators, thereby facilitating B C R signaling (reviewed in Matsuuchi and Gold, 2001; Pierce, 2002; Dykstra et al, 2003). 19  Within the lipid rafts the B C R initiates a protein tyrosine phosphorylation cascade (Gold et al, 1990; Gold et al., 1991).  This cascade involves several protein tyrosine kinases (PTKs)  including the Src family kinase (SFK) members, Blk, Fyn, Lck and Lyn, the Tec family member, Bruton's tyrosine kinase (BTK), and the SH2 domain containing non-receptor protein tyrosine kinase, Syk (reviewed in Kurosaki et al., 2000). Initially, the translocated B C R associates with Lyn (reviewed in Kurosaki et al., 2000). Lyn is constitutively associated with and active within lipid rafts where it may be protected from its negative regulator Csk (Alland et al.., 1994; Koegl et al, 1994; Sigal et al, 1994; Cheng et al, 1999; reviewed in Resh, 1999). Furthermore, Lyn associates with the non-phosphorylated Iga/p motifs via its N-terminal domain (Plieman et al, 1994). Thus, Lyn is ideally positioned to associate with and phosphorylate the Iga/p ITAMs of the translocated BCR. Such phosphorylation facilitates further recruitment of Lyn, Blk, Fyn and Syk via their respective SH2 domains (Yamanashi et al, 1991; L i n and Justemant, 1992; Campbell and Sefton, 1992; Law et al, 1993; Burg et al, 1994; Clark et al, 1994; Johnson et al, 1995). This recruitment brings the SFKs into close proximity such that they can undergo autophosphorylation and activation (Kurosaki et al, 1994; Johnson et al, 1995; Siderenko et al, 1995; Takata and Kurosaki, 1995).  The active SFKs then further phosphorylate the Iga/p  heterodimer in a positive feedback loop. As well the SFKs phosphorylate and activate the recruited Syk and then, together with Syk, contribute to B T K phosphorylation and activation (Mahajan et al, 1995; Rawling et al, 1996; Afar et al, 1996; Kurosaki and Kurosaki 1997; Baba et al, 2001).  The active SFKs, Syk and B T K then go on to perpetuate a protein tyrosine  phosphorylation cascade that ultimately activates the PI3K, Ras/MAPK and PLCy pathways (Summarized in Fig. 1.6) (Nel et al., 1984; Campbell and Sefton, 1990; Gold et al, 1991; Law et al, 1992; reviewed in Kurosaki, 1999; Kurosaki, 2000; Niiro and Clark, 2002; Gold, 2002; Pierce, 2002).  20  Resting Mature B Cell  I  AjT7 l b Z1E7  /s^7  cytoplasm  /BTK7 ( PI3K )  Activated Mature B Cell  PIP  p7  3  s;| / ! s  1  0  <•>  §  1 1  V  CE1D  Legend  /  /  — © - ->  protein kinase  CZD  lipid kinase  CO  phospholipase  0  multistep positive interaction  phosphorylated tyrosine  Figure 1.6. Summary of the Initial Events of B C R Signaling. In resting mature B cells the B C R is dispersed throughout the plasma membrane. While an active portion of Lyn is localized to lipid rafts the majority of the SFKs as well as Syk, B T K , P I 3 K , M A P K and PLCy remain dispersed throughout the cytoplasm and inactive (top panel). Following B C R cross-linking the B C R translocates into lipid rafts where it associates with active Lyn and becomes tyrosine phosphorylated. The phosphorylated tyrosines act as docking sites for SH2 domain containing proteins including the SFKs and Syk. Once so docked the SFKs autophosphorylate and activate one another as well as phosphorylating and activating Syk. The SFKs and Syk then phosphorylate and activate B T K , which may be recruited to the plasma membrane and/or the B C R signaling complex by way of its SH2 domain binding to membrane-bound PIP . The active SFKs, Syk and B T K then go on to perpetuate a tyrosine phosphorylation cascade that ultimately activates the P I 3 K , Ras/MAPK and PLCy pathways. 3  21  1.6.2  The P I 3 K Pathway  The J3CR/PI3K pathway (Fig. 1.7) is required both for B cell development and for activation of mature B cells (Gold and Aebersold, 1994; Fruman et al, 1999; Suzuki et al, 1999; Clayton et al, 2002; reviewed in Gold, 2002; Okkenhaug and Vanhaesenbrock, 2003). In mature B cells, B C R cross-linking leads to plasma membrane recruitment and activation of PI3K (reviewed in Gold, 2002; Okkenhaug and Vanhaesenbrock, 2003). PI3K subsequently phosphorylates its membrane-associated substrates, phosphatidyl inositol-4-phosphate (PI-4-P) and phosphatidyl inositol-4,5-bisphosphate (PI-4,5-P2), to produce phosphatidyl inositol-3,4-bisphosphate (PI-3,4P2) and phosphatidyl inositol-3,4,5-trisphosphate (PI-3,4,5-P3) (reviewed in Gold, 2002; Okkenhaug and Vanhaesenbrock, 2003).  In turn, these products recruit various cytosolic  signaling proteins to the plasma membrane and the vicinity of the B C R signaling complex by way of their pleckstrin homology (PH) domains and thereby facilitate the activation of several downstream signaling pathways.  To date it remains unclear as to how BCR-induced PI3K plasma membrane recruitment and activation is mediated.  Nonetheless numerous different mechanisms have been proposed to  contribute to this process (reviewed in Gold et al, 1999). The predominant mechanism appears to involve PI3K being membrane-recruited via its SH2 domain. In particular, PI3K has been shown to bind to the membrane-associated, tyrosine phosphorylated proteins CD 19 (Tuveson et al, 1993), Gabl (Ingham et al, 2001), Gab2 (Gu et al, 1998) and B cell adapter for PI3K (BCAP) (Okada et al,  2000) following B C R cross-linking (reviewed in Gold, 2002).  Alternatively, PI3K may be membrane-recruited via its proline-rich domains binding to the Src homology 3 (SH3) domains of plasma membrane-associated SFKs (Prasad et al, 1993; Plieman et al, 1994b). Still, PI3K may be membrane-recruited via an interaction with membrane-bound, active Ras (Rodriquez-Viciana et al, 1994).  Regardless of the mechanism, PI3K plasma  membrane recruitment is necessary to bring PI3K in close proximity to its membrane-bound substrates which it then converts into PI-3,4-P2 and PI-3,4,5-P3.  22  Figure 1.7. Summary of the BCR/PI3K Pathway. B C R cross-linking initiates a protein tyrosine phosphorylation cascade. Among those proteins phosphorylated is the C D 19 co-receptor. PI3K is recruited to tyrosine phosphorylated C D 19 by way of its SH2 domains. Once recruited and activated PI3K phosphorylated its membrane-bound substrate PIP to produce PIP which then serves as a membrane docking site for numerous P H domain containing proteins. In particular, the Gabl and Gab2 (Gabl/2) adapter proteins are recruited to the plasma membrane by way of their SH2 domains. Gabl/2 subsequently become phosphorylated and serve as additional sites for PI3K membrane docking, thereby functioning as a positive feedback loop. As well, the kinases, PDK1, PDK2 and A K T are recruited to the membrane by way of their P H domains. PDK1 and PDK2 (PDK1/2) then phosphorylate and activate A K T which subsequently phosphorylates GSK-3J3, B A D , A S K 1 , M L K 3 , Caspase 9 and IKK to prevent apoptosis and promote cell survival. As well, PI3K activity has been shown to contribute to activation of the P K C which subsequently activates M A P K s (Erkl/2) and SAPKs (JNK and p38) (not shown). 2  3  23  Plasma membrane levels of PI-3,4-P and PI-3,4,5-P3 are tightly regulated as these molecules 2  influence numerous downstream signaling pathways. In terms of B C R signaling, the Gabl and Gab2 adapter proteins associate with PI-3,4,5-P3 via their respective P H domains (Gu et al, 1998; Ingham et al, 2001; reviewed in Gold, 2002). Once recruited to the plasma membrane, these proteins become tyrosine phosphorylated by BCR-associated and activated PTKs (Gu et al, 1998; Ingham et al, 2001; reviewed in Gold, 2002). These proteins then amplify the PI3K signal by further recruiting PI3K to the plasma membrane by way of its SH2 domain (Gu et al, 1998; Ingham et al, 2001; reviewed in Gold, 2002).  As well, PI-3,4,5-P facilitates the 3  membrane recruitment of phosphoinositide-dependent kinase 1 and 2 (PDK1 and PDK2) (Alessi et al, 1997a; Alessi et al, 1997b) and Akt (Burgering and Coffer, 1995; Stokoe et al., 1997) via their respective P H domains (Alessi et al, 1997a; Alessi et al, 1997b). Once recruited to the plasma membrane, PDK1 and PDK2 phosphorylate Akt and contribute to its activation (Alessi et al, 1997a; Alessi et al, 1997b; Stokoe et al, 1997; Anderson et al, 1998; Filipaa et al, 2000; Wick et al, 2000). Activated Akt then promotes B cell survival via several pathways, only some of which are detailed here. Along one pathway, Akt phosphorylates glycogen synthase 3p (GSK-3P) (Hajduch et al, 1998; van Weeren et al, 1998) preventing it from associating with and phosphorylating the P-catenin transcription factor (Cross et al, 1995; Pap and Cooper, 1998). This allows P-catenin to escape degradation and to translocate into the nucleus where it induces the transcription of cell-survival genes (Rubinfeld et al, 1996; Monick et al, 2001). Along another pathway Akt phosphorylates and inactivates the pro-apoptotic protein B A D thus further promoting B cell survival (Datta et al, 1997). Additionally, Akt phosphorylates caspase 9 preventing it from initiating a pro-apoptotic proteolytic cascade (Cardone et al, 1998). Moreover, Akt phosphorylates the IK kinase (IKK) which subsequently phosphorylates the inhibitor of N F - K B (IKB) targeting it for degradation and releasing the transcription factor, N F KB such that it can translocate into the nucleus where it too induces the transcription of cellsurvival genes (Scott et al, 1998; Kane et al, 1999; Ozes et al, 1999; Romashkova and Makarov, 1999). As well, Akt phosphorylates the apoptosis-signaling regulating kinase (ASK1) and M L K 3 , inhibiting their activity which ultimately results in the inhibition of a pro-apoptotic JNK pathway (Kim et al, 2001; Suhara et al, 2002; Barthwal et al, 2003). Thus, B C R crosslinking can be seen to promote B cell survival, in part, through the PI3K pathway.  24  1.6.3  The P L C y Pathway  The BCR/PLCy pathway (Fig. 1.8) is required both for B cell development and for mature B cell proliferation (Hashimoto et al, 2000; Wang et al, 2000). B C R cross-linking in mature B cells leads to plasma membrane-recruitment, tyrosine phosphorylation and activation of both P L C y l and PLCy2 (Bijsterbosch et al, 1985; Fahey and DeFranco, 1987; Carter et al, 1992; Coggeshall et al, 1991; Coggeshall et al, 1992; Hempel et al, 1992; Roifman and Wang, 1992; Siderenko et al, 1995; DeBell et al, 1999). The PLCy isoforms then hydrolyze plasma membrane-bound phosphatidyl inositol-3,4-bisphosphate (PI-3,4-P or, more simply, P I P 2 ) to produce plasma 2  membrane-bound diacylglycerol (DAG) and cytosolic inositol trisphosphate (IP3). In rum, these products initiate several well-characterized signaling pathways that eventually culminate in the nucleus (reviewed in Kurosaki et al, 2000; Gold; 2002) where they assist in promoting B cell survival and proliferation.  As recently as a decade ago the mechanisms coupling B C R cross-linking to PLCy membrane recruitment, phosphorylation and activation were enigmas. PLCy is proposed to be recruited to the B C R signaling complex via the B cell linker protein (BLNK) (Fu and Chan., 1997; Fu et al, 1998; Goitsuka et al, 1998; Wienands et al, 1998; Ishiai et al, 1999a; Ishiai et al, 1999b; reviewed in Kurosaki, 2000; Gold, 2002). In the simplest model B L N K is recruited to the B C R complex by way of its SH2 domain which binds to phosphorylated Iga tyrosine residues located outside of the I T A M (Engels et al, 2000, Kabak et al, 2002; reviewed in Gold, 2002). This brings B L N K into close proximity with BCR-associated Syk which subsequently phosphorylates B L N K on multiple tyrosine residues (Fu et al, 1998; Goitsuka et al, 1998; Chui et al, 2002). Tyrosine phosphorylated B L N K then associates with B T K (Hashimoto et al, 1999; Su et al, 1999; Chui et al, 2002) and PLCy (Fu and Chan, 1997; Fu et al, 1998; Ishiai et al, 1999; Chui et al, 2002) by way of their respective SH2 domains.  This results in the formation of a  BCR/PLCy signaling complex where the BCR, B L N K , the SFKs, Syk, B T K and PLCy are all in close association. This in rum, enables the SFKs (Mahajan et al, 1995; Rawlings et al, 1996; Afar et al, 1996) and Syk (Kurosaki et al, 1997; Baba et al, 2001) to contribute to B T K phosphorylation and activation.  Then B T K , (Takata and Kurosaki, 1996; Rawlings, 1999)  together with Syk (Takata et al, 1994), phosphorylates PLCy and contributes to its activation (reviewed in Kurosaki and Tsukada, 2000; Gold, 2002).  25  Once activated, PLCy hydrolyses its plasma membrane-bound substrate,  PIP2  to produce the  second messengers, IP3 and D A G (Bijsterbosch et al, 1985; Fahey and DeFranco, 1987; Rhee et al, 1989; reviewed in Marshall et al, 2000; Kurosaki  et al, 2000).  IP then binds to IP 3  3  receptors on the endoplasmic reticulum to induce an intracellular calcium flux that is followed by an extracellular calcium flux (Sugawara et al, 1997; Miyakawa et al, 1999; reviewed in Kurosaki  et al, 2000).  The elevated intracellular calcium levels activate several calcium-  dependent enzymes including the serine/threonine phosphatase, calcineurin. Active calcineurin dephosphorylates the cytosolic component of nuclear factor of activated T cells (NFAT ) which C  then translocates into the nucleus where it forms a variety of transcriptional activation complexes that ultimately promote B cell survival and proliferation (Timmerman et al, 1996; Dolmetsch et al, 1997; reviewed in Marshall et al, 2000; Gold, 2002).  The elevated intracellular calcium levels, along with D A G , also activate several isoforms of protein kinase C (PKC) (Sidorenko et al, 1996; Barbazuk and Gold, 1999; reviewed in Marshall et al, 2000).  Active P K C then contributes to the phosphorylation of IKB, targeting it for  degradation (DiDonato et al, 1997) and affecting the release of N F - K B .  NF-KB  then translocates  into the nucleus where it forms a variety of transcriptional activation complexes that ultimately contribute to B cell survival and proliferation (Lenardo and Baltimore, 1989; reviewed in Gold, 2002).  Elevated calcium and D A G levels also influence the Ras/MAPK pathway (pathway  detailed below in Chapter 1.5.4) (Casillas et al, 1991; Gold et al, 1992; reviewed in Gold, 2000). These effects can be mediated either by D A G binding to and activating the Ras guanyl nucleotide-releasing protein (RasGRP) that subsequently activates the pathway (Tognon et al, 1998) or by P K C phosphorylating and activating downstream components of the pathway such as Raf-1, M E K 1 or M E K 2 (reviewed in Gold, 2000). Regardless of the mechanism, activation of Ras/MAPK pathway ultimately contributes to B cell survival and proliferation. Thus, the BCR/PLCy pathway can be seen to promote B cell survival and proliferation both directly and indirectly.  26  Activated Mature B Cell  Figure 1.8. Summary of the B C R / P L C y Pathway. B C R cross-linking initiates a protein tyrosine phosphorylation cascade. Among the first proteins phosphorylated is the Iga/p heterodimer. The cytoplasmic adapter protein, B L N K is then recruited to the B C R signaling complex by way of its SH2 domain binding to phosphorylated Iga. Such recruitment is thought to facilitate B L N K ' s phosphorylation by Syk. B T K and PLCy are then recruited to the B C R complex by way of their SH2 domains binding to phosphorylated B L N K . The BCR-associated Syk and SFKs then phosphorylate the recruited B T K , contributing to its activation. In turn, B T K and Syk phosphorylate the recruited PLCy, contributing to its activation. Active PLCy then hydrolyzes its membrane bound substrate, PIP2, to produce the second messengers, IP and D A G . Subsequently, IP3 binds to the IP3 receptors, resulting in the release of intracellular calcium stores. The resultant flux in intracellular calcium leads to the activation of the protein kinase, calmodulin, which in turn phosphorylates and activates the protein phosphatase, calcineurin. Calcineurin then dephosphorylates the transcription factor, N F - A T , enabling it to translocate to the nucleus where it promotes the transcription of pro-survival genes. D A G , alone or in conjunction with the calcium flux, activates several forms of P K C which in turn contribute to the activation of the M A P K pathway (indicated by Erk 1/2 in the above) which also promotes cell survival (refer to Chapter 1.5.4 for further details of the M A P K pathway). 3  27  1.6.4 The R a s / M A P K Pathway  The BCPv/Ras/MAPK pathway is required for B cell development (Iritani et al, 1997) and for inducing proliferation of mature B cells (Richards et al., 2001). In mature B cells, BCR crosslinking leads to activation of the plasma membrane-associated GTPase, Ras, which in turn mediates the activation of a well characterized kinase cascade (Harwood and Cambier, 1993; Lazarus et al. 1993; Saxton et al., 1994; Tordai et al, 1994). Ras activation itself is mediated both by a PLCy-dependent pathway and by a PI3K-dependent pathway (reviewed in Gold, 2002) (Fig. 1.9).  In the PLCy-dependent pathway, the Ras guanyl nucleotide-releasing protein  (RasGRP) is recruited to the plasma membrane and activated via its interaction with D A G , a product of active PLCy (Tognon et al., 1998; Oh-hora et al., 2003). RasGRP then interacts with GDP-bound Ras where it exchanges guanine triphosphate (GTP) for the guanine diphosphate (GDP) thereby, activating Ras (reviewed in Gold, 2002). In the PI3K-dependent pathway, Ras is activated by the Son of Sevenless (SOS) guanine nucleotide exchange factor which is recruited to the plasma membrane via the formation of a PI3K-dependent multi-adapter complex (reviewed in Gold, 2002).  Initially, SOS associates with the adapter protein, Grb2.  The  SOS/Grb2 complex is then recruited from the cytosol to the plasma membrane via Grb2's SH2 domain which associates with either the tyrosine phosphorylated adapter protein, She (Saxton et al., 1994) or with the tyrosine phosphorylated Src homology 2 domain-containing protein tyrosine phosphatase 2 (SHP-2) (reviewed in Gold, 2002).  She and SHP-2 are themselves  localized to the plasma membrane via their SH2 domains which interact with tyrosine phosphorylated Gabl and/or Gab2 (Gu et al, 1998; Gold et al, 2000). And ultimately, Gabl and Gab2 are localized to the plasma via their P H domains which associate with the PI3K product, PI-3,4,5-P3 (as described in Chapter 1.5.2). Thus, BCR cross-linking can activate the Ras/MAPK pathway via a PLCy-dependent pathway or a PI3K-dependent pathway.  Once activated Ras associates with the 14-3-3/Raf complexes and dephosphorylates the 14-3-3 protein such that it can no longer inhibit Rafs activity (Cook and McCormick, 1993). Subsequently, the active Rafl phosphorylates and activates the dual-specificity M A P K kinases M E K 1 and M E K 2 . In turn, M E K 1 and M E K 2 phosphorylate the M A P K s , ERK1 and ERK2 (reviewed in Su and Karin, 1996). The phosphorylated ERKs then translocate into the nucleus where they phosphorylate multiple transcription factors that ultimately up-regulate  28  the  expression of proliferative genes (reviewed in Su and Karin, 1996). Thereby, B C R cross-linking can be seen to promote B cell proliferation, in part, through the Ras/MAPK pathway.  29  PLC-Dependent Pathway  PI3K-Dependent Pathway  /Erk1/2X  Figure 1.9. The R a s / M A P K Pathway. Ras activation is mediated both by a PLCy-dependent pathway and by a PI3K-dependent pathway (reviewed in Gold, 2002). In the PLCy-dependent pathway, RasGRP is recruited to the plasma membrane and activated via its interaction with D A G , a product of active PLCy. RasGRP then interacts with GDP-bound Ras where it exchanges G T P for G D P thereby activating Ras. In the PI3K-dependent pathway, Ras is activated by the SOS guanine nucleotide exchange factor which is recruited to the plasma membrane via the formation of a PI3K-dependent multi-adapter complex (reviewed in Gold, 2002). Initially, SOS associates with the adapter protein, Grb2. The SOS/Grb2 complex is then recruited from the cytosol to the plasma membrane via Grb2's SH2 domain which associates with either tyrosine phosphorylated She or tyrosine phosphorylated SHP-2. She and SHP-2 are themselves localized to the plasma membrane via their SH2 domains which interact with tyrosine phosphorylated Gabl and Gab2. Gabl and Gab2 are themselves localized to the plasma via their P H domains which associate with the PI3K product, PI-3,4,5-P . Interestingly, B L N K has also been shown to associating with the Grb2 adapter protein forming a complex of unknown consequence. Once activated, Ras phosphorylates and activates the M A P K kinase kinase ( M A P K K K ) , Rafl. Subsequently, Rafl phosphorylates activates the dual-specificity M A P K kinases MEK1 and M E K 2 . M E K 1 and M E K 2 then phosphorylate the M A P K s , ERK1 and E R K 2 enabling them to translocate into the nucleus where they phosphorylate multiple transcription factors that ultimately up-regulate the expression of proliferative genes. A n d thus, B C R cross-linking promotes B cell proliferation, in part, through a PLCy-dependent and a PI3K-dependent Ras/MAPK pathway. 3  30  1.7 The Enigmas of BCR Signaling  As evidenced by the above discussion, our knowledge of B C R signaling has greatly advanced in the past quarter century. Loss-of-fiinction studies have clearly established the role of PTKs and the Ras/MAPK, PI3K, and PLCy pathways within this process.  As well, the downstream  components and consequences of these pathways have been well-elucidated. Nonetheless, many aspects of B C R signaling remain enigmatic and thus the investigation into B C R signaling continues. This thesis has focused primarily on elucidating the questions that remain regarding the BCR/PLCy pathway. Therefore, to provide the reader with a better understanding of these questions and the purpose of this thesis, a more detailed account of our current understanding of the BCR/PLCy pathway is provided below.  1.8 The BCR/PLCy Pathway in Detail  1.8.1  PLCy Structure and Function  The phospholipase C (PLC) family is divided into three subfamilies termed PLC(3, PLC5, and PLCy (reviewed in Katan, 1998; Sekiya et al, 1999; Rhee, 2001). Common to all P L C family members are, from the N-terminus to the C-terminus, a P H domain, several EF-hand domains, a C2 domain, an X domain and a Y domain (Fig. 1.9). The P H domain binds to membrane-bound phosphoinositides and may assist in targeting the typically cytosolic P L C family members to the plasma membrane where they have access to their plasma membrane-bound substrate. The C2 domain binds to phosphoinositides in a calcium-dependent manner and may target P L C family members to the plasma membrane. The EF-hand domains appear to confer flexibility on the proteins such that they can undergo changes in their tertiary structure and consequently, their activity. And finally, the X and Y domains together form the catalytic site that is responsible for hydrolyzing PIP2 to produce the second messengers, D A G and IP3.  The PLCy subfamily structurally differs from the PLCP and PLC5 subfamilies by containing an approximately 400 amino acid insert between the X and Y domains (reviewed in Katan, 1998; Sekiya et al, 1999). This insert contains multiple interaction domains that include two tandem SH2 domains and an SH3 domain that are flanked by a fragmented P H domain. The SH2 domains bind to tyrosine-phosphorylated proteins, the SH3 domain binds to proline-rich proteins 31  and the fragmented P H domain is proposed to form a functional domain in the tertiary structure that also binds to phosphoinositides. Together these domains mediate PLCy's association with various signaling proteins and ultimately contribute to regulating its activation (reviewed in Rhee and Bae, 1997; Sekiya et al, 1999; Kurosaki et al, 2000; Marshall et al, 2000).  The PLCy subfamily includes P L C y l (-150 kD) and PLCy2 (-145 kD). These proteins are similarly organized, share 50% amino acid sequence identity and share very similar tertiary structures (Rhee et al, 1989; reviewed in Sekiya et al, 1999; Rhee, 2001).  Despite these  similarities, the PLCy isoforms show differences in their expression patterns and in their roles in cellular signaling (reviewed in Carpenter and Ji, 1999; Sekiya et al, 1999; Kurosaki et al, 2000; Marshall et al, 2000; Rhee, 2001). P L C y l is ubiquitously expressed and is absolutely required for mammalian growth and development (Ji et al, 1997; reviewed in Rhee and Bae, 1997; Carpenter and Ji, 1999; Sekiya et al, 1999). In contrast, PLCy2 expression appears to be largely restricted to B cells where it is required for B C R signaling (Takata et al, 1995; reviewed in Noh et al, 1995, Sekiya et al, 1999; Rhee, 2001).  A l l told, the P L C family is responsive to over one hundred cellular receptors and as such is involved in numerous cellular processes (reviewed in Rhee, 2001).  The combination and  specific amino acid sequence of the various regulatory regions (i.e., the P H , SH2, SH3, EF-hand and C2 domains) determines the responsiveness of the various P L C isoforms. Based on this, PLCP isoforms respond primarily to G protein-coupled receptors, while PLC8 isoforms respond primarily to increases in intracellular calcium levels and PLCy isoforms (Fig. 1.10) respond primarily to receptor and non-receptor tyrosine kinases (reviewed in Rhee and Bae, 1997; Carpenter and Ji, 1999; Sekiya et al, 1999; Rhee, 2001). O f particular interest to this thesis is the responsiveness of the PLCy isoforms to the BCR.  32  PH  •.X....  EF  P|SH2  SH3|!L_  SH2  Y.  - —  •  ,.C2  .  PLCyl  PLCy2  I  1 « 100 amino acids  Figure 1.10. General Structure of P L C y l and PLCy2. From the N-terminal to the C-terminal, P L C y l and PLCy2 contain a phosphoinositide binding PH domain, followed by a calcium binding E F hand, followed by the X domain (a highly conserved domain of P L C isozymes that, together with the Y domain, forms the catalytic core of the enzyme), followed by a second PH domain that is interrupted by two phosphotyrosine-binding SH2 domains and a proline-binding SH3 domain, followed by the Y domain, followed by a C2 domain that is proposed to bind to calcium and to mediate calcium-dependent association with phosphoinositides.  P L C activity was implicated in B C R signaling when it was observed that B C R cross-linking results in D A G and IP3 production (Coggeshall and Cambier, 1984; Bijsterbosch et al, 1985; Klaus et al, 1985; Fahey and DeFranco; 1987). The PLCy isoform was further implicated as P L C y l and PLCy2 were found to be tyrosine phosphorylated following B C R cross-linking (Carter et al, 1991; Coggeshall et al, 1992; Hempel et al, 1992; Roifman and Wang, 1992). And finally, PLCy2 was identified as the primary mediator of the B C R / P L C pathway based on loss-of-functions studies (Takata et al, 1995; Hashimoto et al., 2000; Wang et al, 2000). Having established the involvement and importance of PLCy in B C R signaling, ensuing investigations focused on defining the proximal events in the BCR/PLCy pathway.  1.8.2  BCR-Induced Tyrosine Phosphorylation of P L C y  Plasma membrane recruitment and tyrosine phosphorylation are requisite steps in the pathway to PLCy activation.  Unfortunately, two steps a pathway does not make.  Thus, today's B C R  researchers proceed as cartographers seeking to map out the molecular pathways that connect the B C R to PLCy membrane recruitment, tyrosine phosphorylation and activation.  B C R cross-linking initiates a protein tyrosine phosphorylation cascade that is manifested by members of the SFK, Syk and B T K (Gold et al, 1990; Gold et al, 1991; reviewed in Kurosaki et al, 2000; Marshall et al, 2000).  This cascade is required to connect the B C R to PLCy 33  activation (Padeh et al., 1991).  While all of these PTKs appear to contribute to optimal  BCR/PLCy signaling, Syk and B T K both have a direct role in phosphorylating and activating PLCy (reviewed in Kurosaki et al, 2000; Marshall et al, 2000). Genetic ablation of Syk results in an almost complete loss of PLCy phosphorylation and activation (as indicated by abrogated IP3 generation and calcium flux) following B C R cross-linking in the chicken DT40 cell line suggesting that it is absolutely required for PLCy activation. (Takata et al., 1994). Additionally, in vitro studies demonstrate that Syk can directly phosphorylate P L C y l on the key regulatory residue, tyrosine 783 (Y783) (Law et al, 1996). Syk ablation does not completely inhibit PLCy phosphorylation, suggesting that another P T K may contribute to this process. This P T K could be Lyn as in vitro studies indicate that Lyn can directly phosphorylate P L C y l on tyrosine 771. However, genetic ablation of Lyn has a minimal impact on PLCy phosphorylation and activation following B C R cross-linking in the chicken DT40 cell line (Takata et ah, 1994). Interestingly, genetic ablation of B T K only slightly diminishes PLCy phosphorylation while significantly inhibiting PLCy activation (as evidenced by abrogated IP3 production) (Takata and Kurosaki, 1996). Concordantly, human B cell lines from X-linked agammaglobulinemia patients lacking a functional B T K gene show impaired BCR-induced IP3 production and calcium mobilization (Fluckiger et al, 1998). Moreover, reconstitution studies demonstrate that ectopic expression of B T K is sufficient to rescue these pathways in these cells (Fluckiger et ah, 1998) Finally, in vitro studies demonstrate that human B T K can directly phosphorylate rat PLCy2 on several tyrosine residues including tyrosines 753, 759, 1197 and 1217. Thus, it is concluded that PLCy becomes directly phosphorylated and activated by the co-operative activities of Syk and B T K following B C R cross-linking. Having established the role Syk and B T K in BCR-induced PLCy phosphorylation, the questions naturally arise as to how these PTKs themselves are activated and as to how they come to associate with PLCy following B C R cross-linking. The former question will be addressed here while the latter question is addressed in detail in Chapter 1.9.3. Syk is a member of the ZAP-70/Syk family of non-receptor PTKs and is primarily expressed in B cells (Hutchcroft et al, 1991; Hutchroft et al, 1992; Kong et al, 1995). Structurally Syk is a 72 kD, cytosolic protein that possesses two N-terminal SH2 domains and a C-terminal kinase domain (also known as a Src Homology 1 Domain or SHI domain) (Fig. 1.11) (Zioncheck et al, 1986; Zioncheck et al, 1988; Taniguchi et al, 1991). The SH2 domains are proposed to bind to the phosphorylated Iga/p heterodimer and thus recruit Syk to the B C R signaling complex 34  following BCR-cross linking (Law et al,  1994; Rowley et al,  1995).  This recruitment  juxtaposes Syk with the BCR-associated SFKs. The SFKs then phosphorylate and activate Syk. In particular, L y n is implicated in this process as genetic ablation of Lyn dramatically inhibits Syk activation following B C R cross-linking in the chicken DT40 B cells (Kurosaki et al, 1994). However, it is unclear whether Lyn directly phosphorylates Syk or whether Lyn is simply required upstream of this process to phosphorylate the Iga/p heterodimer and thus facilitate Syk's membrane recruitment. Notably, some Syk activity persists in the L y n knockout system suggesting that a Lyn-independent mechanism also exists for coupling the B C R to Syk activation (Kurosaki et al, 1994). While this Lyn-independent mechanism remains unclear it is important as it helps us to reconcile the findings that Syk, but not Lyn, is required for PLCy activation (reviewed in Kurosaki et al, 2000; Takata et al., 1994). Regardless of mechanism, B C R crosslinking clearly leads to plasma membrane recruitment, phosphorylation and activation of Syk (Kurosaki et al, 1994; Law et al., 1994; Rowley et al., 1995) and Syk activity is clearly required to maintain a functional BCR/PLCy pathway (Takata et al, 1994; reviewed in Kurosaki et al., 2000). B T K is a member of the Tec family of PTKs involved in B T K signaling (Saouaf et al, 1994; Aoki et al, 1994; De Weers et al, 1994). Structurally B T K is a 76 kD, cytosolic protein that possesses an N-terminal P H domain followed by an SH3 domain, followed by an SH2 domain, followed by a C-terminal kinase domain (Fig. 1.11) (reviewed in Lewis et al, 2001). Mutational studies show that the P H domain and the SH2 domain are required to maintain B T K ' s role in the BCR/PLCy pathway (Takata and Kurosaki, 1996).  The P H domain is proposed to bind to  membrane-bound PI-3,4,5-P3 and thereby recruit B T K to the plasma membrane where it will be in close proximity to the B C R signaling complex (Salim Varnai  et al, 1999; reviewed in Kurosaki  et al, 1996; Kojima et al, 1997;  et al ., 2000).  This recruitment is thought to  facilitate B T K ' s phosphorylation by the BCR-associated SFKs and/or Syk (Mahajan et al, 1995; Afar et al, 1996; Rawlings et al, 1996; Kurosaki and Kurosaki, 1997; Baba et al, 2001). However, the precise mechanism of B T K phosphorylation remains controversial.  Loss-of-  function studies clearly indicate that Syk and Lyn are required for this process yet it is unknown if this requirement reflects indirect or direct interactions.  For example Lyn may simply be  required to activate Syk as discussed above. Alternatively, L y n could directly phosphorylate B T K itself as is suggested by reconstitution studies in various non-lymphoid cells (Afar et al., 1996; Rawlings et al, 1996).  Similarly, Syk could directly phosphorylate B T K or it may 35  phosphorylate some other unidentified P T K that then goes on to phosphorylate B T K . Whatever the case, ablation of either Syk or Lyn only partially inhibits B T K phosphorylation while ablation of both Syk and Lyn completely inhibits B T K phosphorylation (Kurosaki and Kurosaki, 1997). Thus, both proteins are proposed to function cooperatively to activate B T K (reviewed in Kurosaki, 2000).  ,  SH2  ||SH2  Tec  ,, , .-, „  - ... —~T—  r  r-—r-r  kinase  .•  , kinase  SH3.I ISH2  Syk  BTK  = 100 amino acids  Figure 1.11. General Structure of Syk and BTK. Syk is a 72 kD, cytosolic protein that possesses two N-terminal SH2 domains and a C-terminal kinase domain (also known as an Src Homology 1 Domain or SHI domain). B T K is a 76 kD, cytosolic protein that possesses an N-terminal P H domain followed by an conserved Tec domain (common to all Tec family members), followed by an SH3 domain, followed by an SH2 domain, followed by a C-terminal kinase domain. Recall that SH2 domains specifically associate with phosphorylated tyrosine residues, P H domains specifically associate with membrane bound phosphoinositides and SH3 domains specifically associate with prolinerich sequences. Note that the N-terminal is to the left and the C-terminal is to the right in the diagram.  1.8.3  BCR-Induced Membrane Recruitment of P L C y  Thus sketched, is our current understanding of the pathways that connect the B C R to Syk and B T K activation. Yet the question remains, "How do the membrane-localized, active Syk and B T K come to associate with and activate cytosolic PLCy?" Recent loss-of-function studies have made significant contributions to an emerging model of this process. Initial models suggested that PLCy may be recruited to the plasma membrane by way of its P H domains binding to membrane-bound  PIP3.  These models were based on analogy to the platelet-derived growth  factor receptor (PDGFR)/PLCy pathway where loss-of-function studies demonstrated that P L C y l ' s N-terminal P H domain is required for its membrane recruitment and activation (Falasca et al, 1998; reviewed in Marshall et al, 2000; Rhee, 2001). According to these models, PLCy recruitment and activation would lie downstream of PI3K as PIP3 is a PI3K product. Indeed, inhibitors of PI3K have been shown to have a negative effect of PLCy activation in B cells  36  (reviewed in Rhee, 2001). indirect effect.  However, it remains unclear whether this represents a direct or  The possibility of an indirect effect exists as PI3K and its product,  PIP3,  may  recruit and activate B T K , which itself is proposed to contribute to PLCy phosphorylation and activation (refer to Chapter 1.9.2). Thus, it remains unclear whether or not PLCy's P H domain plays a significant role in mediating PLCy's membrane recruitment and activation in the BCR/PLCy pathway.  Alternate models propose that PLCy may be recruited to the plasma membrane by way of its SH2 domain, similar to what is observed in several receptor PTK/PLCy pathways (Anderson et al, 1990; Kashishian and Cooper, 1993; Margolis et al, 1990; Valius et al, 1993; Zhu et al, 1992).  At the outset, BCR-associated Syk was proposed to facilitate PLCy  recruitment because P L C y l ' s  membrane  SH2 domain binds to tyrosine phosphorylated Syk in an  association that is required for P L C y l ' s membrane recruitment, phosphorylation and activation (Sillman and Monroe, 1995; Law et al, 1996).  Unfortunately, this model does not seem  applicable to PLCy2, the predominate isoform of the BCR/PLCy pathway (reviewed in Kurosaki et al, 2000).  This necessitated further studies that eventually identified B L N K as a key  component in the BCR/PLCy pathway.  B L N K was first isolated from a human B cell line based on its ability to associate with the Cterminal SH2 domain of P L C y l (Fu and Chan, 1997). Interestingly, human B L N K exists in two alternatively spliced forms termed h B L N K (~ 70 kD when phosphorylated) and hBLNK-s (~ 68 kD when phosphorylated) that do not overtly differ in expression pattern or function (Fu and Chan; 1997; Fu et al, 1998). Concurrently, the mouse homolog of B L N K , termed the SH2 domain containing leukocyte protein of 65 kD (Slp-65) or murine B L N K (mBLNK) was isolated from a murine B cell line and shares 82 % amino acid identity with h B L N K (Wienands et al, 1998; Fu et al, 1998). As well, the chicken homologue of B L N K , termed the B lymphocyte adapter protein containing a SH2 domain (BASH) or the chicken phosphorylated protein of 80 kD (pp80) has been isolated (Goitsuka et al, 1998; Ishiai et al, 1999).  The structure of h B L N K is detailed herein because h B L N K is used throughout this thesis (Fig. 1.11). Nonetheless, the general structure of B L N K is conserved across all its homologues (Fu and Chan, 1997; Fu et al, 1998; Goitsuka et al, 1998; Wienands et al, 1998; Ishiai et al, 1999a; Ishiai et al, 1999b). From the N-terminus to the C-terminus, h B L N K is comprised of a 37  basic domain (approximately 50 amino acids), followed by an acidic domain (approximately 70 amino acids), followed by a proline-rich domain (approximately 250 amino acids), followed by an N-terminal SH2 domain (approximately 110 amino acids) (Fig. 1.12). As well, these domains are interspersed with thirteen tyrosine phosphorylation motifs (Chiu et al, 2002).  Thus,  structurally h B L N K appears to be an ideal candidate for an adapter protein as it contains multiple protein interaction domains while lacking a catalytic domain.  This supposition is further  supported by numerous functional studies (discussed below).  -A H* OH'  l  P-rich  SH2  BLNK  1 « 100 amino acids  Figure 1.12. General Structure of h B L N K . From the N-terminus to the C-terminus (left to right in the above diagram) h B L N K is comprised of a basic domain (approximately 50 amino acids in length), followed by an acidic domain (approximately 70 amino acids length), followed by a proline-rich domain (approximately 250 amino acids in length), followed by an N-terminal SH2 domain (approximately 110 amino acids in length). Recall that prolinerich domains specifically associate with P H domains and SH2 domains specifically associate with phosphorylated tyrosine residues.  B L N K was first identified as a component of B C R signaling pathways when it was observed to become membrane-recruited and tyrosine phosphorylated following B C R cross-linking (Fu and Chan, 1997; Fu et al., 1998; Goitsuka et al, 1998; Wienands et al, 1998). Loss-of-function studies demonstrate that B L N K is required for B cell development and activation (Minegishi et al, 1999; Pappu et al, 1999; Hayashi et al, 2000; X u et al, 2000; Jumaa et al, 2001; Tan et al, 2001; Schebesta et al, 2002 X u and Lam, 2002; Hayashi et al, 2004; Taguchi et al, 2004). These studies, combined with mutational and protein association studies, have led to a model of B C R signaling wherein B L N K lies at the junction of several key signaling cross-roads including the PLCy pathway and the M A P K pathway. Of particular interest to this thesis is B L N K ' s role in the BCR/PLCy pathway. B L N K was first implicated in the PLCy pathway when it was found to associate with both P L C y l and PLCy2 38  following B C R cross-linking (Fu and Chan, 1997; Fu et al, 1998; Ishiai et al, 1999a; Ishiai et al, 1999b; Chui et al, 2002). Since then studies have established a model whereby B L N K is required to facilitate PLCy's membrane recruitment, tyrosine phosphorylation and activation following B C R cross-linking (Fig. 1.12) (Ishiai et al, 1999; reviewed in Kurosaki and Tsukada, 2000; Kurosaki et al, 2000; Marshall et al, 2000). As described in Chapter 1.5.3, this model envisions cytoplasmic B L N K being recruited to the B C R signaling complex by way of its SH2 domain binding to the non-ITAM, phosphorylated tyrosine residue 204 of Iga following B C R cross-linking (Engels et al, 2001; Kabak et al, 2002). This brings B L N K into.close proximity with BCR-associated Syk which subsequently phosphorylates B L N K on multiple tyrosine residues (Fu et al, 1998; Goitsuka et al, 1998; Chui et al, 2002). Tyrosine phosphorylated B L N K then associates with B T K (Hashimoto et al, 1999; Su et al, 1999; Chui et al, 2002) and PLCy (Fu and Chan, 1997; Fu et al, 1998; Ishiai et al, 1999; Chui et al, 2002) by way of their respective SH2 domains. This results in a BCR/PLCy signaling complex with the BCR, B L N K , the SFKs, Syk, B T K and PLCy all in close association. This in turn, enables the SFKs (Mahajan et al, 1995; Rawlings et al, 1996; Afar et al, 1996) and Syk (Kurosaki et al, 1997; Baba et al, 2001) to contribute to B T K phosphorylation and activation. Then, B T K (Takata and Kurosaki, 1996; Rawlings, 1999) together with Syk (Takata et al, 1994), phosphorylates PLCy and contributes to its activation (reviewed in Kurosaki and Tsukada, 2000; Gold, 2002). PLCy then hydrolyzes  PIP2  to IP3 and D A G .  These secondary messengers subsequently activate several  downstream signaling pathways that ultimately influence transcription and contribute to directing B cell activation, proliferation and differentiation (as described in Chapter 1.5.3).  As mentioned above, B L N K has also been implicated in several other B C R signaling pathways including the M A P K and stress-activated protein kinase (SAPK; i.e., p38 and Jnk) pathways. At first, B L N K ' s involvement in this pathway may seem intuitive as the PLCy pathway has been shown to intersect with the M A P K pathway at the location of DAG-mediated recruitment of the Ras-GRP protein (refer to chapter 1.5.3 and Fig. 1.13). However, B L N K may also directly affect the M A P K pathway through its association with the Grb2 adapter protein (Fu et al, 1998). While the physiological significance of this association remains unclear, B L N K may recruit Grb2 and its associated Ras guanine nucleotide exchange factor, Son of Sevenless (SoS), to the proximity of Ras, thereby facilitating Ras' activation and subsequently, the phosphorylation and activation of Erk (Fig. 1.13) (Fu et al, 1998; reviewed in Gold, 2002). Thus, it appears that B L N K may facilitate Erk phosphorylation via a PLC-dependent and a PLC-independent 39  pathway. This is of particular consequence to this thesis as Erk phosphorylation is used as a downstream indicator of PLCy activation in a reconstitution system involving B L N K .  To  distinguish between PLC-dependent and PLC-independent contributions to Erk phosphorylation I used the P L C inhibitor, U73122. phosphorylation  should be  In particular, PLC-dependent reconstitution of Erk  inhibited by U73122 treatment  whereas PLC-independent  reconstitution of Erk phosphorylation should not be inhibited by U73122.  Tyrosine phosphorylated B L N K also associates with the adapter protein Cbl (Yasuda et al., 2000).  This protein is a negative regulator of the BCR/PLCy pathway as it binds to  phosphorylated B L N K and competitively inhibits B L N K from associating with PLCy (Yasuda et ah, 2000). Consequently, PLCy is neither recruited to the B C R complex, phosphorylated nor activated (Yasuda et al., 2000). Given this, B L N K may serve as a site of both negative and positive regulation in the BCR/PLCy pathway.  When all the above is considered a map of the pathways connecting the B C R to PLCy begins to emerge. Yet at times our map reads more as a list of signposts, proposing steps along the way rather than as a clearly defined pathway. Thus, the endeavor to fill in the details of the pathways continues in an effort to better understand these pathways and how they may go astray.  The first portion of the pathway requiring further definition is that connecting the B C R to B L N K membrane recruitment. At the onset of this thesis this pathway was completely obscure. But, by way of analogy with Slp-76, B L N K ' s homologue in the TCR/PLCy signaling pathway, B L N K might be recruited to the plasma membrane by way of its SH2 domain binding to a tyrosine phosphorylated, transmembrane adapter protein (akin to the Linker of Activated T cells [LAT] in TCR signaling). Several possible candidates for this protein were identified including the Linker for Activated B cells (LAB) (Janssen et al, 2003).  L A B is a transmembrane protein that  contains multiple tyrosines within its cytoplasmic domain that become phosphorylated upon B C R cross-linking suggesting that it may function as a membrane docking site for SH2 domaincontaining proteins (Janssen et al., 2003). Furthermore, genetic ablation of L A B reduces BCRmediated calcium flux and Erk activation suggesting a role for it in the BCR/PLCy pathway. Thus, L A B may mediate B L N K membrane recruitment. However, co-association studies failed to detect an association between endogenous L A B and endogenous B L N K (Janssen et al., 2003). In contrast, in vitro studies did indicate that L A B can bind the SH2 domains of PLCy suggesting 40  the possibility that L A B directly recruits PLCy to the membrane (Janssen et al, 2003). Yet, this observation could not be repeated in vivo, calling into question the validity of this model. Interestingly, L A B does associate with Grb2 in vivo (Jansenn et al., 2003). This suggests a mechanism for LAB-mediated membrane recruitment of B L N K as proline-rich B L N K associates with Grb2's SH3 domain (Wienands et al, 1998). Still, further studies are required to validate this model. In particular, mutational analysis can be performed to determine i f mutations of Grb2's SH3 domain and/or B L N K ' s proline rich domain have a significant impact on BCRmediated B L N K  membrane recruitment and phosphorylation and on PLCy  membrane  recruitment, phosphorylation and activation. A n alternative model for B L N K membrane recruitment was suggested based on its association with B T K . In particular, B T K could itself be recruited to the plasma membrane by way of its P H domain binding to membrane-bound phosphoinositides (Salim et al, 1996; Kojima et al, 1997; Varnai et al, 1999; reviewed in Kurosaki et al., 2000). In turn, B T K could serve as a docking site for B L N K .  However, this seems unlikely given that B T K ' s association with B L N K is  mediated by B T K ' s SH2 domain binding to tyrosine phosphorylated B L N K (Hashimoto et al, 1999; Su et al, 1999). Such a model would require B L N K to become tyrosine phosphorylated prior to being membrane recruited, which is unlikely given that B L N K is presumably phosphorylated by BCR-associated, membrane localized Syk.  A third model for B L N K membrane recruitment is suggested based on its association with phosphorylated Syk (Engels and Wienands, unpublished results; reported in Engels et al, 2001). This association suggests B L N K could be recruited to the plasma membrane by way of its SH2 domain binding to BCR-associated, phosphorylated Syk. However, further studies supporting this model have not been reported.  Finally, Iga may be another candidate membrane docking protein for B L N K .  Mutational  analysis of Iga tyrosines indicated that the non-ITAM tyrosines, Y176 and Y204, are required for B L N K membrane recruitment and phosphorylation as well as for PLCy phosphorylation and calcium mobilization (Kabak et al, 2002). Moreover, Y204 is required for Iga association with B L N K (Engels et al, 2001; Kabak et al, 2002). However, it remains debatable whether or not this association is direct. Kabak and colleagues (2002) claim that this association is direct based on in vitro based studies. First, Kabak and colleagues (2002) demonstrated that a Sepharose41  coupled Iga phospho-Y204 peptide precipitates B L N K from lysates of stimulated B cells (Kabak et al, 2002). Second, a GST-BLNK-SH2 domain fusion protein precipitates the PDGFRp/Iga chimera containing the Iga cytoplasmic domain with an intact I T A M and Y204. Finally, the GST-BLNK-SH2 domain fusion protein directly binds to the PDGFRp/Iga chimera in a far western blot (Kabak et a l , 2002). However, while the first two studies do indicate that B L N K can associate with the cytoplasmic domain of Iga, they do not rule out the possibility that this association is mediated  by another cellular component  contained within the  lysates.  Furthermore, the PDGFRp/Iga chimera and truncated GST-BLNK-SH2 domain association may not necessarily represent the Iga-BLNK interaction as it would occur within the cellular context. Given this, it still remains to be determined if B L N K ' s association with Iga is direct or indirect.  The second aspect of the BCR/PLCy pathway that needs to be better defined involves the determination as to whether all of the components thus far identified are sufficient to reconstitute the pathway.  In particular, loss-of-function studies have clearly shown that Syk, B T K and  B L N K are all required to link the B C R to PLCy. Yet, the possibility remains that other, yet to be identified, lymphoid specific components are required to complete this pathway.  Thus, a  reconstitution approach has been employed in this thesis to determine i f these components are indeed sufficient to reconstitute the BCR/PLCy pathway and to further define the underlying mechanisms of this pathway (detailed below).  1.9  Thesis Goals  Initial Hypothesis: Co-expression of the B C R , Syk, B L N K and B T K will be sufficient to reconstitute B C R mediated activation of endogenous P L C y l in the non-lymphoid AtT20 reconstitution system.  Subsequent Objectives: I.  To investigate the mechanisms that regulate B C R lipid raft translocation and the role of such translocation in B C R signaling (Chapter 3).  II.  To determine i f co-expression of the B C R , Syk, B L N K and B T K is sufficient to reconstitute BCR-induced activation of endogenous P L C y l in the non-lymphoid  42  AtT20 reconstitution system as determined by monitoring PLC-dependent, B C R mediated Erk phosphorylation (Chapter 4).  III.  To determine why co-expression of the BCR, Syk, B L N K and B T K appears to have a limited ability to reconstitute BCR-induced activation of endogenous P L C y l in the AtT20 system. This includes the ancillary objectives:  To determine i f co-expression of these various components is sufficient to reconstitute BCR-induced tyrosine phosphorylation of B L N K , B T K and PLCy in the AtT20 system (Chapter 5).  ii.  To determine i f co-expression of these various components is sufficient to reconstitute BCR-induced co-association of B L N K , B T K and/or PLCy in the AtT20 system (Chapter 5 and Appendix 3).  iii.  To determine i f co-expression of these various components is sufficient to reconstitute BCR-induced membrane recruitment of B L N K , B T K and/or PLCy in the AtT20 system (Chapter 5).  To determine i f co-expression of the BCR, Syk and constitutively membranetargeted B L N K and/or PLCy is sufficient to reconstitute BCR-induced activation of PLCy in the non-lymphoid AfT20 reconstitution system as determined by monitoring PLC-dependent, BCR-mediated Erk phosphorylation (Chapter 6).  Experimental System: To date, the BCR/PLCy pathway has been mapped out primarily based on data from loss-offiinction studies.  Loss-of-function studies are invaluable for identifying key landmarks in a  pathway as they can determine whether or not a protein is necessary to a pathway's function. However, such studies often miss the more subtle, intervening topography of a pathway as they are often confounded by the existence of functionally redundant proteins and/or pathways that compensate for and ultimately mask the contributions of the genetically ablated protein being studied. As well, loss-of-function studies may fail to identify all of the components involved in a pathway.  Thus, any concerted mapping effort typically couples the techniques of loss-of43  function studies with those of reconstitution studies. Reconstitution studies are invaluable as they can determine whether or not a particular set of proteins are sufficient to reconstitute a given pathway." As well, such studies allow components to be investigated in isolation, alleviating the concerns of functional redundancy and enabling a more directed and detailed investigation of the pathway. Thus, this thesis employs the non-lymphoid AtT20 reconstitution system to further investigate the BCR/PLCy pathway.  The AtT20 reconstitution system is derived from the murine pituitary cell line, AtT20 (described in Moore et al, 1983). Being non-lymphoid, the AtT20 system does not endogenously express the B C R nor the lymphoid specific components required to couple the B C R to the PI3K, Ras/MAPK or PLCy pathways. Nonetheless, the AtT20 system does express endogenous PI3K, Ras, PLCy, and Fyn (Matsuuchi et al, 1992 and Richards et al, 1996). Moreover, the AtT20 system has proven amenable to transfection (Moore et al,  1983; Matsuuchi et al, 1992;  Richards et al, 1996). Thus, the AtT20 system lends itself well to B C R reconstitution studies. In particular, various lymphoid specific components can be transfected into this system in an effort to determine their sufficiency to reconstitute BCR-induced activation of the PI3K, Ras/MAPK and/or PLCy pathways.  Two AtT20-derived cell lines have previously been established (Matsuuchi et al, 1992; Richards et al, 1996). The 100.33 cell line (termed " B C R " herein) expresses intact exogenous BCRs on its cell surface (Matsuuchi et al, 1992). In contrast, the Sykl3 cell line (termed "BCR/Syk" herein) expresses the B C R along with exogenous Syk (Richards et al, 1996). Studies with these cell lines clearly indicate that expression of the B C R alone is sufficient to reconstitute BCRinduced Iga/p heterodimer phosphorylation and PI3K activation in this system (Matsuuchi et al, 1992). Furthermore, co-expression of the B C R and Syk is sufficient to reconstitute a robust BCR-induced tyrosine phosphorylation cascade as well as Ras/MAPK activity in this system (Richards et al, 1996). However, expression of the BCR, alone or in conjunction with Syk, is not sufficient to reconstitute BCR-induced PLCy activation. Thus, it appears that additional lymphoid specific components may be required to link the B C R to PLCy in this system (Richards et al, 1996).  As such, this thesis concentrates on attempting to reconstitute the  BCR/PLCy within this system in an effort to better understand how this pathway may function or fail to function to contribute to the immune response.  44  1.10 Thesis S u m m a r y  This  thesis  contains  two  sections,  each  of which  compartmentalization with respect to B C R signaling.  considers  a  distinct  aspect of  In the first section the mechanisms  involved in regulating the translocation of the B C R into lipid rafts were investigated. Herein it was demonstrated that the B C R can translocate into lipid rafts following B C R cross-linking in the immature B cell lines, WEHI 231 and CH31 (Jackson et al, 2005). Additionally, it was demonstrated that the Iga/p heterodimer, in the absence of the mlgM subunit, can translocate into lipid rafts following cross-linking in the m-IgM deficient immature B cell line, WEHI 303.1.5 (Jackson et al., 2005). This finding is significant as it suggests that Iga/p, in the absence of the mlgM subunit, contains sufficient structural information to facilitate its association with lipid rafts. Similarly, previous studies have demonstrated that the mlgM subunit, in the absence of Iga/p, contains sufficient structural information to facilitate its association with lipid rafts (Cheng et al, 2001).  Considered together, these findings may be used to help define any  structural regions involved in mediating lipid raft association.  In particular, it may be  hypothesized that these subunits use a similar mechanism to associate with lipid rafts and then it may be further hypothesized that the subunits share common structural feature to mediate this association. Given the structures of mlgM and the Iga/p, such a common feature would most likely lie within their transmembrane or membrane proximal domains. And thus, the Matsuuchi Lab is performing mutational analyses within these regions to further define any structural features that may help govern BCR lipid raft association and translocation.  In the second section, the molecular requirements for the BCR/PLCy pathway were investigated. In particular, the AtT20 reconstitution system was used to determine i f co-expression of the BCR, Syk, B L N K and B T K is sufficient to reconstitute the BCR/PLCy pathway. Herein, it was determined that co-expression of the BCR, Syk and B L N K is sufficient to reconstitute BCRinduced PLCy activation in the AtT20 system.  However, this activation is hypothesized to  represent only a partial reconstitution of the pathway as neither B L N K nor PLCy are recruited to the plasma membrane upon B C R cross-linking and as PLCy phosphorylation appears very limited compared to what is typically observed in B cell lines. This limited reconstitution was hypothesized to be due to the absence of B T K ; however, further expression of B T K within the system did not enhance PLCy activation, but rather inhibited it.  Further investigations  determined that B T K is constitutively activated within this system and as such, may be 45  inappropriately influencing the components of the BCR/PLCy pathway. Thus, further studies are in order to address this issue.  Alternatively, it was hypothesized that the limited reconstitution of BCR-induced PLCy activation may be a result of the apparent inability to reconstitute BCR-induced B L N K and PLCy membrane recruitment.  Thus, B L N K and PLCy were constitutively targeted to the plasma  membrane within the AtT20 system. From this, it was determined that membrane-targeting of PLCy is sufficient to reconstitute BCR-induced Syk-dependent PLCy activation, despite the absence  of B L N K .  This  finding  compartmentalization to its activation.  highlights the  importance  of PLCy  membrane  In contrast, membrane-targeting of B L N K is not  sufficient to reconstitute BCR-induced PLCy membrane recruitment or to enhance BCR-induced PLCy activation within this system. This suggests that there may be an additional defect in the system that is preventing the formation of a functional B C R / B L N K / P L C y signaling complex. In particular, it is hypothesized that Iga and B L N K phosphorylation may not be fully reconstituted within this system. Furthermore, it is hypothesized that the SFKs may be required to reconstitute such phosphorylation. Thus, future studies will focus on addressing these possibilities.  Through the course of this thesis it was also determined that PLCy phosphorylation does not predicate PLCy activation. This finding is taken to further emphasize the requirement for appropriate membrane compartmentalization in the process of PLCy activation. Additionally, B L N K that is targeted to the plasma membrane in general (TmBLNK) was found to have a different effect on BCR-induced PLCy phosphorylation than B L N K that is targeted to lipid rafts within the plasma membrane.  This suggests that all membrane compartmentalization is not  equivalent and alludes to the precision required to regulate appropriate compartmentalization. And thus, the work in this thesis demonstrates the importance of compartmentalization in B C R signaling both with respect to compartmentalization of the B C R to lipid rafts and the subsequent compartmentalization of B L N K and PLCy to the BCR.  46  This page is meant to be blank.  CHAPTER 2 Materials and Methods 2.1 2.1.1  Reagents Antibodies  The rabbit polyclonal antibodies; anti-BLNK (H-80), anti-PLCyl (1249), anti-PLCy2 (Q-20) and anti-ERK 1 (K-23); the mouse monoclonal antibodies, anti-CD 16 (GRM1) and anti-BLNK (2B11); and the goat polyclonal antibodies, anti-BLNK (C-19) and anti-BTK (C-20) were all purchased from Santa Cruz Biotechnology (Santa Cruz, California). antiserum (described in Richards et al,  The anti-Syk rabbit  1996) and the anti-mouse Iga rabbit  antiserum  (described in Gold et al, 1991) were produced by the Matsuuchi Laboratory (University of British Columbia; Vancouver, British Columbia). The anti-Akt, anti-phospho-AKT (S473), antiphospho-BLNK (Y96), anti-phospho-BTK (Y223), and anti-phospho-Erkl/2 (T202/Y204) were purchased from Cell Signaling Technology (Beverly, Massachusetts). The anti-phospho-tyrosine mouse monoclonal antibody (4G10) was purchased from Upstate Biotechnology (Charlottesville, Virginia). The anti-Ig|3 cytoplasmic tail rabbit antiserum was kindly provided by Dr. Marcus Clark (University of Chicago; Chicago, Illinois).  The anti-K light chain rabbit polyclonal  antibody and the anti-actin mouse monoclonal antibody were purchased from ICN Biomedicals (Irvine, California). The anti-mouse u heavy chain rabbit polyclonal antibody was purchased from  Jackson  ImmunoResearch  Laboratories  Incorporated  (West  Grove, Pennsylvania;  distributed by Bio/Can Scientific, Mississauga, Ontario). The stimulating antibodies (antibodies used to cross-link the B C R on B cells and/or AtT20derived cell lines) included the anti-mouse IgM (u chain specific) and anti-human IgM (u chain specific) goat polyclonal antibodies purchased from Jackson ImmunoResearch Incorporated.  Laboratories  The HM79-16 monoclonal antibody that recognizes the extracellular domain of  murine IgP was purified from HM79-16 hamster hybridoma cells that were a kind gift from Dr. T.  Nakamura (University of Tokyo; Tokyo, Japan).  The antibody was purified from the  hybridoma culture media using a Protein-G-agarose affinity column and biotinylated using N hydroxysuccinimido biotin (NHS-biotin; Pierce Biotechnology; Rockford, Illinois).  Further  cross-linking of the biotinylated HM79-16 monoclonalantibody was achieved using streptavidin (Molecular Probes; Eugene, Oregon). The horseradish peroxidase (HRP)-conjugated secondary reagents, Protein A-HRP and Protein G-HRP, were purchased from Amersham Biosciences Canada (Baie d'Urfe, Quebec). The goatanti-rabbit IgG-HRP was purchased from Jackson ImmunoResearch Laboratories Incorporated, the goat anti-mouse IgG-HRP was purchased from Invitrogen Life Technologies (Burlington, Ontario) and the donkey anti-goat IgG-HRP, was purchased from Santa Cruz Biotechnology. 48  2.1.2 Plasmids The pWZL Blast 1 and pWZL Blast 2 vectors that encode for blasticidin resistance were kindly provided by Dr. Stephen Robbins (University of Calgary; Calgary, Alberta). The LPCsrf vector that encodes for puromycin resistance was kindly provided by Dr.  Anthony DeFranco  (University of California, San Francisco; San Francisco, California). The pMSCV-puro vector and the pSVneo vector were purchased from B D Biosciences Clonetech (Palo Alto, California). ThepBSmyc-human-BLNK (pp70) vector (described in Fu et al, 1998) that encodes for the myctagged human B L N K (pp70) fusion protein and the pApuro-mPLCy2 vector that encodes for the membrane-targeted human FcyRIIlA/CD 16 extracellular domain/human  T cell receptor C,  transmembrane domain/rat PLCy2 fusion protein (described in Ishiai et al, 1999) were kindly provided by Dr. Andrew Chan (while at Washington University School of Medicine; St. Louis, Missouri; currently at Genetech, California). The pBS-human-PLCy2 vector that encodes for human PLCy2 and the LPCX-BTK-2  vector that encodes for puromycin resistance and murine  B T K were kindly provided by Dr.  David Rawlings (University of Washington; Seattle,  Washington). The RSVpLpA vector (kindly provided by Dr. Marika Walter and Dr. David Standring of the University of California, San Francisco; San Francisco, California) was used as a parental plasmid to create several of the vectors described below.  The RSVpLpA vector contains a  multiple cloning site flanked by the Rous Sarcoma Virus (RSV) long terminal repeat (LTR) promoter and enhancer elements on the 5' side and the SV40 polyadenylation site and splice site on the 3' side. This particular combination of regulatory elements has proven successful in producing high levels of exogenous protein expression within transfected AtT20-derived cell lines (Matsuuchi and Kelly, 1991; Matsuuchi et al, 1992). 2.1.3  Plasmids Created for This Thesis  RSVpLpA-myc-human-BLNK (pp 70) The sequence encoding the myc-tagged human B L N K was excised from the pBSmyc-human BLNK (pp70) vector with Xba I and ligated into the Xba I digested RSVpLpA vector (Fig.2.1). Correct orientation of the myc-tagged human B L N K sequence within the RSVpLpA vector was confirmed by restriction enzyme digests and nucleotide sequencing. Nucleotide sequencing was conducted by the Nucleic Acid Protein Service Unit (NAPS Unit) at the University of British Columbia (UBC). RSVpLpA-human-PLCy2 The sequence encoding human PLCy2 was amplified from the pBS-human-PLCy2 vector using the polymerase chain reaction (PCR). Two primers, TJ1 and TJ2 (Table 2.1), were created to amplify the human PLCy2 sequence from the pBS-human-PLCy2 vector such that the resultant 49  PCR product encoded for human PLCy2 flanked by Xba I sites. The P C R product was then digested with Xba I and ligated into Xba I digested RSVpLpA vector (Fig.2.2). Correct orientation of the human PLCy2 sequence within the resultant RSVpLpA-human-PLCy2 vector was confirmed by restriction enzyme digests and by nucleotide sequencing by the NAPS Unit at UBC. Table 2.1. Summary of Primers Used To Create the Various Plasmids.  Description  Primer Sequence  Primer Name  sense sequence, bold = X b a I site, italics  TJ1 ' ATC AAC TAG TT*C TAG AGC CAA ACC CGG GGC '  5  3  are complimentary to the 5' end of human PLCy2  a s found  in the  pBS-human-  PLCy2 vector anti-sense sequence, bold = X b a I site,  TJ2  italics are complimentary to the 3' end of  ' GG GGC GGC CGC T*CT AGA GTT TTT TTT TTT  5  3  human  PLCy2  human-PLCy2  a s found in the pBSvector  sense sequence, bold = Mlu i site, /Ya//cs  TJ3 5  ' C G A * C G C G T T CCA CCA CGG TCA ATG TAG ATT ' 3  = 5' end of human P L C y 2 beginning with the second codon (i.e., A T G removed) anti-sense sequence, bold = Mlu I site,  TJ4 5  ' C G A * C G C G T C TTC TCT CTTAAC  bold = stop codon, italics = 3' end of  CTC TTG TTG ' 3  human P L C y 2 sense sequence, bold = X b a I site, italics  TJTM5 5  ' T G C T * C T A G A G C A ATG TGG CAG CTG CTC CTC CCA  = 5' e n d of T M construct (human F c R illa/CD16)  ACT ' 3  anti-sense sequence, bold = X b a I site,  TJTM3 5  ' T G C T * C T A G A T C A A * C G C G T GCT CCT GCT GAA CTT  bold = Mlu I site, rta//'cs = 3' end of T M construct (human T C R Q  CAC ' 2  sense sequence, bold = Mlu I site, italics  TJ7 5  = 5' end of human B L N K beginning with  ' C G A * C G C G T G ACA AGC TTA ATA AAA TAA CCG ' 3  the s e c o n d codon (i.e., myc tag and A T G removed) anti-sense sequence, bold = Mlu I site,  TJ8 5  bold = stop codon, italics = 3' end of  ' C G A * C G C G T C C C C CTT TAT GAA ACT TTA ACT ' 3  human B L N K sense sequence, bold = X b a I site, italics  5' A c 5  = 5' e n d murine  ' T G C T * C T A G A G C A ATG GGA TGT ATT AAA TCA ' 3  Lyn encoding  amino  acids 1- 6) anti-sense sequence, bold = X b a I site,  3' A c 5  ' T G C T * C T A G A T C A A * C G C G T GAG ATT GTC TTT ' 3  bold = Mlu I site, italics = 5' end of murine Lyn, encoding amino acids 9-12  RSVpLpA-TM For subsequent cloning procedures an RSVpLpA-TM vector was created.  This vector allows  sequences to be cloned in-frame to a transmembrane-targeting (TM) construct.  Specifically,  RSVpLpA-TM encodes for the extracellular domain of human FcRIIIa/CD16 (amino acids 1-208) 50  fused in-frame to transmembrane domain of the human T cell receptor (TCR) C, chain (amino acids 31-58). In RSVpLpA-TM the T M construct can be further fused to any sequence of interest via an in-frame Mlu I site. The resulting sequence should then encode for a TM-fusion protein that should be constitutively targeted to the plasma membrane. The RSVpLpA-TM vector was created by first digesting the RSV-pLpA vector with Mlu I, bluntending the digested product and then re-ligating the vector to create RSV-pLpA-Mlu I(-) (Fig.2.3).  This procedure effectively eliminated an extraneous Mlu I site that would have  interfered with the cloning strategy.  The sequence encoding the T M construct was then  amplified from the pApuro-mPLCy2 vector (Chapter 2.1.2) using PCR.  1  Two primers, TJTM5  and TJTM3 (Table 2.1), were created to amplify the T M construct from the pApuro-mPLCy2 vector such that the resultant PCR product encoded for the T M construct with flanking Xba I sites and an in-frame 3' Mlu I cloning site. The P C R product was then digested with Xba I and ligated into the Xba I digested RSV-pLpA-Mlu I(-) vector (Fig.2.3). Correct orientation of the T M construct within the resultant RSV-pLpA-TM vector was confirmed by restriction enzyme digests and by nucleotide sequencing by the NAPS Unit at UBC. RS VpLpA - TM-h um an-BLNK The sequence encoding human B L N K was amplified from the pBSmyc-human-BLNK (pp70) vector using PCR.  Two primers, TJ7 and TJ8 (Table 2.1), were created to amplify the human  B L N K sequence from the pBSmyc-human-BLNK (pp70) vector such that the resultant PCR product encoded for human B L N K (myc-tag and first codon removed) flanked by Mlu I sites (with the 5' site being in-frame). The PCR product was then digested with Mlu I and ligated into the Mlu I digested RSV-pLpA-TM vector (Fig.2.4). Correct orientation of human B L N K within the resultant RSV-pLpA-TM-human-BLNK vector was confirmed by restriction enzyme digests and by nucleotide sequencing by the NAPS Unit at U B C . RSVpLpA-Ac For subsequent cloning procedures an RSVpLpA-Ac vector was created.  This vector allows  sequences to be cloned, in-frame to the acylation-targeting (Ac) sequence of murine Lyn.  The  sequence encoding the Ac sequence (amino acids 1-12) was amplified from the RSVpLpAmurine-Lyn vector (described in Santos, 2003) using PCR. Two primers, 5'Ac and 3'Ac (Table 2.1), were created to amplify the Ac sequence from the RSVpLpA-Lyn vector such that the resultant PCR product encodes for the Ac sequence with flanking Xba I sites and an in-frame 3' Mlu I cloning site. The P C R product was then digested with Xba I and ligated into the Xba I digested RSVpLpA-Mlu I(-) vector (Fig.2.5). Correct orientation of the A c sequence within the ' It should be noted that the Ishiai et al. (1999) reports the T M construct encodes for human FcRIIIa/CD16 amino acids 1-212 fused in-frame to human T C R zeta amino acids 30-58. However, our sequencing results indicate that the T M construct actually encodes for human FcRIIIa/CD16 amino acids 1-208 fused in-frame to human T C R zeta amino acids 31-58.  51  resultant RSVpLpA-Ac vector was confirmed by restriction enzyme digests and by nucleotide sequencing by the NAPS Unit at U B C . RSVpLpA-Ac-human-BLNK The sequence encoding human B L N K was amplified from the pBSmyc-human-BLNK (pp70) vector using PCR.  Two primers, TJ7 and TJ8 (Table 2.1), were created to amplify the human  B L N K sequence from the pBSmyc-human-BLNK (pp70) vector such that the resultant PCR product encoded for human B L N K (myc-tag and first codon removed) flanked by Mlu I sites (with the 5' site being in-frame). The PCR product was then digested with Mlu / and ligated into the Mlu I digested RSVpLpA-Ac vector (Fig.2.6). Correct orientation of human B L N K within the resultant RSVpLpA-Ac-human-BLNK vector was confirmed by restriction enzyme digests and by nucleotide sequencing by the NAPS Unit at U B C . RSVpLpA-Ac-human-PLCy2 The sequence encoding human PLCy2 was amplified from the RSVpLpA-human-PLCy2 vector (described above) using PCR.  Two primers, TJ3 and TJ4 (Table 2.1), were created to amplify  the human PLCy2 sequence from the RSVpLpA-human-PLCy2 vector such that the resultant PCR product encoded for human PLCy 2 flanked by Mlu I sites (with the 5' site being in-frame). The PCR product was then digested with Mlu I and ligated into the Mlu I digested RSVpLpA-Ac vector (Fig.2.6). Correct orientation of human PLCy2 within the resultant RSVpLpA-Ac-humanPLCyl vector was confirmed by restriction enzyme digests and by nucleotide sequencing by the NAPS Unit at U B C .  52  Xba /  Xba I  Xba I  Figure 2.1. Schematic representation of how the RSVpLpA-myc-human-BLNK (pp70) expression vector was developed. The pBS-myc-human-BLNK vector (~ 4.6 kB) was digested with Xba I to drop out the sequence encoding myc-tagged human B L N K (~ 1.6 kB). This sequence was then purified and ligated into the mammalian expression vector, RSVpLpA (~ 4.5 kB), that had similarly been digested with Xba I. Correct orientation of the myctagged human B L N K insert within the resultant RSVpLpA-myc-human-BLNK vector (~ 6.1 kB) was then confirmed by restriction enzyme digests and nucleotide sequencing.  53  Xba I  Xba I  U s e P C R to create a P L C y 2 product that has flanking X b a I sites  Xba I  PLCy2 human-PLCy2 PCR product (- 4.0 kB)  Digest with Xba I  Digest with Xba I  Xba I sticky end  Xba I sticky end  Xba I sticky end  Xba I sticky end  PLCy2 human-PLCy2 (-4.0 kB)  Ligate Xba /  Xba /  Figure 2.2. Schematic representation of how the RSVpLpA-human-PLCy2 expression vector (~ 8.5 kB) was developed. The sequence for human PLCy2 (~ 4.0 kB) was amplified from the pBS-human-PLCy2 vector (~ 6.9 kB) using PCR. The resultant P C R product (~ 4.0 kB) was digested with Xba /, purified and ligated into the mammalian expression vector, RSVpLpA (~ 4.5 kB) that had similarly been digested with Xba I. Correct orientation of the human PLCy2 insert within the resultant RSVpLpA-human-PLCy2 vector was then confirmed by restriction enzyme digests and nucleotide sequencing.  54  Mlu I  sticky ends  Digest with Mlu I  |  RSVpLpA (~4.5kB)  Fill in Mlu I sticky ends to create blunt ends  I  Blunted Mlu I ends  !  RSVpLpA (~4.5kB)  Use PCR to create an EC/TM product that has a downstream, in-frame Mlu I cloning site and flanking Xba I sites  Blunt-end ligation  In-frame Mlu i  I  Xba I  Re-ligated blunt ends, extraneous Mlu I site destroyed  Xba I  I  ^  Transmembrane (TM) PCR product (~ 0.7 kB)  Digest with Xba I  Digest with Xba I  Xba I sticky end  In-frame Mlu I  —I  Xba I sticky end  E C TM  Xba / sticky end  Xba / sticky end  Figure 2.3. Schematic representation of how the RSVpLpA-TM  expression vector (~ 5.2 kB) was developed.  Refer to Chapter 2.3.1 for a detailed explanation of how the vector was created. 55  In-frame Mlu I  In-frame Mlu I  = |  Mlu I  BLNK  | =  human-BLNK PCR product (~ 1.6 kB)  Digest with Mlu I  Digest with Mlu I  Figure 2.4. Schematic representation of how the RSVpLpA-TM-human-BLNK expression vector (~ 6.8 kB) was created. The sequence for human B L N K was amplified from the pBS-myc-human-BLNK vector using PCR. The primers, TJ7 and TJ8, were used such that the resultant PCR product (~ 1.6 kB) would encode for human B L N K (with the myc-tag and first codon removed) flanked by Mlu I sites (with the 5' site being in-frame with the B L N K sequence). The PCR product was then digested with Mlu /, purified and ligated into the RSVpLpA-TM vector (~ 5.2 kB) that had similarly been digested with Mlu I. Correct orientation of the human B L N K insert within the resulting RSVpLpA-TM-human-BLNK vector was then confirmed by restriction enzyme digests and nucleotide sequencing.  56  Digest with Mlu I  Fill in Mlu I sticky ends to create blunt ends  acylation sites  H Blunted Mlu I ends  L<^  RSVpLpA (~4.5kB)  U s e P C R to create a product that encodes for the Lyn acylation sites (Ac) with a downstream, in-frame Mlu I cloning site and flanking Xba I sites.  I  Blunt-end ligation  t In-frame Mlu I  Re-ligated blunt ends, extraneous Mlu I site destroyed  ^  =[~A7> Xba /  Xba /  Acylation (Ac) PCR product (~ 0.06 kB) Digest with Xba I ^  Digest with Xba I Xba I sticky end  Xba / sticky end  In-frame Mlu I  t  t  Xba / sticky ends  Figure 2.5. Schematic representation of how the RSVpLpA-Ac  expression vector (~ 4.56 kB) was developed.  Refer to Chapter 2.3.1 for a detailed explanation of how the vector was created  57  In-frame Mlu I  Xba I  I  Xba I  In-frame Mlu I  BLNK human-BLNK PCR product (- 1.6 kB)  Figure 2.6. Schematic representation of how the RSVpLpA-Ac-human-BLNK  expression vector (~ 6.16 kB)  was developed. The sequence for human B L N K was amplified from the pBS-myc-human-BLNK vector using PCR. The primers, TJ7 and TJ8, were used such that the resultant PCR product (~ 1.6 kB) would encode for human B L N K (with the myc-tag and first codon removed) flanked by Mlu I sites (with the 5' site being in-frame with the B L N K sequence). The PCR product was then digested with Mlu /, purified and ligated into the RSVpLpA,-Ac vector (~ 4.56 kB) that had similarly been digested with Mlu I. Correct orientation of the human B L N K insert within the resultant RSVpLpA-Ac-human-BLNK vector was then confirmed by restriction enzyme digests and nucleotide sequencing.  58  In-frame Mlu I  U s e P C R to create a PLCY2 product that has flanking Mlu I sites (5' site in-frame)  In-frame Mlu I  =  PLC 2 Y  Human-PLCy2 PCR  =  product  (-4.0 kB)  Digest with Mlu I  Digest with Mlu I  In-frame Mlu I sticky end  Mlu I sticky end  PLCy2  =_  Figure 2.7. Schematic representation of how the RSVpLpA-Ac-lunnun-PLCy2 expression vector (~ 8.56 kB) was developed. The sequence for human PLCy2 was amplified from the RSVpLpA-human-PLCy2 vector (~ 8.5 kB) using PCR. The primers, TJ3 and TJ4, were used such that the resultant PCR product (~ 4.0 kB) would encode for human PLCy2 (with the first codon removed) flanked by Mlu I sites (with the 5' site being in-frame with the PLCy2 sequence). The PCR product was then digested with Mlu I, purified and ligated into the RSVpLpA,-Ac vector (~ 4.56 kB) that had similarly been digested with Mlu I. Correct orientation of the human PLCy2 insert within the resultant RSVpLpA-Ac-human-PLCy2 vector was then confirmed by restriction enzyme digests and nucleotide sequencing  59  2.2  Molecular biology methods  2.2.1  Restriction endonuclease digests  Restriction enzymes were purchased from Roche Diagnostics (Laval, Quebec), Invitrogen Life Technologies, New England Biolabs (Pickering, Ontario) and Promega (Madison, Wisconsin) and were used according to the manufacturers' instructions.  In general, the total amount of  restriction enzyme(s) added to any given reaction was equal to or less than 10 % of the total volume of the digest. Digests were performed for 2-10 hours at 37 °C. Where necessary, the digested products were isolated by agarose gel purification. 2.2.2  Alkaline phosphatase reactions  Linearized vectors where treated with alkaline phosphatase (Roche Diagnostics) to prevent selfligation. Where required, alkaline phosphatase was added to the restriction endonuclease digest for the final hour of the digest (using 20 units of alkaline phosphatase/1 ug of digested vector). Reactions were terminated by adding E D T A to a final concentration of 1 m M and heating samples in a heating block at 65 °C for 5 minutes. The digested D N A was then purified either by agarose gel purification or by phenol/chloroform extraction. 2.2.3 Phenol/Chloroform extraction Phenol/chloroform extraction was performed essentially as described in the second edition of Molecular Cloning: A Laboratory Manual (Sambrook et al, 1989). A n equal volume of phenol was added to the D N A sample. The DNA/phenol solution was vortexed for 30 seconds and then centrifuged at room temperature at 14 000 rpm for 30 seconds. The aqueous phase containing the D N A (top layer) was then carefully removed, transferred to a fresh tube and again mixed with an equal volume of phenol, vortexed and centrifuged as before. The aqueous phase was then carefully removed and transferred to a fresh tube where it was mixed with an equal volume of Sevag (24:1 chloroform:isoamyl alcohol), vortexed for 30 seconds and then centrifuged as before.  The aqueous phase containing the D N A was then carefully removed, transferred to a  fresh tube and again mixed with an equal volume of Sevag, vortexed and centrifuged as before. The aqueous phase containing the D N A was then carefully removed, and transferred to a fresh tube. The phenol fractions and Sevag fractions were then sequentially back extracted with an equal volume of distilled water to recover any remaining D N A . The distilled water/DNA extract was then pooled with the aqueous phase containing the D N A .  The D N A was then ethanol-  precipitated. Sodium acetate was added to the D N A sample to a final concentration of 0.3 M to facilitate precipitation of the D N A . Two and one-half volumes of ice-cold 95 % ethanol were then added to the solution. Following a 1 hour incubation at -20 °C the solution was then centrifuged at 4 °C at 14 000 rpm for 10 minutes. The supernatant was carefully removed and the remaining D N A pellet was washed with 750 u.1 of ice-cold 95 % ethanol. The ethanol was then carefully 60  removed and the D N A pellet was left to air dry for 10 minutes before being re-suspended in an appropriate volume of the appropriate buffer (generally 25-200 ul of distilled water). Purified D N A samples were stored at -20 °C. 2.2.4  Agarose gel electrophoresis  D N A samples were resolved on agarose gels.  Agarose gels contained 0.5-2.0g agarose  (GibcoBRL; Grand Island, New York) and 25 ug/ml ethidium bromide (GibcoBRL) in 100 ml of Tris/Boric Acid/EDTA (TBE; Chapter 2.14.1). D N A samples were mixed with 6x D N A sample buffer (5 parts sample to 1 part 6x sample buffer; Chapter 2.14.2) prior to loading on gels. Gels were then electrophoresed at a constant voltage of 100 volts in T B E for varying lengths of time dependant of the size and agarose percentage of the gel. Resolved D N A was then examined using ultraviolet light. 2.2.5  G e l purification of D N A  D N A was purified from agarose gels using either the Qiaquick Gel Extraction Kit (Qiagen; Mississauga, Ontario) or the Elutip-D Gel Extraction Kit (Schleicher and Schuell Bioscience GmbH; Keene, New Hampshire) as per the manufacturers' recommended instructions. 2.2.6  D N A ligation reactions  D N A ligations were performed with the Rapid D N A Ligation kit (Roche Diagnostics) as per the manufacturer's instructions. Generally, 0.5 ug of digested, purified vector D N A was mixed with 1-10 ug of digested, purified insert D N A (a molecular ratio of 2:1 of insertplasmid D N A was used) along with 2 ul of the 5x D N A dilution buffer (kit component) and distilled water to a total volume of 10 ul. 10 ul of 2x Ligation Buffer and 1 ul of T4 D N A Ligase (kit components) were then added to the D N A solution. The reaction was then incubated at room temperature for 1 hour prior to being used to transform competent bacteria. 2.2.7  Transformation of competent  Escherichia coli bacteria  Competent Escherichia coli (E. coli) bacteria were prepared by May Dang-Lawson (Matsuuchi Laboratory Technician; University of British Columbia; Vancouver, British Columbia). Competent bacteria were removed from -80 °C and allowed to thaw on ice. One hundred microlitres of thawed competent bacteria were then mixed with 100-300 rig of plasmid D N A and returned to ice for 20 minutes. The bacteria/DNA mixture was then incubated in a 42 °C water bath for 2 minutes before being returned to ice for 5 minutes. Finally, the bacteria/DNA mixture was plated on Luria-Bertani (LB) agar plates (Chapter 2.14.4) containing 100 ug/ml ampicillin (Sigma-Aldrich Canada; Oakville, Ontario). Plated bacteria were then incubated upside down at 37 °C overnight or until bacteria colonies were visible to the naked-eye.  61  2.2.8  Polymerase chain reactions (PCR)  Oligonucleotide primers for PCR were purchased through the NAPS Unit at U B C . Primers were re-suspended at a concentration of 10 pmo 1/ul in distilled, autoclaved water. Template D N A was similarly diluted to a concentration of 0.002 ug/ul. 2.5 ul of each primer (forward and reverse), 2.5 ul of template D N A and 17.5 ul of distilled autoclaved water were then added per tube of puReTaq Ready-To-Go-Beads (Amersham Biosciences Canada).  The P C R mix was then  overlaid with 25 ul of autoclaved mineral oil. Tubes were then placed in a D N A Thermocycler 480 (Perkin Elmer CETUS) and subjected to 30 cycles of 95 °C for 45 seconds, 55 °C for 120 seconds and 72 °C for 120 seconds.  Finally, P C R products were collected from below the  mineral oil layer and gel purified (described in Chapter 2.2.4 and 2.2.5). 2.2.9  Qiagen-mediated preparation of DNA (Small Quantity preparations)  Individual bacteria colonies were used to inoculate 4 ml of L B broth (Chapter 2.14.3) containing 100 ug/ml ampicillin. Inoculated L B broth was then incubated overnight at 37 °C with gentle shaking. Plasmid D N A was then isolated from transformed bacterial cultures using the Qiaprep Spin Miniprep Kit (Qiagen) as per the manufacturer's instructions. 2.2.10 Cesium chloride-mediated preparation of DNA (Large Quantity Preparations) Individual bacteria colonies were used to inoculate 40 ml of L B broth (Chapter 2.14.3) containing 100 ug/ml ampicillin (adapted from Clewell and Helinski, 1972). Inoculated L B broth was then incubated overnight at 37 °C with gentle shaking.  This culture was then  transferred to 1 L of M9 growth media (Chapter 2.14.5) and incubated overnight at 37 °C with gentle shaking. Bacteria were pelleted by centrifiigation at room temperature at 7500 rpm for 10 minutes in a JLA-16.250 rotor and Avanti centrifuge J-25I (Beckman Coulter; Fullerton, California). The supernatant was then removed and the bacterial pellet frozen at -20 °C for at least 1 hour. The pellet was subsequently thawed on ice and re-suspended in 15 ml of sucrose solution (Chapter 2.14.6). The bacteria were lysed by the addition of 3 ml of bacteria lysis buffer I (Chapter 2.14.7) for 5 minutes on ice followed by the addition of 1.5 ml of 250 m M E D T A (pH 8.0) for 15 minute on ice followed by the addition of 15 ml triton lytic mix (Chapter 2.14.8) for 30 minutes on ice. The resulting lysate was then cleared by centrifugation at room temperature at 18 000 rpm for 90 minutes in a JA 25.50 rotor and Avanti centrifuge J-25I. The resulting supernatant was collected and adjusted to a final concentration of 1.6 g/ml of cesium chloride. The supernatant was then transferred to Beckman Quick-Seal centrifuge tubes (16 x 76 mm) (Beckman Coulter) containing 1 mg of ethidiurri bromide/tube. Once full, the tubes were sealed and centrifuged at room temperature at 48 000 rpm for 48 hours in a 70.1 T i rotor and L8-70M Ultracentrifuge (Beckman Coulter). Following centrifugation, plasmid D N A (indicated as the bottom of two ethidium bromide containing bands) was carefully removed from the gradient, extracted with cesium chloride-saturated isopropanol and dialysed extensively against 6 L of Tris/EDTA (TE; Chapter 2.14.11) over 2-3 days at 4 °C. Plasmid D N A was then precipitated 62  with ethanol (described in Chapter 2.2.14), centrifuged and re-suspended in 1 ml of distilled, autoclaved water. The optical density was measured at 260 nm and 280 nm to determine the OD260nm/OD 80nm2  This was then used to determine the concentration of D N A based on the  equation: Concentration of double-stranded D N A = 50 fig/ml (OD 60nm/OD280nm) 2  2.3 2.3.1  Tissue culture Tissue culture cell lines  The AtT20 murine endocrine cell line (described in Matsuuchi and Kelly, 1991) was kindly provided by Dr.  Regis Kelly (University of California, San Francisco; San Francisco,  California). The CH31 murine B lymphoma, K40B-1 pro-B cell and K40B-2 pre-B cell lines were kindly provided by Dr. Anthony DeFranco (University of California, San Francisco; San Francisco, California). The WEHI 231 murine B lymphoma, Ramos human B lymphoma and Daudi human B lymphoma were purchased from the American Type Tissue Culture Collection (Manassas, Virginia). The WEHI 231 murine B lymphoma variant, WEHI 303.1.5 (described as WEHI 303.1.5LM in Condon et al, 2000) was characterized by the Matsuuchi Laboratory. 2.3.2  Maintenance of tissue culture of cell lines  AtT20-derived murine endocrine cell lines were maintained in complete Dulbecco's Modified Eagle Medium (complete D M E M ; Invitrogen Life Technologies; Chapter 2.14.12). Cells were plated on 10 cm polystyrene tissue culture plates (Falcon; Franklin Lakes, New Jersey) and maintained at 37 °C in a 10 % C O 2 atmosphere water-jacketed incubator. Cells were generally grown to 90 % confluence before being passaged.  To passage cells, the media was removed  from the plate, cells were washed with 8 ml of Dulbecco's Phosphate-Buffered Saline (D-PBS) (Invitrogen Life Technologies) and then incubated at room temperature with 1 ml of trypsin solution (Invitrogen Life Technologies; Chapter 2.14.16) for 1-2 minutes. Cells were then resuspended in 8 ml of fresh complete D M E M and transferred to a fresh culture dish. Cells were passaged at various dilutions ranging from 1:20-1:2 dependent on experimental requirements. Human and murine B lymphoma cell lines were maintained in complete Roswell Park Memorial Institute (RPMI)-1640 media (Invitrogen Life Technologies; Chapter 2.14.14).  Cells were  grown either in 10 cm polystyrene tissue culture plates or in tissue culture flasks of various sizes (Falcon) and maintained at 37 °C in a 5% C O 2 atmosphere water-jacketed incubator. Cells were passaged every 3-4 days at various dilutions ranging from 1:20-1:2 dependent on experimental requirements. Tissue culture cell lines were generally not passaged for longer than three months as extended passages may encourage the accumulation of mutations within the cell populations.  Thus,  multiple copies of the various cell lines were stored in 2 ml cryovials (Simport Plastics; Beloell, 63  Quebec) in liquid nitrogen. AtT20-derived cell lines were frozen down in complete D M E M further supplemented with 10 % dimethyl sulfoxide (DMSO; Sigma-Aldrich Canada). Human and murine B lymphoma cell lines were frozen down in fetal calf serum supplemented with 10 % DMSO. 2.3.3  Calcium phosphate-mediated transfections of AtT20-derived cell lines  AtT20-derived cell lines were transfected essentially as described by Matsuuchi and Kelly (1991). The desired AtT20-derived cell line was grown to approximately 70 % confluency on 10 cm polystyrene tissue culture plates. D N A was prepared for transfection by first mixing 50 ug of the desired plasmid(s), 20 ug of a selectable drug resistance plasmid, 94 ul of 2 M CaCl and 2  distilled water (such that the final volume of reagents totaled 750 ul) together. The entire 750ul solution was then added drop-wise to 750 ul of 2x HEPES-buffered saline (Chapter 2.14.17) while vortexing the solution at maximum speed. The DNA/HEPES solution was then incubated at room temperature for 40 minutes to allow a calcium phosphate/DNA precipitate to form. Cells to be transfected were washed twice with l x HEPES-buffered saline (Chapter 2.14.18). The precipitated DNA/HEPES solution (should appear slightly cloudy) was then added dropwise to the washed cells (1.5 ml/plate). Cells were then returned to the 37 °C, 10 % C 0  2  incubator for 20 minutes. Following 20 minutes, complete D M E M media was added to the cells (8 ml/plate). Cells were then returned to the 37 °C, 10 % C 0 incubator. Six to eight hours later 2  the DNA/HEPES/media solution was aspirated from the cells. The cells were then glycerolshocked for 1 minute with a room temperature 25 % glycerol/75 % complete D M E M media solution (lml/plate). After 1 minute the solution was diluted with room temperature D-PBS (8 ml/plate) and removed by aspiration. Cells were then washed twice with room temperature DPBS (8ml/plate). Complete D M E M media was then added to the cells (8 ml/plate) and the cells were returned to the 37 °C, 10 % C 0 incubator for 2-3 days. 2  2.3.4  Drug selection of transfected cells and isolation of individual clones  Two to three days following transfection the cells were split 1:3 into selection media (i.e., one 10 cm polystyrene tissue culture plate of cells was passaged to 3 fresh 10 cm polystyrene tissue culture plates (described in Section 2.14.2). Selection media comprised of complete D M E M supplemented with either 2 ug/ml blasticidin (Invitrogen Life Technologies), 0.4 mg/ml G418 neomycin (Invitrogen Life Technologies) or 0.4 u.g/ml puromycin (Calbiochem; La Jolla, California), as appropriate to the transfection. Once passaged, cells were maintained within the same plates for approximately 3-6 weeks. The transfected cells were washed every 3-4 days with D-PBS (8 ml/plate) to remove any cells that had perished. Individual drug-resistant clones were isolated when they became visible to the eye (approximately at the 64-128 cell stage). To isolate clones the cells were gently washed with room temperature D-PBS (8 ml/plate). The DPBS was then aspirated from the plate and sterile 9 mm Teflon cloning rings (Fisher Scientific; Ottawa, Canada) were placed over well-isolated, individual clones (note: the bottom of the 64  cloning rings were coated with a thin layer of high vacuum grease [Dow Corning; Midland, Michigan] so that a seal could be established between the cloning ring and the polystyrene tissue culture plate). One drop of trypsin solution was then added into the cloning ring to facilitate removal of the individual clone from the plate. Two minutes later 4 drops of the appropriate selection media were added to the cloning ring and the cells were re-suspended by gently pipetting the solution up and down. The re-suspended cells were then transferred to a well in a 24-well polystyrene tissue culture plate (Falcon) and topped up with the appropriate selection media (1 ml/well). As clones grew confluent they were successively passaged to larger wells (from a 24-well plate to a 12-well plate to duplicate 6 well plates). Once in six well plates, the clones of one duplicate plate were lysed while the clones in the other duplicate plate were maintained for further passage and experiments. Lysis was achieved essentially as described below (Chapter 2.4) with the exception that 150-300 ul of Triton X-100 lysis buffer (Chapter 2.14.21) was used per well of a six well plate. Successful expression of the desired plasmids' given products were confirmed by immunoblot analysis of the lysates. 2.4  Stimulation and lysis of cell lines  AtT20-derived cell lines were grown to approximately 90 % confluency on 10 cm polystyrene tissue culture plates. Prior to stimulation the cells were washed twice with 8 ml/plate of D-PBS containing lg/L dextrose (Fisher Scientific). Cells were then incubated in 8 ml/plate of modified HEPES-buffered saline (modified HBS; Chapter 2.1.19) for 15 minutes at 37 °C and 10 % C 0 . 2  BCR-mediated cell stimulation was achieved by cross-linking the B C R with 200 ug/plate of affinity purified goat anti-mouse IgM (u chain specific) antibody. Cells were then incubated at 37 °C and 10 % C 0 for the desired length of stimulation. To terminate stimulation cells were 2  washed twice with ice-cold D-PBS containing 1 m M Na3V04 (also known as sodium pervanadate; from Sigma-Aldrich Canada). Cells were then immediately lysed on ice for 20 minutes with 0.5-1.0 ml/plate of ice-cold Triton X-100 lysis buffer.  The lysate was then  transferred from the 10 cm plate to a 1.5 ml eppendorf tube (Eppendorf/Brinkmann Instruments; Mississauga, Ontario) and centrifuged at 14 000 rpm at 4 °C for 15 minutes to pellet out nuclei and other detergent insoluble material. The cleared lysate (supernatant) was then transferred to a fresh 1.5 ml eppendorf tube where sodium dodecyl sulfate (SDS; Bio-Rad Laboratories; Hercules, California) and deoxycholate (DOC; Fisher Scientific) were added to a final concentration of 0.3% and 0.4 %, respectively. The protein concentration of each sample was determined using the bicinchoninic acid (BCA) protein assay kit (Pierce Biotechnology). Lysates were stored at -20 °C. B cell lymphoma cell lines were grown to confluency either in 10 cm polystyrene tissue culture plates or in tissue culture flasks of various sizes (Falcon). Prior to stimulation cells were washed twice with modified HBS and then re-suspended in this buffer at a concentration of 25 x 10  6  cells/ml. Re-suspended cells were warmed to and maintained at 37 °C in a water bath for the 65  duration of the stimulation. BCR-mediated stimulation was achieved by cross-linking the B C R with either 100 ug/ml of affinity purified goat anti-mouse IgM (LI chain specific) antibodies or with 100 ug/ml affinity purified goat anti-human IgM (p. chain specific) antibodies, as appropriate.  To terminate the stimulation cells were washed twice with ice-cold D-PBS  containing 1 m M Na3V04. Cells were then lysed on ice for 20 minutes in 0.5-1.0 ml of ice-cold Triton X-100 lysis buffer. The lysate was then centrifuge at 14 000 rpm at 4 °C for 15 minutes to pellet out nuclei and other detergent insoluble material. The cleared lysate (supernatant) was then transferred to a fresh 1.5 ml eppendorf tube where SDS and D O C were added to a final concentration of 0.3% and 0.4%, respectively. The protein concentration of each sample was determined using the B C A protein assay kit. Lysates were stored at -20 °C. Non-stimulated cell lysates were prepared by washing the cells twice with room temperature DPBS. Cells were then lysed as described above. 2.5  SDS-PAGE and immunoblot analysis  Samples (whole cell lysates, immunoprecipitates and fractions) were mixed with 5x running sample buffer (5x RSB; mixed as 4 parts sample to 1 part 5x RSB; Chapter 2.14.28). Samples were then heated in a boiling water bath for 5 minutes prior to sodium dodecyl sulfatepolyacrylamide gel electrophoresis (SDS-PAGE).  Once boiled, samples were loaded into  individual wells of a polyacrylamide (Bio-Rad Laboratories) mini-gel (1.0 mm or 1.5 mm thick; 9%, 10 %, 12% or 15%). A 5 pi sample of Benchmark pre-stained protein molecular weight standard (Invitrogen Life Technologies) was also added to one well of the gel. Samples were then resolved at a constant current of 20 milliamps/gel for 2-3 hours in a dual vertical mini-gel apparatus (CSB Scientific; Del Mar, California) in running buffer (Chapter 2.14.29).  The  resolved proteins were then transferred from the gel to a Protran nitrocellulose filter (VWR International; Delta, British Columbia) in a Transblotter Transfer Apparatus (Bio-Rad Laboratories) at a constant voltage of 125 volts for 2 hours in transfer buffer (Chapter 2.14.30). Nitrocellulose filters were generally blocked for 1-2 hours at room temperature with Trisbuffered saline (TBS; Chapter 2.14.31) containing 5% skim milk powder. However, filters that were immunoblotted with the phospho-Erk antibody were blocked for only 15 minutes at room temperature with TBS containing 5% skim milk powder and filters that were immunoblotted with the 4G10 antibody were blocked for a minimum of 2 hours at room temperature with TBS containing 5% bovine serum albumin (BSA).  Blocked filters were rinsed once with TBS  containing 0.1% Tween 20 (TBST; Chapter 2.14.33) before being rocked overnight at 4°C in the desired primary antibody (10 ml solution/filter). Primary antibodies were diluted 1:500-1:2000 in TBST containing either 0-5% B S A or 0-5% skim milk powder (the ideal conditions for each antibody were determined empirically). The next day, filters were washed at room temperature for 4 x 15 minutes with TBST (50 ml/filter/wash). Filters were then rocked at room temperature 66  for 1 hour with the appropriate HRP-conjugated secondary reagent (10 ml solution/filter). Secondary reagents were diluted 1:5000 - 1:10 000 in TBST containing either 0-5% B S A or 05% skim milk powder (again, ideal conditions were determined empirically). Filters were then washed at room temperature for 4 x 15 minutes with TBST (50 ml/filter/wash).  The  immunoreactive bands on the filters were then visualized by enhanced chemiluminescence (ECL; Amersham Biosciences Canada) coupled with exposure of the filters to Kodak autoradiography film (Mandel Scientific; Guelph, Ontario). To re-immunoblot, filters were briefly re-hydrated in distilled water (50 ml/filter). Previously bound antibodies were removed from the re-hydrated filter by incubating the filter at room temperature for 15 minutes with Stripping TBS (Chapter 2.14.32) (50 ml/filter). Filters were then washed at room temperature for 2 x 15 minutes with TBS (50 ml/filter). Finally, filters were blocked and re-immunoblotted as described above. 2.6  Immunoprecipitation studies  Cell lysates were prepared essentially as described above (in Chapter 2.4) with cells being lysed with either Triton X-100 lysis buffer or n-Dodecyl-P-d-maltoside lysis buffer (DM Lysis Buffer; Chapter 2.14.22) or Nonidet P40 lysis buffer (NP40 Lysis Buffer; Chapter 2.14.23). 500-3000 jLXg  of lysate were pre-cleared for 30 minutes at 4 °C with either 20-50 |al of 50 % (v:v in D-PBS)  Protein A - or Protein G-Sepharose 4B beads (Sigma-Aldrich Canada). Pre-cleared lysates were then immunoprecipitated for 1 hour at 4 °C with 1-3 ug of immunoprecipitating antibody and 40100 ul of 50 % Protein A - or Protein G-Sepharose 4B beads. The beads were then collected by centrifugation and washed twice with 500 ul of the lysis buffer that corresponded with the original lysate preparation. The immunoprecipitated proteins were eluted from the beads with 35 ul of l x running sample buffer (Chapter 2.14.27). The immunoprecipitated samples were then resolved by SDS-PAGE mini-gels (as described in Chapter 2.5) with co-immunoprecipitation (i.e., co-association) of various proteins being investigated by immunoblotting (as described in Chapter 2.5). 2.7  Membrane enrichment of cell lines  AtT20-derived cell lines were grown to confluency on 10 cm polystyrene tissue culture plates (as described in Chapter 2.3.2). One plate of cells was used per treatment. Cells were removed from tissue culture plates using 5 ml of D-PBS containing 10 m M E D T A . Cells were incubated with the D-PBS/EDTA solution on a shaker at room temperature for 5-10 minutes. Cells were then collected in a 15 ml polypropylene conical centrifuge tubes (Falcon), centrifuged at low speed (1500 rpm at room temperature for 3 minutes in a tabletop centrifuge) and washed twice with 5 ml of room temperature D-PBS containing 1 g/1 glucose and once with 10 ml of modified HBS. Cells were then re-suspended in 1 ml of 37 °C modified HBS and warmed for 15 minutes in a 37 °C water bath. Cells were stimulated by cross-linking the B C R with affinity purified goat 67  anti-mouse IgM (fi chain specific) antibodies (200 ug/ml). After 5 minutes the stimulation was terminated by adding ice-cold D-PBS containing 1 m M of Na3VC>4 to the reaction (500 u,l/tube), centrifuging at low speed (1500 rpm at 4 °C for 3 minutes) and removing the supernatant from the pelleted cells.  Cells were washed once more with ice-cold D-PBS containing 1 m M of  Na3VG"4 (1 ml/tube). Cells were then re-pelleted, the supernatant was removed and the pelleted cells were frozen in liquid nitrogen for 15 seconds. The cells were then incubated on ice for 5 minutes with non-detergent lysis buffer (200 pi/tube) (Chapter 2.14.24). Following incubation, cells were re-suspended and then centrifuged at 4 °C for 5 minutes at 14 000 rpm.  The  supernatant (cytosolic fraction) was then transferred to a new 1.5 ml eppendorf tube containing 20 ul of 10 % Triton X-100 (v:v in distilled water). The remaining pellet was washed twice with 300 ul of non-detergent lysis buffer before being re-suspended in 220 ul of Triton X-100 lysis buffer and then centrifuged at 4 °C for 5 minutes at 14 000 rpm. The supernatant (membraneenriched fraction) was then transferred to a new 1.5 ml eppendorf tube and the remaining pellet was discarded. The protein concentration of each sample was determined using the B C A protein assay kit. Samples were stored at -20 °C. 2.8  Cytoskeletal-based lipid raft preparation  Cells were cultured for 16-20 hours in low serum RPMI-1640 media (Chapter 2.14.25). Cells were then washed twice with modified H B S and then re-suspended in this buffer at a concentration of 1.25 x 10 cells/ml. 7  For each condition, 2 ml of re-suspended cells were  warmed to and maintained at 37 °C in a water bath for the duration of the stimulation. Stimulation was achieved by cross-linking the components of the B C R with either 50 ug/ml of affinity purified goat anti-mouse IgM (u chain specific) antibodies or with 300 ug/ml of biotinylated anti-Ig(3 antibody (monoclonal HM79-16 antibody) plus 40ug/ml streptavidin, as appropriate. The cells were then washed twice with ice-cold D-PBS containing 1 mM Na3VC>4 to stop the stimulation. Cells were then subjected to cytoskeletal-based lipid raft preparation. Cytoskeletal-based lipid rafts were prepared essentially as described by Weintraub et al. (2000). Cells were re-suspended at a concentration of 1.25 x 10 cells/375 ul of ice cold Low Salt 7  Cytoskeleton Stabilization Buffer (Low Salt CSB; Chapter 2.14.25). After 10 minutes on ice, cells were centrifuged for 10 minutes at 2000 rpm in a 4 °C microfuge. The supernatant was collected and retained as the "detergent-soluble" supernatant fraction.  The pellet was re-  suspended in 375 ul of ice cold Low Salt CSB buffer and then centrifuged for 15 minutes at 2000 rpm in a 4 °C microfuge. The supernatant was then discarded as a wash and the pellet was re-suspended in 110 ul of ice-cold High Salt CSB (Chapter 2.14.26). The re-suspended pellet was centrifuged for 15 minutes at 2000 rpm in a 4 °C microfuge. The supernatant was collected and retained as the "detergent-insoluble, salt extractable" lipid raft fraction.  68  The protein  concentration of each fraction was determined using the B C A protein assay kit. Samples were stored at -20 °C until undergoing SDS-PAGE and immunoblot analysis. 2.9  Inositol phosphate assay  Inositol phosphate production was measured essentially as described by Matsuuchi et al., (1992). Please refer to Appendix 4 for further information regarding attempts to optimize this assay. Cells were plated out in six well plates (Falcon) by adding 2 x 10 cells in complete D M E M to 5  each well. Cells were then returned to a 10 % CO2, 37 °C incubator. After 48 hours the media was aspirated from the cells and replaced with 2 ml/well of low serum D M E M (Chapter 2.14.13). Cells were then returned to a 10 % C 0 , 37 °C incubator overnight. The next day cells 2  were washed twice with 2 ml/well of PBS containing l g / L glucose. Modified HBS containing 10 m M L i C l was then added back to the cells (1.5 ml/well). After 15 min at 10 % C 0 and 37 2  °C the cells were again washed twice with 2 ml/well of PBS containing lg/L glucose. Modified HBS containing 10 m M L i C l was then added back to the cells (1.5 ml/well) along with the appropriate stimulus (either 200 ul/well of 10 % fetal calf serum, 15 pi of 100 u M serotonin/well or 60 ug of affinity purified goat anti-mouse IgM [p chain specific] antibodies).  After 15  minutes at 10 % C 0 and 37 °C, the reactions were terminated by washing the cells twice with 2  PBS and then adding 1 ml/well of ice cold 10 % trichloroacetic acid. Cells were then removed from the plate using a rubber policeman and the contents of each well were transferred to a 1.5 ml eppendorf tube. Insoluble material was removed by centrifuging for 5 minutes at 14 000 rpm in a 4 °C microfuge. The supernatant was then transferred to a glass tube and extracted six times with ice cold water-saturated ethyl ether. The inositol phosphates were then collected from the extracted solution by running the solution over an A G 1-X8 Resin formate column (Bio-Rad Laboratories). The inositol phosphates were then eluted from the column with 20 ml of a 1 M ammonium formate/0.1 M formic acid solution. Five millilitres of the resulting eluate were then transferred to a scintillation vial and mixed with 5 ml of Ready Gel liquid scintillation cocktail (Beckman Coulter).  As well, the insoluble material (pellet) was re-suspended in 1 ml of  methanol, transferred to a scintillation vial and mixed with 5 ml of Ready Gel liquid scintillation cocktail. Samples were then counted for H-inositol using a Beckman LS 5000TA scintillation 3  counter (Beckman Coulter). 2.10 Population-based calcium flux assay Population-based calcium flux assays were performed essentially as described below. Please refer to Appendix 5 for further information regarding attempts to optimize this assay. Initially, 20 x 10 AtT20-derived cells were plated on adherent 10 cm tissue culture plates 6  (Falcon) and returned to a 10 % C 0 , 37 °C incubator overnight. The next day the media was 2  aspirated from the cells and replaced with 1.0 ml of serum-free D M E M containing 10.0 u M Fura-2 acetoxymethyl ester (Fura-2 A M ; Molecular Probes). Cells were returned to a 10 % C 0 , 2  69  37 °C incubator for 40 minutes. Following this, cells were gently washed from the plate with the existing media and pelleted by centrifugation for 5 minutes at 1500 rpm at 4 °C in an IEC Centra-8R Tabletop Centrifuge (International  Equipment Company; Needham Heights,  Massachusetts) and re-suspended in 3 ml of modified HBS. The re-suspended cell sample was then placed in a 3 ml cuvette that was then placed in a Perkin Elmer LS50B Luminescence Spectrometer and allowed to warm to 37 °C for 5 minutes. The appropriate stimulus (6.0 u M serotonin or 50-100 ug of affinity purified goat anti-mouse IgM [u chain specific] antibody) was then added to the sample. Two minutes later 2.0 u M of ionomycin (an ionophore) were added to the sample as a positive control. Following another two minutes the cells were lysed by the addition of 0.1 m M digitonin as a further positive control. Finally, two minutes later, 70 u M of Tris and 10 u M of E D T A were added to the sample to chelate the calcium as yet another control. Throughout the experiment calcium flux was monitored by determining the ratio of the fluorescence intensities at 510 nm (emission wavelength) that was produced by excitation of Fura-2 A M at 340 nm and 380 nm (excitation wavelengths) as an increase in the ratio of fluorescence intensities (340 nm/380 nm) is indicative of increased calcium-binding by Fura-2 AM. 2.11 Single cell-based calcium assay Single cell-based calcium assays were performed essentially as described in Church et al. (1998). Cells were initially plated on 15 mm poly-D-lysine coated coverslips. Poly-D-lysine coated coverslips were prepared by submerging the coverslips in a poly-D-lysine solution (1 mg/ml poly-D-lysine [Sigma-Aldrich Canada], 50 m M sodium borate pH 8.5) for two hours at room temperature.  Coverslips were then rinsed three times with PBS followed by three times  with distilled water. Once confluent, cells were loaded with Fura-2 A M by placing the coverslip (cell side up) in 2 ml of modified HBS containing 7.5 u M Fura-2 A M for 1 hour at 10 % C 0  2  and 37 °C. Cells were then rinsed for 10 minutes at room temperature in 2 ml of modified HBS. The coverslip was then mounted in a temperature controlled perfusion chamber such that the coverslip (cell side up) formed the base of the chamber. The cells were then superfused with the modified HBS at a rate of 2 ml/minute. After 2 minutes the cells were superfused with calciumfree modified H B S (Chapter 2.14.20) at a rate of 2 ml/minute. After 5 minutes the flow was stopped and serotonin (or the agonist of choice) was added to the solution to a final concentration of 6 u M . After 2 minutes the flow of calcium free modified HBS was resumed to rinse the serotonin from the chamber. Two minutes later the flow was switched to modified HBS. Cells were superfused with the calcium containing solution for 10 minutes to enable reloading of the cells' calcium stores. Cells were then once again superfused with calcium-free modified HBS at a rate of 2 ml/minute. After 5 minutes the flow was stopped and the PLC activator, m-3M3FBS, (or agonist of choice) was added to the solution to a final concentration of 50 u M . After 2 minutes the flow of calcium free modified HBS was resumed to rinse the P L C activator from the  70  chamber. Finally, after 2 minutes, flow was stopped and ionomycin was added to the solution to a final concentration of 10 u M . Throughout the experiment calcium flux was monitored utilizing a digital fluorescence microscopy system (Atto Instruments, Incorporated; Rockville, Maryland; and Carl Zeiss Canada Limited; Don Mills, Ontario). This system allows for the simultaneous real-time monitoring of fluorescence emissions from single cells. In particular, the intensity of fluorescence emissions at 510 nm were monitored following excitation of Fura-2 A M at 334 nm and 380 nm where an increase in the ratio of emission intensities (334 nm/380 nm) indicates of increased calciumbinding by Fura-2 A M . 2.12 Production of anti-BLNK polyclonal antibodies Rabbit polyclonal antibodies were produced against peptides corresponding to specific regions of BLNK.  Vectors encoding the peptides fused in-frame with Glutathione S-Transferase (GST)  were kindly provided by Rob Ingham (from Anthony Pawson's Lab at the Lunenfeld Institute at the University of Toronto; Ontario, Canada) (refer to Table 2.2). The GST-fusion peptides were prepared and purified by Gabe Woollam, a former undergraduate student of the Matsuuchi Laboratory. Briefly, a single bacteria colony was used to inoculate 20 ml of L B Broth containing 100 pg/ml ampicillin. The culture was then incubated overnight at 37 °C before being used to further inoculate 1 L of L B broth containing 100 g/ml ampicillin.  This culture was then  incubated at 37 °C until the optical density (ODgoonm) of the culture was between 0.8-1.0 (approximately 4-6 hours) at which point isopropylthio-P-galactopyranoside (Invitrogen Life Technologies) was added to a final concentration of 100 u M . The culture was then incubated overnight at 37 °C with shaking. The next day, bacteria were pelleted by centrifugation and resuspended in 10 ml of bacteria lysis buffer II (Chapter 2.14.9) and incubated on ice for 30 minutes. The lysate was then sonicated for 2 minutes on ice before being centrifuged for 45 minutes at 30,000 rpm at 4 °C in a SW70 Ti rotor and a Beckman L8-70 Ultracentrifuge. The GST-fusion peptide was then purified from the cleared lysate using glutathione-Sepharose 4B beads (Sigma-Aldrich Canada). Lysate and beads were incubated together for 1 hour at 4 °C to allow the GST-fusion peptide to bind to the beads.  The beads were then collected by  centrifugation and washed three times with bead wash buffer (Chapter 2.14.10). Bound GSTfusion peptide was then eluted from the beads using a mild non-denaturing elution buffer (50 m M Tris base, 20 m M glutathione [Amersham Biosciences Canada], pH 8.0). Subsequently, the eluate was dialyzed against 10 mM Tris-HCl (pH 8.0) at 4 °C to remove any free glutathione. The concentration of the GST-fusion peptide was determined by measuring the OD 8o 2  (OD28o/l-46 = mg/ml of protein) and its purity confirmed by SDS-PAGE and Coomassie Blue staining.  71  Once purified, the GST-fusion protein was emulsified in complete Freund's Adjuvant (Difco Laboratories; Detroit, Michigan) for the initial injection and then in Incomplete Freund's Adjuvant (Difco Laboratories) for subsequent injections. Rabbits, TJ1 and TJ2, were injected with the G S T - B L N K SH2 fusion peptide (Table 2) while TJ3 was injected with the G S T - B L N K Proline Rich fusion peptide (Table 2). A l l injections were intramuscular and were carried out approximately every 18 days. 7-10 days following injection, approximately 10 ml of blood was collected from the ear vein of each rabbit. The blood sample was then stirred and left overnight at 4 °C to promote clotting. The next day the clot was carefully removed and the remaining serum was stored at -80 °C. Working aliquots were stored in 50 % glycerol at -20 °C. Table 2.2. Summary of Fusion Peptides Used to Immunize Rabbits to Produce A n t i - B L N K Antibodies. Note that PB refers to prebleed, B to Bleed and T B to termination Bleed. TJxBy refers to the rabbit number (x) and bleed number (y). Description  Fusion Peptide GST  G S T - B L N K SH2  (aa 1-242) fused in-frame to B L N K ' s  Corresponding  Resulting  Rabbit  Antibodies  TJ1 andTJ2  TJ1PB, TJB1-10, TJTB 1-7  SH2 domain (aa 386-507)  TJ2PB, TJ2B1-11, TJ2TB1-6 GST-BLNK Rich (P-Rich)  Proline  GST  (aa 1-242) fused in-frame to B L N K ' s  TJ3  TJ3PB,TJ3Bl-9, TJ3TB1-7  Proline rich domain (aa 152-388)  2.13 Surface biotinylation of cells AtT20-derived cell lines were grown to confluency on 10 cm polystyrene tissue culture plates (as described in Chapter 2.3.2). Cells to be labeled were rinsed five times with l x ice-cold PBS. After rinsing, 2.5 ml of ice-cold Sulfo-NHS-biotin/lx PBS solution was added per 10 cm plate (Sulfo-NHS-biotin from Pierce Biotechnology; Sulfo-NHS-biotin/lx PBS solution made by initially dissolving 25 mg of Sulfo-NHS-biotin in 125ul of D M S O and then further diluting 25 ul of the Sulfo-NHS-biotin/DMSO solution per 10 ml of lx PBS). The plate was then placed on ice on a rocker for 20 minutes. After 20 minutes the reaction was quenched by adding 5 ml of icecold D M E M containing 2mg/ml lysine was per plate.  The D M E M solution was then  immediately aspirated off, the plate rinsed with 10 ml of l x PBS containing 2 mg/ml lysine, and 10 ml of fresh D M E M containing 2 mg/ml lysine added back to the plate. The plate was then left to rock for an additional 5 minutes. Subsequently, the plate was rinsed 8x with ice-cold l x PBS containing 2mg/ml lysine prior to be lysed as described above in Chapter 2.4. . 2.14 Summary of solutions 2.14.1 Tris/Boric Acid/EDTA (TBE) 90 m M  Tris-HCl (pH 8.2)  90 m M  boric acid (Fisher Scientific) 72  2 mM  ethylenedinitrilotetraacetic acid (EDTA)  2.14.2 6x DNA Sample Buffer 0.24 %  bromophenol blue (Sigma-Aldrich Canada)  0.24 %  xylene cyanol FF (Sigma-Aldrich Canada)  60 % (w/v)  sucrose  2.14.3 Lauria-Bertani Broth (LB Broth) 5g  NaCl  5g  yeast extract (Difco Laboratories)  Total volume made up to 1 L with distilled water 2.14.4 Lauria-Bertani Agar (LB Agar) 5g  NaCl  5g  yeast extract (Difco Laboratories)  15 g  agar (Difco, Laboratories)  Total volume made up to 1 L with distilled water 2.14.5 M9 Growth Media 100 ug/ml 1L  ampicillin (Sigma-Aldrich Canada) M9 salts (refer to Molecular Cloning Laboratory Manual, Second Edition by Sambrook, Fritsch and Maniatis)  1 uM  CaCl  1 mM  MgSCU  0.4 %  glucose  0.4 %  cosamino acids  2  0.004 %  proline  0.004 %  leucine  0.004 %  threonine  2.14.6 Sucrose Solution 25 % 50 m M 1 mM  sucrose Tris (pH 8.0) (ICN Biomedicals) E D T A (pH 8.0)  2.14.7 Bacteria Lysis Buffer I 5 mg/ml  lysozyme (Sigma-Aldrich Canada)  25 m M  Tris (pH 8.0)  2.14.8 Triton Lytic Mix 73  0.1 % 62.5 m M 50 m M  Triton X - 100 E D T A (pH 8.0) Tris (pH 8.0)  2.14.9 Bacteria Lysis Buffer II 50 m M 150 m M 1% 1 mg/ml 0.1 mg/ml  Tris-HCl (pH 8.0) NaCl Triton-X 100 (Sigma-Aldrich Canada) Lysozyme (Sigma-Aldrich Canada) DNAsel  10 mg/ml  soybean trypsin inhibitor (Roche Diagnostics)  10 ug/ml  leupeptin (Roche Diagnostics)  1 ug/ml  aprotinin (Roche Diagnostics)  1 mM  phenymethyl'sulfonyl fluoride (Roche Diagnostics)  2.14.10 Bead Wash Buffer 25 m M 150 m M 0.1 %  Tris-HCl (pH 8.0) NaCl Triton-X 100 (Sigma-Aldrich Canada)  10 mg/ml  soybean trypsin inhibitor  10 ug/ml  leupeptin (Roche Diagnostics)  1 ug/ml  aprotinin (Roche Diagnostics)  1 mM  phenymethylsulfonyl fluoride (Roche Diagnostics)  2.14.11 Tris/EDTA (TE) Tris (pH 8.0) (ICN Biomedicals)  10 m M 1 mM  E D T A (pH 8.0)  2.14.12 Complete D M E M 500 ml  D M E M (containing 4.5g/L glucose, 2 m M L-glutamine, 110 mg/L sodium pyruvate; Invitrogen Life Technologies)  50 ml 50 units/ml 50 ug/ml  heat-inactivated fetal calf serum (Invitrogen Life Technologies) penicillin (Invitrogen Life Technologies) streptomycin sulfate (Invitrogen Life Technologies)  2.14.13 Low Serum D M E M 500 ml  D M E M (containing 4.5g/L glucose, 2 m M L-glutamine, 110 mg/L sodium pyruvate; Invitrogen Life Technologies)  1 ml 50 units/ml  heat-inactivated fetal calf serum (Invitrogen Life Technologies) penicillin (Invitrogen Life Technologies) 74  50 ug/ml  streptomycin sulfate (Invitrogen Life Technologies)  2.14.14 Complete RPMI-1640 500 ml  RPMI-1640 (Invitrogen Life Technologies)  50 ml  heat-inactivated fetal calf serum (Invitrogen Life Technologies)  2 mM  L-glutamine (Invitrogen Life Technologies)  1 mM  sodium pyruvate (Invitrogen Life Technologies)  50 uM 50 units/ml 50 ug/ml  2-(3-mercaptoethanol (Sigma-Aldrich Canada) penicillin (Invitrogen Life Technologies) streptomycin sulfate (Invitrogen Life Technologies)  2.14.15 Low Serum RPMI-1640 500 ml 1 ml  RPMI-1640 (Invitrogen Life Technologies) heat-inactivated fetal calf serum (Invitrogen Life Technologies)  2 mM  L-glutamine (Invitrogen Life Technologies)  1 mM  sodium pyruvate (Invitrogen Life Technologies)  50 u M 50 units/ml 50 ug/ml  2-P-mercaptoethanol (Sigma-Aldrich Canada) penicillin (Invitrogen Life Technologies) streptomycin sulfate (Invitrogen Life Technologies)  2.14.16 Trypsin Solution (Invitrogen Life Technologies) 0.25 % 1 mM  trypsin in D M E M EDTA  2.14.17 2x HEPES-Buffered Saline (2x HBS) 50 m M 10 m M 12 m M 280 m M 1.5 m M  HEPES (pH 7.2) (Sigma-Aldrich Canada) KC1 glucose NaCl Na HP0 2  4  2.14.18 lx HEPES-Buffered Saline (lx HBS) 25 m M 5 mM 6 mM  HEPES (pH 7.2) (Sigma-Aldrich Canada) KC1 glucose  140 m M  NaCl  750 uM  Na HP0 2  4  75  2.14.19 Modified HBS 25 m M 125 m M  HEPES (pH 7.2) (Sigma-Aldrich Canada) NaCl  5 mM  KC1  1 mM  CaCl  1 mM  Na HP0  2  2  500 u M  MgSCU  1 mg/ml  glucose  4  2mM  L-glutamine  1 mM  sodium pyruvate  2%  B S A (ICN Biomedicals).  2.14.20 Calcium Free Modified HBS HEPES (pH 7.2) (Sigma-Aldrich Canada) 25 m M 125 m M  NaCl  5 mM  KC1  1 mM  Na HP0 2  1.5 m M  MgSCU  1 mg/ml  glucose  4  2mM  L-glutamine  1 mM  sodium pyruvate  2%  B S A (ICN Biomedicals)  2.14.21 Triton X-100 Lysis Buffer 20 m M 137 m M 1 % 2mM 10%  Tris-HCl (pH 8.0) NaCl Triton X-100 (Sigma-Alrdich Canada) EDTA glycerol  10 ug/ml  leupeptin (Roche Diagnostics)  1 ug/ml 1 mM  aprotinin (Roche Diagnostics) pepstanin A (Sigma-Aldrich Canada)  1 mM  N a V 0 (Sigma-Aldrich Canada)  1 mM  phenylmethylsulfonyl fluoride (Roche Diagnostics)  3  4  2.14.22 D M Lysis Buffer 20 m M 137 m M 1%  Tris-HCl (pH 8.0) NaCl n-dodecyl-P-d-maltoside (Sigma-Alrdich Canada) 76  2 mM  EDTA  10 %  glycerol  10 ug/ml  leupeptin (Roche Diagnostics)  1 ug/ml  aprotinin (Roche Diagnostics)  1 mM  pepstanin A (Sigma-Aldrich Canada)  1 mM  N a V 0 (Sigma-Aldrich Canada)  1 mM  phenylmethylsulfonyl fluoride (Roche Diagnostics)  3  4  2.14.23 NP40 Lysis Buffer 10 m M 150 m M 1%  Tris-HCl (pH 8.0) NaCl Nonidet P40 (NP40) (BDH Laboratory Supplies; distributed by V W R Canlab; Mississauga, Ontario)  10 ug/ml  leupeptin (Roche Diagnostics)  1 ug/ml  aprotinin (Roche Diagnostics)  1 mM  pepstanin A (Sigma-Aldrich Canada)  1 mM  N a V 0 (Sigma-Aldrich Canada)  1 mM  phenylmethylsulfonyl fluoride (Roche Diagnostics)  3  4  2.14.24 Non-Detergent Lysis Buffer 10 m M 137 m M 10% 2 mM  Tris-HCl (pH 8.0) NaCl glycerol EDTA  10 ug/ml  leupeptin (Roche Diagnostics)  1 ug/ml  aprotinin (Roche Diagnostics)  1 mM  pepstanin A (Sigma-Aldrich Canada)  1 mM  Na3V04 (Sigma-Aldrich Canada)  1 mM  phenylmethylsulfonyl fluoride (Roche Diagnostics)  2.14.25 Low Salt Cytoskeletal Stabilization Buffer 20 m M 0.5 % 10.3 % 20 m M 5 mM  HEPES (pH 7.6) Triton X-100 (Sigma-Alrdich Canada) sucrose NaCl  0.1 mg/ml  MgCl B S A (ICN Biomedicals)  500 ug/ml  leupeptin (Roche Diagnostics)  500 ug./ml  aprotinin (Roche Diagnostics)  2  77  0.8 ug/ml 2mM 2.5 m M  pepstanin A (Sigma-Aldrich Canada) sodium orthovanadate phenylmethylsulfonyl fluoride (Roche Diagnostics)  2.14.26 High Salt Cytoskeletal Stabilization Buffer 20 m M 0.5 % 10.3 % 150 m M 5 mM  HEPES (pH 7.6) Triton X - 100 (Sigma-Alrdich Canada) sucrose NaCl MgCl  2  0.1 mg/ml  B S A (ICN Biomedicals)  500 ug/ml  leupeptin (Roche Diagnostics)  500 ug./ml  aprotinin (Roche Diagnostics)  0.8 ug/ml 2 mM 2.5 m M  pepstanin A (Sigma-Aldrich Canada) sodium orthovanadate phenylmethylsulfonyl fluoride (Roche Diagnostics)  2.14.27 lx Running Sample Buffer (RSB) 62.5 m M 4% 2.5% 0.02 % 100 m M  Tris-HCl (pH 6.8) glycerol SDS • bromophenol blue (Sigma-Aldrich Canada) dithiothreitol (DTT) (Sigma-Aldrich Canada)  2.14.28 5x Running Sample Buffer (RSB) 312.5 m M 20 % 12.5 % 0.1 % 500 m M  Tris-HCl (pH 6.8) glycerol SDS bromophenol blue (Sigma-Aldrich Canada) dithiothreitol (DTT) (Sigma-Aldrich Canada)  2.14.29 Running Buffer 50 m M  Tris (ICN Biomedicals)  0.4 M  glycine (Fisher Scientific)  0.1 %  SDS  2.14.30 Transfer Buffer 20 m M  Tris-HCl (pH 8.0) 78  150 m M 20 %  glycine Methanol  2.14.31 Tris Buffered Saline (TBS) lOmM 150 m M  Tris-HCl (pH 8.0) NaCl  2.14.32 Stripping Tris Buffered Saline (TBS) lOmM 150 m M  Tris-HCl (pH 2.0) NaCl  2.14.33 Tris Buffered Saline with Tween 20 (TBST) lOmM  Tris-HCl (pH 8.0)  150 m M  NaCL  0.1 % 2.15.  Tween 20 (Sigma-Aldrich Canada)  Mean Pixel Intensity Analysis of BCR-Induced Erk Phosphorylation Immunoblots  To calculate the mean pixel intensity of the phosphorylated Erk "band" it was first necessary to define the "band". This was done by visually identifying the largest band in the immunoblot and then manually selecting a square area around that band that captured all the pixels of the band yet minimized the capture of background pixels. A n equivalent square area was then selected around the remaining bands again ensuring that all the pixels of the band were captured. The mean pixel intensity of the "bands" was then calculated using Matlab (The Mathworks; Natick, Massachusetts). By this method the mean pixel intensity should correlate to the band size and intensity, and this in turn should correlate with the amount of protein actually present in the immunoblot. Thus, a larger mean pixel intensity should represent a larger and more intense band, and this in turn should represent a larger amount of protein present in the immunoblot. With this method "bands" within the same immunoblot can be compared. However, this method can not be used to compare "bands" across different immunoblots as pixel intensity can vary across immunoblots due to differences in exposure quality as well as differences in the actual amount of protein present in the immunoblot. Nonetheless, this method allows one to determine whether a trend in band intensity is apparent within an immunoblot and to determine i f this trend is consistent across various immunoblots.  79  CHAPTER 3  The Solo Iga/p Heterodimer Can Localize to L i p i d Rafts  3.1  Introduction  Spatiotemporal regulation of protein interactions is an essential aspect of cellular signaling. Such regulation is often achieved via compartmentalization whereby specific proteins are sequestered within defined cellular locations.  These defined cellular locations include  specialized membrane microdomains within the plasma membrane termed lipid rafts (detailed in Chapter 1.5.2). In the past decade, lipid rafts have gained prominence in the B cell field where they appear to play a role in regulating certain aspects of B C R signaling (reviewed in Dykstra et al; 2003, Pierce, 2002; Dykstra et al., 2001; Cherukuri et al, 2001; Matsuuchi and Gold, 2001).  Numerous groups have reported that the B C R translocates into lipid rafts following B C R crosslinking and that such translocation is necessary for B C R uptake, for antigen targeting and for B C R signaling (Su et al., 2002; Stoddart et al,  2002; Cheng et al,  2001; Aman and  Ravichandran, 2000; Guo et al, 2000; Petrie et al, 2000; Wientraub et al, 2000; Cheng et al, 1999). Nonetheless, some reports suggest that the B C R does not always translocate into lipid rafts upon cross-linking (Sproul et al, 2000; Wientraub et al, 2000). In particular, Wientraub and his colleagues (2000) demonstrated that the B C R translocates into lipid rafts following cross-linking in naive B cells but not in tolerant B cells. Furthermore, they confirmed that BCR signaling was robust in naive B cells where the B C R translocates but limited in tolerant cells where the B C R did not translocate. Thus, it was suggested that lipid raft inclusion/exclusion of the B C R may serve as a mechanism for regulating B C R signaling in different B cell types. Additionally, Sproul and his colleagues (2000) reported that B C R signaling can occur outside of lipid rafts in immature B cell and lead to apoptosis. Similarly, Trujillo and his colleagues (2003) demonstrated that lipid raft disruption did not inhibit BCR-induced apoptosis in a variety of B cell lines. Thus, lipid raft inclusion/exclusion may not serve merely as an on/off switch for B C R signaling but rather as a mechanism that couples the B C R to different signaling pathways. Indeed, it may be that inclusion within lipid rafts would compartmentalize the B C R with signaling components that lead to B cell proliferation and differentiation whereas exclusion from  80  lipid rafts may compartmentalize the B C R with signaling components that lead to B cell anergy and apoptosis.  If lipid raft inclusion/exclusion does truly represent a mechanism for regulating BCR signaling the question arises as to how lipid raft inclusion/exclusion is itself regulated. In one scenario, it is hypothesized that B C R lipid raft translocation is regulated as a consequence of BCR signaling itself. In this case, cross-linking of the B C R may induce signal-dependent protein associations that facilitate B C R lipid raft translocation. However, this scenario seems unlikely as the B C R has been found to translocate into lipid rafts prior to any detectable B C R signaling events (Weintraub et al, 2000) and even in the complete absence of B C R signaling (Cheng et al, 2001). In an alternate scenario it is hypothesized that B C R lipid raft translocation is regulated as a consequence of B C R cross-linking itself. In this case, B C R cross-linking is envisioned to cause a structural change within the B C R that facilitates its association with lipid rafts. Such structural changes may involve exposing previously hidden "lipid raft affinity domains" (LRADs) within the B C R or involve the assembling of L R A D s via oligomerization of the BCR. Interestingly, previous studies have demonstrated that the mlgM subunit alone is able to translocate into lipid rafts upon cross-linking (Cheng et al, 2001) That is to say, that the mlgM subunit does not have to be associated with the Iga/p signaling subunit in order to translocate 1  into lipid rafts (Cheng et al, 2001). This suggests first, that mlgM lipid raft translocation occurs independent of signaling and second, that the mlgM itself may contain structural features that facilitate lipid raft translocation upon cross-linking.  However, the precise nature of these  structural features remains to be determined as does whether or not these structural features are sufficient and/or necessary to direct the intact B C R to the lipid rafts. In this chapter, I further investigate the mechanisms involved in regulating B C R lipid raft translocation.  Central to these investigations was the possession of the WEHI 303.1.5 B cell line, a unique B cell line that expresses the Iga/p signaling subunit on its cell surface despite the absence of mlgM (Condon et al, 2000).  Using this cell line, the ability of the Iga/p subunit to  independently translocate into lipid rafts was assessed and compared to that of the mlgM subunit (as reported by Cheng et al, 2001) and of the intact B C R itself. Essentially, by comparing the independent translocation abilities of the mlgM subunit and the Iga/p subunit we are performing a gross deletion analysis of the B C R . With this approach it is possible to compare the translocation abilities and the structures of the two subunits in an attempt to narrow down the 81  possible regions of the proposed LRADs.  For example, if it is found that each subunit can  translocate into lipid rafts independently it may be suggested that the subunits share a common structural feature that facilitates this translocation.  Thus, it would be advisable to perform  further mutational analyses within any common structural regions (e.g., the transmembrane domains or the membrane proximal regions) in an effort to identify a possible L R A D . Alternatively, i f only one subunit is found to translocate it may be suggested that that subunit contains a unique structural feature (as compared to the other subunit) that facilitates its translocation. Thus, it would be advisable to perform further mutational analyses within the unique structural regions of the translocatable subunit.  Using this approach it was determined that both the antigen-binding subunit (mlgM) (previously reported by Cheng et al., 2001) and the signaling subunit (Iga/Igp) (reported here) are capable of localizing to lipid rafts, independent of each other. This finding is important in several aspects. First, as described above, it suggests that the common structural domains (e.g., the transmembrane domains and membrane proximal regions) of these subunits should be further investigated in an attempt to identify a possible L R A D . And second, the finding that solo Iga/p' can localize to lipid rafts is biologically relevant as the solo Iga/p of Pro-B cells has been proposed to function as a pro-BCR. Given this, the finding that Iga/p can localize to lipid rafts in the absence of mlgM may suggest that lipid rafts play a role in pro-BCR signaling  3.2  3.2.1  Results  The B C R Translocates into Lipid Rafts in the Immature B Cell Lines, W E H I 231  and CH31  The mlgM-deficient variant of the WEHI 231 immature B cell line, WEHI 303.1.5, was used to assess the ability of Iga/p to translocate into lipid rafts.  WEHI 303.1.5 has previously been  shown to express Iga/p on its surface despite the absence of mlgM expression (Fig. 3.1) (Condon et al, 2000). Surface expression of solo Iga/p is abnormal in immature and mature B cells as quality control mechanisms typically ensure that incompletely assembled forms of the BCR are retained in the endoplasmic reticulum (ER).  However, the Iga/p of the WEHI 303.1.5 cell line  Solo Iga/p is used throughout this thesis to denote Iga/p heterodimers that are expressed on the cell surface despite the absence of mlgM expression. 1  • 82  contains a point mutation that encodes for a proline to leucine change at amino acid 126 within the extracellular domain of Iga. This change is proposed to inhibit Iga's interaction with ER chaperone proteins such that the solo Iga/p subunit is able to escape ER retention. While this mutation may make the WEHI 303.1.5 cell line amenable for the studies discussed herein it also presents a caveat to any findings. In particular, this mutation may alter the structure of the Iga/p subunit such that the mutant form may associate with lipid rafts in a manner different from that of the wild-type. To control for this possibility, investigations were also performed with the K40B-1 pro-B cell line that naturally expresses solo Iga/p on its cell surface in association with the chaperone protein, calnexin (Shapiro et al., 1993; Matsuuchi, unpublished observations). Finally, the WEHI 231 immature B cell line, the CH31 immature B cell line and the K40B-2 preB cell line, all of which express fully intact BCRs on their cell surfaces, were used as positive controls to ensure that the B C R cross-linking and lipid raft isolation protocols were functional (these cell lines have been previously described in Condon et al, 2000; Shapiro et al, 1993, respectively) (please note that the B C R expression patterns of the above-mentioned cell lines are confirmed in Fig. 3.1).  Before solo Iga/p lipid raft association could be considered it was necessary to first confirm lipid raft association of the intact B C R in immature B cell lines. While many studies have indicated that the B C R translocates into lipid rafts following B C R cross-linking (Cheng et al., 2001; Petrie et al, 2000; Weintraub et al., 2000; Cheng et al., 1999) two recent studies have suggested that this may not be the case for immature B cell lines (Sproul et al, 2000; Cheng et al, 2001). Thus the WEHI 231 and CH31 immature B cell lines were used to investigate BCR/lipid raft translocation in the immature B cells. These cell lines were initially stimulated by cross-linking the B C R with a u heavy chain specific antibody for the indicated length of time (time "0" indicates non-stimulated cells). Lipid rafts were then collected from the cell lines using a saltextractable lipid raft isolation method. This same method was previously used to demonstrate that the B C R translocates into lipid rafts in naive splenic B cells following B C R cross-linking (described in Weintraub et al., 2000). The fractions were subsequently resolved by SDS-PAGE, transferred to nitrocellulose filters and immunoblotted with Iga, IgP, u heavy chain and K light chain specific antibodies (note that the u heavy chain and K light chain together form the mlgM subunit in these cell lines). Filters were also immunoblotted with a CD45 specific antibody (data not shown) as a control to ensure that the lipid raft fraction was not contaminated with a non-raft  83  fraction. The protein tyrosine phosphatase, CD45, is well-known to be excluded from lipid rafts (Rogers and Rose, 1996) and thus, should not be found in pure lipid raft fractions.  Using the above approach, the levels of Iga/p, u heavy chain and K light chain decrease over time in the supernatant fraction of the WEHI 231 immature B cell line following receptor crosslinking (Fig. 3.2; left portion of the image). Concomitantly, the levels of these proteins increase over time in the lipid raft fraction (Figs. 3.2; right portion of the image). Similarly, Iga, IgP, u heavy chain and K light chain also appear to translocate into the lipid raft fraction following ubased B C R cross-linking in the CH31 immature B cell line (Fig. 3.3). Thus, these findings suggest that the B C R translocates into lipid rafts following anti-u-based B C R cross-linking in these immature B cell lines. As these findings are controversial (refer to Sproul et al, 2000; Cheng et al, 2001), a second method, namely the discontinuous sucrose gradient method (described in Deans et al, 1998), was used to isolate lipid rafts in an attempt to support these findings. The results obtained with this second methodology are in agreement with those of the salt-extractable lipid raft method discussed above (experiments were performed by Lorna Santos, formerly of the Matsuuchi lab; published in Jackson et al, 2005). Additionally, the WEHI 231 and CH31 cell lines were confirmed to undergo apoptosis in response to B C R cross-linking as is expected of healthy immature B cell lines (data not shown). Thus, given the findings from the two different immature B cell lines and the two different lipid raft isolation methods, it appears that a portion of the intact B C R is indeed localized within lipid rafts.  Moreover, the B C R  appears able to translocate into lipid rafts following receptor cross-linking in immature B cell lines.  84  m o  CM  CO o CO  LU  X LU  ro  CN  00  ro  o  U  ~c E  X  5  i9a  — 25 kD  — 36 kD  — 25 kD — 80 kD  u heavy chain Akt  — 64 kD  Figure  3.1. Characterization of Iga, IgP and mlgM (as determined by u, chain) expression in the experimental B cell lines. 40 ug o f whole cell lysate was resolved by S D S - P A G E for the K 4 0 B cell lines. In contrast, o n l y 2.0 ug o f whole cell lysate was resolved by S D S - P A G E for the W E H I 231, W E H I 303.1.5 and C H 3 1 cell lines. F o l l o w i n g electrophoresis, the gels were transferred to nitrocellulose filters that were subsequently immunoblotted with I g a , IgP or u chain specific antibodies (as indicated on the left-hand side o f the panels). Similarly A k t expression was characterized in these cell lines using 15 u.g o f whole cell lysate. A k t expression was characterized as increases in A k t phosphorylation are indicative o f successful B C R cross-linking (Ingham et al., 2001).  85  WEHI 231 Cell Line supernatant Anti mlgM (min) ->.|o  3 5  10 15  301  lipid raft 0  5  10 15  30  mm — m» *•»  igp  p. heavy chain —£  ft  N 11 M tl M  K light chain -  P-Akt  37 kD  MiilJ  Iga »~  3  25 kD -  50 kD  -  37 kD  -  80 kD  -  25 kD  64 kD  Figure 3.2. The B C R translocates into a detergent-insoluble, salt-extractable lipid raft fraction in the W E H I 231 immature B cell line following B C R cross-linking. 25 x 10 cells were stimulated with 100 ug of anti mlgM antibody for the indicated time points. The detergent soluble fraction (supernatant) and the detergent-insoluble, saltextractable fraction (lipid raft) were then collected. Equivalent amounts of protein for each fraction were resolved by S D S - P A G E . Following electrophoresis the gels were transferred to nitrocellulose filters that were subsequently immunoblotted with Iga, Igp\ K light chain or u chain specific antibodies (the latter two being components of mlgM). Successful receptor cross-linking was confirmed by monitoring for increased Akt phosphorylation (bottom panel). Data are representative of three independent experiments. 6  86  CH31 Cell Line supernatant Anti mlgM (min) Iga  lipid raft  0 3 5 10 15 30 0 3 5 1015 30  —  - 25 kD igp  •C  - 36 kD - 25 kD  heavy chain — | " ^ "  - 80 kD  K light chain  - 25 kD  P-Akt  - 64 kD  Figure 3.3. The B C R translocates into a detergent-insoluble, salt-extractable lipid raft fraction in the CH31 immature B cell line Following B C R cross-linking. 25 x 10 cells were stimulated with 100 itg o f anti m l g M antibody for the indicated time points. The detergent soluble fraction (supernatant) and the detergent-insoluble, saltextractable fraction (lipid raft) were then collected. Equivalent amounts o f protein for each fraction were resolved by S D S - P A G E . F o l l o w i n g electrophoresis the gels were transferred to nitrocellulose filters that were subsequently immunoblotted with I g a , Igf3, K light chain or u chain specific antibodies (the latter two being components o f m l g M ) . Successful receptor cross-linking was confirmed by monitoring for increased A k t phosphorylation (bottom panel). Data are representative o f three independent experiments. 6  87  3.2.2  A Portion of Solo Iga/p Localizes to Lipid Rafts in the Mutant Immature B Cell  Line, W E H I 303.1.5.  As cross-linking appears necessary to initiate B C R translocation (Cheng et al., 2001; Petrie et al., 2000; Weintraub et al., 2000; Cheng et al, 1999) it was hypothesized that cross-linking would similarly be required to induce solo Iga/p translocation. Thus, it was necessary to develop a method to successfully cross-link Iga/p. To this end, the WEHI 231 cell line was stimulated by cross-linking the IgP with a biotinylated anti-IgP mAb and streptavidin. Lipid rafts were then collected from the cell lines using the salt-extractable lipid raft method.  The fractions were  subsequently resolved by SDS-PAGE, transferred to nitrocellulose filters and immunoblotted with Iga, IgP, [i heavy chain, K light chain and phospho-Akt specific antibodies. Samples were analyzed for phosphorylated Akt as increased Akt phosphorylation is a well-characterized downstream consequence of B C R cross-linking (Astoul et al., 1999; Craxton et al., 1999; Gold et al., 1999) and thus, is indicates successful receptor cross-linking and signaling.  Treatment of the control cell line, WEHI 231, with biotinylated anti-IgP mAb and streptavidin leads to increased Akt phosphorylation (Fig. 3.4, bottom panel). Therefore, this treatment is sufficient to induce BCR-like signaling and likely reflects successful cross-linking of the B C R through IgP. Following Igp-based receptor cross-linking, the levels of Iga, IgP and the u heavy chain appear to slightly decrease in the supernatant fraction while they coincidentally increase in the lipid raft fraction (Fig. 3.4; comparing the left side of image to the right side). While this trend is comparable to that seen following u heavy chain-based receptor cross-linking it should be noted that the relative decreases and increases of protein levels in the respective fractions do not appear to be as significant with the former method as compared to the latter (IgP versus u heavy chain-based receptor cross-linking) (compare Figs.  3.4 and 3.2, respectively). These  differences may reflect differences in the IgP and u heavy chain specific antibodies' abilities to efficiently bind to and cross-link the receptor (refer to discussion).  Regardless, IgP-based  receptor cross-linking appears sufficient to cross-link the receptor and to induce both BCR-like signaling and B C R translocation into lipid rafts in the control cell line, WEHI 231. Therefore, IgP-based cross-linking was then used to investigate whether or not solo Iga/p is able to associate with and translocate into lipid rafts in mlgM deficient cell lines.  88  The mlgM deficient experimental cell line, WEHI 303.1.5, was stimulated by cross-linking the IgP with a biotinylated anti-IgP mAb plus streptavidin for the indicated length of time. Lipid rafts were then collected from the cell lines using the aforementioned salt-extractable lipid raft method (Weintraub et al, 2000). The fractions were subsequently resolved by SDS-PAGE, transferred to nitrocellulose filters and immunoblotted with phospho-Akt, CD45, Iga and IgP specific antibodies. Samples were analyzed for phosphorylated Akt as increased Akt phosphorylation indicates successful B C R cross-linking (Astoul et al., 1999; Craxton et al., 1999; Gold et al, 1999). However, it should be noted that increased Akt phosphorylation would be indicative of successful solo Iga/p cross-linking only if solo Iga/p can in fact induce BCR-like signaling in the absence of mlgM. Fortunately, IgP-based cross-linking leads to an increase in Akt phosphorylation in the WEHI 303.1.5 cell line (Fig. 3.5). This suggests that the anti-Igp based method is sufficient to cross-link the solo Iga/p and furthermore, that cross-linking of solo Iga/p is sufficient to induce BCR-like signaling. Interestingly, a small proportion of Iga and Igp is clearly evident within the lipid raft fraction prior to IgP cross-linking (Fig. 3.5, the right side of the image, time zero). Additionally, the levels of Iga and IgP appear to slightly decrease in the supernatant fraction while they coincidentally increase in the lipid raft fraction following IgP cross-linking (Fig. 3.5, comparing the left side of the image to the right side of the image). Thus, it appears that a portion of solo Iga/p is able to constitutively associate with lipid rafts and that a further portion of solo Iga/p is able to translocate into lipid rafts following IgP cross-linking in the WEHI 303.1.5 cell line.  Finally, solo Iga/p lipid raft association was investigated in the K40B-1 cell line to ensure that the ability of Iga/p to associate with lipid rafts in the absence of mlgM is not unique to the WEHI 303.1.5. The K40B-1 cell line is a pro-B-like cell line that expresses low levels of solo Iga/p on its cell surface in association with calnexin (Fig. 3.1). The solo Iga/p of the K40B-1 cell line may function as a pro-B cell receptor and its ability to associate with lipid rafts may reflect its ability to function as a receptor. As a control, Iga/p lipid raft association was also investigated in the K40B-2 cell line which is the developmental successor of the K40B-1 cell that expresses low levels of the intact pre-BCR on its cell surface. Unfortunately, the relatively low expression levels of Iga and IgP within the KB40 cell lines (refer to Fig. 3.1) made it unfeasible to assess Iga/p lipid raft association with the sucrose density gradient method. Rather, Iga/p lipid raft association within the K40B cell lines was assessed using only the salt-extraction method (Weintraub et al., 2000) as this method allows for greater recovery of the various 89  fractions. Thus, the K40B cell lines were stimulated by cross-linking the IgP with a biotinylated anti-IgP mAb plus streptavidin for the indicated length of time. Lipid rafts were then collected from the cell lines using the salt-extractable lipid raft method (Weintraub et al., 2000) and analyzed as before. As with the WEHI cell lines, Ig/p cross-linking was successful as evidenced by an increase in Akt phosphorylation following treatment with biotinylated anti-IgP mAb plus streptavidin (Fig. 3.6 and 3.7 bottom panel).  Additionally, it can be seen that, despite their  relatively low expression levels, a portion of both Iga and IgP is constitutively associated with lipid raft fraction in the K40B cell lines (Fig. 3.6 and 3.7).  However, this portion does not  appear to significantly increase following IgP cross-linking in the K40B-1 cell line (Fig. 6a). In contrast, this portion appears to increase ever-so-slightly in the K40B2 cell line following Iga/p cross-linking (Fig. 3.7).  These findings may suggest that solo Iga/p does not efficiently  translocate into lipid rafts in the K40B cell lines. Alternatively, it may simply be difficult to elicit or detect lipid raft translocation of the solo Iga/p in the K40B cell lines due to the low expression levels of Iga/p. Nonetheless, it is clearly evident that solo Iga/p is able to associate with lipid rafts in the K40B cell lines and therefore, such associations are not unique to the WEHI 303.1.5 cell line.  90  WEHI 231 Cell Line supernatant  | 0 Iga  I  5  B  lipid raft  10 15  I  0 5  I  10  -  37 kD  -  25 kD  -  50 kD  _  37  -  80 kD  -  173kD  SB iiii  ~C llli u heavy chain  15  * • .•-<#  •*  kD  CD45 P-Akt  -  64 kD  Figure 3.4. The BCR translocates into a detergent-insoluble, salt-extractable lipid raft fraction in the WEHI 231 immature B cell line following IgP cross-linking. 25 x 10 cells were stimulated with 60 ug of biotinylated 6  anti-IgP antibody along with 80 ug of streptavidin for the indicated time points. The detergent soluble fraction (supernatant) and the detergent-insoluble, salt-extractable fraction (lipid raft) were then collected. Equivalent amounts of protein for each fraction were resolved by SDS-PAGE. Following electrophoresis the gels were transferred to nitrocellulose filters that were subsequently immunoblotted with Iga, IgP, u chain (component of mlgM) or CD45 specific antibodies (as indicated to the left of each panel). Fractions were analyzed for CD45 content as a control as CD45 has been shown to be excluded from B Cell lipid raft fractions. Successful receptor cross-linking was confirmed by monitoring for increased Akt phosphorylation (bottom panel). Data are representative of three independent experiments.  01  WEHI 303.1 Cell Line supernatant Bio Anti Igp/Strep (min)  0  lipid raft  5 10 15  0 5  10 15 -  Iga  -f~^  M A Mm  37 kD  i _  t flail  25 kD  50 kD ' 37 kD  CD45 173 kD  P-Akt  -  64 kD  Figure 3.5. A portion of the solo Iga/IgP heterodimer localizes to detergent-insoluble, salt-extractable lipid raft fraction in the mutant W E H I 303.1.5 cell line (mlgM negative). A further portion of the solo Iga/p heterodimer appears to translocate into the detergent-insoluble, salt-extractable lipid raft fraction in the mutant WEHI 303.1.5 cell line (mlgM negative) upon Igp cross-linking. 25 x 10' cells were stimulated with 60 ug of biotinylated anti-IgP antibody along with 80 ug of streptavidin for the indicated time points. The detergent soluble fraction (supernatant) and the detergent-insoluble, salt-extractable fraction (lipid raft) were then collected. Equivalent amounts of protein for each fraction were resolved by S D S - P A G E . Following electrophoresis the gels were transferred to nitrocellulose filters that were subsequently immunoblotted with Iga, IgP, or CD45 specific antibodies (as indicated to the left of each panel). Fractions were analyzed for CD45 content as a control as CD45 has been shown to be excluded from B Cell lipid raft fractions. Successful receptor cross-linking was confirmed by monitoring for increased Akt phosphorylation (bottom panel). Data are representative of three independent experiments. (  92  K40B-1 Cell Line supernatant  - -- -  \o Iga  igp  P-Akt  salt fraction  -C  5  15  30 | 0  5  —  -  15 30  -  25 kD  -  50 kD  _  37 kD  64 kD  Figure 3.6. A portion of the solo Iga/IgP heterodimer localizes to the detergent-insoluble, salt-extractable lipid raft fraction in the K40B-1 pro-B-like cell line (mlgM negative). 25 x 10 cells were stimulated with 60 ug of biotinylated anti-IgP antibody along with 80 p,g of streptavidin for the indicated time points. The detergent soluble fraction (supernatant) and the detergent-insoluble, salt-extractable fraction (lipid raft) were then collected. Equivalent amounts of protein for each fraction were resolved by S D S - P A G E . Following electrophoresis the gels were transferred to nitrocellulose filters that were subsequently immunoblotted with Iga or Igp (as indicated to the left of each panel). Successful receptor cross-linking was confirmed by monitoring for increased Akt phosphorylation (bottom panel). Data are representative of three independent experiments using similar time courses. 6  93  K40B-2 Cell Line supernatant Bio Anti Igp/Strep (min)  Iga  0  5  salt fraction  15 30 | 0  5  15  30  — 25 kD  igp  -  ~C  50 kD 1  u heavy chain  -  P-Akt  -  37 kD  80 kD  64 kD  Figure 3.7. A portion of the Iga/IgP heterodimer localizes to the detergent-insoluble, salt-extractable lipid raft fraction in the K40B-2 pre-B-like cell line (mlgM positive). 25 x 10 cells were stimulated with 60 ug o f biotinylated anti-Igp antibody along with 80 (.ig o f streptavidin for the indicated time points. The detergent soluble fraction (supernatant) and the detergent-insoluble, salt-extractable fraction (lipid raft) were then collected. Equivalent amounts o f protein for each fraction were resolved by S D S - P A G E . F o l l o w i n g electrophoresis the gels were transferred to nitrocellulose filters that were subsequently immunoblotted with Iga, IgP, or |a heavy chain specific antibodies (as indicated to the left o f each panel). Successful receptor cross-linking was confirmed by monitoring for increased A k t phosphorylation (bottom panel). Data are representative o f three independent experiments using similar time courses. 6  94  3.3  Discussion  As mentioned in the introduction, previous studies have shown that the B C R translocates to lipid rafts following B C R engagement (Cheng et al, 2001; Petrie et al., 2000; Weintraub et al, 2000; Cheng et al., 1999). Furthermore, these studies suggest that such translocation is necessary for appropriate B C R signaling (Su et al, 2002; Cheng et al, 2001; Aman and Ravichandran, 2000; Guo et al, 2000; Petrie et al, 2000). Thus, it appears that a comprehensive understanding of B C R signaling will require a further understanding of BCR lipid raft translocation. Accordingly, this chapter of my thesis focused on investigating the molecular mechanisms that govern BCR translocation into lipid rafts.  It is clear that B C R lipid raft association is enhanced following B C R engagement (Cheng et al, 2001; Petrie et al, 2000; Weintraub et al, 2000; Cheng et al, 1999). However, it is unclear whether this enhanced association is a consequence of B C R signaling itself (e.g., induced association with raft associated proteins), a consequence of a structural change within the engaged B C R or a consequence of some combination of structural and signaling events initiated by B C R engagement. It seems unlikely that enhanced lipid raft association is mediated strictly by B C R signaling as the B C R has been found to translocate into lipid rafts in the absence of B C R signaling (Cheng et al, 2001). Thus, we are left to consider the alternative, that enhanced lipid raft association is mediated by a structural change within the engaged BCR. As mentioned above, such structural changes may involve exposing previously hidden "lipid raft affinity domains" (LRADs) within the B C R or likewise may involve the assembling of LRADs via oligomerization of the B C R . Thus, the initial hypothesis was that the B C R itself contains structural information of some form (termed here as LRADs) that, upon receptor engagement, directly promotes its translocation to lipid rafts.  As previous studies have demonstrated that the solo antigen-binding subunit of the B C R (mlgM) translocates into lipid rafts (Cheng et al, 2001) it was of interest to determine if the solo Iga/(3 subunit could likewise translocate into lipid rafts. This investigation would help to further define structural regions within the subunits (Iga/p and mlgM) that may contain the proposed LRADs. Furthermore, this investigation could help to shed light on BCR-like signaling processes in pro-B cells which express solo Iga/p on its surface in association with calnexin (Nagata et al, 1997). 95  Lipid raft association and translocation of solo Iga/(3 was investigated in the mutant, mlgMdeficient immature B cell line, WEHI 303.1.5.  However, before proceeding with these  investigations, a control investigation was performed to confirm that the intact BCR does indeed associate with and translocate into lipid rafts in wild-type immature B cell lines. This was necessary as Susan Pierce's group has reported that such translocation may not occur (Sproul et al., 2000; Chung et al., 2001). If this were to be the case our experimental system would be inappropriate for investigating B C R lipid raft translocation. Fortunately, using the Weintraub salt-based lipid raft isolation method, the intact B C R was found to associate with and translocate into lipid rafts in the wild-type immature B cell line, WEHI 231 (Fig. 3.2; published in Jackson et al., 2005). Due to the controversial nature of this finding, B C R lipid raft association and translocation was further investigated using a second immature B cell line (CH31) and a second lipid raft extraction method (sucrose density gradient method performed by Lorna Santos). The findings from these secondary approaches confirmed that the B C R does in fact translocate into lipid rafts in at least two immature B cell lines following B C R cross-linking (Fig. 3.3 and data not shown; published in Jackson et al, 2005). The reported differences between the Pierce Lab's findings (Sproul et al., 2000; Cheng et al., 2001) and the Matsuuchi Lab's findings may reflect differences in lipid raft preparations, subtle differences in the cell lines used and/or differences in the sensitivity of the methods used to detect the B C R within the lipid raft  fractions.  Interestingly, while the Pierce lab reports that "B cell antigen receptor signaling occurs outside lipid rafts in immature B cells" based on their inability to detect B C R lipid raft association, they did not report performing the loss-of-function study whereby they disrupt the lipid rafts with chemical agents to demonstrate that lipid rafts are not required for B C R signaling in immature B cell lines (Sproul et al, 2000; Cheng et al, 2001). Indeed, B C R association and/or lipid raft translocation, while being difficult to detect, may yet contribute to B C R signaling in immature B cells. Regardless, the ability to detect B C R lipid raft association and translocation within wildtype immature B cell lines within the Matsuuchi Lab enabled us to further investigate solo Iga/p lipid raft association and translocation within the mutant, immature B cell line, WEHI 303.1.5.  To investigate solo Iga/p lipid raft translocation it was first necessary to establish a protocol for cross-linking the solo Iga/p. To this end, the WEHI 231 cell line was treated with a biotinylated IgP specific antibody plus streptavidin. IgP-based cross-linking appears to induce B C R lipid raft translocation in the WEHI 231 immature B cell line (Fig. 3.4). However, this translocation does 96  not appear as efficient as that induced by mlgM-based cross-linking (compare Fig. 3.4 to 3.2). The apparent differences in translocation efficiency may reflect differences in the antibodies' abilities to cross-link the receptor.  In particular, the mlgM specific antibody is a polyclonal  antibody that can bind to multiple sites  within the mlgM  to induce B C R cross-  linking/oligomerization whereas the Igp antibody is a monoclonal antibody that can only bind one specific site within the IgP to induce B C R cross-linking/oligomerization. Regardless of the apparent differences in efficiency, treatment with a biotinylated IgP specific antibody plus streptavidin appears sufficient to cross-link the B C R and as such this treatment was used to cross-link the solo Iga/p in subsequent investigations.  Solo Iga/p lipid raft association was investigated in the mlgM-deficient, mutant immature B cell line, WEHI 303.1.5.  These investigations found that solo Iga/p can associate with and  translocate into lipid rafts.  This suggests that, similar to solo mlgM, solo Iga/p contains  structural information that is sufficient to mediate its association with lipid rafts. Hypothesizing that the two subunits employ similar mechanisms to associate with lipid rafts, it may be predicted that they share common structural features that mediate this association. Given the structures of mlgM and the Iga/p such a common feature would most likely lie within their transmembrane or membrane proximal domains. Thus, the Matsuuchi Lab is performing numerous mutational analyses within these particular regions to further define any structural features that may help govern B C R lipid raft association and translocation.  Second, the finding that solo Iga/p can translocate into lipid rafts and induce BCR-like signaling may be physiologically relevant. In particular, the putative pro-B cell receptor is expressed on the surface of pro-B cells as a solo Iga/p in association with calnexin (Nagata et al., 1997). While this receptor has been proposed to be involved in development-dependent signaling it has been unclear whether such signaling could occur in the absence of an antigen-binding subunit. The findings here suggest that such signaling could indeed occur, and may occur from within lipid rafts, providing that there is a physiological ligand capable of cross-linking the solo Iga/p. While such a ligand has yet to be identified, it has been speculated that the ligand may have lectin-like properties such that it is able to bind and cross-link the putative pro-B cell receptor via the carbohydrate chains of the Iga/p heterodimer.  97  To ensure that these findings are not a unique phenomenon of the mutant WEHI 303.1 cell line, similar investigations were performed with the naturally mlgM-deficient K40B-1 pro-B cell line and the mlgM positive K40B-2 control cell line.  Similar to the WEHI 303.1.5 cell line, a  significant portion of solo Iga/p was found to be constitutively associated with lipid rafts in the K40B-1 cell line (Fig. 3.6; published in Jackson et al., 2005). However, unlike in the WEHI 303.1.5 cell line, the Iga and Igp do not appear to translocate into the lipid rafts following IgPbased cross-linking (Fig. 3.5; published in Jackson et al, 2005). This finding may indicate that solo Iga/p does not translocate into lipid rafts upon cross-linking in the K40B-1 cell line. Alternatively, it may also reflect an inability to induce or detect such translocation with the approaches used.  As noted above, cross-linking of IgP may not be ideally efficient.  Additionally, the K40B-1 cell line expresses very low levels of Iga/p (Fig. 3.1). Together, these two factors may make it difficult to detect inducible lipid raft translocation of the solo Iga/p in the K40B-1 cell line. This would be especially true if the translocating Iga/p represented a very small fraction of the total Iga/p (as was seen with the WEHI 303.1.5 cell line). Nonetheless, the fact that a portion of solo Iga/p is constitutively associated with the lipid rafts in the K40B-1 cell line indicates that this association may occur in wild-type cell lines and is not a phenomenon unique to the mutant WEHI 303.1.5 immature B cell line. Moreover, cross-linking of solo Iga/p was found to induce BCR-like signaling in the K40B-1 cell line (Fig. 3.6; published in Jackson et al., 2005), further supporting the notion that solo Iga/p may indeed be able to act as a signaling receptor to mediate pro-B cell development.  Given these findings several future investigations are suggested.  First, mutational analysis  should be performed on the transmembrane and membrane proximal domains of the mlgM subunit and the Iga/p. Such analysis may help to identify structures within the subunits that enable them to associate with lipid raft (i.e., identify the proposed L R A D ) . These studies are currently being pursued within the Matsuuchi Lab by Steven Machtaler.  Second, further  investigations should be performed to determine whether or not lipid raft translocation is necessary for solo Iga/p signaling. In particular, loss of function studies could be performed using cholesterol sequestering agents to disrupt lipid rafts. Such studies would help to further determine if solo Iga/p lipid raft translocation is physiologically relevant and to further define the role of lipid raft translocation in BCR and BCR-like signaling.  98  CHAPTER 4  Co-Expression of the BCR, Syk, and BLNK Is Sufficient to Reconstitute BCR-Induced PLCy Activation in AtT20-Derived Cell Lines  4.1  Introduction  The ability to recruit and/or sequester proteins within defined cellular locations is essential to intracellular signaling.  Such compartmentalization  enables the  efficient  assembly of  macromolecular signaling complexes and aids in the regulation of cross-talk between various signaling pathways. Adapter proteins represent one mechanism by which compartmentalization can be achieved.  These proteins are typically non-enzymatic proteins that contain multiple  protein interaction domains that enable specific and multiple protein associations (detailed in Chapter 1.5 and 1.5.1). Depending on the nature of the protein interaction domains, the resulting protein associations may be constitutive or inducible.  Thus, similar to lipid rafts, adapter  proteins are postulated to serve as platforms upon which macromolecular signaling complexes can be specifically assembled and regulated.  Within the past decade the B cell linker protein (BLNK) has been identified as a key adapter protein in B cell signaling (Fu et al., 1997; Fu et al, 1998; Goitsuka et al., 1998; Wienands et al., 1998; Ishiai et al, 1999a; Ishiai et al, 1999b; reviewed in Kurosaki, 2000; Gold, 2002). In particular, B L N K has been suggested to play a pivotal role in coupling the engaged B C R to PLCy activation (Fu et al., 1997; Fu et al, 1998; Goitsuka et al, 1998; Wienands et al, 1998; Ishiai  et al,  1999a; Ishiai  et al,  1999b;  reviewed in Kurosaki, 2000; Gold, 2002).  Understanding how the B C R is coupled to PLCy activation is important as defects in this pathway have been shown to contribute to several immune system disorders including autosomal recessive agammaglobulinemia ( B L N K defects) and X-linked agammaglobulinemia (BTK defects) which render sufferers unable to mount an effective humoral response against pathogens (reviewed in Fischer, 2004 and the online Immunodeficiency Resource at http://bioinf.uta.fi/idr/).  Our current understanding of how the B C R is coupled to PLCy activation is predominantly based on loss-of-function studies (detailed in Chapter 1.5.3). These studies have clearly indicated that Syk is necessary to couple the B C R to PLCy activation and that B T K and B L N K are necessary 99  for this process to occur efficiently. Given the current loss-of-function data, prevailing models of the BCR/PLCy signaling pathway envision that B L N K is recruited to the cross-linked B C R at the plasma membrane where it is phosphorylated by BCR-associated Syk. Once phosphorylated, B L N K is proposed to bind to both B T K and PLCy via their respective SH2 domains. This process is thought to facilitate the co-localization of Syk, B T K and PLCy such that Syk and B T K can phosphorylate and activate PLCy.  While these models (summarized in Fig. 4.1) are  suggestive, many details of the pathway remain to be elucidated and/or confirmed.  Most  significantly, it remains to be determined whether Syk, B T K and B L N K are sufficient to couple the engaged B C R to PLCy activation or whether additional lymphoid specific components are required to mediate this signaling pathway.  Thus, I have used a reconstitution approach to  investigate the molecular requirements and mechanisms underlying BCR-induced PLCy activation.  As discussed in Chapter 1.10, the AtT20 reconstitution system is derived from a non-lymphoid, endocrine cell line that endogenously expresses the PI3K, Ras, PLCy, and Fyn signaling enzymes (Matsuuchi et al, 1992 and Richards et al., 1996). While these enzymes obviously function in non-lymphoid signaling pathways within the AtT20 system they have also been identified as key downstream signaling components in B C R signaling pathways. This, along with the fact that the AtT20 cell line is amenable to transfection, enables it to be used as a reconstitution system into which lymphoid specific components can be transfected in an effort to determine the sufficiency of these components to reconstitute BCR-induced activation of the various downstream signaling enzymes.  AtT20 cells have previously been transfected with and shown to express the intact B C R on their cell surface (Matsuuchi et al., 1992). Cross-linking the transfected BCRs has proven sufficient to induce some aspects of B C R signaling within these cells including phosphorylation of the Iga/p heterodimer and phosphorylation and activation of PI3K (Matsuuchi et al., 1992). However, expression of the transfected BCRs alone has proved insufficient to reconstitute a robust tyrosine phosphorylation cascade or PLCy activation in these cells (Matsuuchi et al., 1992). This situation was partially rectified by further transfecting the cells with the lymphoid specific protein tyrosine kinase, Syk (Richards et al, 1996). Co-expression of the B C R and Syk proved sufficient to reconstitute a robust BCR-induced protein tyrosine phosphorylation cascade as well as to reconstitute BCR-induced Erk activation (Richards et al., 1996). 100  Still, co-  expression of the B C R and Syk proved insufficient to reconstitute BCR-induced PLCy activation in the transfected cells suggesting that additional lymphoid specific components may be required to couple the B C R to PLCy (Richards et al., 1996). These findings are in agreement with lossof-function studies which, as mentioned above, suggest that B L N K and B T K may be necessary to effectively couple the B C R to PLCy activation. Thus, I further transfected the B C R and BCR/Syk AtT20-derived cell lines with B T K and/or B L N K to determine if these components are sufficient to reconstitute BCR-induced PLCy activation within this system.  4.2  4.2.1  Results  Expression of the B C R , Syk, B T K and B L N K within AtT20-Derived Cell Lines.  A reconstitution system expressing transfected lymphoid specific components was established to investigate the molecular mechanism underlying BCR-induced PLCy activation.  Initial  transfections were performed into two previously established AtT20-derived cell lines that will be termed the " B C R " and "BCR/Syk" cell lines throughout this thesis.  The B C R cell line  expresses transfected BCRs on its cell surface (Matsuuchi et al., 1992; previously referred to as 100.33) while the BCR/Syk cell line expresses transfected BCRs on its cell surface and transfected Syk within its cytoplasm (Richards et al., 1996; previously referred to as Sykl3). For the purpose of this thesis, the B C R and BCR/Syk cell lines were further transfected with a plasmid encoding for a drug resistance marker along with a plasmid encoding for myc-tagged human B L N K and/or a plasmid encoding for human B T K (refer to Chapter 2.1.2 and 2.1.3 for details). Following the transfections, drug-resistant clones were isolated and screened for their ability to express the desired proteins (detailed in Chapter 2..3.3 and 2.3.4). From this process multiple AtT20-derived cell  lines were established including B C R / B L N K ,  BCR/BTK,  B C R / B L N K / B T K , BCR/Syk/BLNK, BCR/Syk/BTK and B C R / S y k / B L N K / B T K , respectively. As may be anticipated, the names of the various cell lines reflect the transfected proteins that they have been demonstrated to successfully express (Figure 4.2). It should be noted that several clones were obtained for each described cell line and that initial experiments were performed with multiple clones (data not shown). As the results from the various clones were consistent, and for the sake of clarity, only one clone has been shown for each described cell line (Figure 4.2).  101  Abbreviations  Interaction Arrows  Legend /  /  Kinase  \  \  Phosphatase  -©—>  Positive Regulation  •> C a R e l e a s e  B cell antigen receptor B cell linker protein Bruton's Tyrosine Kinase calcium calmodulin calcenurin diacylglycerol inositol trisphosphate I P receptor/Ca * channel nuclear factor of activated T cells phospholipase C g a m m a  2  Adaptor  ©  BCR BLNK BTK Ca * CaM Cn DAG IP IP3R NFAT 3  3  PLCy  Phosphorylation  Figure 4.1. Basic Overview of the B C R / P L C y Signaling Pathway. Cross-linking of the B C R initiates the activation of several PTKs (e.g., Src family members, Syk and B T K ) that coordinate the activation of the Ras/MAPK, PI3K and PLCy signaling pathways. Following B C R cross-linking Iga/p becomes phosphorylated. B L N K is then proposed to be recruited to phosphorylated Iga via its SH2 domain. B L N K is then phosphorylated by Syk such that it can then associate with B T K and PLCy via their respective SH2 domains. Thus, B L N K acts as an adapter protein facilitating the co-localization of B T K , Syk and PLCy such that the PTKs can efficiently phosphorylate PLCy. Once activated, PLCy hydrolyzes PIP to D A G and IP . IP then binds to IP R causing the release of intracellular calcium stores. Calcium and D A G then contribute to the activation of several enzymes (including P K C and calcineurin) which in turn influence the status of numerous transcription factors which ultimately assist in the regulation of the differentiation and proliferation of the B cell. Additionally, PLCy has been shown to activate the M A P K s , Erk 1/2 via the Ras pathway (refer to Fig. 4.3). Adapted from "AFCS Nature, The Signaling Gateway" (online at http://www.sianalinjj_gatewav.orK/molecule/maQs/bcr.html). 2  102  3  3  3  CO  z-I j*± _zl  h-  ffl * I— 0£  O CO  oo o CO  Z -I  00  00  o  o  00  00  00  CO  CO CQ  s Ss  >« > > w w w  S S £  O O o CQ  00  CO  o  CO  myc-tagged B L N K  80 kD  BTK  —  - 80 kD  —  Syk  64 kD  ^  u heavy chain  80 kD  25 kD  endogenous P L C y l  - 173 kD  IB: B L N K (H80)  IB: B T K ( C 2 0 ) IB: Syk  IB: u heavy chain  IB: a  |B: PLCyl  (1249)  Figure 4.2. Characterization of a, u, Syk, B L N K , B T K and P L C y l Expression in Transfected AtT20 Cell Lines. 35 ug of whole cell lysate were resolved by SDS-PAGE. Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with antibodies specific for a, u, B L N K , B T K , Syk and P L C y l . Note that a is a component of the Iga/p signaling subunit of the B C R whereas u is a component of the mlgM antigen-binding subunit of the BCR. IB indicates the immunoblotting antibody.  103  4.2.2  Co-Expression of the BCR, BLNK and/or BTK Is Not Sufficient to Reconstitute  BCR-Induced Erk Phosphorylation in AtT20-Derived Cell Lines.  Having established a reconstitution system that expresses various combinations of the desired proteins, it was next necessary to establish a method to assay for BCR-induced PLCy activation. While membrane-recruitment  and tyrosine phosphorylation of PLCy are suggestive of its  activation these events have yet to be established as hallmarks of PLCy activation. Thus, PLCy activation is best assessed by monitoring downstream targets and/or consequences of PLCy activation. Initially, it was proposed that PLCy activation could be assayed by monitoring for increased IP3 production and induced calcium fluxes as these are well-known downstream consequences of PLCy activation (refer to Chapter 1.5.3 and Figure 4.1). Unfortunately, these assays proved inconclusive in the AtT20 system (summarized in Appendix II and Appendix III, respectively).  Alternatively, BCR-induced PLCy activation was assessed by monitoring for  BCR-induced Erk phosphorylation as increased Erk phosphorylation has been demonstrated to be a downstream consequence of BCR-induced PLCy activation (refer to Chapter 1.5.3 and 1.5.4). However, the B C R can induce Erk phosphorylation both via PLC-dependent and P L C independent pathways (reviewed in Gold, 2002; refer to Fig. 4.3).  Thus, a PLC inhibitor  (U73122) was used to distinguish between PLC-dependent and PLC-independent changes in BCR-induced Erk phosphorylation.  According to previous studies and models, co-expression of the B C R along with B L N K and/or B T K should not be sufficient to reconstitute BCR-induced Erk phosphorylation within the AtT20 system.  Rather, Syk is proposed to be necessary to initiate most B C R signaling pathways  including both the PLC-independent and the PLC-dependent pathways that contribute to Erk phosphorylation. Nonetheless, it is possible that Syk and B T K may perform partially redundant functions or that some unidentified AtT20-specific PTK may exist that, along with B L N K , could initiate BCR-induced PLCy activation within this system. Thus, to address these possibilities, the sufficiency of B L N K and/or B T K , to reconstitute BCR-induced Erk phosphorylation was investigated first.  As can be seen in Figure 4.4, no apparent increase in E R K phosphorylation is found following B C R cross-linking in AtT20-derived cell lines that co-express the B C R along with B L N K and/or B T K (Figure 4.4). This suggests that co-expression of the B C R along with B L N K and/or B T K 104  4>—• / B T K /  AlEK/  AlEK/  /ErkT/2X PLC-independent Pathway  PLC-Dependent Pathway  /  /  Kinase  \  \  Phosphatase  -©—>  \  /  <\___7 o  Positive Regulation  Positive Regulation Suspected  | Adaptor  I  Key Abbreviations  Interaction Arrows  Legend  BCR BLNK BTK Erk  B cell antigen receptor B cell linker protein Bruton's Tyrosine Kinase Extracellular regulated kinase  GEF GTPase  ..>  Translocation  Phosphorylation  Figure 4.3. Basic Overview of BCR-Induced Erk Phosphorylation via the PLCy-Independent and PLCyDependent Signaling Pathways. Note that B L N K can associate with Grb2 of the PLC-independent pathway (Fu et al., 1998). While this is believed to be a positive association the physiological significance remains to be determined. Furthermore, BLNK-associated Grb2 does not appear to associate with She (Fu et al., 1998) as would be expected for the PLC-independent pathway. Reviewed in Gold (2002). Adapted from "AFCS Nature, The Signaling Gateway" (online at http://www.signaline_gateway.org/molecule/maps/bcr.html). 105  BCR Anti m l g M (min) - » |  0  3  5  10  BCR/BLNK 15  30  |  0  3  5  10  15  30  50 kD -  IB: ( P ) E r k 37 kD -  50 kD •  S & R : Erk 37 kD •  BCR/BTK Anti m l g M (min) - • ~ |6  3  5  10  15  BCR/BLNK/BTK 30  | 0  50 kD -  3  5  10  15  30  |  IB: ( P ) E r k  37 kD -  50 kD -  S & R : Erk  37 kD -  Figure 4.4 Co-Expression of the B C R , B L N K and/or B T K Is Not Sufficient to Reconstitute BCR-Induced Erk Phosphorylation in AtT20-Derived Cell Lines. The B C R was cross-linked with anti mlgM antibodies at 37 °C for the indicated length of time. Cells were then lysed and 50 (ag of whole cell lysate were resolved by SDS-PAGE. Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-Erk (Thr202/Tyr204) specific antibody. Finally, filters were stripped and reprobed with an Erk specific antibody. IB indicates the immunoblotting antibody while S & R indicates the antibody used to reprobe the filter. Data are representative of three independent experiments. Note that these experiments were performed concurrently with those shown in Figures 4.5 and 4.6. Thus, the experiments shown in Figure 4.5 and 4.6 serve as positive controls that demonstrate that Erk phosphorylation is indeed detectable in the reconstitution system  106  is not sufficient to reconstitute BCR-induced Erk phosphorylation in the AtT20-derived system. This in turn suggests that co-expression of these components is not sufficient to reconstitute BCR-induced PLCy activation.  As mentioned above, this result is as expected as Syk, the  putatively essential protein tyrosine kinase, is absent in these cell lines (refer to discussion).  4.2.3  Co-Expression of the BCR and Syk Is Sufficient to Reconstitute BCR-Induced,  PLC-Independent Erk Phosphorylation in AtT20-Derived Cell Lines.  As previous studies and models suggest that Syk is essential to initiate most B C R signaling pathways, the sufficiency of Syk to reconstitute BCR-induced Erk phosphorylation was investigated next. As can be seen in Figure 4.5, an increase in E R K phosphorylation is evident following B C R cross-linking in AtT20-derived cell lines that co-express the B C R and Syk. The increase in E R K phosphorylation is evident within 3 minutes of B C R engagement and appears to be sustained for at least 15 minutes before beginning to decrease by 30 minutes (Fig. 4.5 a and b). This increase in Erk phosphorylation suggests that co-expression of the B C R and Syk is sufficient to reconstitute BCR-induced Erk phosphorylation in the AtT20-derived system. The PLC inhibitor, U73122, was then used to determine whether or not the Erk phosphorylation was being mediated by a PLC-dependent and/or a PLC-independent pathway.  If the Erk  phosphorylation is being mediated by a PLC-independent pathway one would predict that addition of the inhibitor should not affect BCR-induced Erk phosphorylation. In contrast, if the Erk phosphorylation is being mediated by a PLC-dependent pathway one would predict that addition of the inhibitor will repress Erk phosphorylation. In this case, it was found that addition of the inhibitor does not appear to significantly repress BCR-induced Erk phosphorylation in the BCR/Syk cell line (Figure 4.5 a and b). This suggests that co-expression of the BCR and Syk is sufficient to reconstitute BCR-induced, PLC-independent Erk phosphorylation in the AtT20derived system. Furthermore, these findings suggest that co-expression of the B C R and Syk is not sufficient to reconstitute BCR-induced P L C y l activation within this system as assayed by Erk phosphorylation. Again, these finding are not unexpected as the prevailing models suggest that B L N K and/or B T K are required, along with Syk, to couple the B C R to PLCy. Thus, the sufficiency of Syk, B L N K and/or B T K to reconstitute BCR-induced Erk phosphorylation in AtT20-derived cell lines was investigated next.  107  BCR/Syk P L C Inhibitor  DMSO Anti mlgM (min)-» | 0  3  5  10  15  30 | 0  3  5  10  15  30  |  50 kD -  IB: ( P ) E r k  37 kD 50 kD  S & R : Erk  37 kD •  Figure 4.5. (a) Co-Expression of the B C R and Syk Is Sufficient to Reconstitute PLC-Independent, B C R Induced E r k Phosphorylation in AtT20-Derived Cell Lines. Cells were treated either with D M S O (negative control) or the PLC inhibitor, U73122 (in D M S O , at a final concentration of 10 uM) for 1 hour. The BCR was then cross-linked with anti mlgM antibodies at 37 °C for the indicated length of time. Cells were then lysed and 50 ug of whole cell lysate were resolved by SDS-PAGE. Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-Erk (Thr202/Tyr204) specific antibody. Finally, filters were stripped and reprobed with an Erk specific antibody. IB indicates the immunoblotting antibody while S&R indicates the antibody used to reprobe the filter. Data are representative of at least four independent experiments.  108  BCR-Induced Erk (p44) Phosphorylation in the BCR/Syk Cell Line 40 35 •|  30  Ic  25  •5 x  20  a.  15  g  10  EI Non-linhibited • Inhibited  5  S  0  0 min  3 min  5 min  10 min  15 min  30 min  Time  BCR-Induced Erk (p42) Phosphorylation in the BCR/Syk Cell Line 70  Z  6 0  _  50  £  40  ._  30  c  20  5  10  o  ra  1  B Non-Inhibited • Inhibited  0 min  3 min  5 min  10 min  15 min  30 min  Time  Figure 4.5. (b) Comparison of Mean Pixel Intensity of the Phosphorylated E r k "Bands" in the BCR/Syk Cell Line in Non-Inhibited and PLC-Inhibited Co-Expression Samples. To calculate the mean pixel intensity of the phosphorylated Erk "band" it was first necessary to define the "band". This was done by visually identifying the largest band in the immunoblot and then manually selecting a square area around that band that captured all the pixels of the band yet minimized the capture of background pixels. An equivalent square area was then selected around the remaining bands again ensuring that all the pixels of the band were captured. The mean pixel intensity of the "bands" was then calculated using Matlab (refer to Chapter 2.15 for further details). By this method the mean pixel intensity should correlate to the band size and intensity such that the larger the mean pixel intensity the larger and more intense the band. The top graph considers the band intensities of the top phosphorylated Erk band (p44) and the bottom graph considers the band intensities of the bottom phosphorylated Erk band (p42).  109  4.2.4  Co-Expression of the BCR, Syk and BLNK Is Sufficient to Reconstitute BCR-  Induced, PLC-Dependent Erk Phosphorylation in AtT20-Derived Cell Lines.  Existing models suggest that B L N K is required to couple the B C R to PLCy activation. Thus, coexpression of the BCR, Syk and B L N K may be predicted to reconstitute BCR-induced PLCy activation and therefore, enhance BCR-induced Erk phosphorylation in this system. However, as can be seen in figure 4.6a, BCR-induced E R K phosphorylation does not appear to be enhanced in the B C R / S y k / B L N K cell line as compared to the BCR/Syk cell line. If anything, B C R induced E R K phosphorylation appears to be slightly, yet reproducibly, inhibited in the B C R / S y k / B L N K cell line (Fig. 4.6a). Thus, it appears that co-expression of B L N K may inhibit BCR-induced Erk phosphorylation in the AtT20-derived system.  Given the complexity of the system and the pathways contributing to BCR-induced Erk phosphorylation these results must be interpreted with caution. In fact, it is possible that B L N K may be enhancing BCR-induced Erk phosphorylation by activating the PLC-dependent pathway while at the same time inhibiting BCR-induced Erk phosphorylation by repressing the P L C independent pathway (refer to discussion). If this is the case, BCR-induced increases in Erk phosphorylation may or may not be evident, depending on the relative contributions of the two pathways. Thus, the P L C inhibitor, U73122 was used to clarify whether or not PLC-dependent changes in BCR-induced Erk phosphorylation were occurring in the B C R / S y k / B L N K cell line. As can be seen in Figure 4.6b and 4.6c, U73122, does appear to inhibit BCR-induced Erk phosphorylation. This suggests that co-expression of the BCR, Syk and B L N K may be sufficient to reconstitute BCR-induced, PLC-dependent Erk phosphorylation in the AfT20-derived system. Although these components appear sufficient to reconstitute BCR-induced PLCy activation in the AfT20-system it is unclear whether or not maximal PLCy activity has been achieved. Indeed, as  mentioned  above,  B T K expression  may  be required  to  achieve  maximal  PLCy  phosphorylation and activation. Thus, the effect of B T K expression within this system was investigated next.  It is important to note that B L N K appears to be concomitantly inhibiting PLC-independent Erk phosphorylation while enhancing PLC-dependent Erk phosphorylation.  This is based on the  observations that while co-expression of the BCR, Syk and B L N K appears to enhance PLC-  110  a. BCR/Syk Anti m l g M (min) - •  0  3  5 10 15 30  BCR/Syk/BLNK  10 3  5 10 15 30  50 k D -  IB: ( P ) E r k 37 kD -  50 kD -  S & R : Erk 37 kD -  b.  BCR/Syk/BLNK DMSO Anti mlgM (min) -+  0  3  P L C Inhibitor  I  5 10 15 30 | 0 3  5 10 15 30  50 kD -  IB: ( P ) E r k 37 kD -  50 k D -  S & R : Erk 37 k D -  Figure 4.6. (a) Co-Expression of B L N K , Along with Syk and the B C R , Appears to Slightly Inhibit B C R Induced E r k Phosphorylation in AtT20-Derived Cell Lines. The B C R was cross-linked with anti mlgM antibodies at 37 °C for the indicated length of time. Cells were then lysed and 50 p.g of whole cell lysate were resolved by S D S - P A G E . Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-Erk (Thr202/Tyr204) specific antibody. Finally, filters were stripped and reprobed with an Erk specific antibody. IB indicates the immunoblotting antibody while S&R indicates the antibody used to reprobe the filter. Data are representative of three independent experiments, (b) Co-Expression of B L N K , Along with Syk and the B C R , Is Sufficient to Reconstitute BCR-Induced, PLC-Dependent E r k Phosphorylation in AtT20-Derived Cell Lines. Cells were treated either with D M S O (negative control) or the P L C inhibitor, U73122 (in D M S O , at a final concentration of 10 uM) for 1 hour. The B C R was then cross-linked with anti mlgM antibodies at 37 °C for the indicated length of time. Cells were then lysed and 50 ug of whole cell lysate were resolved by S D S - P A G E . Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-Erk (Thr202/Tyr204) specific antibody. Finally, filters were stripped and reprobed with an Erk specific antibody. IB indicates the immunoblotting antibody while S&R indicates the antibody used to reprobe the filter. Data are representative of three independent experiments.  Ill  BCR-Induced (p44) Phosphorylation in the BCR/Syk/BLNK Cell Line  BCR-Induced (p42) Phosphorylation in the BCR/Syk/BLNK Cell Line  Figure 4.6. (c) Comparison of Mean Pixel Intensity of the Phosphorylated E r k "Bands" in the B C R / S y k / B L N K Cell Line in Non-Inhibited and PLC-Inhibited Co-Expression Samples. To calculate the mean pixel intensity of the phosphorylated Erk "band" it was first necessary to define the "band". This was done by visually identifying the largest band in the immunoblot and then manually selecting a square area around that band that captured all the pixels of the band yet minimized the capture of background pixels. A n equivalent square area was then selected around the remaining bands again ensuring that all the pixels of the band were captured. The mean pixel intensity of the "bands" was then calculated using Matlab (refer to Chapter 2.15 for further details). By this method the mean pixel intensity should correlate to the band size and intensity such that the larger the mean pixel intensity the larger and more intense the band. The top graph considers the band intensities of the top phosphorylated Erk band (p44) and the bottom graph considers the band intensities of the bottom phosphorylated Erk band (p42).  112  dependent Erk phosphorylation (Fig. 4.6b) it also appears to slightly inhibit overall Erk phosphorylation as compared to the BCR/Syk expressing cell line (Fig. 4.6a). This finding was quite unexpected and is discussed further in Chapter 4.3.  4.2.5  Co-Expression of the BCR, Syk and B T K Is Not Sufficient to Reconstitute BCR-  Induced PLC-Dependent Erk Phosphorylation in AtT20-Derived Cell Lines.  As existing models suggest that B T K is required for efficient BCR-induced PLCy activation it may be predicted that co-expression of B T K will enhance BCR-induced, Syk-dependent PLCy activation and therefore BCR-induced Erk phosphorylation within this system. However, as can be seen in figure 4.7a, BCR-induced E R K phosphorylation does not appear to be enhanced in the BCR/Syk/BTK cell line as compared to the BCR/Syk cell line. Rather, BCR-induced E R K phosphorylation appears to be reproducibly inhibited in the BCR/Syk/BTK cell line (Fig. 4.7a). Thus, it appears that co-expression of B T K may inhibit BCR-induced Erk phosphorylation in the AtT20-derived system.  Again, given the complexity of the system and the pathways contributing to BCR-induced Erk phosphorylation these results must be interpreted with caution. It is possible that B T K may be enhancing BCR-induced Erk phosphorylation by activating the PLC-dependent pathway while at the same time inhibiting BCR-induced Erk phosphorylation by repressing the PLC-independent pathway (refer to discussion). Thus, the P L C inhibitor, U73122 was used to clarify whether or not PLC-dependent changes in BCR-induced Erk phosphorylation were occurring in the BCR/Syk/BTK cell line. As can be seen in Figure 4.7b and 4.7c, inhibitor treatment does not appear to affect BCR-induced Erk phosphorylation in the BCR/Syk/BTK cell line. This suggests that co-expression of the B C R , Syk and B T K is not sufficient to reconstitute BCR-induced P L C y l activation in the AtT20 system.  This finding is not surprising given that numerous  studies have suggested that B L N K is required to efficiently couple Syk and B T K to PLCy. In contrast, the finding that B T K expression inhibits PLC-independent Erk phosphorylation within this system was unanticipated and thus, is discussed further in Chapter 4.3.  113  a. BCR/Syk Anti mlgM (min) -  [o  3  5  10  15  BCR/Syk/BTK 30  10  3  5  10  15  30  50 kD -  IB: ( P ) E r k 37 kD -  50 kD -  S & R : Erk  37 kD -  BCR/Syk/BTK  DMSO Anti mlgM (min) -•  P L C Inhibitor  0 3 5 10 15 30 | 0 3 5 10 15 30  50 kD-  IB: ( P ) E r k 37 kD -  50 kD •  S & R : Erk  37 kD •  Figure 4.7. (a) Co-Expression of B T K , Along with Syk and the B C R , Appears to Slightly Inhibit B C R Induced E r k Phosphorylation in AtT20-Derived Cell Lines. The B C R was cross-linked with anti mlgM antibodies at 37 °C for the indicated length of time. Cells were then lysed and 50 ug of whole cell lysate were resolved by S D S - P A G E . Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-Erk (Thr202/Tyr204) specific antibody. Finally, filters were stripped and reprobed with an Erk specific antibody. IB indicates the immunoblotting antibody while S&R indicates the antibody used to reprobe the filter. Data are representative of three independent experiments, (b) Co-Expression of B T K , Along with Syk and the B C R , Is Not Sufficient to Reconstitute BCR-Induced, PLC-Dependent E r k Phosphorylation in AtT20-Derived Cell Lines. Cells were treated either with D M S O (negative control) or the P L C inhibitor, U73122 (in D M S O , at a final concentration of 10 uM) for 1 hour. The B C R was then cross-linked with anti mlgM antibodies at 37 °C for the indicated length of time. Cells were then lysed and 50 ug of whole cell lysate were resolved by S D S - P A G E . Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-Erk (Thr202/Tyr204) specific antibody. Finally, filters were stripped and reprobed with an Erk specific antibody. IB indicates the immunoblotting antibody while S&R indicates the antibody used to reprobe the filter. Data are representative of three independent experiments.  114  BCR-Induced (p44) Phosphorylation in the BCR/Syk/BTK Cell Line  BCR-Induced (p42) Phosphorylation in the BCR/Syk/BTK Cell Line  Figure 4.7. (c) Comparison of Mean Pixel Intensity of the Phosphorylated Erk "Bands" in the BCR/Syk/BTK Cell Line in Non-Inhibited and PLC-Inhibited Co-Expression Samples. To calculate the mean pixel intensity of the phosphorylated Erk "band" it was first necessary to define the "band". This was done by visually identifying the largest band in the immunoblot and then manually selecting a square area around that band that captured all the pixels of the band yet minimized the capture of background pixels. An equivalent square area was then selected around the remaining bands again ensuring that all the pixels of the band were captured. The mean pixel intensity of the "bands" was then calculated using Matlab (refer to Chapter 2.15 for further details). By this method the mean pixel intensity should correlate to the band size and intensity such that the larger the mean pixel intensity the larger and more intense the band. The top graph considers the band intensities of the top phosphorylated Erk band (p44) and the bottom graph considers the band intensities of the bottom phosphorylated Erk band (p.42). 115  4.2.6  Co-Expression of the BCR, Syk, BTK and BLNK Is Not Sufficient to Reconstitute  BCR-Induced, PLC-Dependent Erk Phosphorylation in AtT20-Derived Ceil Lines.  Current models suggest that B L N K , Syk and B T K are all required to couple the B C R to PLCy activation. Thus it is hypothesized that maximal PLCy activation would be achieved in a cell line reconstituted with all these components as compared to a cell line reconstituted with only some of these components. To test this hypothesis the B C R / S y k / B L N K / B T K cell was created and assessed for its ability to reconstitute BCR-induced PLCy activation. As can be seen in Figure 4.8a the P L C inhibitor does not appear to inhibit BCR-induced Erk phosphorylation in the B C R / S y k / B L N K / B T K cell line. This suggests that co-expression of the BCR, Syk, B L N K and B T K is not sufficient to reconstitute BCR-induced PLCy activation. This is very surprising both given the proposed models of BCR/PLCy signaling and the previous finding that co-expression of the BCR, Syk and B L N K alone is sufficient to reconstitute BCR-induced PLCy activation (Fig. 4.6b). Considered together, these findings suggest that expression of B T K inhibits, rather than enhances, BCR-induced PLCy activity within this system.  As these findings are quite  unexpected they are discussed at greater length in Chapter 4.3.  Interestingly, while B T K expression appears to inhibit BCR-induced PLC-dependent Erk phosphorylation  it also appears to enhance the  overall Erk phosphorylation in the  B C R / S y k / B L N K / B T K cell line as compared to either the B C R / S y k / B L N K cell line or the BCR/Syk/BTK cell line (compare Fig. 4.8b to Fig. 4.6a and 4.7a, respectively). In fact, coexpression of the BCR, Syk, B L N K and B T K appears sufficient to restore overall BCR-induced Erk phosphorylation to levels comparable to those observed for the BCR/Syk cell line (Fig. 4.8b). This apparent restoration likely reflects a release of the inhibition of the PLC-independent pathway that is apparent in the BCR/Syk/BLNK and BCR/Syk/BTK cell lines.  While, the  reasons for this initial inhibition and its subsequent release are unclear some possible explanations are presented in Chapter 4.3.  116  a. BCR/Syk/BLNK/BTK DMSO Anti mlgM (min) -*  0  3  |  P L C Inhibitor  5 10 15 30 | 0 3  5 10 15 30  50 kD -  IB: ( P ) E r k  37 kD -  S&R:  50 k D -  Erk  37 kD -  b. BCR/Syk Anti mlgM (min) -  [7j  3  BCR/Syk/BLNK/BTK  5 10 15 30 |0 3  5 10 15 30  50 kD -  IB: ( P ) E r k  37 kD -  50 kD -  S&R:  Erk  37 kD -  Figure 4.8. (a) Co-Expression of the B C R , Syk, B L N K and B T K Is Not Sufficient to Reconstitute B C R Induced, PLC-Dependent E r k Phosphorylation in AtT20-Derived Cell Lines. Cells were treated either with D M S O (negative control) or the P L C inhibitor, U73122 (in D M S O , at a final concentration of 10 uM) for 1 hour. The B C R was then cross-linked with anti mlgM antibodies at 37 °C for the indicated length of time. Cells were then lysed and 50 p.g of whole cell lysate were resolved by SDS-PAGE. Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-Erk (Thr202/Tyr204) specific antibody. Finally, filters were stripped and reprobed with an Erk specific antibody. IB indicates the immunoblotting antibody while S & R indicates the antibody used to reprobe the filter. Data are representative of three independent experiments, (b) Overall BCR-Induced E r k Phosphorylation is Comparable in the B C R / S y k and B C R / S y k / B L N K / B T K Cell Lines. The B C R was cross-linked with anti m l g M antibodies at 37 °C for the indicated length of time. Cells were then lysed and 50 ug of whole cell lysate were resolved by SDS-PAGE. Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-Erk (Thr202/Tyr204) specific antibody. Finally, filters were stripped and reprobed with an Erk specific antibody. IB indicates the immunoblotting antibody while S&R indicates the antibody used to reprobe the filter. Data are representative of three independent experiments.  117  BCR-Induced (p44) Phosphorylation in the BCR/Syk/BTK/BLNK Cell Line  B Non-Inhibited • Inhibited  0 min  3 min  5 min  10 min  15 min  30 min  Time  BCR-Induced (p42) Phosphorylation in the BCR/Syk/BTK/BLNK Cell Line  200 180 160  'Si 140 | c  120  El Non-Inhibited  100  • Inhibited  80 60 -I 40 20 0 -M 0 min  3 min  5 min  10 min  15 min  30 min  Time  Figure 4.8. (c) Comparison of Mean Pixel Intensity of the Phosphorylated Erk "Bands" in the BCR/Syk/BTK/BLNK Cell Line in Non-Inhibited and PLC-Inhibited Co-Expression Samples. To calculate the mean pixel intensity of the phosphorylated Erk "band" it was first necessary to define the "band". This was done by visually identifying the largest band in the immunoblot and then manually selecting a square area around that band that captured all the pixels of the band yet minimized the capture of background pixels. A n equivalent square area was then selected around the remaining bands again ensuring that all the pixels of the band were captured. The mean pixel intensity of the "bands" was then calculated using Matlab (refer to Chapter 2.15 for further details). By this method the mean pixel intensity should correlate to the band size and intensity such that the larger the mean pixel intensity the larger and more intense the band. The top graph considers the band intensities of the top phosphorylated Erk band (p44) and the bottom graph considers the band intensities of the bottom phosphorylated Erk band (p42). 118  4.3 Discussion  4.3.1  Syk is Necessary and Sufficient to Reconstitute BCR-Induced, PLC-independent  Erk Phosphorylation in the AtT20 System.  Previous loss-of-function studies clearly suggest that Syk, B T K and B L N K are required to efficiently couple the B C R to PLCy activation (Fig. 4.8). However, loss-of-function studies can not determine whether these proteins are sufficient to achieve a functional pathway or whether additional components are required. Thus, I employed a reconstitution approach to investigate the molecular requirements and mechanisms underlying BCR-induced PLCy activation.  Reconstitution experiments were performed in AfT20-derived cell lines where BCR-induced activation of PLCy was assayed by monitoring BCR-induced increases in Erk phosphorylation. Increases in Erk phosphorylation can be indicative of PLCy activation as Erk phosphorylation is a downstream consequence of PLCy activation.  However, it should be noted that Erk  phosphorylation can also be achieved as a downstream consequence of the BCR-induced PLCindependent pathway (Figure 4.3).  Therefore, a P L C inhibitor was employed to distinguish  between PLC-dependent and PLC-independent increases in BCR-induced Erk phosphorylation.  Initial reconstitution studies were performed in AtT20-derived cell lines that co-expressed the B C R along with B L N K and/or BTK. As mentioned above, previous studies clearly suggest that Syk is required to couple the B C R to downstream signaling pathways including the P L C dependent and PLC-independent pathways that regulate BCR-induced Erk phosphorylation (Fig. 4.8; reviewed in Weiss and Littman, 1994, Gold and Matsuuchi, 1995; Reth and Weinands, 1997, Gold et al, 2000). Thus, one may predict that co-expression of the B C R along with B L N K and/or B T K would be insufficient to reconstitute BCR-induced Erk phosphorylation. Nonetheless, it is prudent to confirm these predictions within the AtT20 system and to ensure that B T K or some unidentified AtT20-specific P T K is not able to compensate for the absence of Syk.  As predicted, co-expression of the B C R along with B L N K and/or B T K is not sufficient to reconstitute BCR-induced Erk phosphorylation within the AtT20 system (Fig. 4.2).  These  findings are in agreement with previous loss-of-function studies that clearly demonstrated that 119  Syk is required for activation of the PI3K, Ras and PLCy pathways in B cells (reviewed in Weiss and Littman, 1994, Gold and Matsuuchi, 1995; Reth and Weinands, 1997, Gold etal, 2000). In contrast, co-expression of the B C R and Syk is sufficient to reconstitute BCR-induced Erk phosphorylation (Fig. 4.4a).  However, as the P L C inhibitor does not appear to inhibit this  process (Fig. 4.4b), it appears that Syk is reconstituting BCR-induced Erk phosphorylation via a PLC-independent pathway within this system.  Again, these findings are in agreement with  previous studies that have shown that co-expression of the B C R and Syk is sufficient to reconstitute BCR-induced Ras phosphorylation and Erk activation, yet insufficient to reconstitute PLCy activation (as assayed via IP3 production) within the AtT20 system (Matsuuchi et al, 1992; Richards et al, 1996). Thus, it appears that Syk is necessary and sufficient to reconstitute BCR-induced, PLC-independent Erk phosphorylation in the AtT20 system.  4.3.2  The Effects  of Syk, B T K and/or  BLNK  Expression on BCR-Induced E r k  Phosphorylation in the AtT20 System.  The effects of Syk, B T K and/or B L N K expression on BCR-induced Erk phosphorylation are quite varied and complex within the AtT20 System. Thus, these effects are summarized in Table 4.3.1 below in an effort to provide the reader with some clarity. Several unexpected findings were made during the course of the abovementioned investigations. These include:  (1) When comparing the BCR/Syk cell line to the B C R / S y k / B L N K cell line, B L N K addition appears to enhance the PLC-dependent Erk pathway while concomitantly inhibiting the P L C independent Erk pathway.  Upon initial inspection it may appear contradictory that B L N K  should enhance Erk phosphorylation via the PLC-dependent pathway while at the same time inhibiting it via the PLC-independent pathway.  Indeed, a parsimonious model would propose  that B L N K , if anything, should enhance the PLC-independent pathway as well as the P L C dependent pathway. However, natural selection (being blissfully unaware of scientists' penchant for parsimony) may have rendered a situation where B L N K may differentially influence the pathways as ultimately each pathway influences a multitude of outcomes beyond mere Erk phosphorylation. Unfortunately, the AtT20 reconstitution system is not ideal for investigating whether this hypothesis is physiologically accurate. Rather, the DT40 knockout system may be better suited to such an investigation. In particular, BCR-induced Erk phosphorylation should be compared in the PLCy knockout versus the B L N K / P L C y double knockout. If one were to 120  Table 4.1. The Effects of Syk, B T K and/or BLNK Expression on BCR-Induced Erk Phosphorylation in the AtT20 System. Note that arrow indicates that BCR-induced phosphorylation was detected. The colour of the arrow is intended to roughly indicate the intensity of the observed phosphorylation relative to the other cell lines. B C R induced PLC-dependent Erk phosphorylation is distinguished from BCR-induced PLC-independent with the aid of the P L C inhibitor, U73122. The intensity of the PLC-independent phosphorylation is inferred by comparing the overall Erk phosphorylation to the PLC-dependent Erk phosphorylation. BCR-induced PLCy activation is inferred to occur if BCR-induced PLC-dependent Erk phosphorylation is observed.  PLCindependent Erk (P) (inferred)  PLCy Activation  (observed)  PLCdependent Erk (P) (observed)  none  none  none  none  BCR/Syk  t  none  t  none  BCR/BTK  none  none  none  none  BCR/BLNK  none  none  none  none  AtT20-Derived Cell Line  BCR  Overall Erk (P)  BCR/Syk/BTK (Fig. 4.7) BCR/Syk/BLNK (Fig. 4.6) BCR/BTK7BLNK  BCR/Syk/BLNK/BTK (Fig. 4.8)  (inferred)  none  none  t  yes  none  none  none  none  t  none  t  none  121  hypothesize that B L N K positively influences BCR-induced Erk-phosphorylation solely via the PLCy-dependent pathway then one would predict that BCR-induced Erk phosphorylation would be equally impaired in the PLCy as compared to the B L N K / P L C y double knockout. In contrast, if one were to hypothesize that B L N K positively influences BCR-induced Erk-phosphorylation via both the PLCy-dependent and PLC-independent pathways one would then predict that BCRinduced Erk phosphorylation would more significantly impaired in the B L N K / P L C y double knock-out as compared to the PLCy knockout. Finally, i f one were to hypothesize that B L N K positively influences BCR-induced Erk-phosphorylation via the PLCy-dependent pathway while negatively influencing it via the PLC-independent pathway one would predict that BCR-induced Erk phosphorylation would be less impaired in the B L N K / P L C y double knockout as compared to the PLCy knockout.  Alternatively, parsimony may have been victorious and B L N K may well only enhance BCRinduced Erk phosphorylation, either via the PLC-dependent pathway alone or via both the PLCdependent and PLC-independent pathways. If this is the physiological reality, then the observed results may merely be a reflection of some inadequacy or anomaly within the AtT20 system. In particular, B L N K could be associating with a component of the PLC-independent pathway in a way that is inhibiting that component's normal function.  Indeed, B L N K has been shown to  associate with Grb2, a putative component of the PLCy-independent pathway (Fu et al, 1998). And while this association has been proposed to have a positive influence on the pathway (Fu et al, 1998), the true physiological significance of the association has yet to be confirmed.  (2) When comparing the BCR/Syk cell line to the BCR/Syk/BTK cell line, B T K addition appears to inhibit the PLC-independent Erk pathway while not affecting the PLC-dependent pathway. Similar to B L N K , B T K may be associating with a component of the PLC-independent pathway in a way that is inhibiting that component's normal function. Interestingly, there is little in the literature to connect B T K to the BCR-induced PLC-independent Erk pathway.  Nonetheless,  given these results, it may be prudent to revisit the literature regarding B T K in an effort to determine if any physiological connection is hinted at and furthermore, such a connection should be further investigated within this system.  (3)  When B L N K and B T K are added together in the B C R / S y k / B L N K / B T K  cell line, they  appear to release the inhibition of the PLC-independent pathway that is observed when they are 122  expressed independently in the BCR/Syk/BLNK and BCR/Syk/BTK cell lines. At first glance this observation may appear contradictory to the above two observations however, it is possible to explain. Primarily, B T K and B L N K , may preferentially interact with each other when coexpressed within the AfT20 system.  Now, based on the aforementioned hypothesis, this  interaction may prevent B L N K and/or B T K from forming inhibitory/non-productive associations with components of the PLC-independent pathway and may thereby release the inhibition of the PLC-independent Erk pathway.  Due to time constraints the hypotheses regarding possible inhibitory interactions between B L N K and/or B T K and components of the PLC-independent pathways were not investigated. However, these hypotheses could be investigated, in part, via co-immunoprecipitation studies.  In  particular, co-immunoprecipitation studies can be performed to determine whether or not B L N K associates with Grb2 and/or any other components of the PLC-independent pathway within this system. While such studies cannot indicate whether or not such an association is inhibitory they would be suggestive; especially if further co-immunoprecipitation studies were to find that that association of a particular component with B L N K correlated with decreased association between that component and others of the PLC-independent pathway. Such an observation would suggest that B L N K may be disrupting the formation of functional PLC-independent signaling complexes which in turn could inhibit BCR-induced PLC-dependent Erk phosphorylation.  (4)  When the B C R , Syk and B L N K are co-expressed, BCR-induced PLC-dependent Erk  phosphorylation appears to be reconstituted in the AtT20 system. However, when the BCR, Syk, B L N K and B T K are co-expressed, BCR-induced PLC-dependent Erk phosphorylation no longer appears to be reconstituted.  This finding is quite surprising as loss-of-function studies clearly  indicate that B T K has a positive influence on the BCR/PLCy pathway in a physiological setting (Takata and Kurosaki, 1996; Fluckinger et al, 1998; detailed in Chapter 1.9). This finding may be explained i f B T K and B L N K are preferentially associating with each other (as suggested above) to form a non-functional signaling complex. As single proteins, B L N K and B T K may be able to exercise a positive influence on the PLC-dependent Erk pathway. However, as a nonfunctional complex they may lose this ability. Given this, the question arises as to why B T K and B L N K may be forming a non-functional as opposed to a functional signaling complex within the AfT20 system  123  Ultimately, co-expression of the BCR, Syk, and B L N K is sufficient to reconstitute BCR-induced PLCy activity (as assayed by increased PLC-dependent Erk phosphorylation) within the AtT20 system. However, it remains unclear whether or not these components alone are sufficient to achieve maximal reconstitution of the pathway. Indeed, loss-of-function studies clearly suggest that B T K should also be required to achieve such maximal reconstitution.  However, as  discussed above, B T K expression appears to have a negative rather than a positive affect on the BCR/PLCy pathway within the AtT20 system.  Thus, the BCR/PLCy pathway is further  dissected within the AtT20 system in an effort to determine where the pathway may be aligning with or diverging from the proposed B cell model.  124  CHAPTER 5  Phosphorylation, Protein Association and Compartmentalization Status of BLNK, BTK and PLCy in the AtT20 System  5.1  Introduction  Based on loss-of-function studies Syk, B T K and B L N K are all proposed to be necessary to couple the B C R to PLCy activation. Furthermore, in recent models (detailed in Chapter 1.9 and reviewed in Fig. 5.1) it has been hypothesize that these components should be sufficient to couple the B C R to PLCy activation. Thus, the initial focus of this thesis was to employ a reconstitution approach to determine i f these components are indeed sufficient to couple the B C R to PLCy activation. Having confirmed sufficiency, subsequent investigations would then be performed to further dissect the molecular mechanisms that govern this process. However, as determined and discussed in Chapter Four, while co-expression of the B C R , Syk, and B L N K is sufficient to reconstitute BCR-induced PLCy activation within the AtT20 reconstitution system it is unclear whether or not maximal activation is achieved. Moreover, co-expression of the BCR, Syk, B L N K and B T K is not sufficient to reconstitute the BCR-induced PLCy pathway. Hence, subsequent investigations focused on determining why these components may not be sufficient to effectively couple the B C R to PLCy activation within this system.  Recalling the proposed model (reviewed in Fig. 5.1), numerous events are required to couple the B C R to PLCy activation. Included in these steps are the phosphorylation of B L N K , Syk, B T K and PLCy. As well, B T K and PLCy must associate with B L N K to form a functional BCR/PLCy signaling complex. And ultimately, these components must be recruited from the cytosol to the plasma membrane. Should one of these steps fail to occur, it is likely that PLCy activation itself would fail to occur. Thus, the phosphorylation, association and compartmentalization status of these various components were assayed within the AfT20 system.  125  BCR  BCR  In resting B lymphocytes the BCR is distributed throughout the plasma membrane (PM). As well, there are some Src farmily kinases constitutively associate with the Iga/p heterodimer of the BCR.  BCR  t  cytosol  b.  Following antigen (Ag) engagement the BCRs become cross-linked and aggregated. This enables the SFKs to autophosphorylate and activate each other as well as to phosphorylate the Iga/p heterodimer.  Syk is then recruited from the cytosol to the phosphorylated Iga/p heterodimer via its SH2 domain. Subsequently, Syk is phosphorylated and activated by the BCRassociated SFKs.  15  Similarly, BLNK is proposed to be recruited from the cytosol to the phosphorylated Iga/p heterodimer via its SH2 domain. BLNK is then phosphorylated by the BCR-associated Syk. Phosphorylated BLNK is then able to recruit a variety of SH2-containing proteins to itself thereby acting as a scaffolding or adapter protein.  ©  BTK is proposed to be recruited from the cytosol to phosphorylated BLNK via its SH2 domain. BTK is then phosphorylated and activated by the BCR-associated and/or the BCR-associated Syk.  f.  Additionally, PLCy is proposed to be recruited from the cytosol to phosphorylated BLNK via its SH2 domain. Such recruitment facilitates PLCy's association with Syk and BTK such that these kinases can then phosphorylate it.  Recruitment of PLCy to BLNK not only facilitates its phosphorylation, it also facilitates PLCy's association with its membrane-associated substrate, PIP which it hydrolyzes to the second messengers DAG and IP . 3  3  Figure 5.1. Review of Proposed Model of BCR-Induced PLCy Activation.  126  5.2  5.2.1  Results  Co-Expression of the BCR and BLNK Is Not Sufficient to Reconstitute BCR-  Induced PLCy Phosphorylation in AtT20-Derived Cell Lines  PLCy activation is dependent, in part, on its phosphorylation (Carter et al, 1991; Coggeshall et al, 1992; Hempel et al., 1992; reviewed in Rhee, 2001). Given this, any inability to reconstitute BCR-induced PLCy activation may reflect an inability to appropriately phosphorylate PLCy. Thus, investigations were performed to assay PLCy's phosphorylation status within this system. Prior studies have indicated that Syk is necessary to mediate PLCy phosphorylation (Takata et al, 1994). Nonetheless, endogenous PTKs, such as Fyn, may be able to compensate for Syk in the AtT20 system. To investigate this possibility PLCy phosphorylation was initially assayed in the B C R and B C R / B L N K , cell lines. To do this, the cell lines were first stimulated by crosslinking the B C R with anti mlgM antibodies for a set length of time (as indicated in the figures). Cells were then lysed and P L C y l was immunoprecipitated from 1000 ug of whole cell lysate using a PLCyl-specific antibody. The immunoprecipitates were then resolved by SDS-PAGE and the resultant gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a pan-phospho-tyrosine specific antibody (4G10) (Figure 5.2).  As can be seen in Figure 5.2, P L C y l does not appear to become phosphorylated following B C R cross-linking in either the B C R or the B C R / B L N K cell lines. This suggests that co-expression of the B C R and B L N K is not sufficient to reconstitute BCR-induced PLCy phosphorylation in the AtT20 system. Moreover, it suggests that there are not any endogenous PTKs within the system that are able to compensate for Syk and phosphorylate P L C y l . (Fig. 5.2). These findings are not unexpected as Syk has repeatedly been demonstrated to be necessary to BCR-induced PLCy phosphorylation and activation (Takata et al, 1994). Thus, studies progressed to investigate whether or not co-expression of the B C R and Syk is sufficient to reconstitute BCR-induced PLCy phosphorylation.  127  BCR Anti mlgM (min) -> | 0  3  5  10  BCR/BLNK 15 301 0  3  5  10  15  301  IP: P L C y l IB: (P) Tyrosine  183 kD  183kD.  S&R:  PLCyl  Figure 5.2. Co-Expression of the B C R and B L N K Is Not Sufficient to Reconstitute BCR-Induced PLCyl Phosphorylation in AfT20-Derived Cell Lines. The BCR was cross-linked with anti m l g M antibodies at 37 °C for the indicated length of time. Cells were then lysed and PLCyl was immunoprecipitated from 1000 ug of whole cell lysate using Protein A-sepharose and a P L C y l specific antibody. Immunoprecipitates were resolved by SDS-PAGE. Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a pan-phospho-tyrosine specific antibody (4G10 monoclonal antibody). Finally, filters were stripped and reprobed with a P L C y l specific antibody. IP indicates the immunoprecipitating antibody, IB indicates the immunoblotting antibody and S & R indicates the antibody that was used to reprobe the filter. Data are representative of three independent experiments. Note that these experiments were performed concurrently with those shown in Figures 5.3 and 5.4. Thus, the experiments shown in Figure 5.3 and 5.4 serve as positive controls that demonstrate that PLCy phosphorylation is indeed detectable in the reconstitution system  128  5.2.2  Co-expression of the BCR with Syk and/or B T K is Sufficient to At Least Partially  Reconstitute BCR-Induced PLCy Phosphorylation in AtT20-Derived Cell Lines  As stated above, previous studies have indicated that Syk is necessary to mediate PLCy phosphorylation (Takata et al, 1994).  Additional studies have also suggested that B T K is  required to achieve maximal BCR-induced PLCy phosphorylation and activation (Takata and Kurosaki, 1996).  Thus, the ability of Syk and/or B T K to reconstitute BCR-induced PLCy  phosphorylation was investigated within this system. To do this, the B C R , BCR/Syk, B C R / B T K and BCR/Syk/BTK cell lines were stimulated by cross-linking the B C R with anti mlgM antibodies for a set length of time (as indicated in the figures). Cells were then lysed and P L C y l was immunoprecipitated from 1000 ug of whole cell lysate using a PLCyl-specific antibody. The immunoprecipitates were then resolved by SDS-PAGE, and the resultant gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a pan-phosphotyrosine specific antibody (4G10) (Figure 5.3).  As can be seen in Figure 5.3, there appears to be an increase in phosphorylated P L C y l following B C R cross-linking in the B C R / B T K , BCR/Syk and BCR/Syk/BTK cell lines. However, this increase is barely detectable in the B C R / B T K cell line (Fig. 5.3c) and appears minimal in the BCR/Syk and BCR/Syk/BTK cell lines (Fig. 5.3b and d). Interestingly, while both Syk and B T K appear independently sufficient to reconstitute BCR-induced PLCy phosphorylation they do not appear to act synergisticalty to enhance PLCy phosphorylation within this system (Fig. 5.3b and c compared to Fig. 5.3d). There are several possible explanations for this observation. First, Syk may phosphorylate the same PLCy sites as B T K such that a synergistic effect is not apparent. Alternatively, phosphorylation of PLCy by one kinase may preclude further phosphorylation by the other kinase. However, neither of these explanations seem likely given that loss-of-function studies clearly suggest that Syk and B T K function co-operatively to phosphorylate PLCy (Takata et al, 1994; Takata and Kurosaki, 1996).  Else wise, Syk and B T K may well be acting  synergistically to enhance BCR-induced PLCy phosphorylation but this may not be readily observable within this system due to the fact that BTK-dependent phosphorylation itself is barely detectable (Fig. 5.3c).  129  The above findings suggest that both Syk and B T K may be independently sufficient to reconstitute BCR-induced P L C y l phosphorylation within this system. However, it remains to be determined i f such a minute increase in phosphorylation is sufficient to be physiologically relevant. This would seem unlikely given that BCR-induced PLCy phosphorylation is typically very robust in B cell lines (Carter et al, 1991; Coggeshall et al., 1992; Hempel et al, 1992; Roifman and Wang, 1992; DeBell et al., 1999).  Furthermore, this appears unlikely as the  observed reconstitution of P L C y l phosphorylation within the BCR/Syk, B C R / B T K and BCR/Syk/BTK cell lines does not appear to correlate with reconstitution of BCR-induced PLCydependent Erk phosphorylation within said cell lines (refer to chapter 4).  The above findings are not surprising given the proposed model of BCR-induced PLCy activation. In fact, B L N K is expected to be necessary to facilitate robust BCR-induced PLCy phosphorylation. Thus, further studies were performed to determine i f co-expression of B L N K could assist in further reconstituting BCR-induced PLCy phosphorylation within the AtT20 system.  130  BCR Anti mlgM (min) -  |fj  3  5  10  15  30 |  183 kD •  IP: PLCyl IB: (P) Tyrosine  183 kD •  S & R : PLCyl  BCR/Syk Anti mlgM (min) -  |o  3  5  10  15  30 |  IP: PLCyl IB: (P) Tyrosine  183kD183kD.  S & R : PLCyl  BCR/BTK Anti mlgM (min) - » |o  3  5  10  15  30 |  183kD-  IP: PLCyl IB: (P) Tyrosine  183 kD •  S & R : PLCyl  BCR/Syk/BTK Anti mlgM (min) -  |Q  3  5  10  15  30 |  183 k D -  IP: PLCyl IB: (P) Tyrosine  183 k D -  S&R:  PLCyl  Figure 5.3. Co-Expression of the BCR, along with Syk or BTK, Is Sufficient to Reconstitute BCR-Induced PLCyl Phosphorylation in AtT20-Derived Cell Lines. The BCR was cross-linked with anti mlgM antibodies at  37 °C for the indicated length of time. Cells were then lysed and PLCyl was immunoprecipitated from 1000 ug of whole cell lysate using Protein A-Sepharose and a PLCyl specific antibody. Immunoprecipitates were resolved by SDS-PAGE. Following electrophoresis, gels were transferred to nitrocellulosefiltersthat were subsequently immunoblotted with a pan-phospho-tyrosine specific antibody (4G10 monoclonal antibody). Finally,filterswere stripped and reprobed with a PLCyl specific antibody. IP indicates the immunoprecipitating antibody, IB indicates the immunoblotting antibody and S&R indicates the antibody that was used to reprobe the filter. Data are representative of three independent experiments.  131  5.2.3  Co-expression of the BCR and BLNK with Syk and/or BTK Does Not Significantly  Enhance BCR-Induced PLCy Phosphorylation  The B C R / B T K / B L N K , B C R / S y k / B L N K and B C R / S y k / B L N K / B T K cell lines were used to determine i f co-expression of B L N K could further assist in reconstituting BCR-induced PLCy phosphorylation. These cell lines were stimulated by cross-linking the B C R with anti mlgM antibodies for a set length of time (as indicated in the figures). Cells were then lysed and P L C y l was immunoprecipitated from 1000 ug of whole cell lysate using a PLCyl-specific antibody. The immunoprecipitates were then resolved by SDS-PAGE with the resolved gels being transferred to nitrocellulose filters that were subsequently immunoblotted with a pan-phosphotyrosine specific antibody (4G10) (Figure 5.4).  Comparing the B C R / B T K cell line to the B C R / B T K / B L N K cell line, co-expression of B L N K does not appear to enhance BCR-induced P L C y l phosphorylation (Fig. 5.4a).  Similarly,  comparing the BCR/Syk cell line to the B C R / S y k / B L N K cell line, co-expression of B L N K does not appear to enhance BCR-induced P L C y l phosphorylation (Fig. 5.4b). In contrast, comparing the BCR/Syk/BTK cell line to the B C R / S y k / B T K / B L N K cell line, co-expression of B L N K appears to slightly enhance BCR-induced P L C y l phosphorylation (Fig. 5.4c). However, this enhanced P L C y l phosphorylation still pales in comparison to what is typically observed following B C R cross-linking in B cell lines (Carter et al, 1991; Coggeshall et al., 1992; Hempel et al, 1992; Roifman and Wang, 1992; DeBell et al, 1999) and thus, again calls into question whether or not such as small increase is physiologically relevant.  Recalling Chapter 4, PLCy activation is reconstituted in the B C R / S y k / B L N K cell line. Thus, a very small amount of PLCy phosphorylation does appear to be sufficient to reconstitute PLCy activation, at least in this cell line. This finding is surprising as it was hypothesized that a much more robust level of PLCy phosphorylation would be required to reconstitute measurable PLCy activation based on the observation that PLCy phosphorylation is very robust in activated B cells (Carter et al, 1991; Coggeshall et al, 1992; Hempel et al, 1992; Roifman and Wang, 1992; DeBell et al, 1999).  While this finding may suggest that the downstream Erk pathway is  sensitive to very subtle changes in PLCy activation the question arises as to whether or not maximal PLCy phosphorylation and activation is being achieved in this system, and i f not, why. 132  Interestingly, while a minimal amount of PLCy phosphorylation appears sufficient to reconstitute PLCy activity in the B C R / S y k / B L N K cell line a comparable or even slightly greater amount of PLCy phosphorylation does not appear sufficient to reconstitute PLCy activation in the B C R / S y k / B L N K / B T K cell line. This suggests that PLCy activation may not be predicated solely on its phosphorylation status. Indeed, additional factors such as specific protein associations and specific localization within the cell (i.e., compartmentalization) may also influence PLCy activity. Thus these factors, as they relate to the BCR, Syk, B L N K , B T K and PLCy were further investigated within the AfT20 system (refer to Chapter 5.2.8 and 5.2.9)  133  a. BCR/BTK Anti mlgM (min) -> | 0  3  5  10  15 301  BCR/BTK/BLNK 0  3  5  10  15  30 |  183kD-  IP: IB:  183kD.  S&R:  BCR/Syk Anti m l g M (min) -  | 0  3  5  10  PLCyl  BCR/Syk/BLNK  15 30 | 0  3  5  10  15  30  183 kD -  183kD.  IP:  PLCyl  IB:  (P) Tyrosine  S&R:  BCR/Syk/BTK Anti m l g M (min) -1~0  PLCyl (P) Tyrosine  3  5  10  PLCyl  BCR/Syk/BTK/BLNK  15 30 | 0  3  5  10  15  30  183 kD  IP: PLCyl IB: (P) Tyrosine  183 kD -  S&R:  PLCyl  Figure 5.4. Co-Expression of the BCR and Syk Is Sufficient to Reconstitute BCR-Induced PLCyl Phosphorylation in AtT20-Derived Cell Lines. The B C R was cross-linked w i t h anti m l g M antibodies at 37 ° C for the indicated length o f time. C e l l s were then lysed and P L C y l was immunoprecipitated from 1000 ug o f whole cell lysate using Protein A-sepharose and a P L C y l specific antibody. Immunoprecipitates were resolved by S D S - P A G E . F o l l o w i n g electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a pan-phospho-tyrosine specific antibody (4G10 monoclonal antibody). F i n a l l y , filters were stripped and reprobed with a P L C y l specific antibody. IP indicates the immunoprecipitating antibody, I B indicates the immunoblotting antibody and S & R indicates the antibody that was used to reprobe the filter. D a t a are representative o f three independent experiments.  134  5.2.4  B T K Is Constitutively Phosphorylated in AtT20-Derived Cell Lines  As observed in the previous section, BCR-induced PLCy phosphorylation and/or activation may not be maximally reconstituted within the AtT20 system.  Considering the proposed model,  attempts to reconstitute this pathway could be limited i f any of Syk, B T K or B L N K were to be inappropriately phosphorylated.  Thus, the phosphorylation status of these proteins was  investigated.  Previous studies have indicated that co-expression of the B C R and Syk is sufficient to reconstitute BCR-induced Syk phosphorylation and activation with the AtT20 system (Richards et al, 1996) and therefore, studies investigating Syk phosphorylation status have been done previously and were not further pursued here. In contrast, B T K ' s phosphorylation status was assayed in the B C R / B T K , B C R / B T K / B L N K , BCR/Syk/BTK and B C R / S y k / B L N K / B T K cell lines. To do this, the cell lines were stimulated by cross-linking the B C R with anti mlgM antibodies for a set length of time (as indicated in the figures). Cells were then lysed and B T K was immunoprecipitated from 1000 ug of whole cell lysate using a BTK-specific antibody. The immunoprecipitates were then resolved by SDS-PAGE and the gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-BTK (Tyr223) specific antibody (Figure 5.5).  It should be noted that B T K becomes  sequentially  phosphorylated on tyrosine residues 551 and 223, and that phosphorylation of this latter residue is generally considered to be indicative of active B T K (Rawlings et al, 1996; Park et al., 1996).  As can be seen in Figure 5.5, B T K appears to be constitutively phosphorylated on tyrosine 223 in the B C R / B T K , B C R / B L N K / B T K ,  BCR/Syk/BTK  and B C R / S y k / B L N K / B T K  cell lines.  Furthermore, B T K ' s phosphorylation does not appear to increase following B C R cross-linking (Fig. 5.5). The finding that co-expression of the B C R and B T K is sufficient to reconstitute B T K phosphorylation is not unexpected as B T K phosphorylation has been shown to be mediated by Src family members (Afar et al., 1996; Rawlings et al., 1996; detailed in Chapter 1.9.2) and as the AtT20 system endogenously expresses the Src family member, Fyn (Richards et al., 1996). However, the observation that B T K phosphorylation is constitutive as opposed to inducible is somewhat surprising.  While it is not unusual to observe constitutive activation in over-  expression systems (Kurosaki et al, 1994; Rawlings et al, 1996) one of the advantages of the AtT20 system is that, to date, it has proven to be a system where over-expression does not 135  BCR/BTK Anti mlgM (min) -  0  3  5  10 15  30  IP: B T K IB: (P) B T K  80 kD • 80 kD.  S & R : BTK  BCR/BTK/BLNK Anti mlgM (min) -  fj  3  5  10  15  30  IP: B T K IB: (P) B T K  80 kD •  S & R : BTK  80 kD • BCR/Syk/BTK Anti mlgM (min) -  "Q  3  5  10  15  30~|  IP: B T K IB: (P) B T K  80 kD80 kD •  S & R : BTK  BCR/Syk/BLNK/BTK Anti mlgM (min) -  0  3  5  10  15  30  IP: B T K IB: (P) B T K  80 kD> 80 kD •  S & R : BTK  Figure 5.5. Co-Expression of the BCR and BTK is Sufficient to Reconstitute BCR-Induced BTK Phosphorylation in AfT20-Derived Cell Lines. The B C R was cross-linked with anti m l g M antibodies at 37 °C for the indicated length of time. Cells were then lysed and B T K was immunoprecipitated from 1000 ug of whole cell lysate using Protein G-Sepharose and a B T K specific antibody. Immunoprecipitates were resolved by S D S - P A G E . Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-BTK (Tyr223) specific antibody. Finally, filters were stripped and reprobed with a B T K specific antibody. IP indicates the immunoprecipitating antibody, IB indicates the immunoblotting antibody and S&R indicates the antibody that was used to reprobe the filter. Data are representative of three independent experiments.  136  necessitate activation and thus it lends itself well to studies that are concerned with investigating inducible pathways (Richards et al, 1996).  Several controls were performed to confirm that the observed B T K phosphorylation was not merely a methodological artifact.  First, it was confirmed that the phospho-BTK (Tyr223)  specific antibody can indeed distinguish between phosphorylated B T K and non-phosphorylated B T K . To do this the Daudi and Ramos human B cell lines were used as controls. These cell lines, along with the B C R / B L N K / B T K and B C R / S y k / B L N K / B T K cell lines, were left either non-stimulated or were stimulated for 10 minutes by cross-linking the B C R with an anti mlgM specific antibody. Cells were then lysed and 50 ug of whole cell lysate were resolved by SDSP A G E . The resolved gels were then transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-BTK (Tyr223) specific antibody (Figure 5.6a).  The Daudi and Ramos B cell lines were used as controls as it has previously been established that there are small amounts of phosphorylated B T K in resting/non-stimulated B cells and that this amount dramatically increases following B C R engagement (Mahajan et al., 1995; Rawlings et al, 1996; Afar et al, 1996; Kurosaki et al, 1997; Baba et al, 2001). Thus, if the phosphoB T K (Tyr223) specific antibody is specific for phosphorylated B T K a small amount of phosphorylated B T K should be observed in the non-stimulated cells while a much greater amount should be observed in stimulated cells.  In contrast, i f the phospho-BTK (Tyr223)  specific antibody is unable to distinguish between phosphorylated and non-phosphorylated B T K the similar amounts of supposedly phosphorylated B T K should be observe in the non-stimulated and stimulated cells as the antibody will likely be detecting the entire pool of B T K (a pool that is not anticipated to change in response to stimulation). As can be seen in Figure 5.6a, a small amount of phosphorylated B T K is observed in the non-stimulated B cell lines and this amount dramatically increases following stimulation. In contrast, there appears to be a relatively large and consistent amount of phosphorylated B T K in both the non-stimulated and stimulated AtT20derived cell lines. The former observation suggests that the phospho-BTK (Tyr223) specific antibody can indeed distinguish between phosphorylated B T K and non-phosphorylated B T K while the latter observation, coupled with the former, suggests that B T K is indeed constitutively phosphorylated in the AtT20-derived cell lines.  137  While the above findings are suggestive it is still possible that the phospho-BTK (Tyr223) specific antibody is non-specifically cross-reacting with a protein within the AtT20 system such that it is creating a background band that could be misinterpreted as being representative of phosphorylated B T K . This possibility was addressed using the B C R / S y k / B L N K cell line as a negative control (5.6b). Similar to before, the B C R / S y k / B L N K and the B C R / S y k / B L N K / B T K cell lines were first stimulated by cross-linking the B C R with anti mlgM antibodies for a set length of time (as indicated in the figures).  Cells were then lysed and B T K was  immunoprecipitated from 1000 ug of whole cell lysate using a BTK-specific antibody. The immunoprecipitates were then resolved by SDS-PAGE and the resolved gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-BTK (Tyr223) specific antibody (Figure 5.6b).  If the phospho-BTK (Tyr223) specific antibody is non-  specifically cross-reacting with a protein within the AtT20 system it is expected that the subsequent background band will coincide with the supposed phospho-BTK band and be apparent in both the BCR/Syk/BLNK cell line and the B C R / S y k / B L N K / B T K cell lines. As can be seen in Figure 5.6b, there are not any background bands coinciding with the phospho-BTK band in the B C R / S y k / B L N K cell line as compared to the B C R / S y k / B L N K / B T K cell line. Thus, when considered with the above findings, it appears that the phospho-BTK (Tyr223) specific antibody is indeed specific for phospho-BTK and that B T K is indeed constitutively phosphorylated within the AtT20 system.  Having determined that B T K is constitutively  phosphorylated within the AtT20 system, the mechanism of this phosphorylation was briefly investigated. In particular, studies were performed to determine i f B T K phosphorylation was attributable to Fyn activity.  138  a.  T3  Anti mlgM (min) •  | 0 10  80 k D -  o m  at  0  10 0  10  K/BTK  BLNK/ 22  >>  CO  o  o CO  00  10  10  m  m  80 k D -  r2  ro  z  _J CQ  co  E  ro Q  1CQ  0  —  .  10 | IP: B T K IB: (P) B T K  m  I  mm  S & R : BTK  b.  BCR/Syk/BLNK Anti mlgM (min) -  |~0  3  5  10  15  BCR/Syk/BLNK/BTK 301  0  3  5  10  15  30  80 k D -  IP: B T K IB: (P) BTK  80 kD -  S & R : BTK  Figure 5.6. The phospho-BTK (Tyr223) specific antibody is specific for phosphorylated B T K . (a) The phospho-BTK (Tyr223) specific antibody does not appear to detect non-phosphorylated B T K . The B C R was crosslinked with anti mlgM antibodies at 37 °C for 10 minutes. Cells were then lysed and 50 ug of whole cell lysate were resolved by S D S - P A G E . Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-BTK (Tyr223) specific antibody. Finally, filters were stripped and reprobed with a B T K specific antibody. IB indicates the immunoblotting antibody and S & R indicates the antibody that was used to reprobe the filter. The Daudi and Ramos cell lines are human B cell lines that should be observed to express low levels of phosphorylated B T K when not stimulated and high levels of phosphorylated B T K when stimulated, (b) The phospho-BTK (Tyr223) specific antibody does not appear to cross-react with any other proteins in the AtT20 system). The B C R was cross-linked with anti mlgM antibodies at 37 °C for the indicated length of time. Cells were then lysed and B T K was immunoprecipitated from 1000 p.g of whole cell lysate using Protein G Sepharose and a B T K specific antibody. Immunoprecipitates were resolved by SDS-PAGE. Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phosphoB T K (Tyr223) specific antibody. Finally, filters were stripped and reprobed with a B T K specific antibody. IP indicates the immunoprecipitating antibody, IB indicates the immunoblotting antibody and S & R indicates the antibody that was used to reprobe the filter. Data are representative of three independent experiments. The B C R / S y k / B L N K cell line serves as a negative control for the phospho-BTK (Tyr223) specific antibody as it is an AtT20-derived cell line that does not express B T K .  139  5.2.5  B T K Phosphorylation Is At Least Partially Dependent on Fyn Activity in AtT20-  Derived Cell Lines  Previous studies have indicated that the Src kinases are responsible for trans-phosphorylating B T K in B cells (Afar et al, 1996; Rawlings et al, 1996). Once so phosphorylated B T K then auto-phosphorylates such that it is believed to have maximal kinase activity (Rawlings et al, 1996; Kurosaki et al, 1997; Wahl et al, 1997).  Thus, investigations were performed to  determine if the endogenous Fyn (a Src family member) was mediating the observed constitutive phosphorylation of B T K within the AtT20 system. To do this the B C R / B T K , B C R / B T K / B L N K , BCR/Syk/BTK and B C R / S y k / B L N K / B T K cell lines were treated overnight with PP2, a Src family kinase inhibitor (added to a final concentration of 5 uM). The following morning, the cell lines were stimulated by cross-linking the B C R with anti mlgM antibodies for 10 minutes. Cells were then lysed and 50 ug of whole cell lysate were resolved by SDS-PAGE.  The gels were  transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-BTK (Tyr223) specific antibody (Fig. 5.7). As can be seen in Figure 5.7, treatment with PP2 appears to decrease the amount of phosphorylated B T K observed in both the non-stimulated and stimulated AtT20-derived cell lines. However, PP2 treatment does not appear to completely inhibit B T K phosphorylation in these cell lines (Fig. 5.7). This may suggest either that PP2 treatment is not sufficient to completely inhibit Fyn activity or that an unidentified P T K is contributing to B T K phosphorylation within this system. However, this second explanation is unlikely given that no other Src family kinase members appear to be expressed within the AtT20-derived system (Richards et al, 1996).  It should be noted that the results for the BCR/Syk/BTK and B C R / S y k / B L N K / B T K cell lines are somewhat unusual as compared to the two cell lines that do not express Syk (Fig. 5.7). Specifically, multiple bands are apparent above the band that is taken to represent phosphorylated B T K (Fig. 5.7). These bands could be background bands that become evident only in the presence of Syk. However, if these bands are attributable to Syk it might be expected that these bands would appear more prevalent in the B C R / S y k / B L N K / B T K cell  line as  compared to the BCR/Syk/BTK cell line as the former appears to express more Syk than the latter (Fig. 4.2). Yet, the opposite observation is apparent in that the background bands appear more prevalent in the BCR/Syk/BTK cell line as compared to the B C R / S y k / B L N K / B T K cell line (Fig. 5.7). Arguably, the presence of B L N K could somehow be counteracting the effect of Syk 140  on these proposed background bands.  Alternatively, these bands may represent differentially  phosphorylated B T K and may only exist within the Syk expressing cell lines due to the differential phosphorylation being mediated by Syk. This explanation could help to account for the difference in intensity of the bands between the two cell lines as the B C R / S y k / B T K / B L N K cell line expresses considerably less B T K than the BCR/Syk/BTK cell line (Fig. 4.2). This theory could be better supported by using cell lines that show similar expression levels of both Syk and B T K . Unfortunately, it was not reasonably possible to obtain a selection of clones that expressed similar levels these proteins and thus, such a comparison can not be made.  The  proposal that Syk may be contributing B T K phosphorylation is not novel. In fact, Kurosaki and his colleagues have recently shown that Syk may contribute to B T K phosphorylation (Kurosaki and Kurosaki, 1997; Baba et al, 2001)  Thus, the above findings hint at a mechanism for constitutive B T K phosphorylation within the AtT20 system. In particular, it appears that Fyn is at least partially responsible for mediating constitutive B T K phosphorylation.  Furthermore, Syk may be able to contribute to B T K  phosphorylation. Ideally, further studies could be performed within this system to further define the mechanisms involved in phosphorylating B T K including performing studies with a Syk inhibitor. However, central to this thesis is the ability to reconstitute B T K phosphorylation as a step towards reconstituting BCR-induced PLCy activation.  As this appears to have been  achieved, the focus of the study returned to more direct investigations of BCR-induced PLCy activation including investigating B L N K ' s phosphorylation status.  141  BCR/BTK  BCR/BLNK/BTK  | DMSO | PP2 | DMSOl PP2 | Anti mlgM (min) - | 0  10 | 0  10 | 0  10 I 0  10 |  80 kD-  IB: (P) B T K  80 kD-  S&R:  BTK  BCR/Syk/ BCR/Syk/BTK  BLNK/BTK  DMSO | PP2 | DMSO | PP2 Anti mlgM (min) -  | 0  10 | 0  10 | 0  10 | 0  10  80 kD •  IB: (P) B T K  80 kD-  S&R:  BTK  Figure 5.7. Constitutive Phosphorylation of B T K is, At Least, Partly Dependent on F y n Activity in AtT20Derived Cell Lines. Cells were treated overnight with either D M S O (negative control) or the Src Family Kinase inhibitor, PP2 (at a final concentration of 5 uM). The BCR was then cross-linked with anti mlgM antibodies at 37 °C for the indicated length of time. Cells were then lysed and 50 pg of whole cell lysate were resolved by SDSPAGE. Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-BTK (Tyr223) specific antibody. Finally, filters were stripped and reprobed with a B T K specific antibody. IB indicates the immunoblotting antibody while S & R indicates the antibody used to reprobe the filter. Data are representative of at least three similar independent experiments.  142  5.2.6  Co-Expression of the BCR, BLNK and BTK Is Not Sufficient to Reconstitute BCR-  Induced BLNK Phosphorylation in AtT20-Derived Cell Lines  Previous studies have clearly shown that B T K and PLCy associate with phosphorylated B L N K via their SH2 domains (Hashimoto et al., 1999; Su et al, 1999; and Ishiai et al, 1999; respectively). Thus, i f B L N K is to function as an adapter protein within this system it must become appropriately phosphorylated. Failure to do so could contribute to the apparent inability to reconstitute robust BCR-induced PLCy phosphorylation (refer to Chapter 5.2.1-5.2.3) and PLCy activation within this system (refer to chapter 4). Given this, studies were performed to investigate whether or not BCR-induced B L N K phosphorylation was reconstituted within the AfT20-derived system.  Prior studies have indicated that Syk is necessary to mediate BCR-induced  BLNK  phosphorylation (Fu et al, 1998). Nonetheless, endogenous PTKs, such as Fyn, may be able to mediate BCR-induced B L N K phosphorylation in the AfT20-derived system.  Alternatively,  exogenous B T K may be able to mediate BCR-induced B L N K phosphorylation in this system. To  investigate these possibilities B L N K  phosphorylation was initially assayed in the  B C R / B L N K and B C R / B L N K / B T K cell lines. To do this, the cell lines were first stimulated by cross-linking the B C R with anti mlgM antibodies for a set length of time (as indicated in the figures). Cells were then lysed and myc-tagged B L N K was immunoprecipitated from 1000 ug of whole cell lysate using a myc specific antibody (9E10). The immunoprecipitates were then resolved by SDS-PAGE and the resultant gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a pan-phospho-tyrosine specific antibody (4G10) (Fig. 5.8) before being stripped and reprobed with a B L N K specific antibody (AC). Alternatively, 50 ug of whole cell lysate were resolved by SDS-PAGE.  The resolved gels were then transferred to  nitrocellulose filters that were subsequently immunoblotted with a phospho-BLNK (Tyr96) specific antibody (Fig. 5.9).  Based on these experiments, B C R engagement does not. appear sufficient to induce B L N K phosphorylation in either the B C R / B L N K cell line (Fig. 5.8) or the B C R / B L N K / B T K cell line (Fig. 5.9). The former result suggests that there are not any endogenous PTKs sufficient to reconstitute BCR-induced B L N K phosphorylation within the AtT20-derived system. Similarly, 143  the latter result suggests that B T K alone is not sufficient to reconstitute BCR-induced B L N K phosphorylation within the AtT20-derived system.  Given that the aforementioned model of  BCR-induced PLCy activation envisions B L N K being phosphorylated in a Syk-dependent manner, the above findings are as expected. Thus, studies progressed to investigate whether or not co-expression of the B C R , B L N K and Syk was sufficient to reconstitute BCR-induced B L N K phosphorylation.  144  BCR/BLNK Anti mlgM (min) - » [ " -  +1  1  I  so kD  i  IP:  myc(9E10)  IB:pan-phosphotyrosine(4G10)  BOkD-EZD  S & R : BLNK (AC)  b. BCR/BLNK Anti m l g M (min) -»  I -  50 kD -  +  IB: pan-phosphotyrosine (4G10)  37 kDFigure 5.8. Co-Expression of the BCR and BLNK is Not Sufficient to Reconstitute BCR-Induced BLNK Phosphorylation in AtT20-Derived Cell Lines. In Contrast, Co-Expression of the BCR, Syk and B L N K is Sufficient to Reconstitute BCR-Induced B L N K Phosphorylation in AtT20-Derived Cell Lines, (a) The B C R was not cross-linked in the non-stimulated samples (-) and was cross-linked with anti m l g M antibodies at 37 °C for the 5 minutes in the stimulated samples (+). Cells were then lysed and myc-tagged human B L N K was immunoprecipitated from 1000 ug of whole cell lysate using a myc specific antibody (9E10). Immunoprecipitates were resolved by S D S - P A G E . The gels were then transferred to nitrocellulose filters that were subsequently immunoblotted with a pan-phospho-tyrosine specific antibody (4G10). Finally, filters were stripped and reprobed with a B L N K specific antibody (AC). IP indicates the immunoprecipitating antibody, IB indicates the immunoblotting antibody while S & R indicates the antibody that was used to reprobe the filter. Data are representative of three independent experiments. Similar results were obtained with multiple clones expressing the same combination of proteins, (b) Control blot demonstrating that cross-linking of the B C R was successful. Ten micrograms of whole cell lysate from the above samples were resolved by S D S - P A G E . The resolved gels were then transferred to nitrocellulose filters that were subsequently immunoblotted with a pan-phospho-tyrosine specific antibody (4G10). Following B C R cross-linking an increase in total tyrosine phosphorylation is evident indicating that B C R cross-linking was successful.  145  BCR/BTK/BLNK Anti mlgM (min) -  0  3  5  10 15 30  80 kD-  IB: (P) BLNK  80 k D -  S & R: BLNK  Figure 5.9. Co-Expression of the BCR, B T K and BLNK is Not Sufficient to Reconstitute BCR-Induced B L N K Phosphorylation in AtT20-Derived Cell Lines. The B C R was cross-linked w i t h anti m l g M antibodies at 37 ° C for the indicated length o f time. Cells were then lysed and 50 p g o f whole cell lysate were resolved b y S D S PAGE. F o l l o w i n g electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a p h o s p h o - B L N K (Tyr96) specific antibody. F i n a l l y , filters were stripped and reprobed with a B L N K specific antibody. I B indicates the immunoblotting antibody w h i l e S & R indicates the antibody that was used to reprobe the filter. Data are representative o f three independent experiments.  146  5.2.7  Co-Expression of the BCR, Syk and BLNK is Sufficient to Reconstitute BCR-  Induced BLNK Phosphorylation in AtT20-Derived Cell Lines  As observed in the previous section, BCR-induced PLCy phosphorylation and/or activation may not be maximally reconstituted within the AtT20 system.  Considering the proposed model,  attempts to reconstitute this pathway could be limited i f any of Syk, B T K or B L N K were to be inappropriately phosphorylated.  Thus, the phosphorylation status of these proteins was  investigated.  To determine i f co-expression of the BCR, Syk and B L N K is sufficient to reconstitute BCRinduced B L N K phosphorylation the BCR/Syk/BLNK and B C R / S y k / B L N K / B T K cell lines were first stimulated by cross-linking the B C R with anti mlgM antibodies for a set length of time (as indicated in the figures). Cells were then lysed and 50 ug of whole cell lysate were resolved by SDS-PAGE.  The resolved gels were then transferred to nitrocellulose filters that were  subsequently immunoblotted with a phospho-BLNK (Tyr96) specific antibody (Fig. 5.10a and 5.11). Alternatively, B L N K was immunoprecipitated from 1000 ug of whole cell lysate using a BLNK-specific antibody (2B11 or H80, as indicated in the figures). The immunoprecipitates were then resolved by SDS-PAGE and the resolved gels were then transferred to nitrocellulose filters that were subsequently immunoblotted with a pan-phospho-tyrosine specific antibody (4G10) (Fig. 5.10b; also refer to Fig. 5.8).  BCR-induced B L N K phosphorylation appears to be reconstituted in both the BCR/Syk/BLNK cell line (Fig. 5.8 and 5.10) and the B C R / S y k / B L N K / B T K cell line (Figure 5.11). Upon initial inspection, it appears that BCR-induced B L N K  phosphorylation is enhanced  in the  B C R / S y k / B L N K / B T K cell line as compared to the B C R / S y k / B L N K cell line (Fig. 5.10b compared to 5.11). From this it may be tempting to conclude that co-expression of B T K , along with Syk, enhances BCR-induced B L N K phosphorylation. However, the apparent enhanced B L N K phosphorylation may merely reflect the fact that more B L N K is expressed in the B C R / S y k / B L N K / B T K cell line as compared to the B C R / S y k / B L N K cell (refer to Fig. 4.2). This fact may not be readily apparent as, according to the strip and reprobe, there appears to be less B L N K in the B C R / S y k / B L N K / B T K cell line as compared to the B C R / S y k / B L N K cell (compare bottom panels of Fig 5.10b and 5.11). However, this is likely an artifact of the strip and reprobe process as it has been a consistent observation within the Matsuuchi lab that the intensity of 147  reprobed bands appear to be inversely proportional to the intensity of the initial phospho-tyrosine band (for example compare the zero minute time point to the thirty minute time point in Fig. 5.11). Additionally, it should be noted that a basal level of B L N K tyrosine phosphorylation is evident prior to B C R engagement within this system (refer to the zero time points in Figs. 5.10 and 5.11).  Such basal phosphorylation may be an artifact of the various exogenous B cell  components being over-expressed however; basal phosphorylation of B L N K is also evident in B cell lines (Hashimoto et al, 1999). Regardless, co-expression of the BCR, Syk and B L N K appears sufficient to reconstitute BCR-induced B L N K phosphorylation within the AtT20-derived system (Fig. 5.8, Fig. 5.10 and 5.11).  Furthermore, BCR-induced B L N K phosphorylation  appears to be Syk-dependent as neither endogenous Fyn (Fig. 5.8) nor exogenous B T K (Fig. 5.9) appear sufficient to reconstitute this event in this system.  Given that B L N K phosphorylation appears to be reconstituted it could be predicted that B L N K should be able to associate with B T K and PLCy within this system. However, B L N K ' s ability to associate with various proteins appears dependent upon which of its tyrosine residues become phosphorylated  (Chiu  et ah,  2002).  While  reconstitution  of BCR-induced  BLNK  phosphorylation is evident within this system, beyond tyrosine 96 (Fig. 5.10b and 5.11), it is not clear which tyrosines are becoming phosphorylated. It may be that all the tyrosine residues are becoming phosphorylated or it may be that only a subset of the residues is becoming phosphorylated. Thus, while B L N K phosphorylation may appear to be reconstituted within this system it does not necessarily follow that B L N K ' s association with B T K and PLCy has likewise been reconstituted.  Accordingly, investigations were performed to determine i f the ability to  reconstitute BCR-induced B L N K phosphorylation correlated with an ability to reconstitute BCRinduced B L N K / P L C y / B T K association.  148  a. BCR/Syk/BLNK Anti mlgM (min) -  |~0  3  5  10 15  30 I  80 kD.  IB: (P) B L N K  80 kD.  S & R : BLNK  BCR/Syk/BLNK Anti mlgM (min) -  0  3  5  10 15  30  80 kD-  IP: B L N K IB: (P) Y  80 kD-  S & R : BLNK  Figure 5.10. Co-Expression of the BCR, Syk and BLNK is Sufficient to Reconstitute BCR-Induced BLNK Phosphorylation in AtT20-Derived Cell Lines, (a) The B C R was cross-linked with anti m l g M antibodies at 37 °C for the indicated length of time. Cells were then lysed and 50 ug of whole cell lysate were resolved by SDS-PAGE. Following electrophoresis, gels were transferred to nitrocellulose fdters that were subsequently immunoblotted with a phospho-BLNK (Tyr96) specific antibody. Finally, filters were stripped and reprobed with a B L N K specific antibody. IB indicates the immunoblotting antibody while S&R indicates the antibody that was used to reprobe the filter. Data are representative of three independent experiments, (b) The B C R was cross-linked with anti mlgM antibodies at 37 °C for the indicated length of time. Cells were then lysed and B L N K was immunoprecipitated from 1000 pg of whole cell lysate using Protein A-Sepharose and a B L N K specific antibody (H80). Immunoprecipitates were resolved by S D S - P A G E . Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a pan-phospho-tyrosine specific antibody (4G10 monoclonal antibody). Finally, filters were stripped and reprobed with a B L N K specific antibody (H80). IP indicates the immunoprecipitating antibody, IB indicates the immunoblotting antibody and S & R indicates the antibody that was used to reprobe the filter. Data are representative of three independent experiments.  149  BCR/Syk/BTK/BLNK Anti mlgM (min) -  I fj  3  5  10 15  30  80 kD-  IB: (P) BLNK  80 k D -  S & R : BLNK  Figure 5.11. Co-Expression of the BCR, Syk, BLNK and B T K is Sufficient to Reconstitute BCR-Induced B L N K Phosphorylation in AtT20-Derived Cell Lines. The B C R was cross-linked with anti mlgM antibodies at 37 °C for the indicated length of time. Cells were then lysed and 50 ug of whole cell lysate were resolved by SDSPAGE. Following electrophoresis, gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a phospho-BLNK (Tyr96) specific antibody. Finally, filters were stripped and reprobed with a B L N K specific antibody. IB indicates the immunoblotting antibody while S & R indicates the antibody that was used to reprobe the filter. Data are representative of three independent experiments.  150  5.2.8  Protein Association Studies are Inconclusive in Lymphoid and AtT20-Derived Cell  Lines (refer to Appendix II)  As noted above, BCR-induced PLCy phosphorylation and/or activation may not be maximally reconstituted within the AtT20 system.  According to the proposed model, attempts to  reconstitute this pathway could be limited i f any of the BCR, Syk, B T K , B L N K or PLCy did not properly associate to form a functional BCR/PLCy signaling complex. O f particular concern is B L N K ' s ability to associate with the various proteins as B L N K is proposed to nucleate the BCR/PLCy signaling complex. Thus, the ability of B L N K to associate with B T K and PLCy was assayed within this system. To do this, the various cell lines were initially stimulated by crosslinking the B C R with anti mlgM antibodies.  Cells were then lysed and B L N K was  immunoprecipitated from 1000 ug of whole cell lysate using a BLNK-specific antibody. The immunoprecipitates were then resolved by SDS-PAGE and the resultant gels were transferred to nitrocellulose filters.  Subsequently, the filters were immunoblotted with a P L C y l specific  antibody to determine i f P L C y l had co-immunoprecipitated with B L N K as this would be indicative of an association. Finally, the filters were striped and re-probed with a B L N K specific antibody to confirm that the immunoprecipitation had been successful. Alternately, the filters were first immunoblotted with a B T K specific antibody to determine i f B T K had coimmunoprecipitated with B L N K .  Furthermore, reciprocal immunoprecipitation studies were  performed with antibodies specific for P L C y l and B T K .  Based on the above approach, B L N K , B T K and P L C y l do not appear to co-immunoprecipitate either before or following B C R cross-linking in this system (data not shown; summarized in Appendix II). These results may suggest that these proteins do not co-associate within this system. This could suggest either that B L N K is inappropriately/incompletely phosphorylated such that it can not serve its function as an adapter protein and/or that additional lymphoid components may be required to facilitate these associations. Alternatively, B L N K may not be able to associate with the P L C y l isoform that is endogenously expressed within the AtT20 system. However, this seems unlikely as previous studies have demonstrated that B L N K can associate with both P L C y l and PLCy2 (Fu and Chan; 1997; Fu et al, 1998). Nonetheless, the AtT20 system was transfected with PLCy2 and further immunoprecipitation assays were performed to address this possibility. Unfortunately, the findings for these assays were as before in that B L N K , B T K and PLCy2 do not appear to co-immunoprecipitate (data not shown; 151  summarized to Appendix II). Thus, again these results may suggest that these proteins do not coassociate within this system.  The above results may also suggest that the immunoprecipitation assay is not optimized for detecting protein associations within this system. If the immunoprecipitation assay is functional it should be able to replicate the findings of previous immunoprecipitation studies.  Thus,  attempts were made to replicate the findings that B L N K and PLCy co-associate in the human Daudi B cell line following B C R cross-linking (previously demonstrated by Fu et al., 1998). To do this the Daudi B cell line was stimulated by cross-linking the B C R with anti mlgM antibodies for a set length of time (as indicated in the figures). Cells were then lysed and P L C y l and PLCy2 were immunoprecipitated from 1000 pg of whole cell lysate using PLCy specific antibodies. The immunoprecipitates were then resolved by SDS-PAGE and the resolved gels were transferred to nitrocellulose filters. The filters were then immunoblotted with a B L N K specific antibody to determine i f B L N K had co-immunoprecipitated with PLCy (Fig. 5.12). Subsequently the filters were striped and re-probed with PLCy specific antibodies to confirm that the  immunoprecipitation had  been  successful  (Fig. 5.12).  Furthermore, reciprocal  immunoprecipitation studies were performed with antibodies specific for B L N K (Fig. 5.13).  As can be seen in figure 5.12a, a very small amount of B L N K can be seen to coimmunoprecipitate with P L C y l in the Daudi B cell line.  However, this amount is barely  detectable and does not appear to significantly increase following B C R cross-linking (Fig. 5.12a). Similarly, only a very small amount of B L N K can be seen to co-immunoprecipitate with PLCy2 in the Daudi B cell line (Fig. 5.12b). And again, this amount is barely detectable and does not appear to significantly increase following B C R cross-linking (Fig. 5.12b). Furthermore, only a very small amount of PLCy2, while no P L C y l , appears to co-immunoprecipitate with B L N K in the reciprocal immunoprecipitation assays (Fig. 15.14 and 5.13, respectively). Thus, B L N K ' s association with P L C y l and PLCy2 appears to be very difficult to detect with this assay. While these findings appear to be contrary to what is reported in the literature (Fu et al., 1998), personal communications with Dr. T. Kurosaki (a co-author of the original paper reporting B L N K / P L C y 1/PLCy2 association) has confirmed that this association is indeed very difficult to detect. This is not to suggest that the association does not occur or that it is not significant, but rather that the association is difficult to detect, and as such, the apparently negative coimmunoprecipitation results within the AfT20-system should be interpreted with caution. 152  a.  Daudi  Anti mlgM -»  I- •  wl|  80 k D -  184 k D -  IP: P L C y l IB: B L N K (AC)  r. n  S&R:  PLCyl  Daudi Anti mlgM ->  \ -  +  |wl |  80 kD  IP: P L C y 2 IB: B L N K (H80)  184kD-  S&R:  PLCy2  Figure 5.12. BLNK Does Not Appear to Co-Immunoprecipitate with Either P L C y l or PLCy2 in the Daudi B Cell Line, (a) The B C R was not cross-linked in non-stimulated samples (-) and was cross-linked with anti mlgM antibodies at 37 °C for 5 minutes in stimulated samples (+). Cells were then lysed in TX-100 buffer and P L C y l was immunoprecipitated from 300 pg of whole cell lysate using Protein A-sepharose and a P L C y l specific antibody. Immunoprecipitates were resolved by S D S - P A G E . The resolved gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a B L N K (AC) specific antibody. Finally, filters were stripped and reprobed with a P L C y l specific antibody. IP indicates the immunoprecipitating antibody, IB indicates the immunoblotting antibody and S & R indicates the antibody that was used to reprobe the filter. 25 pg of non-immunoprecipitated, whole cell lysate was resolved in the lane labeled "wl". (b) Experiments were performed as described for (a) except that a PLCy2 specific antibody was used to immunoprecipitate PLCy2 from 3000 p_ of whole cell lysate, the B L N K (H80) specific antibody was used for the B L N K immunoblot and 50 pg of non-immunoprecipitated, whole cell lysate was resolved in the lane labeled "wl".  153  CO  "D  o  Q Anti mlgM ->  CQ +  | wl |  184 kD  IP: BLNK (2B11) IB: PLCyl  80 kD.  S&R:  Figure 5.13.  BLNK (AC)  P L C y l Does Not Appear to Co-Immunoprecipitate with B L N K in the Daudi B Cell Line. The  B C R was not cross-linked in non-stimulated samples (-) and was cross-linked with anti m l g M antibodies at 37 °C for 5 minutes in stimulated samples (+). Cells were then lysed in TX-100 buffer and B L N K was immunoprecipitated from 300 ug whole cell lysate using Protein A-sepharose and a mixture of two B L N K specific antibodies (2B11 and H80). Immunoprecipitates were resolved by S D S - P A G E . The resolved gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a P L C y l specific antibody. Finally, filters were stripped and reprobed with a mixture of two B L N K antibodies (2B11 and H80). IP indicates the immunoprecipitating antibody, IB indicates the immunoblotting antibody and S & R indicates the antibody that was used to reprobe the filter. 25 ug of non-immunoprecipitated, whole cell lysate from the B C R / S y k / B L N K cell line was resolved in the lane labeled "wl" as a control for the B L N K antibody as this cell line expresses a myc-tagged version of exogenous human B L N K that is slightly heavier than the endogenous B L N K isoforms expressed in the Daudi cell line (68 kD and 70 kD).  154  CM  o  CO  3 TO Q Anti  mlgM -> | -  o  m + | wl  184 kD • 80 k D .  3  IP: BLNK(2B11/H80) IB: PLCy2 S&R:  BLNK(2B11/H80)  Figure 5.14. PLCy2 Does Not Appear to Effectively Co-Immunoprecipitate with B L N K in the Daudi B Cell Line. The B C R was not cross-linked in non-stimulated samples (-) and was cross-linked with anti mlgM antibodies at 37 ° C for 90 seconds in stimulated samples (+). Cells were then lysed in NP-40 buffer and B L N K was immunoprecipitated from whole cell lysate from 2x 10 cells using Protein A-sepharose and a mixture of two B L N K specific antibodies (2B11 and H80). Immunoprecipitates were resolved by S D S - P A G E . The resolved gels were transferred to nitrocellulose filters that were subsequently immunoblotted with a P L C y l specific antibody. Finally, filters were stripped and reprobed with a mixture of two B L N K antibodies (2B11 and H80). IP indicates the immunoprecipitating antibody, IB indicates the immunoblotting antibody and S & R indicates the antibody that was used to reprobe the filter. 25 pg of non-immunoprecipitated, whole cell lysate from the BCR/Syk/BLNK/PLCy2 cell line was resolved in the lane labeled "wl" as a control for the PLCy2 antibody as this cell line expresses exogenous human PLCy2. 7  155  Given the above results, the immunoprecipitation assay does not appear to be optimized. In particular, this assay may not be able to detect protein associations i f the association levels are low, or i f the associations were relatively weak such that they can be disturbed by the detergents required by the experiment, or i f the associations occur in a manner that occlude the immunoprecipitating  antibody  from  binding to the  protein  of interest.  Thus,  the  immunoprecipitation protocol was adapted to address these possibilities including using more whole cell lysate per immunoprecipitation, using a variety of detergent conditions and using a variety of antibodies specific for different regions of the proteins of interest (summarized in Appendix II). Regardless of the approach, the majority of these studies were unable to detect a reproducible association between B L N K , B T K and/or PLCy (summarized in Appendix II) in either the AtT20 system or in B cell lines. Thus, the immunoprecipitation assays were deemed inconclusive and were not pursued further. Rather, investigations were performed to assay the compartmentalization status of B L N K , Syk, B T K and PLCy within the AfT20 system.  5.2.9  BCR-Induced Membrane Recruitment of Syk, BLNK, B T K and PLCy Is Not  Reconstituted in the AtT20-Derived Cell Lines  Despite their difficulties, the co-immunoprecipitation assays did not completely confound this thesis.  Rather, progress was made by investigating the compartmentalization status of the  various proteins. In particular, BCR-induced membrane recruitment of Syk, B L N K , B T K and PLCy was investigated as such recruitment is proposed to be necessary to BCR-induced PLCy phosphorylation and activation. Indeed, inappropriate compartmentalization could explain the apparently limited BCR-induced PLCy phosphorylation that was observed within this system. Moreover, because compartmentalization is generally dependent on regulated and specific protein associations, failed compartmentalization could also hint at failed protein associations.  To investigate membrane recruitment, the various cell lines were stimulated by cross-linking the B C R with anti mlgM antibodies for five minutes. The cells were then fractioned into membraneenriched fractions and cytosolic fractions (refer to Chapter 2.7 for details). The fractions were then resolved by SDS-PAGE and the resolved gels were transferred to nitrocellulose filters that were subsequently immunoblotted with B L N K , B T K , Syk, P L C y l , Iga or tubulin specific antibodies (Figure 5.15).  It should be noted that this approach is limited to distinguishing  whether a protein is localized to the cytosol and/or to cellular membranes. This approach can not 156  distinguish whether a protein is localized to the plasma membrane and/or to other cellular membranes.  Nonetheless, this approach can be suggestive as to whether or not a cytosolic  protein is being recruited to the membrane-fraction following B C R cross-linking.  As can be seen in Figure 5.15, Syk appears to be associated with both the cytosolic and membrane-enriched fractions in the AtT20-derived cell lines. Similar results are observed for B T K , B L N K and PLCy (Fig. 5.15).  However, while Syk and B T K appear to be equally  distributed between the cytosolic and membrane fractions, the vast majority of B L N K and P L C y l appear to be distributed in the cytosolic fraction (Fig. 5.15). Moreover, the distribution of the various components does not appear to change following B C R cross-linking (Fig. 5.15). Thus, Syk, B T K , B L N K and P L C y l do not appear to be recruited to the membrane-enriched fraction, following B C R cross-linking in this system. This suggests that these components may not be appropriately compartmentalized within this system and this may explain why coexpression of these components does not appear sufficient to reconstitute robust BCR-induced P L C y l phosphorylation (refer to Chapter 5.2.1 - 5.2.3).  157  H  DO  z  m z -1  o Fraction  Anti mlgM (min) -  m  m  _i to  >» CO  o  o GO  o m  2  o m  CO  z  H  m  00 m  m  Z  _J  m  to  oo  _l EQ  CO  CO  >  CO  o CO  o  o co.  >.  m  DO  m  m  m  |o 5|o 5 | 0 5|o 5 | 0 5 |o 5 | 0 5|o 5 |o 5 |o 5 | 0 5|o 5 | 0 5|fJ 5 |o 5 |o 5  IB: BLNK  IB: BTK  IB: Syk  IB: PLCyl  IB: Iga  ,»|  M|  « * |  i i i  l a  iP  •II  •«  IB: Tubulin  Figure 5.15. BCR-Induced Membrane Recruitment of Syk, B L N K , B T K and/or P L C y is not Apparent in AtT20 Derived Cell Lines. The B C R was cross-linked with anti mlgM antibodies at 37 ° C for 5 minutes. Cells were then fractionated according to the membrane-enrichment protocol (detailed in Chapter 2.7). 50 pg of each fraction was then resolved by SDS-PAGE. The resolved gels were then transferred to nitrocellulose filters that were subsequently immunoblotted with B L N K , B T K , Syk, P L C y l , Iga or Tubulin specific antibodies. Iga and Tubulin localization was used to confirm that the fractions were not cross-contaminated. Iga is a transmembrane protein and as such is expected to be apparent only in the membrane-enriched fraction whereas Tubulin is a cytosolic protein and as such is expected to be apparent only in the cytosolic fraction. IB indicates the immunoblotting antibody. C indicates the cytosolic fraction and M indicates the membrane-enriched fraction. Data are representative of three independent experiments.  158  5.3  5.3.1  Discussion  Recalling the Proposed Model of the BCR/PLCy Pathway  Recall the aforementioned model of the BCR/PLCy pathway. It is proposed that: 1. B C R cross-linking leads the B C R to translocate into lipid rafts. 2. Lipid raft translocation brings the BCRs into close proximity with active Lyn. 3. Active Lyn then phosphorylates the B C R on the Iga/p ITAMs. 4. The SFKs (including Lyn) are then recruited to the B C R by way of their SH2 domains binding to the phosphorylated Iga/p ITAMs 5. Such recruitment brings the SFK into close proximity with each other enabling them to trans-phosphorylate and activate each other as well as to further phosphorylate the Iga/p ITAMs, creating a positive feedback loop. 6. Syk is also recruited from the cytosol to plasma membrane/BCR by way of its SH2 domain binding to the phosphorylated Iga/p ITAMs. 7. Such recruitment brings Syk into close proximity with the active SFK such that they can phosphorylate and activate Syk. 8. B L N K is also recruited from the cytosol to the plasma membrane/BCR by way of its SH2 domain binding to phosphorylated Iga tyrosines that exist outside of the ITAMs. 9. Such recruitment brings B L N K into close proximity with active Syk such that Syk can phosphorylate B L N K . 10. B T K and PLCy are then recruited from the cytosol to the plasma membrane/BCR signaling complex by way of their SH2 domains binding to the BCR-associated, phosphorylated B L N K 11. Such recruitment brings PLCy into close proximity with B T K and Syk such that they can phosphorylate PLCy. 12. Such recruitment also brings PLCy into close proximity with its plasma membranebound substrate,  PIP2.  13. The combined phosphorylation and membrane-recruitment of PLCy is proposed to facilitate its activation such that it can then hydrolyze  PIP2  to produce the second  messengers, IP3 and D A G . 14. Ultimately, PLCy activation leads to increased IP3 and D A G levels, to increased intracellular calcium levels and to the activation of several downstream signaling 159  pathways including the M A P K pathway that leads to an increase in phosphorylated Erk levels.  Based on this model it was hypothesized that co-expression of the B C R , Syk, B L N K and B T K would be both necessary and sufficient to maximally reconstitute BCR-induced PLCy activation in the AfT20 system. Additionally, it was hypothesized that co-expression of the BCR, Syk and B L N K may be sufficient to at least partially reconstitute BCR-induced PLCy activation with the AtT20 system. This latter hypothesis was based on the observation that genetic ablation of B T K does not completely inhibit PLCy phosphorylation and/or activation (Takata and Kurosaki, 1996).  Now recall the two key findings in Chapter Four: 1. BCR, Syk and B L N K co-expression is sufficient to reconstitute BCR-induced PLCy activation in the AfT20 system (as determined by monitoring BCR-induced P L C dependent increases in Erk phosphorylation). 2. B C R , Syk, B L N K and B T K co-expression is not sufficient to reconstitute B C R induced PLCy activation in the AtT20 system (as determined by monitoring BCRinduced PLC-dependent increases in Erk phosphorylation).  While the former finding is as expected with respect to the initial hypothesis the latter is not. Thus, the AfT20 system was further analyzed in an attempt to determine how or why the results were diverging from the hypothesis, as such an understanding could reveal any shortcomings in the AtT20 system and/or the proposed model of the BCR/PLCy signaling pathway. In particular, the protein phosphorylation, association and compartmentalization status of Syk, B T K , B L N K and PLCy were investigated, as "defects" within these processes could explain a divergence from the hypothesis and/or model.  5.3.2  BCR-Induced PLCyl Phosphorylation in the AtT20 System  Before discussing the implications of the findings it may be helpful to first recap the key findings regarding PLCy phosphorylation and activation within this system. This is done below in Table 5.1.  160  Table 5.1. Summary of Key Findings Regarding BCR-Induced P L C y l Phosphorylation and Activation in the AtT20 System. Please note that the colour of the arrow is meant to roughly represent the intensity of the observed phosphorylation where the darker the colour the greater the phosphorylation (e.g. = very weak phosphorylation, t = weak phosphorylation, t = strong phosphorylation, t = very strong phosphorylation). Cell Line BCR  PLCy Phosphorylation  PLCy Activation  No phosphorylation  No activation  BCR/Syk  No activation  BCR/BTK  No activation  BCR/Syk/BTK  f  BCR/Syk/BLNK  A  1  No activation Activation No Activation  BCR/BTK/BLNK BCR/Syk/BTK/BLNK  t  No Activation  The first finding of note is the observation that co-expression of the B C R with Syk and/or B T K is sufficient to reconstitute BCR-induced P L C y l phosphorylation.  This finding is quite  surprising as, according to the model, B L N K is proposed to be necessary to couple these PTKs to PLCy. This finding suggests that these PTKs may be able to associate with PLCy via a B L N K independent  pathway.  If this is indeed the case, some residual BCR-induced PLCy  phosphorylation should be evident in B L N K knockout cells. Interestingly, while such a cell line exists the authors only reported on the status of BCR-induced PLCy activation, which was found to be completely ablated, and not on the status of PLCy phosphorylation (Ishiai et al., 1999). On the other hand, the observed independence from B L N K could be an artifact of Syk and B T K over-expression within the AtT20 system.  Importantly, co-expression of the B C R with Syk and/or B T K is sufficient to reconstitute only a minimal amount of BCR-induced PLCy phosphorylation and it is not sufficient to reconstitute BCR-induced PLCy activation.  This finding has several implications.  First, the minimal  phosphorylation may suggest that B L N K is necessary to efficiently couple these PTKs to PLCy. Second, the inability to activate PLCy may suggest either that PLCy is not being sufficiently phosphorylated to be active or that PLCy is being activated but not at level sufficient to be detected.  Alternatively, the inability to activate PLCy could suggest that phosphorylation in  itself is not sufficient to activate PLCy. Thus, we go back to considering the role of B L N K in this process.  Interestingly, BCR-induced PLCy phosphorylation is comparable in the BCR/Syk versus the B C R / S y k / B L N K cell line (Fig. 5.4b) and in the B C R / B T K versus the B C R / B T K / B L N K cell line 161  (Fig. 5.4a). This suggests that B L N K is not assisting Syk and B T K to phosphorylate PLCy in this system. This may suggest that B L N K is not appropriately associating with these proteins which in turn may suggest that B L N K is not becoming appropriately phosphorylated in this system or that an additional lymphoid component may be required for B L N K to associate with these proteins.  However, while B L N K co-expression does not appear sufficient to enhance  BCR-induced PLCy phosphorylation it does appear sufficient to reconstitute PLCy activation (Figs. 4.6b and 4.7b). This finding is intriguing as it suggests that B L N K may be involved in a component of PLCy activation that is independent from its phosphorylation. For example, it could be that B L N K is associating with PLCy and thereby localizing it such that it can act on its substrate. However, i f this is the case the question arises as to why B L N K may be associating with PLCy yet not facilitating its phosphorylation as predicted by the aforementioned model. One possible explanation could be that, contrary to the proposed model, B L N K only facilitates PLCy's phosphorylation by B T K and not by Syk. Thus, when the B C R , Syk and B L N K are coexpressed you may not see enhanced PLCy phosphorylation but you may see enhanced PLCy activation due to a change in the localization of PLCy due to its association with B L N K . Alternatively, when the B C R , B T K and B L N K are co-expressed you may not see enhanced PLCy phosphorylation or activation as B T K is not likely to associate with B L N K given that B L N K is not predicted to be phosphorylated in the absence of Syk expression. Initially, this explanation appears to challenge the existing model of the BCR/PLCy signaling. Indeed, it appears contradictory to publications such as those by Ishiai and colleagues (1999) that report that " B L N K is required for coupling Syk to PLCy".  However, upon further inspection, it is  evident that these reports only establish that B L N K is required to couple the B C R to PLCy activation while they do little to convince the reader that this requirement involves B L N K facilitating Syk's phosphorylation of PLCy.  Indeed, Ishiai (1999) and colleagues themselves  point out that PLCy may well directly associate with phosphorylated Syk via its C-terminal SH2 domain. Thus, B L N K ' s predominant roles in the BCR/PLCy pathway may be to appropriately localize PLCy to the plasma membrane/BCR signaling complex and to couple B T K to PLCy phosphorylation and thereby, facilitate PLCy's activation (refer to Chapter 7.4.2 for further discussion).  Two final findings worth noting in this section are:  162  1.  That co-expression of the B C R , Syk, B T K and B L N K appears to slightly enhance P L C y l phosphorylation as compared to co-expression of the B C R , Syk and B L N K alone or the BCR, B T K or B L N K alone.  2.  That co-expression of the BCR, Syk, B T K and B L N K is no longer sufficient to reconstitute BCR-induced PLCy activation whereas co-expression of the BCR, Syk and B L N K is sufficient.  These findings have three major implications. First, they support the proposed model in that coexpression of B L N K  with  Syk and B T K appears to facilitate BCR-induced  PLCy  phosphorylation. Second, they suggest that the phosphorylation status of PLCy is not necessarily indicative of its activation. This finding is not completely unexpected as PLCy phosphorylation has long been accepted as a required step in PLCy activation yet rejected as a hallmark of its activation. Rather, it has been suspected that other factors, such as specific protein associations and appropriate localization (i.e., compartmentalization), may also contribute to its activation. Third, they suggest that B T K is inhibiting rather than enhancing PLCy activation in this system. This is contrary to all expectations and suggests that the proposed model and/or the system is deficient in some manner.  As suggested in Chapter Four, B L N K and B T K may be coming  together in this system to form a non-functional signaling complex. And while this complex may be able to facilitate the phosphorylation of PLCy it may not be able to facilitate the correct localization and therefore, activation of PLCy. To address this possibility the protein association and compartmentalization status of PLCy was investigated. As well, it was hypothesized that PLCy activation may not be being maximally reconstituted within this system as BCR-induced PLCy phosphorylation appears greatly limited as compared to what is observed in B cells. Thus, the phosphorylation, protein association and compartmentalization statuses of B L N K , Syk and B T K were also further investigated to determine i f there where any defects in these processes that may be limiting PLCy phosphorylation and/or activation.  5.3.3  BCR-Induced Syk and BTK Phosphorylation in the AtT20 System  Syk and B T K must be tyrosine phosphorylated and activated i f they are to contribute to PLCy phosphorylation (Kurosaki et al, 1994; Zoller et al, 1997; and Afar et al, 1996; Rawlings et al, 1996; respectively). Thus, an inability to appropriately phosphorylate either Syk or B T K could be a contributing factor to the apparently limited ability to reconstitute BCR-induced PLCy 163  phosphorylation and/or activation within the AtT20 system. As such, the phosphorylation status of these components was investigated.  Previous studies have demonstrated that co-expression of the B C R and Syk is sufficient to reconstitute  BCR-induced Syk phosphorylation and activation within the AtT20 system  (Richards et al., 1996). Likewise, co-expression of the B C R and B T K appears sufficient to reconstitute B T K phosphorylation and activation within the AtT20 system (Fig. 5.5) . Moreover, 1  this phosphorylation appears to be constitutive, at least partly Fyn-dependent (Fig.5.7) and does not appear to be enhanced either by B C R cross-linking or by the co-expression of Syk and/or B L N K (Fig. 5.5). Thus, in terms of their respective kinase activities, Syk and B T K should be able to contribute to PLCy phosphorylation and activation in this system.  However, these  components may be unable to contribute to PLCy phosphorylation and/or activation i f they are unable to appropriately associate with the BCR/PLCy signaling complex.  As such, the  compartmentalization status of these components was investigated (refer to 5.3.4).  It is important to note that the apparently constitutive phosphorylation of B T K does not appear to be inhibiting its ability to phosphorylate PLCy within this system following B C R cross-linking. Nonetheless, the constitutive phosphorylation may explain the unusual observation that B T K appears to be negatively influencing the BCR-ind