The canonical function of small heat-shock proteins (sHSPs) is to interact with proteins destabilized under conditions of cellular stress. While the breadth of interactions made by many sHSPs is well-known, there is currently little knowledge about what structural features of the interactors form the basis for their recognition. Here, we have identified 83 in vivo interactors of the sole sHSP in the cyanobacterium Synechocystis sp. PCC 6803, HSP16.6, reflective of stable associations with soluble proteins made under heat-shock conditions. By performing bioinformatic analyses on these interactors, we identify primary and secondary structural elements that are enriched relative to expectations from the cyanobacterial genome. In addition, by examining the Synechocystis interactors and comparing them with those identified to bind sHSPs in other prokaryotes, we show that sHSPs associate with specific proteins and biological processes. Our data are therefore consistent with a picture of sHSPs being broadly specific molecular chaperones that act to protect multiple cellular pathways. . . . . Keywords Small heat-shock protein α-Crystallins Molecular chaperone Cyanobacteria Bioinformatics Introduction (Basha et al. 2012; Hilton et al. 2013; McHaourab et al. 2009). Although sHSPs are relatively small as monomers (12 to Small heat-shock proteins (sHSPs) are a diverse family of pro- 42 kDa), the majority assemble into large oligomers. These teins that share a conserved ≈ 90-residue α-crystallin domain range in size from 12 to > 40 subunits, with some family mem- (ACD) that is flanked by variable N- and C-terminal regions bers being monodisperse and others forming polydisperse en- sembles (Basha et al. 2012;Hiltonetal. 2013; McHaourab et al. 2009). Found in all kingdoms of life, many sHSPs have been The original version of this article was revised: Table 1 needed corrections. demonstrated in vitro to act as ATP-independent molecular The DOI of the Erratum is: https://doi.org/10.1007/s12192-018-0901-6 chaperones with the ability to capture denaturing proteins in a partially unfolded form such that they can be reactivated by the Electronic supplementary material The online version of this article (https://doi.org/10.1007/s12192-018-0884-3) contains supplementary cell’s ATP-dependent chaperones. Recent reviews have de- material, which is available to authorized users. scribed models for this canonical mechanism of sHSP chaper- one action; however, details are derived primarily from in vitro * Justin L. P. Benesch studies with recombinant proteins and model interactors from email@example.com non-homologous organisms (Haslbeck and Vierling 2015; * Elizabeth Vierling Treweek et al. 2015). Thus, a major gap in our understanding firstname.lastname@example.org of sHSP mechanism is the considerable lack of information about which substrates they protect in the cell. Department of Chemistry, Physical & Theoretical Chemistry In order to investigate the properties of proteins that are Laboratory, University of Oxford, Oxford OX1 3QZ, UK sHSP interactors, we identified HSP16.6 from the single- Department of Chemistry – BMC, Uppsala University, Box 576, celled cyanobacterium Synechocystis sp. PCC 6803 (hereafter Uppsala 75123, Sweden 3 Synechocystis) as an ideal system to interrogate. HSP16.6 is Department of Biochemistry & Molecular Biology, University of the only sHSP in Synechocystis (Giese and Vierling 2002;Lee Massachusetts, Amherst, MA 01003, USA 4 et al. 2000). It is strongly induced at high temperature, and Present address: Alorica, Inc., Irvine, CA, USA cells deleted for HSP16.6 (Δ16.6) grow normally at optimal Present address: Department of Molecular and Cellular Biology, growth temperature but are sensitive to heat stress (Giese and University of Arizona, Tucson, AZ 85721, USA 724 E. G. Marklund et al. Vierling 2002, 2004). The temperature-sensitivity phenotype spectinomycin gene and HSP16.6 carrying a Strep-tag II af- of Δ16.6 cells has enabled studies of sHSP properties required finity tag (WSHPQFEK) on the C-terminus (HSP16.6-Strep for activity in vivo in a homologous system. Crucially, point strain) (Basha et al. 2004). This HSP16.6-Strep strain had mutations in the N-terminal domain were found to decrease been shown previously to behave like wild type in assays of heat tolerance in vivo, but to have no effect on the efficiency heat tolerance (Basha et al. 2004), and recombinant HSP16.6- of chaperone function in assays with model substrates in vitro strep protein was equivalent to untagged protein in assays of (Giese et al. 2005). This observation emphasizes the need to chaperone activity in vitro (Friedrich et al. 2004). identify native interactors of sHSPs and renders Synechocystis Cells were grown in 50-mL cultures at 30 °C as described an excellent system with which to do so. previously to A ≈ 0.2 (Basha et al. 2004) and then subjected We previously used immunoprecipitation and mass spectrom- to treatmentat42°Cfor2hfollowedby1 hrecovery at 30 °C, etry (MS)-based proteomics to identify 13 proteins associated to allow accumulation of HSP16.6-Strep protein. Control sam- in vivo with HSP16.6 from Synechocystis cells that had been ples were prepared directly after this treatment, while heat- heat-stressed prior to cell lysis (Basha et al. 2004). Notably, these stressed samples were treated for an additional 30 min at 13 proteins were not detected in equivalent pull-downs from 46 °C. To control for interaction of HSP16.6-Strep protein dur- cells that had not been heat-stressed, or when recombinant ing sample processing, recombinant HSP16.6-Strep protein was HSP16.6 was added to heat-stressed Δ16.6 cells before lysis added to heat-stressed samples of the ΔHSP16.6 strain directly (to control for sHSP-protein interactions that might occur in the after heat treatment at a concentration matching that in heat- lysate, as opposed to during heat stress in vivo). Although these stressed cells. Cells were harvested, suspended in 1.5 mL lysis proteins were associated with the sHSP in the soluble cell frac- buffer (25 mM HEPES-KOH, 0.2 M NaCl, 0.5% Triton X-100, −1 tion, they were also found in the insoluble cell fraction after heat 5mM ϵ-aminocaproic acid, 1 mM benzamidine, 1 μgmL stress (Basha et al. 2004). All of these proteins, whose functions leupeptin, and 1 mM EDTA, pH 7.5), and opened as described span a variety of cellular processes, including translation, tran- previously (Basha et al. 2004). The soluble fraction was mixed scription, secondary metabolism, and cell signaling, could be with 30 μL of Strep-Tactin resin (Sigma) at 4 °C for 2 h. Resin released from the immunoprecipitate by addition of DnaK, co- was washed six times in lysis buffer, and bound proteins were chaperones, and ATP (Basha et al. 2004). In addition, one of eluted using either sample buffer (for SDS-PAGE) or isoelectric these interactors, a serine esterase, when purified, was shown focusing (IEF) rehydration buffer (for 2D gels) (7.0 M urea, to be heat sensitive and to associate with HSP16.6 and thereby 2.5 M thiourea, 2% CHAPS, 2% IPG buffer pH 3–10 NL −1 be protected from insolubilization (Basha et al. 2004). While (Amersham Biotech), and 3 mg mL dithiothreitol). these data identified 13 proteins as potential interactors for ca- For 2D gel analysis, pH 3–10 NL first dimension strips nonical sHSP chaperone function, their relatively small number (18 cm; Amersham Biotech) were rehydrated overnight at meant it was not possible to derive any common protein features room temperature using 600 μL of sample in IEF rehydra- that might dictate interaction with the sHSP. tion buffer. IEF was carried out for 2 h at 150 V, 2 h at Here, we have extended the identification of HSP16.6- 300 V, 5 h at 500 V, and 7 h at 3500 V. The second dimen- interactors to a total of 83 proteins by performing an affinity sion was separated by 11–17% SDS-PAGE for 30 min at pull-down from heat-stressed Synechocystis.Byperforming 15 mA and then for 7 h at 25 mA. Samples were also rigorous bioinformatic analyses, we provide new insights into separated by SDS-PAGE according to standard protocols, the primary and secondary structural properties of proteins using 8% acrylamide gels in order to afford good separa- that interact with sHSPs in the soluble cell fraction during tion of proteins above 100 kDa, which are typically not stress. We also catalogue the functions of the interactors and well resolved on the 2D system. Gels were silver stained compare these to sHSP interactors previously identified in two according to a previous protocol (Rabilloud 2012). other prokaryotes, Escherichia coli and Deinococcus radiodurans (Bepperling et al. 2012; Fu et al. 2013). Our Protein identification by means of mass spectrometry combined results indicate that sHSPs protect a specific yet diverse set of proteins from aggregation in the cell. Proteins unique to the heat-stressed HSP16.6-Strep sam- ple were excised from 1D or 2D gels and digested with trypsin, and peptides were prepared for MS as described Methods previously (Basha et al. 2004). Peptideextractswerein- troduced onto a 100-μm I.D. × 5-cm C18 column using an Affinity isolation of HSP16.6-interacting proteins autosampler and separated with a 25-min gradient of 2– 100% acetonitrile in 0.5% formic acid. The column eluate Isogenic Synechocystis strains were used in which the wild- wasdirectedinto aThermoFinniganLCQ Decaiontrap type HSP16.6 gene had been replaced with a spectinomycin mass spectrometer. The mass range scanned was 400 to resistance gene (aadA gene) (ΔHSP16.6 strain) or with the 1500 m/z, and data-dependent scanning was used to select Structural and functional aspects of the interaction partners of the small heat-shock protein in... 725 the three most abundant ions in each parent scan for tan- Statistical significance testing and representation dem MS. Peptides were searched using SEQUEST and allowed for static modification of Cys (57 Da; A bootstrapping approach was employed to assess the statis- iodoacetamidation), and differential modification of Met tical significance of any differences between I and G.First, a (16 Da; oxidation) was considered. X correlation cutoffs random subset, R, was taken from G by arbitrarily picking, of 2.0 for 2+ ions, 3.0 for 3+ ions, and delta Xcorr > 0.05 with replacement, of 83 proteins (i.e., R ⊆ G and |R|=|I|). The were applied, and data were sorted using DTASelect mean, Q , was then calculated for the given quantity of inter- (Tabb et al. 2002). The complete list of 83 proteins iden- est Q, to allow comparison with Q , the mean calculated from tified as HSP16.6 interaction partners from these and our I for the same quantity. This was repeated N times, after which previous experiments (Basha et al. 2004) is given in the p value was calculated as the frequency by which Q ≥Q R I Supplemental Table 1. For the purpose of comparisons or Q ≤Q , in the respective cases of Q > Q and Q < Q . R I I G I G and calculations, this set is considered to represent sHSP For each quantity, a total of N = 100,000 iterations was run, interactors and denoted I,where |I| = 83. Known protein- and the statistical significance was tested at the 0.01 level. protein interactions (PPIs) from yeast-2-hybrid experi- Kernel density estimates were plotted for all quantities ments are available for Synechocystis (Sato et al. 2007). where a statistically significant difference was found. A We identified all PPIs made by members of I Gaussian kernel with a bandwidth equal to 2% of the visible (Supplemental Table 1), excluding PPIs that were not range was used in all cases and the amplitude was set such that identified with multiple positive prey clones, in order to the integrated density was equal to the number of proteins in avoid false positives. each set. As such, the amplitudes are inversely proportional to the ranges along the x-axis, and their heights can thus differ Bioinformatic analyses substantially between distributions. Moreover, the y-axes’ ranges were chosen to make the I and G distributions occupy The Synechocystis sp. PCC 6803 genome (Kaneko et al. 1995; the same visible area in the resulting plot. Kotani et al. 1995) was obtained from CyanoBase, http:// genome.microbedb.jp/cyanobase/ (Nakamura et al. 1998). A set G representing the genome, containing all proteins such Biological function analysis that I ⊆ G, was created from the protein-coding sequences in the genome. Only proteins with estimated isoelectric point (pI) A PANTHER Overrepresentation Test (release 20170413) within the range 4–9.5 and mass m between 10 and 200 kDa, against the GO Ontology database (release 20170926) was corresponding to the range of proteins that could be identified made for all proteins in I, using the Synechocystis reference in either the 1D or 2D gels, were included (see Supporting list and the BGO biological process complete^ annotation data Information). This filtering resulted in G comprising 3021 set. Bonferroni correction was applied for multiple testing, proteins (i.e., |G| = 3021), which amounts to > 80% of the and a p value cut-off of 0.05 was used to filter the results. proteins encoded in the genome. Proteins that were not mapped to any entry in the reference The mass, sequence length n , and abundance (abso- list were addedtothe setof Bunclassified^ proteins. aa lute numbers n and frequencies f = n /n ) of various Enrichment was defined as n /E(n ), where n is the number F F F aa p p p sequence features F were determined for every protein. of proteins in I being ascribed to biological process p, and These were DnaK-binding motifs; VQL, IXI, and [I/L/ E(n ) is the expected number of such proteins based on their V]X[I/L/V] motifs (where X refers to any amino acid); frequency in G and the size of I.Proteinsthat wereassigned to charged (D,E,H,K,R), positive (H,K,R), negative (D,E), the GO-class Bbiological process^ but not to any of its sub- and hydrophobic (C,F,I,L,M,V,W) residues. DnaK- classes were given the collective label Bother biological binding motifs were identified using a previously de- process.^ Since a single protein can have multiple classifica- scribed algorithm (Van Durme et al. 2009), and the other tions, the sum of proteins in the different classes exceeds |I|. motifs were found through regexp pattern-matching using Protein BLAST (Altschul et al. 1990) was used to find the Python Standard Library. Long-range disorder was orthologs among the interactors identified for HSP16.6 in predicted with IUPred (Dosztanyi et al. 2005a, b)using Synechocystis,IbpB in E. coli (Fu et al. 2013), and HSP20.2 default parameters, and residues with a score > 0.5 were in D. radiodurans (Bepperling et al. 2012). Three pairwise considered unstructured. For the remainder, secondary comparisons were made to define the overlap between the sets structure was predicted from the sequences using the of interactors, where the list of interactors from one organism EMBOSS (Rice et al. 2000) implementation of the GOR was used as the Bdatabase^ and the list of interactors from the method. β-strands and β-turns were pooled together into other as the Bquery.^ Using E. coli as the database yielded Bβ-structures.^ Average abundances were calculated sep- poorly annotated hits; hence, the primary database was set to arately for I and G. be Synechocystis and the secondary database to be 726 E. G. Marklund et al. −10 D. radiodurans.An E value cut-off of 10 was used for all hybrid data (Sato et al. 2007) (Supplementary Table 1), nota- BLAST searches, and whenever a protein in the query yielded bly there are only three described pairwise PPIs within I,and several matches the one with the lowest E value was chosen. all three of these are self-associations. To see if this low count Lastly, the overlap between E. coli and D. radiodurans was was an artifact from our conservative approach of excluding used as a query against Synechocystis in order to find the PPIs that were identified with only one prey clone, we also overlap between the interactors in all three organisms. The tested including the latter, which presumably yields more false triply overlapping set of proteins were also analyzed for an positives. This increased the number of pairwise PPIs within I overrepresentation test in Synechocystis as described above, to 12, including six self-interactions, which is still a small but without imposing a p value because of the small number of subset of I. Consequently, the proteins in I appear largely proteins in the query. independent of each other in their interaction with HSP16.6, consistent with our affinity-isolate methodology being sensi- tive to stable interactors. Results Primary- and secondary-structure features of HSP16.6 Identification of proteins associated with HSP16.6 interactors during heat stress in vivo We first compared the average mass and sequence lengths of To identify a larger number of HSP16.6-associated proteins the interactors to the genome. We found that these were very than we did previously (Basha et al. 2004), we developed a different, with the interactors being about 60% larger on av- Synechocystis strain in which the wild-type HSP16.6 gene erage (Table 1,Fig. 2a, b). While this is informative about the was replaced with an HSP16.6 gene modified to encode a interactor profile of HSP16.6, it also means that the absolute Strep-tag II at the C-terminus. HSP16.6-Strep was shown to number, n of any feature F, is likely to be larger for the complement HSP16.6 in vivo in thermotolerance assays interactors. To account for this, all subsequent analyses are (Basha et al. 2004), as well as functioning in vitro to protect consequently focused on fractional quantities, f , which are model interactors from irreversible heat-denaturation normalized by sequence length in order to reveal distinctive (Friedrich et al. 2004). features for the proteins associated with HSP16.6. The HSP16.6-Strep strain and an isogenic strain carrying We judged that certain sequence motifs might be im- wild-type HSP16.6 were subjected to mild heat stress to allow plicated in the association of interactors with sHSPs. To accumulation of the sHSP and then to a short, more severe develop hypotheses for testing, we considered a model in heat stress to maximize association of thermally unstable pro- which interfaces that allow the sHSP to self-assemble teins with the sHSP. The soluble cell fraction from control and might be the same as interactor binding sites (Jacobs heat-stressed cells of the HSP16.6-Strep and HSP16.6 strains et al. 2016). In this context, the inter-monomer contact was subjected to Strep-Tactin affinity chromatography and the made between the highly conserved BIXI^ motif in the recovered proteins compared by means of 2D electrophoresis C-terminal region and the β4–β8 groove of the ACD (or, to examine high molecular mass proteins, by using 1D hasbeenproposedasanauto-inhibitory interface (Jehle electrophoresis) (Fig. 1). Individual spots or bands unique to et al. 2010; van Montfort et al. 2001). Theorizing that IXI proteins affinity-purified with HSP16.6-Strep from the heat motifs might mediate contacts with the sHSPs, we there- stress samples were excised and subjected to MS analysis. fore asked whether they were differentially represented in We identified a total of 72 proteins in these experiments the interactors. We also posed this question in a more which, when combined with others we had identified previ- general form, by searching for motifs matching the re- ously (Basha et al. 2004), expanded to a total of 83. Notably, quirement [I/L/V]X[I/L/V], which is more encompassing the proteins were recovered from the soluble fraction, so they across the breadth of sHSPs (Poulain et al. 2010). do not represent those that underwent excessive aggregation, Furthermore, we searched for VQL motifs, as this corre- or associations with membranes and cytoskeletal elements sponds to the specific manifestation of the BIXI^ in that may have led to partitioning into the pellet. As such, these HSP16.6. Comparing the fractional abundance of these proteins represent potential sHSP interactors that have been motifs (f , f , f , respectively) between the IXI [ILV]X[ILV] VQL prevented from insolubilization by interaction with HSP16.6. interactors and the genome, we found there to be no We denote this set of interactors I, representing a subset of the meaningful difference for IXI and VQL, but the general genome G detectable in our experiments. This allows us to test form [I/L/V]X[I/L/V] was significantly under-represented hypotheses about the features of these interactors to shed light in the interactors (Table 1,Fig. 2c). on what distinguishes them from the other proteins in sHSPs are thought to transfer interactors to the DnaK Synechocystis.Thoughmanyofthe interactorshaveknown (HSP70 in eukaryotes) system for ATP-dependent refolding PPIs, based on cross-referencing to genome-wide yeast-2- (Haslbeck and Vierling 2015). We therefore hypothesized that Structural and functional aspects of the interaction partners of the small heat-shock protein in... 727 Fig. 1 Identification of HSP16.6 interactors. a SDS-PAGE separation of Double-width dashes indicate bands that gave hits for proteins associated proteins recovered in association with HSP16.6-Strep in cells grown at with protein-folding processes. b 2D gel separation of samples prepared 30 °C and treated at 42 °C for 2 h plus 1 h recovery at 30 °C to allow as described in a. The position of molecular mass markers and the acidic sHSP accumulation (control sample, C) or further treated with an addi- (+) and basic (−) sides of the silver-stained 2D gels are indicated. Spots tional 30 min at 46 °C (heat-stressed sample, HS). To recover proteins in that were excised and yielded the reported data are annotated with red the high molecular mass range, separation was performed using an 8% circles (right panel). The ellipse in each panel indicates the spots due to acrylamide gel, and the position of molecular mass markers is indicated. HSP16.6 Bands that were excised for analysis are annotated with red dashes. the presence of DnaK-binding motifs (Rudiger et al. 1997), (f ), we discovered it to be higher in the interactors. By Charged which mediate association with this downstream chaperone, investigating negatively and positively charged residues sep- might be different between the interactors and the genome. We arately (f and f , respectively), we found this difference to be − + found the fractional abundance of DnaK motifs (f )tobe> due to the former, with negatively charged residues > 16% DnaK 30% lower in the interactors (Table 1,Fig. 2d). more abundant in the interactors. Conversely, the genome We next considered electrochemical properties of the pro- contains a higher fractional abundance of hydrophobic resi- teins. The difference in pI between the interactors and genome dues (f )(Table 1, Fig. 2e–g). H-phobic was just outside our significance criterion (p =0.036 >0.01). Lastly, we asked whether predicted secondary structure However, when examining the fraction of charged residues differed between the two sets. The fraction of residues in disordered regions (f ) is insignificantly higher in the interactors, albeit very near our threshold (p = Table 1 Comparison of various primary- and secondary-structure fea- 0.015 ≈ 0.01). For the structured regions, on average, tures between interactors of HSP16.6 in Synechocystis with the wider the interactors had a higher fraction of residues in helices genome. Mean values obtained for the proteins in I and G,along with p values for the differences between them (f ) and lower fraction in β-structures (f ), compared to α β the proteins in the wider genome (Table 1,Fig. 2h, i). Quantity Interactors, I Genome,Gp value −5 Functional classification of HSP16.6-associated m/Da 57,860 36,561 < 10 −5 proteins n 525 336 < 10 aa −5 f 0.0198 0.285 < 10 DNAK Where possible, interactors were classified according to their f 0.000335 0.000274 0.27 VQL gene-ontology annotation into either Bmetabolic process,^ f 0.00349 0.00305 0.15 IXI Bcellular process,^ or Bother biological process.^ Many pro- f 0.0378 0.0426 0.002 [ILV]X[ILV] teins were assigned to multiple classes, and 15 proteins could pI 5.22 5.63 0.036 −5 not be matched to the reference list and were added to the set f 0.252 0.230 6.0∙10 Charged of unclassified proteins, which then comprised 24 proteins. f 0.118 0.115 0.24 −5 This classification yielded different distributions of processes f 0.134 0.114 < 10 −5 in I and G (Fig. 3a), indicating that HSP16.6 has an interaction f 0.309 0.331 1.0∙10 H-phobic profile that reflects the biological function of its interactors. To f 0.086 0.058 0.015 quantify the differences, we calculated the overrepresentation −5 f 0.355 0.415 < 10 −5 of proteins involved in the various biological processes (Fig. f 0.383 0.338 3.1∙10 3b). The data reveal statistically significant enrichment of pro- Bold text indicates statistically significant differences, defined as p <0.01 teins ascribed to certain biological processes in the interactors, 728 E. G. Marklund et al. Fig. 2 Probability distributions of the statistically significant differences negatively charged residues and a lower fraction of hydrophobic residues. identified in Table 1. a, b The distributions of protein mass (a) and h, i Fraction of residues with predominately helical (α and 3 , h)pro- sequence length (b)for I and G. The proteins in I are on average pensity and β-structure (sheet and turn, i). The helix content is higher in I approximately 60% larger than those in G, both in terms of mass and than in G, and conversely, the β-structure content is lower in I. The sequence length. c, d Distributions of frequencies of [I/L/V]X[I/L/V] distributions were normalized such that their integral equals the number motifs (c) and DnaK-binding motifs (d). Both sequence features are less of proteins in each set. Consequently, the amplitudes are inversely pro- frequent and more narrowly distributed in I. e–g The fraction of hydro- portional to the width of the distributions, and the amplitudes of the two phobic (e), charged (f), and negative (g) residues. Charged residues are distributions in each panel reflect the different sizes of the two sets more frequent in I, which can be attributed to a higher fraction of suggesting that HSP16.6 makes function-specific interactions. one would expect by chance (approximately 3 for each The most striking association was for proteins involved in pairwise overlap, and fewer than 1 for the triple overlap). protein folding, with 6 out of the 19 known such proteins Interestingly, these proteins were also diverse, spanning being found in I (Table 2), corresponding to a thirteen-fold multiple biological processes, with only one eluding clas- enrichment. sification (Table 3,Fig. 3b inset). With the exception of the To compare HSP16.6-interactors with those identified in Bprotein folding^ and Bother biological process,^ which other prokaryotes, we cross-referenced our list with those were not represented at all in this subset, all categories reported as IbpB interactors in E. coli (Fu et al. 2013), and were even more overrepresented than in the complete list HSP20.2 in D. radiodurans (Bepperling et al. 2012)(Fig. of HSP16.6 interactors. We note that the small number of 3c). There were unique orthologs for 17 HSP16.6 proteins precluded low p values for the levels of enrich- interactors among the 113 IbpB interactors and 17 for the ment for the individual categories. Taken together, they 101 HSP20.2 interactors. The overlap between IbpB and nonetheless indicate that the enrichment pattern seen for HSP20.2 interactors was larger still, comprising 36 unique the Synechocystis interactors is particularly prominent for orthologs. A total of 10 proteins were found in all three sets the interactors that are common for all three sHSPs, with of interactors. Notably, these overlaps are much larger than the striking exception of the protein-folding interactors, Structural and functional aspects of the interaction partners of the small heat-shock protein in... 729 Fig. 3 Classification of proteins involved in different gene-ontology an- Proteins involved in protein folding are enriched thirteen-fold, with 6 of notations of biological processes. a Pie charts show the extent of different the 19 such proteins known being found among the interactors. Inset: classes in I and G. The most fundamental classes have labels in bold face. Same analysis performed for the 10 overlapping proteins from the anal- Note that Bcellular metabolic process^ belongs to both Bmetabolic ysis in (c). In all featured classes, the fold-enrichment is higher. c Venn process^ and Bcellular process^ and is therefore represented by two diagram showing the overlap of sHSP interactor ranges from colors. b Enrichment within I of proteins taking part in the various bio- Synechocystis, E. coli,and D. radiodurans. Note that, with the exception logical processes. Circle areas reflect the number of proteins in I,and of the intersection of the three sets, all areas of the diagram reflect the numbers indicate proteins in I and G. I contains a smaller fraction of number of elements within unclassified proteins than G, and all classes are somewhat enriched in I. which might be a species- or sHSP-specific phenomenon, Discussion or the result of differences in the methods used for recov- ering interacting proteins. Here, we have examined the properties of 83 proteins that associate in vivo with HSP16.6 under conditions of heat stress. Given that the proteins were obtained from the soluble supernatant after centrifugation, they are likely to under-represent membrane- and cytoskeleton- Table 2 The six interactors of Synechocystis HSP16.6 annotated as belonging to the Bprotein folding^ category associated proteins. Furthermore, as our experiment in- volves affinity pull-downs, these interactors are inevitably Gene UniProt ID Name restricted to those that form interactions that are stable on the timescale of the experiment. In the context of the sll0058 Q55154 DnaK 1 model proposed for sHSPs wherein they display both a sll0170 P22358 DnaK 2 low-affinity mode with high capacity, and a high-affinity sll1932 P73098 DnaK 3 mode with low capacity (McHaourab et al. 2009), our slr2076 Q05972 60 kDa chaperonin 1 interactors are likely representative of the latter. sll0533 Q55511 Trigger factor (TF) Notwithstanding these potential biases of the experiment, slr1251 P73789 Peptidyl-prolyl cis-trans isomerase we have shown that the interactors were on average larger 730 E. G. Marklund et al. Table 3 Proteins that we associated to all three of HSP16.6 process (BP), organic substance metabolic process (OSMP), cellular met- (Synechocystis), IbpB (E. coli), and HSP20.2 (D. radiodurans). The abolic process (CMP), and unclassified (U). In some cases, two distinct GO annotations for biological processes are coded as follows: IbpB or HSP20.2 interactors would correspond to an HSP16.6 interactor, metabolic process (MP), cellular process (CP), nitrogen-compound met- in which case, both UniProt IDs were included in the table abolic process (NCMP), primary metabolic process (PMP), biosynthetic Synechocystis gene UniProd ID Name GO biological process Synechocystis E. coli D. radiodurans sll0018 Q55664 Fructose-bisphosphate aldolase, class II MP, CP, NCMP, PMP, OSMP, CMP G64976n NP_295312.1 sll1099 P74227 Elongation factor Tu MP, CP, NCMP, PMP, BP, OSMP, CMP NP_289744.1, pdb|1EFC|A NP_295522.1 sll1180 P74176 Toxin secretion ABC transporter CP, NCMP, PMP, OSMP NP_287490.1 ATP-binding protein NP_295291.1 sll1326 P27179 ATP synthase alpha chain MP, CP, NCMP, PMP, BP, OSMP, CMP CAA23519.1 NP_294424.1 sll1787 P77965 RNA polymerase beta subunit MP, CP, NCMP, PMP, BP, OSMP, CMP AAC43085.1 NP_294636.1 sll1789 P73334 RNA polymerase beta’ subunit MP, CP, NCMP, PMP, BP, OSMP, CMP NP_290619.1 NP_294635.1 sll1818 P73297 RNA polymerase alpha subunit MP, CP, NCMP, PMP, BP, OSMP, CMP CAA37838.1 NP_295851.1 sll1841 P74510 Pyruvate dehydrogenase dihydrolipoamide MP NP_285811.1, NP_286443.1 acetyltransferase component (E2) NP_293809.1, NP_293979.1 slr0542 P54416 ATP-dependent protease ClpP MP, NCMP, PMP, OSMP NP_286179.1 NP_295695.1 slr1105 P72749 GTP-binding protein TypA/BipA homolog U NP_289127.1 NP_294922.1 than the proteins in the genome, have a distinct electro- 2013), similar to observations made for other molecular chemical profile, an increased fraction of helical second- chaperones (Huang et al. 2016;Saioet al. 2014). ary structure, and a lower fraction of [I/L/V]X[I/L/V] and Upon considering amino acid motifs and composition, DnaK-binding motifs. we found a lower fraction of [I/L/V]X[I/L/V] motifs in the We observed that HSP16.6 preferentially binds longer, interactors. This suggests that the β4–β8 groove, which more massive, proteins. This is in agreement with analysis binds this motif intra-molecularly in sHSP oligomers of sHSP interactors E. coli and D. radiodurans (Fu et al. (Basha et al. 2012; Hilton et al. 2013), is not the binding 2014) and is interesting in light of recent data noting that site for these stable interactors. However, this does not thermally unstable proteins in cells are typically longer preclude the β4–β8groovebeinga sitefor low-affinity, than those that are stable (Leuenberger et al. 2017). or transient, interactions. This is consistent with the ob- Longer proteins might therefore be overrepresented in the servationthattheexcisedACD candisplay potent chap- interactors by virtue of being more likely to be destabilized erone activity (Cox et al. 2016;Hochberget al. 2014). We by the heat-shock condition assayed here. Alternatively, or also identified an overabundance of charged and, in par- in addition, it is possible that longer proteins, by virtue of ticular negatively charged, residues in the interactors. A having more binding sites, might be held tighter by the preponderance of charged residues was also observed for sHSPs. This would stem from avidity effects resulting sHSP interactors in E. coli and D. radiodurans (Fu et al. 2014). Notably, aspartates have been shown to be from the multivalency of sHSP oligomers (Hilton et al. Structural and functional aspects of the interaction partners of the small heat-shock protein in... 731 enriched in thermally unstable proteins (Leuenberger et al. In sum, our study provides an initial view of the functional 2017), again hinting that thermal stability could be a key interactome of prokaryotic sHSPs and of Synechocystis in attribute for recognition by sHSPs. It is also interesting to particular. In addition, the statistical framework we have im- consider the electrochemical profile of the sHSPs them- plemented for examining sequence determinants can be ap- selves, which have an overabundance of charged residues plied to the analysis of the likely future profusion of proteomic in the ACD and C-terminal region (Kriehuber et al. 2010). data identifying molecular chaperone interactors in cells. As such, it is possible that there may be charge- Acknowledgements We thank Linda Breci (University of Arizona) for complementarity aspects to binding. performing MS experiments and Georg Hochberg (University of The depletion of DnaK-binding motifs in the HSP16.6 Chicago) for helpful discussions. This work was supported by the interactors is striking, particularly when considering that Swedish Research Council and the European Commission for a Marie DnaK is able to release interactors from the complexes made Skodowska Curie International Career Grant (2015-00559) to EGM, the Biotechnology and Biological Sciences Research Council (BB/K004247/ with HSP16.6. This suggests that the DnaK-binding motif is 1) to JLPB, and the National Institutes of Health (RO1 GM42762) to EV. not responsible for the recognition events that mediate interactor transfer between the chaperones. Instead, the Open Access This article is distributed under the terms of the Creative DnaK-binding motif may be more reflective of DnaK’s Commons Attribution 4.0 International License (http:// holdase, rather than refoldase activity. In this way, proteins that creativecommons.org/licenses/by/4.0/), which permits unrestricted use, are not protected by the sHSPs are captured by HSP70 instead distribution, and reproduction in any medium, provided you give (Mayer and Bukau 2005). The interactors are also enriched in appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. α-helical propensity and depleted in β-structure. It is possible that, based on the observation that there is little cooperativity in the folding of β-sheets (Wu and Zhao 2001), this may be References reflective of physico-chemical differences in re- or unfolding. Gene-ontology analysis demonstrates that, while capa- Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local ble of associating with many interactors, HSP16.6 none- alignment search tool. J Mol Biol 215(3):403–410. https://doi.org/ theless does so with statistically significant specificity, 10.1016/S0022-2836(05)80360-2 evidenced by varying enrichments for different biological Basha E, Lee GJ, Breci LA, Hausrath AC, Buan NR, Giese KC, Vierling processes. This observation is validated by the overlap E (2004) The identity of proteins associated with a small heat shock between Synechocystis, E. coli,and D. radiodurans protein during heat stress in vivo indicates that these chaperones protect a wide range of cellular functions. J Biol Chem 279(9): sHSP interactors. The notion that sHSPs have specific 7566–7575. https://doi.org/10.1074/jbc.M310684200 interactors in the cell also extends to eukaryotes, where Basha E, O’Neill H, Vierling E (2012) Small heat shock proteins different sHSPs found in the same cellular compartment and alpha-crystallins: dynamic proteins with flexible functions. have differing interactor profiles (Fleckenstein et al. 2015; Trends Biochem Sci 37(3):106–117. https://doi.org/10.1016/j. tibs.2011.11.005 McLoughlin et al. 2016; Mymrikov et al. 2017). Bepperling A, Alte F, Kriehuber T, Braun N, Weinkauf S, Groll M, The most enriched groups of proteins associated with Haslbeck M, Buchner J (2012) Alternative bacterial two-component HSP16.6 were other components of the protein folding small heat shock protein systems. Proc Natl Acad Sci U S A 109(50): machinery. We interpret this as due to HSP16.6 being part 20407–20412. https://doi.org/10.1073/pnas.1209565109 Cox D, Selig E, Griffin MD, Carver JA, Ecroyd H (2016) Small heat- of a tightly linked molecular chaperone network (Gong shock proteins prevent alpha-synuclein aggregation via transient et al. 2009), collaborating to prevent and reverse improper interactions and their efficacy is affected by the rate of aggregation. protein interactions in the wider heat-shock response of the J Biol Chem 291(43):22618–22629. https://doi.org/10.1074/jbc. cell (Richter et al. 2010). Possibly, these interactions are M116.739250 Dosztanyi Z, Csizmok V, Tompa P, Simon I (2005a) IUPred: web server indirect, captured due to HSP16.6 and other protein-folding for the prediction of intrinsically unstructured regions of proteins components acting on the same substrates. An indirect in- based on estimated energy content. Bioinformatics 21(16):3433– teraction with protein-folding components could also ex- 3434. https://doi.org/10.1093/bioinformatics/bti541 plain the lack of equivalent proteins in the E. coli sHSP Dosztanyi Z, Csizmok V, Tompa P, Simon I (2005b) The pairwise energy content estimated from amino acid composition discriminates be- interactors (Fu et al. 2013), as the previous report tween folded and intrinsically unstructured proteins. J Mol Biol employed covalent-crosslinking and urea solubilization pri- 347(4):827–839. https://doi.org/10.1016/j.jmb.2005.01.071 or to immunoprecipitation. The D. radiodurans interactors Fleckenstein T, Kastenmuller A, Stein ML, Peters C, Daake M, were identified by a different method, employing ex vivo Krause M, Weinfurtner D, Haslbeck M, Weinkauf S, Groll M et al (2015) The chaperone activity of the developmental small addition of purified HSP20.2 to cell lysates, prior to heat heat shock protein Sip1 is regulated by pH-dependent confor- stress and immunoprecipitation. Given the differences in mational changes. Mol Cell 58(6):1067–1078. https://doi.org/ methodology between these studies, we suggest that those 10.1016/j.molcel.2015.04.019 proteins comprising common interactors are highly signif- Friedrich KL, Giese KC, Buan NR, Vierling E (2004) Interactions be- icant (Table 3). tween small heat shock protein subunits and substrate in small heat 732 E. G. Marklund et al. shock protein-substrate complexes. J Biol Chem 279(2):1080–1089. Leuenberger P, Ganscha S, Kahraman A, Cappelletti V, Boersema PJ, von Mering C, Claassen M, Picotti P (2017) Cell-wide https://doi.org/10.1074/jbc.M311104200 Fu X, Shi X, Yan L, Zhang H, Chang Z (2013) In vivo substrate diversity analysis of protein thermal unfolding reveals determinants of and preference of small heat shock protein IbpB as revealed by using thermostability. Science 355(6327):eaai7825. https://doi.org/10. a genetically incorporated photo-cross-linker. J Biol Chem 288(44): 1126/science.aai7825 31646–31654. https://doi.org/10.1074/jbc.M113.501817 Mayer MP, Bukau B (2005) Hsp70 chaperones: cellular functions and Fu X, Chang Z, Shi X, Bu D, Wang C (2014) Multilevel structural charac- molecular mechanism. Cell Mol Life Sci 62(6):670–684. https://doi. teristics for the natural substrate proteins of bacterial small heat shock org/10.1007/s00018-004-4464-6 proteins. Protein Sci 23(2):229–237. https://doi.org/10.1002/pro.2404 McHaourab HS, Godar JA, Stewart PL (2009) Structure and mech- Giese KC, Vierling E (2002) Changes in oligomerization are essential for anism of protein stability sensors: chaperone activity of small the chaperone activity of a small heat shock protein in vivo and heat shock proteins. Biochemistry 48(18):3828–3837. https:// in vitro. J Biol Chem 277(48):46310–46318. https://doi.org/10. doi.org/10.1021/bi900212j 1074/jbc.M208926200 McLoughlin F, Basha E, Fowler ME, Kim M, Bordowitz J, Katiyar- Giese KC, Vierling E (2004) Mutants in a small heat shock protein that Agarwal S, Vierling E (2016) Class I and II small heat shock proteins affect the oligomeric state. Analysis and allele-specific suppression. together with HSP101 protect eukaryotic protein translation factors J Biol Chem 279(31):32674–32683. https://doi.org/10.1074/jbc. during heat stress. Plant Physiol 172:1221-1236. https://doi.org/10. M404455200 1104/pp.16.00536 Giese KC, Basha E, Catague BY, Vierling E (2005) Evidence for an essen- Mymrikov EV, Daake M, Richter B, Haslbeck M, Buchner J (2017) The tial function of the N terminus of a small heat shock protein in vivo, chaperone activity and substrate spectrum of human small heat independent of in vitro chaperone activity. Proc Natl Acad Sci U S A shock proteins. J Biol Chem 292(2):672–684. https://doi.org/10. 102(52):18896–18901. https://doi.org/10.1073/pnas.0506169103 1074/jbc.M116.760413 Gong Y, Kakihara Y, Krogan N, Greenblatt J, Emili A, Zhang Z, Houry Nakamura Y, Kaneko T, Hirosawa M, Miyajima N, Tabata S (1998) WA (2009) An atlas of chaperone-protein interactions in CyanoBase, a www database containing the complete nucleotide Saccharomyces cerevisiae: implications to protein folding pathways sequence of the genome of Synechocystis sp. strain PCC6803. in the cell. Mol Syst Biol 5:275 Nucleic Acids Res 26(1):63–67. https://doi.org/10.1093/nar/26.1.63 Haslbeck M, Vierling E (2015) A first line of stress defense: small heat Poulain P, Gelly JC, Flatters D (2010) Detection and architecture of small shock proteins and their function in protein homeostasis. J Mol Biol heat shock protein monomers. PLoS One 5(4):e9990. https://doi. 427(7):1537–1548. https://doi.org/10.1016/j.jmb.2015.02.002 org/10.1371/journal.pone.0009990 Hilton GR, Lioe H, Stengel F, Baldwin AJ, Benesch JLP (2013) Small Rabilloud T (2012) Silver staining of 2D electrophoresis gels. Methods heat-shock proteins: paramedics of the cell. Top Curr Chem 328:69– Mol Biol 893:61–73. https://doi.org/10.1007/978-1-61779-885-6_5 98. https://doi.org/10.1007/128_2012_324 Rice P, Longden I, Bleasby A (2000) EMBOSS: the European Molecular Hochberg GK, Ecroyd H, Liu C, Cox D, Cascio D, Sawaya MR, Collier Biology Open Software Suite. Trends Genet 16(6):276–277. https:// MP, Stroud J, Carver JA, Baldwin AJ et al (2014) The structured doi.org/10.1016/S0168-9525(00)02024-2 core domain of alphaB-crystallin can prevent amyloid fibrillation Richter K, Haslbeck M, Buchner J (2010) The heat shock response: life and associated toxicity. Proc Natl Acad Sci U S A 111(16):E1562– on the verge of death. Mol Cell 40(2):253–266. https://doi.org/10. E1570. https://doi.org/10.1073/pnas.1322673111 1016/j.molcel.2010.10.006 Huang C, Rossi P, Saio T, Kalodimos CG (2016) Structural basis for the Rudiger S, Germeroth L, Schneider-Mergener J, Bukau B (1997) antifolding activity of a molecular chaperone. Nature 537(7619): Substrate specificity of the DnaK chaperone determined by screen- 202–206. https://doi.org/10.1038/nature18965 ing cellulose-bound peptide libraries. EMBO J 16(7):1501–1507. Jacobs WM, Knowles TP, Frenkel D (2016) Oligomers of heat-shock https://doi.org/10.1093/emboj/16.7.1501 proteins: structures that don’t imply function. PLoS Comput Biol Saio T, Guan X, Rossi P, Economou A, Kalodimos CG (2014) Structural 12(2):e1004756. https://doi.org/10.1371/journal.pcbi.1004756 basis for protein antiaggregation activity of the trigger factor chaperone. Jehle S, Rajagopal P, Bardiaux B, Markovic S, Kuhne R, Stout JR, Science 344(6184):1250494. https://doi.org/10.1126/science.1250494 Higman VA, Klevit RE, van Rossum BJ, Oschkinat H (2010) Sato S, Shimoda Y, Muraki A, Kohara M, Nakamura Y, Tabata S (2007) Solid-state NMR and SAXS studies provide a structural basis for A large-scale protein–protein interaction analysis in Synechocystis the activation of alphaB-crystallin oligomers. Nat Struct Mol Biol sp. PCC6803. DNA Res 14(5):207–216. https://doi.org/10.1093/ 17(99):1037–1042. https://doi.org/10.1038/nsmb.1891 dnares/dsm021 Kaneko T, Tanaka A, Sato S, Kotani H, Sazuka T, Miyajima N, Sugiura Tabb DL, McDonald WH, Yates JR 3rd (2002) DTASelect and Contrast: M, Tabata S (1995) Sequence analysis of the genome of the unicel- tools for assembling and comparing protein identifications from lular cyanobacterium Synechocystis sp. strain PCC6803. I. shotgun proteomics. J Proteome Res 1(1):21–26. https://doi.org/ Sequence features in the 1 Mb region from map positions 64% to 10.1021/pr015504q 92% of the genome. DNA Res 2(153–166):191–158. https://doi.org/ Treweek TM, Meehan S, Ecroyd H, Carver JA (2015) Small heat-shock 10.1093/dnares/2.4.191 proteins: important players in regulating cellular proteostasis. Cell Mol Kotani H, Tanaka A, Kaneko T, Sato S, Sugiura M, Tabata S (1995) Life Sci 72(3):429–451. https://doi.org/10.1007/s00018-014-1754-5 Assignment of 82 known genes and gene clusters on the genome of Van Durme J, Maurer-Stroh S, Gallardo R, Wilkinson H, Rousseau F, the unicellular cyanobacterium Synechocystis sp. strain PCC6803. Schymkowitz J (2009) Accurate prediction of DnaK-peptide bind- DNA Res 2(3):133–142. https://doi.org/10.1093/dnares/2.3.133 ing via homology modelling and experimental data. PLoS Comput Kriehuber T, Rattei T, Weinmaier T, Bepperling A, Haslbeck M, Buchner Biol 5(8):e1000475. https://doi.org/10.1371/journal.pcbi.1000475 J (2010) Independent evolution of the core domain and its flanking van Montfort RL, Basha E, Friedrich KL, Slingsby C, Vierling E (2001) sequences in small heat shock proteins. FASEB J 24(10):3633– Crystal structure and assembly of a eukaryotic small heat shock protein. 3642. https://doi.org/10.1096/fj.10-156992 Nat Struct Biol 8(12):1025–1030. https://doi.org/10.1038/nsb722 Lee S, Owen HA, Prochaska DJ, Barnum SR (2000) HSP16.6 is involved in the development of thermotolerance and thylakoid stability in the Wu YD, Zhao YL (2001) A theoretical study on the origin of unicellular cyanobacterium, Synechocystis sp. PCC 6803. Curr cooperativity in the formation of 3(10)- and alpha-helices. J Am Microbiol 40(4):283–287. https://doi.org/10.1007/s002849910056 Chem Soc 123(22):5313–5319. https://doi.org/10.1021/ja003482n
Cell Stress and Chaperones – Springer Journals
Published: Feb 23, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera