Molecular Vision 1999; 5:15 <http://www.molvis.org/molvis/v5/p15/> Received 20 January 1999 | Accepted 5 August 1999 | Published 10 August 1999

Download Reprint

Maintenance of Chaperone-like Activity Despite Mutations in a Conserved Region of Murine Lens [alpha]B Crystallin

Henry W. Hepburne-Scott, M. James C. Crabbe
 
 

Division of Cell and Molecular Biology, School of Animal and Microbial Sciences, The University of Reading, Whiteknights, Reading, Berkshire RG6 6AJ, UK

Correspondence to: M. James C. Crabbe, Division of Cell and Molecular Biology, School of Animal and Microbial Sciences, The University of Reading, P. O. Box 228, Whiteknights, Reading, Berkshire RG6 6AJ, UK; Phone and FAX: +44 (0)1189 318894; email: m.j.c.crabbe@rdg.ac.uk

Abstract

Purpose: To understand the relationship between certain conserved residues in [alpha]B crystallin and the chaperone-like function of the protein.

Methods: In [alpha]B crystallin, residues H101 to R120 are highly conserved between [alpha]B crystallin and [alpha]A crystallin (85% identity), and between [alpha]B crystallin and the small heat shock protein hsp 27 (80% identity). We made three substitution mutants of [alpha]B crystallin: the single mutant F118A, and the double mutants K103L/H104I, and E110H/H111E.

Results: Polyacrylamide gel electrophoresis revealed no decrease in aggregate size or measureable structural changes between them and the native structure. Using the insulin aggregation assay, all three mutants had identical chaperone-like activity to the wild-type recombinant [alpha]B crystallin.

Conclusions: Despite the high conservation in this area of sequence between [alpha]B crystallin, [alpha]A crystallin, and the small heat shock protein hsp 27, mutations F118A, K103L/H104I, and E110H/H111E did not significantly alter chaperone-like activity.

Introduction

The mammalian small heat shock protein [alpha]crystallin is emerging as a key protein in a remarkable variety of cellular processes including: oxidative stress responses in heart and respiratory tissue [1], development of receptivity in the secretory phase endometrium [2], cellular differentiation in the eye, and in a variety of neurodegenerative disorders. It is the major antigen in the autoimmune response observed in multiple sclerosis [3], and is expressed by both oligodendrocytes and astrocytes in MS lesions [4].

[alpha]Crystallin is composed of 2 subunits, [alpha]A and [alpha]B, of approximately 20 kDa in mass each but the active protein exists as a large aggregate of varying mass between 300 and 1000 kDa [5,6]. [alpha]B crystallin can co-aggregate with [alpha]A crystallin, and with the small heat shock protein hsp27. The mechanism(s) involved in [alpha]crystallin activity remain unclear. It has been shown to confer increased stress resistance, enhanced adherence to surfaces and increased cytoskeleton on mammalian cell cultures [7] and promotion of proliferation, and reorganisation of the cytoskeleton appears a probable mechanism in [alpha]crystallin activity. Observations of stretch responses in human trabecular meshwork cells have provided further evidence of the association of [alpha]crystallin with actin filaments and of specific proteolysis of [alpha]crystallin during reorganisation of actin filaments [8]. It has also been shown to exhibit temperature dependent interaction with tubulin [9] and with intermediate filaments vimentin and peripherin during stress in vivo [10,11]. [alpha]B Crystallin is normally expressed in neurons, and may interact with intermediate filaments in NIH 3T3 cells [11].

Observation of [alpha]crystallin behaviour in vitro and in vivo suggests that other pathways may also be involved. A number of researchers have proposed that [alpha]crystallin is a molecular chaperone. It is certainly able to prevent protein aggregation in vitro by sequestering denatured proteins [12,13]. We have shown that mutations of [alpha]B crystallin which influence chaperone-like activity [13] may also influence viability of transfected N1E-115 neural cells under stress, while not influencing the distribution of the protein within the neural cells subject to heat shock [14].

Under certain conditions, we and others have found that [alpha]crystallin can assist in the refolding of proteins, including citrate synthase, [alpha]-glucosidase [15] and prochymosin [16]. Recent evidence appears to suggest that [alpha]crystallin can function as a true molecular chaperone in vitro in the presence of ATP [17]. Peptide binding by molecular chaperones is thought to be largely mediated by hydrophobic interactions, particularly involving region(s) such as the conserved RLFDQFF near the N-terminus [18]. Using site-directed mutagenesis and heterologous expression in E. coli, we showed that this region was indeed important in [alpha]crystallin chaperone-like activity, both in vitro and in vivo [13]. Recently published data [17] has confirmed our previous findings that recombinant [alpha]crystallin also confers increased viability on E. coli cells expressing the protein.

In [alpha]B crystallin, residues H101 to R120 are highly conserved between [alpha]B crystallin and [alpha]A crystallin (85% identity), and between [alpha]B crystallin and the small heat shock protein hsp27 (80% identity). We considered that residues in this region of [alpha]B crystallin might influence either or both the chaperone-like activity and the aggregation state of the protein. We therefore made three substitution mutants of [alpha]B crystallin: the single mutant F118A, and the double mutants K103L/H104I, and E110H/H111E. An aim of each mutation was to potentially reduce protein aggregation and/or chaperone-like activity. This paper shows that all three mutants had identical chaperone-like activity to the wild-type recombinant [alpha]B crystallin, and did not show any alterations in gross structural properties.

Methods

Plasmids and bacterial strains used were as described previously [13,19]. Restriction enzymes were obtained from Life Technologies Inc, Gathersburg, MD, USA, Taq polymerase was from Perkin-Elmer Cetus (Perkin-Elmer Corp., Warrington, UK). T4 ligase was from Gibco BRL, Paisley, Renfrewshire, UK. Media and chemicals were from Sigma, London, UK, and were of molecular biology grade as appropriate.

Subcloning and preparations of plasmids

This was as described previously [13], using pBlueScript SK. Prior to ligation, pBluescript was subjected to EcoRV digestion to create blunt ended sites for ligation.

Site-directed mutagenesis

This was carried out by PCR overlap extension mutagenesis [13]. Substitutions were made by incorporating mismatches in the PCR primers. DNA sequences of mutant amplicons were verified as described previously [13]. Mutagenic Primer sequences were designed as shown in Table 1.

Subcloning of Mutant Genes into Expression Vector pET3-d

This was as described previously [13], using linearised and dephosphorylated pET3-d. Linearisation was performed using restriction endonuclease NcoI, and dephosphorylation was by shrimp alkaline phosphatase, which lacks exonuclease activity. Plasmid DNA from the ligation reaction was used to transform competent E. coli BLR (DE3) cells. Transformants were screened for the presence of correctly orientated mutant [alpha]B crystallins using restriction endonuclease BamHI before being selected for use in expression experiments.

Expression of Mutant [alpha]B Crystallin

To express mutant [alpha]B crystallin, single well isolated colonies of recombinant E. coli BLR (DE3) cells were used as inoculum for 50 ml expression cultures. Following expression, cells were lysed by the freeze thaw method or the French Press method. Bacterial lysate was subjected to dot blotting and SDS PAGE analysis.

To prepare a starter inoculum for the 50 ml culture, single well isolated colonies of E. coli BLR (DE3) expression cells were used to inoculate Luria Broth (LB, 10 ml) which had been supplemented with ampicillin (final concentration 1 µg/ml). This was incubated overnight at 37 °C in a shaker (250 cycles per min). The overnight culture (1 ml) was used to inoculate 50 ml of sterile LB which was contained in a conical flask and had been supplemented with ampicillin (final concentration 1 µg/ml). This was incubated at 37 °C on a shaker until an OD₆₀₀ of between 0.6 and 1.0 had been reached (typically 2 to 3 h). At this time induction was performed by the addition of IPTG (final concentration 0.4 mM). After a further 2 h incubation the cultures were centrifuged at 3000 x g for 5 min in order to sediment the bacterial cells. The supernatant was discarded and the pellet resuspended in PBS (7 ml) in preparation for lysis.

Freeze Thaw Method for cell lysis

Cell resuspension (1 ml) was added to a 1.5 ml microcentrifuge tube. This was subjected to 3 freeze thaw cycles. Freeze thaw lysate was stored at -20 °C for up to 2 months.

French Press Method for cell lysis

Cell resuspension (5 ml) was added to the chamber of the French Press mini-cell and subjected to 2 presses. French Press lysate was stored at -20 °C for up to 2 months.

Dot blotting

Dot blotting was used to demonstrate the expression of [alpha]B crystallin. The antibody used was rabbit polyclonal antibody which had previously been shown to bind wild-type and mutant [alpha]B crystallin protein but not E. coli proteins. Lysates were pre-heated at 90 °C for 10 min in order to inactivate host enzymes prior to dot blotting. Dot blotting then proceeded as described [13].

SDS-polyacrylamide gel electrophoresis (PAGE) and western blotting

These were performed as described previously [13,19].

Protein purification

This was carried out essentially as described previously [13], using Sephacryl S-400 HR (Pharmacia, Uppsala, Sweden). Immunoreactive fractions were collected and pooled prior to concentration in an Ultrafree-15 centrifugal filter device (Millipore, Watford, Herts., UK).

Assays for Chaperone-like Activity

The reduced insulin B-chain method was used [13], as it is conducted at a temperature (25 °C) that does not risk heat denaturation of the protein. Bovine pancreatic insulin (0.1 mg) was reduced with 20 mM dithiothreitol in the presence or absence of either 0.1 mg or 1.0 mg of wild-type or mutant [alpha]crystallin. The final reaction volume of 400 µl was contained in a quartz cuvette and A₃₆₀ measurements were recorded at 30 s intervals using a Unicam UVB spectrophotometer with a water-heated cuvette holder.

Results

Expression and purification of mutant [alpha]B crystallins

DNA sequencing showed that all three mutants within pBluescript were of the desired sequence. Subcloning of the mutants into pET3-d was successful, and at least one correctly orientated gene for each mutant was identified and selected for expression. Dot blots of post-expression E. coli lysates bearing mutant [alpha]B crystallin genes were positive, indicating the presence of a protein which was immunoreactive for anti [alpha]crystallin antibodies. A control dot blot of wild type E. coli lysate was negative.

SDS PAGE analysis revealed an increase in the level of a 20 kDa protein in post-expression E. coli lysates as compared with pre-expression lysates (Figure 1). Western blot analysis verified that this 20 kDa protein was immunoreactive to anti [alpha]crystallin antibody.

Gel filtration chromatography succeeded in purifying the expressed mutant [alpha]B crystallins to a purity of 85-90% with a yield of 1-2 mg/ml (estimated from A₂₈₀). Figure 2 shows a typical SDS PAGE gel for two of the mutants and bovine [alpha]crystallin; the third mutant gave identical results. The higher molecular band is always observed in [alpha]crystallin recombinant as well as native preparations [13], and probably represents protein that has remained partially aggregated despite being boiled in SDS prior to loading.

Aggregate size of recombinant wild-type and mutant [alpha]B crystallins

The aggregate size of wild-type [alpha]crystallins is very high (approx 800 kDa) and difficult to estimate by PAGE or gel filtration. However comparison of the immuno-dot blot-positive fractions from the Sephacryl S-400 HR column demonstrated that control bovine [alpha]crystallin, and all 3 mutants eluted in the 88-92 ml fractions with no lower molecular weight [alpha]crystallin species detected in later fractions. This was confirmed by the non-denaturing PAGE which demonstrated no observable low molecular weight species in the purified [alpha]B crystallin samples. In all samples (wild-type and mutant) the protein was entirely found as a very high MW species which sat on the top of the 5% separating gel. Thus it appears that all of the mutations studied have failed to alter the aggregation behaviour of the [alpha]B crystallin in such a way as to produce monomers, tetramers or other low molecular weight species.

Chaperone-like Activity

Wild-type [alpha]crystallin was shown to possess chaperone-like activity (Figure 3A) at both 1.0 and 0.1 mg concentrations. Each mutant showed identical chaperone-like activity to the wild-type protein at both concentrations (Figure 3B-D). In the presence of mutant [alpha]crystallin the aggregation of insulin was also prevented; this was the case for each of the three mutants tested.

Discussion

The results of the non-denaturing PAGE and the size exclusion chromatography suggest that none of the mutations made resulted in a decrease in aggregate size of [alpha]B crystallin, since the mutant proteins co-eluted with the wild-type protein. Results from the insulin aggregation assay show that the chaperone-like activity of [alpha]B crystallin was unaffected by any of the mutations made. In mutant F118A, the amino acid phenylalanine was replaced with alanine which, in lacking the phenyl moiety from its side chain, is much less hydrophobic. The fact that this mutation had no effect on [alpha]crystallin's function is in contrast to previous findings in which a hydrophobic amino acid (F27) was found to be crucial to this protein's chaperone-like activity. This suggests that only a proportion of hydrophobic residues are involved in chaperone-like activity, probably dependent upon the surface exposure of the hydrophobic groups [6]. The fact that neither mutation had an effect on the aggregation property of [alpha]B crystallin suggests that any charge:charge interaction between subunits is unlikely to involve these amino acids. The question persists as to the reason for these residues being so well conserved in proteins with which [alpha]B crystallin co-aggregates. One possibility is that these residues may be involved in another mechanism for [alpha]B crystallin/small heat shock protein activity. It is of interest that this conserved region has high sequence homology to residues encountered in helix 3/4 of homeodomain proteins [14], which bind DNA and are involved in regulation of cell development.

A crystal structure for hsp16.5 from the bacterium Methanococcus jannaschii has recently been published [6]. Not only does this suggest a GroEL-like structure for this small heat shock protein, but also it provides evidence as to the regions which are important for intersubunit interaction. The most extensive intersubunit contacts are found between the backbone atoms of the ß2-strand of one hsp16.5 subunit and the ß6-strand of another. The ß2-strand of hsp16.5 spans from I45 to E49. It has an equivalent in [alpha]A crystallin (S63 to S67) and [alpha]B crystallin (S66 to L70). [alpha]B Crystallin amino acids S66, E67 and R69 are conserved in [alpha]A crystallin and hsp25. This raises the possibility that these amino acids are important for the intersubunit interactions, which give rise to its quaternary structure. The fact that they differ from those at equivalent positions in hsp16.5 may simply reflect the differing environmental demands made on the stability of these proteins (M. jannaschii is a hypothermophile). Recent analysis using cryo-electron microscopy indicates a variable pattern of [alpha]B crystallin subunit assembly resulting in a variable quaternary structure for this protein [20]. In the lens, it is important to ensure that proteins are not of repeating regular structure, and so do not crystallise at the high protein concentrations encountered in this tissue.

It is clear that [alpha]B crystallin is remarkably resilient to muations without modification of biophysical properties or chaperone-like activity [21]. That is not to say that key residues are not important in certain species and under certain conditions [22]. Recent experiments using fluorescent probes to detect hydrophobic binding suggested that the sequence of residues 93-107 was one of two substrate-binding sites in [alpha]B crystallin [23]. The mutant K103L/H104I, with mutation sites within this sequence, might have increased chaperone-like activity because of enhanced hydrophobicity in this region. We did not demonstrate this effect either because the assay is not sensitive enough, or because, as we and others have suggested previously, hydrophobic protein interactions near the N-terminal residues are quantitatively more important in chaperone-like activity. Based on this suggestion, experimental design should target the N-terminal residues in mutation studies and future work on the protein may yield important results on its true function or functions within the cell.

Acknowledgements

We thank the Wellcome Trust, the RNIB, and the University of Reading for grant support.

References

1. Iwaki T, Iwaki A, Goldman JE. Alpha B crystallin in oxidative muscle fibers and its accumulation in ragged-red fibers: a comparative immunohistochemical and histochemical study in human skeletal muscle. Acta Neuropathol (Berl) 1993; 85:475-80.

2. Gruidl M, Buyuksal A, Babaknia A, Fazleabas AT, Sivarajah S, Satyaswaroop PG, Tabibzadeh S. The progressive rise in the expression of alpha crystallin B chain in human endometrium is initiated during the implantation window: modulation of gene expression by steroid hormones. Mol Hum Reprod 1997; 3:333-42.

3. van Noort JM, van Sechel AC, Bajramovic JJ, el Ouagmiri M, Polman CH, Lassnamm H, Ravid R. The small heat-shock protein alpha B-crystallin as candidate autoantigen in multiple sclerosis. Nature 1996; 375:798-801.

4. Bajramovic JJ, Lassmann H, van Noort JM. Expression of alphaB-crystallin in glia cells during lesional development in multiple sclerosis. J Neuroimmunol 1997; 78:143-51.

5. de Jong WW, Leunissen JA, Voorter CE. Evolution of the alpha-crystallin/small heat-shock protein family. Mol Biol Evol 1993; 10:103-26.

6. Kim KK, Kim R, Kim SH. Crystal structure of a small heat-shock protein. Nature 1998; 394:595-9.

7. Iwaki T, Iwaki A, Tateishi J, Goldman JE. Sense and antisense modification of glial alpha B-crystallin production results in alterations of stress fiber formation and thermoresistance. J Cell Biol 1994; 125:1385-93.

8. Mitton KP, Tumminia SJ, Arora J, Zelenka P, Epstein DL, Russell P. Transient loss of alphaB-crystallin: an early cellular response to mechanical stretch. Biochem Biophys Res Commun 1997; 235:69-73.

9. Arai H, Atomi Y. Chaperone activity of alpha B-crystallin suppresses tubulin aggregation through complex formation. Cell Struct Funct 1997; 22:539-44.

10. Nicholl ID, Quinlan RA. Chaperone activity of alpha-crystallins modulates intermediate filament assembly. EMBO J 1994; 13:945-53.

11. Djabali K, de Nechunda B, Landon F, Portier MM. AlphaB-crystallin interacts with intermediate filaments in response to stress. J Cell Sci 1997; 110:2759-69.

12. Horwitz J. Alpha-crystallin can function as a molecular chaperone. Proc Natl Acad Sci U S A 1992; 89:10449-53.

13. Plater ML, Goode D, Crabbe MJ. Effects of site-directed mutations on the chaperone-like activity of alphaB-crystallin. J Biol Chem 1996; 271:28558-66.

14. Wiesmann KE, Coop A, Goode D, Hepburne-Scott HW, Crabbe MJ. Effect of mutations of murine lens alphaB crystallin on transfected neural cell viability and cellular translocation in response to stress. FEBS Lett 1998; 438:25-31.

15. Jakob U, Gaestel M, Engel K, Buchner J. Small heat shock proteins are molecular chaperones. J Biol Chem 1993; 268:1517-20.

16. Chitpinityol S, Goode D, Crabbe MJ. Studies on the binding of alpha-crystallin to recombinant prochymosins and chymosin. Mol Vis 1998; 4:1 <http://www.molvis.org/molvis/v4/p1/>.

17. Muchowski PJ, Clarke JI. ATP-enhanced molecular chaperone functions of the small heat shock protein human alphaB crystallin. Proc Nat Acad Sci U S A 1998; 95:1004-9.

18. Crabbe MJ, Goode D. alpha-Crystallin: chaperoning and aggregation. Biochem J 1994; 297:653-4.

19. Goode D, Crabbe MJ. Heterologous expression of gamma E-crystallin produces protein with aberrant tertiary structure. Arch Biochem Biophys 1994; 315:104-10.

20. Haley DA, Horwitz J, Stewart PL. The small heat-shock protein, alphaB-crystallin, has a variable quaternary structure. J Mol Biol 1998; 277:27-35.

21. Horwitz J, Bova M, Huang QL, Ding L, Yaron O, Lowman S. Mutation of alpha B-crystallin: effects on chaperone-like activity. Int J Biol Macromol 1998; 22:263-9.

22. Vicart P, Caron A, Guicheney P, Li Z, Prevost MC, Faure A, Chateau D, Chapon F, Tome F, Dupret JM, Paulin D, Fardeau M. A missense mutation in the alphaB-crystallin chaperone gene causes a desmin-related myopathy. Nat Genet 1998; 20:92-5.

23. Sharma KK, Kumar GS, Murphy AS, Kester K. Identification of 1,1'-bi(4-anilino)naphthalene-5,5'-disulfonic acid binding sequences in alpha-crystallin. J Biol Chem 1998; 273:15474-8.

Typographical corrections