UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Analysis of two members of the EXT family, EXT1 and EXT2 bearing ER-, Golgi-, and secretory vesicle-sorting… Lee, Tracy Phek Hui 2001

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

Item Metadata

Download

Media
831-ubc_2001-0446.pdf [ 9.49MB ]
Metadata
JSON: 831-1.0090120.json
JSON-LD: 831-1.0090120-ld.json
RDF/XML (Pretty): 831-1.0090120-rdf.xml
RDF/JSON: 831-1.0090120-rdf.json
Turtle: 831-1.0090120-turtle.txt
N-Triples: 831-1.0090120-rdf-ntriples.txt
Original Record: 831-1.0090120-source.json
Full Text
831-1.0090120-fulltext.txt
Citation
831-1.0090120.ris

Full Text

Analysis of Two Members of the E X T Family, EXT1 and EXT2 Bearing ER-, Golgi-, and Secretory Vesicle-Sorting Motifs by TRACY PHEK HUI LEE B.Sc. (Honours), The University of British Columbia, 1998 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES (Department of Microbiology and Immunology) We accept this thesis as CQirfb1:rmng~toTi^^ standard THE UNIVERSITY OF BRITISH COLUMBIA SEPTEMBER, 2001 copyright, Tracy Phek Hiii Lee, 2001 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department The University of British Columbia Vancouver, Canada DE-6 (2/88) A B S T R A C T EXT1 and EXT2 are type II transmembrane glycosyltransferases involved in the biosynthesis of heparan sulfate (HS) polysaccharide chains (McCormick et al., 2000). These two glycoproteins are found predominantly in the ER when overexpressed individually but form a Golgi-localized heterocomplex when co-transfected together (Kobayashi et al., 2000; McCormick et al., 2000). To examine the roles of the N-terminal domains of EXT 1 and EXT2, a series of chimeric cDNAs were constructed with ER-, Golgi- and secretory vesicle-sorting motifs. These constructs were transiently expressed in mammalian cells and tested for in vivo HS biosynthetic activity, heterocomplex formation, and subcellular localization. A l l EXT1 chimeras were able to synthesize HS in vivo as determined by their ability to restore herpes simplex virus infectivity to normally resistant HS-deficient Sog9 cells, though with varying efficiencies. Sec-EXTl and Sec-EXT2 were able to form a heterocomplex together despite the loss of their wild-type (wt) cytoplasmic N-termini, putative transmembrane domains and 10 aa of their stem regions, suggesting that the N-termini of EXT1 and EXT2 are not essential for heterocomplex formation. The steady state subcellular distribution of the chimeras was determined by indirect immunofluorescence microscopy. As specified by their sorting motifs, ER-EXT1 and ER-EXT2 distributed in staining patterns distinctive of the ER, and Sec-EXTl and Sec-EXT2 displayed staining patterns indicative of secretory vesicles. Though Golgi-EXTl and Golgi-EXT2 did result in perinuclear distribution, a majority of transfected cells exhibited staining patterns typically observed for ERGIC- or early Golgi-localization. Interestingly, when each of the EXT1 chimeric constructs was co-transfected with wt EXT2, both proteins co-localized together and exhibited a staining pattern characteristic of the Golgi regardless of their respective sorting signals. This suggests that when a wt anchoring partner is present, chimeric EXT Is are able to form a heterocomplex with wt EXT2 and localize to the Golgi. Taken together, the results presented in this thesis suggest that the major Golgi-localization mechanism used by EXT1 and EXT2 is hetero-oligomer formation as described in the kin recognition model (Nilsson et al., 1993) which hypothesizes that Golgi-resident proteins form complexes too large to enter transport vesicles and are thus retained in the Golgi. iv T A B L E OF CONTENTS Abstract ii Table of Contents iv List of Tables vi List of Figures vii List of Abbreviations viii Acknowledgements ix CHAPTER 1 Introduction 1 1.1 Hereditary Multiple Exostoses (HME) 1 1.2 EXT1 andEXT2 2 1.3 Heparan Sulfate Proteoglycans 7 1.4 Herpes Simplex Virus Type 1 & Type 2 11 1.5 Secretory Pathway 14 1.6 Subcellular Localization 17 1.6.1 ER-localization 17 1.6.2 Golgi-localization ; 18 1.6.2.1 Bilayer thickness model 20 1.6.2.2 Kin recognition model 22 1.6.3 Secretory Vesicles '. 24 1.7 Research Objectives 24 CHAPTER 2 Materials and Methods 26 2.1 Cell Culture 26 2.2 EXT Constructs 26 2.2.1 EXTmyc constructs 26 2.2.1.1 Sec-EXTlmyc 26 2.2.1.2 Sec-EXT2myc 27 2.2.1.3 ER-EXTlmyc 27 2.2.1.4 ER-EXT2myc 28 2.2.1.5 Golgi-EXTlmyc 28 2.2.1.6 Golgi-EXT2myc 29 2.2.2 EXTlgfp constructs 30 2.2.2.1 Sec-EXTl gfp 30 2.2.2.2 ER-EXTlgfp 30 2.2.2.3 Golgi-EXTlgfp 30 2.3 Transient Transfections 31 V 2.4 Immunofluorescence staining 31 2.5 Immunoblotting 32 2.6 Immunoprecipitation 33 2.7 HSV Assay 33 2.8 Sandwich Enzyme-linked Immunosorbent Assay (ELISA) 34 CHAPTER 3 Results 35 3.1 EXT constructs 35 3.2 Transient expression of EXT constructs in mammalian cells 38 3.3 Golgi-EXTl, ER-EXT1 and Sec-EXTl were all able to restore HSV susceptibility in Sog9 cells 40 3.4 Subcellular localization patterns of mono-transfected EXT constructs 44 3.5 Co-transfection of chimeric EXT l s with wild-type (wt) EXT2 46 3.5.1 Immunofluorescence analysis of co-transfected chimeric EXT1 with wild-type EXT2 46 3.5.2 A l l chimeric EXT l s were able to form a heterocomplex with wild-type EXT2 49 3.6 Co-transfection of chimeric EXT1 with chimeric EXT2 51 3.7 No other proteins involved in the EXT1/EXT2 heterocomplex were detected when either EXT1 or EXT2 were immunoprecipitated 54 3.8 Co-transfection of chimeric EXT1 with chimeric EXT1 56 CHAPTER 4 Discussion 58 4.1 Subcellular localization of chimeric constructs 58 4.2 In vivo heparan sulfate activity 60 4.3 Heterocomplex formation 62 4.4 Involvement of other proteins 62 4.5 Model of Golgi retention 64 4.6 Future directions 68 References 69 ( vi LIST OF TABLES Table I Heparan sulfate-ligand interactions 10 Table II Microbial pathogens that bind to heparan sulfate 10 ) Vll LIST OF FIGURES Figure 1 EXT 1 and EXT2 structure 3 Figure 2 Heparan sulfate polysaccharide chain 8 Figure 3 Secretory pathway 16 Figure 4 Bilayer thickness model 21 Figure 5 Kin recognition model 23 Figure 6 Chimeric EXT constructs 37 Figure 7 Western blot analysis of EXT constructs 39 Figure 8 HSV assay of EXT2myc constructs in Sog9 cells 42 Figure 9 HSV assay of EXTlgfp constructs in Sog9 cells 43 Figure 10 Subcellular localization patterns of mono-transfected chimeric EXT proteins 45 Figure 11 Subcellular localization patterns of co-transfected chimeric EXT1 with wild-type EXT2 48 Figure 12 Heterocomplex analysis of chimeric EXT1 with wild-type EXT2 50 Figure 13 Subcellular localization patterns of co-transfected chimeric EXT1 with chimeric EXT2 53 Figure 14 Radiolabeled immunoprecipitation analysis of EXT1-EXT2 heterocomplex formation 55 Figure 15 Subcellular localization patterns of co-transfected chimeric EXT1 and chimeric EXT1 57 Figure 16 Model of EXT 1 and EXT2 Golgi-localization mechanism 67 List of Abbreviations aa amino acid BSA bovine serum albumin COPI coat protein I COPII coat protein U ER endoplasmic reticulum ERGIC ER-Golgi intermediate compartment G A G glycoaminoglycan GlcA D-glucuronic acid GlcA-T D-glucuronic acid transferase GlcNAc N-acetylglucosamine GlcNAc-T N-acetylglucosamine transferase HME hereditary multiple exostoses HPLC high performance liquid chromatography HS heparan sulfate HSPG heparan sulfate proteoglycan HSV-1 herpes simplex virus type 1 HSV-2 herpes simplex virus type 2 Ii invariant chain Iip31 invariant chain protein isoform 31 Iip33 invariant chain protein isoform 33 LOH loss of heterozygosity M.O.I. multiplicity of infection NDST N-deacetylase/N-sulfotransferase PBS phosphate buffered saline PCR polymerase chain reaction PenStrep penicillin-streptomycin pfu plaque forming unit P M plasma membrane TBS Tris buffered saline TGN trans-Golgi network TMB 3,3' ,5,5' -tetramethylbenzidine wt wild-type r Acknowledgements I would like to thank my supervisor, Dr. Frank Tufaro for his guidance, patience and support throughout the course of my graduate studies. I would also like to thank the past and present members of the Tufaro lab, Peter Cheung, Dr. Gillian Duncan, Dr. Craig McCormick, Dong Qiang and Dr. Sonia Yeung for their help and encouragement. Many thanks to my committee members, Dr. Tom Beatty and Dr. Francois Jean for their advice and input. Finally, I would like to thank my parents and brother for all their love and support. 1 .1. Introduction 1.1 Hereditary Multiple Exostoses (HME) Hereditary multiple exostoses (HME) is one of the most common hereditary skeletal dysplasias in humans with an estimated prevalence of 1:50,000 (Schmale et al, 1994) though this increases to 1:1,000 among the Chamorros of Guam (Krooth et al, 1961). It is characterized by the formation of multiple benign cartilage-capped tumours (exostoses) at the growth plate of the epiphyses of long bones (Solomon, 1963). These exostoses are present or develop shortly after birth and continue to grow into puberty. They can cause a variety of orthopedic deformities such as limb length discrepancies and short stature. Most exostoses are non-problematic but surgical removal of exostoses may be required in instances where the exostoses are pressing on nerves or surrounding soft tissues (Paik et al, 2000). In severe presentation of the disorder, limb length equalization procedures may also be necessary (Dahl, 1993). Another serious complication of HME occurs in 0.5-2% of HME patients in which the benign tumours undergo malignant transformation into chondrosarcomas (Hennekam, 1991). Linkage studies show that mutations in either EXT1 on chromosome 8q24.1 (Cook et al., 1993) or EXT2 on chromosome 1 l p l l-pl2 (Wu et al., 1994) are responsible for the majority of HME cases though EXT3 on chromosome 19p (Le Merrer et al., 1994) has also been implicated in the disease as well. 2 1.2 EXT1 and EXT2 The human and mouse EXT1 and EXT2 genes have been cloned and sequenced (Ahn et al., 1995; Lin and Wells, 1997; Stickens et al, 1996; Stickens and Evans, 1997) as have the Drosophila melanogaster homologues of EXT1 and EXT2 (Bellaiche et al., 1998; The et al., 1999), and the Caenorhabditis elegans (C. elegans) homologue of EXT2 (Clines et al., 1997). Amino acid (aa) comparisons between the human, mouse and Drosophila EXT1 homologues show the greatest similarity in the C-terminal half of the protein (Cheung et al., 2001). Human, mouse and C. elegans homologues of EXT2 also show the C-terminal end of the aa sequence as being the most conserved (Clines et al., 1997). The finding that the C-terminal domains of EXT1 and EXT2 is the most conserved part of both proteins implies that these two domains are important, and in agreement with this, all etiological missense mutations identified to date are localized in the C-termini of EXT 1 and EXT2 (Philippe et al, 1997; Wuyts and Van Hul, 2000). EXT1 and EXT2 are expressed ubiquitously (Ahn et al., 1995; Stickens et al, 1996) and encode the exostosin-1 (EXT1) and exostosin-2 (EXT2) proteins, respectively (Fig. 1 A). Further analysis of these two proteins revealed that they have a type II transmembrane configuration (Fig. IB), with a cytoplasmic N-terminus, a single putative transmembrane domain, a stem region with little/no predicted secondary structure followed by a large lumenal C-terminus (McCormick et al., 2000). They are 31% identical (Wuyts et al., 1996) with EXT1 predicted to encode a 746 aa protein, and EXT2, a 718 aa protein. 3 A E X T l amino acid sequence 1 8 27 M Q A K K R Y F i L L S A G S C t A L L F Y F G G I Q F R A S R S H S R R E E H S G R N G L H H P S P D H F W P R F P E P L R P F V P 100 R M E S C F D F T L C K K N G F K V Y V Y P Q Q K G E K I A E S Y Q N I L A A I E G S R F Y T S D P S Q A C L F V L S L D T L D R D Q L S P Q Y V H N L R S K V Q S L H L W N N G R N H L I F N L Y S G T W P D Y T E D V G F D I G Q A M L A K A S I S T E N F R P N F D V S I P L F S K D H P R T G G E R G F L K F N T I P P L R K Y M L V F K G K R Y L T G I G S D T R N A L Y H V H N G E D W L L T T C K H G K D W Q K H K D S R C D R D N T E Y E K Y D Y R E M L H N A T F C L V P R G R R L G S F R F L E A L Q A A C V P V M L S N G W E L P F S E V I N W N Q A A V I G D E R L L L Q I P S T I R S I H Q D K I L A L R Q Q T Q F L W E A Y F S S V E K I V L T T L E I I Q D R I F K H I S R N S L I W N K H P G G L F V L P Q Y S S Y L G D F P Y Y Y A N L G L K P P S K F T A V I H A V T P L V S Q S Q P V L K L L V A A A K S Q Y C A Q I I V L W N C D K P L P A K H R W P A T A V P V W I E G E S K V M S S R F L P Y D N I I T D A V L S L D E D T V L S T T E V D F A F T V W Q S F P E R I V G Y P A R S H F W D N S K E R W G Y T S K W T N D Y S M V L T G A A I Y H K Y Y H Y L Y S H Y L P A S L K N M V D Q L A N C E DILM N F L V S A V T K L P P I K V T Q K K Q Y K E T M M G Q T S R A S R W A D P D H F A Q R Q S C M N T F A S W F G Y M P L I H S Q M R L D P V L F K D Q V S I L R K K Y R D I E R L EXT2 amino acid sequence 1 25 44 M C A S V K S N I R G P A L I P R M K T K H R I Y Y V T L F S I V L L G L I A T G M F Q F W P H S I E S S S D G G V E K R T I R E V P W R L P T D S P I P E R G D L S C R M H T C F D V Y R C G F N P K 104 N K I K V Y I Y P L K K Y V D D A G V P V S S A I S R E Y N E L L T A I S D S D Y Y T D D I N R A C L F V P S I D W N Q N P L R I K E T A Q A L A Q L S R W D R G T N H L L F N M L P G A P P D Y N T A L D V P R D R A L L A G G G F S T W T Y R Q G Y D V S I P V F S P L S A E M A L P E K A P G P R R Y F L L S S Q M A I H P E Y R E E L E A L Q A K H Q E S V L V L D K C T N L S E G V L S V R K R C H Q H Q V F D Y P Q V L Q E A T F C T V L R R A R L G Q A V L S D V L Q A G C V P W I A D S Y I L P F S E I L D W K K A S N A A / P E E K M S D W S I L Q N I P Q R Q I E E M H R Q A R W F W E A Y F Q S I K A I A L A T L Q I I N D R I Y P Y A A I S Y E E W N D P P A V K W A S V S N P L F L P L I P P Q S Q G F T A I V L T Y D R V E S L F R V I T E V S K V P S L S K L L W W N N Q N K N P P E E S L W P K I R V P L K W R T A E N K L S N R F F P Y D E I E T E A V L A I D D D I I M L T S D E L Q F G Y E V W R E F P D R L V G Y P G R L H L W D H E M N K W r O r - E S E V \ n " N E V S M V L T G A A F Y H K Y F N Y L Y T Y K M P G D I K N W V D A H M N C E D I A M N F L V A N V T G K A V I K V T P R K K F K C P E C T A I D G L S L D Q T H M V E R S E C I N K F A S V F G T M P L K W E H R A D P V L Y K D D F P E K L K S F P N I G S L letters in bold denote the cytoplasmic N-terminal domain letters shaded in gray denote the putative transmembrane domain letters underlined denote the predicted stem region numbers denote amino acid residues B E X T l (746 aa) cytoplasmic N-terminal domain EXT2 (718 aa) cytoplasmic N-terminal domain cytoplasm 100 putative GlcA-T catalytic domain \ gg^Kgtem/* lurnmal C-terminal domain C-terminal domain lumen ER membrane numbers in black denote amino acid residue Figure 1 EXT l and EXT2. A) The amino acid (aa) sequence of E X T l and E X T 2 are presented. Important domains are indicated by different colored letters. Letters in bold indicate the cytoplasmic N-terminal domain while letters shaded in gray indicate the putative transmembrane domain. Finally, the letters underlined indicate the stem region (area of little/no secondary structure) as predicted by J-PRED (a secondary structure prediction server). B) E X T l and E X T 2 are both type II transmembrane (TM) proteins with a short N-terminal cytoplasmic domain, a single hydrophobic TM-spanning domain followed by a large C-terminal domain containing the stem region and catalytic domain(s). Amino acid residue numbers indicate the start of the different putative domains. 5 Loss of heterozygosity (LOH) studies using polymorphic markers linked to EXTl and EXTl indicate that these loci are lost in the chondrosarcomas of HME patients, suggesting these two genes encode proteins with tumour suppressor functions (Hecht et al, 1995; Raskind et al., 1995). But their function was unknown until recently when E X T l was demonstrated to be involved in in vivo heparan sulfate (HS) biosynthesis (McCormick et al., 1998). Subsequently, both E X T l and EXT2 were discovered to harbour N-acetylglucosamine transferase (GlcNAc-T) and D-glucuronic acid transferase (GlcA-T) activities in vitro (Lind et al., 1998). Remarkably, E X T l and EXT2 were shown to form a hetero-oligomeric complex in vivo and that this complex possesses substantially higher glycosyltransferase activity in vitro than either E X T l or EXT2 alone (McCormick et al., 2000). This finding has since been confirmed by studies in yeast, which do not exhibit endogenous HS polymerase activity (Senay et al, 2000). Interestingly, both E X T l and EXT2 are found predominantly in the ER when overexpressed alone but re-localize to the Golgi when co-expressed in the same cell (Kobayashi et al., 2000; McCormick et ai, 2000). EXTl and EXT2 are members of the EXT family of putative tumor suppressors which also currently includes three EXT-like genes, EXTL1 (Wise et al., 1997), EXTL2 (Wuyts et al, 1997), and EXTL3 (Van Hul et al, 1998). These three EXTL genes encode proteins shown to possess GlcNAc-T in vitro activity as well (Kim et al., 2001; Kitagawa et al., 1999), suggesting that the members of the EXT family may all be involved in polysaccharide chain elongation. Most HME cases have been attributed to missense or frameshift mutations in either EXTl or EXTl (Philippe et al, 1997; Wuyts and Van Hul, 2000), and until recently, it was believed that HME was due to deficient HS polymerase activity. However, recent data have shown that some etiological E X T l mutants still possess HS enzymatic activity as determined by the HSV assay and high performance liquid chromatography (HPLC) analysis of cell surface glycosaminoglycans (GAGs) (Cheung et al., 2001). This implies that other proteins may be involved in the disorder or that E X T l and E X T 2 encode another function yet undescribed. It is also possible that the etiological mutants produce abnormal levels/types of HS which are undetectable in our lab assays, and that adversely affect the controlled signalling required for proper proliferation, differentiation and apoptosis of chondrocytes. 7 1.3 H e p a r a n Sulfate Proteoglycans Heparan sulfate proteoglycans (HSPGs) are found ubiquitously on cell surfaces and in the extracellular matrix. These complex molecules of which there are three major families, the syndecans, the glypicans and the perlecans consist of a protein core covalently attached via a tetrasaccharide linker to HS chains. Perlecans are components of cartilage and basement membranes, and they, along with the transmembrane syndecans and the glycosylphosphoinositide-linked glypicans, mediate a variety of biological processes through the binding of their HS chains to different cellular ligands. HS is a member of the GAG family of unbranched polysaccharide chains composed of repeating disaccharide units. Depending on the disaccharide composition, GAGs can be grouped into three different groups: chondroitin/dermatan sulfate, heparin/heparan sulfate and keratan sulfate. The biosynthesis of HS chains occurs in the Golgi apparatus with the initial attachment of a tetrasaccharide linker region (xylose, galactose, galactose and glucuronic acid) to a serine residue of a core protein (Fig. 2). The EXTL2 protein is then believed to add the first N-acetylglucosamine (GlcNAc) to the linker to start the HS chain (Kitagawa et al., 1999). Repeating disaccharide units of D-glucuronic acid (GlcA) and GlcNAc are then added on by the copolymerase E X T l and EXT2 (Lind et al., 1998; McCormick et al, 2000; McCormick et al, 1998). Furthermore, recent data indicate that the other two members of the EXTL proteins, EXTL1 and EXTL3 harbour in vitro GlcNAc-T activity likely involved in the synthesis of HS chains (Kim et al, 2001). The HS polysaccharide chains are modified by N-deacetylase-N-sulfotransferases (NDSTs), 6-O-sulfotransferases, 3-O-sulfotransferases, 2-O-sulfotransferases and epimerases to create highly variable and complex sulfated structures. X "3H B o o o S-l H t—1 CD T3 cD o o CD y/1 CD o CD o cd 00 CD cd CO O O _ 3 00 o X ~ a -e. s .g a O ' f H <D CD CD X ! co •a c CD T3 CD co O H r-fi. « <D B >-. N c _ 0 — c i c3 X o o S- i CD M fi X ! o o cd PS CD « t^ 'Ti G *3 cd ' > T3 CD X ! I .5 cd S •a X ! O Xi TZt <D N •c CD B >> "o cD S- l cd X ! o xi s$ o <c o o -a o ci 8 rt c u < CD o X ! rs o fi o CD 3 T3 <D X ! H CD cd X i o W X CD X5 H-> 2 a o o < o O CU "2 "C 03 CJ (O Cw 3 un S CS J-a cu sc CU J-•e Q .2 _o *c» TJ T 3 cd CD •a CD SJ "cd cd o IS 8.12 xV 00 c o •£S co rt 13 cD ra OH >> CD cd CD X ! 3 2 oo ^ cd O oo <zt o c r—I H 8 9 These remarkably diverse HS molecules interact with a variety of different proteins to regulate a diverse range of cellular functions. They have been shown to mediate cell-cell adhesion, serve as co-receptors for soluble factors, mediate protein internalization, and effect conformational change thus enhancing enzymatic activity (Table I). Furthermore, cell surface HSPGs are also exploited by microbial pathogens from bacteria to viruses as ligands for attachment to host cells (Table II). HSPGs are also shed from the cell surface thus they can further regulate cell surface HSPG-ligand interactions by competing for available ligands in the extracellular matrix. 10 Table I Examples of heparan sulfate-ligand interactions. Protein ligand Effect of binding HS Fibroblast growth factor-2 Mediate interaction with fibroblast growth factor-2 receptor (Yayon et al, 1991, Rapraeger, 1991 #23) Lipoprotein lipase Internalization (Fuki et al., 1997; Saxena et al, 1990) Laminin Adhesion (Utani et al, 2001) Antithrombin III Induce conformational change thus increasing its activity (Olson et al, 1992) Table II Microbial pathogens that bind to heparan sulfate. Microbial pathogen Protein ligand Herpes simplex virus glycoprotein C (Herold et al, 1991) Listeria monocytogenes ActA (Alvarez-Dominguez et al, 1997) Bordetella pertussis major subunit of fimbriae (Geuijen et al, 1996) Pseudorabies virus glycoprotein C (Karger and Mettenleiter, 1996) 11 With all these regulatory functions it is not surprising that mutating or knocking out HS biosynthetic enzymes in Drosophila or mice would produce dramatic in vivo effects. Mutations in tout velu (ttv) which encodes the Drosophila homologue of EXTl result in defects in wing anterior-posterior patterning (Bellaiche et al., 1998) while homozygous ttv mutants die during the pupal stage (The et al., 1999). Sugarless (sgl), also a Drosophila gene, encodes a protein homologous to mammalian UDP-glucose dehydrogenase which generates UDP-GlcA for HS chain polymerization. Mutations in sgl result in abnormal segmental patterning of the embryonic epidermis as it affects the signalling of Wingless, a secreted heparin-binding morphogen involved in patterning during embryogenesis (Binari et al., 1997; Hacker et al., 1997; Haerry et al., 1997). Further evidence of HS's regulatory role in development can also be found in mammals. EXTl homozygous mutants in mice die during the embryonic stage due to defects in mesoderm and extraembryonic tissue formation (Lin et al., 2000). Furthermore, mice homozygous for a gene trap mutation that disrupts the HS 2-O-sulfotransferase gene exhibit kidney agenesis, abnormal eye and skeletal growth, and die in the neonatal stage (Bullock et al., 1998). A l l these in vivo data clearly show that HS plays an important role in controlling embryogenesis and is essential for proper development in invertebrates and vertebrates. 1.4 Herpes Simplex Virus Type 1 & Type 2 Herpesviruses are eukaryotic viruses responsible for several important human and animal diseases. A l l herpesviruses have a linear double-stranded DNA genome, icosahedral capsid, tegument (amorphous proteinaceous layer surrounding the capsid), and a lipid envelope containing numerous viral glycoproteins involved in attachment and entry. 12 The family Herpesviridae is divided into three subfamilies of alpha-, beta-, and gammaherpesvirinae based on biological characteristics and genomic analysis. Though herpesviruses have been isolated from a wide range of vertebrates there is a high degree of host specificity with eight human herpesviruses identified to date. They are the herpes simplex virus type 1 (HSV-1), herpes simplex virus type 2 (HSV-2), and varicella-zoster virus of the alphaherpesvirinae, human cytomegalovirus, human herpesvirus-6 and human herpesvirus-7 of the beta subfamily, and Epstein-Barr virus and human herpesvirus-8 of the gammaherpesvirinae. The complete sequence of the 152.3 kb genome of HSV-1 was published in 1988 (McGeoch et al., 1988) while HSV-2's 154.7 kb genome sequence was only published recently (Dolan et al., 1998). The two genomes are closely related to each other with an 83% nucleotide sequence homology (Dolan et al., 1998). The entry of HSV into host cells is a multistep process beginning with the initial attachment of HSV to cell surface GAGs followed by a stable attachment between HSV and secondary receptors. The final step in the process is the fusion of the viral envelope with the PM of the host cell. The initial attachment of both HSV-1 and HSV-2 to the host cell surface is primarily mediated by the binding of viral envelope glycoproteins (gB & gC) to HS which serves as the host-cell primary receptor (Herold et al., 1994; Herold et al., 1991; Williams and Straus, 1997). The absence of cell surface HS GAGs due to either heparanase digestion or defects in HS-biosynthesis (i.e. mutant cell lines) results in dramatically reduced infectivity by HSV of these host cells (Banfield etal., 1995; WuDunn and Spear, 1989). Stable attachment of HSV to host cells is mediated by the binding of gD to one of several cell surface receptors. Herpes virus entry mediator (Hve) A, a member of 13 the tumour necrosis receptor family, and HveC, a member of the immunoglobulin (Ig) family can mediate stable attachment between the gD of both HSV-1 and HSV-2 to the host cell (Geraghty et al, 1998; Montgomery et al, 1996). In constrast, HveB, also a member of the Ig family can only mediate adhesion of HSV-2 and some laboratory strains of HSV-1 (Warner et al, 1998). Finally, viral penetration of the host cell involves the viral glycoproteins gB, gH-gL required for the fusion of the viral envelope with the cell P M (Cai et al, 1988; Ful ler^ al, 1989; Roop et al, 1993). Although both HSV-1 and HSV-2 use HS as a receptor for initial attachment, and HveA and HveC for stable attachment, the two serotypes differ epidemiologically with HSV-1 mainly causing oral lesions and HSV-2 primarily associated with genital lesions. However, the results of a recent study suggests that HSV-1 is increasingly responsible for genital infections (Nilsen and Myrmel, 2000). Differences in the ability of HSV-2 to bind HveB which most strains of HSV-1 are unable to use may help explain some of the epidemiological differences seen between the two serotypes. Additionally, HS GAGs are very heterogeneous with varying degrees of N-acetylation, N-sulfation, O-sulfation and epimerization of glucuronic acid to iduronic acid. Thus another possibility is that differences in the interaction with these chain modifications between HSV-1 and HSV-2 may further account for the differences seen between the two serotypes. Indeed, when the addition of O-desulfated heparins on HSV-1 and HSV-2 infection was examined, it was found that HSV-2 infection was inhibited while there was little or no inhibitory effects on HSV-1 infection (Herold et al, 1996). Furthermore, Herold et al. found that this phenotype was due to glycoprotein C of HSV-2 (gC-2) and that when gC-2 was transferred to an HSV-1 background this led to the mutant HSV-1 being inhibited by O-desulfated heparins. 14 Thus, the expression level of HveB or the degree of O-sulfation in various cell types may partly account for the different cell trophism seen between these two serotypes. 1.5 Secretory Pathway E X T l and EXT2 must traverse the secretory pathway to reach the Golgi where they polymerize the polysaccharide chains of HS (Fernandez and Warren, 1998). Eventually these glycosyltransferases are secreted out of the cell where their truncated forms can be isolated from serum (Lind et al., 1998). Intracellular trafficking and eventual secretion of proteins is one of the most complex biological phenomena in the field of biology. The transport pathway of proteins destined for secretion or the plasma membrane (PM) originate in the lumen of the rough ER and then pass through the ER-Golgi intermediate compartment (ERGIC) to the Golgi apparatus before being packaged into secretory vesicles (Fig. 3). Movement of proteins from the ER to the cell surface is mediated by vesicles which bud from the donor membrane then dock and fuse with the acceptor membrane thus delivering their cargo to the targetted location (Nickel et al., 1998). Transport vesicles examined to date have been shown to contain cytosolic coat proteins such as clathrin, coat protein I (COPI) and coat protein II (COPII). Clathrin-coated vesicles contain clathrin attached to the donor membrane via adaptor proteins and are involved in transporting proteins between the PM and the trans-Go\g'\ network (TGN) (Pearse and Robinson, 1990). COPII-coated vesicles are non-clathrin-coated vesicles which bud from the ER and are involved in the forward movement (anterograde transport) of proteins to the ERGIC (Aridor et al., 1995; Barlowe et al., 1994). COPI-coated vesicles are also non-clathrin-coated but their role in the 15 transport pathway is less clear. They have been shown to be involved in anterograde transport from the ER-Golgi intermediate compartment (ERGIC) to the Golgi as well as in intracisternal protein transport through the Golgi (Aridor et al., 1995; Orci et al, 1986). Furthermore, COPI vesicles have also been implicated in the retrieval of escaped ER resident proteins back from the Golgi as COPI protein has been demonstrated to bind the C-terminal Lys-Lys-Xaa-Xaa motif which serves as an ER-retrieval signal in the cytoplasmic tail of type I transmembrane proteins (Cosson and Letourneur, 1994; Teasdale and Jackson, 1996). Sorting of proteins within the secretory pathway is accomplished by the selective packaging of cargo molecules into coated vesicles (Gu et al., 2001). These transport vesicles then bud and fuse with the target membranes. Targetting proteins used by transport vesicles include members of the SNAp-REceptor (SNARE) family which are also required for membrane fusion between the vesicle and the target organelle (Sollner et al., 1993). 16 PM other transport vesicles o O ^ \^ COPI vesicles ERGIC COPII vesicles ER Figure 3 Secretory pathway. Proteins ctestined for secretion or the plasma membrane (PM) travel through the transport pathway to reach the cell surface. Proteins translated in the endoplasmic reticulum (ER) are packaged into COPII vesicles which carry them to the ER-Golgi intermediate (ERGIC). From there, they travel via COPI vesicles to the cis-Golgi network (CGN) and through the Golgi stacks to the trans-Golgi network (TGN). Once in the TGN, they are packaged into secretory/exocytic vesicles destined for the cell surface. Other transport vesicles can also be found cycling between the TGN and the PM. Escaped proteins may be retrieved back to their compartments by COPI vesicles. 17 1.6 Subcellular Localization A typical eukaryotic cell has a diverse number of organelles and compartments each containing its own unique set of proteins. Thus the sorting of proteins to the appropriate location within the cell is vital to its function. This is usually accomplished by signal sequences encoded in the gene. 1.6.1 ER-localization The ER is a network of membranes extending throughout the cytoplasm of a cell (Solomon et al., 1993). The ER is composed of two morphologically distinct regions called the rough ER (RER) and the smooth ER (SER). The RER is a series of parallel flattened sacs characterized by the presence of ribosomes on the outer surface (cytosolic side) of its membrane. The SER on the other hand does not have ribosomes attached to its membranes and resembles a branched tubular network rather than flattened sacs. Many ER resident proteins or those destined for compartments along the secretory pathway are synthesized on the ribosomes of the RER and translocated into the lumen of the ER. ER resident proteins contain ER localization motifs that allow them to be maintained in the ER either by retention or retrieval via a retrograde transport flow from the Golgi back to the ER by COPI vesicles (Teasdale and Jackson, 1996). Human invariant chain (Ii) is a type II transmembrane protein involved in the assembly of the major histocompatibility complex class II in the ER (Janeway Jr. and Travers, 1996). Due to alternative initiator methionines in the Ii mRNA, 2 different protein products are produced; Iip31 and Iip33 (Strubin et al., 1986). Iip31 is rapidly transported out of the ER while Iip33 is maintained in the ER (Lotteau et al., 1990). The transplantation of the cytoplasmic N-terminus of the ER-localized isoform onto N-18 acetylglucosaminyltransferase (GlcNAc-T I) relocalized it from the medial Golgi to the ER (Nilsson et al., 1994) indicating that the N-terminus of Iip33 contained an ER-localization signal. In agreement with this, a double arginine motif in close proximity to the N-terminus of Iip33 was identified as an ER-localization signal sufficient to localize a chimeric transferrin receptor (a type II cell surface membrane protein) to the ER (Schutze et al., 1994). 1.6.2 Golgi Localization The Golgi apparatus consists of a stack of flattened membraneous sacs and was first described by Camillo Golgi in 1898 (Solomon et al., 1993). The Golgi serves as a site of post-translational modifications as well as a sorting point for proteins and lipids to different organelles (Gu et al., 2001). These two functions are performed by various Golgi resident proteins, some of which are the glycosyltransferases responsible for the addition of oligosaccharide chains onto nascent proteins to produce glycoproteins. A l l Golgi glycosyltransferases cloned to date are type II transmembrane proteins with a short cytoplasmic amino-terminus, a single transmembrane domain, a stem region of little secondary structure followed by a large carboxy-terminus in the lumen of the Golgi apparatus (Colley, 1997). There is very little sequence homology among the different glycosyltransferases which suggest that the Golgi localization signal used by these enzymes must involve structural/conformational features instead of a specific amino acid sequence. There are currently two main models to explain how Golgi residents are retained within the Golgi: the bilayer thickness model (Bretscher and Munro, 1993) and the kin-recognition model (Nilsson et al., 1993). Though both models are supported by experimental evidences, neither model alone can account for how all Golgi-resident glycosyltransferases retained within the Gol 20 1.6.2.1 Bilayer thickness model Differences in the length of the transmembrane domain of Golgi proteins (shorter) and plasma membrane proteins (longer), and the finding that increasing the amount of cholesterol resulted in increased membrane thickness (Nezil and Bloom, 1992) led to the proposal of the bilayer thickness model by Bretscher and Munro in 1993. This model (Fig. 4) postulates that the shorter length of the transmembrane domain of Golgi-resident proteins excludes the Golgi proteins from entering cholestrol-rich transport vesicles destined for the PM (Bretscher and Munro, 1993). The membrane spanning domain of Golgi residents have been shown to be sufficient in localizing reporter molecules to the Golgi (Munro, 1991; Nilsson et al, 1991). In addition, Munro demonstrated that the amino acids of the transmembrane domain of Golgi residents could be altered without changing the Golgi localization, but that extending the transmembrane domain length resulted in PM localization (Munro, 1995). f\)X— vAA-SD > t-H O CD *-i CD o «D c a 0> CD • 5 .a <*H » © -g o O O Cl> s-B CD E 00 J CD e o cn CD J 3 tu 3 -a ob CD c -a CD c CD & 00 .s n o OH c CD 13 ' cn CD C+H '"HH o oo *3 e to 00 s a H3 oo CD N '55 •§ a s j ? c» _cD CD " O I I •a M — t n * * CD a £ f cc? CU § o E J3 a> cn J3 CD * - C M M -a ^ 1 m >-O N cn O N 3 - f j 0 -o 5 ccj S J "° -CO CD t- > 1 2 © 8 cn QO .i s o +^  OH CD CD g 1 - 1 ci3 " i § CD C S O -a S '55 C t U 1-1 •HH-S? <+H o CD CD t n J3 '5 s 5 OH C c3 <D > •a o o T 3 T 3 CD CD g g CD CD £ 6 cn cn § i 43 4= IZ 22 1.6.2.2 Kin recognition model Initial work on the Golgi localization of coronavirus M glycoprotein E l demonstrated that Golgi retention of this protein was associated with the formation of insoluble homo-oligomers (Weisz et ah, 1993). This led to the kin-recognition model (Fig. 5), which proposes that resident proteins of the Golgi oligomerize to form large homo- and hetero-oligomers whose large structure prevents their entry into transport vesicles destined for the cell surface, and are thus retained in the Golgi (Nilsson et al., 1993). In support of this model, the substitution of an ER localization signal for the cytoplasmic N-terminus of GlcNAc-T I not only resulted in its relocalizaion to the ER but also caused another medial-Golgi protein to relocalize there as well (Nilsson et al., 1994). Furthermore, recent data on fj-galactoside a2,6-sialyltransferase have demonstrated that the major mechanism of Golgi localization used for this glycosyltransferase is formation of Triton-insoluble mono-oligomers (Chen ef al, 2000). \f\T4 C/) o „ <u c > o <» OO OH i | ° -2 oo £ ? s o .-fa OO CU . _ T Ic3 T3 CU • s o — s s cu p ccj <o fi CU O N 2 tU N tu s • ^  "3 s -T3 £2 S -a OO 3 CU cU ] M 2 —. tu W 3*1 o0 cxS cn cu N 'oo . tu "55 •£ -o o o — B Til O g « o « -JP a | l * u CU m 1-1 « .5 .1.3 CO X) g " O "tu cU O E <u " M C CA o M cj T3 cu cn OO <U CS — 00 CU CU £ > ON >> ON 43 „ tu 2 -s 24 1.6.3 Secretory Vesicles Once processed by the Golgi apparatus, secretory proteins and other substances destined for transport out of the cell are packaged into secretory vesicles. These vesicles move from the Golgi to the PM where they fuse with the PM and release their contents into the extracellular space. Several cytosolic coat proteins have been described for vesicles exiting from the TGN (Gu et al, 2001; Jones et al, 1993; Ladinsky et al, 1994) though other studies have not found evidence of a distinct class of coat complex (Nickel et al, 1994). Thus at present, the biogenesis of secretory vesicles remains unclear. 1.7 Research Objectives The glycosyltransferases E X T l and EXT2 are type II transmembrane proteins involved in the biosynthesis of heparan sulfate (HS) polysaccharide chains (McCormick et al, 2000). They are found predominantly in the ER when overexpressed individually but form a Golgi-localized heterocomplex when co-expressed in the same cell (Kobayashi et al, 2000; McCormick et al, 2000). A l l published E X T l and EXT2 missense mutations identified in HME patients are located in their lumenal C-termini (Philippe et al, 1997; Wuyts and Van Hul, 2000), and studies on these E X T l mutants show that some of these mutations are capable of abrogating in vivo HS biosynthetic activity (Cheung et al, 2001). Furthermore, the GlcA-T catalytic domain has been localized to the central region of the E X T l protein, indicating that the lumenal domain is important for enzymatic activity (McCormick et al, 2000; Wei et al, 2000). By contrast, little is known regarding the role of the N-terminal domains of E X T l and EXT2. Thus I decided to investigate the importance of the N-terminal domains of these two proteins in i) in vivo HS biosynthesis, 25 ii) heterocomplex formation, and iii) subcellular localization. Chimeric constructs of E X T l and EXT2 were made in which different subcellular localization signals were grafted onto their cytoplasmic N-termini. HS biosynthesis is thought to occur in the Golgi (Fernandez and Warren, 1998) and as the lumenal C-terminus is considered to contain the catalytic domain, the N-terminal region should not be involved in the enzymatic activity of the EXT proteins. I therefore hypothesized that only subcellular localization signals directing the E X T l proteins to compartments other than the Golgi would abrogate in vivo function as determined by our HSV assay. To test this hypothesis, E X T l and EXT2 chimeras containing ER-, and Golgi-localization signals as well as secretory forms of the two EXT proteins were created and analyzed to determine the roles of their respective N-terminal domains. 26 2. Materials and Methods 2.1 Cell Culture BHK and Sog9 [HS-deficient murine L-cell line (Banfield et al, 1995)] cells were maintained as monolayers in a 37°C, 5% C 0 2 humidified incubator. A l l cells were cultured in Dulbecco's Modified Eagle Medium supplemented with 10% (v/v) fetal bovine serum, 50 U/mL penicillin and 50 ug/mL streptomycin. A l l cell culture reagents were from Life Technologies, Burlington, Ont., Canada. 2.2 EXT Constructs Al l primers used in the creation of chimeric EXTmyc or EXTlgfp constructs were obtained from Life Technologies, Burlington, Ont., Canada. 2.2.1 EXTmyc constructs The chimeric E X T l constructs were created by PCR of the human E X T l gene from pEXTlmyc (McCormick et al, 1998) while the chimeric EXT2 constructs were created by PCR of the murine EXT2 gene from pmEXT2myc (McCormick et al, 2000). The constructs were created with a myc tag downstream of the gene to enable easy detection of the protein products. 2.2.1.1 Sec-EXTlmyc Sec-EXTl myc was constructed by PCR of the E X T l coding region starting 10 aa downstream of the putative transmembrane domain by using the forward primer BamHI-EXTlstem (5'-CCG GAT CCC GAG AGA AGA ACA CAG CGG TAG G-3') and the reverse primer EcoRI-EXTlrev (5'-GGA ATT CCA AGT CGC TCA ATG TCT CG-3'). The forward primer contains a BamRl restriction enzyme site directly upstream of the E X T l sequence, and the reverse primer contains an EcoRl site excluding the E X T l stop codon to allow for in-frame tagging. After digestion with BamHI (Life Technologies, Burlington, Ont., Canada) and EcoRl (Life Technologies, Burlington, Ont., Canada), the Sec-EXTl PCR product was then ligated into the BamHI and EcoRl site of pSecTag2 "C" (Invitrogen, Burlington, Ont., Canada) such that the E X T l coding region was in frame with pSecTag2's secretion signal (immunoglobulin kappa chain leader sequence). The PCR reaction was performed as follows: 2 min initial denaturation at 94°C, 30 cycles of 94°C for 30 sec, 54°C for 30 sec, 72°C for 2 min 30 sec with an additional 10 min elongation step at 72°C at the end. 2.2.1.2 Sec-EXT2myc Sec-EXT2myc was constructed in a similar fashion as described for Sec-EXTl myc using the primers BamHI-mEXT2stem (5'-CGG GAT CCC GAG TGA TGG GGG TGT GGA G A A ACG-3') containing a BamHI site, and EcoRI-mEXT2rev (5'-GGA ATT CCT A A G CTG CCA ATG TTG GGG A A G CTC TTC-3') containing an EcoRl site. The Sec-EXT2 PCR product was also ligated in the pSecTag2 vector. 2.2.1.3 ER-EXTlmyc ER-EXTlmyc was constructed by PCR of the E X T l coding region by using the primers hEXTl-ERfor (5'-GCG GAT CCT CCC ACC ATG CAC AGG AGG A G A AGC TAT TTC ATC CTG TCA GCT GGC-3') containing a BamHI site in the 5'-end, and NotTagclone8 (5'-TTT TTC TTT TGC GGC CGC TTT TTT CCT TAA GTC GCT C A A TGT CTC GGT A-3') containing a Notl site in the 3'-end. After digestion with BamHI (Life Technologies, Burlington, Ont., Canada) and Notl (Life Technologies, Burlington, 28 Ont., Canada), the ER-EXT1 PCR product was then ligated into the BamHl and Noil site of pcDNA3.1mycHis "B" (Invitrogen, Burlington, Ont., Canada). 2.2.1.4 ER-EXT2myc ER-EXT2myc was constructed by ligating the amplified chimeric EXT2 PCR product into the BamHl and Sstll (Life Technologies, Burlington, Ont., Canada) site of pcDNA3.1mycHis "B". The primers used were mEXT-ERfor (5'-GCG GAT CCG CCC ACC ATG CAC AGG AGG AGA AGC ATC TAC TAC GTC ACC CTG TTC TCC-3') and SstIImEXT2rev (5'-TCC CCG CGG GGA TAA GCT GCC AAT GTT GGG GAA-3'). The forward primer, mEXT-ERfor contains a BamHl restriction enzyme site while the reverse primer, SstIImEXT2rev contains a Sstll site. 2.2.1.5 Golgi-EXTlmyc Golgi-EXTlmyc is a chimeric construct containing the N-terminus, transmembrane domain and part of the stem region of murine N-deacetylase/N-sulfotransferase 2 (NDST2) (Kusche-Gullberg et al., 1998) fused to the stem region and catalytic domain of E X T l . It was constructed in 2 fragments, the first PCR fragment containing the mNDST N-terminus, transmembrane domain and 129 nucleotides of the stem region along with 24 nucleotides of the E X T l gene at the 3'-end [constructed using the primers mHSNST-Bgl-fwd (5'-GAA GAT CTT CCC ACC ATG CTC CAG CTG TGG A A G GT-3') containing the Bglll site, and mNDST2/hEXT 1 -Golgi(rev) (5' -GCT GTG TTC TTC TCT CCG GCT GTG GCT CCT AGT TGT TTC TGG AGG CCT TTG GGG-3'), a hybrid primer containing 24 nucleotides of the E X T l gene]. The second PCR fragment contained the C-terminus of E X T l starting in the stem region along with 24 nucleotides of the NDST2 gene at the 5'-end. It was created using mNDST2/hEXTl-Golgi(for) (5'-CCT CGA CCC C A A A G G CCT CCA GAA ACA ACT AGG AGC CAC AGC CGG AGA G A A GAA-3'), a hybrid primer with containing 24 nucleotides of the NDST2 gene, and EXT1-L360rev (5'-ATA GTT TAG CGG CCG CAT TCT TAT TCA GAG CAT CAC AGG GAC GCA GGC-3') containing the PpuMl site. The two PCR products were mixed together, denatured for 5 min at 92°C, and allowed to slowly reanneal at room temperature. The 3'-ends of the reaction products were extended using Taq DNA polymerase (72°C for 20 min) to form double stranded "polished" ends. This was then used as the template for a PCR reaction using the primers rnHSNST-Bgl-fwd and EXT1-L360rev. The PCR reaction was performed as follows: a 2 min initial denaturation step at 94°C followed by 5 cycles of 94°C for 30 sec, 30°C for 30 sec, 72°C for 2 min 30 sec then 25 cycles of 94°C for 30 sec, 52°C for 30 sec, 72°C for 2 min 30 sec with an additional 10 min elongation step at 72°C at the end. The PCR product was then digested with Bglll (Life Technologies, Burlington, Ont., Canada) and PpuMl (New England Biolabs, Mississauga, Ont., Canada) and ligated into a BamHI and PpuMl digested pEXTlmyc (McCormick et al., 1998). 2.2.1.6 Golgi-EXT2myc Golgi-EXT2myc was constructed in the same manner as described above. The primers used were rnHSNST-Bgl-fwd (sequence as listed above) and mNDST2/mEXT2-Golgi(rev) (5'-CAC ACC CCC ATC ACT GGA GGA CTC AAT GGA AGT TGT TTC TGG AGG CCT TTG GGG-3'), containing 24 nucleotides of the EXT2 gene in the 3'-end, for the first fragment of the chimeric protein. The NDST2/EXT2 hybrid primer, mNDST2/mEXT2-Golgi(for) (5'-CCT CTA CCC CAA AGG CCT CCA GAA A C A ACT TCC ATT GAG TCC TCC AGT GAT GGG-3'), and EXT2-E367rev (5'-CCG CCG CTC GAG CGG TCA CTC TGG AAC GAC CAC AGA TGC-3') containing the Blpl site were 30 used for the second fragment. After the final PCR with the primers mHSNST-Bgl-fwd and EXT2-E367rev, the PCR fragments were digested with Bglll and Blpl (New England Biolabs, Mississauga, Ont., Canada) and ligated into BamHl and Blpl digested pEXT2myc (McCormick et al., 1998). PCR was performed as described for Golgi-EXTlmyc. 2.2.2 EXT lg fp constructs The myc-tagged chimeric E X T l constructs were re-created with a gfp tag downstream of the gene to enable differentiation of cotransfected chimeric proteins. 2.2.2.1 Sec-EXTlgfp Sec-EXTlgfp was constructed by amplifying the E X T l gene containing the N-terminal immunoglobulin kappa chain leader sequence from Sec-EXTmyc using the primers Secl&2gfp(for) (5 '-GAA GAT CTT CTA GCC ACC ATG G A G A C A GAC A C A CTC-3') and EcoRI-EXTlrev (sequence as listed above). The PCR fragment was then digested with Bglll and EcoRI, and subcloned into the Bglll and EcoRl restriction enzyme sites of the pEGFP-Nl vector (Clontech, Palo Alto, CA, USA). 2.2.2.2 ER-EXTlgfp ER-EXTlgfp was also created in a similar fashion, the gene was amplified with the forward primer hEXTl-ERfor (sequence listed above) and the same reverse primer as the one used in the construction of Sec-EXTlgfp. The PCR fragment was digested with BamHl and EcoRl and subcloned into the Bglll- and EcoRI-digested pEGFP-Nl . 2.2.2.3 Golgi-EXTlgfp The chimeric NDST2/EXT1 gene was amplified from Golgi-EXTlmyc with mHSNST-Bgl-fwd and EcoRI-EXTlrev (both sequences listed above). Golgi-EXTlgfp 31 was created by digesting with Bglll and EcoRl followed by ligation of the PCR fragment into pEGFP-Nl . 2.3 Transient Transfections Cells were transfected using LipofectAmine Plus (Life Technologies, Burlington, Ont., Canada) according to manufacturer's instructions. Briefly, for the transfection of cells in one well of a 6-well dish, 1 |ig of DNA was complexed with 6 pi of PLUS Reagent in 100 pi of DMEM. 15 min later, the addition of 4 ul of LipofectAMINE in 100 pi of D M E M was combined with the PLUS-complexed DNA. After another 15 min, this mixture was added to the well containing 0.8 ml of D M E M and incubated for 3 hrs at 37°C in a 5% C 0 2 humidified incubator. The transfection media was then replaced with D M E M containing 10% FBS and 1% PenStrep. 2.4 Immunofluorescence staining Al l oc-myc and a-gfp antibodies used in this and other experiments involving these two antibodies were from Invitrogen, Burlington, Ont., Canada. The antibodies were provided at a concentration of approximately lmg/mL. The a-myc antibodies were mouse monoclonal antibodies (9E10.2), and the a-gfp antibodies were rabbit polyclonal antibodies. Approximately 175,000 BHK cells were seeded on acid-etched coverslips (Fisher, Nepean, Ont., Canada) and transiently transfected (as described above) the following day. The cells were fixed with 4% paraformaldehyde, EM-grade (Canemco Inc., St. Laurent, Que., Canada) 24-28 hrs following transfection and blocked overnight with 1% BSA/PBS. 32 Cells were then incubated with a-myc antibodies diluted at 1:200 in 0.25% saponin/1% BSA/PBS for 1 hr at room temperature in a humidified chamber. Bound primary antibodies were visualized using goat a-mouse antibodies conjugated to Texas Red (Cedar Lane, Hornby, Ont., Canada), Cy5 (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) or FITC (Kirkegaard and Perry Laboratories, Gaithersburg, MD, USA) at 1:100 in 1% BS A/PBS. Cells were visualized and images acquired digitally using either the Zeiss DVC fluorescent microscope or the Bio-Radiance Plus Confocal microscope. 2.5 Immunoblotting Subconfluent BHK cells were transiently transfected and lysed 40 hrs later in 100 uL of Triton X-100 lysis buffer (2% Triton X-100/20 mM Tris-HCl pH 7.4/150 mM NaCl containing complete protease inhibitors [Boehringer Mannheim, Indianaplis, IN, USA]) for 5 min at 4°C. The cell lysates were then centrifuged at 15,000 x g for 15 min and the supernatant separated by 6% SDS-PAGE. Following separation, the proteins were transferred onto Immobilon-P membrane (Millipore, Bedford, MA, USA) using a semi-dry transfer apparatus (Pharmacia Biotech, Baie d'Urfe, Que., Canada) according to manufacturer's instructions. The membrane was blocked with 5% skim milk/TBS at 4°C O/N then incubated with a-myc and/or a-gfp antibodies diluted 1:2500 in 3% skim milk/TBS for 3 hrs. The membrane was then washed 3X (5 min each time) in 0.05% Tween-20/TBS followed by the addition of horseradish peroxidase-conjugated (HRP) oc-mouse (Kirkegoard and Perry Laboratories, Gaithersburg, MD, USA) or HRP-conjugated a-rabbit (Kirkegoard and Perry Laboratories, Gaithersburg, MD, USA) antibodies at 1:2000 in 3% skim milk/TBS. Following another 3 washes, the proteins were detected 33 using the LumiGLO Chemiluminescent Substrate Kit (Kirkegoard and Perry Laboratories, Gaithersburg, MD, USA) followed by exposure to B ioMAX Light Film (Kodak, Toronto, Ont., Canada). J 2.6 Immunoprecipitation At 16-18 hrs post-transfection, the BHK cells were radiolabeled with 3.7 MBq/mL L-[35S]methionine (trans 35S-label, ICN, Montreal, Que., Canada) in methionine-free D M E M (ICN, Montreal, Que., Canada) supplemented with 4% dialyzed fetal bovine serum and 2 mM L-glutamine (Life Technologies, Burlington, Ont., Canada). Cells were then lysed in 1 mL of Triton X-100 lysis buffer and centrifuged at 15,000 x g for 15 min at 4°C. The supernatant was pre-cleared with Protein G-sepharose 4 Fast Flow (Amersham Pharmacia Biotech, Baie d'Urfe, Que., Canada) for 1 hr at 4°C on a rotator before being centrifuged for 15 sec at 15,000 x g. One pg of either a-gfp or a-myc was added to the pre-cleared supernatant and allowed to bind overnight while rotating at 4°C. The following day, protein G-sepharose was added and allowed to rotate for an additional 2 hrs at 4°C followed by centrifugation. The supernatant from this was then separated by 10% SDS-PAGE (Hoeffer apparatus) and transferred onto Immobilon-P membrane (Millipore, Bedford, MA , USA) using the semi-dry transfer apparatus followed by exposure to B i oMAX MR Film (Kodak, Toronto, Ont., Canada). 2.7 HSV Assay Approximately 1 million Sog9 cells were seeded and transfected with various EXT constructs the following day. At 40 hrs post-transfection, the transfected Sog9 cells were 34 washed in PBS and infected with (3-galactosidase expressing herpes simplex viruses (either HSV type 1 [G207] or HSV type 2 [LIBR1]) at an M.O.I, of 0.1 pfu/cell for 1 hr in serum-free media. Cells were then washed 3X in PBS and incubated in 0.1% IgG/10% FBS/DMEM for 10 hrs (for HSV-1) or 6 hrs (for HSV-2). The cells were then fixed in 4% paraformaldehyde, blocked with 1% BSA/PBS, and stained with X-gal. 2.8 Sandwich Enzyme Linked Immunosorbent Assay (ELISA) The bottom of 96-well plastic plates were coated with a-myc antibodies at 1:100 in TBS for 5 hrs at room temperature then blocked overnight with 3% skim milk/TBS at 4°C. Approximately 500,000 BHK cells were transiently transfected and lysed in 150 LiL of Triton X-100 lysis buffer, centrifuged at 15,000 x g for 15 min at 4°C and the supernatant added to the a-myc antibody-coated wells for 2 hrs at room temperature. Wells were washed 3X each time with 0.05% Tween 20/TBS before the addition of the next substance, a-gfp antibodies at a dilution of 1:100 in TBS were added to the wells next for 2 hrs followed by the addition of HRP-conjugated secondary a-rabbit antibodies at 1:100 in TBS for 1 hr. Homo-/hetero-complex formation was detected by the addition of 100 (iL of 3,3',5,5'-tetramethylbenzidine (TMB) liquid substrate system (Sigma, Oakville, Ont., Canada). The reaction was stopped with the addition of 100 |iL of 0.5M H 2 S0 4 and the relative amount of chromophore generated was determined by spectrophotometry (A,abs: 450 nm) using the Ultrospec 3000 spectrophotometer (Pharmacia Biotech, Baie d'Urfe, Que., Canada). < 35 3. Results 3.1 EXT constructs To determine the importance of the N-terminal regions of E X T l and EXT2 for in vivo HS biosynthetic activity as well as to study its role in EXT1/EXT2 heterocomplex formation, several chimeric EXT constructs were made (Fig. 6). Specifically, the N-terminal sequences of E X T l and EXT2 were replaced with other known subcellular-localization signal sequences, which have been shown to direct chimeric proteins to the ER, Golgi or for secretion. A l l E X T l and EXT2 chimeric constructs were C-terminally tagged with either myc or gfp for easy detection of the chimeras. To construct the ER-EXT1 and ER-EXT2 constructs (Fig. 6A), the cytoplasmic N-termini of E X T l and EXT2 were deleted up to 1 aa before the putative transmembrane domain and substituted with MHRRRS, an ER-localization motif from human invariant chain protein 33. These 6 aa residues were sufficient to retain the human transferrin receptor, a type II membrane protein, to the ER and thus abrogate normal cell-surface expression (Schutze et al., 1994). The transmembrane domain and adjacent sequences of oc2,6-sialyltransferase and GlcNAc-TI, both Golgi-localized glycosyltransferases, have been shown to be sufficient to localize non-Golgi proteins to the Golgi (Burke et al., 1994; Munro, 1991). Thus to localize E X T l and EXT2 to the Golgi, the N-terminus, putative transmembrane domain and part of the stem region of both E X T l and EXT2 were replaced with the N-terminus, transmembrane domain and part of the stem region of murine N-deacetylase/N-36 sulfotransferase 2, an isozyme of human NDST1 which localizes to the trans-Golgi network (Humphries etal., 1997; Kusche-Gullberg etal., 1998) (Fig. 6B). For the secretory constructs of E X T l and EXT2, the cytoplasmic N-terminus, putative transmembrane domain along with 10 amino acids of the stem region were excised and substituted with the immunoglobulin kappa chain secretory signal tag supplied in the pSecTag2 vector (Invitrogen) thus targetting the chimeric protein for secretion (Fig. 6C). 37 E X T l EXT2 L Iip33 ER-EXT1 ER-EXT2 1 8 27 100 1 25 MHRRRS TM 44 104 TM 1 47 72 MHRRRS 1 7 8 MHRRRS \ 7 27 100 TM 24 25 44 104 NDST2 Golgi-EXTl Golgi-EXT2 c Ig kappa chain | leader sequence I Sec-EXTl I I Sec-EXT2 39 85 1 19 39 81 32 100 1 19 39 81 49 104 42 42 37 42 54 100 104 746 aa 718aa 231 aa 746 aa 701 aa 882 aa 796 aa 751 aa 752 aa 707 aa Figure 6 Chimeric EXT constructs. Al l chimeric constructs were C-terminally tagged with either myc or gfp for easy detection. A) The N-termini of EXT l and EXT2 were replaced with the ER-localization signal MHRRRS (Schutze, M. et al, 1994) to generate the ER-EXT1 and ER-EXT2 chimeric constructs. B) Munro, S., 1991 proposed that the cytoplasmic N-terminus, transmembrane domain and part of the stem region of type II Golgi-residents were sufficient to localize type II membrane proteins to the Golgi. Thus to generate the Golgi-EXTl and Golgi-EXT2 chimeric constructs, the N-termini of E X T l and EXT2 were substituted with the NDST cytoplasmic N-terminus, the transmembrane domain and part of the stem region. C) EXT l and EXT2 minus the cytoplasmic N-terminus, transmembrane domain and 10 aa of the stem region were subcloned into the secretory vector, pSecTag2 (InVitrogen) which contains the immunoglobulin kappa chain leader sequence for efficient secretion of expressed proteins (Coloma, MJ et al., 1992). 38 3.2 Transient expression of EXT constructs in mammalian cells Al l EXT constructs were verified by DNA sequencing and tested for expression in mammalian BHK cells. Expression of chimeric EXT proteins was detected via their myc or gfp C-terminal tags. ER-EXTlgfp and Sec-EXTlgfp were predicted to have a molecular weight of 114 and 116 kDa, respectively, while Golgi-EXTlgfp was expected to be slightly larger at 120 kDa. Analysis of the chimeric E X T l gfp expression showed protein bands with molecular masses that were larger than predicted (Fig. 7A). As there are two putative N-glycosylation sites on E X T l , it is possible that they were glycosylated thus causing the increase in molecular weight observed. The myc-tagged EXT2 chimeras were expected to be 83, 85 and 88 kDa for ER-EXT2, Golgi-EXT2 and Sec-EXT2 respectively. Bands migrating higher than expected were observed for the chimeric EXT2myc proteins (Fig. 7B). Like E X T l , there are also two putative N-glycosylation sites on EXT2 which may become N-glycosylated thereby resulting in an increase in molecular'weight. Thus within the resolution of the gel system used, all chimeric EXT constructs were expressed in mammalian cells, and were larger than expected likely due to glycosylation of the proteins. CO CO Q 0 5 CO T — T — W CM CO CD TJ CD «—' CO i_ ro CL cn c CD J C CD co CD CO co 0) o CD CO •+•»» o C L TJ CD CD CJ) CO 1 5 | 1 CO TJ £ I p CO 5 co Q. CD CO c o CD O T J CD « 2 1 = £ : CD a, o c ^ O X O CD K- o X I H - CD O T5 (fl £ •— cn tn c _ > > CO E § 2> i i TJ S» T 3 CO CD « 1 = 1 8 <D i j? -• g CD CD CO E I c ® o >> _Q o E E c CD c CO CD O CD TJ 5 ?> <o CO +>» O J2 C s_ CD o CD CO c ro 0 i— CD tfl 5 <U co 3 u LL, LU to i : c to CO c 4= o TJ | = " i - -8 * y UJ is II CD S i CD JZ CO i_ -Q E CD C o L_ J! CO c CO D) c 1 CO CD TJ o c ro o >> E JZ TJ CD -Q 2 CL CD CD % co I CD O TJ CL I g "E E CO CM -Q | TJ . i CD JZ 0 O CD ^ 1 IE 6e 40 3.3 Golgi-EXTl, ER -EXT1 and Sec-EXTl were all able to restore HSV susceptibility in Sog9 cells To determine whether the substitution of different subcellular localization signals would perturb the in vivo enzymatic function of E X T l , all E X T l constructs were transiently expressed in Sog9 cells. Sog9 cells are a murine L-cell derived line (Banfield et al., 1995) that contains a specific defect in the E X T l gene, predicting a truncated E X T l protein (McCormick et al., 2000) and are thus deficient in heparan sulfate biosynthesis. Due to the lack of cell surface heparan sulfate expression, and therefore of the initial attachment receptor, Sog9 cells are 99.5% resistant to HSV infection (Banfield et al., 1995). At 40 hrs following transfection, Sog9 cells expressing the various EXT constructs were infected with HSV strains expressing (3-galactosidase, and 6-10 hrs following infection, the cells were fixed and stained with the p-galactosidase substrate, X-gal. EXT2 is unable to compensate for the defect in HS biosynthesis present in Sog9 cells (McCormick et al., 2000) and in agreement with this, none of the chimeric EXT2myc constructs were capable of restoring Sog9 susceptibility to HSV infection (Fig. 8). By contrast, all the EXTlgfp constructs tested were able to synthesize HS in vivo as determined by the HSV assay though at different efficiencies (Fig. 9). As GlcA-T and GlcNAc-T activity is believed to occur in the Golgi (Fernandez and Warren, 1998), the transfection of Golgi-EXTl into Sog9 cells was expected to allow HSV to infect Sog9 cells. Not only was this observed, but the Golgi construct allowed 15-20% more Sog9 cells to be infected by HSV than wt E X T l (Fig. 9D). Interestingly, ER-EXT1 also restored HS biosynthesis in Sog9 cells despite an ER localization signal (Fig. 9C). The HSV infectivity of ER-EXT1-transfected Sog9 cells was slightly lower at 90-95% of wt E X T l level. The 41 ability of ER-EXT1 to allow HSV to infect transfected Sog9 cells may be an artifact of overexpression as the ER may be overwhelmed with the amount of proteins present leading to leakage of ER-EXT1 to the Golgi. Remarkably, the secretory construct of E X T l (Sec-EXT1) was also able to synthesize HS, and thus enabled HSV to infect Sog9 cells although at a greatly reduced level (Fig. 9E). The infectivity of the Sec-EXTl-transfected Sog9 cells was only 20-30% of wt E X T l level but this still suggests that the Sec-EXTl protein was functional and likely being retained within the cell long enough for the polymerization of heparan sulfate chains to occur on proteoglycans. Since HSV-1 and HSV-2 differ slightly in their interactions with HS (Herold et al, 1996), all the E X T l constructs were tested with both HSV serotypes. Similar results were obtained with HSV-1 as with the HSV-2 assay (data not shown). 00 o >» E CM P -X UJ o UJ LU o o >* E OJ fe UJ I o o >» E CM X UJ _ j _ o o >> E CM fe UJ I Dd UJ E CN - S £ LU cu CO 0) o T J 0 O IM CO c 2 *•* CD 1— CD c o g 0 O O >< E CM h-X LU _i_ J3J O jfl o I O > 9. co ^ 5 E c 2 6 x cn LU LU > . CO DQ O =! | • | 13 5 .N "S3 "ro CM w > > co a) X -c CO H .S £ f _ £ ro P CO CD .C CD f to > CD "O TJ (j) CO C ro ~ O CO co p c o CO T J c CO CD CD m CO CO (« co gj -(0 "5 C CD C CO CO OJ CO $ •fa co TJ <0 >« .2 o >< E x UJ c TJ >. JZ CD TJ 1 - CO * i co £ 1 1 CJ CO cu a CM CO _h i m o *S o o CD > (A OJ (0 Si CO CO o c * O XJ "*= CD CO C CO Q. .£3 00 J2 o >> E ^ CD CO <£• o 5 CL 5 CD o X 3 OJ LU LL W S CD o CD zt C D CM I-X UJ a-j— X LU it- Cl LU . 6 CD on '•m > ,gt Q_ *— X UJ o o O) UJ ill ex. JZ ™ 16 XJ 111 i i £ to co — CO —» *» CD ° £- Q-CD I ^  Tj LU CD O TJ CD O 0 j C ;> . CO oo o ^ § tZ *i •— CO 5 N "to "co > > CO CD X JZ • i ^ co 2 cn x CD > CO > CO XJ TJ CD CO C 2 CO i * lo •§ CD CD £ I S CO "CJ CD CD CJ) 0 TJ m >> M=- JZ £ CD 1 1 co £ 1 i O CO CD e-- JZ JZ •b fe p ** m > co X o -«-' *~* c CO -t—' CO CO CD _D) oo Q. cn i— X LU .2? LU Ci o o> Ui to O ZJ tO £ c JZ ~~ 2> (j CO o 3 (0 c o o O) i— X UJ CO TJ c CO CO CO CD JZ I £ -Q X CO LU X c o o CD > CO cn CD c m c CD o s s CO CO CO > CO o 5_ V c 5 0 TJ 1 i CO C 5 £ cn CD 5 to CL m JL CD •as F " # * 3 CD LU O) §) — u. co S JZ fl) o S cu «M JZ < r ~ 3.4 Subcellular localization patterns of mono-transfected EXT constructs Immunofluorescence studies of all chimeric EXT constructs were done to determine the effects the sorting signals were having on the subcellular distribution of the chimeras. Each EXT construct was overexpressed in BHK cells and its subcellular localization pattern observed by immunofluorescence microscopy. Immunofluorescence images depicting the predominant staining patterns of the chimeric proteins are shown in Figure 10. Mono-transfected E X T l and EXT2 (Fig. 10 A, B) displayed a staining pattern indicative of the ER in agreement with previously published data that the overexpression of either E X T l or EXT2 leads to ER-localization (McCormick et al., 2000; McCormick et al., 1998). Similarly, ER-EXT1 and ER-EXT2 (Fig. 10 C, D) also distributed in a staining pattern distinctive of ER proteins. Interestingly, Golgi-EXTl and Golgi-EXT2 (Fig. 10 F, G) exhibited staining patterns characteristic of the early Golgi cisternae or ERGIC, though some chimeras did distribute to a perinuclear location. Yet NDST2 (Fig. 10 E), which supplied the Golgi-localization signal used in the Golgi-EXT chimeras, localized predominantly to the Golgi as indicated by the immunofluorescence pattern and not to the ERGIC or ER. Mono-transfected Sec-EXTl and Sec-EXT2 (Fig. 10 I, J) showed a staining pattern characteristic of secretory vesicles and SecPSA (prostate specific antigen with the same immunoglobulin kappa chain leader sequence in its N-terminus), a secretory control provided by Invitrogen (Fig. 10 H). 45 4k' EXT 1 gfp C E X T 2 m y c D E R - EXT1g fp E E R - E X T 2 m y c F NDST2g fp Golg i -EXT1gfp SecPSAmyc Sec -EXT1gfp Sec -EXT2rr yc Figure 10 Subcellular localization patterns of mono-transfected chimeric EXT proteins. Monolayers of BHK cells were mono-transfected with A) EXTIgfp, B) EXT2myc, C) ER-EXT1gfp, D) ER-EXT2myc, E) NDSTgfp, F) Golgi-EXT1gfp, G) Golgi-EXT2myc, H) SecPSAmyc. I) Sec-EXT1gfp and, J) Sec-EXT2myc. Myc-tagged constructs were detected with cx-myc antibodies followed by FITC-conjugated secondary antibodies. The subellular distribution of myc-tagged or gfp-tagged proteins were observed by immunofluorescence and images acquired using either the Zeiss DVC fluorescent microscope or the Bio-Radiance Plus confocal microscope 46 3.5 Co-transfection of E X T l constructs with wild-type (wt) EXT2 constructs E X T l and EXT2 localize to the ER when overexpressed individually but re-localize to the Golgi when transfected together (McCormick et al, 2000). Furthermore, McCormick et al. have also shown that E X T l and EXT2 are capable of forming a heterocomplex in vivo. Therefore, the subcellular localization patterns of chimeric E X T l when co-expressed with wt EXT2 was investigated. 3.5.1 Immunofluorescence analysis of co-transfected chimeric E X T l with wild-type EXT2 Data thus far showed that Sec-EXTl was able to synthesize HS in vivo as demonstrated by HSV's ability to infect SecEXTl-transfected Sog9 cells, therefore it must be retained within the cell long enough for HS chain polymerization to occur. Hence the effects of co-transfecting gfp-tagged chimeric E X T l with myc-tagged wt EXT2 was examined. The distribution of cells with both co-transfected EXTs co-localizing in staining patterns characteristic of the ER, ERGIC, Golgi or secretory vesicles were assessed, and images depicting the immunofluorescence patterns of the majority of co-transfected cells are presented in Figure 11. The immunofluorescence images of chimeric E X T l gfp and wt EXT2myc were overlaid and shown in the third panels of each row. ER-EXTlgfp + EXT2myc (Fig. 1 IB) co-localized in a staining pattern characteristic of the Golgi despite the ER-localization motif in the cytoplasmic N-terminus of ER-EXTlgfp. The co-transfection of Golgi-EXTlgfp with wt EXT2myc (Fig. 11C) also resulted in both proteins localizing to what appears to be the Golgi. Interestingly though, Sec-EXTl gfp displayed a J 47 staining pattern typical of Golgi-localization when co-transfected with wt EXT2myc (Fig. 1 ID). The data showed that all chimeric E X T l constructs including the secretory construct of E X T l re-localized predominantly to the Golgi when co-transfected with wt E X T 2 , suggesting that they are able to hetero-oligomerize. 4 8 Chimeric EXTIgfp of co-transfected cells EXT2myc of co-transfected cells Overlay of chimeric EXT1gfp+EXT2myc A EXTIgfp + EXT2myc B ER-EXT1gfp + EXT2myc EXT2myc l l i i l i i • c c • cr K •4 \ A , ; . , cr -Sec-EXT1gfp + EXT2myc Figure 11 Subcellular localization patterns of co-transfected chimeric EXT1 with wild type EXT2. BHK monolayers are co-transfected with A) EXTIgfp + EXT2myc, B) ER-EXT1gfp + EXT2myc, C) Golgi-EXTlgfp + EXT2myc and D) Sec-EXT1gfp + EXT2myc. Wild-type EXT2myc proteins were detected with a-myc antibodies followed by Cy5-conjugated secondary antibodies. The subcellular distribution of myc-tagged or gfp-tagged proteins were observed by immunofluorescence and images acquired using either the Zeiss DVC fluorescent microscope or the Bio-Radiance Plus confocal microscope. 49 3.5.2 A l l E X T l constructs were able to form a heterocomplex with wild-type EXT2 To confirm that the co-localization of the E X T l constructs with wild-type EXT2 was due to heterocomplex formation between the two proteins and not an artifact, sandwich enzyme-linked immunosorbent assay (ELISA) was performed. A 96-well plastic plate was coated with cc-myc antibodies and cell lysates from either mono- or co-transfected BHK cells were added to the wells in triplicates. Rabbit a-gfp antibodies were then added followed by HRP-conjugated secondary a-rabbit antibodies. Heterocomplex formation between the E X T l constructs and wild-type EXT2 was detected by the addition of the liquid substrate, TMB, and the soluble coloured product read at 450 nm. As can be seen in Figure 12, the substitution of the different N-terminal localization signals did not affect the oligomerization ability of chimeric E X T l proteins with wt EXT2, suggesting that the N-terminal sequences was not necessary for hetero-oligomerization. CN CN H H co *-> >< n a <u pq z + + <-<o o o =3 5-t i-i S-i j *jr 0(0 00 C fi c fi o O o O u U U 0 • • • H 00 O © —f-d o « 0 5 fi O u CU u Cm CA fi H I T ) d d —f-d o u< -<-> fi o o fN d O C f - eoiieqjiosqy o 6 5 CO cd C4 cu . cj ^ u «i F* is c o c 3 H k c cu c: W cn" - 3 cn I « ? t f l f c° o 3 S a T J >» ^ a jo ? rs y O >J cd 2 **a cu S cu 3 M c £ o K S u 2 2 ^ 2 cu a 2 H W u CU U o ° "2 3cu 2 « JS tj T3 CU *9 U C M CO S "cu CO </j 6 0 rrl «' 3 o .5 o o 3 © ° f i cu o u o -CU *-> CU £ C M O CO & JS " S o CU CU - S % H 5 cu ?r> a st B ra cn g cn cn Js >> 3 g — <_> g 13 | | cL -5 co (U 1X1 O, 6 "2 cu s 2 S iS £ 2 -11 l & 3 ces oo D-j *3 C50 * > e cu « M S 0 3 o C CU B a CU cn & a. XJ e 3 CO 1 "2 cU M c CQ S Q E N c CU __ XJ E g 5 cn C 2 S cU & cn 2 I S E s s JO CU XJ cd CU g i S3 o XJ CU •S 5 XJ x3 B o cn 00 JJ o CJ cn CU i i l i iJ c CU E c <U CX XI CU o it 3 cn fi CU > '3 a 3 CU cn fi &, cu CU o XJ CU o CU OS 51 3.6 Co-transfection of chimeric E X T l with chimeric EXT2 As the above data indicated, chimeric E X T l re-localized predominantly to the Golgi when co-transfected with wt EXT2 as suggested by the immunofluorescence patterns observed. Thus the ability of chimeric EXT 1 to hetero-oligomerize with chimeric EXT2 and localize to the Golgi despite both proteins containing different subcellular localization signals was determined. Immunofluorescence studies were done to observe the effects on subcellular localization of co-transfecting chimeric E X T l and chimeric EXT2 together. The proportion of cells containing the two co-localized chimeric constructs displaying staining patterns characteristic of each subcellular compartment was ascertained, and Figure 13 shows the predominant immunofluorescence patterns of the chimeric proteins. When ER-EXT1 was co-transfected with either ER-EXT2 or Sec-EXT2, the staining patterns observed suggested the chimeras were co-localized among the ER, ERGIC and the Golgi with no obvious predominant localization. For the Golgi-chimeric constructs, the co-transfection of either Golgi-EXTl or Golgi-EXT2 with another chimeric partner allowed both chimeras to immunofluoresce to a perinuclear localization indicative of the Golgi. Without an anchoring domain, the majority of co-transfected Sec-EXTl and Sec-EXT2 exhibited staining patterns suggesting the chimeras were found in secretory vesicles when overexpressed. Thus the results of transfecting chimeric E X T l with chimeric EXT2 were not as clear as the results of co-transfecting with wt EXT2. To investigate whether the non-Golgi localization staining patterns observed for several of the chimeric EXT constructs were due to a lack of hetero-oligomerization, heterocomplex formation was tested using ELISA as described earlier. The absorbance 52 results obtained for the co-transfection of all combinations of chimeric E X T l with chimeric EXT2 were similar to each other and to the absorbance reading observed for wt E X T l and wt EXT2 (data not shown). This indicated that all the chimeric E X T l constructs were able to form a heterocomplex with all the chimeric EXT2 constructs, suggesting that the inability of ER-EXT1 + ER-EXT2, ER-EXT1 + Sec-EXT2, Sec-EXTl + ER-EXT2 and Sec-EXTl + EXT2 to localize to the Golgi was not due to an inability to hetero-oligomerize. 53 Figure 13 Subcellular localization patterns of co-transfected chimeric EXT1 with chimeric EXT2. BHK monolayers were co-transfected with the listed chimeric EXTIgfp and chimeric EXT2myc constructs. Chimeric EXT2myc proteins were detected with a-myc antibodies followed by Cy5-conjugated secondary antibodies. The subellular distribution of myc-tagged or gfp-tagged proteins were observed by immunofluorescence and images acquired using either the Zeiss DVC fluorescent microscope or the Bio-Radiance Plus confocal microscope. Individual gfp- or myc-immunofluoresence images were overlaid together and presented here. 54 3.7 No other proteins involved in the EXT1/EXT2 heterocomplex were detected when either E X T l or EXT2 were immunoprecipitated To investigate whether there were other proteins involved in heterocomplex formation, a radiolabelled immunoprecipitation assay was performed. Approximately 18 hrs after the BHK cells were transfected with EXTIgfp and/or EXT2myc, they were radiolabelled overnight with [35S]methionine. The cell lysates were collected and the tagged-proteins immunoprecipitated with either oc-gfp or a-myc antibodies. The immunoprecipitated proteins were then separated by SDS-PAGE, transferred onto an Immobilon-P membrane and exposed to film. No other proteins were observed to be immunoprecipitated along with the EXT1/EXT2 heterocomplex within the detection sensitivity of the radiolabelled immunoprecipitation assay (Fig. 14). To further complicate the search for other proteins in the EXT1/EXT2 heterocomplex, a high degree of background bands were present despite several attempts to reduce non-specific binding. 55 Immuno-precipated with oc-gfp a-myc a-gfp a-myc a-gfp a-myc Transfected with EXTIgfp EXT2myc EXTIgfp + EXT2myc EXTIgfp + EXTtmyc EXTIgfp EXT2myc Figure 14 Radiolabelled immunoprecipitation analysis of EXT1-EXT2 heterocomplex formation. BHK cells were transfected with the listed constructs and radiolabelled overnight with rsjmethionine. Cells were then lysed and immunoprecipitated with the listed antibody followed by electrophoresis and exposure to film. 56 3.8 Co-transfection of chimeric E X T l with chimeric E X T l The kin recognition model hypothesizes that Golgi residents form homo- and/or hetero-oligomers in the Golgi preventing them from being packaged into vesicles destined for the P M (Nilsson et al, 1993). Thus the ability of two different chimeric EXT l s to homo-oligomerize and the effect of this homo-oligomerization on subcellular localization were investigated. BHK cells were co-transfected with different combinations of the E X T l constructs, their subcellular localization patterns were analyzed using immunofluorescence microscopy, and the results presented in Figure 15. The staining patterns of co-transfected E X T l + E X T l (Fig. 15B), E X T l + ER-EXT1 (Fig. 15C), E X T l + Sec-EXTl (Fig. 15E), ER-EXT1 +ER-EXT1 (Fig. 15F), ER-EXT1 + Sec-EXTl (Fig. 15H) and Sec-EXTl + Sec-EXTl (Fig. 15K) were not characteristic of the Golgi, suggesting that they localized to subcellular compartments other than the Golgi. When Golgi-EXTl was co-transfected with E X T l (Fig. 15D), ER-EXT1 (Fig. 15G), Golgi-EXTl (Fig. 151) and Sec-EXTl (Fig. 15J), it was able to partially co-localize the other chimeric E X T l to the ERGIC or early-Golgi as suggested by the staining patterns observed. The immunofluorescence data indicated that homo-oligomer formation does not play a major role in Golgi localization in the case of E X T l as the co-transfection of chimeric E X T l constructs containing non-Golgi localization signals were not observed to re-localize in a staining pattern characteristic of the Golgi. To confirm that chimeric EXT l s were capable of forming complexes with other chimeric EXT ls , homocomplex formation was studied by ELISA. The results indicated that the chimeric EXT l s were still able to form complexes with all the other chimeric E X T l proteins (data not shown). 57 EXT1gfp+ER-EXT1myc | EXTl gfp+Golgi-EXT1 myc | EXT'gfp+Sec-EXT1 myc F ER-EXT1gfp+ER-EXT1 myc G • . . .^ ER-EXT1gfp+Golgi-EXT1myc H ER-EXT1gfp+Sec-EXT1myc I > H p * d i • GoIgi-EXT1gf||KGolgi-EXT1myc Golgi-EXTlgfp+Sec-EXT1 myc K Sec-EXT1gfp-*-Sec-EXT1myc Figure 15 Subcellular localization patterns of co-transfected chimeric EXT1 with chimeric EXTL BHK monolayers were cotransfected with the following constructs: A) EXT1+EXT2, B) EXT1+EXT1, C) EXT1+ER-EXT1, D) EXT1+Golgi-EXT1, E) EXT1 +Sec-EXT1, F) ER-EXT1+ER-EXT1, G) ER-EXT1+Golgi-EXT1, H) ER-EXT1+Sec-EXT1. I) Golgi-EXT1+GolghEXT1, J) Golgi-EXT1+Sec-EXT1, and K) Sec-EXT1+Sec-EXT1. One chimeric EXT1 partner was detected with a-myc antibodies followed by Cy5-conjugated secondary antibodies. The subellular distribution of myc-tagged or gfp-tagged proteins were observed by immunofluorescence and images acquired using either the Zeiss DVC fluorescent microscope or the Bio-Radiance Plus confocal microscope. Individual gfp- or myc-immunofluoresence images were overlaid together and presented here. 58 Discussion An ER-, Golgi- and secretory form of EXT l and EXT2 were created to determine the importance of the N-terminal region for in vivo HS polymerase activity, and its role in complex formation. The effects that the substituted localization signals had on Golgi-localization as indicated by immunofluorescence staining when chimeric EXT constructs were co-transfected together were also studied. 4.1 Subcellular localization of chimeric constructs For type II transmembrane proteins, a double arginine motif in the cytoplasmic N-terminus of the ER-localized isoform of the human invariant chain has been identified as an ER-localization signal (Schutze et al., 1994). This motif localized a type II transmembrane protein to the ER when transplanted onto the cell surface transferrin receptor. When this same motif was substituted into the cytoplasmic N-termini of E X T l and EXT2, both chimeric constructs localized to the ER when overexpressed individually as suggested by the staining patterns observed (Fig. 10). But when co-transfected with their respective wt partners, they were unable to retain both themselves and their partner in the ER as might have been expected based on previously published results (Nilsson etal, 1994). Nilsson et al. demonstrated that the substitution of an ER localization signal onto the medial Golgi protein, NAGTI caused another endogenous medial Golgi protein, Mannll to relocalize to the ER. The results described in the present study suggest that the Golgi-localization mechanism used by E X T l and EXT2 overrides the "RR" ER-localization motif. 59 The transmembrane domain and adjacent sequences of Golgi glycosyltransferases have been shown to be sufficient to localize type II transmembrane proteins to the Golgi (Burke et al., 1994; Munro, 1991). Furthermore, the first 169 aa of the N-terminus of human NDST1 was able to localize a reporter protein to the Golgi (Humphries et al., 1997). Thus the first 81 aa encompassing the transmembrane domain and adjacent sequences of murine NDST2, an isozyme of NDST1 were substituted into the N-termini of E X T l and EXT2. From the staining patterns observed, this putative Golgi-localization signal was unable to localize the chimeric constructs of E X T l and EXT2 strictly to the Golgi though it greatly improved the Golgi-localization of other chimeric constructs co-transfected with it (Fig. 11). This result suggests that some mechanism other than the T M domain and amino acids located immediately on either side is also involved in the Golgi-localization of E X T l and EXT2. Variations in subcellular localization for the same Golgi localization signal have been seen in different cell types (Tang et al., 1995). Thus it is possible that the NDST2-Golgi-localization signal is not sufficient to localize the chimeric constructs to the Golgi in BHK cells but would be adequate in other cell types though similar results were observed for the Golgi constructs in both BHK and Sog9 cells (data not shown). The immunoglobulin kappa chain leader sequence in the SecTag2 vector appears to allow E X T l and EXT2 to be efficiently packaged into secretory vesicles (Fig. 10). Interestingly, despite the secretory signal and the loss of its cytoplasmic N-terminus, T M domain and 10 aa of the stem region, it still localized to the Golgi as suggested by the perinuclear staining observed when co-transfected with its wt partner (Fig. 11). This suggests that it is forming a heterocomplex with its wt partner and its partner is anchoring both itself and the secretory EXT protein in the Golgi. 60 The subcellular localization as suggested by the different staining patterns observed when the chimeric constructs were mono-transfected or when various combinations of chimeric EXTs were co-transfected together should be confirmed in the future by co-localization with known subcellular markers. However, this study does provide information as to the effect the different sorting signals have on the chimeras as well as providing intriguing clues to Golgi localization due to heterocomplex formation. 4.2 In vivo heparan sulfate activity Despite the substitutions of different subcellular signals on E X T l , all chimeric E X T l constructs were still able to synthesize HS in vivo as determined by the HSV assay though at different efficiencies (Fig. 9). The transient expression of Golgi-EXTl (Fig. 9D) allowed better infection of Sog9 cells by HSV compared to wt E X T l (Fig. 9A). This further supports the belief that HS polysaccharide chain polymerization occurs in the Golgi as suggested by the finding that GlcA-T and GlcNAc-T in vitro activity is coupled to the Golgi (Fernandez and Warren, 1998). In order for E X T l to localize to the Golgi efficiently, heterocomplex formation with EXT2 appears to be required thus overexpressed wt E X T l is constrained by the limiting amounts of endogenous EXT2 available. In contrast, Golgi-EXTl is able to partially bypass the EXT2 requirement for Golgi-localization with its putative Golgi-localization signal thereby increasing the availability of the enzyme in the Golgi for HS biosynthesis. Interestingly, ER-EXT1 also restored HS biosynthesis in Sog9 cells despite an ER localization signal (Fig. 9C). It is possible that the ability of ER-EXT1 to restore Sog9 susceptibility to HSV was an artifact of overexpression. The large amounts of proteins 61 expressed may have overwhelmed the ER leading to leakage of the chimera to the Golgi where it is then able to synthesize HS chains. Another possibility is that ER-EXT1 formed a heterocomplex with endogenous EXT2 thus localizing to the Golgi as suggested by the results of co-transfecting ER-EXT1 with wt EXT2 in which both proteins displayed immunofluorescence patterns characteristic of the Golgi when both partners are abundantly available. In particular, the Sec-EXTl construct was still able to synthesize HS though its putative transmembrane domain and adjacent sequences were replaced with a secretory signal. The GlcA-T activity has been localized to the middle of the lumenal C-terminus and the GlcNAc-T catalytic site is believed to be in the C-terminal region as well, thus Sec-EXT l ' s N-terminal substitution may not be expected to abrogate in vitro activity but surprisingly, it still did not abolish in vivo HS biosynthesis. Since it is able to synthesize HS in vivo, it further suggests that it is forming a heterocomplex with wt EXT2 in the Golgi and that this heterocomplex is stable enough for polysaccharide chain elongation to occur. The other possibility is that Sec-EXTl is secreted out of the cell and polymerizing the HS polysaccharide chain onto nascent proteins in the extracellular space. This however, is highly unlikely as the substrates would have to be present in sufficient quantities for the enzyme to encounter. Furthermore, after the polysaccharide chains were synthesized, they would still have to undergo modification by other enzymes (i.e. NDST) to become sulfated. The HSV assay is a useful assay for determining whether HS is being produced. As long as HS is present, HSV can use it to attach to the cell surface and infect the cells. However, the assay can't detect differences in the amount, length or in the type (i.e. degree 62 of sulfation) of HS being produced. Nevertheless, it is a useful assay for investigating the in vivo HS biosynthetic function of mutagenized E X T l . 4.3 Heterocomplex formation From the ELISA results (Fig. 14, and data not shown), I conclude that the cytoplasmic N-terminus, T M domain and at least 10 aa of the stem region are not essential for heterocomplex formation. It appears that the lumenal C-termini of E X T l and EXT2 alone are sufficient for heterocomplex formation as evidenced by the ability of Sec-EXTl and Sec-EXT2 to oligomerize. The hetero-oligomerization between Sec-EXTl and wt EXT2 is consistent with the results obtained in the HSV assay and the subcellular localization pattern analysis. Taken together, the data indicate that Sec-EXTl is retained in the Golgi due to heterocomplex formation with endogenous wt EXT2 and is thus able to synthesize HS chains for HSV to bind and infect Sog9 cells. Further support that the lumenal portions of E X T l and EXT2 are sufficient to generate the heterocomplex is provided by data showing an increase in the in vitro GlcNAc-T and GlcA-T activity of co-expressed E X T l and EXT2 constructs lacking the TM domains, compared to individually expressed E X T l or EXT2 (Senay et ai, 2000). 4.4 Involvement of other proteins Yeast two-hybrid screens using a fragment containing a conserved portion of the E X T l or EXT2 C-terminus with the human keratinocyte and lymphocyte cDNA libraries revealed 2 interacting proteins (Simmons et ai, 1999). EXT2 was found to interact with both GalNAc-T5 and TRAP1 (tumor necrosis factor type 1 associated protein) while E X T l 63 only interacted with TRAP1. Furthermore, they found that some etiological mutants failed to interact with TRAP1 and GalNAc-T5, suggesting a role for these two proteins in HME. GalNAc-T5 may be involved in HS biosynthesis while TRAP1, a heat shock protein with homology to the hsp90 family, has been demonstrated to interact with retinoblastoma protein (Rb) and type 1 receptor for tumor necrosis factor (Chen et ah, 1996; Song et al., 1995). A recent study on TRAP1 suggests that in addition to playing a role in transforming growth factor-beta signalling, it acts as a chaperone for Smad2 as well (Wurthner et al., 2001). Thus I decided to see if I could detect any other proteins involved in the EXT1-EXT2 heterocomplex. Cells co-transfected with E X T l and EXT2 were radiolabelled, the EXT proteins were immunoprecipitated, electrophoretically separated and analyzed. No other protein bands were observed compared to the negative controls (Fig. 14). E X T l and EXT2 were able to form a heterocomplex resulting in Golgi localization despite being overexpressed in cells (McCormick et al., 2000). This suggests that no other proteins are required for formation of the heterocomplex unless the other members of the complex were also being expressed in large quantities by the cell. Unfortunately, there were a lot of background bands present which may interfere with the detection of other proteins involved in the heterocomplex. The background bands remained despite attempts to clear them up thus running a 2-D SDS-PAGE of the immunoprecipitated proteins may separate the different proteins enough to identify novel partners in the heterocomplex. Another approach to confirm that TRAP-1 and GalNAc-T5 are involved in the heterocomplex may be to immunoprecipitate E X T l or EXT2 and probe specifically for TRAP-1 or GalNAc-T5. It is also possible that TRAP1, GalNAc-T5 64 or other interacting proteins are expressed at levels too low for detection by the radiolabeled immunoprecipitation assay or that the interactions between these proteins are only transient. 4.5 Model of Golgi retention Golgi resident proteins must be maintained in the Golgi despite the large flow of proteins passing through heading for various destinations. Thus the mechanism of Golgi retention must be able to distinguish between Golgi residents and non-residents. To account for the maintenance of Golgi-residents but not transient proteins in the Golgi, two main models have been proposed: the bilayer thickness model and the kin-recognition model. The bilayer thickness model postulates that the shorter length of the transmembrane domain of Golgi-resident proteins versus those of the PM proteins prevents the protein from entering secretory vesicles and is thus retained within the Golgi (Bretscher and Munro, 1993). The kin-recognition model proposes that Golgi-residents oligomerize to form structures too large to enter transport vesicles, and are thus retained within the Golgi (Nilsson et al., 1993). Although both models are supported by experimental evidence, neither model alone can account for how all Golgi-resident glycosyltransferases are retained. It is possible that different Golgi residents use different mechanisms to remain in the Golgi or that both models contribute to Golgi retention. The two models may work in concert, with the shorter length of the transmembrane of Golgi residents functioning to slow entry into transport vesicles thus allowing the proteins to interact with other Golgi proteins, thereby forming oligomers too large to exit the Golgi. 65 To further complicate the matter, the Golgi is composed of several discrete cisternae and individual Golgi residents are found only in certain cisternae and not others. In fact, many modification enzymes are found in the order in which they act on substrates. Thus further research needs to be done to account for the maintenance of distinct subpopulations of Golgi residents in specific compartments of the Golgi. The co-localization patterns observed for the chimeric E X T l and wt EXT2 proteins in the Golgi, especially with the chimeric Sec-EXTl and wt EXT2, suggest that the major Golgi-localization mechanism used by EXT l and EXT2 is hetero-oligomer formation as described by the kin recognition model (Nilsson et al., 1993). Though the kin recognition model predicts that homo-oligomers also contribute to Golgi retention, they do not appear to contribute significantly to Golgi-localization in the case of E X T l (data not shown). This is in agreement with the fact that overexpression of either E X T l or EXT2 have been shown to lead to ER- and not Golgi-localization (McCormick et al, 2000). Although E X T l forms homo-oligomers, they may not be able to interact with other Golgi proteins (i.e. TRAP1) without the help of EXT2, or they may be unable to form a large enough complex to prevent entry into transport vesicles. From all the data currently available, heterocomplex formation of E X T l with EXT2 either allows a chaperone to bind and take the complex to the Golgi, or else E X T l and EXT2 normally cycle between the ER and Golgi but upon heterocomplex formation, a conformational change occurs in one or both partners allowing the binding of more EXT1/EXT2 heterocomplexes or other Golgi-localized proteins in a multicomplex (Fig. 16). This multicomplex would then be retained in the Golgi and prevented from entering 66 secretory vesicles destined for the cell surface, or COPI vesicles for retrograde transport back to the ER. • • | l ! • M B —1 a t o Q I 0 « S s - -2 ,5 68 4.6 Future directions Sec-EXTl and Sec-EXT2 were able to form a heterocomplex despite the loss of their cytoplasmic N-termini, putative transmembrane domains and 10 aa of their stem regions. Thus it would be very interesting to investigate which amino acids in the lumenal C-terimini of E X T l and EXT2 are involved in heterocomplex formation. The heterocomplex appears to be relatively stable as it is able to withstand cell lysis and handling during the immunoprecipitation step, therefore disulfide bonds may be involved in the interaction between the two proteins. Destroying the ability of E X T l and EXT2 to form intermolecular disulfide bonds by mutating cysteine residues in the lumenal domains of E X T l and EXT2 may provide further information about heterocomplex formation. Three dimensional crystal structure analysis of the heterocomplex versus EXT 1 or EXT2 alone should also provide clues to structure-function relationship as the heterocomplex contains higher glycosyltransferase activity in vitro than individual E X T l or EXT2 alone (McCormick et al., 2000). Furthermore, future experiments to determine the ability of etiological E X T l mutants to hetero-oligomerize may help clarify the importance of heterocomplex formation in HME. 69 References Ahn, J., Ludecke, H. J., Lindow, S., Horton, W. A., Lee, B., Wagner, M. J., Horsthemke, B., and Wells, D. E. (1995). Cloning of the putative tumour suppressor gene for hereditary multiple exostoses (EXTl). Nat Genet 11, 137-143. Alvarez-Dominguez, C , Vazquez-Boland, J. A., Carrasco-Marin, E., Lopez-Mato, P., and Leyva-Cobian, F. (1997). Host cell heparan sulfate proteoglycans mediate attachment and entry of Listeria monocytogenes, and the listerial surface protein ActA is involved in heparan sulfate receptor recognition. Infection and Immunity 65, 78-88. Aridor, M., Bannykh, S. I., Rowe, T., and Balch, W. E. (1995). Sequential coupling between COPTJ and COPI vesicle coats in endoplasmic reticulum to Golgi transport. J Cell Biol 131, 875-893. Banfield, B. W., Leduc, Y., Esford, L., Schubert, K., and Tufaro, F. (1995). Sequential isolation of proteoglycan synthesis mutants by using herpes simplex virus as a selective agent: evidence for a proteoglycan-independent virus entry pathway. J Virol 69, 3290-3298. Barlowe, C , Orci, L., Yeung, T., Hosobuchi, M., Hamamoto, S., Salama, N., Rexach, M. F., Ravazzola, M., Amherdt, M., and Schekman, R. (1994). COPII: a membrane coat formed by Sec proteins that drive vesicle budding from the endoplasmic reticulum. Cell 11, 895-907. Bellaiche, Y., The, I., and Perrimon, N. (1998). Tout-velu is a Drosophila homologue of the putative tumour suppressor EXT-1 and is needed for Hh diffusion. Nature 394, 85-88. Binari, R. C , Staveley, B. E., Johnson, W. A., Godavarti, R., Sasisekharan, R., and Manoukian, A. S. (1997). Genetic evidence that heparin-like glycosaminoglycans are involved in wingless signaling. Development 124, 2623-32. Bretscher, M. S., and Munro, S. (1993). Cholesterol and the Golgi apparatus. Science 261, 1280-1281. Bullock, S. L., Fletcher, J. M., Beddington, R. S., and Wilson, V. A. (1998). Renal agenesis in mice homozygous for a gene trap mutation in the gene encoding heparan sulfate 2-sulfotransferase. Genes Dev 12, 1894-1906. Burke, J., Pettitt, J. M., Humphris, D., and Gleeson, P. A. (1994). Medial-Golgi retention of N-acetylglucosaminyltransferase I. Contribution from all domains of the enzyme. J Biol Chem 269, 12049-12059. Cai, W. H., Gu, B., and Person, S. (1988). Role of glycoprotein B of herpes simplex virus type 1 in viral entry and cell fusion. J Virol 62, 2596-2604. 70 Chen, C , Ma, J., Lazic, A., Backovic, M., and Colley, K. J. (2000). Formation of insoluble oligomers correlates with ST6Gal I stable localization in the golgi. J Biol Chem 275, 13819-13826. Chen, C. F., Chen, Y., Dai, K., Chen, P. L., Riley, D. J., and Lee, W. H. (1996). A new member of the hsp90 family of molecular chaperones interacts with the retinoblastoma protein during mitosis and after heat shock. Mol Cell Biol 16, 4691-4699. Cheung, P. K., McCormick, C , Crawford, B. E., Esko, J. D., Tufaro, F., and Duncan, G. (2001). Etiological point mutations in the hereditary multiple exostoses gene E X T l : a functional analysis of heparan sulfate polymerase activity. Am J Hum Genet 69, 55-66. Clines, G. A., Ashley, J. A., Shah, S., and Lovett, M. (1997). The structure of the human multiple exostoses 2 gene and characterization of homologs in mouse and Caenorhabditis elegans. Gen Res 7, 359-367. Colley, K. J. (1997). Golgi localization of glycosyltransferases - more questions than answers. Glycobioll, 1-13. Cook, A., Raskind, W., Blanton, S. FL, Pauli, R. M., Gregg, R. G., Francomano, C. A., Puffenberger, E., Conrad, E. U., Schmale, G., Schellenberg, G., Wijsman, E., Hecht, J. T., Wells, D., and Wagner, M. J. (1993). Genetic heterogeneity in families with hereditary multiple exostoses. Am J Hum Genet 53, 71-79. Cosson, P., and Letourneur, F. (1994). Coatomer interaction with di-lysine endoplasmic reticulum retention motifs. Science 263, 1629-1631. Dahl, M. T. (1993). The gradual correction of forearm deformities in multiple hereditary exostoses. Hand Clin 9, 707-718. Dolan, A., Jamieson, F. E., Cunningham, C , Barnett, B. C , and McGeoch, D. J. (1998). The genome sequence of herpes simplex virus type 2. Journal of Virology 72, 2010-2021. Fernandez, C. J., and Warren, G. (1998). In vitro synthesis of sulfated glycosaminoglycans coupled to inter-compartmental Golgi transport. J Biol Chem 273, 19030-19039. Fuki, I. V., Kuhn, K. M., Lomazov, I. R., Rothman, V. L., Tuszynski, G. P., Iozzo, R. V., Swenson, T. L., Fisher, E. A., and Williams, K. J. (1997). The syndecan family of proteoglycans. Novel receptors mediating internalization of atherogenic lipoproteins in vitro. J Clin Invest 100, 1611-1622. Fuller, A. O., Santos, R. E., and Spear, P. G. (1989). Neutralizing antibodies specific for glycoprotein H of herpes simplex virus permit viral attachment to cells but prevent penetration. / Virol 63, 3435-3443. 71 Geraghty, R. J., Krurnmenacher, C., Cohen, G. FL, Eisenberg, R. J., and Spear, P. G. (1998). Entry of alphaherpesviruses mediated by poliovirus receptor-related protein 1 and polio virus receptor. Science 280, 1618-1620. Geuijen, C. A. W., Willems, R. J. L., and Mooi, F. R. (1996). The major fimbrial subunit of Bordetella pertussis binds to sulfated sugars. Infection and Immunity 64, 2657-2665. Gu, F., Crump, C. M., and Thomas, G. (2001). Trans-Golgi network sorting. Cell. Mol. Life Sci. 58, 1067-1084. Hacker, U., Lin, X., and Perrimon, N. (1997). The Drosophila sugarless gene modulates Wingless signaling and encodes an enzyme involved in polysaccharide biosynthesis. Development 124, 3565-3573. Haerry, T. E., Heslip, T. R., Marsh, J. L., and O'Connor, M. B. (1997). Defects in glucuronate biosynthesis disrupt Wingless signaling in Drosophila. Development 124, 3055-3064. Hecht, J. T., Hogue, D., Strong, L. C , Hansen, M. F., Blanton, S. H., and Wagner, M. (1995). Hereditary multiple exostosis and chondrosarcoma: linkage to chromosome 11 and loss of heterozygosity for EXT-linked markers on chromosomes 11 and 8. Am J Hum Genet 56, 1125-1131. Hennekam, R. C. (1991). Hereditary multiple exostoses. J of Med Genet 28, 262-266. Herold, B. C , Gerber, S. I., Belval, B. J., Siston, A. M., and Shulman, N. (1996). Differences in the susceptibility of herpes simplex virus types 1 and 2 to modified heparin compounds suggest serotype differences in viral entry. Journal of Virology 70, 3461-3469. Herold, B. C , Visalli, R. J., Susmarski, N., Brandt, C. R., and Spear, P. G. (1994). Glycoprotein C-independent binding of herpes simplex virus to cells requires cell surface heparan sulphate and glycoprotein B. J Gen Virol 75, 1211-1222. Herold, B. C , WuDunn, D., Soltys, N., and Spear, P. G. (1991). Glycoprotein C of herpes simplex virus type 1 plays a principal role in the adsorption of virus to cells and in infectivity. J Virol 65, 1090-1098. Humphries, D. E., Sullivan, B. M., Aleixo, M. D., and Stow, J. L. (1997). Localization of human heparan glucosaminyl N-deacetylase/N-sulphotransferase to the trans-Golgi network. Biochem J 325, 351-357. Janeway Jr., C. A., and Travers, P. (1996). "Immunobiology: the immune system in health and disease." Current Biology Ltd./Garland Publishing Inc., 72 Jones, S. M., Crosby, J. R., Salamero, J., and Howell, K. E. (1993). A cytosolic complex of p62 and rab6 associates with TGN38/41 and is involved in budding of exocytic vesicles from the trans-Golgi network. / Cell Biol 122, 775-788. Karger, A., and Mettenleiter, T. C. (1996). Identification of cell surface molecules that interact with pseudorabies virus. Journal of Virology 70, 2138-2145. Kim, B. T., Kitagawa, H., Tamura, J., Saito, T., Kusche-Gullberg, M., Lindahl, U., and Sugahara, K. (2001). Human tumor suppressor EXT gene family members EXTL1 and EXTL3 encode alpha 1,4- N-acetylglucosaminyltransferases that likely are involved in heparan sulfate/ heparin biosynthesis. Proc Natl Acad Sci U S A 98, 7176-7181. Kitagawa, H., Shimakawa, H., and Sugahara, K. (1999). The tumor suppressor EXT-like gene EXTL2 encodes an a l , 4-TY-acetylhexosaminytransferase that transfers N-acetylgalactosamine and TV-acetylglucosamine to the common glycosaminoglycan-protein linkage region. J. Biol. Chem. 274, 13933-13937. Kobayashi, S., Morimoto, K., Shimizu, T., Takahashi, M., Kurosawa, H., and Shirasawa, T. (2000). Association of E X T l and EXT2, hereditary multiple exostoses gene products, in Golgi apparatus. Biochem Biophys Res Commun 268, 860-867. Krooth, R. S., Macklin, M. T., and Kilbish, T. F. (1961). Diaphyseal aclasis (multiple exostoses) on Guam. J. Hum. Genet. 13, 340-347. Kusche-Gullberg, ML, Eriksson, I., Pikas, D. S., and Kjellen, L. (1998). Identification and expression in mouse of two heparan sulfate glucosaminyl N-deacetylase/N-sulfotransferase genes. J Biol Chem 273, 11902-11907. Ladinsky, M. S., Kremer, J. R., Furcinitti, P. S., Mcintosh, J. R., and Howell, K. E. (1994). H V E M tomography of the trans-Golgi network: structural insights and identification of a lace-like vesicle coat. J Cell Biol 127, 29-38. Le Merrer, M., Legeai-Mallet, L., Jeannin, P. M., Horsthemke, B., Schinzel, A., Plauchu, H., Toutain, A., Achard, F., Munnich, A., and Maroteaux, P. (1994). A gene for hereditary multiple exostoses maps to chromosome 19p. Hum Mol Genet 3, 717-722. Lin, X., Wei, G., Shi, Z., Dryer, L., Esko, J. D., Wells, D. E., and Matzuk, M. M. (2000). Disruption of gastrulation and heparan sulfate biosynthesis in EXTl-deficient mice. Dev Biol 224, 299-311. Lin, X., and Wells, D. (1997). Isolation of the mouse cDNA homologous to the human E X T l gene responsible for Hereditary Multiple Exostoses. DNA Sequence 7, 199-202. Lind, T., Tufaro, F., McCormick, C , Lindahl, U., and Lidholt, K. (1998). The putative tumor suppressors E X T l and EXT2 are glycosyltransferases required for the biosynthesis of heparan sulfate. Journal of Biological Chemistry 273, 26265-26268. 73 Lotteau, V., Teyton, L., Peleraux, A., Nilsson, T., Karlsson, L., Schmid, S. L., Quaranta, V., and Peterson, P. A. (1990). Intracellular transport of class I IMHC molecules directed by invariant chain. Nature 348, 600-605. McCormick, C , Duncan, G., Goutsos, K. T., and Tufaro, F. (2000). The putative tumor suppressors E X T l and EXT2 form a stable complex that accumulates in the Golgi apparatus and catalyzes the synthesis of heparan sulfate. Proceedings of the National Academy of Sciences of the United States of America 97, 668-673. McCormick, C , Leduc, Y., Martindale, D., Mattison, K., Esford, L. E., Dyer, A. P., and Tufaro, F. (1998). The putative tumour suppressor E X T l alters the expression of cell-surface heparan sulfate. Nature Genetics 19, 158-161. McGeoch, D. J., Dalrymple, M. A., Davison, A. J., Dolan, A., Frame, M. C , McNab, D., J., P. L., Scott, J. E., and Taylor, P. (1988). The complete DNA sequence of the long unique region in the genome of herpes simplex virus type I. Journal of General Virology 69, 1531-1574. Montgomery, R. I., Warner, M. S., Lum, B. J., and Spear, P. G. (1996). Herpes simplex virus-1 entry into cells mediated by a novel member of the TNF/NGF receptor family. Cell 87, 427-436. Munro, S. (1991). Sequences within and adjacent to the transmembrane segment of alpha-2,6-sialyltransferase specify Golgi retention. Embo J10, 3577-3588. Munro, S. (1995). An investigation of the role of transmembrane domains in Golgi protein retention. Embo J14, 4695-4704. Nezil, F. A., and Bloom, M. (1992). Combined influence of cholesterol and synthetic amphiphillic peptides upon bilayer thickness in model membranes. Biophys J 61, 1176-1183. Nickel, W., Brugger, B., and Wieland, F. T. (1998). Protein and lipid sorting between the endoplasmic reticulum and the Golgi complex. Semin Cell Dev Biol 9, 493-501. Nickel, W., Huber, L. A., Kahn, R. A., Kipper, N., Barthel, A., Fasshauer, D., and Soling, H. D. (1994). ADP ribosylation factor and a 14-kD polypeptide are associated with heparan sulfate-carrying post-trans-Golgi network secretory vesicles in rat hepatocytes. J Cell Biol 125, 721-732. Nilsen, A., and Myrmel, H. (2000). Changing trends in genital herpes simplex virus infection in Bergen, Norway. Acta Obstet Gynecol Scand 79, 693-696. Nilsson, T., Hoe, M. H., Slusarewicz, P., Rabouille, C , Watson, R., Hunte, F., Watzele, G., Berger, E. G., and Warren, G. (1994). Kin recognition between medial Golgi enzymes in HeLa cells. EMBO J. 13, 562-574. 74 Nilsson, T., Lucocq, J. M., Mackay, D., and Warren, G. (1991). The membrane spanning domain of p-l,4-galactosyltransferase specifies trans Golgi localization. EMBO Journal 10, 3567-3575. Nilsson, T., Slusarewicz, P., Hoe, M. H., and Warren, G. (1993). Kin recognition: a model for the retention of Golgi enzymes. FEBS Lett. 330, 1-4. Olson, S. T., Bjork, I., Sheffer, R., Craig, P. A., Shore, J. D., and Choay, J. (1992). Role of the antithrombin-binding pentasaccharide in heparin acceleration of antithrombin-proteinase reactions. Resolution of the antithrombin conformational change contribution to heparin rate enhancement. Journal of Biological Chemistry 267, 12528-12538. Orci, L., Glick, B. S., and Rothman, J. E. (1986). A new type of coated vesicular carrier that appears not to contain clathrin: its possible role in protein transport within the Golgi stack. Cell 46, 171-184. Paik, N. J., Han, T. R., and Lim, S. J. (2000). Multiple peripheral nerve compressions related to malignantly transformed hereditary multiple exostoses. Muscle Nerve 23, 1290-1294. Pearse, B. M., and Robinson, M. S. (1990). Clathrin, adaptors, and sorting. Annu Rev Cell Biol 6, 151-171. Philippe, C , Porter, D. E., Emerton, M. E., Wells, D. E., Simpson, A. H., and Monaco, A. P. (1997). Mutation screening of the E X T l and EXT2 genes in patients with hereditary multiple exostoses. Am J Hum Genet 61, 520-528. Raskind, W. H., Conrad, E. U., Chansky, H., and Matsushita, M. (1995). Loss of heterozygosity in chondrosarcomas for markers linked to hereditary multiple exostoses loci on chromosomes 8 and 11. Am J Hum Genet 56, 1132-1139. Roop, C , Hutchinson, L., and Johnson, D. C. (1993). A mutant herpes simplex virus type 1 unable to express glycoprotein L cannot enter cells, and its particles lack glycoprotein H. J Virol 67, 2285-2297. Saxena, U., Klein, M. G., and Goldberg, I. J. (1990). Metabolism of endothelial cell-bound lipoprotein lipase. Evidence for heparan sulfate proteoglycan-mediated internalization and recycling. J Biol Chem 265, 12880-12886. Schmale, G. A., Conrad, E. U., and Raskind, W. H. (1994). The natural history of hereditary multiple exostoses. J Bone J Surg 76, 986-992. Schutze, M. P., Peterson, P. A., and Jackson, M. R. (1994). An N-terminal double-arginine motif maintains type II membrane proteins in the endoplasmic retiuculum. EMBO J. 13, 1696-1705. 75 Senay, C , Lind, T., Muguruma, K., Tone, Y., Kitagawa, H., Sugahara, K., Lidholt, K., Lindahl, U., and Kusche-Gullberg, M. (2000). The EXT1/EXT2 tumor suppressors: catalytic activities and role in heparan sulfate biosynthesis. EMBO Reports 1, 282-286. Simmons, A. D., Musy, M. M., Lopes, C. S., Hwang, L. Y., Yang, Y. P., and Lovett, M. (1999). A direct interaction between EXT proteins and glycosyltransferases is defective in hereditary multiple exostoses. Hum Mol Genet 8, 2155-2164. Sollner, T., Whiteheart, S. W., Brunner, M., Erdjument-Bromage, H., Geromanos, S., Tempst, P., and Rothman, J. E. (1993). SNAP receptors implicated in vesicle targeting and fusion. Nature 362, 318-324. Solomon, E. P., Berg, L. R., Martin, D. W., and Villee, C. (1993). "Biology." Saunders College Publishing, Solomon, L. (1963). Hereditary multiple exostosis. J Bone J Surg 45, 292-304. Song, H. Y., Dunbar, J. D., Zhang, Y. X., Guo, D., and Donner, D. B. (1995). Identification of a protein with homology to hsp90 that binds the type 1 tumor necrosis factor receptor. J Biol Chem 270, 3574-3581. Stickens, D., Clines, G., Burbee, D., Ramos, P., Thomas, S., Hogue, D., Hecht, J. T., Lovett, M., and Evans, G. A. (1996). The EXT2 multiple exostoses gene defines a family of putative tumour suppressor genes. Nat Genet 14, 25-32. Stickens, D., and Evans, G. A. (1997). Isolation and characterization of the murine homolog of the human EXT2 multiple exostoses gene. Biochem Mol Med 61, 16-21. Strubin, M., Long, E. O., and Mach, B. (1986). Two forms of the la antigen-associated invariant chain result from alternative initiations at two in-phase AUGs. Cell 47, 619-625. Tang, B. L., Low, S. H., Wong, S. H., and Hong, W. (1995). Cell type differences in Golgi retention signals for transmembrane proteins. Eur J Cell Biol 66, 365-374. Teasdale, R. D., and Jackson, M. R. (1996). Signal-mediated sorting of membrane proteins between the endoplasmic reticulum and the Golgi apparatus. Anna Rev Cell Dev Biol 12, 27-54. The, I., Bellaiche, Y., and Perrimon, N. (1999). Hedgehog movement is regulated through tout velu-dependent synthesis of a heparan sulfate proteoglycan. Mol Cell 4, 633-639. Utani, A., Nomizu, M., Matsuura, H., Kato, K., Kobayashi, T., Takeda, U., Aota, S., Nielsen, P. K, and Shinkai, H. (2001). A unique sequence of the laminin alpha 3-G domain binds to heparin and promotes cell adhesion through syndecan-2 and -4. J Biol Chem 276, 28779-28788. 76 Van Hul, W., Wuyts, W., Hendrickx, J., Speleman, F., Wauters, J., De Boulle, K. , Van Roy, N . , Bossuyt, P., and Willems, P. J. (1998). Identification of a third EXT-like gene (EXTL3) belonging to the E X T gene family. Genomics 47, 230-237. Warner, M . S., Geraghty, R. J., Martinez, W. M . , Montgomery, R. I., Whitbeck, J. C., Xu , R., Eisenberg, R. J., Cohen, G. H. , and Spear, P. G. (1998). A cell surface protein with herpesvirus entry activity (HveB) confers susceptibility to infection by mutants of herpes simplex virus type 1, herpes simplex virus type 2, and pseudorabies virus. Virology 246, 179-189. Wei, G., Bai, X . , Gabb, M . M . , Bame, K. J., Koshy, T. I., Spear, P. G., and Esko, J. D. (2000). Location of the glucuronosyltransferase domain in the heparan sulfate copolymerase E X T l by analysis of Chinese hamster ovary cell mutants. Journal of Biological Chemistry 275, 27733-27740. Weisz, O. A. , Swift, A . M . , and Machamer, C. E. (1993). Oligomerization of a membrane protein correlates with its retention in the Golgi complex. J Cell Biol 122, 1185-1196. Williams, R. K. , and Straus, S. E. (1997). Specificity and affinity of binding of herpes simplex virus type 2 glycoprotein B to glycosaminoglycans. J Virol 71, 1375-1380. Wise, C. A. , Clines, G. A. , Massa, H., Trask, B. J., and Lovett, M . (1997). Identification and localization of the gene for E X T L , a third member of the multiple exostoses gene family. Gen Res 7, 10-16. Wu, Y . Q., Heutink, P., de Vries, B. B. , Sandkuijl, L . A. , van den Ouweland, A . M . , Niermeijer, M . F., Galjaard, H. , Reyniers, E., Willems, P. J., and Halley, D. J. (1994). Assignment of a second locus for multiple exostoses to the pericentromeric region of chromosome 11. Hum Mol Genet 3, 167-171. WuDunn, D., and Spear, P. G. (1989). Initial interaction of herpes simplex virus with cells is binding to heparan sulfate. J. Virol. 63, 52-58. Wurthner, J. U . , Frank, D. B., Felici, A. , Green, H . M . , Cao, Z., Schneider, M . D., McNally, J. G., Lechleider, R. J., and Roberts, A . B. (2001). Transforming growth factor-beta receptor-associated protein 1 is a Smad4 chaperone. J Biol Chem 276, 19495-19502. Wuyts, W., and Van Hul, W. (2000). Molecular basis of multiple exostoses: mutations in the E X T l and EXT2 genes. Human Mutation 15, 220-227. Wuyts, W., Van Hul, W., Hendrickx, J., Speleman, F., Wauters, J., De Boulle, K. , Van Roy, N . , Van Agtmael, T., Bossuyt, P., and Willems, P. J. (1997). Identification and characterization of a novel member of the E X T gene family, E X T L 2 . Eur J Hum Genet 5, 382-389. 77 Wuyts, W., Van Hul, W., Wauters, J., Nemtsova, M . , Reyniers, E., Van Hul, E. V. , De Boulle, K., de Vries, B. B., Hendrickx, J., Herrygers, I., Bossuyt, P., Balemans, W., Fransen, E., Vits, L . , Coucke, P., Nowak, N . J., Shows, T. B. , Mallet, L . , van den Ouweland, A . M . , McGaughran, J., Halley, D. J., and Willems, P. J. (1996). Positional cloning of a gene involved in hereditary multiple exostoses. Hum Mol Genet 5, 1547-1557. Yayon, A. , Klagsbrun, M . , Esko, J. D., Leder, P., and Ornitz, D. M . (1991). Cell surface, heparin-like molecules are required for binding of basic fibroblast growth factor to its high affinity receptor. Cell 64, 841-848. 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0090120/manifest

Comment

Related Items