Functional analysis of a small heat shock/a-crystallin protein from Artemia franciscana Oligomerization and thermotolerance Julie A. Crack, Marc Mansour, Yu Sun and Thomas H. MacRae Department of Biology, Dalhousie University, Halifax, Nova Scotia, Canada Oviparously developing embryos of the brine shrimp, Artemia franciscana, synthesize abundant quantities of a small heat shock/a-crystallin protein, termed p26. Wild-typ e p26 functions as a molecular chaperone in vitro and is thought to help encysted Artemia embryos survive severe physiological stress encountered during diapause and anoxia. Full-length and truncated p26 cDNA derivatives were generated by PCR amplification of p26-3-6-3, then cloned in either pET21(+) or pRSETC and expressed in Escherichia c oli BL21 (DE3). A ll constr ucts gave a polypep- tide detectable on Western blots with either p26 specific antibody, o r w ith antibody to the H is 6 epitope tag encod ed by pRSETC. Full-length p26 in cell-free extracts of E. coli was a bout equal i n mass to t hat f ound in Artemia embryos, but p26 lacking N- and C-terminal r esidues remained e ither as monomers or small multimers. All p26 constructs conferred thermotolerance on transformed E. coli, although not all formed oligomers, and cells expressing N-terminal truncated derivatives of p26 were more heat resistant than bacteria expressing p26 with C-terminal d eletions. The C-terminal extension o f p26 is seemingly more i mportant for thermotolerance than is the N-terminus, and p26 protects E. coli against heat shock when oligomer size and protein concentration are low. The findings have important impli- cations for understanding the functional mechanisms of small h eat s hock/a-crystallin proteins. Keywords: small heat shock/a-crystallin protein; oligomeri- zation; thermotolerance; diapause; Artemia f ranciscana. Cells respond to stress by the enhanced synthesis of heat shock or stress p roteins, which are also developmentally regulated under normal physiological conditions. Stress proteins are divided into several families o n the basis of s ize and a mino-acid s equence [ 1–5]. M oreover, they function as molecular c haperones, facilitating proper folding, transport and multimerization of nascent proteins, as well as preventing the irreversible aggregation of denaturing proteins. The small heat shock/a-crystallin proteins consti- tute a structurally divergent, ubiquitous group within the chaperone superfamily, ranging in mole cular m ass f rom 12 to 43 kDa [6]. A conserved region, termed the a-crystallin domain, distinguishes a ll small heat shock/a-crystallin proteins, and a two or three domain structure is proposed for these proteins [7,8]. The a-crystallin domain, located toward the C-terminus of the protein monomer, consists of 80–100 amino-acid residues and is important for oligomer formation and chaperoning [9–13]. F lexible C-terminal extensions of small heat shock/a-crystallin proteins, enriched in polar and charged amino-acid residues, vary in length and sequence [8,14,15]. Loss or modification of the C-terminal extension has the potential to perturb function and reduce solubility of these proteins and their complexes with target proteins [15–19]. The N-terminus, which may be partly buried within the mature protein, promotes oligomer formation, subunit exchange, and capture of unfolding proteins [12,18,20–26]. Small heat shock/a-crystallin proteins confer thermotol- erance upon cells [27–33], protect against apoptotic death [34,35] and have chaperone activity in vitro, wherein the aggregation of client proteins is prevented [36–38]. Chap- eroning is thought to depend upon formation of oligomers that reach 800 kDa in mass and possess quaternary structure modifiable by environmental parameters [8,18,20,22,39,40]. Oligomers exhibit dynamic equilibrium with constituent subunits, which can affect chaperoning but is not in itself sufficient to ensure chaperone activity [25,41,42]. A small h eat shock/a-crystallin protein from Methanococcus jannaschii,termedMjhspl6.5,hasbeen crystallized, revealing highly o rdered oligomers of 24 subunits with a hollow center [9]. Cryoelectron microscopy of small heat shock/a-crystallin proteins from several sources has shown, however, that oligomer structure ranges from well d efined to variable, leading to the idea that structural plasticity elicits low specificity and permits binding of different target proteins [10,24,42]. Several molecules of denaturing proteins, present in an unstable molten globule state, i nteract with a single oligomer when chaperoning occurs. The proteins are protected from irreversible aggregation under stress, their activity may be preserved, and they either refold s pontaneously o r with the assistance of other chaperones upon release [38,43–46]. Embryos of the brine shrimp, Artemia franciscana, develop ovoviviparously, leading to release o f swimming Correspondence to T. H. MacRae, Department of Biology, Dalhousie University, Halifax, Nova Scotia, B3H 4J1, Canada. Fax: + 902 494 3736, Tel.: + 902 494 6525, E-mail: tmacrae@i s.dal.ca Abbreviations:Gp4G,guanosine5¢-tetraphospho-5¢-guanosine; IPTG, isopropyl thio-b- D -galactoside; HRP, horseradish peroxidase. (Received 12 October 2001, revised 3 December 2001, accepted 5 December 2001) Eur. J. Biochem. 269, 933–942 (2002) Ó FEBS 2002 nauplii from females. Alternatively, oviparous development occurs, embryos encyst as gastrulae composed of about 4000 cells and are discharged from females enclosed in a shell permeable only to volatile molecules [47–49]. Subse- quent to release, encysted embryos enter a dormant state known as diapause [50], wherein metabolic activity is difficult to detect [51,52]. Diapause continues, even under favourable growth conditions, until the appropriate activa- tion signal. The embryos are very tolerant of physical and chemical insults such as exposure to organic solvents, a-irradiation, temperature extremes and desiccation, the latter probably a cue that terminates diapause [53]. As one remarkable example of stress resistance, fully hydrated cysts survive several years at physiological temperature in the complete absence o f o xygen [48,51,52,54], an unusual degree of tolerance for any animal. These observations contradict the general belief that under ordinary hydration and temperature, cell maintenance entails a constant and substantial free energy flow [51,52]. Anoxic cysts m ay acquire s ufficient energy to survive b y utilization of guanosine 5¢-tetraphospho-5¢-guanosine (Gp4G), an abun- dant nucleotide at this developmental stage [55]. Previous work has revealed p26, a small heat shock/ a-crystallin protein found o nly in Artemia undergoing oviparous development [47–49,56–59]. p26 has chaperone activity in vitro and i mparts thermotolerance to trans- formed bacteria [49,57]. Although chaperoning and ther- motolerance are not necessarily equivalent activities, the results indicate that p26 prevents irreversible denaturation of proteins in diapause/encysted Artemia embryos. This permits spontaneous and/or assisted refolding of proteins, the former a llowing rapid resumption of d evelopment under limiting e nergy r eserves, perhaps t o e xploit the transient occurrence of favourable environmental condi- tions encountered by Artemia. In the current study, functions of p26 N- and C-terminal regions were explored through deletion mutagenesis. Specifically, protein solubil- ity, oligomerization, and t he thermotolerance of t rans- formed bacteria were examined. Such information may illuminate the mechanism by which p26 protects Artemia from physiological stress experien ced during diapause and anoxia, thereby enhancing our appreciation of small heat shock/ a-crystallin proteins. EXPERIMENTAL PROCEDURES Cloning of full-length and truncated p26 cDNAs Full-length and truncated p26 cDNAs were generated by PCR using p26-3-6-3 cDNA (GenBank accession no. AF031367) [58] as template, and custom primers possessing BamHI and XhoI restriction sites on the sense and antisense oligoneucleotides, r espectively (CyberSyn, Inc., Lenni, PA, USA) (Table 1). Fifty microliter PCR mixtures included 38 n g of template DNA, 0.01 lgÁmL )1 each of sense and antisense primers, 5 lL of PCR buffer (ID Laboratories, London, Ontario, Canada), 1 m M dNTP, 5 0 m M Mg 2+ , 40 lLofH 2 O and 0.01 U of proof-reading Taq polymerase (ID Laboratories). Reaction mixtures, covered with mineral oil, were incubated for 2 min at 94 °C prior to five cycles of 30 s at 94 °C, 45 s at 40 °C, 30 s at 72 °C , then 30 cycles of 30 s at 94 °C, 30 s at 55 °Cand30sat72°C, followed by 10 min at 72 °C. PCR products were analyzed in 1.0% agarose gels in TAE buffer (0.04 M Tris HC1, 0.02 M glacial acetic acid, 0.001 M EDTA) using 100-bp standards (Amersham-Pharmacia Biotech or BioÁRad). DNA frag- ments of appropriate length were ligated into the T /A vector, pCRII (Invitrogen, San Diego, CA, USA), using T4 DNA ligase overnight at 14 °C, and E. coli DH5a made competent by the calcium chloride procedure were trans- formed with the recombinant DNA [60]. Putative p26 cDNA containing clones were s elected by blue/white screening using the LacZ system, propagated in LB broth, and examined by restriction analysis for plasmids incorpo- rating inserts of t he appropriate size, which were subcloned into the prokaryotic expression vector pET21(+) (Nov- agen, Inc., Madison, WI, USA). Briefly, pCRII constructs and pET21(+) were d igested with BamHI and XhoIbefore electrophoresis in 1% agarose gels. Linearized pET21(+) and p26 cDNAs were excised and purified with the GFX TM PCR DNA and Gel Band Purification Kit (Amersham- Pharmacia Biotech). Each p26 cDNA was ligated into pET21(+) using T4 DNA ligase, and competent E. coli DH5a were transformed w ith the constructs [60]. Bacteria containing p26 cDNA of the correct length were identified by restriction digestion of constituent plasmids followed by electrophoresis in 1% agarose gels. The p26 cDNAs were Table 1. Full-length and truncated p26 cDNAs generated by PCR. The p rimers are listed in the 5 ¢fi3¢ direction and restriction sites are underlined. ATG, start codon; TTA, termination codon; bp, base pair. p26 cDNAs Amino acid residues deleted Designations Primer sequences Length (bp/amino acids) p26-full None (p26-1Bam-s) GCGCGGATCCACCATGGCACTTAACCCATG 576/192 (p26-192Xho-as) CGCGCCTCGAGTTAAGCTGCACCTCCTGATCT p26-ND36 1–36 (p26-36Bam-s) GCGCGGATCCACCATGCCCTTCCGGAGAAGA 468/156 (p26-192Xho-as) CGCGCCTCGAGTTAAGCTGCACCTCCTGATCT p26-ND60 1–60 (p26-60Bam-s) GCGCGGATCCACCATGTCCTTGAGGGACACA 396/132 (p26-192Xho-as) CGCGCCTCGAGTTAAGCTGCACCTCCTGATCT p26-CD40 153–192 (p26-1Bam-s) GCGCGGATCCACCATGGCACTTAACCCATG 459/153 (p26-153Xho-as) CGCGCCTCGAGTTAACGTTCTGTTGGTGAGCT p26-CD10 183–192 (p26-1Bam-s) GCGCGGATCCACCATGGCACTTAACCCATG 546/182 (P26-182 Xho-as) CGCGCCTCGAGTTATGGAGTTGAACTAGCTGT p26-alpha 1–60 and (p26-60Bam-s) GCGCGGATCCACCATGTCCTTGAGGGACACA 297/93 153–192 (p26-153Xho-as) CGCGCCTCGAGTTAACGTTCTGTTGGTGAGCT 934 J. A. Crack et al.(Eur. J. Biochem. 269) Ó FEBS 2002 sequenced in both directions, either at the Hospital for Sick Children (Toronto, Ontario, Canada) or t he National Research Council Laboratory (Halifax, Nova Scotia, Canada). Sequence similarity amongst the p26 constructs was analyzed by CLUSTAL W , with files viewed and edited in Microsoft WORD . Selected p26 cDNA fragments recovered from pET21(+) constructs were also cloned into pRSETC (Invitrogen). Full-length p26 cDNA cloned in pRSETC, termed pRSET-p26-3-6-3, was prepared previously [49]. Expression of p26 in E. coli BL21(DE3) Two ml of LB medium supplemented with 50 lgÁmL )1 ampicillin was inoculated with a single colony consisting of transformed bacteria possessing either full-length or trun- cated p26 cDNA, and incubated with shaking at 37 °C until the D 600 reached 0.6–1.0. Cultures w ere stored a t 4 °C overnight before incubation with shaking in 50 mL of f resh LB medium containing 50 lgÁmL )1 ampicillin at either 30 °Cor37°C. Isopropyl t hio-b- D -galactoside (IPTG) wa s added w hen the culture reached a D 600 of 0.6–1.0 a nd incubation continued for 5 h, followed by 5 min on ice and centrifugation at 5000 g for 5 min at 4 °C. Growth rates of E. coli transformed with wild-typ e and mutated p26 were not determined, but the D 600 increases for all cultures were similar indicating that expressed p26 had no effect on cell division. The pelleted cells were washed twice with cold buffer (50 m M Tris/HC1, 2 m M EDTA, pH 8.0) and resuspended in 5 mL of the same buffer, before adding lysozyme and Triton X -100 to final concentrations of 100 lgÁmL )1 and 0.01%, respectively. Triton X-100 was omitted from s ome preparations to learn if detergent influenced p26 oligomerization. Mixtures were incubated at 30 °C for 15 min, sonicated twice for 10 s at the high output setting with a Branson Sonifier cell d isruptor 200 fitted with a microtip, and centrifuged at 12 000 g for 15 min at 4 °C. Supernatants were either used immediately or frozen at )70 °C until required. The p ellets, when retained, were resuspended in 500 lLofSDS/PAGE treatment buffer, placed in a boiling water bath for 3 min and either electrophoresed immediately or stored at )70 °C. Immunodetection and quantitation of p26 Protein samples electrophoresed in 12.5% SDS polyacryl- amide gels were either stained with Coomassie blue or transferred to nitrocellulose. Blots were rinsed briefly with Tris/NaCl/P i (0.01 M Tris/HC1, 0.14 M NaC1, pH 7.4) and stained with 0.2% Ponceau-S in water to confirm protein transfer. For immunodetection, membranes were blocked 45 min in 5% milk powder dissolved in Tris/NaCl/P i / Tween (Tris/NaCl/P i with 0.1% Tween 20) , followed by incubation for 30 min at room temperature with either anti-p26 Ig [57] or anti-(His 6 tag) Ig (Santa Cruz Biotech- nology, Inc., S anta Cruz, CA, USA) diluted in HST buffer (0.01 M Tris/HC1, 1 M NaC1,0.5%Tween20,pH7.4). Blots were washed twice in HST buffer, then in Tris/NaCl/ P i /Tween, prior to incubation for 30 min with horseradish peroxidase (HRP)-conjugated goat anti-(rabbit IgG) I g (Jackson Immunochemicals, Inc.) diluted in HST buffer. Membranes were washed twice in HST buffer, twice in Tris/NaCl/P i /Tween and once in T ris/NaCl/P i ,witheach wash for 5 min. Immunoconjugates were detected by the enhanced chemiluminescence (ECL) p rocedure (Amersham Pharmacia Biotech) following manufacturer’s instructions. The p 26 bands were scanned with a Bio ÁRad Model GS-670 Imaging Densitometer and analyzed in MOLECU- LAR ANALYST . Values so obtained w ere compared with those comprising the linear portion of a standard curve established for quantitation o f p26. The standard curve was prepared by electrophoresing different amounts of cell free extract from Artemia cysts containing p26 in 12.5% SDS polyacrylamide gels, blotting t o nitrocellulose and probing with antip26 antibody before scanning. Each densitometer value (arbitrary units) was plotted against the amount of cell free extract protein in the gel lane from which the density measurement was made. Centrifugation of p26 in sucrose gradients Sucrose gradients were formed in 0.1 M Tris/glycine (pH 7.4) by layering 5 mL of 10% sucrose on 5 mL of 50% sucrose and centrifuging at 200 000 g for 3 h at 15 °C. Four-hundred microliters of cell free extract from bacteria grown at 30 °C was loaded per g radient and centrifuged at 200 000 g for 21 h at 4 °CinaBeckmanSW41Tirotor. Additionally, 400 lL of p26 purified from Artemia cysts [57], and molecular mass markers o f 29 kDa (carbonic anhydrase), 66 k Da (bovine serum albumin), 150 kDa (alcohol dehydrogenase), 200 kDa (a-amylase), 443 kDa (apoferritin), and 669 kDa (thyroglobulin) (Sigma) were centrifuged on gradients. T ube bottoms were punctured with a 25-gauge needle, 1 mL samples were collected, and 75 lL from each fraction was mixed with 25 lLof4· SDS polyacrylamide gel treatment buffer. Twenty microliters of each sample was then electrophoresed in 12.5% SDS polyacrylamide gels, blotted to nitrocellulose and probed with antibody to p26. Each molecular mass marker, located by reading the A 280 of gradient fractions, tended to occur in seve ral samples, t hus each marker w as centrifuged separately. The position of the peak tube for each marker is indicated in the figures. Thermotolerance of E. coli BL21(DE3) expressing p26 Two m illiliters of Luria–Bertani broth containing 50 lgÁmL )1 ampicillin and 1 m M IPTG was inoculated with a single colony of E. coli BL21(DE3) transformed with either full-length or truncated p26 cDNA in pET21(+), and incubated at 30 °C f or 8–9 h. Immediately before heat shock, 0.5 mL of culture was diluted 1 : 10 in fresh medium supplemented with 25 lgÁmL )1 ampicillin. Cultures were incubated at 5 4 °C in a water bath, 100 lL samples were removed after 0, 15, 30, 45 and 60 min of heat shock, diluted in cold LB broth and maintained on ice prior to plating in duplicate o n LB agar. Colonies were counted after 20–24 h at 37 °C and all p26 constructs were tested a m inimum of three times for t hermotolerance induction. To verify the presence of p26, 500 lLofeach IPTG induced culture was removed prior to heating, cells were collected by centrifugation for 20 s at top speed in a microcentrifuge, re suspended in 50 lL of treatment buffer, placed in a boiling w ater bath for 3 min, and frozen at )20 °C before e lectrophoresis in SDS/polyacrylamide gels, blotting to nitrocellulose and immunodetection. Ó FEBS 2002 Small heat shock/a-crystallin protein from Artemia (Eur. J. Biochem. 269) 935 RESULTS Cloning of full-length and truncated p26 cDNAs Six cDNA products were generated by PCR using selected primers and p26-3-6-3 as template (Table 1). The cDNAs i nclude: p26-full, the full-length p26 cDNA; p26-ND36, lacks N-terminal residues 1–36; p26-ND60 , lacks N-terminal residues 1–60; p26-CD40, lacks C-termi- nal residues 153–192; p26-CD10, lacks C-terminal residues 183–192; p26-alpha, lacks residues 1–60 and 153–192, thereby corresponding to t he a-crystallin domain. That the cDNA fragments in pET21(+) were of the proper size was confirmed by restriction digestion with BamHI and XhoI. Additionally, primers p26-1Bam-s and p26-192Xho- as (Table 1) amplified only the cDNA in p26-full, and amplification of constructs with the primers employed for production of their respective inserts yielded PCR prod- ucts of the expected length. That is, the fragments were the same s ize as those r eleased from pET21(+) by restriction d igestion and to those obtained during t he initial PCR. As final verification of identity, and to see if errors were introduced during PCR amplification, each p26 cDNA cloned in pET21(+) was sequenced and its deduced amino-acid sequence determined (not shown). With one exception, deduced amino-acid sequences of cDNA products were identical, exclusive of engineered deletions, to full-length p26. Construct p26-ND60 had a modified nucleotide at position 407 (numbered as in full- length p26-3-6-3) that caused a Val136Ala substitution. Each p26 cDNA had cytosine at position 324, whereas adenine was reported for p26-3-6-3, and cytosine at position 354 was replaced by thymine. Neither of these changes modified the deduced am ino-acid sequence o f p26. The p26 cDNAs cloned in pET21(+) and utilized in subsequent experiments are represented schematically in Fig. 1. Synthesis of full-length and truncated p26 in E. coli BL21(DE3) Cell free extracts prepared from E. coli transformed with p26 cDNAs in pET21(+) and induced at 30 °CwithIPTG were electrophoresed in SDS/polyacrylamide gels and either stained with Coomassie blue (Fig. 2A) or blotted to nitrocellulose and probed with antibody to p26 (Fig. 2B). Only p26-ND36 yielded an additional band visible in stained gels (Fig. 2A, lane 2). In agreement with this observation, immunostaining of blots with antibody to p26 gave a strong reaction with p26-ND36, while bands of lesser intensity were obtained for p26-full, p26-CD40 and p26-CD10 (Fig. 2B). Extracts from bacteria transformed with p26-ND60 and p26-alpha in pET21(+) usually failed to produce visible bands when Western blots were stained with anti-p26 Ig (Fig. 2B), although v ery weak bands appeared occasionally (not shown). The relative amounts of p26 in lysates of transformed E. coli were determined by incubating Western blots with anti-p26 Ig, taking care to ensure that density measurements were within the linear r ange of film e xposure (Fig. 2C). The ratio of p 26-full: p26-ND36: p26-CD40: p26- CD10 was 2 : 16 : 1 : 2.5. Similar quantities of each p26 variant were produced by transformed b acteria incubated at 30 an d 37 °C, but much of the p26 at the higher t emperature pelleted upon centrifugation at 12 000 g for 15 min (Fig. 3). In contrast, full-length and truncated p26 polypeptides synthesized at 30 °C were almost completely soluble, thus bacteria used for subsequent analysis were grown at this temperature. Occasionally, expressed proteins appeared as doublets (Fig. 3B, lane 2, 3D, lane 4) but the reason is unknown. Centrifugation of bacterially expressed p26 in sucrose gradients The sedimentation patterns of p26 proteins e ncoded by pET21(+) constructs, and detectable on Western blots with anti-p26 Ig, varied upon centrifugation of cell free extracts in sucrose gradients (Fig. 4). For example, p26-CD40 (Fig. 4C) existed mainly as monomers, whereas p26-CD10 (Fig. 4D) and p 26-ND36 (Fig. 4B) were in protein com- plexes larger than monomers but smaller on average than Fig. 1. Schematic representation of p26 cDNAs cloned in pET21(+). Results obtained by cloning full-length and truncated derivatives o f p26 in the prokaryotic expression vector, pET21(+), are summarized. The p 26 cDNAs were cloned into t he BamHI and XhoIsitesof pET21(+), and the constructs were use d to transform E. coli BL21(DE3). MCS, multiple c loning site; Amp, ampicillin resistance; ori, origin o f replic ation; lac 1, lac operator repressor gene; f1 origin, filamentous phage origin of replication. Additional description of clones is available i n Table 1. 936 J. A. Crack et al.(Eur. J. Biochem. 269) Ó FEBS 2002 those seen in E. coli transformed w ith p ET21(+) containing full-length p26 cDNA (Fig. 4A). p26 purified from Artemia cysts ocurred as oligomers (not shown) and in cell extracts from cysts (Fig. 4E) the p26 complexes were slightly larger in size than these produced by recombinant p 26. As revealed by sedimentation patterns therefore truncated p26 variants detectable by antip26 antibody exhibited limited ability t o oligomerize and/or to interact with other proteins, whereas full-length p26 in cell free extracts from E. coli and Artemia formed much larger protein complexes, perhaps due to oligomer assembly as demonstrated for purified p26. Thermotolerance of E. coli expressing full-length and truncated p26 cloned in pET21(+) Bacteria transformed with p26 containing constructs dem- onstrated greater thermotolerance than cells that had incorporated only pET21(+) (Fig. 5 ). Maximum tolerance occurred in bacteria expressing p26-full, but this was only marginally better than protection conferred by p26-ND36 and p26-ND60, w hich in turn was g reater than t he resistance afforded by p26-CD40, p26-CD10 and p26-alpha. The insert (Fig. 5 ) indicates e ither t hat p26 occ urred in cell free extracts from only four transformed cultures, although a ll exhibited enhanced heat tolerance, or that p26 variants produced by p26-ND60 and p26-alpha were pr esent but recognized poorly by antip26 antibody. To determine if recognition of p26-ND60 and p26-alpha encoded polypeptides by antibody to p26 was problematic, corresponding cDNA fragments were inserted into pRSETC and used to transform E. coli. IPTG induced bacteria containing full-length p26, p26- ND60 and p26-alpha in pRSETC produced polypeptides of appropriate size that reacted with antibody to the (His) 6 epitope tag, while only the product of p26-full reacted with anti-p26 Ig (Fig. 6). Thus, antibody to p26 reacted poorly with polypeptides encoded by p26-ND60 and p26-alpha, demonstrating that enhanced thermotolerance conferred by these two constructs correlated with synthesis of p26 polypeptides. Additionally, when tested by centrifugation on sucrose gradients, the size of full-length p26-His 6 (Fig. 7 A) was c lose to those produced by recombinant full-length p26 lacking His 6 , while results with the truncated p26 polypeptides (Fig. 7B,C) were similar to those for p26-CD40, yielding mostly monomers. As with the other truncated derivatives of p26, the induction of thermotoler- ance by p26-ND60 and p26-alpha did not depend on oligomer formation. Fig. 2. Expression of pET21(+) containing p26 cDNA in transformed bacteria. Cell free protein extracts were prepared from E. coli trans- formed with pET21(+) containing full-length and truncated p26 cDNAs and grown in the presence of IPTG at 30 °C. Samples were electrophorese d in 12.5% SDS polyacrylamide gels and either stained with Coomassie blue ( A) or transferred to nitrocellulose and probed with antibod y t o p26 using the ECL proc edure (B) . E ach lane received 15 lL of cell free extract in A and 10 lLinB.M,molecularmass markers of 97, 66, 43, 31, 22 and 14 kDa; 1, p26-full; 2, p26-ND36; 3, p26-ND60; 4, p26-CD40; 5, p26-CD10;6,p26-alpha;7,pET21(+). Arrowhead, p26-N D36. Panel C, Western blots containing lysates of transformed E. coli BL21(DE3) grown at 30 °C an d induced with IPTG were probed with antibody to p26. Film s were scanned and absorbance of the p26 band in each lane, in arbitrary units, dete r- mined. The amounts of sample applied t o the gel were: p26-full, 5 lL; p26-ND36, 1 lL; p26-CD40, 10 lL; p26-CD10, 5 lL. The lanes in which p26 is not v isible each received 10 lL of l ysate. Fig. 3. Solubility of p26 synthesized in transformed bacteria. E. coli BL21(DE3) transformed with p26 cDNA in pET21(+) and ind uced with IPTG were grown at either 30 °C(A,B)or37°C(C,D)for5h, following which soluble (A,C) and insoluble (B,D) fractions were prepared. Twenty microliters of each sample was electrophoresed in 12.5% polyacrylamide gels, blotted to n itrocellulose and probed with antibody to p26 by the ECL p rocedure. Lane 1, 5 lg of cell free e xtract protein from Artemia cysts; 2, p26-full; 3, p26-ND36; 4, p26-CD40; 5, p26-CD10. Ó FEBS 2002 Small heat shock/a-crystallin protein from Artemia (Eur. J. Biochem. 269) 937 DISCUSSION Restriction digestion, PCR amplification and sequencing confirmed the identity of p26 cDNA fragments cloned in pET21(+) and pRSETC. The deduced amino-acid sequence fo r each construct was identical to the corre- sponding region encoded by p26-3-6-3 [58], except for p26- ND60, which had a Val136Ala substitution. Alanine a nd valine are similar amino acids, indicating this modification is unlikely to affect p26. Two other nucleotides were altered in all constructs, but neither rendered an amino-acid conver- sion. Because each construct was amplified from the same p26-3-6-3 preparation, these changes probably represent errors in the original sequence. SDS/PAGE of extracts from IPTG induced bacteria gave a visible pro tein band o f the e xpected size for p26-ND36, but not for other cDNA fragments cloned in pET21(+), upon Fig. 4. Centrifugation of p26 in sucrose gradients. Four hundred llof extract from E. coli BL21(DE3) transformed w ith p26 cDNA cloned in pET21(+) and 400 lL of cell free extract from Artemia cysts were centrifuged in sucrose gradients. Samples f rom gradient fractions were electrophoresed in 12.5% SDS polyacrylamide ge ls, transferred to nitrocellulose, and reacted with antibody to p26 using th e ECL pro- cedure. Th e top of each gradient is to the left of the figure, and numbers across the top indicate successive samples t aken from gradi- ents. (A) p26-full; (B) p26-ND36; (C) p26-CD40; (D) p26-CD10; (E) p26 i n cell fre e extract from Artemia. The p ositions from l eft t o right of molecular mass markers re presenting 29, 66, 150 , 200, 443 and 669 kDa are indicated by arrows. Fig. 5. Thermotolerance of transformed bacteria. Transformed E. coli BL21(DE3), grown as described in Materials and methods, were incubated at 54 °C for the times indicated, plated in duplicate on LB agar and incubated at 37 °C for 20 h. Colonies were counted and the log 10 values of colony forming units (cfu) per mL were plotted against the length of heat shock in min. The results shown are the average of three independent experiments. Bacteria containing only pET21(+) did not survive 60 min of heat shock and the curve was terminated at 45 m in I nsert , 10 lL of cell lysate from e ach heat shoc ked culture was electrophoresed in 12.5% SDS polyacrylamide gels, blotted to nitro- cellulose and probed with anti-p26 Ig, a procedure re peated for each heat shock experiment. 1, p26-full; 2, p26-ND36; 3, p26-ND60; 4, p26- CD40;5,p26-CD10; p26-alpha. Fig. 6. Expression of p26 cDNA cloned in pRSETC. Cell free protein extracts were prepared from E. coli transformedwitheitherpRSETC (A,C) or pET21(+) (B,D) containing p26 cDNA. Dup licate samples were electrophoresed in 12.5% SDS polyacrylamide gels, transferred to nitrocellulose and p robed with antibody to either p26 (A,B) or the (His) 6 epitope tag encoded by pRSETC (C,D). Each lane received 7.5 lL of e xtract. 1, p 26-full; 2, p26-N D60; 3, p26-alpha; 4, vector only. 938 J. A. Crack et al.(Eur. J. Biochem. 269) Ó FEBS 2002 Coomassie blue staining. Probing of Western blots with p26 specific antibodies demonstrated the products of four constructs in extracts of IPTG induced bacteria, but polypeptides corresponding to p26-ND60 and p26-alpha were either absent or recognized poorly by anti-p26 Ig. Western blots of extracts from E. coli transformed w ith p 26- full, p26-ND36 and p26-alpha cloned in pRSETC and probed with antip26 antibody gave results identical to those just described. However, antibody to His 6 revealed the epitope tag, and thus p26-ND36 and p26-alpha, in duplicate samples from pRSETC transformed E. co li. Interestingly, mammalian Cos1 cells transiently transfected with p26- ND60 and p26-alpha were immunofluorescently labeled with antip26 antibody, even though the p26 variants were not detectable on Western blots of extracts from these cells (data not shown). The combined results support the conclusion that bacteria transformed with p ET21(+) containing p26-ND60 and p26-alpha produced p26, but it was poorly recognized by anti-p26 Ig after electrophoresis and transfer to nitrocellulose. Equally important, full-length and truncated derivatives of p26 from transformed E. coli were almost completely soluble at 30 °C, but less so at 37 °C, indicating that bacteria grown at the lower tempe- rature are more likely to give an accurate portrayal of p26 function than are cells incubated at 37 °C. Small heat shock/a-crystallin proteins generally exist as large oligomers when purified [4,8,39,40,57], but there are exceptions [19,20,22]. In this study, p26, either purified from Artemia embryos (not shown) or in cell free extracts, was shown by sucrose gradient centrifugation to exist in protein complexes as large as 670 kDa. Liang et al. [57] demon- strated that purified p26 assembled into oligomers o f 670 kDa, and full-length p26 in c ell free e xtracts from transfected Cos1 cells is also in a complex of similar mass (unpublished data). Full-length p26 synthesized in E. coli yielded protein complexes somewhat smaller on average than those in extracts f rom Arte mia, and by way of comparison, small h eat shock/a-crystallin proteins pro- duced in transformed bacteria usually reside as oligomers similar in size to those in cells from which the expressed cDNA was obtained [17,24,37,61]. The reluctance of full- length p26 to form complexes as large as those in Artemia may reflect improper post-translational processing of the protein in E. coli. On the other hand, trivial explanations for the slightly reduced mass are either that Triton X-100 used during protein preparation affects quaternary structure or that oligomerization of p26 and/or its interaction with other proteins is concentration dependent. The former possibility was not investigated systematically, but preliminary data suggest detergent does not affect the ability of p26 to form large complexes in bacterial extracts. Published results vary in terms of how the concentration of small heat shock/ a-crystallin proteins influe nces oligomer assembly. F or example, a-crystallin tends to oligomerize readily, even at low concentrations [40,42], while Hsp20 oligomerization is concentration dependent [19]. Expression of full-length p26 cDNA in pRSETC was more than for pET21(+), and the average size of p26 complexes resolved in sucrose gradients increased, perhaps as a consequence of greater oligomer- ization due to higher p26 concentration. The a bsence in cell free extracts of high molecular mass complexes when p26 lacks either part or all of the N-terminus favours a role for this region in oligomer assembly. R einforcing th is proposal, t etramers are the maximum size attained by Hspl2.2 and 12.3 from Caenor- habditis elegans [22], and like p26-ND36, the N-terminal domains of these proteins are short. High molecular mass oligomers a re not detected afte r N -terminal deletion of C. elegans Hsp16.2, although dimers and possibly tetramers are present [18]. Additionally, H spl2.6 from C. elegans, with 16 fewer N-terminal residues than Hspl6.2, is mono- meric [20]. Eliminating the 56 N-terminal residues from aA-c rystallin, but not the first 19, reduced oligomer mass, as did removal of 87 N-terminal residues from H sp27 [25] and 33 residues from Hsp25 [33], although the latter modifica- tion was small. In contrast, loss of 42 residues from the N-terminus of a rice small heat shock/a-crystallin protein increased oligomer size [23]. Deleting the last 16 C-terminal residues o f C. elegans Hspl6.2 had limite d effect on quaternary structure [18], as is true for dispensing with 10 C-terminal residues from aA-crystallin [25] and 18 C-terminal residues from Hsp25 [15]. In contrast, oligomer formation by p26 lacking C-terminal residues was compro- mised, signifying this region is important for oligomeriza- tion. Although caution is required because the function of p26, a eukaryotic protein, was examined in E. coli,the Fig. 7. Centrifugation o f p26 in sucrose gradients. Fou r hundred microliters of extract from E. coli BL21(DE3) transformed with p26 cDNA cloned in pET21(+) was centrifuged in sucrose gradients. Samples from gradient factions were electrophoresed in 12.5% SDS polyacrylamide gels, transferred to nitrocellulose, and reacted with antibody to the H is 6 epitope tag. Th e t op o f eac h gradie nt is t o th e l eft of th e figure, and numbers across the top indicate successive samples from gradients. A, p26-full, B, p26-ND60; C, p26-alpha. The positions from left to right of molecular mass markers representing 29, 66, 150, 200, 443 and 669 kDa are indicated by arrows. Ó FEBS 2002 Small heat shock/a-crystallin protein from Artemia (Eur. J. Biochem. 269) 939 results corroborate the idea that N-terminal domains of small heat shock/a-crystallin proteins aid construction o f oligomers from smaller building blocks arising by interac - tions between residues within a-crystallin domains [7,8,21,62]. Contrary to other reports, the C-term inal extension is a lso implicated in oligomerization, but the exact nature of its role is uncertain. Small heat shock/a-crystallin proteins confer thermotol- erance on prokaryotic and eukaryotic organisms [23,27–33]. The construct used p reviously to examine induction of thermotolerance by full-length Artemia p26 encoded N -terminal, nonp26 residues, missi ng from the construct employed h erein [ 49], but the outcome was similar in each case. Moreover, loss of N-terminal residues did not drastically change the ability of p 26 to con fer thermotolerance o n E. coli, suggesting this domain and the assembly of large oligomers are not required for protection. In agreement, bacteria transformed w ith Hsp25, and Hsp25 lacking 33 N-terminal amino-acid residues, are equally heat tolerant [33]. Removal of C-terminal exten- sions from C. elegans Hspl6.2, murine Hsp25 and human aA-crystallin reduced, but did not extinguish small heat shock/ a-crystallin protein chaperone activity in vitro [15,16,18], as is true when hydrophobic residues are placed in the region [17]. Loss of t he C-terminal extension lowered protein s olubility, con sistent with the notion that this region is a solubilizing agent [14–17]. E liminating t he C-terminal extension had little effect on p26 solubility when bacteria were grown at 30 °Cand37°C, although testing at higher temperatures may be informative. Additionally, the C-terminal extension of p26 is required for full induction of thermotolerance in E. coli and thus may be necessary for chaperoning in vitro. Oligomerization in the context of thermotolerance, as described in this s tudy, is not as thoroughly investigated as the association between quaternary structure and chaper- oning in vitro, where an increase in oligomer mass gene- rally enhances protection of client proteins [18–20,22,63]. However, oligomer mass is not the only determinant of small heat shock/a-crystallin protein function. As a case in point, o ligomers l acking chaperone activity arise from chimeric a-crystallins [63]. Also, insertion of a peptide [41,64] and change of a single residue [10,11,13,65] lead to enlarged oligomers with curtailed chap erone action in vitro. In other work, dissociation of oligomers was a prerequisite for chaperoning in vitro [66], and disassembly of active units from an oligomeric (storage) state of a-crystallin was proposed, upon structural analysis of aA-crystallin by site- directed spin labelling, as a model for chaperone function [67]. Such r esults downplay oligomerization as a prerequisite for protection again st stress. In t his vein, monomeric a-crystallin at low chaperone to target ratios protects lens sorbitol dehydrogenase enzyme activity upon heating [46], while the N-terminal portion of a small heat shock/ a-crystallin protein confers thermotolerance on cells [29] and prevents aggregation of stressed proteins in vitro [12]. These polypeptides are unlikely to oligomerize, either in vivo or in vitro. Clearly, our data st rengthen the notion that small heat shock/a-crystallin proteins function in vivo when not in large oligomers. Whether this signals nonspecific effects on proteins, an interplay with membranes as reported recently [68], or mechanistic differences between thermotolerance and molecular chaperoning in vitro awaits purification of truncated p26 derivatives and testing in a defined system, experiments now in progress. ACKNOWLEDGEMENTS The work was supported by a Natural Sciences and E ngineering Research Co uncil of Canada Re search Grant and a Nova Sc otia Health Research Foundation New O pportunity Grant to T. H. M. REFERENCES 1. Bukau, B. & Horwich, A.L. ( 1998 ) The Hsp70 and H sp 60 chap- erone machines. Cell 92 , 351–366. 2. Wickner, S., Maurizi, M.R. & Gottesman, S. (1999) Posttransla- tional quality control: fold ing, refolding, and de grading proteins. Science 286, 1888–1893. 3. Kimmins, S. & MacRae, T.H. (2000) Maturation of steroid receptors: an example of functional cooperation amo ng m olecular chaperones and their assoc iat ed proteins. Cell Stress Chaperones 5, 76–86. 4. MacRae, T.H. (2000) Structure and function of small heat shock/ a-crysta llin proteins: es tablish ed concepts and emerging ideas. Cell. Mol. Life Sci. 57 , 899–913. 5. Easton, D.P., Kaneko, Y . & Subjeck, J.R. ( 2000) The Hsp110 and Grp170 stress proteins: newly recognized relatives of t he Hsp70s. Cell Stress Chaperones 5, 276–290. 6. de Jong, W.W., Caspers, G J. & Leunissen, J.A.M. (1998) Genealogy of the a-crystallinCsmall heat-shock protein super- family. Int. J. Biol. Macromol. 22, 151–162. 7. Merck, K.B., De Haar d-Hoekman, W.A., Essink, B.B.O., Bloemendal, H . & De Jong, W.W. (1992) Expression and aggre- gation of recombinant aA-crystallin and its two domains. Bi ochim. Biophys. Acta 1130, 267–276. 8. Augusteyn, R.C. (1998) a-Crystallin polymers and polymerization: the view from down under. Int. J. Biol. Macromol. 22, 253–262. 9. Kim, K.K., Kim, R. & Kim. S H. (1998) Crystal structure of a small heat-shock protein. Nature 394, 595–599. 10. Bova, M.P., Yaron, O., Huang, Q., Ding, L., Haley, D.A., Stewart, P .L. & Horwitz, J. (1999) Mutation R120G in aA-crys- tallin, which is linked to a desmin-related myopathy, results in an irregular structure and defective chape rone-like function. Proc. Natl Acad. Sci. USA 96, 6137–6142. 11. Kumar, L.V.S., Ramakrishna, T. & Rao, ChM. (1999) Structural and functional conse quences o f the mu tation o f a co nserved arginine residue in aAandaB crystallins. J. Biol. Chem. 274, 24137–24141. 12. Sharma, K.K., Kumr, R.S., Kumar, G.S. & Quinn, P.T. (2000) Synthesis and characterization of a pep tide identified as a fu nc- tional element in aA-crystallin. J. Biol. Chem. 275, 3767–3771. 13. Shroff,N.P.,Cherian-Shaw,M.,Bera,S.&Abraham,E.C.(2000) Mutation of R116C results in h ighly oligomerized aA-crystallin with modified structure and defective chaperone-like function. Biochemistry 39, 420–1426. 14. Carver, J .A., Esposito, G., Schwedersky, G. & Gaestel, M . (1995) 1 H NM R spectroscopy reveals that mouse H sp25 has a flexible C-terminal extension of 18 amino acids. FEBS Lett. 369, 305–310. 15. Lindner, R.A., Carver, J.A., Ehrnsperger, M., Buchner, J., Esposito,G.,Behlke,J.,Lutsch,G.,Kotlyarov,A.&Gaestel,M. (2000) Mouse Hsp25, a small heat shock protein. The role of its C-terminal extension in oligomerization and chaperone action. Eur. J. Biochem. 267, 1923–1932. 16. Andley, U.P., Mathur, S., Griest, T.A. & Petrash, J.M. (1996) Cloning, expression, and chaperone-like activity of human aA-crystallin. J. Biol. Chem. 271, 31973–31980. 17. Smulders, R.H.P.H., Carver, J.A., Lindner, R.A., van Boekel, M.A.M., B loemendal, H. & de Jong, W.W. (1996) Immobilization 940 J. A. Crack et al.(Eur. J. Biochem. 269) Ó FEBS 2002 of the C-terminal extension of bovine aA-crystallin r educes chaperone-like activity. J. Biol. Chem. 271, 2 9060–29066. 18. Leroux, M.R., Melki, R., Gordon , B., Bate lier, G. & Candido, E.P.M. (1997) Structure-function studies on small heat shock protein oligomeric assembly and interaction with unfolded poly- peptides. J. Biol. Chem. 272, 24646–24656. 19. van de Klundert, F.A.J.M., Smu lders, R.H.P.H., Gijsen, M.L.J ., Lindner, R.A., Jaenicke, R., Carver, J.A. & de Jong, W.W. (1998) The mammalian small heat-shock protein Hsp20 forms dimers and is a poor chaperone. Eur. J. Biochem. 258, 1014– 1021. 20.Leroux,M.R.,Ma,B.J.,Batelier,G.,Melki,R.&Candido, E.P.M. (1997) Unique structural features of a novel class of small heat shock proteins. J. Biol. Chem. 272, 12847–12853. 21. Boelens, W.C., Croes, Y., de Ruwe, M., de Reu, L. & de Jon g, W.W. (1998) Negative charges in the C-terminal domain stabilize the aB-crystallin complex. J. Biol. Chem. 273, 28085– 28090. 22. Kokke,B.P.A.,Leroux,M.R.,Candido,E.P.M.,Boelens,W.C.& de Jong, W.W. (1998) Caenorhabditis elegans small heat-shock proteins Hsp12.2 and Hsp12.3 f orm tetramers and have no chaperone-like activity. FEBS Lett. 433, 228–232. 23. Young, L S., Yeh, C H., Chen, Y M. & Lin, C Y. (1999) Molecular ch aracterization of Oryza sativa 16.9 k Da h eat s hock protein. Bi oche m. J. 344, 31–38. 24. Haley, D.A., Bova, M .P., Huang, Q L., Mchaourab, H.S. & Stewart, P.L . (2000) Small heat-shock p rotein structures reveal a continuum from symmetric to variable assemblies. J. Mol. Biol. 298, 261–272. 25. Bova, M.P., Mchaourab, H.S., Han, Y. & Fung, B.K K. (2000) Subunit e xchange of small h eat sho ck proteins. Analysis of olig- omer formation of aA-crystallin and Hsp27 by fluorescence res- onance energy transfer and site-directed truncations. J. Biol. Chem. 275, 1035–1042. 26. Ha ¨ rndahl, U., Kokke, B .P.A., Gustavsson, N., Linse, S., Berg- gren, K., Tjerneld, F., Boelens, W.C. & Sundby, C. (2001) The chaperone-like activity of a small heat shock protein is lost after sulfoxidation of conserved methionines in a surface-exposed amphipathic a-helix. Biochim. Biophys. Acta 1545, 227–237. 27. van den I Jssel, P.R.L.A., Overkamp, P., K nuaf, U., Ga estel, M. & de Jong, W.W. (1994) aA-crystallin confers cellular t hermoresis- tance. FEBS Lett. 355, 54–56. 28. Linder, B., Zhijun, J., Freedman, J.H. & Rubin, C .S. (1996) Molecular characterization of a novel, developmentally regulated small e mbry onic chaperone f rom Caenorhabditis elegans. J. Bio l. Chem. 271, 30158–30166. 29. Yeh,C H.,Chang,P F.L.,Yeh,K W.,Lin,W C.,Chen,Y M. & Lin, C Y. (1997) Expression of a gene encoding a 16.9-kDa heat-shock protein, Oshsp16.9. Escherichia coli enhances the r- motolerance. Proc. Natl Acad. Sci. USA 94 , 10967–10972. 30. Wiesmann, K.E.H., C oop, A., Goode, D., Hepburne -Scott, H .W . & Crabbe, M.J.C. (1998) Effect of mutations of murine lens aB crystallin on transfected neural cell viability and cellular translo- cation in response to stress. FEBS Lett. 438, 25–31. 31. Muchowski, P.J. & Clark, J.I. (1998) ATP-enhanced molecular chaperone functions of t he small heat s hock protein human aB crystallin. Proc. Natl Acad. Sci. USA 95, 1004–1009. 32. Muchowski, P .J., Wu, G.J.S., Liang, J.J.N., Adman, E.T. & Clark, J.I. (1999) Site-directed mu tations within the core Ôa- crystallinÕ domain of the small heat shock protein, human aB- crystallin, decrease molecular chaperone functions. J. Mol. Biol. 289, 397–411. 33. Guo, Z. & Cooper, L.F. (2000) An N-terminal 33-amino-acid- deletion variant of hsp25 retains oligomerization a nd functional properties. Biochem. Biophys. Res. Comm. 270, 183–189. 34. Andley, U.P., Song, Z., Wawrousek, E.F., Fleming, T.P. & Bassnett, S. (2000) Differential protective activity of aA- and aB-crystallin in lens epithelial cells. J. Biol. Chem. 275, 36823– 36831. 35. Samali, A., Robertson, J.D., Peterson, E., Manero, F., van Zeijl, L., Paul, C., Cotgreave, I.A., Arrigo, A P. & Orrenius, S. (2001) Hsp27 protects m itochondria of thermotolerant cells aga inst apoptotic stimuli. Cell Stress Chaperones 6, 49–58. 36. Rajaraman, K., Raman, B. & Rao, ChM. (1996) Molten-globule state of carbonic anhydrase binds to the chaperone-like a-crys- tallin. J. Biol. Chem. 271, 27595–27600. 37. Sun, T X., Das, B.K. & Liang, J.J N. (1997) Conformational and functional differences b etween recombinant human len s aA- and aB-crystallin. J. Biol. Chem. 272, 6220–6225. 38. Wang, K. & S pector, A. (2000) a-Crystallin prevents irreversible protein d en aturat ion and acts co op eratively with other heat-shock proteins to renature t he stabilized p artially de natured protein in an ATP-dependent manner. Eur. J. Biochem. 267, 4705–4712. 39. Vanhoudt, J., Aerts, T., Abgar, S. & Clauwaert, J. (1998) Quar- ternary structure of bovine a-crystallin: influence o f temperature. Int. J. Biol. Macromol. 22, 229–237. 40. Vanhoudt, J., Abgar, S., Aerts, T. & Clauwaert, J. (2000) Native quaternary structure of bovine a-crystallin. Biochemistry 39, 4483– 4492. 41. van Rijk, A.F., van den Hurk, M.J.J., Renkema, W., Boelens, W.C., de Jong, W.W. & Bloemendal, H. (2000) Characteristics of super aA-crystallin, a product of in vitro exon shuffling. F EBS Lett. 480, 79–83. 42. Datta, S .A. & Rao, ChM. (2000) Packing-induced conformational and functio nal changes in the subunits of a-crystallin. J. Biol. Chem. 275, 41004–41010. 43. Ehrnsperger, M., Gra ¨ ber, S., Gaestel, M . & Buchner, J. (1997) Binding of non-native protein to Hsp25 during heat shock creates a reservoir of folding intermediates for reactivation. EMBO J. 16, 221–229. 44. Lindner, R .A., Ka pur, A. & C arver, J.A . (1997) The interaction of the molecular chaperone, a-crystallin, with molten globule states of bovine a-lactalbumin. J. Biol. Chem. 27 2, 27722–27729. 45. Lindner, R.A., Treweek, T.M. & Carver, J.A. (2001) The molec- ular chaperone a-crystallin is in kinetic c ompetition with aggre- gation to s tabilize a monomeric molten-globule form o f a-lactalbumin. Biochem. J. 354, 7 9–87. 46. Marini,I.,Moschini,R.,Corso,A.D.&Mura,U.(2000)Com- plete protection by a-crystallin of lens sorbitol dehydrogenase undergoing thermal stress. J. Bio l. Chem. 275, 32559–32565. 47. Jackson, S.A. & Clegg, J.S. (1996) Ontogeny of low molecular weight stress protein p26 during early development of the brine shrimp, Artemia franciscana . Dev. Growth Differ. 38, 153–160. 48. Clegg, J.S., Willsie, J.K. & Jackson, S.A. ( 1999) Adaptive signifi- canceofasmallheatshock/a-crystallin protein (p26) in encysted embryos of the brine s hrimp, Artemia franciscana. Am. Zool. 39, 836–847. 49. Liang, P. & MacRae, T.H. (1999) The synthesis of a small heat shock/a-crystallin protein in Artemia and its relationship to stress tolerance during development. Dev. Biol. 207, 445–456. 50. MacRae, T.H. (2001) Do stress proteins protect embryos during metabolic arrest and diapause? I n Molecular Mechanisms o f Metabolic Arre st. Life in Limbo (Storey, K.B., ed.), pp. 169–186. BIOS Scientific Publishers Ltd., Oxford, UK 51. Clegg, J.S. (1997) Embryos of Artem ia franciscana su rvi ve four years o f continuous a noxia: the c ase for comple te metabolic rate depression. J. Exp. Biol. 200 , 467–475. 52. Clegg, J .S. & Jackson, S.A. ( 1998) The metabolic status of qui- escent and diapause embryos of Artemia franciscana (Kellogg). Arch. Hydrobiol. 52, 425–439. 53. Drinkwater, L.E. & Clegg, J.S. (1991) Experimental biology of cyst diapause. In Artemia Biology (Browne,R.A.,Sorgeloos,P.& Trotman, C.N.A., eds), pp. 93–117. CRC P ress, I nc., Boca Raton, FL. Ó FEBS 2002 Small heat shock/a-crystallin protein from Artemia (Eur. J. Biochem. 269) 941 54. Clegg, J.S., Jackson, S .A. & Po pov, V.I. (2000) Long-term anoxia in encysted embryos o f the c rusta cean, Artemia franciscana: v iabil- ity, ultrastructure and stress proteins. Cell Tiss . R es. 301, 4 33–446. 55. Warner, A.H. & Clegg, J.S. (2001) Diguanosine nucleotide me- tabolism and the survival of Artemia embryos during years of continuous anoxia. Eur. J. Biochem. 268, 1568–1576. 56. Clegg, J.S., Jackson, S.A., Liang, P. & MacRae, T.H. (1995) Nuclear-cytoplasmic translocations of p rote in p26 during ae robic- anoxic transitions in embryos o f Artemia franciscana. Exp. Cell Res. 219, 1–7. 57. Liang, P., Amons, R., MacRae, T.H. & Clegg, J.S. (1997) Puri- fication, structure and in vitro molecular-chaperone activity of Artemia p26, a small heat shock/a-crystallin protein. Eur. J. Bio- chem. 243, 225–232. 58. Liang, P., Amons, R., Clegg, J.S. & MacRae, T.H. (1997) Mole- cular characterizat ion o f a sm all he at sho ck/a-crystallin protein in encysted Artemia embryos. J. Biol. Chem. 272, 19051–19058. 59. MacRae, T.H. & Liang, P. (1998) Molecular characterization of p26, a cyst-specific, small heat shock/a-crystallin protein from Artemia franciscana. Arch. Hydrobiol. 52, 3 93–409. 60. Sambrook, J., Fritsch, E.F. & Maniatis, T. (1989) Molecular Cloning – A Laboratory Manual Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York. 61. Muchowski, P.J., Bassuk, J.A., Lubsen, N.H. & Clark, J.I. (1997) Human aB-crystallin. Small heat shoc k protein and molecular chaperone. J. Biol. Chem. 272, 2578–2582. 62. Lambert, H., Cha rette, S.J., Bernier, A .F., Guimond, A. & Lan- dry, J. (1999) H SP27 multimerization m ediated by phosphoryl- ation–sensitive intermolecular interactions at the amino terminus. J. Biol. C hem. 274, 9378–9385. 63. Kumar, L.V.S. & Rao, ChM. (2000) Domain swapping in human aAandaB crystallins affects oligomerization and enhances chaperone-like activity. J. Biol. Chem. 275, 22009–22013. 64. Smulders, R .H.P.H., van Geel, I.G., Gerards, W.L.H., Bloemen- dal, H. & de Jong, W.W. (1995) Reduced chaperone-like activity of aA ins -crystallin, an alternative splicing product containing a large insert peptide. J. Biol. Chem. 270, 13916–13924. 65. Cobb, B.A. & Petrash, J.M. (2000) Struct ural and functional changes in the aA-crystallin R116C mutant in hereditary cata- racts. Biochemistry 39, 15791–15798. 66. Haslbeck,M.,Walke,S.,Stromer,T.,Ehrnsperger,M.,White, H.E., Chen, S., Saibil, H.R. & Buchner, J. ( 1999) Hsp26: a tem- perature-regulated chaperone. EMBO J. 18 , 6744–6751. 67. Koteiche, H.A. & Mchaourab, H.S. (1999) Folding pattern of the a-crystallin domain in aA-crystallin determined by site-directed spin labelling. J. Mol. B iol. 294, 561–577. 68. To ¨ ro ¨ k, Z., Goloubinoff, P., Horva ´ th, I., Tsvetkova, N.M., Glatz, A.,Balogh,G.,Varvasovszki,V.,Los,D.A.,Vierling,E.,Crowe, J.H. & Vigh, L. (2001) Synechocystis HSP17 is an a mphitropic protein that stabilizes heat-stressed membranes and binds dena- tured proteins for su bsequent chaperone-mediated refolding. Proc. Natl Acad. Sci. USA 98, 3098–3103. 942 J. A. Crack et al.(Eur. J. Biochem. 269) Ó FEBS 2002 . Functional analysis of a small heat shock /a- crystallin protein from Artemia franciscana Oligomerization and thermotolerance Julie A. Crack, Marc Mansour,. MacRae, T.H. & Liang, P. (1998) Molecular characterization of p26, a cyst-specific, small heat shock /a- crystallin protein from Artemia franciscana. Arch.