Biochemical and biophysical characterization of human small heat shock proteins

Small heat shock proteins play an important role in maintaining the quality of all proteins in a cell, both under normal and stress conditions and are part of a large network called the protein quality control network. sHSPs can be considered as a first line of defense against unfolding protein species under stress conditions, this is because of their ATP-independent mode of action. These chaperones can bind partially unfolded proteins and trap them in a folding competent state. Because of their importance in proteostasis, it is not surprising that these proteins have been associated with many human diseases. Many sHSPs co-localize with amyloid deposits in debilitating neurodegenerative diseases such as Parkinson’s and Alzheimer’s disease. Mutation of sHSPs lead to congenital forms of myopathies, neuropathies and cataract and upregulation of sHSP genes is seen during cancer development. This makes them an interesting target for developing new drugs, although to date attempts have been limited. All sHSPs share the same architectural arrangement: a central conserved domain of approximately 90 residues called the α-crystallin domain, flanked by unstructured and N- and C-terminal arms that are considered to be poorly conserved. All sHSPs assemble into dimers via their α-crystallin domain and assemble further into large oligomers that can comprise up to 40 subunits. This assembly is dependent on interactions mediated by the N- and C-terminal arms, although details about the interacting residues are missing. Most mammalian sHSPs assemble into polydisperse oligomers that can vary in both size, shape and the number of subunits and can even contain different sHSPs expressed in the same tissue, which makes these proteins nearly impossible to crystallize. The chaperone activity of these proteins is also linked to the N-terminal region, but due to the low conservation and lack of secondary structure, studies regarding the structure and function of these proteins have been an arduous task. This is mainly due to the overlapping role the N-terminal region plays in both structure and function, so truncation or mutation usually has a concerted effect on size and chaperone activity, making it hard to delineate the sequence properties. This research project focuses on characterizing the structure and function two human small heat shock proteins, HSPB1 and HSPB6. Both proteins reach high levels of expression in smooth muscle tissue and are known to hetero-oligomerize. In order to gain insights into the specific sequences necessary for both chaperoning and assembly, we have used HSPB6 as a model sHSP. This particular sHSP does not form the typical large oligomeric assemblies but only forms dimers in solution. This property makes HSPB6 an excellent model to study the functional epitopes, as mutation or truncation is unlikely to affect the size of the protein. Furthermore, this protein hetero-oligomerizes with HSPB1, so the sequences involved in this interaction can also be investigated. As a first objective, the sequence determinants that define the chaperone activity of HSPB6 were determined by creating a library of deletion constructs in which 10 amino acids were deleted stepwise. These truncations were all characterized using size-exclusion chromatography and small-angle X-ray scattering to determine the effect of truncation on the structure and we have found that all were still dimeric in solution. Therefore, the activity of each was assessed using standard chaperone assays in which an aggregation-prone protein was incubated with different amounts of a sHSP. All of the truncations, except for a complete removal of the NTR were still capable of protecting insulin and yeast alcohol dehydrogenase – two standard substrates – from aggregation. Deletion of residues 41 to 60 of the NTR led to a reduced activity when compared to the wild type, although differences in the chaperoning profile for the two substrates were observed. This suggests that likely multiple regions within the NTR are necessary for chaperoning and that these regions may even be redundant, or display substrate specificity. Surprisingly, deletion of a central conserved stretch in the NTR, encompassing residues 31-40 led to a vast increase in activity. Further mapping using smaller deletions and point mutations, have shown that the glutamate at position 31 functions as a negative regulator of activity, and that the residues surrounding it also affect its regulatory function. Even though HSPB6 does not assemble into multisubunit oligomers, it does display concentration dependent self-association in solution. This non-ideal behavior is also regulated by the conserved residues found in the 31 to 35 region, where the phenylalanine at position 33 regulates this specific property. Again, this feature is also modulated by the surrounding residues, as mutation of Glu31 leads to increased self-association. It thus seems that residues in the N-terminal region of HSPB6 that are conserved throughout metazoan sHSPs, have an important regulatory role. The results described in Chapter 3, hint at a fine-tuned interplay between these residues in regulating both assembly and function of sHSPs. To further investigate the effect of the NTR in defining the structure of sHSPs, the same library of deletion constructs and mutants was used to assess hetero-oligomerization with HSPB1. HSPB6 and HSPB1 have been shown to form hetero-oligomers in vivo and in vitro. Size exclusion chromatography and disulfide crosslinking (using a mutant of HSPB6 that is capable of forming a disulfide crosslinked dimer) have shown that this complex consists of a highly polydisperse mix containing two main species of approximately 500 kDa and 150 kDa, both built up of heterodimers. We have shown, using native mass spectrometry that this heterocomplex is built up of 100% heterodimers, even though both proteins on their own are capable of exchanging subunits in a stochastic fashion. Large truncations, where only the NTR (ΔN) or both the NTR and CTR (ACD) was removed, showed that the NTR is the key player in dictating this preferential interaction between the wild type proteins, as both truncations led to a mere stochastic exchange between the two sHSPs. The conserved region, identified as negative regulator of activity in HSPB6 before, was also found to be essential for complex formation with HSPB1. Deletion of this region in HSPB6 prevented the formation of a complex, whereas deletion of the residues adjacent to this region (36 to 40) were necessary for the preferential interaction between HSPB1 and HSPB6. These results are described in Chapter 4. Thus, overall we have found an essential role for the NTR in regulating both chaperone activity and assembly of a human sHSP. An evolutionary conserved region found in the middle of the NTR functions as both a negative regulator of activity and contains the necessary residues for interaction with another sHSP. Using the same set of technique, the effect of mutations in the ACD of HSPB1 that cause hereditary neuropathies was also investigated. The results described in Chapter 5 clearly show that most of these mutations increased the size of the protein and some, especially S135F had an increased chaperone activity. By investigating the same mutants using the ACD only constructs, we have shown that the effect of mutation does not lie within the ACD itself, as its structure was unaffected by mutation. The S135F however, had an unclear effect on the ACD structure as shown by small-angle X-ray scattering although the molecular mass still corresponded to an ACD dimer. Nonetheless, it thus seems that again the NTR is affected by these mutations as both the size and chaperone activity are regulated by sequences found in the N-terminal region. In summary, we have extensively characterized two human small heat shock proteins: HSPB1 and HSPB6. We have investigated the role of the NTR in defining two traits most commonly associated with sHSPs – chaperoning and assembly – and have found an essential role for a conserved region within the NTR. In an attempt to characterize disease-causing mutants of HSPB1 localized in the ACD, we found that although most did not have a profound effect on the ACD core structure, all full length proteins were larger and had a changed chaperoning profile against our standard substrate proteins. This suggests again an essential role for the NTR in disease pathology and we therefore suggest that future experiments should focus on the effect of mutation on the properties defined by the ACD-flanking regions. Furthermore, the techniques outlined in this thesis provide a useful toolbox for characterizing sHSPs and could serve as a guide for future experiments.nrpages: 187status: publishe