Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Characterization of the 16 kDalton heat shock proteins of Caenorhabditis elegans Hockertz, Michael Karl 1991

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

Item Metadata


831-UBC_1991_A6_7 H62.pdf [ 6.11MB ]
JSON: 831-1.0098485.json
JSON-LD: 831-1.0098485-ld.json
RDF/XML (Pretty): 831-1.0098485-rdf.xml
RDF/JSON: 831-1.0098485-rdf.json
Turtle: 831-1.0098485-turtle.txt
N-Triples: 831-1.0098485-rdf-ntriples.txt
Original Record: 831-1.0098485-source.json
Full Text

Full Text

Characterization of the 16 kDalton Heat Shock Proteins of Caenorhahditis elegans by Michael Karl Hockertz B.Sc.(Honours), The University of British Columbia, 1986 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Biochemistry We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA My, 1991 (c) Michael K. Hockertz, 1991 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of \ P \^>AKUUX\^W y The University of British Columbia Vancouver, Canada Date DE-6 (2/88) Abstract ABSTRACT The 16 kDa heat shock proteins of Q. elegans have been partially purified and characterized using antipeptide antibodies and standard chromatographic techniques. Based upon the sequences of the previously isolated hspl6 genes, peptides were synthesized against which polyclonal antibodies were produced in rabbits. The antibodies were used to detect fractionated hspl6 by the western blot technique and to purify hspl6 by affinity chromatography. The specificity of the anti-peptide antibodies indicated that the four hspl6 genes may encoded both 16 kDa and the 18 kDa polypeptides. The latter were previously thought to be derived from distinct genes. Two-dimensional gel electrophoresis showed that the 16 kDa hsps exist in multiple isoforms, probably due to their posttranslational modification. Hspl6 was purified by affinity chromatography with immobilized antibody to a C-terminal 36 residue peptide of hspl6-2. The hspl6s did not bind to the antibody in a low salt buffer, but did so in the presence 4M urea. Hspl6 was fractionated by hydroxylapatite and gel exclusion chromatography and shown to exist as a large reasonably uniform complex of 460±55 kDa. ii Table of Contents TABLE OF CONTENTS Abstract -ii Table of Contents .' iii List of Figures vi List of Abreviations vii Acknowledgements x I. Introduction 1 A. The Heat Shock Response 1 1. The Heat Shock Genes 2 2. Heat Shock Induced Expression of the Heat Shock Genes 4 a. Transcriptional Control of Heat-Induced Heat Shock Gene Expression 4 b. Translational Control of Heat Shock Gene Expression 5 3. Non-Heat Shock Induction of the Heat Shock Genes 6 4. Developmental Expression of the Hsps 7 a. Hormonal Induction of the Heat Shock Proteins 8 B. Functional Aspects of the Heat Shock Proteins 8 1. Hsp70 8 a. Hsp70 in the Non-Stressed Cell 8 b. Hsp70 in the Heat-Shocked Cell 10 c. Hsp70 and Intrinsic Therrnoresistance 11 d. Hsp70 and Acquired Thermotolerance 12 2. Hsp90 13 3. Hspl04/110 14 4. The Small Heat Shock Proteins 14 a. Homology to cc-Crystallin 14 b. The Small Hsps as Large Particles 16 c. Intracellular Localization of the Small Heat Shock Proteins 18 d. The Small Heat Shock Proteins and Thermotolerance 19 e. The Role of Small Hsp Phosphorylation in Thermotolerance 20 £ The Role of Calcium in Small Hsp Phosphorylation and Thermotolerance 22 C. The Nematode Caenorhabditis elegans '. 23 D. The Small Hsps of C. elegans 24 E . The Present Study 26 II. Materials and Methods 28 A. Growth of Caenorhabditis eleeans 28 1. Collection of Gravid Adult Nematodes 29 2. Preparation of Nematode Embryos 29 B. Heat Shock Conditions 29 iii Table of Contents C. 35S-Labeling of £ , ej£gMS_ 3 0 1. Labeling of E. coli 30 2. Labeling of C. elegans . 30 D. Homogenization 31 1. Sonication 31 2. French Press 31 E. SDS-poly acrylamide Gel Electrophoresis . 31 1. lD-Electrophoresis 31 2. 2D-Electrophoresis 32 F. Detection of protein on SDS gels 32 1. Coomassie Blue Stain • 32 2. Autoradiography 33 G. Production of anti-hspl6 antiserum 33 1. Synthesis of Peptides 33 2. Immunization of Rabbits 33 3. Ammonium Sulfate Precipitation of Sera 34 4. Affinity Purification 34 a. Peptide Column Preparation 34 b. Serum Purification 35 H. Western Blot Analysis of Proteins 35 1. Blotting Procedure 35 2. Immunostaining of Blots 36 I. Column Chromatography 36 1. Hydroxylapatite Chromatography . 36 2. Gel Filtration Chromatography 37 3. Affinity Chromatography 37 a. Synthesis of Antibody-Sepharose 37 b. Binding and Elution of Antigen 38 J. Protein Quantitation 39 IH. Results and Discussion 40 A. Production of Antibodies Specific for hspl6 40 B. Specificity of anti-hspl6 Antibodies 40 C. Heat Shock Induction of C. elegans 42 . 1. Optimal Time for Heat Induction 42 2. Stability of Hspl6 Isolated from Adult vs. Embryonic Nematodes 45 3. Induction of Hspl6 When Periods of Heat Shock are Followed by Periods of Recovery 47 D. Analysis of Control and Heat Shock Induced Nematode Extracts by Two-Dimensional Gel Electrophoresis 49 E. Immunoaffinity Chromatography of Hsp 16 53 F. Protein Sequencing of the 16 and 18 kDa Heat Shock Polypeptides 54 G. Adjusting the Conditions of the Heat Shock Protocol 56 H. Affinity Purification of Hsp 16 57 I. Affinity Purification of Hsp 16 in the Presence of Urea 60 J. Analysis of the Hsp 16 Complex by Hydroxylapatite and Gel Exclusion Chromatography 62 iv Table of Contents 1. Purification by Hydroxylapatite Fractionation 62 2. Purification and Analysis by Gel Exclusion Chromatography 65 3. Estimation of the Size of the Hspl6 Complex 69 IV. References 74 v List of Figures LIST OF FIGURES Fig. 1. The hsp!6 loci of C. elegans 25 Fig. 2 Western blot analysis of 35S-labeled control and heat shocked nematode extracts 41 Fig. 3 Amino acid sequence of the hsp 16 genes in the region to which synthetic peptides were produced 43 Fig. 4 Heat shock induction of Q. elegans for varying lengths of time 44 Fig. 5 Stability of hspl6 in Embryos vs. Adults with and without protease inhibitors 46 Fig. 6 Induction of hsp 16 by two successive periods of heat shock, each followed by a period of recovery 48 Fig. 7 Two-dimensional gel electrophoretic analysis of extracts from control [Cont] and heat shocked [HS] 35S-labeled nematodes 50 Fig. 8 Affinity chromatography of hsp 16 with anti-2/110-145 antibody 55 Fig. 9 Affinity chromatography of hspl6 with anti-2/110-145 antibody-protein A-Sepharose in the absence and presence of 4 M urea 58 Fig. 10 Hydroxylapatite chromatography of a heat shocked embryo extract 63 Fig. 11 Gel filtration of heat shocked nematode extracts 66 Fig. 12 Calibration curve of the S-300 gel exclusion column 71 vi List of Abreviations LIST OF ABREVIATIONS [x] square brackets refer to a figure lane marker Ab antibody ATP adenosine-5'-triphosphate BCIP/NBT bromochloroindolyl phosphate/nitro blue tetrazolium BiP Binding Protein bp base pair(s) bsa bovine serum albumin Buffer A 20 mM Tris pH 7.4, 20 mM NaCl, 1 mM EDTA, 0.1 mM MgCh, 10% glycerol Buffer B Buffer A with no MgCh Buffer G 10 mM potassium phosphate pH 7.4, 20 mM NaCl, 1 mM MgCl 2, 0.1 mM EDTA, 10% glycerol Buffer G/50 Buffer G except containing 50 mM potassium phosphate Buffer G/200 Buffer G containing 200 mM potassium phosphate C-terminus carboxy-terminus C. elegans Caenorhabditis elegans CEF chicken embryo fibroblasts CNBr cyanogen bromide cpm counts per minute (IH2O distilled water E. coli Escherichia coli EDTA ethylenediamine tetraacetic acid EGTA Ethylene glycol-0,0'-bis(2-aminoethyl)-N,N,N',N'-tetraacetic acid ELIZA Enzyme Linked Immunosorbent Assay »vii List of Abreviations ER Endoplasmic Reticulum GRP78 78 kDa glucose regulated protein HAP hydroxylapatite hsc70 70 kDa heat shock cognate protein HSE heat shock element HSG heat shock granule hsp heat shock protein hsp 16, 70, etc. 16 kDa heat shock protein hsps heat shock proteins HSTF heat shock transcription factor IgG immunoglobulin G kb kilobases kDa kilo Daltons (103) KLH Keyhole limpet hemocyanin LSM low sulfate medium mCi milliCurie MDa Mega Daltons (106) mhsp70 mitochondrial hsp70 M r molecular weight mRNA messenger RNA N-terminus amino-terminus NGM nutrient growth medium PBS 20 mM potassium phosphate pH 7.4, 0.12 M NaCl PMA phorbol-12-myristate-13-acetate List of Abreviations PMSF phenylmethylsulfonylfluoride psi pounds per square inch RNA ribonucleic acid SDS Sodium dodecylsulfate TE 10 mM Tris-HCl pH 7.4, ImM EDTA TNF tumor necrosis factor-a TPA 12-O-tetradecanoylphorbol-13-acetate Tris tris (hydroxymethyl) aminomethane Vc/Vo ratio of the elution volume to the column void volume V T total column volume xg times gravity °C degrees Celsius ix Acknowledgements ACKNOWLEDGEMENTS I would first like to thank my supervisor, Dr. Peter Candido, for his assistance and guidance throughout my years of study here. Don Jones was of particularly great help during my project as were all as were all my friends and colleagues who shared their ideas, time, and many protocols. My deepest thankyou goes to my family whose love and support have never faultered and to whom I will always be grateful. x Introduction I. INTRODUCTION Elevation of the temperature of a cell to a few degrees above normal results in the induction of a set of genes known as the heat shock genes (for reviews: Lindquist, 1986; Schlesinger el &L, 1982). In addition to being induced by heat shock, these genes are also induced by other treatments, eg. heavy metals such as cadmium, metabolic inhibitors such as arsenite and dinitrophenol, and by certain amino acid analogues. This induction results in the synthesis of a set of proteins, the heat shock proteins (hsps), which fall into three main categories according to size and sequence homologies: hsp90, the hsp70 related proteins, and the small heat shock proteins which range in size from approximately 15 to 35 kDa. Originally observed in Drosophila melanogaster. the heat shock response has since been recognized as universal, homologous systems being found in such diverse groups as bacteria, plants, invertebrates, and mammals. Expression of the hsps is associated with the phenomenon of thermotolerance whereby a prior mild heat shock will protect cells from a subsequent otherwise lethal heat shock. The development of thermotolerance closely parallels the appearance and disappearance of the hsps. Together, the expression of the hsps is thought to be responsible for protecting a cell from stress and aiding its subsequent recovery. A. THE HEAT SHOCK RESPONSE The heat shock response was first observed in the giant polytene chromosomes of Drosophila melanogaster. Upon heat shock at 37 °C several new puffs appeared, indicating heat shock induced transcriptional activity. At the same time there was a marked reduction in the number of puffs normally present at 25 °C (Ritossa, 1962). This suggested that the transcriptional activity represented by the latter puffs was somehow repressed by the heat shock. The heat shock puffs were later observed to be sites of RNA synthesis by 3H-uridine 1 Introduction labeling of heat shocked salivary glands (Tissieres ej aj\> 1974). Heat shock proteins were later found to be synthesized not only in the salivary glands from which polytene chromosomes are obtained but also in other tissues such as brain, malpighian tubules, and wing imaginal disks. Upon more detailed analysis, proteins of 84, 70, 68, 27, 26, 23, and 22 kDa were shown to be induced (Mirault el al-, 1978). Two-dimensional gel electrophoresis indicated the existence of five 70 kDa spots, one of which was present at 25 °C and not particularly inducible by heat shock. Analysis of tryptic cleavage products of the heat shock induced polypeptides indicated that the 70 kDa components were closely related and that the 26 and 27 kDa proteins were related to each other while the rest were distinct proteins. Until 1978 the heat shock response was recognized only in Drosophila. The phenomenon was subsequently observed in many different tissues and organisms, eg. chick embryo fibroblasts (Kelly and Schlesinger, 1978), yeast (Miller et al., 19791. E. coli (Yamamori eial-, 1978), and man (Slater eial., 1981). Moreover, it was shown that antibodies to the chick embryo fibroblast 70 kDa heat shock protein (hsp70) cross-reacted with 70 kDa proteins of Drosophila. Xenopus. yeast, mouse, and man (Kelly and Schlesinger, 1982). These findings suggested that the heat shock response was both universal and highly conserved through evolution. 1. The Heat Shock Genes Two-dimensional electrophoresis of Drosophila vheat shock proteins indicates the existence of several hsp70 species. Accordingly, five genes encoding hsp70 have been found at two different loci in Drosophila (Ingolia gt al., 1980; Moran el al., 1983). Similar hsp70 multigene families are also found in other organisms such as mouse and yeast (Ingolia ej al., 1982). The Drosophila hsp70 is 70% identical to human hsp70 (Hunt and Morimoto, 1985), 2 Introduction and 48% identical to the E.coli dnak protein (Bardwell and Craig, 1984). This high degree of conservation through evolution indicates the importance of hsp70 in the cell. In Drosophila. three genes coding for proteins closely related (75 to 80% identical) to hsp70 are expressed at normal temperatures. These are referred to as the heat shock cognate proteins (hsc70) (Palter el al., 1986), and are also found in other systems such as mouse where some of the members of the hsp70 family are heat inducible and others are constitutively expressed (Lowe and Moran, 1984). In E. coli the dnak gene is constitutively expressed at normal temperatures and is further induced upon heat shock (Bardwell and Craig, 1984). Hsp70 is therefore probably necessary not only in stressed cells, but also for functions in normally growing cells. Another highly conserved family of heat shock proteins is the 83 to 90 kDa class . The Drosophila and yeast proteins share 60% identity at the amino acid level (Hackett and Lis, 1983; Farrelly and Finkelstein, 1984). This protein is highly expressed in most cells at normal temperatures as well as being further heat inducible. The constitutive and inducible hsp90s may be encoded by different genes in both yeast and mouse (Lindquist, 1986; Barnier et al., 1987). The small hsps, ranging in size from approximately 15 to 35 kDa, are the most divergent and heterogeneous of the hsp families. All organisms so far examined contain at least one gene encoding such a protein, eg. hsp26 of yeast (Bossier el al., 1989), and as many as twenty, in the case of soybean (Schoffl and Key, 1982). Drosophila has seven genes coding for small heat shock proteins. Four of these, hsps 22, 23, 26 and 27, were initially characterized at locus 67B and were found to be approximately 50% identical to each other along their entire lengths and up to 71% identical within a region showing homology to bovine a-crystallin (Ingolia and Craig, 1982; Southgate ei al., 1983). The other three small hsps were initially recognized by their non-heat induced developmental expression (Sirotkin and 3 Introduction Davidson, 1982). These genes were later shown to also be heat inducible, although not to the same extent as the other four small hsps (Ayme and Tissieres, 1985). The hsp27 genes of humans (Hickey ei al., 1986) and the hspl6 genes of Caenorhabditis elegans (Russnak and Candido, 1985; Jones ei al., 1986) also share regions of homology with the Drosophila small hsps and mammalian a-crystallin. A unique feature of the C. elegans small hsps is the existence of an intron in their protein coding sequences. Only the hsp83 gene of Drosophila has otherwise been found to contain an intron (Hackett and Lis, 1983). Furthermore, the C. elegans hsp 16 gene introns occur in a position exactly analogous to that of the first intron of the a-crystallin genes. 2. Heat Shock Induced Expression of the Heat Shock Genes a. Transcriptional Control of Heat-Induced Heat Shock Gene Expression Heat shock genes which are transfected into heterologous systems generally retain the integrity of their transcriptional regulation, although the induction temperature shifts to that of the host organism (for review: Nover, 1990). This was first shown with the introduction of the Drosophila hsp70 gene into mouse cultured cells (Corces et al., 1981). This indicated that heat shock regulation was highly conserved and that factors from the mouse probably interacted with the Drosophila gene. Pelham defined the sequences responsible for hsp70 induction in monkey COS cells (Pelham, 1982) using a transient transfection assay. He found a 15 base pair consensus sequence, CTNGAANNTTCNAGA, which was conserved in the 5' non-coding regions of other heat shock genes and was solely responsible for the heat shock induction. The importance of this heat shock element (HSE) was confirmed when a synthetic HSE was used to drive the transcription of a normally non-heat inducible gene (Pelham and Bienz, 1982). 4 Introduction DNA footprinting techniques have shown that a protein, the 'Heat Shock Transcription Factor' (HSTF), binds specifically to the HSEs and can activate heat shock genes in vitro (Wu, 1984). This factor has recently been purified from yeast, Drosophila. and man (Sorger and Pelham, 1987; Wu ei al., 1987; Goldenberg el al., 1988), and cloned from yeast, Drosophila. and tomato (Sorger and Pelham, 1988; Clos eial., 1990; Scharf el al., 1990). HSTF is rather divergent in structure compared to the conservation of the hsps and HSE, varying in size from 150 kDa (yeast) to 83 kDa (human). The proteins share short highly conserved regions, one of them being the DNA binding domain. The Drosophila HSTF has been shown to self-associate into a hexameric complex and to bind cooperatively to multiple HSEs (Clos eial-, 1990). Yeast HSTF becomes highly phosphorylated in proportion to the severity of the heat shock, and this is coincident with an increase in its HSE binding activity (Sorger and Pelham, 1988). Drosophila HSTF is capable of activation in the absence of protein synthesis, suggesting that post-translational modification may be involved in its activation (Zimarino and Wu, 1987). However, the recent demonstration that purified HSTF can be activated in vitro by urea, low pH and interaction with antibodies suggests that activation may involve a conformational change (Mosser, et al-, 1990). b. Translational Control of Heat Shock Gene Expression Heat shock induces the synthesis of hsps but at the same time quickly represses the synthesis of normal proteins. This is due not to degradation of the normal mRNAs but rather to temporary inhibition of their translation (Storti eial., 1980). The preferential translation of Drosophila hsp70 messages was shown to require sequences in the untranslated leader (McGarry and Lindquist, 1985). Hsp70 messages with this leader deleted were not translated at elevated temperatures, ie. reacted to heat shock as expected for a non heat shock message. The advantage to cells of this repression of normal messages has been 5 Introduction suggested to be the fact that RNA splicing is inhibited by a severe heat shock (Yost and Lindquist, 1986). Hsp83, the only Drosophila hsp gene with an intron, is not expressed at high temperatures, due to a block in its RNA splicing. A mild heat pre-treatment that induces hsps, however, allows hsp83 expression , and therefore its RNA splicing, following a subsequent severe heat shock. This implicates hsp expression in the protection of the splicing mechanism during a heat stress and also suggests why most hsp genes do not contain introns. 3. Non-Heat Shock Induction of the Heat Shock Genes By definition, heat shock proteins are inducible by elevated temperature. However, many of the heat shock proteins can also be induced at normal temperatures under a variety of conditions. For example, arsenite has been shown to induce heat shock proteins in many systems such as Salmo gairdnerii (Kothary and Candido, 1982) and Chinese hamster fibroblasts (Li et al., 1982). In Drosophila embryonic cells, arsenate, cadmium, mercury, and copper ions all induce the full set of hsps. Arsenate induces the small hsps more strongly than hsp70, and arsenite specifically induces the hsps 26 and 27 more strongly than hsps 22, 23, and 70. Furthermore, nickel and zinc ions mildly induce only hsps 22 and 23 (Bournais-Vardiabasis elaL, 1990). In human cell lines, cadmium and arsenite ions have also been shown to induce a protein, heme oxygenase, which is not induced by heat shock (Taketani et al., 1989). These results suggest that the different hsps can be expressed independently of each other and that there exist different regulatory mechanisms modulating the heat shock response. 6 Introduction 4. Developmental Expression of the Hsps Expression of the hsps may also be regulated during various stages of early development. Bensaude and Morange demonstrated that mouse embryonal carcinoma cells and ectoderm from day 8 mouse embryos, which are considered to be close in their developmental potential, exhibited high spontaneous expression of hsps 89, 70, and 59 (Bensaude and Morange, 1983). In fact, hsp70 was found to be the first major zygotic product of gene activity, with high levels of expression occurring by the late 2-cell stage (Bensaude e_t al., 1983). Prior to this hsp70 expression, gene activity from the 1-cell stage was repressed. In Drosophila. hsps 83, 28, and 26 were abundant in embryos until the blastoderm stage (Zimmerman eial., 1983). However, the mRNA encoding these hsps originated in the surrounding nurse cells and was transported into the zygote. None of the hsp messages were actually heat-inducible in the pre-blastoderm cells. Hsps 68 and 70 are uninducible in Xenopus laevis embryos until the late blastula stage (Heikkila, 1985). This coincides with the beginning of general mRNA synthesis, and also with the acquisition of thermotolerance: prior to the late blastula stage, heat shock is lethal. This developmental pattern of hsp expression is also observable in rabbit embryos (Heikkila and Schultz, 1984) and even in yeast (Kurtz eial., 1986). Saccharomyces hsps 84 and 26 were strongly induced during ascopore development, while hsp70 was neither induced nor inducible although hsp70 related proteins were expressed. In Xenopus. the lack of hsp expression in the early embryo is not due to a lack of heat shock transcription factor, since maternally derived HSTF is present in the embryo in a form that can bind HSEs upon heat shock activation (Ovsenek and Heikkila, 1990). The conservation of the developmental expression pattern of the hsps seems to indicate that they play an important role in development and that their function goes beyond the protection of cells from environmental stress. 7 Introduction a. Hormonal Induction of the Heat Shock Proteins In Drosophila cultured cells, the small heat shock proteins hsp22, hsp23, hsp26, and hsp27 are induced by the moulting steroid hormone ecdysterone (Ireland and Berger, 1982). Ecdysterone treatment is specific in that none of the other hsps are induced. Hybrid genes containing the upstream promoter regions of hsps 22 or 23 and the protein coding regions of the thymidine kinase or the p-galactosidase reporter genes respectively were induced by either hear shock or ecdysterone in homologous transformation studies (Morganelli et al., 1985; Lawson etai-. 1985). Deletion studies of the regulatory regions showed that separate regulatory elements were responsible for the heat shock induction (HSE) and for the ecdysterone induction (Cohen and Methelson, 1985). These results suggest that different transacting factors are required for the heat-inducible and hormonal inductions. There is also evidence that other hsps may be induced by hormones. Hsp70 was specifically and transiently induced by insulin in human hepatoma cell lines (Ting ei ai., 1989) and hsp68 was induced in a cell cycle specific manner (Gi phase) by various cyclopentone prostaglandins which block cell cycle progression at the Gi phase (Ohno et al., 1988). B. FUNCTIONAL ASPECTS OF THE HEAT SHOCK PROTEINS 1. Hsp70 a. Hsp70 in the Non-Stressed Cell The functions of hsp70 and related proteins in the cell have been extensively studied and are better understood than those of the other hsps. Recently, members of the hsp70 family of proteins have been implicated in a number of roles in the normal unstressed cell. 8 Introduction The protein responsible for the ATP-dependent uncoating of clathrin from clathrin coated vesicles was shown to be the constitutive form of hsp70, hsc70 (Choppell ej al., 1986), which binds to and dissociates the clathrin light chain a (LCa) (Deluca-Flaherty et al., 1990). Hsc70 was also identified as a protein which was altered in mutant Chinese hamster ovary (CHO) cells resistant to the microtubule inhibitors colchicine and podophyllotoxin (Ahmad et al., 1990). A glucose regulated protein (GRP78) was identified as the same protein which binds immunoglobin heavy chains in the interior of the endoplasmic reticulum (BiP; Binding Protein) of mouse pre-B cells, and is related to hsp70 (Munro and Pelham, 1986). On the basis of this homology, these authors proposed that the function of this non-heat inducible protein was to bind to and aid the folding and assembly of proteins translocated across the ER membrane, ie. those destined to be secreted or integral membrane proteins. Munro and Pelham later showed that an amino acid sequence (KDEL) at the carboxy-terminus of BiP/GRP78 which is not found on cytosolic hsp70 is responsible for its retention in the ER and that this sequence is found on other ER lumenal proteins (Munro and Pelham, 1987). Yeast was also found to contain a protein homologous to BiP/GRP78. It is encoded by a gene, KAR2, previously identified by its ability to complement a mutation blocking nuclear fusion (Rose eial-, 1989; Normington eial-, 1989). The indication was that the mutant BiP/GRP78 caused generalized failure of protein folding in the ER, resulting in pleiotropic effects on cellular metabolism. Hsp70 has also been implicated in the translocation of precursor proteins into mitochondria (for reviews: Neupert et al-, 1990; Haiti et ah, 1990). Depletion of hsp70 in yeast leads to accumulation of both mitochondrial and secretory precursor polypeptides in the cytosol (Deshaies eta!-, 1988). Studies with a hybrid precursor protein demonstrated that it could be imported directly into purified mitochondria if it was already denatured but that this ability was lost with time. Hsp70 as part of a large mass protein fraction (200-250 kDa), but 9 I Introduction not hsp70 alone, was able to maintain this import competence of the precursor (Sheffield et al, 1990). A mitochondrial matrix localized form of hsp70 (mhsp70) was also shown to be required for translocation and folding of precursor proteins into mitochondria (Kang ei al., 1990). In fact, mhsp70 was bound tightly on the matrix side of the mitochondria, to a precursor trapped in the mitochondrial protein import site (Scherer ei al., 1990). In addition, mhsp70 was found to bind co-translationally in a transient and ATP-dependent manner to nascent polypeptides in the normally growing cell (Beckmann et ai., 1990). The hsp70 family of proteins therefore appears to play a role in a variety of normal cellular processes, in each case involving the ATP-dependent folding, unfolding, or maintenance of a partially folded state of various polypeptides. Assembly of proteins in the yeast mitochondrial matrix was also shown to require the constitutive and mildly heat-inducible hsp60. Hsp60, a member of the 'chaperonin' family of proteins which is homologous to groEL in R coli and to proteins in plants and humans, is necessary for the assembly of oligomeric protein complexes (Cheng et al, 1989). The assembly seems to be catalyzed on the surface of hsp60 in an ATP-mediated reaction (Ostermann et al., 1989). b. Hsp70 in the Heat-Shocked Cell As discussed above, some members of the hsp70 family are constitutively expressed at normal temperatures and can be found in the cytosol in association with various cellular components and in organelles such as the mitochondria and the ER. Biochemical and indirect immunofluorescence studies corroborate these findings, indicating that hsp70 is dispersed throughout the cytoplasm and nucleus at normal temperatures. Upon heat shock, the hsp70 is associated with the nucleus and localized in particular to the nucleolus. This marked 10 Introduction redistribution was not observed with exposure to amino acid analogues or sodium arsenite (Vincent and Tanguay, 1982; Welch and Feramisco, 1984). After return to normal temperatures, hsp70 expression decreased and its nuclear and nucleolar staining diminished, returning eventually to its normal state. A study of hsp70 in the nucleus found that it was only weakly associated at normal temperatures. After heat shock, the association with nuclear and nucleolar components became very tight but could be released specifically by low concentrations of ATP (Lewis and Pelham, 1985). These authors suggested a model whereby hsp70 binds to exposed hydrophobic regions of heat-denatured nuclear and nucleolar aggregates and aids their disaggregation and refolding, with concomitant hydrolysis of ATP. The hsp70 homolog in IL coli, dnaK. has been shown to be responsible for renaturation of heat denatured X.-phage repressor protein (Gaitanaris el al., 1990) and the reactivation of heat-inactivated RNA-polymerase (Skowyra etal., 1990). E. coli dnaK mutants are unable to refold these proteins while normal dnaK is able to refold and reactivate these proteins in an ATP-dependent manner. The role of hsp70 in the stressed cell therefore appears related and complementary to its role in the non-stressed cell. The main distinction appears to be that in the normal cell hsp70 acts upon particular substrates while in the stressed cell it additionally acts upon heat-denatured proteins. In either case its general substrate is probably exposed hydrophobic regions of proteins from which it releases itself upon ATP hydrolysis. c. Hsp70 and Intrinsic Thermoresistance Heat shock proteins have been implicated in the protection of cells from thermal stress and mediating their intrinsic heat resistance (thermoresistance). In particular some evidence suggests that hsp70 may be involved. The effects of mutating the two heat inducible hsp70 genes of yeast, YG100 and YG102, were analyzed by Craig and Jacobsen. Cells containing 11 Introduction mutations in either of the genes alone exhibited no phenotype. However, the double mutant exhibited temperature sensitive growth, growing slowly at 30°C and not forming any colonies at 37°C (Craig and Jacobsen, 1984). Thermosensitivity was also observed when inducible hsp70 expression was competitively inhibited by the amplification of a vector containing the 5'-promoter region of the hsp70 gene (Johnston and Kucey, 1988). The mammalian cells still synthesized the 73 kDa constitutive and only mildly inducible form of hsp70 as well as hsp70, but allowed at most 10% expression of the inducible 72 kDa hsp70. Thermosensitivity in the absence of hsp70 was also demonstrated when affinity purified anti-hsp70 antibodies specific for the 72 and 73 kDa hsp70s were microinjected into various mammalian cells. These cells were killed by a brief severe heat shock which was survived by cells injected with control antibodies or heat denatured hsp70 antibodies (Riabowol et al., 1988). The role of hsp70 in thermoresistance was positively demonstrated by the stable transfection of rat fibroblasts with a plasmid expressing the human hsp70 gene. Cell survival of a severe heat shock was proportional to the level of expression of hsp70 in the cell prior to the heat shock (Li gt al., 1991). d. Hsp70 and Acquired Thermotolerance The phenomenon of thermotolerance has been observed in most organisms studied and is related to the above discussed thermoresistance. Acquired thermotolerance is the transient resistance to an otherwise lethal heat shock that is induced by a prior non-lethal heat shock and has been correlated with the synthesis of heat shock proteins following the first heat shock, particularly hsp70 (Li and Werb, 1982). The redistribution of hsp70 from the cytoplasm into the nucleus and nucleolus and its subsequent return to the cytoplasm upon recovery was found to be more rapid in tolerant as opposed to non-tolerant cells (Welch and Mizzen, 1988). These authors also found that the heat-induced collapse of the intermediate 12 Introduction filament cytoskeleton was prevented in cells first made thermotolerant. Acquired thermotolerance was also found to prevent the impairment of mitochondrial ATPase activity which otherwise occurred following heat shock in yeast (Patriarca and Maresca, 1990). A study of thermotolerance at the level of translation indicated that the quantity of hsp70 in the stressed cell was strictly controlled, through the initial heat shock, recovery and the subsequent severe heat shock, and that it was the only hsp whose level was so strictly dependent upon the thermal state of the cell (Mizzen and Welch, 1988). This strictly regulated level of hsp70 was also associated with the reduced translational inhibition of normal proteins and the reduced induction of hsps that occurs in stressed cells first made thermotolerant. It is therefore possible that hsp70, with its capacity to bind and protect various protein structures, may in part mediate the protected state of the thermotolerant cell. 2. Hsp90 Hsp90 is an abundant, conserved hsp found in all cell types examined. Its best characterized function involves its association with hormone receptor proteins (Pratt, 1990). A small fraction of cellular hsp90 is bound to the receptors for such hormones as glucocorticoids, progesterone, and estrogen. The current model proposes that hsp90 is bound to the receptors in the absence of hormone. It is bound to a highly conserved region which is also required to prevent proteolytic cleavage of the receptor (DeMarzo et al., 1991). Hsp90 is therefore associated with the non-DNA binding form of the receptor, possibly repressing DNA binding indirectly by blocking receptor dimerization. In addition to hsp90, hsp70 and a minor hsp, hsp56, also bind to the receptor complex prior to hormone binding. Upon hormone binding, hsp56 is released while hsp70 remains attached, possibly to aid the receptor's transport across the nuclear membrane (Sanchez, 1990). 13 Introduction Hsp90 exists in great excess over that required to bind the steroid receptors and therefore probably has more general functions within the cell, similar to what has been found for hsp70. It has been suggested that hsp90 may act as a cellular 'chaperone' involved in protein folding, protection of proteins from proteolytic degradation, and intracellular receptor transport (Pratt, 1990). 3. Hspl04/110 Hspl04 of yeast and hspllO of mammals have not been characterized as well as the other hsps. HspllO is thought to localize to the nucleolus where it may bind RNA or RNA binding proteins (Subjeck et al., 1983). Hspl04 has been shown to be required for thermotolerance in yeast (Sanchez and Lindquist, 1990). Cells mutant for this gene responded identically to high temperature as did non-tolerant wild-type cells but were unable to acquire thermotolerance from a mild heat shock as could wild-type cells. 4. The Small Heat Shock Proteins a. Homology to a-Crystallin The small heat shock proteins are found in all organisms studied but are less highly conserved than the larger hsps discussed above. Little is known about their function in the cell. An intriguing aspect of these hsps is their homology to mammalian oc-crystallins. The a-crystallin proteins are one of the major lens crystallins which make up 80 to 90% of the lens protein which in total is about 35% of the wet weight of the cell. The remaining 65% is virtually all water (Bloemendal, 1977). The properties of the crystallins are such that the lens is transparent to light and thus able to carry out its function of focussing light on the retina. As the majority of the lens cells are enucleated and avascular the crystallins must be 14 Introduction stable for the lifetime of the organism (Wistow, 1985). This implies that an ancestral small heat shock protein was recruited as a very stable protein that was suitable for the role of the lens crystallin. Taxon specific crystallins have been found which are identical or very closely related to stable enzymes such as lactate dehydrogenase, argininosuccinate lyase, enolase, and glutathione S-transferase (Wistow and Piatigorsky, 1987; Piatigorsky and Wistow, 1991). The e-crystallins expressed in the lens of birds are in fact identical to their functional lactate dehydrogenase. Most recently, it was found that ccB-crystallin is expressed in NIH 3T3 mouse fibroblasts in response to heat shock (Klemenz et al., 1991). This protein was also induced by cadmium and arsenite and localized to the nucleus upon induction, indicating that in addition to being expressed in the lens as an a-crystallin it is also a bona fide hsp. A perfect heat shock element was found in the promoter region of the gene. In each of these cases, it appears that stable pre-existing proteins have been utilized in the process of evolutionary engineering during the development of the lens. The quaternary structure of calf lens a-crystallin has been studied in detail (Tardieu el al., 1986). It is a large polydisperse globular particle with an average molecular weight of 800 kDa composed of approximately forty 20 kDa subunits. A three layer tetrahedral model was proposed whereby the first, second and third layers contain 12, 24, and 24 subunits each, respectively. Polydispersity occurs due to incomplete filling of some of the sites, a-crystallin contains two domains, each of which consists of two repeated motifs of about forty residues. The C-terminal domain is the region which is homologous to the small hsps of Drosophila and C. elegans (Wistow, 1985). Wistow found that the hsps also contain the previously unnoticed repeated motif and that all of them have a C-terminal arm which extends from the C-terminal domain and may be involved in intermolecular interactions. The N-terminal domains have diverged considerably and there is no homology between a-crystallin and the hsps or even among the hsps themselves in this region. Although Ingolia and Craig (1982) 15 Introduction suggested that the C-terminal region of homology was an 'aggregation' domain, Wistow (1985) found it more likely that this region represents an extremely thermodynamically stable structure which was 'borrowed' from the small hsps for the lens. b. The Small Hsps as Large Particles The small hsps exist as high molecular weight complexes but the nature and function of these complexes has been controversial. Early studies suggested that the complexes were associated with RNA (Kloetzel and Bautz, 1983). Cytoplasmic ring-shaped particles were isolated from Drosophila cells grown at normal temperatures, and found to contain small RNA molecules ranging in size from 60 to 200 nucleotides. Comparative V8 protease cleavage indicated that the particles contained hsp23, and antibodies raised against them cross-reacted with hsp28/27 and hsp23. A similar particle, termed the prosome, was found to be ubiquitous from yeast to humans and also showed similarity and immunological cross-reactivity to the cytoplasmic structures formed by the small hsps (Arrigo et al., 1985; Arrigo et al., 1987). The 'prosome' and the 'ring-shaped' particle were then shown to be identical to the previously described large multifunctional protease complex, an ATP-dependent, ubiquitin independent protease (Falkenburg et al., 1988; Arrigo et al., 1988). No RNA was found associated with the purest fraction of this complex, suggesting that its previous observation was due to an impurity. In its most native form, this complex was shown to be a component of the 1500 kDa protease complex which degrades ubiquitin-conjugated proteins as well as other non-conjugated peptides (Driscoll and Goldberg, 1990). Association of the small hsps with the 'prosome' was shown to be due to copurification of the hsp complex. Upon extensive purification, the hsps were separable from the 'prosome' (Arrigo ei al-, 1987; Arrigo and Welch, 1987). 16 ) Introduction The purified human hsp28 complex, which is expressed constitutively at low levels, was found to have a native molecular weight of 500 to 550 kDa. Mammalian cells have only one small heat shock protein and accordingly, the purified complex contained only hsp28 (Arrigo and Welch, 1987). The hsp28 was purified by hydroxylapatite, gel filtration, and ion exchange chromatography. Following elution from the hydroxylapatite column the hsp28 complex eluted from the gel filtration column differently than did hsp28 from crude extracts. Although the former eluted at a peak of 500 kDa, it also eluted at more included volumes suggesting that interaction with the hydroxylapatite led to a decrease in the size of the complex. Hsp28 isolated from cells harvested immediately following a heat shock was found to exist in a complex as large as 2 MDa, compared to 200 to 800 kDa under non-shock conditions or following recovery from a shock (Arrigo elaL, 1988). This suggests that hsp28 monomers are able to associate into complexes of varying sizes depending upon cell conditions. Electron microscopy of the negatively stained, purified hsp28 complex revealed its similarity to a-crystallin particles, ie. a rounded globular structure, distinct from the ring-like structure of the previously mentioned prosome. In both chicken embryo fibroblasts (CEF) and tomato cells, the small heat shock proteins have been reported to form very high molecular weight insoluble cytoplasmic aggregates although their properties are somewhat different. The CEF aggregates develop from soluble hsp particles following a second heat shock, contain no RNA and can not be crosslinked to any other proteins (Collier et al-, 1988). The tomato 'Heat Shock Granules' (HSGs) also form from soluble precursors following a heat shock but seem to be associated with a specific subset of mRNAs, particularly those control mRNAs not translated during heat shock. Heat shock mRNAs were polysome associated (Nover et al., 1989). 17 Introduction c. Intracellular Localization of the Small Heat Shock Proteins At normal temperatures (37°C) human hsp28 is a soluble cytosolic complex. Upon heat shock, its distribution changes as it becomes insoluble, and it fractionates with the low speed nuclear pellet (Arrigo and Welch, 1987; Arrigo el al., 1988). Earlier evidence suggested that this apparent nuclear localization might result from cofractionation due to the association of small hsps with the intermediate filament cytoskeleton which collapses onto the nucleus following heat shock (Leight el al., 1986). However, Arrigo and coworkers (1988) found that hsp28 redistribution was independent of an intact intermediate filament network. In addition, immunofluorescence and immunoelectron microscopy demonstrated hsp28 redistribution into the nucleus and the formation of very large aggregates after heat shock. These visible nuclear aggregates were coincident with the hsp28 complex and attained a size in excess of 2 MDa as judged by gel filtration chromatography. During recovery hsp28 re-entered the soluble cytosolic fraction. This redistribution into the cytoplasm was prevented in cells treated with the sodium ionophore monensin (Arrigo, A.-P., 1990a); hsp28 which had localized to the nucleus remained there as large aggregates at normal temperature until after the monensin was removed. It was suggested that sodium active transport was therefore involved in transporting hsp28 across the nuclear membrane. In heat shocked yeast cells, hsp26 was also concentrated in nuclei and remained there until the cells were returned to normal temperature (Rossi and Lindquist, 1989). However, under many other conditions such as stage specific hsp26 expression (non-heat shock) or constitutive expression by heterologous promoters with or without heat shock, hsp26 did not concentrate in the nucleus. It was suggested that hsp26 redistribution was dependent on cellular physiology rather than on heat stress per se. The small hsps of Dictyostelium  discoideum became localized to the nucleus, and in particular, the chromatin. Upon partial 18 Introduction degradation of the chromatin, the small hsps can be released along with the histones (Loomis and Wheeler, 1982). Drosophila hsp27 is localized primarily to the nucleus during heat shock and is detergent insoluble. However, in cells induced by ecdysterone or in recovering cells, hsp27 remains localized to the nucleus but is detergent soluble, suggesting a change in its association within the nucleus (Beaulieu eiaJL, 1989). In organisms such as Drosophila which have more than one small hsp, differences have been demonstrated in their localization. For example, Drosophila hsp23 forms dense cytoplasmic aggregates, is localized to the nucleolus upon heat shock, and rapidly returns to a soluble cytoplasmic form upon recovery (Arrigo and Pauli, 1988). This localization is similar to that seen in tomato cells and chicken embryo fibroblasts. The small hsps in tomato cells form large insoluble cytoplasmic aggregates from soluble precursors (Nover et al-, 1989) and the single small hsp of chicken embryo fibroblasts forms a large insoluble stress granule (Collier et a].., 1988). In both cases the aggregate is only observed following a second heat shock. Hsp30, a small heat shock protein of Neurospora crassa. associates with mitochondria during heat shock. This association is reversed upon return of the cells to normal temperature (Plesofsky-Vig and Brambl, 1990). d. The Small Heat Shock Proteins and Thermotolerance As discussed earlier, the hsps are thought to be involved in thermotolerance and in particular there is evidence for the role of hsp70 in this phenomenon. However, there is considerable evidence suggesting that the small hsps are also involved in thermotolerance. In Dictyostelium discoideum a mutant strain was isolated which was defective in its ability to acquire thermotolerance. While this mutant otherwise synthesized the normal set of heat shock proteins following heat shock, it failed to synthesize any of the four small hsps (Loomis and Wheeler, 1982). As discussed above, the Drosophila small hsps are induced specifically 19 Introduction by the molting hormone ecdysterone. Berger and Woodward (1983) demonstrated that both ecdysterone-stimulated tissue culture cells and day old pupae are substantially thermotolerant compared to control cells or to third instar larvae which have not yet begun synthesizing ecdysterone. More direct evidence of small hsp involvement in thermotolerance comes from genetic and physiological studies in Chinese hamster tissue culture cells (Chretien and Landry, 1988). Cells were mutagenized and selected using a single severe heat shock (4 hours at 44°C). Several surviving colonies were isolated that possessed a stable thermoresistant phenotype. Comparison of each of the heat resistant variants to wild type cells indicated an increased level of hsp27 synthesized at normal temperatures. Only one of the four cell lines showed any increase in the level of other hsps, that being hsp70. Other experiments involved the stable transfection of wild-type cells with a vector that constitutively expressed human hsp27 (Landry eial., 1989). These cells contained elevated levels of the human hsp27, with normal levels of endogenous hsps. The transfection conferred immediate and permanent thermoresistance. Heat shocking the transfectants induced the normal spectrum of hsps and a further small increase in thermotolerance but the level attained was not greater than that seen in heat induced wild type cells. e. The Role of Small Hsp Phosphorylation in Thermotolerance Following heat shock, and concomitant with their nuclear localization, the three small hsps of rat cells are phosphorylated (Kim el al-, 1984). Human hsp28 was found to exist in three isoforms, the two more acidic components being phosphorylated forms of the most basic spot (Arrigo and Welch, 1987). Hsp27 of Chinese hamster lung fibroblasts was also found to be a phosphorylated protein (Chretien and Landry, 1988). In cells grown at normal temperature, hsp27 exists mainly as the unphosphorylated 'a' form with a small amount of the phosphorylated 'b' form. Within 10 minutes after raising the cells' temperature, there is a 20 Introduction large increase in the 'b' and 'c' phosphorylated forms with a concomitant decrease in the 'a' form. This suggests that heat shock induces a rapid phosphorylation of the non-phosphorylated 'a' hsp27 even before its induced synthesis. In thermoresistant variants of these cells which contain elevated hsp27 levels, both the 'b' and 'c' isoforms are present at levels at normal temperatures which approach those of heat shocked wild type cells. After heat shock the 'b' form increased somewhat while the 'c' form increased substantially. As mentioned previously, Landry and coworkers (1989) found that constitutive expression of human hsp28 in Chinese hamster cells conferred a permanent thermoresistant phenotype. In these cells as well, hsp28 exists mainly as an unphosphorylated form at normal temperature but is phosphorylated within minutes of a heat shock. In addition to heat shock, numerous other treatments have been shown to increase the phosphorylation of the small hsps. A tumor promoting phorbol ester, a calcium ionophore, or fresh serum added to quiescent rat embryo fibroblasts all resulted in an increase in hsp28 phosphorylation without an increase in synthesis (Welch, 1985). Crete and Landry (1990) analyzed the induction of phosphorylation and thermotolerance of Chinese hamster hsp27 following exposure to the stress inducer arsenite, the translational inhibitor cycloheximide, the calcium ionophore A23187, or the calcium chelator EGTA. The effect of arsenite was similar to that of heat shock; development of thermotolerance and phosphorylation were simultaneous with hsp synthesis. On the other hand, cycloheximide, A23187, and EGTA treatment each induced phosphorylation and thermotolerance in the absence of hsp synthesis. Based on these and their past results (Landry gt al., 1989), they proposed that phosphorylation of the small hsps was a key determinant in the regulation of thermotolerance. Rapid phosphorylation of the small hsps in the absence of their synthesis was also observed when mammalian cells were stimulated with either of the cytokines interleukin la or tumor necrosis factor-a (TNF) (Kaur elal-, 1989; Arrigo, 1990b). Arrigo showed that 21 Introduction treatments with TNF, the phorbol ester PMA (phorbol-12-myristate-13-acetate), A23187, or fresh serum failed to result in a redistribution of the small hsps to the nucleus as is the case during heat shock. Phosphorylation in response to these agents occurred entirely in the cytoplasm. Arrigo (1990a) also demonstrated that dephosphorylation of hsp28 aggregates trapped in the nucleus by the sodium ionophore monensin occurred as usual upon return of the cells to normal temperature, even though the hsp28 remained as a large aggregate. This suggests that although phosphorylation may play an important role in small hsp mediated thermotolerance, it is independent of small hsp redistribution and aggregation upon heat shock. /. The Role of Calcium in Small Hsp Phosphorylation and Thermotolerance Rat hepatoma cells incubated in Ca+2-free medium containing EGTA were found to be unable to produce heat shock proteins in response to heat shock and were highly thermotolerant, but became responsive immediately upon addition of C a + 2 to the medium (Lamarche eial., 1985). A further association between the heat shock response and C a + 2 was demonstrated by Stevenson et al. (1987). A brief 45°C heat shock was shown to greatly increase 4 5 C a + 2 flux into Chinese hamster cells. The kinetics of the heat induced C a + 2 influx resembled that of the induction of thermotolerance. They also found that the C a + 2 ionophore A23187, which causes an increase in intracellular C a + 2 , led to an increase in thermotolerance. The phorbol ester PMA caused an increase in the phosphorylation of human mammary carcinoma cell hsp27 (Regazzi et al., 1988). The time course of hsp27 phosphorylation paralleled the rapid PMA induced intracellular redistribution and subsequent down regulation of protein kinase C. They concluded that hsp27 was a target of protein kinase C phosphorylation. Another phorbol ester, 12-0-tetradecanoylphorbol-13-acetate (TPA), was shown to have a similar effect to PMA on the small hsps of human breast cancer cells 22 Introduction (Darbon el al., 1990). The effects of both TPA and PMA are known to be mediated by protein kinase C, a Ca+2-dependent protein. Mouse cells that contained a vector expressing elevated levels of either calmodulin or parvalbumin, both Ca+2-binding proteins, showed a decrease in the synthesis of all three hsp26 isoforms following a heat shock while the synthesis of the other hsps was unaffected (Evans eial., 1990). The decrease in hsp26 synthesis was proportional to the level of either Ca+2-binding protein, suggesting that an effective decrease in intracellular C a + 2 rather than a specific function of increased calmodulin was responsible for the effect. However, these cells did not exhibit decreased thermotolerance even though there was a reduced expression of hsp26. Although the significance and the exact role of the interactions between C a + 2 and the small hsps are unknown, a hypothesis proposed by Landry etal- (1988) suggests an involvement between potentially cytotoxic heat induced intracellular C a + 2 fluctuations and the small hsps. If the small hsps serve such a protective role, perhaps mediated by C a + 2 -dependent phosphorylation, this would explain why high levels of Ca+2-binding proteins would reduce the need for small hsps while maintaining or increasing thermotolerance (Evans eial., 1990). This would also explain why cells exposed to EGTA or A23187 become thermotolerant in the absence of hsp synthesis as they reduce gradients existing between various intracellular pools thus preventing a heat induced C a + 2 surge (Landry etal., 1988). However, much more work is required to assess this hypothesis and to integrate the many other observations such as small hsp localization to the nucleus, and to uncover the mechanism by which the small hsps may exert their protective role. C. THE NEMATODE CAENORHABDITIS ELEGANS C_. elegans is a small free-living soil nematode that has been studied extensively using genetic and molecular biological methods at the behavioural, developmental and 23 Introduction physiological level. Dr. Sydney Brenner was able to characterize hundreds of behavioural mutants and chose to work on Q. elegans because of the ease of isolating and maintaining mutants of this self-fertilizing hermaphroditic nematode (Brenner, 1974). Its short 3 to 4 day life cycle, small size (approximately 1 mm for adults) and accessible genetics also contribute to its ease of use. The life cycle of Q. elegans is known in detail from fertilization through embryogenesis, the four larval stages, to the adult (Byerly eial., 1976). In fact, the entire cell lineage from single cell embryo to full adult has been determined (Sulston ejal-, 1983). Many genes have been identified that are directly involved in the development of C. elegans (Emmons, 1987). Work has been done to define such systems as the activities of C. elegans proteases which are of importance when attempting to purify proteins from homogenates (Sarkis et al., 1988a; Sarkis el al., 1988b). A major effort is underway to define the Q. elegans genome in overlapping cloned fragments (Coulson et al, 1986) and eventually to sequence the entire genome. D. THE SMALL HSPS OF C. ELEGANS C. elegans responds to heat shock by producing the standard set of hsps found in other organisms such as Drosophila (Snutch and Baillie, 1983). It synthesizes four different small hsps of 16 kDa which are encoded by six genes found at two loci (Fig. 1) (Russnak and Candido, 1985; Jones eial, 1986). The hspl6-l and hspl6-48 genes are divergently transcribed with only a 347 bp intergenic region. This gene pair is then duplicated to form a near perfect 3.8 kb inverted repeat (Russnak and Candido, 1985). The hspl6-2 and hspl6-41 genes are also divergently transcribed, and include a 346 bp intergenic region (Jones et a].., 1986). The intergenic regions of these two gene pairs contain all sequences, HSEs and TATA boxes, necessary to confer strict heat inducibility (Fig. IC). A large part of the intergenic region can be deleted, leaving only a single HSE between the two TATA boxes, 24 Introduction Fig. 1. The hspl6 loci of C_. elegans. (A) The hsp 16-1/48 gene pair exists as an inverted duplication, the limits of which are shown in the open bars at top. Filled and unfilled bars represent exons and introns respectively. Arrows over the genes indicate the direction of transcription. (B) The hspl6-2/41 pair. (C) Expanded view of the intergenic region of the hspl6-2/41 pair showing locations of TATA and heat shock elements (HSE) for each gene. The intergenic region of the hsp 16-1/48 gene pair is very similar. Data is from Russnak and Candido (1985) and Jones eial. (1986). Figure is adapted from Candido el ai- (1989). 25 Introduction and still confer heat inducible transcription of the hsp 16-1/48 gene pair in a heterologous mouse system (Kay eial., 1986). Sequences comparisons show that hspl6-l and 2, and hspl6-41 and 48 are 92% and 94% identical respectively (Candido etal-, 1989). Hspl6-1 and 2 code for proteins of 145 amino acids while hspl6-41 and 48 code for proteins of 143 amino acids. While this degree of identity is very high, there is considerable divergence between the sequences of hsp 16-1/2 and hspl6-41/48. The first exons of hspl6-l and hspl6-48 share only 7% identity. The second exon is 60% identical. This is consistent with the finding that the homology to the small hsps of other organisms and to a-crystallin is entirely within the second exon. Although the genes and promoters of the two hsp 16 loci are highly conserved, they exhibit considerable differential expression upon heat induction (Jones eial-, 1989). The hspl6-2/41 locus transcribes up to seven times more mRNA than the hspl6-l/48 locus, despite the presence of two copies of the hsp 16-1/48 genes. It was found that this was likely due to differences in transcription and not mRNA stability. This suggestion was further supported by studies of nuclease hypersensitive sites of the two hsp 16 loci (Dixon et al., 1990). The distribution of hypersensitive sites in the chromatin structure changed following heat shock, in parallel with gene activation. However, the hsp 16-1/48 locus returned to a normal, inactive configuration earlier that did the hsp 16-2/41 locus, suggesting that the hsp 16-2/41 locus remained transcriptionally active for a longer period of time thereby producing more mRNA. E. THE PRESENT STUDY Although the studies discussed above had identified and characterized the Q. elegans hsp 16 genes and their transcriptional expression, essentially nothing was known about the properties of hsp 16 proteins themselves. In addition, information on the small hsps was 26 Introduction available from only a few organisms such as Drosophila (Arrigo el al., 1985) and some mammals (Welch, 1985; Arrigo and Welch, 1987). It was therefore of considerable interest to characterize the small hsps of Q. elegans. Thus began the present study which involved the analysis of the £ . elegans small hsps with anti-peptide antibodies and their purification and characterization by classical chromatography and affinity chromatography techniques. 27 Materials and Methods III MATERIALS AND METHODS A. G R O W T H OF CAENORHABDITIS ELEGANS Caenorhabditis elegans Bristol N2 strain was grown either on plates or, more commonly, in liquid culture. Plates were NGM (nutrient growth medium: 0.3% NaCl, 0.25% tryptone, 5 mg/ml cholesterol, 1 mM CaCl2, 1 mM MgS04, and 25 mM KH2PO4 pH 6.0) spread with a feeder lawn of E. coli strain OP50 as described by Brenner(1974). Liquid cultures of Q. elegans from 0.25 liters to 25 liters (commonly 4 liters) were grown according to a procedure adapted from Sulston and Brenner (1974). From 1 to 1.5 grams of wet packed nematode embryos were inoculated per liter of Basal S medium (0.1 M NaCl, 50 mM KH2PO4 pH 6.0) supplemented with 0.01 mg/ml cholesterol, 2 mM potassium citrate pH 6.0, 0.3 mM CaCl2, 0.3 mM MgS04, 1.3 mM FeS04, 2.5 mM EDTA, 0.5 mM ZnS04, 0.5 mM MnCl2, and 0.05 mM CuS04. With heavy aeration, embryos were allowed to hatch overnight (minimum 10 hours) to the L l stage to synchronize the culture. Approximately 50 grams of E. coli strain MRE 600 (obtained washed and frozen from Grain Processing Corporation, Muscatine, Iowa) was then added per gram of inoculum. Antifoam A (Sigma) was added as required to control foaming, although this was greatly reduced with the use of washed E. coli . With aeration maintained as high as possible, the nematodes can grow to gravid adults in 3 days at 20 °C. Overcrowding of the culture increased this length of time up to 4 days and often required addition of more E. coli. The pH of the culture can increase dramatically, especially in overcrowded conditions, and was adjusted with 1 M citric acid or 50% acetic acid. 28 Materials and Methods 1. Collection of Gravid Adult Nematodes Gravid adults were aerated on ice until the culture temperature dropped below 8 °C. The culture was then placed at 4 °C until the nematodes settled out. The supernatant containing much of the remaining bacteria was aspirated and the nematodes were washed once with one culture volume of cold 0.14 M NaCl. Live nematodes were separated from bacteria and debris by flotation in cold 30% sucrose (Sulston and Brenner, 1974). Centrifugation was at 2000 x g for 2 minutes. Floated nematodes were rinsed promptly in cold 0.14 M NaCl to remove sucrose. 2. Preparation of Nematode Embryos Embryos were prepared from floated gravid adults by alkaline hypochlorite treatment (Emmons el al.., 1979). Five grams or less of adults were pelleted in 50 ml polycarbonate tubes (2000 x g, 1 min, 4 °C). The tubes were then filled to 50 ml with 1.2% sodium hypochlorite, 0.5 M NaOH and shaken vigorously for a total of nine minutes with one change of the hypochlorite solution. The pellet was immediately washed twice in 0.14 M NaCl. The washed pellets of nematode embryos were pooled and floated as above. B. HEAT SHOCK CONDITIONS Embryos were heat shocked either on NGM plates or in liquid culture. For the liquid culture, embryos were added to Basal S medium in a baffled Erlenmyer flask and shaken in a water bath. In either case, the heat shock was for 2 hours at 33 °C. This was at times followed by a recovery period at 18 to 20 °C, usually of 8 to 12 hours, and a second 2 hour heat shock and recovery. The nematodes were then collected by centrifugation, rinsed once in cold 0.14 M NaCl, and twice in either Buffer A (20 mM Tris pH 7.4, 20 mM NaCl, 1 mM 29 Materials and Methods EDTA, 0.1 mM MgCl2, 10% glycerol) or Buffer B (Buffer A with no MgCl2). They were then dripped into liquid N2 from a Pasteur pipet as a thick slurry, and stored at -70 °C. c- 35S-LABEUNG OF C. ELEGANS 1. Label ing of E.. coli 0.01 ml of an overnight culture of E. coli strain K12 Hfr H was added to 2 ml of low sulfate M9 medium (LSM) (5.8% Na2HP04, 3% KH2PO4, 0.05% NaCl, 0.1% NH4CI, 1 mM MgCl2, 1% glucose, and 0.05 mM MgS04) of Bretscher and Smith (1972) containing 1 mCi of carrier free 3 5S04 (ICN). After incubation at 37 °C for 4 hours, this culture was added to 98 ml of LSM containing 4 mCi 3 5S04 and incubated for 20 hours at 37 °C. The bacteria were collected by centrifugation at 4000 x g for 20 minutes and stored at -70 °C in 0.2 mCi aliquots. Samples of the cell suspension and supernatant were taken for scintillation counting to determine the efficiency of 3 5 S incorporation. 2. Label ing of C_. elegans 0.5 mCi of 35S-labeled E. coli and 1 gram of nematodes were added to an NGM plate either at 20 °C for control labeling or pre-warmed to 33 °C for heat shock labeling. The plates were incubated for 2 hours before the nematodes were rinsed with 0.14 M NaCl and incubated with unlabeled E. coli on a new plate for 30 minutes. The temperature of the heat shocked nematodes was maintained at 33 °C throughout this time. The nematodes were then collected, rinsed once with 4 °C 0.14 M NaCl and twice with either Buffer A or Buffer B, and frozen and stored as described above. 30 Materials and Methods D. HOMOGENIZATION 1. Sonication Frozen nematode pellets were thawed and diluted to approximately 0.1 gram nematodes per ml of an appropriate buffer in a 15 ml Corex centrifuge tube. The temperature was maintained at 4 °C throughout. PMSF was added to 0.2 mM and the mixture was homogenized with 5 to 6 ten-second full power bursts from a sonicator fitted with a microtip (Heat Systems Inc.). The solution was cooled on ice between bursts. The homogenate was then centrifuged at 20,000 x g for 20 minutes. The supernatant was carefully removed and filtered through a syringe fitted with a 0.22 u.m filter. 2. French Press Frozen nematode pellets were thawed and diluted to approximately 1 gram of nematodes per 8 ml of the appropriate buffer. The temperature was maintained at 4 °C throughout. PMSF was added to 0.2 mM, the mixture was loaded into a 40 ml french press cell (American Instruments Co., Inc., Silver Spring, Maryland, USA) pre-cooled to 4 °C and pressed at 10,000 psi. The homogenate was centrifuged at 20,000 x g for 20 minutes, the supernatant was carefully removed, and filtered through a syringe fitted with a 0.22 u.m filter. E. SDS-POLYACRYLAMIDE GEL ELECTROPHORESIS 1. lD-Electrophoresis The discontinuous SDS-polyacrylamide gel system of Laemmli (1970) was used to separate and analyze various protein samples made 1 X in SDS sample buffer (62.7 mM 31 Materials and Methods Tris-HCl pH 6.8, 1% P-mercaptoethanol, 1% SDS, 10% glycerol, 5 mg/ml bromophenol blue). Samples were heated to 100 °C for 5 minutes prior to loading on the gel. A 'mini-gel' apparatus with 8 cm x 10 cm x 0.75 mm plates was used to pour 18% polyacrylamide separating gels with 4.5% polyacrylamide stacking gels. Gels were run at 80 to 100 Volts until the bromophenol blue ran off the end of the gel. 2. 2D-Electrophoresis Two-dimensional gel electrophoresis was carried out according to the procedure of O'Farrell (1974) with a first dimension pH gradient of 3 to 10 (Pharmacia, 2D Pharmalyte 3-10) and a second dimension SDS gel as described above. The tube gels were 70 mm x 1.5 mm and the second dimension gels were 1.5 mm in thickness. They were run at 400 Volts for 12 hours followed by 800 Volts for 1 hour. Tube gels which were not used immediately were extruded and stored at -70 °C. After thawing, they were equilibrated for 2 minutes in equilibration buffer (0.124 M Tris-HCl pH 6.8, 2% SDS, 2% p-mercaptoethanol, 10% glycerol) prior to loading on the second dimension gels. F. DETECTION OF PROTEIN ON SDS GELS 1. Coomassie Blue Stain Gels were first fixed for 2 to 4 hours in Solution C (50% methanol, 7% acetic acid), stained with Solution C containing 0.05% Coomassie Brilliant Blue R-250 (Biorad) for a minimum of 2 hours, and destained with Solution D (10% methanol, 7% acetic acid). The gels were either stored wet in 7% acetic acid or dried under vacuum onto filter paper for a minimum of 2 hours. On gels which were to be stained with Coomassie Blue, a low molecular weight standard (LMWS) was run which contained Bovine Albumin (M r = 66 kDa), Egg Ovalbumin 32 Materials and Methods (M r = 45 kDa), Glyceraldehyde-3-phosphate dehydrogenase (M r = 36 kDa), Carbonic anhydrase (M r = 29 kDa), Trypsinogen (M r = 24 kDa), Soybean trypsin inhibitor (M r = 20.1 kDa), and a-lactalbumin (M r = 14.2 kDa) (Sigma). 2. Autoradiography Dried SDS gels or Western blots containing 35S-labeled proteins were exposed to Kodak X-Omat AR film. 14C-containing ink was often first spotted onto the filter paper to facilitate the alignment of autoradiographic spots and protein spots. The film was developed in an automated X-Omat developer (Kodak). G. PRODUCTION OF ANTI-HSP16 ANTISERUM 1. Synthesis of Peptides Three peptides were synthesized using tertiary butyloxycarbonyl solid phase chemistry on an Applied Biosystems 430A Peptide Synthesizer and purified by reverse phase HPLC (Ian Clark-Lewis, Biomedical Research Center, UBC, Vancouver) (Clark-Lewis et ai., 1986) corresponding to amino acids 125 to 143 of the hspl6-41 gene, 33 to 50 of the hspl6-l gene, and 110-145 of the hspl6-2 gene. Peptides 16-1 (33-50) and 16-2 (110-145) had a cysteinyl residue added to their N-terminus to facilitate linkage to a solid support. The 16-41 (125-143) peptide includes as part of its natural sequence a cysteinyl residue at its N-terminus. 2. Immunization of Rabbits New Zealand White rabbits were immunized with 0.25 to 0.5 mg of peptide conjugated to keyhole limpet hemocyanin (KLH) in an emulsion of Freund's complete adjuvant and 33 Materials and Methods sterile 0.14 M NaCl (3:1 ratio). The animals were boosted at three week intervals with the same emulsion except containing Freund's incomplete adjuvant. Blood samples were made 0.1% Sodium Azide, allowed to coagulate at 4 °C overnight, and centrifuged at 15,000 x g for 20 minutes. The serum was then stored at -70 °C. 3. Ammonium Sulfate Precipitation of Sera Ammonium sulfate to a final concentration of 45% was slowly added to the serum which was chilled on ice and stirred vigorously. The IgG precipitate was collected by centrifugation (10,000 x g for 10 minutes), made up to the original serum volume in 0.14 M NaCl. It was then reprecipitated two or three times until the solution was colorless, ie. no red color due to hemoglobin. It was then dialyzed twice against 2 liters of PBS (20 mM potassium phosphate pH 7.4, 0.12 M NaCl), and stored at -70 °C. 4. Affinity Purification a. Peptide Column Preparation Peptide columns were prepared by linking the appropriate peptide to thiopropyl-Sepharose 6B (Pharmacia) via the N-terminal cysteinyl residue. The peptide was dissolved in 0.1 M acetic acid (5 mg peptide/100 ml), neutralized with 1 M NaOH (10 ml/100 ml), and diluted 135 fold (to 15 ml) with PBS. Thiopropyl-Sepharose (1 ml/5 mg peptide) was washed in 15 volumes of PBS three times, added to the peptide solution, and incubated overnight at room temperature with shaking. The peptide-Sepharose was then packed into a column and washed successively with PBS, 0.1 M glycine-HCl pH 2.5, PBS, PBS containing 1 mg/ml bsa, and finally PBS containing 0.1% Sodium azide for storage. 34 Materials and Methods b. Serum Purification Ammonium sulfate precipitated serum containing 0.1% Sodium azide was agitated for 2 to 4 hours at room temperature with the corresponding peptide-Sepharose (ie. anti-hspl6-41 (125-143)-KLH serum mixed with hsp 16-41 (125-143)-Sepharose) (10 to 12 ml serum/ml Sepharose). The slurry was then packed into a column, washed extensively with PBS, and eluted with 0.1 M glycine-HCl pH 2.5. Fractions were neutralized with saturated Tris base (12 to 15 |il/ml eluant) and analyzed for protein content by measurement of the absorbance at 280 nm. The IgG containing fractions were pooled, dialyzed twice against 2 liters of PBS, and stored in aliquots at -70 °C. H. WESTERN BLOT ANALYSIS OF PROTEINS I. Blotting Procedure One- or two-dimensional SDS-polyacrylamide gels were electro-blotted to polyvinylidene difluoride (Immobilon-P, Millipore) membranes using a method similar to that for nitrocellulose (Towbin, H. et al., 1979). The gels were slowly agitated in transfer buffer (Buffer E) (25 mM Tris base, 192 mM glycine, 0.05% SDS, 15% methanol, pH 8.2-8.3) for 15 minutes. The membranes were moistened in methanol, soaked in dH20 for 5 minutes, and then soaked in Buffer E for 10 minutes before being placed against the gel between filter papers. This assembly was placed in an electro-blotting apparatus (Trans-blot cell, Biorad) filled with Buffer E, and transfer allowed to occur either for 45 minutes at 250 mA or overnight at 50 mA. Following transfer the blot was rinsed briefly in buffer E followed by df^O and dried at room temperature. On gels which were to be blotted, a pre-stained molecular weight standard (PSMWS) was run which contained Phosphorylase B (apparent M r = 110 kDa), 35 Materials and Methods Bovine Albumin (84 kDa), Egg ovalbumin (47 kDa), Carbonic anhydrase (33 kDa), Soybean trypsin inhibitor (24 kDa), and Lysozyme (16 kDa) (Biorad). 2. Immunostaining of Blots Dried Western blots were first wet in methanol, soaked briefly in wash buffer (Buffer F) (20 mM Tris-HCl pH 7.4, 0.14 M NaCl, 0.1% non-fat dry milk), and then in Blocking buffer (20 mM Tris-HCl pH 7.4, 0.14 M NaCl, 1.0% non-fat dry milk) for a maximum of 1 hour at 37 °C. Blots were incubated with antibody in Buffer F containing 0.05% Tween-20 (Sigma) for 1.5 to 2 hours at room temperature. The primary antibody was diluted 1000 fold and the secondary antibody, goat-anti-rabbit IgG conjugated to either horseradish peroxidase or alkaline phosphatase (BRL), were diluted 3000 fold. The blots were washed in 3 changes of Buffer F for 5 minutes each after each incubation. Blots with the peroxidase conjugate were developed in 10 mM Tris pH 7.4, 0.01% H 2 0 2 , and 0.0025% o-dianisidine. (Sigma) (from a 1% stock in methanol prepared monthly). Blots with the alkaline phosphatase conjugate were developed with bromochloroindolyl phosphate/nitro blue tetrazolium (BCIP/NBT). The BCIP/NBT solution contained 0.033% NBT, 0.017% BCIP (from 5% stocks in 70% and 100% dimethylformamide respectively), 100 mM NaCl, 5 mM MgCl 2, and 100 mM Tris-HCl pH 9.5. Development was stopped with PBS containing 20 mM EDTA. The blots were then dried at room temperature. /. COLUMN CHROMATOGRAPHY 1. Hydroxylapatite Chromatography Hydroxylapatite was hydrated with Buffer A or B, decanted once and packed into a column at 4 °C. Approximately 8 ml of hydroxylapatite was used for the extract from 1 gram 36 Materials and Methods of wet packed nematodes. The filtered extract was passed down the column and the flow-through was collected for future analysis. The column was washed with 1.5 volumes of starting buffer and 1.5 volumes of Buffer G (10 mM potassium phosphate pH 7.4, 20 mM NaCl, 1 mM MgCl 2, 0.1 mM EDTA, 10% glycerol). It was eluted first with 1.5 volumes of Buffer G/50 (Buffer G except containing 50 mM potassium phosphate) followed by the same volume of Buffer G/200 (Buffer G containing 200 mM potassium phosphate). The eluted fractions were concentrated by ultrafiltration (Amicon) for large volumes and by Centricon filtration for volumes less than 6 ml. All buffers were degassed prior to use. 2. Gel Filtration Chromatography Protein from either a crude extract or a hydroxylapatite fraction was fractionated according to size on an S-300 Sephacryl HR gel filtration column (Pharmacia). The 1.6 x 100 cm column was packed in Buffer A or B containing 0.1 M NaCl according to product literature, a flow adapter was attached to the top of the column, and the samples were run in a reverse direction to packing. Two ml samples made 0.1 M in NaCl were applied to the column and eluted at a flow rate of approximately 5 ml/hour. Fractions (2 ml) were collected after the void volume which was first determined by the elution of Blue Dextran 2000 (Sigma). Following elution, the column was washed with 1 volume of starting buffer. The column was calibrated with a set of proteins of defined molecular weight (29 kDa to 669 kDa; Sigma). 3. Affinity Chromatography a. Synthesis of Antibody-Sepharose Affinity purified antibodies were linked to either cyanogen bromide-activated Sepharose (Sigma) (procedure according to Pharmacia publication, Affinity Chromatography. 37 Materials and Methods Uppsala, Sweden) or Protein-A Sepharose (Sigma) (Harlow and Lane, 1988). Antibody (0.5 to 2 mg) in coupling buffer (Buffer H) (0.1 M Na Borate pH 8.3, 0.5 M NaCl) was added to 1 ml of CNBr-activated Sepharose which had been swollen in 1 mM HC1 and equilibrated in Buffer H. This was mixed for 2 hours at room temperature and then transferred to 0.2 M glycine pH 8.0, 0.5 M NaCl for a further 2 hours. The gel was then washed with Buffer H and sodium acetate pH 4.0, 0.5 M NaCl alternately several times, and finally washed and stored in PBS, 0.1% Sodium azide. Antibody in PBS was mixed for 1 hour at room temperature with Protein-A Sepharose (0.3 mg/ml). The beads were washed twice in 10 volumes of 0.2 M Na Borate pH 9.0 and then cross-linked with 20 mM dimethylpimelimidate (Sigma) in the same buffer for 30 minutes. The beads were washed once in 0.2 M ethanolamine pH 8.0 and shaken in the same buffer for 2 hours at room temperature, to block unreacted crosslinker. The gel was then washed and stored in PBS containing 0.1% Sodium azide. The coupling efficiency for these columns was estimated by running samples of antibody before and after coupling on an SDS-polyacrylamide gel and staining with Coomassie Blue. b. Binding and Elution of Antigen A crude or partially purified, filtered extract in a buffer containing 0.5 M NaCl and, in one case, 4 M urea was recycled twice at about 10 ml/hour through a column containing antibody-Sepharose or was mixed with the gel for 2 to 4 hours. The gel was washed in 20 column volumes of starting buffer followed by 10 mM potassium phosphate pH 7.4 (containing 4 M urea if it was present in the starting buffer). The antigen was then eluted in 5 column volumes of 0.1 M glycine-HCl pH 2.5. 1 ml fractions were collected into 1/20 volume of 1 M 38 Materials and Methods potassium phosphate pH 7.4. These were pooled and dialyzed against 10 mM ammonium bicarbonate and lyophilized, or against Buffer B and frozen at -20 °C. J. PROTEIN QUANTITATION The total protein content of a sample was estimated either by the solution's absorbance at 280 nm, or by its binding in solution to Coomassie Brilliant Blue G-250 (Biorad Protein Assay). For the Biorad assay, acetylated bovine serum albumin was used to produce a standard curve. Hsp 16 was roughly quantitated by Western blotting of samples followed by immunodetection. 39 Results and Discussion III. RESULTS AND DISCUSSION A. PRODUCTION OF ANTIBODIES SPECIFIC FOR HSP 16 Initial investigation of the small hsps of Q. elegans involved the autoradiographic analysis of fractionated 35S-labeled nematode extracts. To facilitate the purification and analysis of the hsp 16s antibodies to the proteins were produced. The hydropathy plot (Kyte and Doolittle, 1982) of each of the proteins was analyzed (not shown) for regions of high hydrophilicity. Amino acids 33 to 50 of hsp 16-1 and 125 to 143 of hsp 16-41 were chosen as regions to which peptides were synthesized. Polyclonal antibodies were raised against these peptides and tested in various experiments. They were found to detect hsp 16 polypeptides on a Western blot (Fig. 2A and C), but were not effective for the purpose of isolating or detecting native, soluble hsp 16 in experiments such as immunochromatography or immunoprecipitation (data not shown). The hsp 16 protein sequences were then analyzed with a more comprehensive program Q°C gene, Intelligenetics) and the region of amino acids 110 to 145 of hspl6-2 was selected based on its hydrophilic and a-helical nature, and its high antigenicity rating. The titre of this antibody against its peptide antigen was approximately 1/64000 by ELISA (Enzyme Linked Immunosorbent Assay) and was similar to the titres of the 1/33-50 and 41/125-143 antibodies. The 2/110-145 antibody, as will be detailed below, was found to be the most useful for detecting and purifying soluble hsp 16. B. SPECIFICITY OF ANTI-HSP16 ANTIBODIES The specificity of each of these antipeptide antibodies was examined by Western blot analysis of control and heat shock induced 35S-labeled C. elegans extracts (Fig. 2). Each of 40 Results and Discussion Fig. 2 Western blot analysis of S-labeled control and heat shocked nematode extracts. Western blots (West) of control (Cont) and heat shocked (HS) extracts were developed with (A) anti-1/33-50, (B) anti-2/110-145, and (C) anti-41/125-143 primary antibody and horseradish peroxidase Goat-anti-Rabbit secondary antibody. Autoradiograms (Aut) of the heat shock lane of each blot are aligned vertically to the right of their respective blots. The solid arrowheads mark the 16 kDa polypeptides and the hollow arrowheads mark the 18 kDa polypeptides. 41 Results and Discussion the antibodies detected primarily a single polypeptide which appeared only in the heat shock induced extracts. Upon closer inspection, however, it is clear that each of the antibodies does not detect the same polypeptide. The 1/33-50 and 2/110-145 antibodies detect an 18 kDa band while the 41/125-143 antibodies detect a 16 kDa band. In addition, the 2/110-145 antibody faintly detects the 16 kDa band while the 41/125-143 antibody faintly detects the 18 kDa band. This antibody specificity correlates well with the extent of sequence similarity between the hsp 16 proteins in regions corresponding to the three peptides. As can be seen in Fig. 3, hsp 16-1 and hsp 16-2 are highly similar to each other but are only 50% identical to hspl6-41 and hspl6-48 at positions 110 to 145 and are only 30% identical at positions 33 to 50. Similarly hspl6-41 and hspl6-48 are identical at positions 125 to 143 but are only 45% identical to hsp 16-1 and hsp 16-2. These results suggest that the 18 kDa heat shock polypeptide may not be a distinct small hsp as previously believed (Snutch and Baillie, 1983; Russnak et al., 1983) but rather may be encoded by the hspl6-l and hspl6-2 genes, while the 16 kDa heat shock polypeptide may be encoded by the hsp 16-41 and hsp 16-48 genes. C. HEAT SHOCK INDUCTION OF C. ELEGANS 1. Optimal Time for Heat Induction In order to determine the optimal length of time for the heat shock induction of the small hsps, a culture of adult C. elegans was incubated at 33 °C and samples were removed at various times for analysis. Fig. 4 shows that hsp 16 was first detectable 1 hour after induction. Its concentration increased at 1.5 hours, was greatest at 2 hours, and began to decrease after 3 hours at 33 °C. Experiments which analyzed the heat shock induction of the hsp 16 genes by Northern dot-blot analysis showed that hsp 16 mRNA levels were detectable after 15 minutes and 42 Results and Discussion 33 1/33-50 I HSP16-1 C R G S P S E S S E I V N N D Q K F i HSP16-2 C R G i P S E S S E I V N N D Q K F ! HSP16-41 s f n f s d n i g e I V N d e s K F HSP16-48 s f n f s d n i g e i v n d e s K F 110 i 2/110-145 HSP16-1 1 N L S E D G K L S I E A P K K E A i Q G R S I P I Q Q A p V E q K t s E HSP16-2 N L S E D G K L S I E A P K K E A V Q G R JS I P I Q Q A I V E E K S A E HSP16-41 a i S n e G K L q I E A P K K t n s s - R S I P I n f v a k h HSP16-48 a i S n e G K L q I E A P K K t n s s R S I P I n f v a k h 125 i 41/125-143 HSP16-1 1 A P K K e a i q g R S I P I q q a P V e q k t s e HSP16-2 A P K K e a v q g R S I P I q q a i V e e k s a e HSP16-41 A P K K T N S S - R S I P I N F V A K H HSP16-48 A P K K T N S S - R S I P I N F V A K H Fig. 3 Amino acid sequence of the hspl6 genes in the region to which synthetic peptides were produced. Peptides were synthesized for the genes shown in bold type. Capitalized letters represent those amino acids identical to the peptide, lower case letters represent non-identity to the peptide. Amino acid sequences are from Russnak and Candido (1985) and Jones ej al. (1986). 43 Results and Discussion 0 0.25 0.5 1 1.5 Fig. 4 Heat shock induction of C_. elegans for varying lengths of time. Adult nematodes were heat shocked in aerated 0.14 M NaCl at 33 °C. Samples were removed at various times [in hours], homogenized by sonication in TE, and blotted. Blots were developed with anti-41/125-143 primary antibody and horseradish peroxidase conjugated secondary antibody. Total protein in each lane (other than [E]) is constant. [E] is a more concentrated two hour heat shock sample. The solid arrowhead marks the 16 kDa polypeptide. 44 Results and Discussion peaked and reached a plateau after 45 minutes for hsp 16-2 and 60 minutes for hsp 16-1 (Jones eial., 1989). This indicates that with a more sensitive detection system, hspl6 may be observable within 30 minutes. Although the quantity of mRNA peaks after 45 minutes o f induction, it remains translationally active for at least 2 hours after induction, when the quantity of hsp 16 in the culture is greatest. The decrease in hsp 16 in the culture by 3 hours after induction is likely due to a high nematode mortality rate which is generally observed after this length of time at 33 °C. 2. Stability of Hspl6 Isolated from Adult vs. Embryonic Nematodes For initial experiments on hsp 16, adult nematodes were used as the source material as they were easiest to isolate and provided the greatest mass from a culture. In these experiments, it was found that hsp 16 was subject to proteolysis after a short'period o f time at 4 °C. An experiment was conducted in which adult [Adults] or embryonic [Embryo] nematode extracts were mixed with SDS-sample buffer and boiled either immediately [0] or after two days [2] of storage at 4 °C (Fig. 5). The results indicated that hsp 16 from adults was substantially degraded almost immediately[Adults, N, 0] and was nearly completely degraded after two days [Adults, N, 2]. In comparison, hspl6 from embryos did not appear to be degraded significantly, even two days [Embryo, N, 2] after homogenization. A likely explanation for these observations is that adults contain gut proteases which are not yet active in embryos. The inclusion of the protease inhibitors PMSF, leupeptin, and pepstatin [I] did not appear to affect the rate of degradation of hspl6 in either adults or embryos compared to samples in buffer alone [N] (Fig. 5). However these inhibitors were shown to be particularly effective against acidic proteases (Sarkis eial., 1988a). It is possible that there are proteases in the adult gut which are not affected by these inhibitors. 45 Results and Discussion Adul ts Embryo N 0 I 0 N 2 I 2 N 0 I 0 N 2 I 2 < Fig. 5 Stability of hspl6 in Embryos vs. Adults with and without protease inhibitors. Equal masses of Adult or Embryonic nematodes were sonicated either in T E [N] or in T E containing 67 u.g/ml pepstatin A, 67 u.g/ml leupeptin, and 0.1 mM PMSF [I]. Samples of these extracts were mixed with SDS-sample buffer and boiled either immediately [O] or after two days at 4 °C [2]. [E] is as in Fig. 4. Samples were analyzed by Western blotting as in Fig. 4. The solid arrowhead marks the 16 kDa polypeptide. 46 Results and Discussion The results of these experiments emphasize the necessity of using embryos as the source for the isolation and study of hsp 16. 3. Induction of Hspl6 When Periods of Heat Shock are Followed by Periods of Recovery During a heat shock and subsequent period of recovery hsps are synthesized which are thought to induce a state of thermotolerance against a second severe heat shock. In both • rat fibroblasts (Widelitz eial., 1986) and tomato cells (Nover et al., 1989), it was shown that following a second heat shock, the hsps were further induced to high levels. We therefore set out to examine C. elegans hsp expression in detail, with an outlook to obtaining a maximum yield of hsp 16 from each nematode culture. An experiment to follow hsp 16 levels was therefore conducted with heat shocked C. elegans embryos (Fig. 6). Embryos were heat shocked for two hours, allowed to recover for thirteen hours, heat shocked again for two hours and allowed to recover for a further eight hours. Hspl6 was barely detectable immediately after heat shock (Fig. 6, [2]), but increased gradually during the thirteen hours of recovery [2 to 15]. Following the second heat shock [17] the level of hsp 16 approximately doubled and continued to increase, perhaps doubling again, during the eight hour recovery [17 to 25]. Although faint, these results show that hsp 16 increases considerably during the recovery phase, not just during the heat shock induction. Also it shows that a second heat shock and recovery leads to more than double the amount of hspl6 compared to a single induction and recovery alone. Thus the double heat shock/recovery regimen is potentially useful for hsp 16 purification studies. 47 Results and Discussion 10 14 E < 15 17 19 21 25 Fig. 6 Induction of hspl6 by two successive periods of heat shock, each followed by a period of recovery. Embryos were suspended in Complete 'S' medium in a shaking water bath. They were heat shocked at 33 °C for 2 hours [0 to 2], allowed to recover at 20 °C for 13 hours [2 to 15], heat shocked again for 2 hours [15 to 17], and allowed to recover finally for 8 hours [17 to 25]. Samples were removed at various times and homogenized, blotted, and developed as in Fig. 4. [E] is a concentrated sample of a 2 hour heat shocked extract. The solid arrowheads mark the 16 kDa polypeptides and the hollow arrowheads mark the 18 kDa polypeptides. 48 Results and Discussion D. ANALYSIS OF CONTROL AND HEAT SHOCK INDUCED NEMATODE EXTRACTS BY TWO-DIMENSIONAL GEL ELECTROPHORESIS There are four distinct hsp 16 genes in the Q. elegans genome which produce four different 16 kDa heat shock proteins upon induction. These four proteins can be resolved into two components by one dimensional SDS-gel electrophoresis (Fig. 2; Fig. 7A2, left side of panel). In order to examine this protein family in more detail, they were analyzed by two dimensional gel electrophoresis. Both 35S-labeled control and heat shocked nematode extracts were analyzed in duplicate. One gel of each was dried and the other was blotted and immunostained with the anti-hsp 16-2/110-145 antibody. Fig. 7A shows the autoradiograms of the entire gels, while Fig. 7B and 7C show the portions of the gels containing the hsp 16 peptides. Figs. 7A1 and 7A2 also include samples separated only in the second dimension on the left side of each gel. Comparing the control and heat shock panels (Fig. 7A) one can see that the majority of components are more intensely labeled in the control extract with the exception of a few intense components in the heat shock extract. This is because during heat shock, the translation of non-heat inducible proteins is greatly reduced (see for example: Mizzen and Welch, 1988). Therefore, since the heat shock extract was derived from cells labeled only during heat shock, it is enriched for labeled heat shock proteins. The heat shock sample separated in only the second dimension (SDS-gel electrophoresis) includes the two characteristic 16 and 18 kDa bands which do not appear in the control sample. When separated in two dimensions (isoelectrofocusing electrophoresis and SDS-gel electrophoresis) , these two bands were resolved into approximately thirteen components. This can be seen more clearly in Fig. 7B4 compared to 7B3 where the heat shock peptides are marked by solid dots. There are four major components, two 16 kDa and two 18 kDa 49 • Results and Discussion Fig. 7 Two-dimensional gel electrophoretic analysis of extracts from control [Cont] and heat shocked [HS] 35S-labeIed nematodes. 340,000 cpm and 310,000 cpm of control and heat shocked extracts respectively in TE, 0.2 mM PMSF were electrophoresed in the first dimension. These tube gels, as well as 200,000 cpm each of control or heat shock extract [left lane in (Al) and )A2)] were then separated in the second dimension. (Al) and (A2), autoradiograms of the entire control and heat shock gels, respectively. (B) and (C), portions of the full gel in the region of the small hsps. (Bl) and (B2), Coomassie blue stained gels; (B3) and (B4) autoradiograms of (Bl) and (B2), respectively (same as (Al) and (A2)). (CI) and (C2), Western blots of the gels developed with 2/110-145 primary antibody and alkaline phosphatase conjugated secondary antibody. (C3) and (C4), autoradiographs of (CI) and (C2), respectively. The dots in (B4) indicate protein components not found in (B3). The solid arrowheads mark the 16 kDa polypeptides and the hollow arrowheads mark the 18 kDa polypeptides. The pH gradients in (A) are pH 3 to 10 from left to right and in (B) and (C) approximately pH 6 to 8 from left to right. 50 Results and Discussion Cont-Aut HS-Aut C o n t - W e s t HS-West 2 51 Results and Discussion polypeptides. These four components contain sufficient protein to be clearly visible on the Coomassie blue stained 2-D gel (Fig. 7B2) although the more basic 16 kDa peptide co-migrates with a larger non-heat shock induced protein (Fig. 7B1 and 2). The Western blots in Figs. 7C1 and 7C2 demonstrate that the control extract contains no hspl6 as detectable by autoradiography or the anti-hspl6-2/110-145 antibody. In comparing the heat shock Western blot (7C2) to its autoradiogram (7C4) the specificity of this antibody is particularly evident. While the pattern of detected components in (7C4) is essentially the same as in the autoradiogram (7B4) of the gel, the 2/110-145 antibody reacts predominantly with the two major 18 kDa polypeptides. This suggests that these two components may represent the hsp 16-2 and hsp 16-1 polypeptides. A further clue as to the identities of these four major components may be derived from knowledge of the differential expression of the two hspl6 loci. The hspl6-2/41 locus is expressed at four- to eight-fold greater levels than the hspl6-l/48 locus in rapidly heated Q. elegans embryos (Jones et al., 1989). Close inspection of the heat shocked autoradiograms in Fig. 7B4 and C4 indicates that the two more basic 16 and 18 kDa major polypeptides are present at higher levels than the two more acidic 16 and 18 kDa major polypeptides. It may be tentatively suggested therefore that the two more acidic 16 and 18 kDa polypeptides may correspond to hsp 16-48 and hspl6-l respectively, and the two more basic 16 and 18 kDa polypeptides to hspl6-41 and hsp 16-2 respectively. This assignment will require confirmation by direct means such as amino acid sequencing, or hybrid-selected translation using gene-specific probes. In addition to the four major heat shock components visible on these gels there are also at least nine minor polypeptides which are clearly heat inducible. Three of these have an apparent size of 18 kDa, three are 16 kDa in size, and three others appear to be of an intermediate size, ie. approximately 17 kDa. The most intense of these spots are also visible on the Coomassie blue stained gel (7B2). Some of these hsp 16 components most likely 52 Results and Discussion result from post-translational modifications such as phosphorylation. Phosphorylation has been shown to occur in the small hsps of several organisms thus far examined, eg. human (Arrigo and Welch, 1987) and rat (Welch, 1985). It would therefore be of interest to see if phosphorylation also occurs in an invertebrate such as C_. elegans. Attempts to detect shifts in any of these spots by reacting the samples with alkaline phosphatase were unsuccessful due to proteolysis during the enzyme incubation. The analysis of 32P-labeled nematode extracts would probably provide conclusive evidence as to whether the minor spots are due to phosphorylation. Since the 2/110-145 antibody detects the hspl6-l/2 pair of proteins strongly and the hsp 16-41/48 pair weakly, it should also distinguish between those minor spots which result from modification of the hsp 16-1/2 pair of proteins and those which result from modification of the hspl6-41/48 pair. Several minor components appear more intense on the Western blot (Fig. 7C2) than on the autoradiogram (7C4), as do the two major 18 kDa components. They are the 18 kDa minor spots and the three acidic 17 kDa spots. The three 16 kDa minor components are detected weakly on the Western compared to the autoradiogram. The 18 kDa and 17 kDa minor spots therefore may be modifications of the hsp 16-1/2 pair of proteins. The four hsp 16 proteins all have molecular masses of 16 kDa based on the amino acid sequences derived from their genes (Russnak and Candido, 1985; Jones et al, 1986). The postulated lower mobility of the hsp 16-1/2 pair (ie. 18 kDa) relative to the hsp 16-41/48 pair (16 kDa) would therefore have to be due to the presence of SDS-stable conformations and/or to postsynthetic modifications in these proteins. E. IMMUNOAFFINITY CHROMATOGRAPHY OF HSP 16 Immunoaffinity chromatography, whereby an antibody is attached to a solid support to remove its antigen from solution, was used in an attempt to purify hsp 16 from heat shocked 53 Results and Discussion nematode extracts. Initial experiments used the hspl6-l/33-50 and hspl6-41/125-143 antibodies linked to cyanogen bromide activated Sepharose. The 1/33-50 column did not bind hspl6. The 41/125-143 column had a low capacity and presumably a low affinity for hspl6 since very little purification of the protein resulted (data not shown). Once the hsp 16-2/110-145 antibody became available it was also tested as an affinity reagent. Fig. 8 shows the results of an experiment in which a heat shocked nematode extract was first fractionated by hydroxylapatite chromatography (HAP) followed by an immunoaffinity column. Bound protein was eluted from the HAP column with 20 mM [20], followed by 35 mM [35] phosphate buffer. The 35 mM phosphate fraction was then applied to the 2/110-145 antibody column, washed and eluted with ammonium thiocyanate [A]. The 16 and 18 kDa peptides are barely detectable in the [35] lane. The [A] lane shows that the antibody column resulted in a substantial purification of the 16 and 18 kDa polypeptides. They are present in considerably greater quantity than most of the other proteins in the [A] lane. Therefore under these conditions the hsp 16-2/110-145 antibody bound the hspl6 complex. The amount of hsp 16 in the [A] lane represents approximately 40% of the total eluant purified from five grams of nematodes. It still contains a considerable amount of non-specifically bound protein indicating that the procedure requires further refinement. F. PROTEIN SEQUENCING OF THE 16 AND 18 KDA HEAT SHOCK POLYPEPTIDES There is good evidence that the 16 and 18 kDa heat shock induced polypeptides, as observed by SDS-gel electrophoresis, are encoded by the hsp 16 genes. This evidence includes the reactivity of the anti-peptide antibodies to these bands. A more direct method of identification would be to directly micro-sequence the N-terminal regions of these polypeptides after separation by electrophoresis and blotting (Aebersold el a].., 1986; 54 Results and Discussion M 20 35 A F Fig. 8 Affinity chromatography of hspl6 with anti-2/110-145 antibody. 5g of french press homogenized heat shocked embryos (2 hr HS, 8 hr rec., 2 hr HS, 4 hr rec.) in Buffer A were fractionated by hydroxylapatite chromatography with 20 mM [20] and 35 mM [35] PO4 elution buffers. The 35 mM PO4 fraction was affinity purified on an anti-2/110-145 antibody-Sepharose column and eluted with 2.4 M ammonium thiocyanate. 40% of this eluate is shown in [A]. [20], [35], and [A] contain approximately 0.5%, 0.5%, and 40% of total fraction volume respectively. [M] are molecular weight protein standards with the sizes (kDa) shown at left. The solid arrowhead marks the 16 kDa polypeptide and the hollow arrowhead marks the 18 kDa polypeptide. 55 Results and Discussion Matsudaira, 1987). To do this, an affinity purification of hspl6 with the hsp 16-2/110-145 antibody column was carried out using a variation of the method described in the previous section. The main differences were that 0.1 mM PMSF was added just prior to homogenization, the hydroxylapatite column was eluted with 10 mM and 50 mM phosphate buffers, and the 50 mM phosphate fraction was eluted from the affinity column with 100 mM glycine HC1, pH 2.5 (data not shown). The sample was separated by electrophoresis on a 20 cm gel along with 125I-labeled cc-lactalbumin as a standard and blotted. After staining with Coomassie blue the 16 and 18 kDa bands and the 125I-labeled a-lactalbumin were excised and subjected to N-terminal micro-sequencing by Dr. Reudi Aebersold of the Biomedical Research Center, UBC. The 125I-labeled a-lactalbumin provided good sequence data. This control indicated that the procedure of gel electrophoresis and blotting had not caused blockage of the N-terminals. The 16 and 18 kDa polypeptides gave no sequence data, indicating that they are likely blocked at their N-termini. This is not surprising since it has been estimated that greater than 75% of the intracellular proteins from eukaryotic cells have blocked N-termini as a result of posttranslational modification (Charbonneau elaL, 1989; Brown and Roberts, 1976). Due to this blockage, the more elaborate method of sequencing internal peptides obtained by enzymatic or chemical cleavage of the hsps would have to be used (Aebersold eial., 1987; Kennedy etal , 1988). This was not attempted as sufficient quantities of affinity purified hsp 16 were not available. G. ADJUSTING THE CONDITIONS OF THE HEAT SHOCK PROTOCOL Following the successful affinity purifications of hsp 16 just described, subsequent attempts to reproduce these results were unsuccessful, ie. little or no purification of hsp 16 was achieved. The majority of the hsp 16 did not bind to the affinity column in these experiments (data not shown). The only apparent major variables between these 56 Results and Discussion experiments were the methods of heat shock and the preparation of the C_. elegans embryos. Maintaining consistency in the heat shock protocol between different batches of nematodes is difficult as both temperature and the degree of aeration must be controlled over an extended period of time, as long as twenty four hours. In retrospect, it is possible that the batch of nematodes from which hsp 16 was successfully affinity purified was heat shocked and aerated under conditions that proved ideal. The procedure of two heat shocks, each followed by a period of recovery was adopted as a way to increase the yield of the hsps (after: Nover et al., 1989). It was done in liquid culture in a shaking water bath as this allowed easy handling of large quantities of embryos and the yield was quantitative. Agar plates were avoided because when the nematodes are starved, as is required to reduce gut protease synthesis, they tend to burrow into the agar making recovery difficult and greatly reducing yield. However, the use of agar plates is suitable for small cultures, and could be preferable as this might avoid the development of anoxic conditions. H. AFFINITY PURIFICATION OF HSP16 As a compromise solution to the problems discussed above, embryos for this experiment were heat shocked once for two hours followed by a seven hour recovery on an agar plate. Burrowing after this period of time was negligible. These nematodes were homogenized and the extract was applied directly to the affinity column without prior fractionation by hydroxylapatite chromatography. The Coomassie blue stained gel in Fig. 9A1 shows that the extract [E] applied to the column and the column flow-through [FI] are essentially the same and that no protein is visible in the affinity column eluant [Al]. The hspl6-2/110-145 antibody developed Western blot of these same fractions (Fig. 9A2) shows that all the hsp 16/18 from the extract [E] 57 Results and Discussion Fig. 9 Affinity chromatography of hspl6 with anti-2/110-145 antibody-protein A-Sepharose in the absence and presence of 4 M urea. Fig. 9A. The extract [E] of 0.8 g of 2 hour heat shock, 7 hour recovered nematodes (17 mg total protein) in Buffer A plus 0.3 mM PMSF with no MgCl 2 was passed over the 2/110-145 affinity column, the flow-through [FI] was collected, and the column was eluted [Al] with 100 mM glycine • HCl pH 2.5. Fig. 9B. The column flow-through from (9A), [FI], was made to 4 M in urea and passed over the same column, equilibrated in the above buffer plus 4 M urea. The flow-through [F2] was collected and the column was eluted [A2] as in Fig. 9A. [E], [FI], and [F2] each contain 1/500 of the total extract volume. [Al] contains 1/16, [A2] contains 1/5, and [A2*] contains 1/3.3 of the total extract volume. Fig. 9A1 and 9B1 are Coomassie blue stained gels, Fig. 9A2 and B2 are Western blots developed as in Fig. 7. [M] is as in Fig. 8. The solid arrowheads mark the 16 kDa polypeptides and the hollow arrowheads mark the 18 kDa polypeptides. 58 Results and Discussion 59 Results and Discussion flowed through the column [FI]. The eluant [Al], which contains a sample volume proportionally thirty times greater than [E] and [FI], has only very faint 16 and 18 kDa bands. These results clearly show that under these conditions, ie. low salt Buffer A, the hspl6/18 peptides were not bound by the anti-hsp 16-2/110-145 antibody. /. AFFINITY PURIFICATION OF HSP 16 IN THE PRESENCE OF UREA The 2/110-145 antibody was directed against the thirty-five C-terminal amino acid residues of hsp 16-2 which represents 25% of the total protein sequence. This region is part of the seventy or so residues which are homologous to the small hsps of many other organisms such as Drosophila melanogaster (Ingolia and Craig, 1982), yeast (Bossier et al., 1989), and to the mammalian a-crystallins (Wistow, 1985). This region of homology is thought to be responsible for the aggregation of small hsp monomers into large complexes (Ingolia and Craig, 1982). Therefore, the C-terminal of the native hsp 16 in the complex may be involved in aggregation such that it is unavailable for binding to the 2/110-145 antibody. This may explain why the 2/110-145 antibody was unsuitable for immunochromatography and immunoprecipitation (data not shown), while it detected hspl6 efficiently on Western blots. The hspl6 on Western blots is likely denatured, thus exposing its C-terminus. It therefore seemed possible that the inclusion of a denaturing agent such as urea in the nematode extracts might allow binding of the complex to the 2/110-145 antibody column. Fig. 9B shows the results of such an experiment. The column flow-through [FI] from Fig. 9A, which contained hspl6 as is evident in Fig. 9A2[F1], was made 4 M in urea. This extract was passed over the 2/110-145 antibody column which had been equilibrated in buffer containing 4 M urea. The column was washed with the same urea containing buffer and then eluted ([F2]) with 100 mM glycine pH 2.5. Fig. 9B1 is the Coomassie blue stained gel. It shows [FI] (the extract applied to the column) and [F2] to be similar to [E] and [FI] in Fig. 60 Results and Discussion 9A1. The eluant from this column ([A2] and a 1.5 times greater loading, [A2*]), however, contains two bands of 16 and 18 kDa which were not found in the eluant of the column without urea. The 66 kDa band in [A2] was an impurity of the SDS sample buffer that is not present in [A2*] which was prepared using fresh sample buffer. The Western blot in Fig. 9B2 of these same fractions shows clearly that the bands visible in [A2*] of the stained SDS gel are in fact 16 and 18 kDa heat shock polypeptides. Only faint bands are seen in the column flow-through [F2], indicating that the binding of hspl6 and 18 polypeptides in a single pass through the column was nearly quantitative. The [A2] lane contains a one hundred-fold larger aliquot of the total sample volume than [FI] or [F2]. This is indicated in part by the much more intense bands in Fig. 9B2[A2] compared to Fig. 9B2[F1]. Even with this much greater loading, no protein bands other than hsp 16 and 18 were visible on the entire Western blot (not shown) or in the stained gel (Fig. 9B1[A2*]). In Fig. 9B2 three bands are visible against a dark background of probable hsp 16 degradation products. The upper band is probably the hsp 16-1/2 pair of polypeptides. The lowest band may contain either the hsp 16-41/48 polypeptides with the middle band being intermediate migrating peptides (see: Fig. 7B4 and associated text), or the lower band may be a degradation product with the middle band being the hsp 16-41/48 pair. In either case it appears that the hspl6-41/48 polypeptides were bound by the hsp 16-2/110-145 antibody, as were the hsp 16-1/2 polypeptides. This binding, however, was conditional upon the inclusion of urea in the extract. The hsp 16 was isolated by this method with high specificity and to high purity. Although the concentration of protein in the eluant was not determined due to a lack of material, the high purity and the high specificity was evident by the lack of any bands visible in [A2*] other than the 16 and 18 kDa components. The presence of urea in the extract appears to have resulted in the exposure of the C-terminal antigenic site of the hspl6s. This finding is consistent with the supposition that this 61 Results and Discussion region is an aggregation domain. On the other hand, the N-terminal regions of the hsp 16 proteins of C_. elegans are quite divergent from each other and are completely dissimilar to those of the small hsps of other organisms and to a-crystallin (Candido ei al, 1989). This suggests that the N-terminal region may not be important for maintenance of the structure of the complex. J. ANALYSIS OF THE HSP 16 COMPLEX BY HYDROXYLAPATITE AND GEL EXCLUSION CHROMATOGRAPHY The small hsps of humans (Arrigo and Welch, 1987) as well as organisms such as chicken (Collier eial-, 1988), Drosophila (Arrigo, 1987), and tomatoes (Nover et al., 1989) have been found to exist as large complexes of from 200 kDa to 800 kDa under normal heat shock conditions and as large as 2 MDa following a severe heat shock (Arrigo et al., 1988). The 16 kDa heat shock proteins of £ . elegans share with these other small hsps and with a-crystallin the region of homology thought to be responsible for aggregation. Therefore, it was expected that hsp 16 might also associate into large multimeric complexes. In order to examine this question, heat shocked £ . elegans extracts were fractionated by gel exclusion chromatography. 1. Purification by Hydroxylapatite Fractionation Prior to gel exclusion chromatography, heat shock extracts were first fractionated by hydroxylapatite chromatography. This was done for several reasons. Firstly, was to clarify the crude extract by removing lipids and other particulates not easily removed by centrifugation or filtration. This was required to prevent clogging of the gel exclusion column. Secondly, was to concentrate the extract to be applied to the gel exclusion column. Thirdly, 62 Results and Discussion M C H 5 0 2 0 0 Fig. 10 Hydroxylapatite chromatography of a heat shocked embryo extract. 3 g of heat shocked [H] embryos were homogenized using a french press and the extract, containing 73 mg of protein, was loaded onto a hydroxylapatite column, washed with Buffer G, and eluted with Buffer G/50 [50] and Buffer G/200 [200]. An extract of non-heat shocked [C] embryos was also prepared. The [H] lane contained 1/1000, and [50] and [200] each contained approximately 1/750 of the total extract volume. Left panel, Coomassie blue stained gel; right panel, Western blot developed as in Fig. 7; [M] is as in Fig. 8, and the solid and hollow arrowheads indicate the positions of hsp 16 and 18, respectively. 63 Results and Discussion was to increase the final purity of the hsp 16 eluted from the gel filtration column. Fig. 10 shows the results of fractionating an extract from 3 g of heat shocked nematodes containing 73 mg of protein. The bound protein was eluted from the column with Buffer G/50 followed by Buffer G/200. The Buffer G/50 and Buffer G/200 fractions contained 7.5 and 23 mg of protein, respectively, which corresponded to approximately ten-fold and three-fold reductions in total protein, respectively. This can be seen in the Coomassie blue stained gel of Fig. 10 where the [50] lane contains fewer and some less intense bands than the [200] lane or the [H] lane, even though the [H] lane contains a 1.25 times lower sample loading. The hsp 16 bands in lane [50] of the Western blot are approximately three-fold more intense than in lane [200]; since the Buffer G/50 fraction contained three-fold less protein, the hsp 16 was approximately nine-fold purer in the Buffer G/50 fraction. The Western blot of Fig. 10 also indicates that the control fraction [C] contained some hsp 16. This suggests that in this experiment the control nematodes, as well as the heat shocked nematodes, were stressed. This stress may have resulted from anoxia caused by insufficient aeration. Improved methods of heat shock to avoid this problem and to prevent losses of hsp 16 due to nematode death are discussed in Section III. G. All of the hsp 16 in the crude extract which was loaded on the column was bound, and previous experiments have shown that no hsp 16 eluted in the Buffer G wash. Assuming that very little hsp 16 remained bound after the Buffer G/200 elution, then the Buffer G/50 fraction contained 75% and the Buffer G/200 fraction contained 25% of the hsp 16 loaded on the column. Since 10-fold and 3-fold less protein was eluted in the Buffer G/50 and Buffer G/200 fractions respectively, hspl6 was purified approximately 7.5-fold (10-fold x 0.75) in the Buffer G/50 fraction and approximately 0.8-fold (3-fold x 0.25), ie. not fractionated, in the Buffer G/200 fraction. Hydroxylapatite chromatography is therefore useful for partial purification of hsp 16, although it results in a loss of approximately 25% under these conditions. Previous 64 Results and Discussion experiments have used lower phosphate concentrations for elution of hsp 16. A concentration of 50 mM phosphate appeared to be the best compromise between purity and yield. 2. Purification and Analysis by Gel Exclusion Chromatography i Fig. 11 shows the results of fractionating various hsp 16 containing nematode extracts by Sephacryl S-300 gel exclusion chromatography. Gel exclusion chromatography fractionates macromolecules essentially according to their size, although other factors such as shape may have an effect. Sephacryl S-300 has a fractionation range of 1 x 104 to 1.5 x 106Mr. Both the Buffer G/50 and Buffer G/200 fractions from the hydroxylapatite column discussed above, as well as a crude extract prepared from the same batch of heat shocked nematodes were fractionated on a 1.6 cm x 95 cm S-300 column. A 2 ml sample of the Buffer G/50 fraction containing 2.4 mg of protein was loaded on the column and 2 ml fractions were collected. The absorbance of each fraction at 280 nm, from the void volume (V0, Fig. 11B, fraction [1], 80.2 ml) to the total column volume (VT, Fig. 11B, approximately fraction [57]), is plotted in Fig. 11B. Fractions represented by solid circles were analyzed by gel electrophoresis in Fig. 11A and by Western blot in Fig. 11C. The first sixteen fractions and every third or fourth fraction thereafter are shown. The Western blot (Fig. 1 IC) indicates that hsp 16 eluted from the column in fractions [6] through [14] with the majority eluting in fractions [8], [9], and [10], and peaking in fraction [9]. There is no hspl6 detectable in fractions [2] through [5]. The small amount in fraction [1] is probably due to an overflow from the [E] lane during loading of the gel. The peak of elution is indicated in both Fig. 11A and B by a bar over these fractions. Inspection of the elution profile shows that hsp 16 elutes before the major protein peak which is centered around fraction [22]. A barely visible band is present in fractions [8-10], 65 Results and Discussion Fig. 11 Gel filtration of heat shocked nematode extracts. The Buffer G/50 hydroxylapatite fraction (Fig. 10) containing 2.4 mg of protein in 2 ml was fractionated on a 1.6 cm x 95 cm S-300 gel filtration column. The absorbance at 280 nm of each 2 ml fraction was measured and is plotted against the the fraction number (B). An A28O of 0.01 is equivalent to approximately 7 |!g/ml of protein. 1/200 volume of each fraction marked by a solid circle in (B) was analyzed by Coomassie blue stained gel electrophoresis (A) and Western blot analysis (C) as in Fig. 7. The Buffer G/200 fraction (7.5 mg protein) from Fig. 10 as well as a crude extract were fractionated on the same gel filtration column and analyzed by Western blot analysis (D) and (E) respectively. The brackets in (A) and (B) indicate the elution peak of hsp 16. The elution positions of standard proteins from the column are shown in kDa above the graph in (B). [E] in (A) and (B) are samples of the Buffer G/50 and in (D) is a sample of the Buffer G/200 fraction from Fig. 10. [M] is as in Fig. 8 and the solid and hollow arrowheads indicate the positions of hspl6 and 18, respectively. 66 Results and Discussion i i M E 1 2 3 4 5 6 7 6 9 1 0 1 1 12 1 3 14 1 5 1 6 1 9 22 25 29 33 37 42 47 52 Fraction Number 4 5 6 7 8 9 10 11 12 13 14 15 16 19 22 25 29 33 37 42 47 52 50 mM 200 mM 5 6 7 8 9 1 0 1 1 1 2 1 3 1 4 1 5 1 6 19 22 25 29 33 37 42 47 51 1 2 3 4 5 6 7 8 9 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 Crude "* ^ m 67 Results and Discussion near the 18 kDa position on the gel. This may represent the 16 or 18 kDa heat shock polypeptides as the Western blot indicates their presence in these lanes. The protein concentration of fraction [9] was found to be 24 p:g/ml. Since the concentration of the Buffer G/50 hydroxylapatite fraction loaded onto the S-300 column was about 1200 Ug/ml, the quantity of protein in fraction [9] has been reduced 50-fold. The majority of hsp 16 originally loaded in a volume of 2 ml was diluted over three fractions of 6 ml. This represents a 3-fold dilution. The hsp 16 pooled from fractions [8], [9], and [10] has therefore been purified approximately 16-fold. The yield of hspl6 would be nearly quantitative if fractions [7] and [11] were included in the pool, but the net purification would then be reduced to 10-fold. The yield from fractions [8-10] was estimated to be 80%. The net purification of the hsp 16 in fractions [8-10] by S-300 gel exclusion was therefore approximately 13-fold. The total purification of hspl6 from the crude extract by both hydroxylapatite and gel exclusion chromatography was approximately 100-fold, and the overall yield was approximately 60%. Samples of the Buffer G/200 fraction from the hydroxylapatite column and of a crude extract were also fractionated on the same S-300 gel exclusion column with identical sample loading and collection. Western blots of various fractions from these experiments are shown in Fig. 11D and E respectively. The blots are consistent with the fractionation of the Buffer G/50 fraction (Fig. 11C) in that the peak of hspl6 elutes in fraction [9]. The A280 elution profiles of these two separations (data not shown), however, indicated that the hsp 16 co-eluted with a larger proportion of the total protein applied to the column. In particular, the hsp 16 from the Buffer G/200 fraction co-eluted with a quantity of protein equivalent to that in the protein peak around fraction [22] of the Buffer G/50 fraction (see: Fig. 1 IB). Comparison of the stained gels indicated that the Buffer G/50 fraction contained less protein of a similar molecular weight to hspl6 complex than the Buffer G/200 fraction. This further indicated the 68 Results and Discussion effectiveness of the use of the Buffer G/50 fraction as a purification step prior to the gel exclusion column. The yield of hsp 16 could be increased in two ways. Firstly, the hydroxylapatite column could be eluted with a buffer of higher phosphate concentration than 50 mM. Secondly, more hsp 16 containing fractions from the gel exclusion column could be pooled. Both of these techniques, however, would result in decreased purity. Likewise, the purity of the isolated hsp 16 could be increased by washing the hydroxylapatite column a second time with a 20 mM to 25 mM phosphate buffer, by lowering the concentration of the elution buffer from 50 mM phosphate, and by being more exclusive in the selection of fractions from the gel exclusion column. This however, would result in a reduced hsp 16 yield. A better alternative would be to use gel exclusion chromatography in combination with immunoaffinity chromatography on the hsp 16-2/110-145 column, in the presence of 4M urea as discussed above. This procedure would also give information about the state of the hsp 16 complex following affinity purification in the presence of urea. A shift in the peak of elution could indicate a change in the complex structure caused either by the urea or by the glycine • HC1 pH 2.5 elution. It might also produce hsp 16 of sufficiently high purity for structural analysis by electron microscopy. 3. Estimation of the Size of the Hspl6 Complex The S-300 gel exclusion column has an effective fractionation range of 1 x 104 to 1.5 x 106 M r . The elution positions of the standard proteins used are marked in Fig. 11B, and the calibration curve is shown in Fig. 12. The void volume of the column, as determined by the elution peak of blue dextran (not shown), is 80.2 ml. The elution peak of hsp 16, fraction [9] in Fig. 1 IB, is 93 ± 1.5 ml. The error of ± 1.5 ml is considered to be a reasonable estimate arrived at by examination of the Western blot (Fig. 11C) and its consistency compared to the 69 Results and Discussion hsp 16 elution peak in Fig. 1 ID and E. This yields an elution volume to void volume ratio (Vg/Vo) of 1.16 + 0.02 and a molecular weight for the hsp 16 complex of 460 ± 55 kDa. This size range of between 400 to 500 kDa is slightly lower than the 500 to 550 kDa estimated for human hsp28 (Arrigo and Welch, 1987). The hsp 16 eluted in the Buffer G/50 fraction from the hydroxylapatite column was separated quite sharply by the gel exclusion column. Hsp 16 was not detectable until fraction [6] and was fully eluted by fraction [14] (Fig. 11C). The majority of the hspl6 eluted within a 6 to 8 ml volume, representing 3 to 4-fold dilution of the original sample. This is similar to the elution profiles observed for the standard proteins (data not shown). The range of elution of hsp 16 therefore probably represents the elution of a complex of a more or less uniform size. Analysis of Fig. 11A indicates that other protein bands also elute over a range of six or so consecutive fractions (for example, the approximately 60 and 70 kDa bands in fractions [10] to [15]). The hspl6 from the crude extract and the Buffer G/200 hydroxylapatite fraction also peak in fraction nine. However, they clearly begin to elute from the gel exclusion column immediately after the void volume in fraction [1] (Fig. HD and E). This suggests that although the hsp 16 complex is predominantly in a form with a size of 400 to 500 kDa, it also exists as much larger complexes, ie. at least 1.5 x 106 kDa based upon the fractionation range of the S-300 column. Phosphorylated proteins are known to bind strongly to hydroxylapatite, probably by coordination of the phosphate group to C a + 2 on the hydroxylapatite column (Gorbunoff, 1990). This suggests the possibility that the large complexes bind more strongly to hydroxylapatite and require elution by 200 mM phosphate due to their phosphorylation The small heat shock proteins in other organisms have also been found to exist as very large complexes. In humans, hsp28 was found in complexes of up to 2 MDa following a severe heat shock (Arrigo eial., 1988) and similar very high molecular weight complexes were also shown to be formed in tomatoes (Nover gl al-» 1989) and chicken (Collier gi ai., 70 Results and Discussion Fig. 12 Calibration curve of the S-300 gel exclusion column. The proteins thyroglobulin (669 kDa), apoferritin (443), (3-amylase (200 kDa), alcohol dehydrogenase (150 kDa), bovine serum albumin (66 kDa), and carbonic anhydrase (29 kDa) were fractionated on the S-300 gel exclusion column under the conditions used for the heat shock extracts (see Fig. 11). The equation above the graph describes the calibration curve where y = molecular weight and x = Ve/Vo- The Ve/Vnof hsp 16 from Fig. 11C (1.16+0.02: approximately fraction [9]) is indicated on the graph and yields a molecular weight for hsp 16 of460±55kDa. 71 Results and Discussion 1988). However, while the human 2 MDa complexes were localized to the nucleus, the chicken and tomato complexes remained in the cytoplasm. Tardieu el si. (1986) hypothesized that the quaternary structure of the a-crystallin complex is a three layer structure with twelve subunits in the first layer and twenty four subunits in the middle and outer layers. The 460 kDa hsp 16 complex would consist of about 32 hspl6 monomers. By this model the hspl6 complex would have 12 subunits in the first layer and 20 or so subunits in the middle and outer layers as its most stable conformation. By comparison, the human hsp28 525 kDa complex (Arrigo and Welch, 1987) would have about 19 hsp28 monomers, ie. 12 monomers in the first layer and only approximately 7 in the second/third layer(s). This would suggest that the size of the complex in the normal cell is not dictated by the number of sites filled, but by some critical, stable mass. The change of that stable, critical mass to a size in excess of 2 MDa upon severe heat stress may be mediated by a change in the conformation of the small hsp or by its phosphorylation. Phosphorylation of the large complex was suggested by its elution from hydroxylapatite in 200 mM phosphate (Section III. /. 3.). A simple test of this hypothesis would be to fractionate a 32P-labeled two hour heat shocked C. elegans extract by S-300 gel exclusion chromatography. Comparing a Western blot of the fractions to its autoradiogram would indicate whether the higher molecular weight aggregates are more heavily 32P-labeled. Although purely conjectural, an additional hypothesis can be drawn based upon these results and results found in the literature as discussed in Section I. B. 4.f. It would be interesting to determine whether, upon phosphorylation, small hsps gained the ability to bind C a + 2 . If this were true and hsps bound C a + 2 with an affinity somewhat lower than say calmodulin, then this would immediately suggest an interesting mechanism for protection of cells against heat induced surges in cytoplamic C a + 2 . For example, a C a + 2 surge might lead 72 Results and Discussion to the phosphorylation of the small hsps which could then sequester excess C a + 2 until recovery again reduced levels of C a + 2 to normal levels. 73 References IV. REFERENCES 1. Aebersold, R.H., D.B. Teplow, L.E. Hood and S.B.H. Kent. 1986. J. Biol. Chem. 261: 4229-4238. 2. Aebersold, R.H., J. Leavitt, R. Saavedra, L.E. Hood and S.B.H. Kent. 1987. Proc. Nad. Acad. Sci. USA $4: 6970-6974. 3. Ahmad, S., R. Ahiya, T.J. Venner and R.S. Gupta. 1990. Mol. Cell. Biol. 1£: 5160-5165. 4. Arrigo, A.-P. 1987. Devi. Biol. 122: 39-48. 5. Arrigo, A.-P. 1990a. J. Cell Sci. 9_6_: 419-427. 6. Arrigo, A.-P. 1990b. Mol. Cell. Biol. IQ: 1276-1280. 7. Arrigo, A.-P. and D. Pauli. 1988. Expl. Cell Res. 125_: 169-183. 8. Arrigo, A.-P. and WJ. Welch. 1987. J. Biol. Chem. 262: 15359-15369. 9. Arrigo, A.-P., J.-L. Darlix, E.W. Khandjian, M. Simon and P. -F. Spahr. 1985. EMBO J. 4: 399-406. 10. Arrigo, A.-P., J.P. Suhan and WJ. Welch. 1988. Mol. Cell. Biol. & 5059-5071. 11. Arrigo, A.-P., K. Tanaka, A.L. Goldberg and W.J. Welch. 1988. Nature 221: 192-194. 12. Arrigo, A.-P., M. Simon, J.-L. Darlix and P.-F Spahr. 1987. J. Mol. Evol. 25_: 141-150. 13. Ayme, A. and A. Tissieres. 1985. EMBO J. 4: 2949-2054. 14. Bardwell, J.C. and E.A. Craig. 1984. Proc. Natl. Acad. Sci. USA 8J.: 848-852. 15. Bamier, J.V., O. Benosaude, M. Morange and C. Babinet. 1987, Exp. Cell Res. 170: 186-191. 16. Beaulieu, J.-F, A.-P. Arrigo and R.M. Tanguay. 1989. J. Cell. Sci. 22: 29-36. 17. Beckmann, R.P., L.E. Mizzen and W J . Welch. 1990. Science 248.: 850-854. 18. Bensaude, O. and M. Morange. 1983. EMBO J. 2: 173-177. 19. Bensuade, O., C. Babinet, M. Morange and F. Jacob. 1983. Nature 2Q5_: 331-333. 20. Berger, E.M. and M.P. Woodward. 1983. Exp. Cell Res. 142: 437-442. 21. Bloemendal, H. 1977. Science 197: 127-138. 22. Bossier, P., I.T. Fitch, H. Boucherie and M.F. Tuite. 1989. Gene 7_8_: 323-330. 23. Bournais-Vardiabasis, N., C. Buzin and J. Flores. 1990. Exp. Cell Res. 139_: 177-182. 74 References 24. Brenner, S. 1974. Genetics 22: 71-94. 25. Bretscher, M.S. and A.E. Smith. 1972. Anal. Bioc. 47_: 310-312. 26. Brown J.L. and W.K. Roberts. 1976. J. Biol. Chem. 25J.: 1009-1016. 27. Byerly, L.C., C. Cassada and R L . Russell. 1976. Devi. Biol. 5J.: 23-33. 28. Candido, E.P.M., D. Jones, D.K. Dixon, R.W. Graham, R.H. Russnak and R.J. Kay. 1989. Genome 21:690-697. * 29. Charbonneau, H. 1989. pp. 15-30. in A Practical Guide to Protein and Peptide Purification for  Microsequencing. P. Matsudaira (ed.), Academic Press Inc., San Diego, California. 30. Cheng, M.Y., F.-U. Hartl, J. Martin, R.A. Pollock, F. Kalousek, W. Neupert, E.M. Hallberg, R.L. Hallberg and A.L. Horwich. 1989. Nature 217: 620-625. 31. Choppell, T.G., W.J. Welch, D.M. Schlossman, K.B. Palter, M.J. Schlesinger and J.E. Rothman. 1986. Cell 45_: 3-13. 32. Chretien, P. and J. Landry. 1988. J. Cell. Phys. 127: 157-166. 33. Clark-Lewis, I., R.H. Aebersold, H.J. Ziltener, J.W. Schrader, L.E. Hood and S.B.H. Kent. 1986. Science 231: 134-139. 34. Clos, J., J.T. Westwood, P.B. Becker, S. Wilson, K. Lambert and C. Wu. 1990. Cell 62: 1085-1097. 35. Cohen, R.S. and M. Methelson. 1985. Cell 42: 737-746. 36. Collier, N.C., J. Heuser, M.A. Levy and M J . Schlesinger.1988. J. Cell Biol. 106: 1131-1139. 37. Corces, V., A. Pellicer, R. Axel and M. Meselson. 1981. Proc. Natl. Acad. Sci. USA 7_8_: 7038-7042. 38. Coulson, A., J.E. Sulston, S. Brenner and J. Karn. 1986. Proc. Natl. Acad. Sci. USA 82: 7821-7825. 39. Craig, E.A. and K. Jacobsen. 1984. Cell 2& 841-849. 40. Crete, P. and J. Landry. 1990. Radiat. Res. 121: 320-327. 41. Darbon, J.M., M. Issandou, J.F. Tournier and F. Bayard. 1990. Bioch. Biophys. Res. Commun. 1M: 527-536. 42. DeLuca-Flaherty, C , D.B. McKay, P. Parham and B.L. Hill. 1990. Cell 6.2: 875-887. 43. DeMarzo, A.M., C.A. Beck, S.A. Ofiate and D.P. Edwards. 1991. Proc. Natl. Acad. Sci. USA 8_&: 72-76. 44. Deshaies, RJ. , B.D. Koch, M. Werner-Washburne, E.A. Craig and R. Schekman. 1988. Nature 332: 800-805. 75 References 45. Dixon, D.K., D. Jones and E.P.M. Candido. 1990. DNA Cell Biol. 9_: 177-191. 46. Driscoll, J. and A.L. Goldberg. 1990. J. Biol. Chem. 2£5_: 4789-4792. 47. Emmons, S.W. 1987. Cell 5J_: 881-883. 48. Emmons, S.W. 1988. pp. 47-81. in The Nematode Caenorhabditis elegans. W.B. Wood (ed.), Cold Spring Harbor Laboratory, Cold Spring Harbor, New York. 49. Emmons, S.W., M.R. Klass and D. Hirsh. 1979. Proc. Natl. Acad. Sci. USA 76.: 1333-1337. 50. Evans, D.P., R.A. Simonette, CD. Rasmussen, A.R. Means and S.P. Tomasovic. 1990. J. Cell. Phys. 142: 615-627. 51. Falkenburg, P.-E., C. Haass, P.-M. Kloetzel, B. Niedel, F. Kopp, L. Kuehn and B. Dahlmann. 1988. Nature 3J3JL: 190-192. 52. Farrelly, F.W. and D.B. Finkelstein. 1984. J. Biol. Chem. 25J?: 5745-5751. 53. Gaitanaris, G.A., A.G. Papavassiliou, P. Rubock, S.J. Silverstein and M.E. Gottesman. 1990. Cell £1: 1031-1020. 54. Goldenberg, C.J., Y. Luo, M. Fenna, R. Baler, R. Weinmann and R. Voellmy. 1988. J. Biol. Chem. 263: 19734-19739. 55. Gorbunoff, M.J. 1990. Methods in Enzymology 182: 329-339. 56. Hackett, J.R. and J.T. Lis. 1983. Nucleic Acids Res. H.: 7011-7030. 57. Harlow E. and D.P. Lane. 1988. Antibodies: a laboratory manual. Cold Spring Harbor Laboratory, / Cold Spring Harbor, NY. 58. Hard, F.-U. and W. Neupert. 1990. Science 247: 930-938. 59. Heikkila, JJ. and G.A. Schultz. 1984. Gamete Res. 1£: 45-56. 60. Heikkila, J.J., M. Kloc, J. Bury, G.A. Schultz and L.W. Browder. 1985. Dev. Biol. 107: 493-489. 61. Hickey, E., S.E. Brandon, R. Potter, G. Stein, J. Stein and L.A. Weber. 1986. Nucleic Acids Res. 14: 4127-4145. 62. Housley, P.R., E.R. Sanchez, M. Danielsen, G.M. Ringold and W.B. Pratt. 1990: J. Biol. Chem. 265: 12778-12781. 63. Hunt, C. and R. Morimoto. 1985. Proc. Natl. Acad. Sci. USA 82: 6455-6459. 64. Ingolia, T.D. and E.A. Craig. 1982. Proc. Natl. Acad. Sci. USA 7_9_: 2360-2364. 65. Ingolia, T.D., E.A. Craig and B.J. McCarthy. 1980. Cell 21: 669-679. 76 References 66. Ingolia, T.D., M.R. Slater and E.A. Craig. 1982. Mol. Cell. Biol. 2: 1388-1398. 67. Ireland, R.C. and E.M. Berger. 1982. Proc. Natl. Acad. Sci. USA 29_: 855-859. 68. Johnston, R.N. and B.L. Kucey. 1988. Science 242: 1551-1554. 69. Jones, D., D.K. Dixon, R.W. Graham and E.P.M. Candido. 1989. DNA 8_: 481-490. 70. Jones, D., R.H. Russnak, RJ. Kay, E.P.M. Candido. 1986. J. Biol. Chem. 26J_: 12006-12015. 71. Kang, P.-J., J. Ostermann, J. Schilling, W. Neupert, E.A. Craig and N. Pfanner. 1990. Nature 348.: 137-143. 72. Kaur, P., W.J. Welch and J. Saklatvala. 1989. FEBS Lett. 25J: 269-273. 73. Kay, R.J., R.H. Russnak, D. Jones, C. Mathias and E.P.M. Candido. 1987. Nucleic Acids Res. 15_: 3723-3741. 74. Kay, R.J., R.J. Boissy, R.H. Russnak and E.P.M. Candido. 1986. Mol. Cell. Biol. £: 3134-3143. 75. Kelly, P. and MJ. Schlesinger. 1978. Cell I5_: 1277-1286. 76. Kennedy, T.E., M.A. Gawinowicz, A. Barzilai, E.R. Kandel and J.D. Sweatt. 1988. Proc. Natl. Acad. Sci. USA g5_: 7008-7016. 77. Kim, Y.-J., J. Shuman, M. Sette and A. Przybyla. 1984. Mol. Cell. Biol. 4 : 468-474. 78. Klemenz, R., E. FrOhli, R.H. Steiger, R. Schafer and A. Aoyama. 1991. Proc. Natl. Acad. Sci. USA SS: 3652-3656. 79. Kloetzel, P-M. and E.K.F. Bautz. 1983. EMBO J. 2: 705-710. 80. Kothary, R.K. and E.P.M. Candido. 1982. Can. J. Biochem. £0: 347-355. 81. Kurtz, S., J. Rossi, L. Petko and S. Lindquist. 1986. Science 231.: 1154-1157. 82. Kyte, J. and R.F. Doolittle. 1982. J. Mol. Biol. 157: 105-132. 83. Laemmli, U.D. 1970. Nature 277: 680-685. 84. Lamarche, S., P. Chr6tien and J. Landry. 1985. Bioch. Biophys. Res. Commun. HI: 868-876. 85. Landry, J., P. Chr6tien, H. Lambert, E. Hickey and L.A. Weber. 1989. J. Cell Biol. JJ22: 7-15. 86. Landry, J., P. Crete, S. Lamarche and P. Chretien. 1988. Radiat. Res. I H : 426-436. 87. Lawson, R., R. Menstril, Y. Luo and R. Voellmy. 1985. Dev. Biol. UQ_: 321-330. 88. Leicht, B.G., H. Biessmann, K.B. Palter and JJ. Bonner. 1986. Proc. Natl. Acad. Sci. USA SI : 90-94. 77 References 89. Lewis, M.J. and H.R.B. Pelham. 1985. E M B O J. 4: 3137-3143. 90. L i , G . C . and Z. Werb. 1982. Proc. Nad. Acad. Sci. U S A 29: 3218-3222. 91. L i , G . C , D .C. Shrieve and Z. Werb. 1982. pp. 395-404. in Heat Shock: from Bacteria to Man. M.J. Schlesinger, M . Ashburner and A. Tissieres (eds.), Cold Spring Harbor Laboratory, Cold Spring Harbor, N Y . 92. L i , G . L . , L . L i , Y . - K . Liu, J.Y. Mak, L . Chen and W . M . F. Lee. 1991. Proc. Natl. Acad. Sci. USA 8&: 1681-1685. 93. Lindquist, S. 1986. Annu. Rev. Biochem. 55_: 1151-1191. 94. Loomis, W.F . and S.A. Wheeler. 1982. Devi. Biol. 20_: 412-418. 95. Lowe, D . G . and L . A . Moran. 1984. Proc. Natl. Acad. Sci. U S A E l : 2317-2321. 96. Matsudaira, P. 1987. J. Biol. Chem. 262: 10035-10042. 97. McGarry, I.T. and S. Lindquist 1985. Cell 42: 903-911. 98. Miller, M J . , N - H . Xuong and E.P. Geiduschek. 1979. Proc. Nad. Acad. Sci. U S A 2£:5222-5225. 99. Mirault, M . - E . , M . Goldschmidt-Clermont, L . Moran, A.P. Arrigo and A. Tissieres. 1978. Coldspring Harbor Symp. Quant. Biol. 42: 819-827. 100. Mizzen, L . A . and W J . Welch. 1988. J. Cell Biol. IM : 1105-1116. 101. Moran, L . , M . Chauvin, M . E . Kennedy, M . Korri, D .G. Lowe, R.C. Nicholson and M . D . Perry. 1983. Can. J. Biochem. Cell Biol. 6J.: 488-499. 102. Morganelli, C M . , E . M . Berger and H.R.B. Pelham. 1985. Proc. Natl. Acad. Sci. U S A 82: 5865-5869. 103. Mosser, D.D. , P.T. Kotzbauer, K .D. Sarge and R.I. Morimoto. 1990. Proc. Natl. Acad. Sci. USA 32: 3748-3752. 104. Munro, S. and H.R.B. Pelham. 1986. Cell 46: 291-300. 105. Munro, S. and H.R.B. Pelham. 1987. Cell 48: 899-907. 106. Neupert, W. , F . - U . Hard, E . A . Craig and N . Pfanner. 1990. Cell 62: 447-450. 107. Normington, K . , K. Kohno, Y . Kozotsumi, M. -J . Gething and J. Sambrook. 1989. Cell 5J7: 1223-1236). 108. Nover, L . (ed.). 1990. Heat Shock Response . C R C Press. Boca Raton, F L . 109. Nover, L . , K . -D . Scharf and D. Neumann. 1989. Mol. Cell. Biol. 9_: 1298-1308. 78 References 110. O'Farrell, P.H. 1974. J. Biol. Chem. 25J2: 4007-4021. 111. Ohno, K., M. Fukushima, M. Fujiwara and S. Narumiya. 1988. J. Biol. Chem. 26J1: 19764-19770. 112. Osterman, J., A.L. Horwich, W. Neupert and F.-U. Hard. 1989. Nature 241: 125-130. 113. Ovsenek, N. and JJ. Heikkila. 1990. Development HO: 427-433). 114. Palter, K.B., M. Watanabe, L. Stinson, A.P. Mahowald and E.A. Craig. 1986. Mol. Cell. Biol. 6_: 1187-1203. 115. Patriarca, E.J. and B. Maresca. 1990. Exp. Cell Res. 191): 57-64. 116. Pelham, H.R.B. 1982. Cell 3J): 517-528. 117. Pelham, H.R.B. and M. Bienz. 1982. EMBO J. 1: 1473-1477. 118. Piatigorsky, J. and G. Wistow. 1991. Science 252: 1078-1079. 119. Plesofsky-Vig, N. and R. Brambl. 1990. J. Biol. Chem. 265.: 15432-15440. 120. Pratt, W.B. 1990. Mol. Cell. Endocrinol. 74: C76-C76. 121. Regazzi, R., U. Eppenberger and D. Fabbro. 1988. Bioch. Biophys. Res. Commun. 152: 62-68. 122. Riabowol, K.T., L.A. Mizzen and WJ. Welch. 1988. Science 242: 433-436. 123. Ritossa, F. 1962. Experimentia 18_: 571-573. 124. Rose, M.D., L .M. Misra and J.P. Vogel. 1989. Cell 52: 1211-1221. 125. Rossi, J.M. and S. Lindquist. 1989. J. Cell Biol. Ifj8_: 425-439. 126. Russnak, R.H. and E.P.M. Candido. 1985. Mol. Cell. Biol. 5_: 1268-1278. 127. Russnak, R.H., D. Jones and E. P.M. Candido. 1983. Nucleic Acids Res. 11:3187-3205. 128. Sanchez, E.R. 1990. J. Biol. Chem. 2£5_; 22067-22070. 129. Sanchez, Y. and S.L. Lindquist 1990. Science 248_: 1112-1115. 130. Sarkis, G.J., J.D. Ashcom, J.M. Hawdon and L.A. Jacobson. 1988b. Mech. Age. Devi. 45_: 191-201. 131. Sarkis, G.J., M.R. Kurpiewski, J.D. Ashcom, L. Jen-Jacobson and L.A. Jacobson. 1988a. Arch. Bioch. Biophys. 26_L 80-90. 132. Scharf, K.-D., S. Rose, W. Zott, F. Schoff and L. Nover. 1990. EMBO J. 2: 4495-4501. 133. Scherer, P.E., U.C. Krieg, S.T. Hwang, D. Vestweber and G. Schatz. 1990. EMBO J. 2: 4315-4322. 79 References 134. Schlesinger, MJ., M. Ashburner and A. Tissieres (eds.). 1982. Heat Shock: from Bacteria to Man. Cold Spring Harbor Laboratory, Cold Spring Harbor, NY. 135. SchoTfl, F. and J.C. Key. 1982. J. Mol. Appl. Genet. 1: 301-314. 136. Schuldt, C. and P-M. Kloetzel. 1985. Dev. Biol. UH: 65-74. 137. Sheffield, W.P., G.C. Shore and S.K. Randall. 1990. J. Biol. Chem. 2£5_: 11069-11076. 138. Sirotkin, K. and N. Davidson. 1982. Dev. Biol. 89_: 196-210. 139. Skowyra, D., C. Georgopoulos and M. Zylicz. 1990. Cell £2: 939-944. 140. Slater, A., A.C.B. Cato, G. M. Sillar, J. Kioussis and R.H. Burdon. 1981. Eur. J. Biochem. 117: 341-346. 141. Snutch, T.P. and D.L. Baillie. 1983. Can. J. Biochem. Cell Biol. 6J.: 480-487. 142. Sorger, P.K. and H.R.B. Pelham. 1987. EMBO J. 6: 3035-3041. 143. Sorger, P.K.and H.R.B. Pelham. 1988. Cell 54.: 855-864. 144. Southgate, R., A. Ayme and R. Voellmy. 1983. J Mol. Biol. 16J>: 35-57. 145. Stevenson, M.A., S.K. Calderwood and G.M. Hahn. 1987. Cancer Res. 47: 3712-3717. 146. Storti, R.V., M.P. Scott, A. Rich and M.L. Pardue. 1980. Cell 22: 825-834. 147. SubjeckJ.R., T. Shy, J. Shen and R.J. Johnson. 1983. J. Cell Biol. 9J7: 1389-1395. 148. Sulston, J.E. and S. Brenner. 1974. Genetics 77: 95-104. 149. Sulston, J.E., E. Schierenberg and J.G. White. 1983. Devi. Biol. 100: 64-119. 150. Taketani, S., H. Hogno, T. Yoshinaga and R. Tokunaga. 1989. FEBS Lett. 245: 173-176. 151. Tardieu, A., D. Laporte, P. Licinio, B. Krop and M. Delaye. 1986. J. Mol. Biol. 192: 711-724. 152. Ting, L.-P., C.-L. Tu and C.-K. Chou. 1989. J. Biol. Chem. 264: 3304-3408. 153. Tissieres, A., H.K. Mitchell and U.M. Tracey. 1974. J. Mol. Biol. 34: 389-398. 154. Towbin, H., T. Staehelin and J. Gordon. 1979. Proc. Natl. Acad. Sci. USA 76: 4350-4354. 155. Vincent, M. and R.M. Tanguay. 1982. J. Mol. Biol. 1^ 2: 365-378. 156. Welch, WJ. 1985. J. Biol. Chem. 2£0_: 3058-3062. 157. Welch, WJ. and J.R. Feramisco. 1984. J. Biol. Chem. 252: 4501-4511. 80 References 158. Welch, W.J. and L.A. Mizzen. 1988. J. Cell. Biol. IQ&: 1117-1130. 159. Widelitz, R.B., B.E. Magun and E.W. Gerner. 1986. Mol. Cell. Biol. 6_: 1088-1094. 160. Wistow, G. 1985. FEBS Lett. JM: 1-6. 161. Wistow, G. and J. Piatigorsky. 1987. Science 22£: 1554-1556. 162. Wu, C. 1984. Nature 2Q9_: 229-234. 163. Wu, C , S. Wilson, B. Walker, I. Dawid, T. Paisley, V. Zimarino and H. Ueda. 1987. Science 22&: 1247-1253. 164. Yamamori, T., K. Ito, Y. Nakamura and T. Yura. 1978. J. Bacteriology 124:1133-1140. 165. Yost, H.J. and S. Lindquist. 1986. Cell 45_: 185-193. 166. Zimarino, V. and C. Wu. 1987. Nature 221: 727-730. 167. Zimmerman, J.L., W. Petri and M. Meselson. 1983. Cell 22: 1161-1170. 81 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



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"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items