UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization and crystalization of Gelsolin Wang, Hui 2005

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

Item Metadata

Download

Media
831-ubc_2006-0336.pdf [ 14.83MB ]
Metadata
JSON: 831-1.0061124.json
JSON-LD: 831-1.0061124-ld.json
RDF/XML (Pretty): 831-1.0061124-rdf.xml
RDF/JSON: 831-1.0061124-rdf.json
Turtle: 831-1.0061124-turtle.txt
N-Triples: 831-1.0061124-rdf-ntriples.txt
Original Record: 831-1.0061124-source.json
Full Text
831-1.0061124-fulltext.txt
Citation
831-1.0061124.ris

Full Text

C H A R A C T E R I Z A T I O N A N D C R Y S T A L L I Z A T I O N OF G E L S O L I N : A C T I N C O M P L E X E S by HUI W A N G B.Sc, Nanjing University, 2003 A THESIS SUMBITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE In THE F A C U L T Y OF GRADUATE STUDIES (Chemistry) THE UNIVERSITY OF BRITISH C O L U M B I A Oct. 2005 © Hui Wang, 2005 Abstract Actin, a protein existing in all eukaryotic cells, is involved in such cellular processes as cell movement, cell division, apoptosis and so on. Gelsolin, which consists of six structurally similar domains (G1-G6), can regulate these processes by calcium activated disassembly of actin filaments. In inactive gelsolin, the six domains pack into a compact globule so that none of the actin binding sites are accessible. From inactive gelsolin to activated gelsolin, three latches are unlocked to expose actin binding sites. Although the structure of fully activated, intact gelsolin is not available, those of the C-terminal and N-terminal halves of gelsolin, respectively, each bound to one actin molecule, have been solved. These structures may be added on to a model for F-actin to generate a model for a gelsolin-capped actin filament. In order to test these models, we set about to crystallize complexes in which one, two and three actin units bind to one activated intact gelsolin, i.e. G A , G A 2 and GA3, respectively. This thesis centers on purification, characterization and crystallization of G A , G A 2 and GA3 complexes. We have crystallized G A , G A 2 and GA3, respectively, as confirmed by size exclusion chromatography and gel electrophoresis. We also confirmed our G A 2 and G A 3 complexes by M A L D I mass spectrometry experiments. Although only the N-terminal half of gelsolin bound to one actin is visible in our G A 2 and G A 3 crystals by X-ray diffraction analysis, the crystals possess a unit cell that is sufficiently large to accommodate the second half of gelsolin bound to a second actin in a variety of different positions. We are still awaiting diffraction data of G A crystals and E D C cross-linked GA3 - 3 . crystals. We also soaked G A 2 and GA3 crystals in Tb -containing solutions in order to investigate the exchangeability of C a 2 + in these complexes. i i Table of contents Abstract ii Table of contents iii List of Tables vi List of Figures vii List of Abbreviations xii Acknowledgements xiv Chapter One - Introduction 1 1.1 Actin 1 a. Introduction to actin 1 b. Structure of G-actin 1 c. Models of F-actin 4 d. Transition between G-actin and F-actin 8 1.2 Gelsolin 10 a. Introduction to gelsolin 10 b. Structure of inactive gelsolin 12 c. From inactive gelsolin to activated gelsolin 16 1.3 Gelsolin and Actin interaction 21 a. Gelsolin/actin complexes 21 b. Model of F-actin severing and capping by gelsolin 23 c. Models of gelsolin/actin complexes ..27 Chapter Two - Materials and Methods 31 2.1 Protein purification 31 iii a. Gelsolin purification 31 b. Actin purification 34 2.2 Formation and purification of complexes 35 a. GA 2 complex '• 35 b. GA3 complex 36 c. GA complex 36 2.3 SDS-PAGE 37 a. Introduction to SDS-PAGE 37 b. SDS-PAGE for Gelsolin, Actin, GA, GA 2 , GA 3 38 2.4 Cross-linking of G A 2 and G A 3 39 a. Cross linking regents 39 b. EDC Cross-linking of GA2 and GA 3 complexes 41 2.5 Protein Crystallization 41 a. Introduction 41 b. G A 2 complex 43 c. GA 3 complex 43 d. GA complex 44 e. EDC-cross linked GA 3 complex 44 2.6 X-ray crystallography 44 a. Introduction • 44 b. Freezing of G A 2 and GA 3 46 c. Freezing of GA 46 d. Terbium and Cadmium soaking of GA 2 and GA 3 47 iv e. Data collection 47 2.7 Luminescence spectroscopy 47 Chapter Three — Results and Discussion 49 3.1 Purification of gelsolin and complexes 49 a. Purification of gelsolin 49 b. Purification of actin 51 c. Purification of G A 2 52 d. Purification of GA3 53 e. Purification of GA 54 3.2 SDS-PAGE analyses 56 a. Gelsolin and actin 56 b. GA, GA2 and G A 3 in solution 58 c. G A 2 and G A 3 crystals 61 c. Cleaved gelsolin and actin complex in solution 63 3.3 Mass spectrometry results 66 3.4 GA, G A 2 and G A 3 crystals and X-ray diffraction results 72 3.5 Luminescence Results of T b 3 + titration of gelsolin 79 3.6 Conclusions and Future Work 83 References 87 V List of Tables Table 1. The salt gradient program for H P L C used to elute the gelsolin 32 Table 2. Alternative G A complex elution buffers 37 Table 3. Reactive Cross-linker Groups and Their Functional Group Targets 40 Table 4. Data collection and molecular replacement statistics for terbium soaked crystals. 76 v i List of Figures Fig. 1. A ) Atomic structure of an actin monomer derived from actin/DNase I cocrystals at 2.8 A resolution; B) Atomic structure of an actin monomer derived from uncomplexed actin crystals at 1.54 A resolution. 2 Fig . 2. A ) Closed state of actin monomer (PDB code 2BTF) ; B) Open state of actin monomer (PDB code 1HLU) . 3 Fig. 3. A ) The atomic structure of uncomplexed actin in the A D P state; B) A hydrophobic pocket at the front end of the hydrophobic cleft in actin is accessible to both the G-actin-and F-actin-binding proteins. 4 Fig . 4. A ) Two descriptions of F-actin; B) Holmes model of F-Act in and one actin unit from the Holmes model of F-Act in . 6 Fig. 5. The Holmes model of F-actin. 7 Fig. 6. The assembly of an actin monomer onto an actin filament. 8 Fig . 7. The regulation of actin filaments by A B P s in the cell. 10 Fig. 8. Schematic of the aberrant proteolysis of D187N/Y gelsolin. 12 2"b Fig. 9. The structure of inactive plasma gelsolin in the absence of Ca . 13 Fig. 10. Individual domains of inactive gelsolin excised from the inactive structure of plasma gelsolin. 14 Fig . 11. The N-terminal half and C-terminal half of inactive gelsolin excised from the structure of inactive gelsolin. 15 Vll Fig . 12. Locations of type I calcium (golden spheres) and type II calcium (gray spheres) binding sites in inactive gelsolin. 17 Fig. 13. A ) Activation of gelsolin in presence of calcium; B) Different levels of calcium activation. 19 Fig. 14. A ) Transition from G1-G3 in inactive gelsolin to its active conformation; B) Transition from G4-G6 in inactive gelsolin to its active conformation. 21 Fig. 15. A ) Structure of G4-G6 bound to actin in the presence of calcium; B) Structure of G1-G3 bound to actin in the presence of calcium. 24 Fig. 16. Two possible models for the capped barbed end of F-actin. 25 Fig. 17. The binding to and severing of an actin fdament by gelsolin in the presence of calcium. 26 Fig. 18. Proposed G A 2 model. 28 Fig . 19. Proposed G A 3 model. 29 Fig. 20. M i n i - P R O T E A N 3 Electrophoresis. 39 Fig . 21. Hanging drop crystallization set up. 43 Fig. 22. Elution of gelsolin from a DEAE-Sephadex A-50 ion exchange column (36 x 4.5 cm) by a gradient of N a C l concentration in 25 m M Tr i s -HCl , 1 m M E D T A , p H 7.8, at the rate of 2 ml/min. 49 Fig. 23. Elution of gelsolin from a size exclusion column. 50 v i i i Fig . 24. Elution of actin from a size exclution column. 51 Fig. 25. Elution of G A 2 complex from a size exclusion column. 52 Fig. 26. Elution of a complex formed between actin and a 68 kDa fragment of gelsolin from a size exclusion column. 53 Fig . 27. Elution of GA3 complex from a size exclusion column. 54 Fig. 28. Elution of G A complex from a size exclusion column. 55 Fig. 29. Elution of G A complex (gelsolin used to form G A complex contained gelsolin and cleaved gelsolin) from a size exclusion column. 55 Fig . 30. Gel electrophoresis of gelsolin and G A 2 complex. 57 Fig. 31. Gel electrophoresis of actin, gelsolin and G A complex. 58 Fig. 32. Gel electrophoresis of G A complex. 59 Fig. 33. Gel electrophoresis of GA3 complex. 60 Fig. 34. Gel electrophoresis of gelsolin actin mixtures at different molar ratios, and G A 2 , and GA3 complexes. 61 Fig. 35. Gel electrophoresis of G A 2 and GA3 crystals. 62 Fig. 36. Gel electrophoresis of gelsolin-actin mixtures at different molar ratios, and of samples prepared from G A 2 and GA3 crystals. 63 ix Fig. 37. Gel electrophoresis products eluted from gel-fdtration during a gelsolin preparation. ( 64 Fig. 38. Gel electrophoresis a complex prepared from actin and cleaved gelsolin. 65 Fig. 39. The mechanism of M A L D I . 67 Fig. 40. M A L D I - M S spectrum of G A 2 (10 mg/ml) with the acquisition mass range of 33-200 kDa. 68 Fig. 41. M A L D I - M S spectrum of EDC-crosslinked G A 2 (10 mg/ml) with the acquisition mass range of 10-200 kDa. 69 Fig. 42. M A L D I - M S spectrum of GA3 (10 mg/ml) with the acquisition mass range of 10-250 kDa. 70 Fig . 43. M A L D I - M S spectrum of EDC-crosslinked G A 3 (10 mg/ml) with the acquisition mass range of 35-230 kDa. 71 Fig. 44. M A L D I - M S spectrum of EDC-crosslinked GA3 (10 mg/ml) with the acquisition mass range 45-230 kDa. 71 Fig. 45. M A L D I - M S spectrum of EDC-crosslinked G A 3 (10 mg/ml) with the acquisition mass range of 90-230 kDa. 72 Fig . 46. G A crystals (-0.1 mm) grown at 4 °C from a gel fdtered G A sample at 10 mg/ml mixed at a 1:1 ratio (v/v) with a reservoir solution of 100 m M sodium acetate buffer (pH 4.50), 2% PEG8000 (w/v), 5 m M E G T A . 73 x Fig. 47. A G A 2 crystal (-0.1 mm) grown at 4 °C from a gel filtered G A 2 sample at 10 mg/ml, mixed at a 1:1 ratio (v/v) with a reservoir solution of 100 m M sodium acetate buffer (pH 4.44), 1.5% PEG8000 (w/v), 2 m M C a C l 2 . 73 Fig. 48. G A 2 crystals (-0.1 mm) grown at 4 °C from a gel filtered G A 2 sample at 10 mg/ml mixed at a 1:1 ratio (v/v) with a reservoir solution of 100 m M sodium acetate buffer (pH 4.50), 2% P E G 8000 (w/v), 2 m M C a C l 2 . 74 Fig . 49. G A 2 (-0.1 mm) crystals grown at 4 °C from a gel filtered G A 2 sample at 10 mg/ml mixed at a 1:1 ratio (v/v) with a reservoir solution of 100 m M sodium acetate buffer (pH 4.50), 1.5% PEG8000 (w/v) + 10% PEG400, (v/v), 2 m M C a C l 2 . 74 Fig. 50. G A 3 crystals (-0.1 mm) grown at 4 °C from a gel filtered GA3 sample at 10 mg/ml mixed at a 1:1 ratio (v/v) with a reservoir solution of 100 m M sodium acetate buffer (pH 4.60), 2% PEG8000 (w/v), 2 m M C a C l 2 . 75 Fig. 51. A n EDC-crosslinked G A 3 crystal (-0.1 mm) grown at 4 °C from a gel filtered EDC-crosslinked GA3 sample at 10 mg/ml mixed at a 1:1 ratio (v/v) with a reservoir solution of 100 m M sodium acetate buffer (pH 4.68), 2% PEG8000 (w/v), 2 m M C a C l 2 . 75 Fig. 52. Terbium ion substitution in G1-G3 domains of complexes. 78 Fig. 53. E D T A titration of T b 3 + . 80 Fig. 54. The T b 3 + concentration dependence of the luminescence intensity obtained by indirect excitation of T b 3 + . (preparation 1) 81 Fig . 55. The T b 3 + concentration dependence of the luminescence intensity obtained by indirect excitation of T b 3 + . (preparation 2) 82 x i List of Abbreviations A B P s actin binding proteins A D F actin depolymerizing facor A D P Adenosine diphophate A D F Act in Depolymerization Factor A T P Adenosine triphosphate A U F S Absorbance units full scale Buffer A 2 m M Tr i s -HCl , 0.2 m M C a C l 2 0.2 m M A T P , 1 m M D T T , p H 7.6-7.8 Buffer B 25 m M T r i s - H C l , 1 m M E D T A , 0.1 m M N a N 3 , 1 M N a C l , p H 7.8 Buffer C 25 m M Tr i s -HCl , 1 m M E D T A , 0.1 m M N a N 3 , p H 7.8 D E A E 2-diethylamino-ethyl functional group D M S O Dimethyl sulfoxide DNase I Deoxyribonuclease I D T T Dithiothreitol E D C 1 -Ethyl-3-[3-dimethylaminopropyl]carbodiimide Hydrochloride E D T A Ethylene diamine tetraccetic acid E G T A Ethyleneglycol-bis-(P-aminoethyl)-N,N,N',N' -tetraacetic acid F-actin Filamentous actin F A F Familial amyloidosis (Finnish type) G A Complex of one gelsolin with one actin molecule G A 2 Complex of one gelsolin with two actin molecules G A 3 Complex of one gelsolin with three actin molecules G-actin Act in monomer G1+ Gelsolin residues 25-160 G1-G6 Gelsolin domains 1 through 6 G1-G3 Gelsolin domains 1 through 3 G4-G6 Gelsolin domains 4 through 6 H I Long helix present in each gelsolin domain H P L C High performance liquid chromatography kDa kiloDalton L P A Lysophosphatidic acid M O P S 3-(N-Morpholino)propanesulfonic acid N B D 7-chloro-4-nitrobenzo-2-oxa-l ,3-diazole N M R Nuclear magnetic resonance P D B Protein data bank P E G Polyethyleneglycol PIP 2 Phosphatidylinositol 4,5 bisphosphate P M S F Phenyl methyl sulfonyl fluoride PPIs Polyphosphoinositides S D S - P A G E Sodium Dodecyl Sulphate-Polyacrylamide Gel Electrophoresis S P A Sinapinic acid Tr i s -HCl Tris (hydroymethyl) aminomethane hydrochloride T M R Tetramethyl-rhodamine-5-maleimide v/v volume/volume w/v weight/volume Acknowledgements I would like to thank Dr. Les Burtnick for his guidance and supervision throughout the thesis work and for his comments and advice during the preparation of this thesis. I also wish to thank Dunja Urosev for helping me in the laboratory and answering all my questions. I am also very grateful to Dr. Robert Robinson for the crystallographic data analysis. I appreciated very much the support I received from my family, friends and colleagues. xiv Chapter One ~ Introduction 1.1 Actin a. Introduction to Actin Actin is a highly conserved protein in all eukaryotic cells with a molecular mass of 42 kDa. It is located in the cytoplasm mostly, with some in the nucleus. A monomeric form (G-actin) and a fdamentous form (F-actin) of actin both exist. F-actin is formed by polymerization of G-actin. In eukaryotic cells, actin fdaments constitute one of the three molecular building blocks (actin fdaments, intermediate fdaments and microtubules) of the cytoskeleton. In higher eukaryotes, actin exists as different isoforms: a-, p \ y-isoforms, etc [reviewed by (Dos Remedios et al., 2003)]. These isoforms are functionally diverse and are expressed in different tissues and cells. In muscle cells, actin (a-skeletal muscle isoform) comprises 20-25% of the total protein. In non-muscle cells, actin can control cell shape and polarity, organelle transport, solute transport and various other cellular processes (Sheterline et al., 1998). B y dynamic transition between both forms, F-actin can produce cell movement by assembling G-actins at one end while disassembling them at the other, even in the absence of motor proteins. b. Structure of G-actin The actin structure derived from DNase I/actin cocrystals (Fig. 1A) was regarded as the actual G-actin structure before the uncomplexed G-actin structure (Fig. IB) was obtained. In spite of some structural differences, both structures share several common features. G-actin is composed of a single polypeptide chain that is 375 amino acids long. It has dimensions of 55 A x 55 A x 35 A. Each actin monomer comprises four 1 subdomains (numbered 1,2, 3, 4). The N-terminus and C-terminus of G-actin are both located in subdomain 1. These subdomains come together to surround a bound nucleotide, either A T P or A D P . The A T P form can be converted into the A D P form by hydrolysis. The conformation of G-actin depends on whether there is A T P or A D P in the nucleotide-binding site. Along with nucleotide, a divalent metal ion, Ca or M g , binds within the cleft in a high affinity metal ion-binding site. Fig. 1. A) Atomic structure of actin monomer derived from actin/DNase I cocrystals at 2.8 A resolution (Kabsch et al., 1990); [from (Dos Remedios et al., 2003)] B) Atomic structure of an actin monomer derived from uncomplexed actin crystals at 1.54 A resolution (Otterbein et al., 2001). [from (Dos Remedios et al., 2003)] Because G-actin assembles under crystallization conditions (high salt concentrations) to form F-actin [reviewed by (Dos Remedios et al., 2003)], it is very difficult to crystallize uncomplexed G-actin. In order to prevent polymerization, G-actin may be cocrystallized with Act in Binding Proteins (ABPs) and analyzed by X-ray diffraction. Then the structure of G-actin can be deduced from the structure of the complex. So far, a handful of structures of actin-ABP complexes have been determined: actin-DNase I (Kabsch et al., 1990), actin-profilin (Chik et al., 1996), actin-gelsolin A B 2 (Burtnick et al., 2004; Choe et al., 2002; McLaughl in et al., 1993; Robinson et al., 1999), actin-vitamin D binding protein (DBP) (Otterbein et al., 2002), actin-fhymosin-pM (Irobi et al., 2004), actin-ciboulot (Hertzog et al., 2004). In these structures of G-actin complexes, two categories of actin conformation occur with regard to the nucleotide cleft (Fig. 2): a closed state and an open state, resulting from a relative rotation of subdomains 2 and 4 [reviewed by (Aguda et al., 2005)]. The nucleotide in open state actin has greater solvent accessibility than in the closed state. Only one structure, actin-profilin (Chile et al., 1996), exhibits an open state. A l l others to date are in a closed state. Interestingly, there are some minor differences among the closed state actin structures, which could be attributed to the different isoforms of actin used for crystallization (McLaughlin et al., 1993). Closed Open Fig. 2. A) Closed state of actin monomer (PDB code 2BTF); B) Open state of actin monomer (PDB code 1HLU). [from (Aguda et al., 2005)] In 2001, the crystal structure of G-actin, modified near its COOH-terminus at Cys374 with tetramethylrhodamine-5-maleimide ( T M R ) to inhibit polymerization, and 3 having A D P in its nucleotide cleft, was determined at 1.54 A resolution in the absence of any A B P (Otterbein et al., 2001). In this crystal form, actin also possesses a closed state conformation. When compared to other closed state conformations determined from ABP-act in complexes, the DNase I binding loop within subdomain 2 (His40-Gly48, shown as Fig 3A) is folded as an a-helix, while it is a 3-strand or disordered in structures of complexes. The structural differences may be due to a global nucleotide-dependent conformational change (Otterbein et al., 2001). The DNase I-binding loop probably binds in the hydrophobic cleft (Fig. 3B) of a neighboring monomer in F-actin (Dominguez, 2004). A B Fig 3 A) The atomic structure of uncomplexed actin in the ADP state. The four subdomains of actin are represented as- subdomains 1 (purple), 2 (green), 3 (yellow), and 4 (red). TMR is covalently attached to Cys374 and binds in a hydrophobic pocket near the C-terminus in actin subdomain 1 (Otterbein et al., 2001); B) A hydrophobic pocket (yellow) at the front end of the hydrophobic cleft (green) in actin is accessible to both G-actin- and F-actin-binding proteins (Dominguez, 2004). c. Models of F-actin A n F-actin filament can be described in two ways: either as a single-start tightly wound left-handed helix, or as a two-start long-pitch right-handed helical structure made 4 up of two intertwined strands (Fig. 4). In the single-start description, there are 13 monomers per 6 left-handed turns. Every monomer is related to the preceding one by a translation of 27.5 A and a rotation of 166 degrees around the helical axis. A s for the two-start long-pitch helix, there are 13 monomers which cross each other every half pitch. In each strand, the axial monomer translation is about 55 A. The two helical strands are axially staggered by half the axial monomer translation [reviewed by (Dos Remedios and Moens, 1995; Steinmetz et al., 1997)]. It is reasonable to describe F-actin as a two-start helix because the intersubunit contracts along the two long-pitch helical strands are continuous and much stronger than those between them [reviewed by (Steinmetz et al., 1997)]. Since each actin unit in F-actin is oriented with its cleft toward the same end of fdament, F-actin has polarity with discernibly different ends. One end is called the barbed (or "plus") end, and the other is called the pointed (or "minus") end. The names "barbed" and "pointed" refer to the chevron-like appearance in electron micrographs of an actin fdament when it is decorated by myosin SI fragments. F-actin, under certain conditions, undergoes a process called treadmilling. That is, it adds G-actin to its barbed end and, at the same time, subtracts G-actin from its pointed end. The crystal structure of F-actin is not available because of the inherent disorder in linear aggregations of fdaments [reviewed by (Dos Remedios et al., 2003). So other methods, such as electron microscopy (Egelman, 1985), fiber diffraction (Popp et al., 1987) and cross-linking methods have been used to investigate the structure of F-actin. Based on these results and actin cocrystals structures (Kabsch et al., 1990; Schutt et al., 1993), several models have been proposed. 5 The Holmes model (Holmes et al., 1990) was constructed by fitting the atomic structure of the actin monomer from the actin-DNase I complex (Kabsch et al., 1990) into a low resolution F-actin model determined using fiber diffraction data from ordered arrays of filaments (Popp et al., 1987). The monomers in F-actin were allowed to rotate in all possible orientations. Finally the fiber diffraction data was used to constrain the orientation so that the best-fit orientation could be found. Fig. 4. A) Two descriptions of F-actin: a single helix ('genetic' helix) which is a single-start, left-handed helix passing through the monomers 1 through 6, or a two-start, right-handed helix. The 'back' strand on each helix is shaded.[from (Dos Remedios and Moens, 1995)]. B) Holmes model of F-Actin and one actin unit from the Holmes model of F-Actin (http://wvvw.ncbi.nlm.nih.gov/books/bv.fcgi7calNbv.View.. ShowTOC&rid=stryer.TOC). 6 According to the Holmes model (Fig. 4B and Fig . 5), subdomains 1 and 2 are at larger radius than subdomains 3 and 4, relative to the filament axis. So, in each protomer, subdomains 3 and 4 are located close to the filament axis, while subdomains 1 and 2 are located at the outer edge of the filament. Each actin subunit interacts with four of its neighbours and the contacts involve all four subdomains. The interactions along the filament which stabilize the helical strands are as follows: subdomain 1 with subdomain 4, subdomain 3 with subdomain 4, subdomains 3/1 with subdomain 2 (Holmes et al., 1990). Fig. 5. The Holmes model (Holmes et al., 1990) of F-actin. [from (Dos Remedios et al., 2003)] 7 On the other hand, a different model was proposed by Schutt and his colleagues (Schutt et al., 1993) based on a ribbon-to-helix hypothesis that was derived from crystallographic properties of actin-profdin crystals. The intermolecular contacts in the ribbon structure from actin-profilin crystals are similar to the contacts found in the oligomeric protein assemblies. In this model, the orientations of protomers are changed and the contacts along the single-start left-handed helix are strengthened. That is to say, subdomains 1 and 2 are located closer to the filament axis, with subdomains 3 and 4 placed at the higher radius. d. Transition between G-actin and F-actin Polymerization from G-actin consists of three steps (Dos Remedios et al., 2003): 1) The first two or three monomers aggregate to form a filament nucleus. 2) Free G-actins rapidly assemble onto the filament and the reaction proceeds through the elongation phase (Fig. 6). Fig. 6. The assembly of an actin monomer onto an actin filament (http://www.ncbi.nlm.nih.gov/books/bv.fcgi7calNbv. View. .ShowTOC&rid=strver.TOQ 3) The steady-state phase occurs once the concentration of G-actin reaches a level (the so-called critical concentration) that would be in equilibrium with a population of F-actin. In this phase, G-actins associate with the barbed end and dissociate from the pointed end at the same rate. Note that ATP-act in adds to the barbed end of F-actin, following which A T P hydrolyses to A D P + P i . Subsequent release of P i means that the protomers at the 8 pointed end are in an ADP-bound state. In general, most of the protomers in F-actin contain A D P . The whole process can be expressed by the following scheme [reviewed by (Dos Remedios et al., 2003)]: ATP-G-actin -> ATP-F-actin -» ADP-Pi-F-actin -> ADP-F-actin + Pi Depolymerization is not the opposite of polymerization because the hydrolysis of A T P is not reversible in this scheme. Hence, after the dissociation of G-actin from F-actin, the exchange of A T P for A D P on the actin monomer in solution must occur before the G -actin can again jo in the left side of the scheme. The factors that regulate the transition between G-actin and F-actin include cellular ion concentration (ionic strength), p H and a number of actin-binding proteins (ABPs) . In vitro, G-actin w i l l polymerize to form well-structured and polar actin fdaments under ionic conditions believed to exist in cells (50-100 m M KC1, p H 7.0-7.1, 1 m M M g 2 + ) [reviewed by (Sheterline et al., 1998)]. However, in the cytoplasm of typical nonmuscle cells, 20-50% of actin is still in the form of G-actin. This may be attributed to intervention by A B P s contained in the cytoplasm [reviewed by (Steinmetz et al., 1997)]. These A B P s perform the following functions [reviewed by (Dos Remedios et al., 2003)]: ADF-cofd in family members can promote the disassembly of G-actin and inhibit nucleotide exchange by binding to ADP-act in with high affinity; profilin can accelerate the exchange of A D P for A T P in G-actin released from F-actin; gelsolin can sever F-actin and subsequently cap the barbed end of F-actin; thymosin p4 can inhibit polymerization of F-actin by tightly binding ATP-G-act in; DNase I can depolymerize F-actin by sequestering G-actin; CapZ can slow down the turnover of F-actin by binding to the 9 barbed end of F-actin; Arp2/3 can create branch points by binding to the side of F-actin. These A B P s together regulate the cellular processes involving actin (Fig. 7). Extracellular stimuli 5. Growing filaments push membrane forward 2. Signals activate WASp/Scar proteins Q mi l 11 llll 11 • • • • • • • • • • • • • I l l i l l l l l l l l l l l l i l l l l l l l l l l l l l - l l l r T T T T T T | J j C E i D 1 3. WASp/Scars activate o ^ n * } Q Arp2/3 complex to Q, ' rv O initiate new filaments ^ "O^ Q^ Q # 6. Capping protein limits elongation I PAK 9. UM-kinase inhibits ADF/cofilin .0,9 1.P00I of ATP-acttn bound to profilin • Q» o 10. Profilin catalyzes exchange of AOP for ATP Fig. 7. The regulation of actin filaments by ABPs in the cell, [from (Pollard et al., 2000)] 1.2 Gelsolin a. Introduction to gelsolin The gelsolin superfamily of proteins consists of at least seven members: gelsolin, vi l in , adservin, capG, advillin, supervillin and flightless [reviewed by (Silacci et al., 2004)]. Gelsolin is the founding member of this family, with a molar mass of 82-84 kDa. In vivo, gelsolin exists in two different isoforms: cytoplasmic gelsolin (intracellular) and plasma gelsolin (extracellular). Plasma gelsolin is a 755-amino acid protein, while cytoplasmic gelsolin is 25 residues shorter at its N-terminus. In addition, plasma gelsolin 10 possess a disulfide bond between Cys l88 and Cys201 [reviewed by (Silacci et al., 2004)], which stabilizes the second domain of gelsolin in its activated conformation (Burtnick et al., 1997; Zapun et al., 2000). Cytoplasmic gelsolin regulates the mobility and architecture of cells. Its activities include severing and capping actin filaments, and nucleating the formation of new filaments. These activities are regulated by C a 2 + and P I P 2 . On activation by micromolar Ca , gelsolin changes its conformation to expose previously buried actin-binding sites. Then gelsolin binds (rapidly) and severs (slowly) the actin filament. After severing, gelsolin caps the newly created barbed end of F-actin and prevents its reannealing with short actin filaments. A t the same time, disassembly at the pointed end proceeds unchecked. These actin filaments may be uncapped by P I P 2 - r i c h membrane to generate polymerization-competent ends and seed the assembly of new filaments [reviewed by (McGough etal.,2003)]. Plasma gelsolin, together with vitamin D-binding protein (DBP) contribute to removal actin filaments from the circulatory system. Act in is released into blood plasma upon cell death and tends to polymerize to filaments that can elevate the viscosity of plasma in the absence of opposing factors (e.g., scavenger system proteins). Plasma gelsolin can sever and cap actin filaments as discussed above. D B P can then sequester actin monomers, released from pointed ends, into a actin-DBP complexes that are removed from the circulation in the liver [reviewed by (Dos Remedios et al., 2003)]. A neurodegenerative disease, familial amyloidosis of Finnish type (FAF) , is caused by hereditary mutation in gelsolin domain 2 of Asp 187 to A s n or Tyr. These mutations expose the otherwise masked proteolytically sites and render gelsolin domain 2 11 unable to bind C a 2 + and susceptible to aberrant proteolysis by furin and, subsequently, (3-gelsolinase (Huff et al., 2003). Furin is a kind of Ca -dependent protease and can cleave after an -R-x-x-R- consensus sequence. The resulting gelsolin fragments (residues 173-243 and residues 173-225) assemble into amyloid fibrils that manifest themselves in various neuropathies and ophthalmic disorders. D187N/Y i | z _ J 1 73 ..furin 755 i 1 73 243 244 ..f^gelsolinase(s) 755 1 t 173- 225 226 755 U...L 1 I Amyloid Excretion/Degradation Fig. 8. Schematic of the aberrant proteolysis of D187N/Y gelsolin. The FAF variants are cleaved by furin in trans-Golgi network and P-gelsolinases after secretion from the cell, [from (Huff et al., 2003)] b. Structure of inactive gelsolin Gelsolin contains six analogous domains (G1-G6), each of which consists of 120-130 amino acid residues. Domains 1-3 constitute the N-terminal half and domains 4-6 the C-terminal half of gelsolin. From amino acid sequence analysis, G l and G4, G2 and G5, and G3 and G6 are most closely related when compared pair-wise. The inactive gelsolin 94-structure in the absence of C a was solved by crystallography (Burtnick et al., 1997). The six domains in inactive gelsolin pack into a compact globule with overall dimensions 85 A x 55 A x 36 A, as shown in Fig. 9. Each of G l through G6 share a similar folding topology, consisting of a five or six stranded |3-sheet at its core sandwiched between a 12 3.5-4.5 turn a-helix (HI) and a 1-2 turn a-helix (H2) (Fig. 10). H I and H2 respectively, run parallel to and perpendicular to the strands in the sheet. Fig. 9. The structure of inactive plasma gelsolin in the absence of Ca 2 + . The six domains of are colored as: Gl (red); G2 (light green); G3 (yellow); G4 (pink); G5 (dark green); G6 (orange) (Burtnick et al., 1997). 13 Fig. 10. Individual domains of inactive gelsolin excised from the inactive structure of plasma gelsolin (Burtnick et al., 1997) Proteolytic fragments and sequence analysis identified actin binding regions in gelsolin [reviewed by (McGough et al., 2003)]: G l contains a calcium-independent G -actin binding site; G4 contains a calcium-dependent G-actin binding site; G2 contains calcium-independent F-actin binding sites. Because most residues that are involved in the interaction of G l with actin (McLaughlin et al., 1993) are conserved in G4 (Burtnick et 14 al., 1997), G l and G4 bind actin in a similar manner. The F-actin binding sites (residues 161-172 and 197-226) in G2 involve strand A , the A B loop and the long helix (Fig. 10) which constitute a single, solvent-exposed entity. The PPI-binding sites (residues: 132-149 and 161-172) overlap the actin binding sites so that PPI dislodges and uncaps actin. Inactive gelsolin does not bind actin because all the binding sites for G-actin and F-actin are masked within the globular arrangement of the domains. Fig. 11. The N-terminal half and C-terminal half of inactive gelsolin excised from the structure of inactive gelsolin (Burtnick etal., 1997) The triplet of G1-G3 (N-terminal half) shows broadly similar pseudoquaternary structure with that of G4-G6 (C-terminal half), as shown in Fig . 11. These two haves are connected by a large 53-residues linker, which is necessary for the F-actin binding and severing functions of gelsolin [reviewed by (McGough et al., 2003)]. The relationship of 15 G l to G3 in inactive gelsolin is similar to that of G4 to G6, while the organization of G2 relative to G l and G3 is somewhat different from that of G5 relative to G4 and G6. Also, gelsolin possesses a C-terminal extension (a latch helix), which plays an important role in calcium regulation of actin-binding activity. In Ca -free gelsolin, the latch helix interacts with the long helix of G2 so as to block the actin-binding sites in G2 (Choe et al., 2002). c. From inactive gelsolin to activated gelsolin Although the structure of fully activated, intact gelsolin is still not available, structures of activated fragments of gelsolin have been investigated in the form of complexes with G-actin and in isolation. These structures include: Gl /ac t in complex (McLaughlin et al., 1993), G4-G6/actin (Choe et al., 2002; Robinson et al., 1999), activated G4-G6 (Kolappan et al., 2003; Narayan et al., 2003), Gl-plus/actin (Irobi et al., 2003), Gl-G3/ac t in (Burtnick et al., 2004), Gl-fhymosinp4/actin (Irobi et al., 2004). Comparing these activated gelsolin fragments with inactive gelsolin, a mechanism for calcium activation can be proposed (Burtnick et al., 2004; Choe et al., 2002; Khaitlina et a l , 2004). The activation of gelsolin from an inactive to an active conformation is regulated by C a 2 + . Ca 2 + bind ing causes rearrangement of domains. During the activation process, at least three identifiable latches are opened: a tail latch, the G1-G3 latch and the G4-G6 latch. The tail latch is the C-terminal extension, which terminates in a short a-helix. B y interacting with G2, the tail latch blocks F-actin binding sites on G2.The G1-G3 latch and G4-G6 latch are formed by the shared P-sheets between G l and G3, and between G4 and G6, respectively. The G1-G3 latch is stabilized by interactions between G6 and G1/G3, 16 G l and G3-G4 linker. The G4-G6 latch is stabilized by interaction between the G3-G4 linker and G6, C-terminal tail and G4 (Choe et al., 2002). Fig. 12. Locations of type I calcium (golden spheres) and type II calcium (gray spheres) binding sites in inactive gelsolin (Choe et al., 2002). Gelsolin possesses two types of calcium binding sites: type I and type II (Robinson et al., 1999). Type I sites involve coordination of C a between actin and a gelsolin domain. G l and G4 possess type I sites (Burtnick et al., 2004; McLaughl in et al., 1993; Robinson et al., 1999). Type I C a 2 + ions moderate the affinity between gelsolin and actin by participating at the binding surface. Type I C a 2 + binding also contributes to the breaking of the G1-G3 latch and G4-G6 latch and revealing the actin-binding sites on G l and G4 (Choe et al., 2002). Type II C a 2 + sites are wholly contained within gelsolin domains. Each of the six gelsolin domains possesses one type II site. A l l type II sites in G l (Burtnick et al., 2004; McLaughl in et al., 1993), G2 (Kazmirski et al., 2002), G3 17 (Burtnick et al., 2004), G4 (Robinson et al., 1999) and G5 (Choe et al., 2002), G6 (Choe et al., 2002) have been characterized. Type II calcium binding disrupts the conformation of the inactive form gelsolin and stabilizes activated gelsolin (Choe et al., 2002). Two high-affinity sites located in G4-G6, with Kd values o f 0.2 and 2 u M , have been identified by equilibrium dialysis experiments (Lin et al., 2000; Pope et al., 1995). A third, low affinity, calcium binding site in G4-G6 with a Kd value of 100 u M has also been identified from synchrotron footprinting experiments (Kiselar et al., 2003). Based on these results, a multilevel scheme (Fig. 13) for activation of gelsolin through occupation, at least transiently, of all six type II Ca 2 + -binding sites was proposed (Burtnick et al., 2004). Level A (0.2 u M Ca 2 + ) : calcium binding at G6 destabilizes the G2-G6 tail latch, which dynamically moves between attached and detached states. But gelsolin still keeps its global structural integrity. Level B (2 u M C a 2 + , a typical "resting" 2+ cytoplasmic calcium level): Ca binding to G4 weakens the G 4 - G 6 latch. The tail latch is fully released. Level C (100 u M Ca 2 + ) : C a 2 + binding to G5 stabilizes the G 4 - G 5 interface and opens G 4 - G 6 latch fully. Also , C a binding to G3 straightens its long helix and opens the G 1 - G 3 latch, revealing the Ca binding site on G2 and driving G 2 - G 3 toward an active conformation. Level D (in excess of 0.6 m M C a , typical of plasma calcium levels): Ca 2 + b ind ing to the G l site cocks G l in a position to strike and sever F-actin. In cytoplasmic submicromolar calcium conditions, gelsolin is partly activated and severs actin filaments slowly, relying on a non-equilibrium calcium-binding mechanism; in blood plasma (milllimolar calcium), gelsolin is fully activated and severs F-actin much more efficiently (Choe et al., 2002). 18 A X FURIN 1 mM 100 uM TRANSIENT STATE Fig. 13. A) Activation of gelsolin in presence of calcium;[from (Choe et al., 2002)] B) Different levels of calcium activation, [from (Burtnick et al., 2004)] G4-G6 calcium-activated switch. During transformation to the active form (Robinson et al., 1999), there are obvious changes in the G4-G6 conformation, as shown in Fig. 14B. The P-sheet that runs from G4 to G6 is separated into two and the kinked long helix in G6 is straightened. G6 19 rotates 90 degree around axes in both the horizontal and vertical planes and translates 40 A in the vertical direction away from G4. Thus, new G6 contacts with G5 are established. G1-G3 calcium-activated switch. On going from inactive gelsolin to active Gl-G3/ac t in (Burtnick et al., 2004), as shown in F i g . l 4 A , the P-sheet between G l and G3 in inactive gelsolin is severed. G l translates away from G2 to extend the reach of the G1-G2 linker to 30 A. The G2-G3 linker shortens through adoption of a helical conformation. Comparison of structural changes in G1-G3 and G4-G6 reveals similarities and differences between the two activation processes. Both processes involve severing of the continuous P-sheet and straightening of the kinked helices in G3 and G6, respectively. New contacts between G2 and G3, and G5 and G6 are formed. The differences lie in the relative orientations of G2 and G l vs. that of G5 and G4. G2 is in a position to bind actin, whereas G5 is not even in the proximity of actin. 20 B 0 Fig. 14. A) Transition from G1-G3 in inactive gelsolin to its active conformation (Burtnick et al., 2004); B) Transition from G4-G6 in inactive gelsolin to its active conformation (Choe et al., 2002). Golden spheres and black spheres represent Ca 2 + bound at type I sites and type II sites, respectively. 1.3 Gelsolin and Actin interaction a. Gelsol in/Act in complexes The G A complex is composed of one gelsolin and one actin. It was first extracted by EGTA-treatment o f human platelets (Kurth et al., 1983). The G A complex can affect F-actin assembly in presence of either C a 2 + or E G T A . In other word, G A tends to bind to 21 the barbed end of F-actin in either C a 2 + or E G T A . But the G A complex binds fourfold faster to the barbed end in the presence of C a 2 + than in excess E G T A (Selve and Wegner, 1986). G A has a high affinity for the barbed end even in E G T A , differing in this regard from activated gelsolin (Janmey et al., 1985). So, once gelsolin binds the barbed end of F-actin, it can not be dissociated by the removal of calcium. But the G A complex can not efficiently sever F-actin (Janmey et al., 1985). In addition, the interaction between gelsolin and actin in the G A complex is strongly ligand-mediated and a Ca is reported to be trapped within G A (Bryan and Kurth, 1984). In the G A complex, actin binding is not reversible by treatment with E G T A , but G A can be separated by P I P 2 (Yin et al., 1988). After a second actin binds to gelsolin in the presence o f calcium, the G A complex is converted to a ternary complex, G A 2 (Bryan and Kurth, 1984). The rate constant of o 1 1 formation o f the G A 2 complex from G A and actin is 1 x 10 M " s" , which is 5000-fold higher than the rate constant of binding of actin to activated gelsolin (Ditsch and Wegner, 1995). The second binding is much weaker than the first, and is reversed by decreasing the C a 2 + concentration or adding excess E G T A (Janmey et al., 1985). Although the C a 2 + -sensitive sites that regulate the first actin binding have a greater affinity (Kd = 25 pM) for calcium than those that regulate the second actin binding ( K a = 200 p M ) (Ditsch and Wegner, 1995), the difference is overridden by the much higher rate constant of binding actin to G A compared to the rate constant of binding actin to gelsolin. Furthermore, the two C a 2 + are bound to gelsolin in a highly co-operative manner (Gremm and Wegner, 1999). Thus, G A 2 is formed exclusively when mixing gelsolin and actin in the presence of C a 2 + . The G A 2 complex can also bind to the barbed end of F-actin, but can not sever F-22 actin (Coue and Korn , 1985). Surcose-density gradient and chemical crosslinking studies also show G A 2 to be the actin-gelsolin complex formed preferentially in the presence of micromolar Ca (Yin , 1987). The GA3 complex was found by native gel electrophoresis techniques (Edgar, 1990) on mixing a molar ratio of actin to gelsolin greater than 2:1, in the presence of C a 2 + . After S D S - P A G E in a second dimension, the stoichiometry observed between actin and gelsolin was 3:1, determined by densitometry for this 247 kDa complex. The G A 3 complex increased in proportion as the molar ratio o f actin to gelsolin in the sample increased, and no higher order complexes were found. G A 3 could also dissociate into a G A complex and two actins after C a ions were chelated by E G T A . Two explanations for the formation of G A 3 were proposed (Edgar, 1990): on one hand, the third actin could bind the unoccupied actin-binding site of gelsolin; on the other hand, the third actin also could bind to the other two actins in a G A 2 complex (Edgar, 1990). b. Model of F-actin severing and capping by gelsolin Based on the crystal structures of the C-terminal half of gelsolin bound to one actin (Choe et al., 2002; Robinson et al., 1999) (Fig. 15) and the N-terminal half of gelsolin bound to one actin (Burtnick et al., 2004) (Fig. 15), a model for a gelsolin-capped filament was proposed (Burtnick et al., 2004) (Fig. 16) by superimposition of the G4-G6/actin structure and Gl-G3/ac t in structure onto two protomers within the Holmes F-actin model. This model explains how activated intact gelsolin binds to the side of F-actin, severs F-actin, and then caps the newly generated barbed filament end. 23 Fig. 15. A) Structure of G4-G6 bound to actin in the presence of calcium (Choe et al., 2002). B) Structure of G1-G3 bound to actin in the presence of calcium (Burtnick et al., 2004). [from (Burtnick et al„ 2004)] 24 A model ADP-act in fdament can be constructed by overlaying the uncomplexed G-actin structure (Otterbein et al., 2001) onto the actin protomers in the Holmes F-actin model (Holmes et al., 1990). The Gl-G3/act in and G4-G6/actin structures are laid onto the terminal actins. The model requires that actin subdomains 1 and 2 should be on the outside of F-actin, which agrees with the Holmes model. Fig. 16. Two possible models for the capped barbed end of F-actin. The actin units are represented according to Holmes model. A) Four protomers of F-actin (drawn in blue and gray) with Gl-G3:actin (Gl, red; G2, green; G3, yellow) and G4-G6:actin (G4, pink; G5, dark green; G6, orange) [from (Burtnick et al., 2004)]; B) Five protomers of F-actin (drawn in blue, dark gray and cyan) with Gl-G3:actin (Gl, red; G2, green; G3, yellow) and G4-G6:actin (G4, pink; G5, dark green; G6, orange). Golden spheres and black spheres represent Ca 2 + bound at type I sites and type II sites, respectively. The two blue spheres represent the last residue in G3 and the first residue in G4, respectively. 25 In this capping model (Burtnick et al., 2004), G2-G3 is placed at the junction between two longitudinally neighboring actin protomers. The long helix of G2 lies close to the binding site on the actin occupied by the analogous helix of G l , but on a different • • • 2+ protomer. G l and G4 exert steric hindrance to G-actin joining the barbed end. If C a -free G2 were superimposed onto the activated G2 position, the C-terminal tail would obscure the actin-binding site on G2. So the C-terminal tail must be removed before G2 binds the side of F-actin. Thus, the tail latch hypothesis (Burtnick et al., 1997) is confirmed. The severing o f F-actin can be summarized in three steps: first, G2 binds to the side of the filament; second, the G1-G2 linker directs G l into its binding site, while the G3-G4 linker directs activated G4-G6 to its site at a monomer in the intact strand directly across from the site of attachment of G2; finally, F-actin is severed by the coordinated pincer movement of G l and G4 (Burtnick et al., 2004; Pope et al., 1991). directed to their G2-G3 binds binding sites Severing Fig. 17. The binding to and severing of an actin filament by gelsolin in the presence of calcium, [from (Choe et al., 2002)] 26 But a second capping model can also be proposed (Fig. 16B). The difference is that the G4-G6 half of gelsolin interacts with a different protomer in the actin fdament. Based on the length of the G3-G4 linker, both models could be acceptable. The G3-G4 linker conformation in the capped model also is unknown. Further investigations are needed to test the gelsolin-capped models o f F-actin. c. Models of gelsolin/actin complexes Although the existence of G A , G A 2 and G A 3 has been proved by previous study, the structures of these complexes are unknown. In the three actin binding sites detected in gelsolin, two are located in the N-terminal half of gelsolin and one is in the C-terminal half (Bryan, 1988). Taken together with the crystal structure of C-terminal half of gelsolin bound to one actin (Choe et al., 2002; Robinson et al., 1999), the N-terminal half of gelsolin bound to one actin (Burtnick et al., 2004) and gelsolin-capped actin filament models, we are able to propose models for G A 2 and G A 3 complexes: For G A , because the binding is calcium independent, as is the severing activity of the N-terminal half, it should be the N-terminal part o f gelsolin that is bound to one G -actin monomer with the C-terminal of gelsolin unbound. In G A 2 , considering that the second actin binding is calcium dependent and the activity of the C-terminal half is calcium dependent, the C-terminal of gelsolin is likely to be bound to the second actin monomer, with the N-terminal half bound to the first G-actin. 27 Fig. 18. Proposed GA 2 model. The two actin units are color as cyan and blue. The blue-colored actin is bound to the G4-G6 (pink, dark green and orange) half of intact activated gelsolin and the cyan-colored actin is bound to the G1-G3 (red, light green and yellow) half of intact activated gelsolin. Golden spheres and black spheres represent Ca 2 + bound at type I sites and type II sites, respectively. The two blue spheres represent the C-terminus of G1-G3 and the N-terminus of G4-G6, respectively. The G3-G4 linker (40 residues) is not represented. In G A 3 , the third actin could bind to the F-actin binding sites of G2 and also bind to the original actin attached to the N-terminal end of gelsolin at the same time. The orientations of gelsolin and actin in GA3 could resemble the orientations in the gelsolin-capped actin filament model (Fig. 19). Also, the relative position of the three actins in GA3 could provide direct evidence in favor of or against controversial F-actin models. Furthermore, the G A 2 and G A 3 structures wi l l give the G3-G4 linker conformation and 28 provide a complete picture of the end result of severing and capping of an actin fdament by gelsolin. Fig. 19. Proposed GA 3 model. The three actin units are color as cyan, blue and dark gray. The blue-colored actin is bound to the G4-G6 (pink, dark green and orange) half of intact activated gelsolin and the cyan-colored actin is bound to the G1-G3 (red, light green and yellow) half of intact activated gelsolin. The dark gray-color actin is close to F-acin binding sites of G2 and is positioned as the Holmes model. Golden spheres and black spheres represent Ca 2 + bound at type 1 sites and type II sites, respectively. The G3-G4 linker (40 residues) is not represented. In this work, we set about to crystallize complexes in which one, two and three actin units bind one activated intact gelsolin, i.e. G A , G A 2 and GA3, respectively. The structures of G A , G A 2 and GA3 would provide experimental evidence to help support or refute different models of gelsolin-capped F-actin and for F-actin itself. In the end, we 29 succeeded in crystallizing G A , G A 2 and GA3, as proved by size exclusion chromatograph and gel electrophoresis. We also used M A L D I mass spectrometry experiments to confirm our G A 2 and GA3 complexes. Unfortunately, our G A 2 and G A 3 crystals analyzed by X -ray crystallography have not yielded high resolution structures of the desired complexes. The X-ray results in both cases provide a picture of the N-terminal half of gelsolin binding to one actin, while the C-terminal half of gelsolin and a second actin are not detectable. In order to stabilize floppy ends and loops in our complexes, we attempted to use cross-linking reagents (e.g., E D C ) . We are awaiting diffraction data from E D C cross - l inked crystals and G A crystals. We also soaked G A 2 and G A 3 crystals in Tb 3 +-containing solutions in order to investigate the exchangeability of C a in these complexes. B y X-ray crystallography, we found T b 3 + in the type I site in G l and in all three type II C a 2 + sites of G1-G3, including the type II site in G2, which is not seen to be occupied by a metal ion in the Gl-G3/act in structure (Burtnick et al., 2004). Then the Tb substituted complexes were back-titrated with a high concentration C a 2 + stock solution. 30 Chapter Two « Materials and Methods 2.1 Protein purif ication a. Gelsolin purification Gelsolin was purified from horse serum (Pel-Freez Biologicals, Rogers, A R ) based on a modified method of Bryan (Bryan, 1988). 500 ml of frozen horse serum was thawed under cold running water. Protease inhibitors leupeptin (2 mg/ml in H2O) and pepstatin (2 mg/ml in D M S O ) were thawed as well . Protease inhibitor P M S F solution was newly made by dissolving 2 mg P M S F powder in 1 ml M e O H . 50 pi each of leupeptin, pepstatin and P M S F were added to the thawed horse serum so that the final concentration of the each inhibitor was 2 x 10"4 mg/ml. Then the serum was dialyzed against 4 L of 25 m M T r i s - H C l , 0.5 m M C a C l 2 , p H 7.5 at 4 °C. After three changes over three days, the dialyzed horse serum was centrifuged at 9,000 rpm (RC-5B centrifuge and Sorvall G S A rotor) for 40 mins. After decanting and adjusting to 35 m M N a C l with crystalline salt, the clarified plasma was mixed with 1.5 L anion exchange resin ( D E A E -Sephadex A-50, Pharmacia) that had been equilibrated against 25 m M Tr i s -HCl , 0.5 m M CaCi2, 50 m M N a C l , p H 7.5. The mixture was kept at 4 °C and stirred every 15 min for 2 hours. The binding of C a 2 + to inactive gelsolin changes its surface properties so that it does not bind the anion exchange resin, while serum albumin and other negatively charged proteins do. Thus, only gelsolin and positively charged proteins are left in the liquid phase. The mixture was filtered through Whatmann #3 filter paper, initially by gravity. Vacuum was not used until only wet resin was left in the funnel. E D T A was added to the filtrate to a concentration of 10 m M , the p H of the mixture was adjusted to 7.8 by drop-wise addition of 5 M N a O H . E D T A chelation of C a 2 + returns gelsolin to its 31 inactive condition in which it does bind to anion exchange resin at p H 7.8. Thus, an anion . exchange column was used to purify gelsolin further. Positively charged proteins would not bind to the column, and would elute in the void volume. Negatively charged gelsolin, which binds to the column, and some minor protein contaminants can be eluted by application of a gradient in salt concentration. Two strategies were used at this stage to purify gelsolin further. 1. HPLC-based anion exchange column strategy. If an H P L C anion exchange column were used, the EDTA-adjusted solution was first concentrated to 50 ml by ultrafiltration using a Y M - 3 0 membrane (Amicon). The concentrate was filtered through a 0.22 pm pore size Mil l ipore filter and applied to an H P L C column (150 x 20 mm, BioSep D E A E - P , Phenomenex) at room temperature with a flow rate of 1.5 ml/min. The absorbance of the eluant was monitored 280 nm. Two buffers were used in this chromatography: 25 m M Tr i s -HCl , 1 m M E D T A , 0.1 m M NaN3, p H 7.8 (buffer C) , and 25 m M Tr i s -HCl , 1 m M E D T A , 0.1 m M N a N 3 , 1 M N a C l , p H 7.8 (buffer B) . The gradients used to elute gelsolin were constructed as follows: Time (min) % Buffer B % Buffer C Flow rate (ml/min) 0 0 100 1.5 25 0 100 1.5 27 4.4 95.6 1.5 57 4.4 95.6 1.5 59 6 94 1.5 84 6 94 1.5 86 100 0 1.5 116 100 0 1.5 118 0 100 1.5 148 0 100 1.5 Table 1. The salt gradient program for HPLC used to elute the gelsolin (Max. pressure: 280 psi; Min. pressure: 50 psi). 32 With buffer C only (0-25 min), proteins that did not bind the ion exchanger were eluted. With 44 m M N a C l (27-57 min), the proteins that weakly bound the ion exchanger came out. With 60 m M N a C l (59-84 min), gelsolin eluted together with some minor contaminants. 2. Anion exchange column strategy using DEAE-Sephadex A-50. After the anion exchange column (36 x 4.5 cm) was equilibrated with 2 5 m M Tris-H C l , 50 m M N a C l , 1 m M E D T A , p H 7.8, the EDTA-adjusted protein solution was loaded at 2 ml/min. The column was washed with 1.5 L o f equilibration buffer. A gradient maker was used with 1 L 50 m M N a C l , 2 5 m M Tr i s -HCl , 1 m M E D T A , pH 7.8 on one side and 1 L 350 m M N a C l , 2 5 m M Tr i s -HCl , 1 m M E D T A , p H 7.8 on the other. Gelsolin was eluted by the salt gradient, as monitored with an E C O N O U V monitor (Bio-Rad) at 280 nm. Then, the gelsolin-containing fractions were concentrated to about 20 ml and dialyzed over two days against two changes of 1.5 L 25 m M Tr i s -HCl , 1 m M E D T A , pH 8.0 at 4 °C. In order to ensure the purity of gelsolin for crystallization uses, it was subjected to Aff i -Gel Blue affinity chromatography in 25 m M Tr i s -HCl , 1 m M E D T A , p H 8.0. Gelsolin bound to the dye Cibacron Blue F 3 G A , which was attached to an inert agarose support matrix (Yamamoto et al., 1989). After washing with 150 ml equilibration buffer, 100 ml of 2.5 m M A T P made up in the equilibration buffer was used to elute the gelsolin. A fluorescence spectrometer (Perkin Elmer, model LS-5B) was used to detect gelsolin using an excitation wavelength of 280 nm and an emission wavelength of 340 nm. Finally, the concentrated eluant (20 ml) was dialyzed against three changes of 1 L of 25 m M Tr i s -HCl , 1 m M E D T A , p H 8.0, over three days. The concentration of gelsolin was 33 measured spectrophotometrically (Perkin Elmer Lamda 4B U V - V i s spectrometer) at a wavelength of 280 nm. A n absorption coefficient was used for gelsolin of 1.4 ml mg"1 cm"1 (Silva and Burtnick, 1990). The purity of the gelsolin was characterized by polyacrylamide gel electrophoresis ( P A G E ) in the presence of sodium dodecyl sulfate (SDS) and 2-mercaptoethanol. b. Ac t in purification Act in was purified from rabbit skeletal muscle based on a modified method of Spudich and Watt (Spudich and Watt, 1971). 5 g rabbit skeletal muscle powder was extracted with 60 ml of 2 m M Tr i s -HCl , 0.2 m M C a C l 2 0.2 m M A T P , 1 m M D T T , pH 7.6-7.8 (buffer A ) for 45 mins on ice. Then the extract was filtered through two layers of cheesecloth and Whatman #1 filter paper. The residue was washed with 25 ml of buffer A and filtered again. The combined filtrates were centrifuged at 35,000 rpm (Beckman model optima L-90k ultracentrifuge and Beckman 45 T i rotor) for one hour at 4 °C. After discarding the pellet, the actin in the supernatant was polymerized overnight by the addition of KC1 and M g C l 2 to final concentrations of 50 m M and 2 m M , respectively. The next day, actin binding proteins were dissociated and solubilized by addition of KC1 to 0.8 M . After stirring 2.5 hr at 4 °C, the actin solution was centrifuged at 35,000 rpm (Beckman model optima L-90k ultracentrifuge and Beckman 70 T i rotor) for 3 hours. The pellet was suspended in 10 ml buffer A using a homogenizer, after which the pellet was dialyzed against three changes of 1 L of buffer A over three days. The resulting G -actin solution was centrifuged at 35,000 rpm (Beckman model optima L-90k ultracentrifuge and Beckman 70.1 T i rotor) for 3 hours to get rid of any remaining F-actin. Finally, the G-actin solution was applied to a gel filtration column (Bio-Rad Sephacryl 34 S300; 90 x 2.5 cm) at 4 °C and eluted with buffer A at a rate of 2 ml/min. The actin elution was monitored by an E C O N O U V monitor (Bio Rad) at 280 nm. Act in was eluted as a single peak, but only the second half of the peak was collected for subsequent use. The concentration of G-actin was measured by U V - V i s spectrophotometry (Perkin Elmer Lamda 4B) at 290 nm with an absorption coefficient of 0.63 ml mg"' cm" 1. 2.2 Formation and purification of complexes a. G A 2 complex Gelsolin in 25 m M Tr i s -HCl , 1 m M E D T A , p H 8.0 was incubated with 2 m M C a C b at 4 °C for 5 min. Then gelsolin was mixed with actin in buffer A at a molar ratio of 1:2. The resulting solution was incubated at 4 °C overnight. Then the G A 2 complex solution was run through a gel filtration column (Bio-Rad Sephacryl S300; 90 x 2.5 cm) at 4 °C and eluted with buffer A at a rate of 2 ml/min. The G A 2 elution was monitored by a U V E C O N O U V monitor (Bio-Rad) at 280 nm. The G A 2 fractions were combined and concentrated by R C - 5 B centrifugation at 3,000 rpm at 4 °C, first using Millipore centrifugal concentrators with a Sorvall G S A rotor, and then using Biosep centrifugal concentrators with a Sorvall SS34 rotor. The final concentration of G A 2 was 10 mg/ml as determined by U V - V i s spectrometry (Perkin Elmer Lamda 4B) at 280 nm using a calculated absorption coefficient of 1.25 ml mg"1 cm" 1. The purity of G A 2 was characterized by polyacrylamide gel electrophoresis ( P A G E ) in the presence of sodium dodecyl (SDS) and 2-mercaptoethanol. Two clean bands were observed at 83 kDa and 42 kDa, respectively. 35 b. G A 3 complex A s for GA3 formation, in the presence of 2 m M C a , gelsolin was mixed with actin at a molar ratio of 1:4. The mixed solution was incubated for 30 min at 4 °C. After addition of KC1 and M g C l 2 to 100 m M and 2 m M , respectively, the GA3 solution was incubated at 4 °C overnight. Then the GA3 complex solution was run through a gel filtration column (Bio-Rad Sephacryl S300; 90 x 2.5 cm) at 4 °C and eluted with 2 m M Tr i s -HCl , 0.2 m M C a C l 2 0.2 m M A T P , 1 m M D T T , 100 m M KC1, 2 m M M g C l 2 , p H 7.6-7.8 at a rate of 2 ml/min. The GA3 elution was monitored with an E C O N O U V monitor (Bio Rad) at 280 nm. The G A 3 fraction was concentrated to 10 mg/ml in the same way as the G A 2 complex. The absorption coefficient for the GA3 complex was calculated to be 1.2 ml mg"1 cm"1 at 280 nm. The purity of G A 3 was characterized by polyacrylamide gel electrophoresis ( P A G E ) in the presence of sodium dodecyl (SDS) and 2-mercaptoethanol. c. G A complex Based on a modified method of Gremm (Gremm and Wegner, 1999), gelsolin and actin were mixed at a molar ratio of 1:2 in the presence of 10 m M C a 2 + . After 5 mins, 30 m M E G T A was added to the mixture. After the G A solution was incubated at 4 °C overnight, the complex was purified by gel-filtration (Bio-Rad Sephacryl S300; 90 x 2.5 cm) at 4 °C and eluted with 10 m M M O P S / K O H , 50 u M M g C l 2 , 0.1 m M E G T A , p H 6.8 at a rate of 2 ml/min. The G A fraction was concentrated to 10 mg/ml in the same way as the G A 2 complex. The G A absorption coefficient used was calculated to be 1.3 ml mg"1 cm"1 at 280 nm. The purity of G A was characterized by polyacrylamide gel electrophoresis ( P A G E ) in the presence of sodium dodecyl (SDS) and 2-mercaptoethanol. 36 We also tried other strategies and buffer components to form and purify the G A complex. E G T A was added to G A 2 , which had been formed by adding actin to gelsolin at a molar ratio of 2:1 in the presence of 2 m M C a 2 + , to a final concentration of 5 m M . After 2 hours incubation, the extra G-actin could be polymerized to F-actin by addition of 100 m M KC1 and stirring overnight. The resulting solution was centrifuged at 35,000 rpm (Beckman model optima L-90k ultracentrifuge and Beckman 70.1 T i rotor) for 3 hours to get rid of any F-actin. The G A complex solution was further purified by gel filtration (Bio-Rad Sephacryl S300; 90 x 2.5 cm) at 4 °C and eluted with different buffers (shown in Table 2) at a rate of 2 ml/min. The G A elution was monitored with an E C O N O U V monitor (Bio Rad) at 280 nm. But multiple gel-filtration elution peaks were obtained. Collection and concentration of these peaks failed to form crystals. GA elution buffer components 25 m M Tr i s -HCl , 5 m M E G T A 25 m M Tr i s -HCl , 5 m M E G T A , 100 m M KC1 25 m M Tr i s -HCl , 5 m M E G T A , 0.2 m M A T P , 1 m M D T T 25 m M Tr i s -HCl , 5 m M E G T A , 0.5 m M A D P , 1 m M D T T Table 2. Alternative GA complex elution buffers 2.3 SDS-PAGE a. Introduction to SDS-PAGE S D S - P A G E stands for sodium dodecyl sulfate-polyacrylamide gel electrophoresis. It is a useful technique for molecular weight analysis of proteins. SDS is an anionic detergent that can denature proteins and can bind to them quantitatively at a mass ratio of approximately 1.4:1, adding its negative charges to the amino acids of proteins. The 37 resultant SDS-protein complex is elongated in shape, with an essentially constant charge to length ratio, so separation in a uniform porosity gel is determined by size. 2-mercaptoethanol reduces disulphide bridges which are not disrupted by SDS in proteins so that the observed SDS-protein complexes represent single polypeptide chains [reviewed by (Righetti, 2004)]. The polyacrylamide gel is a cross-linked matrix that functions as a sort of sieve through which the SDS-protein complexes pass as they are transported by the applied electric field. The complexes move from the negative end to the positive end of the gel and encounter different resistances from the polyacrylamide mesh. The smaller molecules are able to navigate the mesh faster than the larger ones. Migration, therefore, can be used to determine molecular weight [reviewed by (Righetti, 2004)]. After electrophoresis, the positions of the SDS-protein bands in the gel are visualized by staining the proteins with Coomasie blue dye. b. S D S - P A G E for Gelsol in , A c t i n , G A , G A 2 , G A 3 Prior to electrophoresis, all protein samples (20 u.L) were mixed with the same volume of sample buffer (80 m M Tr i s -HCl , 4.8% (w/v) sodium dodecyl sulfate, 0.16% (w/v) bromophenol blue, 16% (v/v) glycerol, 8% (v/v) 2-mercaptoethanol). The mixtures were heated in a boiling water bath for 5 minutes. Precast Ready Gels (10% acrylamide) (Bio-Rad) were loaded with samples and electrophoresis was performed in a M i n i -P R O T E A N 3 Electrophoresis apparatus (Bio-Rad), as shown in F ig . 20. 38 Sample w e l l s ^atlxxte -Buffer Stacking 4% Larger proteins / 12% Running gel Plastic frame At + Buffer"' Fig. 20. Mini-PROTEAN 3 Electrophoresis (http://www.life.uiuc.edu/biochem/ 455/notes/Grabner/lOSDSPAGE.pdf). After the gels were run, they were stained with 0.2% (w/v) Coomassie Brilliant Blue ( C B B ) , 45% (v/v) methanol, 45% (v/v) water, 10% (v/v) acetic acid for 1 hour. Then the gels were destained with several changes of 10% (v/v) methanol, 60% (v/v) water, 30% (v/v) acetic acid. 2.4 Cross-linking of GA 2 and GA 3 a. Cross linking regents Cross-linking is the process of chemically linking two or more molecules by a covalent bond or through a covalent bonded reagent. Cross-linking reagents contain reactive ends (succinimidyl esters, maleimides and iodoacetamides, hydrazide, carboxylic acids, etc.) toward proteins or other molecules. Cross-linking reagents have been used to assist in determination of three-dimensional structures o f proteins. Cross-linkers can be either homobifunctional (two identical reactive groups) or heterobifunctional (two different reactive groups). Homobifunctional cross-linkers are used to couple like functional groups — typically two thiols, two amines, two acids or two alcohols — and 39 form intramolecular crosslinks, or to prepare polymers from monomers. Heterobifunctional cross-linkers are used in sequential (two-stage) conjugations, helping to minimize undesirable polymerization or self-conjugation. The two-step strategy allows a protein that can tolerate the modification of its amines to be coupled to a protein or other molecule having different functional groups. (http://www.piercenet.com/ObjectsA^iew.cfm?type=Page&ID=169B5C9F-5024-44C7-B6C6-E03B7FC4A773) . Reactive Cross-linker Groups Functional Group Targets Reactive Cross-linker Groups Functional Group Targets Aryl Azide Non-selective (or primary amine) Maleimide Sulfhydryl Carbodiimide Amine/Carboxyl NHS-ester Amine Hydrazide Carbohydrate (oxidized) PFP-ester Amine Hydroxymethyl Phosphine Amine Psoralen Thymine (photoreactive intercalator) Imidoester Amine Pyridyl Disulfide Sulfhydryl Isocyanate Hydroxyl (non-aqueous) Vinyl Sulfone Sulfhydryl, amine, hydroxyl Table 3. Reactive cross-linker groups and their functional group targets (http://www.piercenet.com/Objects/View.cfm?type=Page&ID=169B5C9F-5024-44C7-B6C6-E03B7FC4A773) E D C (l-ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride), a zero-length cross linking regent reacts with carboxylic acid groups and activates the carboxyl group to form an active O-acylisourea intermediate, allowing it to be coupled to an amino group in the reaction mixture. This intermediate reacts with a primary amine to form an 40 amide derivative. The cross-linking reaction is usually performed between p H 4.5 to 5 and requires only a few minutes (for most applications), b. E D C Cross- l inking of G A 2 and G A 3 complexes The usage of E D C was based on a modified method of Do i (Doi et al., 1987). E D C was added to G A 2 or G A 3 complexes (1 mg/ml) to a final concentration of 20 m M . The mixed solution was incubated at room temperature for 30 min. In order to terminate the reaction, excess 2-mercaptoethanol (about 20 m M ) was added to the cross-linked solution. Then the mixture was filtered through a 0.22 pm M i l l e x filter unit (Millipore) to remove the precipitate, and centrifuged in Biosep centrifugal concentrators with a Sorvall SS34 rotor to remove the excess E D C or mercaptoethanol. 2.5 Protein Crystallization a. Introduction Proteins crystals are made up of a repeating 3D arrangement o f molecules (typically 10 1 5 molecules) with a high degree of symmetry. These molecules are held together in the crystal by weak noncovalent intermolecular interactions. The spaces between these molecules are occupied by disordered solvent molecules with a volume fraction of 30%-90%. The strategy of protein crystallization is to bring a solution of the protein towards a state of minimum solubility and thereby achieve a limited degree of supersaturation. The spontaneous relaxation from the non-equilibrium state of a supersaturated solution to the equilibrium state drives crystallization. In the equilibrium state, the chemical potential of each component in the crystalline state is the same as that in solution. 41 The most popular method of attaining supersaturation is vapor diffusion from hanging drops. A s shows in Fig. 21, about 4 p i of solution containing the protein (generally 5-30 mg/ml) and well solution is suspended from the underside of a microscope cover slip. This is placed over a well containing 1 ml of crystallizing solution (buffer and crystallization agents), which has at least one solute component having a significantly higher concentration than that in the hanging droplet. This system equilibrates by the distillation o f H2O from the sample droplet into the more concentrated reservoir solution. Thus, the salt and protein concentrations in the sample droplet increase, leading to supersaturation of the protein component. Factors such as protein purity, buffer type, p H , temperature, ionic strength, and the presence of organic solvents w i l l affect the crystallization of proteins. A t least 95% chemical purity is required to produce crystals for most proteins. A homogeneous population of conformations of the protein molecule is also a prerequisite for crystallization. B y screening different buffer types (e.g. N a A c / H A c , M O P S - N a O H , T r i s -HCl , etc), different types and concentrations of crystallization agents (salt, organic solvent or polyethyleneglycol) and different temperatures of incubation, a specific condition for protein crystallization may be found. Equilibration and crystal formation may require as much a month or more, depending on the protein involved and the choice of conditions. 42 r e s e r v o i r s d u t x s n Fig. 21. Hanging drop crystallization set up (http: //www. hamptonresearch .com) b. G A 2 complex B y screening the p H of buffers (pH 4.4-8.0), molecular size o f P E G (PEG400, PEG3000 and PEG8000) and the concentrations of buffer components (75-100 mM), P E G and C a 2 + (1-2 m M ) , G A 2 crystals were grown at 4 °C from a gel fdtered G A 2 sample at 10 mg/ml mixed with the following reservoir solutions: 1. 75-100 m M N a A c (pH 4.45-4.50), 1.5-2% P E G 8000 (w/v), 2 m M C a C l 2 at a 1:1 ratio (2 pi + 2 pi) (v/v); 2. 100 m M N a A c (pH 4.50), 1.5% P E G 8000 (w/v), 10% P E G 400 (v/v), 2 m M C a C l 2 at a 1:1 ratio (2 p i + 2 pi) (v/v); 3. 100 m M N a A c (pH 4.50), 4.8% P E G 3350 (w/v), 2 m M C a C l 2 at a 1:1 ratio (2 pi + 2 ul) (v/v). c. GA 3 complex B y screening the pH of buffers, molecular size of P E G and the concentration of buffer component, P E G , C a 2 + and N a C l , G A 3 crystals were grown at 4 °C from a gel fdtered GA3 sample at 10 mg/ml mixed with a reservoir solution of 100 m M N a A c (pH 4.60), 2% P E G 8000 (w/v), 2 m M C a C l 2 at a 1:1 ratio (2 pi + 2 pi) (v/v). 43 d. G A complex B y screening crystallization conditions, G A crystals were grown at 4 °C from a gel fdtered G A sample at 10 mg/ml mixed with a reservoir solution of 100 m M N a A c (pH 4.50-4.70), 2% P E G 8000 (w/v), 5 m M E G T A at a 1:1 ratio (2 pi + 2 pi) (v/v). e. EDC-cross linked GA3 complex E D C - G A 3 crystals were grown at 4 °C from an E D C - G A 3 sample at 10 mg/ml with the following reservoir solutions: 1. 100 m M N a A c (pH 4.68), 2% P E G 8000 (w/v), 2 m M C a C l 2 at a 1:1 ratio (2 pi + 2 pi) (v/v). 2. 100 m M N a A c (pH 4.48), 2.5% P E G 8000 (w/v), 2 m M C a C l 2 at a 1:1 ratio (2 pi + 2 ul) (v/v). 2.6 X-ray crystallography a. Introduction Crystallography is the most often used technique to analyze macromolecular structure at atomic resolution. It is capable of producing the complete 3D structure of a molecule without molecular weight limit. It is well known that when light waves hit the atoms in crystals, the scattered waves contain information about the structural arrangement of the atoms. Thus, when X-rays hit protein crystals, the scattered X-rays from individual atoms in the periodic structure combine constructively where Bragg's law conditions are satisfied and are detected in the form of diffraction spots on film or other image detectors. In order to calculate an electron density map, the amplitudes of the reflections and the relative relationships of their wavefronts are needed. The amplitude of 44 the reflection can be measured directly from the observed intensity, but the phase problem must be solved, for example, by heavy-atom multiple isomorphous replacement or molecular replacement. A plane of electron density through the protein structure is obtained in this way. B y rotating the crystals, every possible set of planes that can satisfy Bragg's law and yield diffraction is detected. Thus the electron densities within the entire volume of the protein structure can be calculated. The combined diffraction spots from every rotation constitute a complete data set and an initial electron-density map can be obtained. B y fitting the known protein amino acid sequence, a preliminary model of protein structure can be interpreted from this first electron density map. This model is then refined until the observed experimental structure factors agree with the calculated structure factors from the model. The final model can be regarded as the actual protein structure. Synchrotron radiation is the electromagnetic emission from charged particles submitted to acceleration. After electrons are accelerated to high speed (high energy) in a linear accelerator, they are injected into a booster ring and storage ring in sequence. A n extremely brilliant full spectrum beam of photons can be produced, which is called synchrotron light. The use o f synchrotron radiation sources (high energy electron or positron beams) can provide extremely intense X-rays, a requirement for obtaining high quality diffraction data from small crystals and to shorten data collection times. Furthermore, different wavelengths of synchrotron light can be selected to facilitate phasing by anomalous diffraction methods. Crystals should be flashed to liquid nitrogen temperature in the presence of a cryoprotectant solution prior to data collection. A t liquid nitrogen temperatures, the 45 diffusion of free radical created by incident X-ray radiation in crystals can be slowed down. Thus when crystals are exposed to an X-ray beam, the rate of radiation damage to crystals w i l l be minimized. The use of cryoprotectant solutions also suppresses crystallization of water present in the crystal when cooled to liquid nitrogen temperatures. Ice formation can damage the protein crystal lattice and overwhelm some regions of the diffraction pattern. The role of cryoprotectant is to cause the solvent to solidify in a vitreous, glass-like state and avoid ice formation. In practice, a crystal is trapped in a small nylon fiber loop glued to a metal pin and transferred to droplet containing about 5 pi of cryoprotectant solution. After several minutes of incubation, the crystal is picked up in a cryoloop, immersed in liquid nitrogen, placed into a Hampton magnetic cryo-cap storage vial and transferred to a cane in a cryogenic transport dewar filled with liquid nitrogen. b. Freezing of GA2 and G A 3 The cryoprotectant solutions for freezing GA2 and GA3 were as follows: 1. 100 m M N a A c , 5 m M C a C l 2 , 2% P E G 8000 (w/v), 20-35% PEG400 (v/v); 2. 100 m M N a A c , 5 m M C a C l 2 , 5% P E G 8000 (w/v), 20-35% PEG400 (v/v); 3. 100 m M N a A c , 5 m M C a C l 2 , 5% P E G 8000 (w/v), 10-25% glycerol (v/v), 400 m M N a C l ; 4. 100 m M N a A c , 5 m M C a C l 2 , 10% P E G 8000 (w/v), 10-25% glycerol (v/v), 400 m M N a C l . 5. Wel l buffer ( G A 2 or G A 3 crystallization condition), 8% Trehalose Dihydrate (w/v), 15-25% glycerol (v/v) c. Freezing of G A The cryoprotectant solutions for freezing G A were as follows: 1. Wel l buffer ( G A crystallization condition), 8% Trehalose Dihydrate (w/v), 20-25% glycerol (v/v); 46 2. 100 m M N a A c , 5 m M E G T A , 2% P E G 8000 (w/v), 8% Trehalose Dihydrate (w/v), 25% glycerol (v/v); 3. 100 m M N a A c , 5 m M E G T A , 10% P E G 8000 (w/v), 20% glycerol (v/v), 400 m M N a C l . d. Terbium and Cadmium soaking of GA 2 and GA3 3+ 2+ The Tb (or C d ) soaking conditions were as follows: 1. 100 m M N a A c , 0.5-1 m M T b C l 3 , 2% P E G 8000 (w/v), 25-35% PEG400 (v/v); 2. 100 m M N a A c , 10 m M T b C l 3 (or Cd(Ac) 2 ) , 2% P E G 8000 (w/v), 25-35% PEG400 (v/v); 3. 100 m M N a A c , 0.5-1 m M T b C l 3 (or Cd(Ac) 2 ) , 10% P E G 8000, 15%-25% glycerol (v/v), 400 m M N a C l ; 4. 100 m M N a A c , 10 m M T b C l 3 ( o r Cd(Ac) 2 ) , 10% P E G 8000, 15%-25% glycerol (v/v), 400 m M N a C l . e. Data collection X-ray diffraction data were collected by Dr. Robinson at 100 K at ID 7-11, M A X - L a b synchrotron facility, Lund, Sweden. 2.7 Luminescence spectroscopy Fluorescence and phosphorescence spectra were generated using an LS-5B Luminescence Spectrometer equipped with a P E 7500 Professional Computer (Perkin Elmer, Norwalk, CT) . Luminescence spectra were measured with excitation and emission monochromator slit widths of 10 nm. Data acquisition in the phosphorescence mode for T b 3 + studies did not begin until after a delay time of 0.1 ms after excitation with a lamp pulse at 296 nm to eliminate detection of any fluorescence signals or light scattering in the emission signal. Data were collected over a gating time of 9.9 ms, prior to the next lamp pulse. Excitation was at 296 nm and luminescence was measured at 554 nm for all titrations. The 296 nm excitation was chosen to selectively excite tryptophan residues so that tryptophan residues could act as an energy transfer donors to terbium. 47 Gelsolin titrations with T b 3 + were performed by adding stock TbCh solution (134 m M ) to 1.2 ml of 1.12 u M gelsolin in 15 m M M O P S , 0.2 m M E D T A , p H 7.1, with or without 0.5 m M A T P present. A similar set of experiments was performed by adding 40 m M TbChj stock solution to 1.2 ml 0.76 p M gelsolin (from another preparation) in 20 m M M O P S , 0.1 m M E D T A , 100 m M N a C l , p H 7.1, both with and without 0.5 m M A T P present. 48 Chapter Three — Results and Discussion 3.1 Purification of gelsolin and complexes a. Purification of gelsolin Both HPLC-based anion exchange chromatography and conventional open-column anion exchange chromatography were used to purify gelsolin from plasma that had been batch-treated with anion exchange resin in the presence of calcium ions, and then swamped with E D T A . H P L C has the advantage of decreased times of separation and improved resolution, but the loading volume is limited by the size of column. For practical purposes, the open-column technique is not limited by loading capacity, but the separation time is much longer. i—•—i—'—i—1—i—•—i—'—i—1 i • i • i • 180 200 220 240 260 280 300 320 340 The concentration of NaCl (mM) Fig. 22. Elution of gelsolin from a DEAE-Sephadex A-50 ion exchange column (36 x 4.5 cm) by a gradient of NaCl concentration in 25 mM Tris-HCl, 1 mM EDTA, pH 7.8, at the rate of 2 ml/min. The EDTA-adjusted plasma was applied to a DEAE-Sephadex A-50 ion exchange column (36 x 4.5 cm) and eluted using a salt gradient (Fig. 22). The absorbance of the 49 eluant was monitored at 280 nm and recorded at a sensitivity of 0.2 A U F S using a chart recorder. In the Aff iGel-Blue affinity column purification procedure, washing with 2.5 m M A T P did not always quantitatively elute gelsolin from the column. In such cases, a further wash with 1 M N a C l , 25 m M Tr i s -HCl , 1 m M E D T A , p H 8.0, removed additional gelsolin. However, this gelsolin contained contaminants, making it necessary to add a further purification step. The eluant was concentrated to 20 ml and dialyzed against two changes of 1.5 L of 25 m M Tr i s -HCl , 1 m M E D T A , p H 8.0, over two days. Finally, the gelsolin solution was applied to a gel filtration column (Bio-Rad Sephacryl S300; 90 x 2.5 cm) at 4 °C and eluted with 25 m M Tr i s -HCl , 1 m M E D T A , p H 8.0, at a rate of 2 ml/min. The gelsolin elution was monitored with an E C O N O U V monitor (Bio Rad) at 280 nm and recorded at a sensitivity of 0.2 A U F S using a chart recorder, as shown in Fig. 23. I—i—i—•—i—•—i—i—i—i—i—i—i—'—i—<—i—>—i—• 0 20 40 60 80 100 120 140 160 180 Time (min) Fig. 23. Elution of gelsolin from a Bio-Rad Sephacryl S300 size exclusion column (90 x 2.5 cm) in 25 mM Tris-HCl, 1 mM EDTA, pH 8.0, at a rate of 2 ml/min 50 From S D S - P A G E results (Fig. 37), this first peak (Fig. 23) contained a component of 200 kDa, and the second peak (Fig. 23) contained gelsolin and/or its cleaved product, b. Purif ication of actin In the final stage of purification of actin, it was eluted from a Sephacryl S300 gel-filtration column (90 x 2.5 cm) as in Fig. 24. The absorbance of the eluant was monitored at 280 nm and recorded at a sensitivity of 0.1 A U F S using a chart recorder. The first small peak was from F-actin. The second peak contained G-actin. However, since the first half of the second peak contained actin dimer or trimer, only the second half was collected and used as pure G-actin. _ J \ —i—•—i—1—i—>—i—•—i—>—i 0 20 40 60 80 100 120 140 160 180 Time (min) Fig. 24. Elution of actin from a Bio-Rad Sephacryl S300 size exclution column (90 x 2.5 cm) in buffer A at a rate of 2 ml/min. 51 c. Purif ication of GA 2 The G A 2 complex was applied to a Sephacryl S300 gel filtration column (90 x 2.5 cm) and eluted as in Fig . 25. The absorbance of the eluant was monitored at 280 nm and recorded at a sensitivity of 0.1 A U F S using a chart recorder. The elution profile shows overlapping peaks. The leading edge shoulder is from F-actin; the main peak is a complex of gelsolin and actin (confirmed by S D S - P A G E ) . Only the central portion of the main peak was collected and used for crystallization trials. T ' 1 ' 1 " 1 • 1 ' 1 ' 1 • 0 20 40 60 80 100 120 140 Time (min) Fig. 25. Elution of GA 2 complex from a Bio-Rad Sephacryl S300 size exclusion column (90 x 2.5 cm) in buffer A at a rate of 2 ml/min. The main impurity to elute first from the AffiGel-Blue column by 1 M N a C l and then from gel-filtration column (Fig. 23) was a 68 kDa fragment of gelsolin, possibly the result of action of a serine protease, such as Furin, during the course of the preparation (Huff et al., 2003). Surprisingly, the 68 kDa fragment also formed a stable complex with 52 G-actin that could be isolated by gel filtration chromatography (Fig. 26). Only the middle part of the main peak was collected and used in crystallization trials. i • 1 • 1 1 1 • 1 ' 1 40 60 80 100 120 140 Time (min) Fig. 26. Elution of a complex formed between actin and a 68 kDa fragment of gelsolin from a Bio-Rad Sephacryl S300 size exclusion column (90 x 2.5 cm) in buffer A at a rate of 2 ml/min. d. Purif ication of GA3 The elution profile for GA3 purification by gel filtration is shown in Fig. 27. Absorbance of the eluant was monitored at 280 nm and recorded at a sensitivity of 0.1 A U F S using a chart recorder. The main peak contained gelsolin and actin (confirmed by S D S - P A G E ) ; the trailing edge was discarded. Only the middle part of the main peak was collected and used for crystallization trials. 53 I 1 , 1 1 1 1 1 1 1 1 1 0 50 100 150 200 250 Time (min) Fig. 27. Elution of GA 3 complex from a Bio-Rad Sephacryl (90 x 2.5 cm) size exclusion column in 2 mM Tri-HCI, 0.2 mM CaCI2, 100 mM KC1, 2 mM MgCl 2, 0.2 mM ATP, 1 mM DTT, pH 7.5-7.8, at a rate of 2 ml/min. e. Purif ication of G A Following formation of G A , this complex was subjected to gel filtration chromatography, as shown in Fig . 28. Absorbance of the eluant was monitored at 280 nm and recorded at a sensitivity of 0.05 A U F S using a chart recorder. According to SDS-P A G E results, the main peak contained a complex of gelsolin and actin. When the gelsolin used to form G A complex contained gelsolin (83 kDa) and cleaved gelsolin (68 kDa), two peaks were eluted, as shown in Fig . 29. Since both of components can form complexes with G-actin, only the first peak was concentrated and used for crystallization trials. 54 I—•—I—I—I—I—I—I—I—'—I—I—I—•—I—'—I—'—I—'—I—•—I 50 60 70 80 90 100 110 120 130 140 150 160 Time (min) Fig. 28. Elution of GA complex from a Bio-Rad Sephacryl (90 x 2.5 cm) size exclusion column in 10 mM MOPS/KOH, 50 uM MgCl 2, 0.1 mM EGTA, pH 6.8, at a rate of 2 ml/min. I—i—i—"—i—•—i—•—i—>—i—1—i—1—i—<—i—•—i—1 40 50 60 70 80 90 100 110 120 130 Time (min) Fig. 29. Elution of GA complex (gelsolin used to form GA complex contained gelsolin and cleaved gelsolin) from a Bio-Rad Sephacryl (90 x 2.5 cm) size exclusion column in 10 mM MOPS/KOH, 50 uM MgCl 2, 0.1 mM EGTA, pH 6.8, at a rate of 2 ml/min. 55 3.3 SDS-PAGE analyses a. Gelsolin and actin We used electrophoresis to test the purity of gelsolin purified by different methods. Addition of P M S F to the freshly thawed plasma as a general inhibitor of serine proteases led to the results shown in Fig . 30. Lane 1 contained standard molecular weight markers (Amersham Biosciences). Lane 2 contained gelsolin eluted by A T P from an Aff i -Gel Blue column. It migrated as a single band, in accordance the molecular mass of gelsolin (83 kDa). Lane 4 contained gelsolin eluted by 1 M N a C l from the Affi-gel Blue column. It migrated as two bands, gelsolin (83 kDa) and cleaved gelsolin (68 kDa). The cleaved gelsolin (68 kDa) could be from proteolysis by furin or another serine protease active in plasma or prior to secretion of gelsolin into the plasma (Huff et al., 2003). Omission of P M S F from the purification procedure had serious consequences, as shown in F ig . 31. Lane 1 contained G-actin with a molar mass of 42 kDa. Lane 2 contained gelsolin (83 kDa) and cleaved gelsolin (68 kDa) in almost equal proportions. Clearly, successful purification of gelsolin in good yield requires use of a serine protease inhibitor during the procedure. 56 k D a 1 4 160-105-75 50-35-• Fig. 30. Gel electrophoresis of gelsolin and GA 2 complex. Lane l: molecular mass markers in kiloDaltons; lane 2, gelsolin eluted by ATP from affinity gel column; lane 3, GA 2 complex purified from gel filtration column; lane 4, gelsolin eluted by 1 M NaCl from affinity gel column (during gelsolin purification). 57 kDa 1 2 3 Fig. 31. Gel electrophoresis of actin, gelsolin and GA complex. Lane 1, actin; lane 2, gelsolin (prepared without PMSF) eluted with 1 M NaCl from the affinity gel column and then subjected to gel-filtration; lane 3, GA complex (prepared with material run in lane 2) subjected to gel filtration. b. GA, GA2 and GA3 in solution We used electrophoresis to test the purity of G A , GA2 and GA3 complex before crystallization. Lane 3 (Fig. 30) contained G A 2 purified by gel filtration, and displayed one band corresponding to gelsolin (83 kDa), another to actin (42 kDa). Lane 3 (Fig. 31) contained a nominal " G A " prepared by gel filtration, and displayed one band corresponding to intact gelsolin (83 kDa), one to cleaved gelsolin (68 58 kDa), and another to actin (42 kDa). The gelsolin used to form this " G A " was prepared without P M S F , shown in lane 2 of Fig. 31. KDa 1 2 250 160 105 75 50 35 Fig. 32. Gel electrophoresis of GA complex. Lane 1, molecular mass markers in kiloDaltons; lane 2: GA complex purified from gel-filtration column (based on strategy 2). When P M S F was used during gelsolin purification, the G A complex from this gelsolin displayed only two bands: one band corresponding to gelsolin, another to actin, as shown in Fig . 32. We also tested the purity our G A 3 complex by electrophoresis before crystallization (Fig. 33). 59 k D a 1 2 25a 160 105-7 5 -50 Fig. 33. Gel electrophoresis of GA 3 complex. Lane l, molecular mass markers in kiloDaltons; lane 2: GA 3 complex purified from gel-filtration column. To confirm the stoichiometry of our gelsolin-actin complexes, we ran a series of standard gelsolin and actin mixtures together with G A 2 and GA3 complexes from the gel filtration columns in the same gel (Fig. 34). In the standards, we mixed different ratios of gelsolin with actin, keeping the loading level of gelsolin the same by loading different volume of mixtures. Thus the intensity of gelsolin was similar in all lanes containing standards and also close to the intensity of gelsolin in G A 2 and G A 3 complexes. B y comparison of the relative intensities of actin bands, the G A 2 bands, as expected, were 60 close to those of 1:2 molar mixture of gelsolin and actin, and the GA3 bands were close to those for 1:3 standard mixture. 1 2 3 4 5 6 7 gelsolin acliti-Fig. 34. Gel electrophoresis of gelsolin actin mixtures at different molar ratios, and GA 2 , and GA 3 complexes. Lane 1, gelsolimactin =1:1; Lane 2, gelsolin:actin = 1:1.5; Lane 3, gelsolin:actin = 1:2; Lane 4, gelsolin:actin = 1:2.5; Lane 5, gelsolimactin = 1:3; Lane 6, GA 2 complex from gel filtration column; Lane 7, GA 3 complex from gel filtration column. c. GA2 and GA3 crystals To test the stoichiometry of our gelsolin-actin crystals grown from G A 2 and G A 3 solutions, we also ran S D S - P A G E analysis of our crystals. The crystals were picked up by fiber loops, washed in the well buffer, and then dissolved in electrophoresis sample buffer. From the results (Fig. 35), the actin intensities in lane 1 (GA3), lane 2 (GA3) and lane 3 ( G A 2 ) were almost the same, while the gelsolin intensity in lane 3 was significantly fainter than those in lanes 1 and 2, in accordance with a G : A mole ratio in G A 2 of 1:2 (or 0.5:1), and a G : A mole ratio in G A 3 of 1:3 (or 0.33:1). Also , the gelsolin 61 intensities in lane 1 ( G A 3 ) and lane 6 (GA2) were similar, while the actin intensity in lane 1 was higher than that in lane 6, consistent with the higher actin content of G A 3 compared to G A 2 . 1 2 3 4 5 6 actin — Fig. 35. Gel electrophoresis of GA 2 and GA 3 crystals. Lanes l and 2, GA 3 complex crystals (crystallization trial initiated on July 27, 2004); Lanes 3 and 4, GA 2 complex crystals (crystallization trial initiated on August 13, 2004); Lanes 5 and 6, GA 2 complex crystals (crystallization trials initiated on July 8, 2004). In order to determine the stoichiometries of our G A 2 and G A 3 crystals accurately, we also ran a series of standard gelsolin and actin mixture together with G A 2 and G A 3 crystals in the same gel (Fig. 36). These standards were made and loaded in the same way as in Fig . 34. Thus, the intensity of gelsolin was similar in all lanes containing standards and also close to the intensity of gelsolin in lanes containing protein from G A 2 and G A 3 crystals. A s expected, the actin intensity in G A 2 crystals was close to that for the 1:2 molar mixture of gelsolin and actin, and the actin intensity in G A 3 was close to that for the 1:3 standard mixture. 62 1 2 3 4 5 6 7 8 9 10 g el soli n-actin-Fig. 36. Gel electrophoresis of gelsolin-actin mixtures at different molar ratios, and of samples prepared from GA 2 and GA 3 crystals. Lane 1, gelsolin:actin = 1:0.5; Lane 2, gelsolin:actin = 1:1; Lane 3, gelsolin:actin = 1:1.5; Lane 4, gelsolin:actin = 1:2; Lane 5, gelsolin:actin = 1:2.5; Lane 6, gelsolin:actin = 1:3; Lanes 7 and 8, GA 2 crystals (crystallization trials initiated on July 8, 2004); Lanes 9 and 10, GA 3 crystals (crystallization trials initiated on July 27, 2004) d. Cleaved gelsolin and actin complex in solution A s discussed previously, preparation of gelsolin without P M S F led to contamination with a 68 kDa fragment of gelsolin (Fig. 31), which could be isolated in essentially pure form (Fig. 37). This cleaved gelsolin also bound to actin to form a complex (Fig. 38) that could be purified by gel filtration in the same conditions as for a G A 2 complex prepared from intact gelsolin. 63 KDa 1 2 3 250-' l i t * - ' ' « 150 100 75 50-37 Fig. 37. Gel electrophoresis products eluted from gel-filtration during a gelsolin preparation. Lane l, protein standards in kiloDaltons; Lane 2, product of the second peak from a gel filtration column; Lane 3, the product of the first peak from a gel filtration column In Fig. 38, lane 1 contained protein standards (Bio-Rad), while lanes 2 and 3 contained a complex formed from actin and a fragment of gelsolin (68 kDa). 64 k D a 1 2 3 250 150 100 75 50 37 Fig. 38. Gel electrophoresis a complex prepared from actin and cleaved gelsolin. Lane 1, protein standards i kiloDaltons; lanes 2 and 3, complex purified by gel filtration. 65 3.3 Mass spectrometry results Matrix-Assisted Laser Desorption/Ionization ( M A L D I ) mass spectrometry ( V O Y A G E R - D E S T R workstation M A L D I - T O F ) was used to try to characterize our G A 2 and GA3 complexes. In M A L D I , a low concentration of analyte is uniformly dispersed throughout the solid or liquid matrix. Energy is able to transfer from the laser to the analyte via the matrix with high efficiency. Since the matrix has a strong absorbance at the laser wavelength, the energy causing the resonant excitation and successful energy transfer from matrix to analyte is lower than the dissociation energy of the analyte. Thus the dissociation of analyte could be reduced [reviewed by (Mann et al., 2001)]. A schematic representation o f a M A L D I experiment is shown in F ig . 39. Firstly, a "solid solution" is formed in which the analyte molecules are distributed throughout the matrix and completely isolated from one other. Secondly, matrix excitation is stimulated by laser energy incident on the solid solution. The solid solution undergoes local disintegration and clusters are formed in which a single analyte molecule is surrounded by neutral and excited matrix molecules. The matrix molecules evaporate away from these clusters to leave excited analyte molecules. Finally, the analyte molecules become ionized by simple protonation involving the photo-excited matrix, leading to the typical formation of [ M + X ] + type species (where X = H for protonation, but might be L i , Na, K , etc. in other situations). In addition, some multiply charged species, dimers and trimers can also be formed [reviewed by (Mann et al., 2001)]. (http://www-methods. ch. cam. ac .uk/meth/ms/theory/maldi .html) 66 T i m e - o f - f l i g h t Mass A n a l y s e r © 2000 PAUL GATES Fig. 39. The mechanism of MALDI. (http://www-methods.ch.cam.ac.uk/meth/ms/theorv/maldi.html) M A L D I - M S analysis of a sample of G A 2 , prepared by gel fdtration, gave clear peaks that could be assigned to actin (around 42 kDa) and gelsolin (around 83 kDa), but not at a mass expected (around 167 kDa) for the intact complex (Fig. 40). It is probable that the laser pulse of the M A L D I technique broke down the non-covalent interaction between gelsolin and actin. 67 Voyager Spec #1[BP = 10103.2,1423] 100 90 80 70 60 a so 40 30 20 10 41914.73 I42332.09 1407.67 1188.98 45 4&19.24 |VE7776.77 |ftl296.38 ^8123.04 50685.37 M490.20Lto047.16 54052.09 ,1.13-2562547.05 84281.66 ,RJ9 2#8060.80 jl 87493.58 ' 8 5 . 4 « # q 9 8 . 8 6 l ° 1 3 2 " 2 1 2 0 272.35 H???^ 0 6 2 t-^! 3 7 : 5 4 138993.U0912.87 23U97l|l*1. rnWMiw^™iMiw:wHimiiiiiwHiuifiililjiti'UlLJi .uk"1' 614. 32990.0 66394.4 99798.8 133203.2 Mass (m/z) 168385.66.,,.,. , 21 166607.6 200012.0 Fig. 40. MALDI-MS spectrum of GA 2 (10 mg/ml). Spot size, 1 ul + 1 ul of Sinapinic acid (SPA) as the matrix. Spectra shot calibrated with Ciphergen lgG. The acquisition mass range was 33-200 kDa. M A L D I - M S analysis of samples of G A 2 that had been chemically crosslinked with E D C did reveal new peaks around 129 kDa and 166 kDa, as would be expected for complexes consisting of one gelsolin and one or two actins, respectively (Fig. 41) 68 Voyager Spec #1[BP = 10032.1, 369] 100 90 80 70 60 2 50 40 30 20 10 10031.60 10501.50 10776.22 11495.87 12018.31 12268.20 12542.38 ,,13171.88 114 >25.93 $223.96 20525.91|(3575| ^2548 Ji:>3! 84861.77 86259.01 86699.48 6466.43 6959.84 369. 187890.10 89487.86 S?571.51 i8639.86 | 9 3 2 5 5 - 9 f t 5 6 2 3 . 4 127974.34 129407.34 l.a82633.34 173608.06 1037.384582.45 ' 83^817.96 9995.0 47999.4 86003.8 124008.2 Mass (m/z) 162012.6 200017.0 Fig. 41. MALDI-MS spectrum of EDC-crosslinked GA 2 (10 mg/ml). Spot size, 1 ul + 10 ul with Sinapinic acid (SPA) as the matrix. Spectra shot calibrated with Ciphergen lgG. The acquisition mass range was 10-200 kDa. M A L D I - M S analysis o f G A 3 purified by gel filtration (Fig. 42) again showed only gelsolin (84 kDa) and actin (42 kDa) peaks. The poor quality of this spectrum may be the result of the high salt content of the sample, which was dissolved in a buffer containing 100 m M KC1 and 2 m M M g C l 2 . 69 Voyager Spec #1[BP = 10035.6, 745] 100 90 80 70 60 S 50 c 40 30 20 10 10036.29 10315.95 10728.96 11317.92 ii 1567.54 ,12133.90 13498.28 14535.03 .745. 41260.06 lo460.26 81384.26 00*1781.03118107.63 •^i?l,r3S!W62^fl,7,76 192120.4ai2016.^95844 9999.0 58001.4 106003.8 154006.2 202008.6 250011.0 Mass (m/zl Fig. 42. MALDI-MS spectrum of GA 3 (10 mg/ml). Spot size, 1 ul + 10 ul with Sinapinic acid (SPA) as the matrix. Spectra shot calibrated with Ciphergen lgG. The acquisition mass range was 10-250 kDa M A L D I - M S analysis of EDC-crosslinked GA3 revealed five peaks: 42 kDa, 84 kDa, 125 kDa, 167 kDa and 209 kDa (Fig. 43). These peaks could be from actin, gelsolin, crosslinked G A , crosslinked GA2 and crosslinked GA3, respectively. Because the relative intensities of the complexes peaks were weak, the acquisition mass range was changed to 45-230 kDa and then to 90-230 kDa to exclude the 42 kDa and 83 kDa peaks Figs. 44 and 45, respectively). In the resulting spectra, the G A , GA2 and GA3 complex peaks are clear. 70 Voyager Spec #1[BP = 41970.9,1644] 41971.37 90 70 60 | 50 30 42417.14 J41613.76 1404.98 1195.41 |t|3613.06 9965.58 84325.43 J '43022.60 al2566.01 3932(1.31*567.39 62724.98 ijl369.pl, . n o o n , ,o125592.84 154471 92 VLiuAW.lii.iLi.uL. .I jTV 9 » 5 6 . 9 4 108803.38 . „ . „ , . . _ , , o w 1 1.37558.17 127835.81 , 196194.6319783.50 i.t iLiliH^i - U l i 34995.0 73996.2 112997.4 151998.6 Mass (m/z) 190999.8 Fig. 43. MALDI-MS spectrum of EDC-crosslinked GA 3 (10 mg/ml). Spot size, 1 ul + 2 ul with Sinapinic acid (SPA) as the matrix. Spectra shot calibrated with Ciphergen lgG. The acquisition mass range was 35-230 kDa Voyager Spec #1[BP = 83156.0,1092] 80 70 2 50 20 62506.09 ||83656.97 2237.25 83870.85 31976.03 31678.14 84342.76 62149.07 y 8 9 4 . 1 7 618 .^88630.39101706.26 I lfeQ429.82 166662.90 l14,„,11.#5.gE7092.55 146415.32 JIP4 5 4 8 ^ 8 7 9 6 ' 8 ? Q r 7 7 Q ,207418.07 44997.0 82001.6 119006.2 156010.8 Mass (m/z) 193015.4 Fig. 44. MALDI-MS spectrum of EDC-crosslinked GA 3 (10 mg/ml). Spot size, 1 ul + 4 ul with Sinapinic acid (SPA) as the matrix. Spectra shot calibrated with Ciphergen lgG. The acquisition mass range was 45-230 kDa 71 Voyager Spec #1[BP = 124496.1, 254] 254 89999.0 118002.2 146005.4 174008.6 202011.8 230015.0 Mass (m/z) Fig. 45. MALDI-MS spectrum of EDC-crosslinked GA 3 (10 mg/ml). Spot size, 1 ul + 6 pi with Sinapinic acid (SPA) as the matrix. Spectra shot calibrated with Ciphergen lgG. The acquisition mass range was 90-230 kDa 3.4 GA, G A 2 and G A 3 crystals and X-ray diffraction results G A , G A 2 and GA3 crystals from different crystallization conditions are shown in Figs. 46-51. A s seen with a stereo microscope, G A , GA2 and GA3 crystals have different geometric shapes. 72 mm Fig. 46. GA crystals (~0.1 mm) grown at 4°C from a gel filtered GA sample at 10 mg/ml mixed at a 1:1 ratio (v/v) with a reservoir solution of 100 mM sodium acetate buffer (pH 4.50), 2% PEG8000 (w/v), 5 mM EGTA. Fig. 47. A G A 2 crystal (-0.1 mm) grown at 4 °C from a gel filtered GA 2 sample at 10 mg/ml, mixed at a 1:1 ratio (v/v) with a reservoir solution of 100 mM sodium acetate buffer (pH 4.44), 1.5% PEG8000 (w/v), 2 mM CaCl 2. 73 Fig. 48. GA 2 crystals (-0.1 mm) grown at 4 C C from a gel filtered GA 2 sample at 10 mg/ml mixed at a 1:1 ratio (v/v) with a reservoir solution of 100 mM sodium acetate buffer (pH 4.50), 2% PEG 8000 (w/v), 2 mM CaCl2. Fig. 49. GA 2 (-0.1 mm) crystals grown at 4 °C from a gel filtered GA 2 sample at 10 mg/ml mixed at a 1:1 ratio (v/v) with a reservoir solution of 100 mM sodium acetate buffer (pH 4.50), 1.5% PEG8000 (w/v) + 10% PEG400, (v/v), 2 mM CaCl2. 74 Fig. 50. GA 3 crystals (-0.1 mm) grown at 4 °C from a gel filtered GA 3 sample at 10 mg/ml mixed at a 1:1 ratio (v/v) with a reservoir solution of 100 mM sodium acetate buffer (pH 4.60), 2% PEG8000 (w/v), 2 mM CaCl2. Fig. 51. An EDC-crosslinked GA 3 crystal (-0.1 mm) grown at 4 °C from a gel filtered EDC-crosslinked GA 3 sample at 10 mg/ml mixed at a 1:1 ratio (v/v) with a reservoir solution of 100 mM sodium acetate buffer (pH 4.68), 2% PEG8000 (w/v), 2 mM CaCl 2. 75 Gl-G3:actin GA 2 Wavelength (A) 0.970 0.974 Space group P3,21 P6522 Unit cell a=b= 144.6, c= 128.3 A a=b= 146.3, c= 388.8 A a =(3= 90°, y= 120° a = (3= 90°, y= 120° Resolution range (A) 30.0-3.50 (3.69-3.50) 30.0-6.80 (7.17-6.80) Total reflections 54943 (4035) 41945 (6261) Unique reflections 17781 (1659) 4492 (659) Redundancy 3.1 (2.4) 9.3 (9.5) Completeness (%) 88.9 (58.1) 96.0(100.0) Average I/CT 15.7 (3.3) 14.7 (5.2) Rmerge" (%) 6.2 (26.3) 13.2 (40.3) Rfactorb (%) 32.4 42.0 Correlation Coefficientb (%) 72.9 61.8 aRmerge(E|I-<I>|/I<I>) b f^actor and correlation coefficient following molecular replacement as defined in the CCP4 distribution of AMoRe Table 4. Data collection and molecular replacement statistics for terbium soaked crystals The X-ray structure from G A 2 crystals reveals the N-terminal half o f gelsolin binding to one actin, but does not reveal the C-terminus of gelsolin or a second actin. Although only one actin is visible in our crystals, our G A 2 crystals exhibit two kinds of space group: P3j21 and P6 5 22, as shown in Table 4. The former is the same as that of G l -G3/actin (Burtnick et al., 2004) and may indicate that proteolysis occurred during the time required for the crystals to nucleate. The second type of crystal possesses a larger unit cell (a= b= 146.3, c= 388.8 A ) , of sufficient volume to accommodate G A 2 , but only the Gl-G3/act in portion appears sufficiently ordered to yield diffraction. 76 Terbium is a good substitute for calcium in many calcium-binding proteins. Firstly, terbium has almost the same ionic radius and prefers similar coordination numbers as calcium, so that Tb could occupy Ca -binding sites without causing serious structural modifications. Also , both ions prefer to bind to charged or uncharged oxygen. T b 3 + ions in calcium-binding sites are easier to observe by X-ray diffraction methods than C a 2 + because Tb is more electron dense than Ca and Tb displays anomalous diffraction. We soaked both types of GA2 crystals in Tb 3 +-containing solutions in order to test the exchangeability of C a 2 + in the complexes. In the case of crystal belonging to space group P3121 (Fig. 52a), three terbium ions positions, characterized by the positive density in this map, are identical to the positions of type II metal ion binding sites in the Gl-G3/act in structure and are labeled G1:II, G2:II and G3:II. A fourth terbium ion ( G l :I) is in a position where a type I calcium ion is found in the structure. A n actin-ATP-associated T b 3 + (labelled " A " ) also is indicated. In the case of the terbium ions soaked into a novel crystal form (space group P6s22) (Fig. 52b), T b 3 + was also able to displace C a 2 + from the type I site between actin and G l . It also occupies all three of the expected type II metal ion binding sites in G1-G3, including the type II site in G2, which was vacant in the Gl-G3/ac t in structure reported in the presence of calcium ions (Burtnick et al., 2004). But no Tb was found associated with the actin-ATP. Interestingly, when the terbium ion-soaked crystals from space group P 6 5 2 2 were backwashed with mother liquor containing 10 m M CaCh in an attempt to displace terbium ions (Fig. 52c), T b 3 + was replaced only at the type II sites in G l and G3. The type II site in G2, which was unoccupied in the structure of Gl-G3/act in , and the type I site sandwiched between actin and G l , remained occupied by terbium ions. 77 Our results indicate the type II C a 2 + in G l to be exchangeable. Previously, this type II C a 2 + in G l was regarded as non-exchangeable. Based on the structures of G l /actin (McLaughlin et al., 1993) and G1-G3/actin (Burtnick et al., 2004), there are two calcium binding sites in G l : one is located in the interface of G l and actin (type I C a ); the other is within the G l moiety (type II Ca 2 + ) . But biochemical analysis detected only one calcium in the Gl /ac t in complex at p H 8.0 (Way et al., 1990). This calcium proved to be the intramolecular (type II) C a 2 + , as inferred by the lack o f effect o f mutation of acidic residues in either actin or G l that are involved in ligation of the intermolecular type I C a 2 + (Weeds et al., 1995). Furthermore, this type II calcium in G l is probably the one trapped in G A complexes (Weeds et al., 1995). Since the trapped calcium in G A is not removed by chelating with E G T A , this type II C a 2 + in G l is regarded as non-exchangeable (Ditsch and Wegner, 1995). Under our conditions, however, this ion does prove to be exchangeable. Furthermore, once this terbium is into the site, it can be removed and replaced by elevating the C a 2 + concentration in the medium bathing the crystals (Fig. 52c). Our results also show Tb fdls the previously vacant type II site in G2. The G2 type II metal ion binding site is vacant in Gl-G3/act in (Burtnick et al., 2004). It was proposed that calcium at this site plays only a transient role in destabilizing the inactive relative to the active structure of gelsolin, and is released prior to or on full activation. 4^- 74-Tb , with its higher charge density and higher binding affinity than Ca , can occupy this type II site (Figs. 52a and 52b) and is not displaced by washing with a higher concentration of calcium-containing solution (Fig. 52c). For the same reason, T b 3 + can 78 replace the C a 2 + at the type I site in G l (Figs. 52a and 52b) and is not displaced by washing with a higher concentration of calcium-containing solution (Fig. 52c). Fig. 52. Terbium ion substitution in G1-G3 domains of complexes. A) A cartoon representation of the structure of G l -G3/actin (PDB i.d. 1RGI). Actin is coloured gray and the domains of gelsolin are coloured: G l , red; G2, light green; and G3, yellow. The terbium anomalous electron density map, contoured at 5 a (black) and derived from terbium soaked crystals of Gl-G3/actin, is superimposed onto the structure. B) A cartoon representation of the structure of G l -G3/actin in GA 2 , with a Tb 3 + anomalous electron density map superimposed, as in (A). In this case the terbium ions were soaked into a novel crystal form (space group P6S22) grown from complexes of GA 2 purified by gel filtration chromatography. C) A cartoon representation of the structure of Gl-G3/actin in GA 2 , with a Tb 3 + anomalous electron density map superimposed, as in (B). In this case, the terbium ion-soaked crystals from (B) were backwashed with mother liquor containing 10 mM CaCl 2 in an attempt to displace terbium ions. 3.5 Luminescence Results of T b 3 + titration of gelsolin The luminescence of terbium could be used to probe the calcium binding sites on gelsolin (Tellam, 1991). In the presence of gelsolin, the terbium emission is remarkably enhanced due to the energy transfer from aromatic amino acids (particularly, tryptophan) to neighboring protein-bound terbium ions. The terbium concentration dependence of the luminescence intensity obtained by indirect exitation of terbium ions indicated the presence of two types of terbium binding site in gelsolin (Tellam, 1991). 79 From the structure of inactive gelsolin containing bound A T P (Urosev et al., unpublished data), A T P binds to gelsolin domains G3, G4 and G5, that is, it spans the two halves of gelsolin. Activation of gelsolin would disperse the cluster of amino acid residues that bind A T P and effectively destroy the binding site. This explains the experimental observation of a drop by more than ten-fold in the affinity of activated gelsolin for A T P relative to inactive gelsolin (Gremm and Wegner, 1999). However, the same authors also reported that inactive gelsolin with A T P bound had a tenfold lower affinity for C a 2 + than when it is devoid of A T P . Therefore, we decided to investigate whether the binding of T b 3 + to gelsolin is affected by the presence of A T P . In the structure of inactive gelsolin (Burtnick et al., 1997), a tryptophan residue lies approximately 5 A from the type II metal ion binding site in G5. We assume, therefore, energy transfer between this tryptophan and a T b 3 + that enters the type II site on G5 wil l be a major contributor to the overall luminescence intensity observed in the titration experiments. We firstly titrated terbium with E D T A by adding 100 m M E D T A stock solution to a 3.00 ml sample of 50 m M M O P S / K O H , 2.5 m M T b C l 3 , p H 7.1 (Fig. 53). The terbium luminescence rose steadily until the concentration of E D T A matched that of T b 3 + , then reached a plateau. The data demonstrate that E D T A can chelate terbium ions at a mole ratio of 1:1 and that energy transfer does occur from E D T A to terbium. This is important to remember in interpreting subsequent experiments. 80 0-\ , 1 1 1 1 1 . 1 1 1 1 1 1 1 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 [EDTA] (mM) Fig. 53. EDTA titration of Tb 3 + . TbCl 3 (2.5 mM) was titrated with increasing concentration of EDTA and the Tb 3 + ion luminescence intensity (at 546 nm), obtained by excitation at 296 nm, was recorded for each concentration. Gelsolin titrations by terbium were performed by adding 134 m M TbCl3 stock solution to 1.2 ml samples of 1.12 p M gelsolin with 0.5 m M A T P and without A T P , both in 15 m M M O P S / K O H , 0.2 m M E D T A , p H 7.1 (Fig. 54). On adding T b 3 + , the luminescence intensities of all samples increase sharply until the concentration of terbium ions matches that of the E D T A . In samples containing gelsolin without A T P , there is a further sharp increase of luminescence on adding more T b 3 + , in agreement with proposed energy transfer from tryptophan to gelsolin-bound terbium. But why the intensity then falls off is not clear. It may be the result of quenching that is introduced by structural changes in gelsolin on activation. Addition of T b 3 + to samples of gelsolin with A T P present, once E D T A had been saturated with T b 3 + , led to formation of a visible precipitate. Precipitate also appeared at this stage in titration of a protein-free control solution containing A T P . Clearly, formation 81 of a T b - A T P precipitate under our experimental conditions renders interpretation of observed luminescence trends extremely difficult. 22 20. & 18 'w | 16 c a 14 c a> § 1 2 I 10 1 8 > I 6H •— ATP but no gelsolin s no ATP or gelsolin A— ATP and gelsoin ••— gelsolin but no ATP 1 1— 0 0.2 0.4 —I— 0.6 0.8 1.0 —I— 1.2 —I— 1.4 [Tb3+](mM) Fig. 54. The Tb 3 + concentration dependence of the luminescence intensity obtained by indirect excitation of Tb 3 + . Gelsolin (1.12 uM) in 15 mM MOPS/KOH, 0.2 mM EDTA, pH 7.1, without ATP and with 0.5 mM ATP was titrated with TbCl3 and the Tb 3 + ion luminescence intensity at 546 nm was determined on excitation at 296 nm. (preparation 1) In order to confirm that our results above are not due to gelsolin quality, the experiments were repeated by adding stock 40 m M T b C b to a 1.2 ml sample of 0.76 p M gelsolin from a different preparation, with 0.5 m M A T P and without A T P , both in 20 m M M O P S / K O H , 0.1 m M E D T A , lOOmM N a C l , p H 7.1 (Fig. 55). Qualitatively similar results were obtained, except that luminescence reached a plateau after saturation of both E D T A and gelsolin, rather than peaking and dropping to a lower level as seen in the previous set of experiments. Unfortunately, the formation of a T b - A T P precipitate prevented us for learning more about the interrelation between metal ion and nucleotide affinities of gelsolin. 82 —•— ATP but no gelsolin no ATP or gelsolin ATP and gelsolin —T— gelsolin but no ATP • 0 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 [Tb ](mM) Fig. 55. The Tb 3 + concentration dependence of the luminescence intensity obtained by indirect excitation of Tb 3 +. Gelsolin (0.76 uM) in 20 mM MOPS/KOH, 0.1 mM EDTA, 100 mM NaCl, pH 7.1, without ATP and with 0.5 mM ATP, was titrated with TbCl 3 and the Tb 3 + ion luminescence intensity at 546 nm was determined on excitation at 296 nm. (preparation 2) 3.6 Conclusions and Future Work Successful purification o f gelsolin in good yield required use o f a serine protease inhibitor (PMSF) during the procedure. If omitted, a cleaved gelsolin with a molar mass of 68 kDa was obtained. What was worse, the cleaved gelsolin could not be eluted by A T P from an Aff i -Gel blue column, but was eluted by N a C l , together with other contaminants. Then, further purification was required. Interestingly, the cleaved gelsolin, like intact gelsolin, could bind to an actin monomer to form a complex that could be purified by gel filtration. Unfortunately, this complex could not be crystallized when subjected to conditions similar to those that had yielded crystals of G A 2 and GA3. 83 Based on our gel filtration, M A L D I mass spectrometry, and gel electrophoresis results, we did obtain G A , G A 2 and GA3 in crystalline form. However, X-ray diffraction results with these crystals have not yielded complete structures for any of the complexes, but rather have produced only images of the Gl-G3/ac t in portions of G A 2 and GA3, respectively. These are essentially identical to the Gl-G3/ac t in structure reported previously from this laboratory. Clearly, further manipulation of the crystals, or of the complexes prior to crystallization, is required i f new information is to be obtained. Soaking crystals of gelsolin-actin complexes in solutions containing terbium ions led to T b 3 + replacing C a 2 + in these structures, as well as fi l l ing the previously vacant type II C a 2 + site in G2. The latter result confirms that the G2 type II site is a metal ion-binding site. It is l ikely that the higher charge density o f T b 3 + relative to C a 2 + explains why the site became occupied. Furthermore, backwashing with calcium salt-containing buffers successfully restored C a to the sites it had previously occupied, but did not displace the T b 3 + bound at the type II site in G2. The terbium ion-binding data also indicate, contrary to predictions based on solution studies with calcium ions, that the type II C a 2 + in G l is exchangeable in gelsolin crystals. In an attempt to investigate whether the binding of T b 3 + to gelsolin is affected by the presence of A T P , we measured the luminescence of terbium added to solutions of gelsolin in the presence and absence of A T P . However, formation of a Tb-ATP precipitate under our experimental conditions renders interpretation of the observed luminescence trends extremely difficult. 84 From my laboratory experiences during the course of this research project, I learned to purify gelsolin from plasma and to extract actin from muscle. This required gaining experience with ion exchange, gel fdtration and affinity chromatography techniques. From purified actin and gelsolin, I was able to form and then to purify and characterize different gelsolin:actin complexes by S D S - P A G E , fluorescence spectroscopy, U V - V i s spectroscopy and mass spectrometry ( M A L D I ) . I was able to set up crystals trays using both hanging drop and sitting drop vapour diffusion methods, and to crystallize gelsolin:actin complexes by screening crystallization conditions. I learned to trap, wash and soak protein crystals in different cryoprotectant solutions, while observing them through a binocular dissecting microscope, then freeze them in liquid nitrogen for storage and subsequent X-ray diffraction analysis. I learned to use cross linking regents to crosslink our gelsolimactin complexes in order to improve the stability and quality of our crystals. Since completion of the work described in this thesis, I also have been able to observe and participate in collection and analysis of protein crystal diffraction data in the new protein crystallography facilities at the U B C Centre for Blood Research. If this project continues, I expect to learn to independently process and refine data using computational tools now available to the Burtnick laboratory. Future work should center on optimizing G A , G A 2 and GA3 crystals for diffraction analysis. This w i l l involve screening different cryoprotectant solutions and testing the effects of chemical crosslinking reagents. In parallel with this goal, it would be useful to express and purify recombinant gelsolin fragments (e.g., G2-G3, G3-G4, etc), crystallize these fragments directly in the presence and absence of calcium ions, and investigate the interaction between these fragments and actin. 85 Additionally, with many details of the activation of gelsolin by calcium now worked out, it would be useful to learn more about the deactivation process. It would be interesting to soak gelsolin-containing crystals in solutions that contain regulators known to inhibit or reverse the interaction between gelsolin and actin, e.g., certain polyphosphoinositides (PPIs) and lysophosphatidic acid ( L P A ) . These structures wi l l contribute to an overall understanding of the mechanism by which gelsolin activity on actin is regulated. 86 Bibliography Aguda, A . H . , Burtnick, L . D . , and Robinson, R. C. (2005). The state of the fdament. E M B O Reports 6, 220-226. Bryan, J. (1988). Gelsolin has 3 actin-binding sites. J Cel l B i o l 106, 1553-1562. Bryan, J., and Kurth, M . C. (1984). Actin-gelsolin interactions - evidence for 2 actin-binding sites. J B i o l Chem 259, 7480-7487. Burtnick, L . D . , Koepf, E . K . , Grimes, J., Jones, E . Y . , Stuart, D . I., McLaughl in , P. J., and Robinson, R. C. (1997). The crystal structure of plasma gelsolin: Implications for actin severing, capping, and nucleation. Cel l 90, 661-670. Burtnick, L . D . , Urosev, D . , Irobi, E . , Narayan, K . , and Robinson, R. C. (2004). Structure of the N-terminal half of gelsolin bound to actin: Roles in severing, apoptosis and F A F . E M B O J 23, 2713-2722. Chik, J. K . , Lindberg, U . , and Schutt, C . E . (1996). The structure of an open state of beta-actin at 2.65 angstrom resolution. J M o l B i o l 263, 607-623. Choe, H . , Burtnick, L . D . , Mejillano, M . , Y i n , H . L . , Robinson, R. C , and Choe, S. (2002). The calcium activation of gelsolin: Insights from the 3 angstrom structure of the G4-G6/actin complex. J M o l B i o l 324, 691-702. Coue, M . , and Korn , E . D . (1985). Interaction of plasma gelsolin with G-actin and F-actin in the presence and absence of calcium-ions. J B i o l Chem 260, 5033-5041. 94-Ditsch, A . , and Wegner, A . (1995). Two low-affinity Ca -binding sites of gelsolin that regulate association with actin. Eur J Biochem 229, 512-516. Doi , Y . , Higashida, M . , and Kido , S. (1987). Plasma-gelsolin-binding sites on the actin sequence. Eur J Biochem 164, 89-94. Dominguez, R. (2004). Actin-binding proteins - a unifying hypothesis. Trends Biochem Sci 29, 572-578. 87 Dos Remedios, C. G . , Chhabra, D. , Kekic , M . , Dedova, I. V . , Tsubakihara, M . , Berry, D . A . , and Nosworthy, N . J. (2003). Act in binding proteins: Regulation of cytoskeletal microfilaments. Physiol Rev 83, 433-473. Dos Remedios, C. G . , and Moens, P. D. J. (1995). Act in and the actomyosin interface - a review. Biochimica Et Biophysica Acta-Bioenergetics 1228, 99-124. Edgar, A . J. (1990). Gel-electrophoresis of native gelsolin and gelsolin actin complexes. J Muscle Res Cel l Mo t i l 11, 323-330. Egelman, E . H . (1985). The structure of F-actin. J Muscle Res Cel l Mo t i l 6, 129-151. Gremm, D. , and Wegner, A . (1999). Co-operative binding of C a 2 + ions to the regulatory binding sites of gelsolin. Eur J Biochem 262, 330-334. Hertzog, M . , van Heijenoort, C , Didry, D. , Gaudier, M . , Coutant, J., Gigant, B . , Didelot, G. , Preat, T., Knossow, M . , Guittet, E . , and Carlier, M . F. (2004). The beta-thymosin/WH2 domain: Structural basis for the switch from inhibition to promotion of actin assembly. Cel l 117, 611-623. Holmes, K . C , Popp, D . , Gebhard, W., and Kabsch, W . (1990). Atomic model of the actin filament. Nature 347, 44-49. Huff, M . E . , Page, L . J., Balch, W. E. , and Kel ly , J. W. (2003). Gelsolin domain 2 C a 2 + affinity determines susceptibility to furin proteolysis and familial amyloidosis of finnish type. J M o l B i o l 334, 119-127. Irobi, E . , Aguda, A . H . , Larsson, M . , Guerin, C , Y i n , H . L . , Burtnick, L . D . , Blanchoin, L . , and Robinson, R. C. (2004). Structural basis of actin sequestration by thymosin-beta 4: Implications for W H 2 proteins. E M B O J 23, 3599-3608. Irobi, E. , Burtnick, L . D . , Urosev, D. , Narayan, K . , and Robinson, R. C. (2003). From the first to the second domain of gelsolin: A common path on the surface of actin? F E B S Lett 552, 86-90. 88 Janmey, P. A . , Chaponnier, C , L ind , S. E . , Zaner, K . S., Stossel, T. P., and Y i n , H . L . (1985). Interactions o f gelsolin and gelsolin actin complexes with actin - effects of calcium on actin nucleation, fdament severing, and end blocking. Biochemistry 24, 3714-3723. Kabsch, W. , Mannherz, H . G . , Suck, D. , Pai, E . F. , and Holmes, K . C. (1990). Atomic-structure of the actin - DNase-I complex. Nature 347, 37-44. Kazmirski , S. L . , Isaacson, R. L . , A n , C. , Buckle, A . , Johnson, C. M . , Daggett, V . , and Fersht, A . R. (2002). Loss of a metal-binding site in gelsolin leads to familial amyloidosis-finnish type. Nat Struct B i o l 9, 112-116. Khaitlina, S., Walloscheck, M . , and Hinssen, H . (2004). Calcium-induced conformational changes in the C-terminal half of gelsolin stabilize its interaction with the actin monomer. Biochemistry (Mosc) 43, 12838-12845. Kiselar, J. G . , Janmey, P. A . , A l m o , S. C , and Chance, M . R. (2003). Visualizing the Ca 2 +-dependent activation of gelsolin by using synchrotron footprinting. Proc Natl Acad Sci U S A 100, 3942-3947. Kolappan, S. L . , Gooch, J. T., Weeds, A . G. , and McLaughl in , P. J. (2003). Gelsolin domains 4-6 in active, actin-free conformation identifies sites of regulatory calcium ions. J M o l B i o l 329, 85-92. Kurth, M . C , Wang, L . L . , Dingus, J., and Bryan, J. (1983). Purification and characterization of a gelsolin-actin complex from human-platelets - evidence for C a 2 + -insensitive functions. J B i o l Chem 258, 895-903. L i n , K . M . , Mejillano, M . , and Y i n , H . L . (2000). C a 2 + regulation of gelsolin by its C-terminal tail. J B i o l Chem 275, 27746-27752. Mann, M . , Hendrickson, R. C , and Pandey, A . (2001). Analysis of proteins and proteomes by mass spectrometry. Annu Rev Biochem 70, A31A12>. 89 McGough, A . M . , Staiger, C. J., M i n , J. K . , and Simonetti, K . D . (2003). The gelsolin family of actin regulatory proteins: Modular structures, versatile functions. F E B S Lett 552, 75-81. McLaughlin, P. J., Gooch, J. T., Mannherz, H . G. , and Weeds, A . G . (1993). Structure of gelsolin segment-1-actin complex and the mechanism of filament severing. Nature 364, 685-692. Narayan, K . , Chumnarnsilpa, S., Choe, H . , Irobi, E . , Urosev, D . , Lindberg, U . , Schutt, C . E . , Burtnick, L . D . , and Robinson, R. C. (2003). Activation in isolation: Exposure of the actin-binding site in the C-terminal half of gelsolin does not require actin. F E B S Lett 552, 82-85. Otterbein, L . R., Cosio, C , Graceffa, P., and Dominguez, R. (2002). Crystal structures of the vitamin D-binding protein and its complex with actin: Structural basis of the actin-scavenger system. Proc Natl Acad Sci U S A 99, 8003-8008. Otterbein, L . R., Graceffa, P., and Dominguez, R. (2001). The crystal structure of uncomplexed actin in the A D P state. Science 293, 708-711. Pollard, T. D. , Blanchoin, L . , and Mull ins, R. D . (2000). Molecular mechanisms controlling actin filament dynamics in nonmuscle cells. Annu Rev Biophys Biomol Struct 29, 545-576. Pope, B . , Maciver, S., and Weeds, A . (1995). Localization of the calcium-sensitive actin monomer binding-site in gelsolin to segment-4 and identification of calcium-binding sites. Biochemistry 34, 1583-1588. Pope, B . , Way, M . , and Weeds, A . G . (1991). Two of the three actin-binding domains of gelsolin bind to the same subdomain of actin - implications for capping and severing mechanisms. F E B S Lett 250, 70-74. Popp, D. , Lednev, V . V . , and Jahn, W. (1987). Methods of preparing well-orientated sols of F-actin containing filaments suitable for X-ray-diffraction. J M o l B i o l 197, 679-684. 90 Righetti, P. G . (2004). Bioanalysis: Its past, present, and some future. Electrophoresis 25, 2111-2127. Robinson, R. C , Mejillano, M . , Le, V . P., Burtnick, L . D . , Y i n , H . L . , and Choe, S. (1999). Domain movement in gelsolin: A calcium-activated switch. Science 286, 1939-1942. Schoepper, B . , and Wegner, A . (1991). Rate constants and equilibrium-constants for binding of actin to the 1-1 gelsolin-actin complex. Eur J Biochem 202, 1127-1131. Schutt, C . E . , Mys l ik , J. C , Rozycki , M . D. , Goonesekere, N . C . W. , and Lindberg, U . (1993). The structure of crystalline profilin beta-actin. Nature 365, 810-816. Selve, N . , and Wegner, A . (1986). Rate constants and equilibrium-constants for binding of the gelsolin-actin complex to the barbed ends of actin-filaments in the presence and absence of calcium. Eur J Biochem 160, 379-387. Selve, N . , and Wegner, A . (1987). PH-dependent rate of formation of the gelsolin-actin complex from gelsolin and monomeric actin. Eur J Biochem 168, 111-115. Sheterline, P., Clayton, J., and Sparrow, J. C. (1998). Act in , 4th edn (New York: Oxford University Press). Silacci, P., Mazzolai , L . , Gauci, C , Stergiopulos, N . , Y i n , H . L . , and Hayoz, D. (2004). Gelsolin superfamily proteins: Key regulators of cellular functions. Ce l l M o l Life Sci 61, 2614-2623. Silva, B . E . R., and Burtnick, L . D . (1990). Characterization of horse plasma gelsolin. Biochem Cell Biol-Biochimie Et Biologie Cellulaire 68, 796-800. Spudich, J. A . , and Watt, S. (1971). Regulation of rabbit skeletal muscle contraction. 1. Biochemical studies of interaction of tropomyosin-troponin complex with actin and proteolytic fragments of myosin. J B i o l Chem 246, 4866-4871. 91 Steinmetz, M . O., Stoffier, D . , Hoenger, A . , Bremer, A . , and Aebi , U . (1997). Actin: From cell biology to atomic detail. J Struct B i o l 119, 295. Tellam, R. L . (1991). The binding of terbium ions to gelsolin reveals 2 classes of metal-ion binding-sites. Arch Biochem Biophys 288, 185-191. Urosev, D. , M a , Q., Tan, A . L . C , Robinson, R. C , and Burtnick, L . D . (unpublished data). The structure of gelsolin bound to A T P . unpublished data. Way, M . , Pope, B . , Gooch, J., Hawkins, M . , and Weeds, A . G . (1990). Identification of a region in segment-1 of gelsolin critical for actin binding. E M B O J 9, 4103-4109. Weeds, A . G . , Gooch, J., McLaughl in , P., Pope, B . , Bengtsdotter, M . , and Karlsson, R. (1995). Identification of the trapped calcium in the gelsolin segment 1-actin complex -implications for the role of calcium in the control of gelsolin activity. F E B S Lett 360, 227-230. Yamamoto, H . , Terabayashi, M . , Egawa, T., Hayashi, E . , Nakamura, H . , and Kishimoto, S. (1989). Affinity separation of human-plasma gelsolin on affi-gel blue. J Biochem (Tokyo) 105, 799-802. Y i n , H . L . (1987). Gelsolin - calcium-regulated and polyphosphoinositide-regulated actin-modulating protein. Bioessays 7, 176-179. Y i n , H . L . , Iida, K . , and Janmey, P. A . (1988). Identification of a polyphosphoinositide-modulated domain in gelsolin which binds to the sides of actin-filaments. J Cel l B i o l 106, 805-812. Zapun, A . , Grammatyka, S., Deral, G . , and Vernet, T. (2000). Calcium-dependent conformational stability of modules 1 and 2 of human gelsolin. Biochem J 350, 873-881. 92 

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.831.1-0061124/manifest

Comment

Related Items