In silico identification of functional divergence between the multiple groEL gene paralogs in Chlamydiae
© McNally and Fares; licensee BioMed Central Ltd. 2007
Received: 25 January 2007
Accepted: 22 May 2007
Published: 22 May 2007
Heat-shock proteins are specialized molecules performing different and essential roles in the cell including protein degradation, folding and trafficking. GroEL is a 60 Kda heat-shock protein ubiquitous in bacteria and has been regarded as an important molecule implicated in chronic inflammatory processes caused by Chlamydiae infections. GroEL in Chlamydiae became duplicated at the origin of the Chlamydiae lineage presenting three distinct molecular chaperones, namely the original protein GroEL1 (Ct110), and its paralogous proteins GroEL2 (Ct604) and GroEL3 (Ct755). These chaperones present differential and independent expressions during the different stages of Chlamydiae infections and have been suggested to present differential physiological and regulatory roles.
In this comprehensive in silico study we show that GroEL protein paralogs have diverged functionally after the different gene duplication events and that this divergence has occurred mainly between GroEL3 and GroEL1. GroEL2 presents an intermediate functional divergence pattern from GroEL1. Our results point to the different protein-protein interaction patterns between GroEL paralogs and known GroEL protein clients supporting their functional divergence after groEL gene duplication. Analysis of selective constraints identifies periods of adaptive evolution after gene duplication that led to the fixation of amino acid replacements in GroEL protein domains involved in the interaction with GroEL protein clients.
We demonstrate that GroEL protein copies in Chlamydiae species have diverged functionally after the gene duplication events. We also show that functional divergence has occurred in important functional regions of these GroEL proteins and that very probably have affected the ancestral GroEL regulatory role and protein-protein interaction patterns with GroEL client proteins. Most of the amino acid replacements that have affected interaction with protein clients and that were responsible for the functional divergence between GroEL paralogs were fixed by adaptive evolution after the groEL gene duplication events.
Cells use several mechanisms to ameliorate the effects of transient changes in the environmental conditions such as heat stress, irradiation, viral infections, etc. For instance, cells have developed a complex family of genes coding for protein-folding machines sharing a wide range of vital functions to buffer the effects of stress on the proteome integrity. These proteins, also called heat-shock proteins or molecular chaperones, are classified in different protein families named on the basis of their members' approximate molecular weight and they assist in the folding, trafficking and degradation of proteins [1–3]. The heat-shock protein GroEL is among the best-studied molecular chaperones in bacteria and belongs to the group I chaperonins. Group I chaperonins are a group of ring-shaped ATPases that assist de novo protein folding in most cellular compartments [4–8]. GroEL is a homotetradecamer that interacts with a ring-shaped cofactor named GroES, which participates in folding proteins into the correct three-dimensional conformation [9, 10], and both proteins are essential for Escherichia coli growth at all range temperatures .
Due to the important functional role played by GroEL in maintaining the proteome integrity of cells, GroEL has become the target of many microbiological studies aimed at uncovering molecules involved in the epidemiology of pathogenic bacteria. GroEL from pathogenic bacteria is a highly immunoadjuvant protein and is recognised by the Toll-like receptors as part of the innate defence system [12, 13]. The fact that GroEL is among the most conserved protein families  and that GroEL isolated from pathogenic bacteria has been reported to have a strong immune eliciting function  has inspired projects aimed at developing vaccines targeting GroEL from pathogens. These studies yielded insightful results implicating GroEL in bacterial disease pathogenesis such as those caused by Chlamydiae infections . GroEL in Chlamydiae trachomatis (also called Ct110) has been implicated in chronic inflammatory processes caused by Chlamydiae infections leading to tissue damage and scarring [16–19]. Interestingly, GroEL in Chlamydiae became duplicated at the origin of the Chlamydiae lineage presenting three distinct molecular chaperones, namely the original protein GroEL1 (Ct110), and its paralogous proteins GroEL2 (Ct604) and GroEL3 (Ct755) . Even though the three Chlamydiae GroEL proteins present substantial amino acid sequence conservation in important regions involved in polypeptide binding when compared to GroEL from the bacterium Escherichia coli, significant differences have been spotted in GroES binding regions and at regions involved in ATP binding and hydrolysis. Among the three groEL genes, only the expression levels of groEL1 and its cochaperone groES increase under heat-stress conditions and only the protein GroEL1 complements the function of a GroEL thermo-sensitive mutation in HeLa cells under heat-stress conditions . Further, a previous report identified differences in the expression levels between the three groEL genes during the developmental stages of C. trachomatis . This study also showed through in vitro models of C. trachomatis infection that the three different groEL genes are differentially and independently expressed during the different infection cycles of this pathogen, with groEL2 being highly expressed during the infectious cycle of Chlamydiae and groEL3 showing the highest expression among the three groEL genes during the persistent infections .
Despite previous efforts invested in unravelling the main functional differences between the three different groEL genes in Chlamydiae, results have brought more questions than they have answered regarding the reasons for this functional divergence. To date, apart from one study in 2003 conducting some computational analyses for these genes , no detailed bioinformatics approach has been performed to aid in understanding the evolutionary dynamic differences between the three groEL genes and to link these differences with functional data.
In this study we conduct state-of-the-art bioinformatics analyses to unravel the main selective constraints leading to the functional differentiation between the Chlamydiae groEL genes. To identify functional divergence between the different GroEL protein copies we test the selective constraints after groEL gene duplication, analyze and phylogenetically map amino acid sites involved in this functional divergence and conduct molecular coevolution analyses within GroEL proteins and between these and proteins known to be obligate E. coli GroEL protein clients. The effects of amino acid sites involved in functional divergence in the stability of GroEL protein structures are also discussed.
GroEL proteins have diverged functionally in Chlamydiae after gene duplication
To test functional divergence between GroEL proteins after gene duplication we applied the program Diverge version 2.0 (See methods for details). Diverge tests for the presence of functional divergence of two types, functional divergence type I and type II. Functional divergence type I is detected when sites conserved (for example, showing no or low number of amino acid replacements when comparing sequences at that particular site) in one of the phylogenetic clusters (protein paralog) are significantly variable in the other related phylogenetic cluster. In other words, functional divergence type I indicates strong selective (and therefore functional) constraints at that site (for example, due to the acquisition or pre-existence of a functional role for that site) in one of the clusters and relaxed constraints (due to the loss or inexistent functional role at that site) in the paralogous cluster. Functional divergence type II is detected when, after gene duplication mutations leading to different amino acids become fixed in both resulting paralogous proteins and these mutations remain conserved after speciation in each cluster. This pattern indicates that amino acid sites diverged functionally between both paralogous clades (showing two distinct amino acids when comparing the two clades) but they were equally important for the protein's function (for example the amino acid remain conserved in each phylogenetic clade). We were interested in testing functional divergence type I to detect loss or acquisition of functional roles in particular amino acid sites in one of the GroEL group paralogs.
Functional divergence type I analysis between GroEL protein paralogs in Chlamydiae species.
θa ± SE(θ)
GroEL1 vs GroEL2
0.371 ± 0.096
GroEL1 vs GroEL3
0.943 ± 0.099
GroEL1 vs GroEL2-3
0.414 ± 0.117
GroEL2 vs GroEL3
0.441 ± 0.073
Functional divergence data is therefore in agreement with the expression divergence shown in previous functional/expression analysis demonstrating that in fact the different groEL gene copies are differently and independently expressed over time post-infection during Chlamydiae infection, and that GroEL3 is the most abundant protein at all time points assessed during the developmental cycle . In their study however, GroEL3 was virtually absent during persistent infections and GroEL2 showed the highest expression levels at that stage . Our results also support, in addition to the differential expression of the different groEL genes, the divergence in the protein function between the three Chlamydiae GroEL proteins.
Comparison of GroEL1 to GroEL3 and GroEL2 to GroEL3 showed a great percentage of sites under functional divergence type I with threshold posterior probability values of PP = 0.75 and PP = 0.95 (Figure 1B). The number of sites detected was greater in the comparison of GroEL1 to GroEL3 than in GroEL2 to GroEL3 comparison. We also studied the pattern of functional divergence and found three different profiles represented by the amino acid sites under functional divergence. The first pattern presented sites conserved for GroEL1 and GroEL2 but variable for GroEL3 (supporting loss of functional constraints at that site in GroEL3) and was represented by 42.18% of the functionally divergent sites. The second pattern was that represented by sites (23.44%) that were variable in GroEL1 and GroEL2 and became conserved in GroEL3 (indicating a gain of functional constraints in GroEL3 as the most parsimonious hypothesis). Finally we also found sites (33.6% of sites) variable in GroEL1 that became conserved in GroEL2 and GroEL3 (indicating the possible loss of functional constraints at that site in GroEL1). In most of the cases hence, GroEL3 presented loss of functional constraints in some sites and gain of constraints in others and these results were more obvious for GroEL3 than for GroEL2 compared to GroEL1. Examination of the sites under functional constraints in GroEL3 provided evidence supporting the involvement of these sites in ATP binding (G86, homologous to G86 in E. coli), substrate and GroES binding (P235, homologous to P236 in E. coli)  and interaction and folding of protein clients in the GroEL central cavity (K362 and D397 homologous to K363 and D397, respectively in E. coli) . These results suggest that functional divergence might have affected the interaction mainly between GroEL3 and its protein clients and to a lesser extent between GroEL2 and GroEL protein clients.
Here, we are comparing the mean distance between amino acids a and b belonging to proteins A and B respectively by comparing their coordinates in the three space axes. This comparison did not detect any significant structural differences among the three GroEL proteins or between them and E. coli GroEL protein (The distances were all below 3.5Å). Results then suggest that amino acid replacements did not involve structural changes but rather may have induced functional shifts between GroEL protein copies.
Although no major structural changes seem to be related to sites under functional constraints we examined whether sites with varying degrees of selective constraints in the different GroEL copies show differences in the folding energy of the local GroEL structures. The performance of different methods to analyze local folding energies has been recently elegantly examined . In their work, Rastogi et al., tested the accuracy of different models to predict the most stable structure or folding for four sets of proteins, Globin-like, SH3 domain, SH2 domain and Flavodoxin-like proteins. We used this methodology to look at folding-energy related differences at those sites under different functional constraints when comparing GroEL copies (for example, highly constrained amino acid sites in one GroEL copy but showing lack of constraints at another GroEL protein copy) and estimated the significance of these differences. We calculated this significance by comparing our folding-energy results with a distribution of folding energies for a 1000 randomly generated set of peptides sharing the same length and composition as the local fold of our proteins. We did the analyses using scripts and programs kindly provided by the group of Prof. Liberles. Our comparisons showed no significant differences in those sites under functional divergence when comparing the different mutant versions of the protein at those sites. In conclusion hence the mutations under varying functional constraints between GroEL copies lineages did not show significant variability in the local folding energies (Data not shown). Although apparently negative, the examination of the effects of mutations on protein structures is anything but straightforward. The reason is that two main factors have to be considered in such analyses. First, structures and folds are very flexible to mutations [26–28] and slight changes on function do not have to imply significant changes on protein-structure or folding. Second, the effect of several mutations on the protein structure may interact, with single mutations having little effect while combined mutations having large effects on the stability of local protein folds. More research is needed to identify the real effects of mutations on protein folds and structures.
Differential coevolution among Chlamydiae GroEL proteins
To further examine the selection shifts between the three GroEL protein copies we also investigated the intra-GroEL molecular coevolution and identified the differences in the coevolutionary relationships between amino acid sites among the three GroEL copies. Comparison of the coevolutionary relationships in GroEL1 to those in GroEL2 showed that while many coevolutionary relationships have been conserved in both copies (For example, amino acid pairs P217-R429, Q347-R429, Q347-P449, I348-R429, I348-P449, I348-A529, N432-P449, taking E. coli GroEL as reference sequence) other relationships have been lost in GroEL2 (D11-A404, L17-A340, L131-A340, A340-A404). Both groups of amino acid coevolving pairs include amino acid sites involved in the interaction with protein clients in the GroEL complex cavity. Interestingly, the level of coevolution between the set of pairs unique to GroEL1 (MIC = 0.225 ± 0.001) was lower than the level of coevolution for the set of pairs of sites conserved in both proteins GroEL1 and GroEL2 (MIC = 0.264 ± 0.034), indicating conservation of the main coevolutionary relationships, which are probably those highly involved in interaction with protein clients. Most interesting is the fact that GroEL3 showed no conservation of any of the intra-molecular amino acid site pairs coevolutionary relationships when compared to GroEL1 or GroEL2, thus pinpointing its unique evolutionary divergence and probable functional divergence from the other GroEL copies.
Analysis of selective constraints in GroEL from Chlamydiae. Mean replacements per non-synonymous (dN) sites and synonymous sites (dS) and the ratio between the two rates (ω) for the pairwise comparisons within GroEL1, GroEL2 and GroEL3 paralogs groups.
d S ± SE (d S )
d N ± SE (d N )
0.731 ± 0.032
0.077 ± 0.006
0.893 ± 0.045
0.548 ± 0.027
0.935 ± 0.039
0.587 ± 0.025
Recurrent adaptive evolution after groEL gene duplication
Examination of sites under adaptive evolution with significant posterior probabilities (PP > 0.95) identified sites involved in substrate binding and sites located in the central cavity of GroEL ring pointing toward the cavity and very probably involved in interaction with GroEL protein clients (Figure 4B). Taking all the results from the functional divergence analyses, protein-protein coevolution analyses and the adaptive evolution in each paralog group, we identified regions in the GroEL1 paralogs, GroEL2 and GroEL3, involved in interaction with proteins that have undergone changes in their selective constraints after gene duplication (Figure 4B). These results suggest that groEL gene duplication in Chlamydiae may have been followed by the GroEL paralogs' functional divergence toward acquiring different regulatory roles and establishing different protein-protein interaction network geometries.
Whether the functional divergence between the duplicated GroEL proteins meant the acquisition of completely novel functions or the subfunctionalization of the proteins copies is unclear. Placing our results into a model that supports subfunctionalization or into one that proposes neofunctionalization as the fate for gene copies after duplication requires taking into account population genetics parameters . In principle, duplicated genes are lost more slowly in organisms with small effective population sizes than in those with large population sizes. The reason is that selection against harmful mutations is weaker in population with small sizes and disadvantageous mutations can drift to fixation. Gene copies resulting from gene duplication have hence more evolutionary time (opportunities) to accumulate advantageous mutations and survive despite the build up of harmful mutations. Because degenerative mutations greatly outnumber beneficial mutations the probability of neofunctionalization in small populations is rare whereas subfunctionalization is more likely to occur in these populations [33, 34].
The effective population sizes of prokaryotes are considered large enough as to preclude any opportunity for subfunctionalization. However in unicellular pathogenic organisms, such as the Chlamydiae species analyzed in this work, their genetic effective population sizes may be greatly dependent on their multicellular hosts, which present significantly lower population sizes. In such a scenario, the genetic drift effect increases and selective constraints strength decreases, incrementing thus the probability of gene copy preservation and subfunctionalization after gene duplication. GroEL protein in Chlamydiae may be a striking example of such process taken to completion at the interactome level.
We have demonstrated that GroEL protein copies in Chlamydiae species have diverged functionally after the gene duplication events. Our comprehensive bioinformatics analysis yields results that are in accordance with previously published experimental and functional data and provides further support to the divergence in the physiological and regulatory roles of the different GroEL protein copies. We also provide evidence that GroEL3 (Ct755) is more divergent from GroEL1 (Ct110) than GroEL2 (Ct604) and that this divergence was due to the fixation of amino acid replacements that modified the functional constraints in specific amino acid sites in GroEL3. Coevolution analyses performed here also support the high divergence of GroEL3 and provide further evidence that the three different GroEL copies have different interaction patterns with previously identified GroEL1 protein clients, further supporting their different regulatory roles. Finally, analysis of selective constraints supports the adaptive fixation of amino acid replacements after gene duplication mainly leading to GroEL3 and that this fixation affected functional sites involved in interaction with protein clients. Based on these analyses and conclusions we propose conducting comprehensive protein-protein interaction analyses between the different GroEL protein copies in Chlamydiae and the known GroEL protein clients to fully understand their functional and regulatory divergence and their role in the epidemiology, developmental and persistent stages of Chlamydiae infections.
The aim of this study is to test the functional divergence between the different GroEL copies in Chlamydiae and provide a list of amino acid sites that may be responsible for such functional divergence, thereby detailing the functional differences among the copies. Aside from in silico testing of the functional divergence between the GroEL protein copies, we are interested in the quantification of such divergence and the identification of the effects of such divergence in the function of each copy. Finally, we test the effect such divergence has on the interaction of GroEL copies with previously identified GroEL-dependent protein clients  and we highlight the selective constraints operating in each GroEL paralog.
Sequence alignments and phylogenetic analysis
Protein sequences coding for GroEL1 (Ct110), GroEL2 (Ct604) and GroEL3 (Ct755) were retrieved from the GeneBank database for the different species of Chlamydiae. The sequences, species names and the protein-coding sequence accession numbers are provided in table 1 of additional file 1. We aligned protein sequences using the program ClustalX  with the default settings. We then aligned nucleotide sequences concatenating triplets of nucleotides according to the multiple protein sequence alignment (alignments are available from the authors on request). Together with the groEL gene sequence we also obtained alignments for client proteins shown to depend on E. coli GroEL to acquire a productive (functional) protein conformation . We obtained the sequences for each one of the Chlamydiae species or strains from GenBank and the accession numbers are provided in Table 2 of additional file 2. We then aligned the sequences for each one of the protein-coding genes following the same procedure detailed above.
Regarding phylogenetic analyses, for each one of the multiple sequence alignments we first used ModelTest 1.3  to determine the best candidate substitution rate matrix for maximum likelihood inference. The program pinpointed TrN + I + G as first option. We used then the output generated by ModelTest as input for the program PAUP  and inferred a maximum-likelihood phylogenetic tree for the alignment containing the three different GroEL protein-coding sequences using the heuristic approach.
Analysis of functional divergence
To identify amino acid replacements responsible for functional divergence between the GroEL proteins, we tested functional divergence Type I [37, 38] in the multiple protein sequence alignment containing the three different GroEL copies of Chlamydiae after each gene duplication event. The Gu method uses a maximum-likelihood procedure to test whether there has been a significant change in the rate of evolution after gene duplication leading to the two paralogs. This method tests for functional divergence by estimating the log-likelihood value of the hypothesis assuming a value for the coefficient of functional divergence (θ > 0) and comparing this likelihood with that under the hypothesis of no functional divergence (θ = 0). Because both models are nested, they can be compared by the Likelihood-ratio test (LRT), which can be approximated to a χ2 distribution with 1 degree of freedom. If the null hypothesis of no-functional divergence is rejected, the program calculates a posterior probability (PP) for a position being classified within the category of functional divergence. We established a cutoff value for the PP according to the effect that the elimination of the sets of amino acid sites having a PP value equal or higher than that cutoff value have on the θ-value test .
We tested functional divergence between GroEL1 and the cluster containing GroEL2 and 3, and between GroEL2 and GroEL3 using the program Diverge version 2.0 . We then mapped the events of functional divergence in the phylogenetic tree including the two duplication events that gave rise to the three GroEL protein copies.
Testing coevolution between GroEL copies
One of the questions we aimed answering was whether GroEL2 and GroEL3 diverged equally from GroEL1 or whether one of them presented less evidence for shared functions with GroEL1. A good way to test this hypothesis is by examining the coevolutionary patterns between the different GroEL copies. The stronger the coevolution between the proteins the greater would be the amount of shared evolutionary pattern and thus the greater the likelihood of sharing more functions. To test the hypothesis of coevolution between proteins we used the non-parametric method based on the mutual information criterion (MIC) developed by Korber and colleagues . The mutual information is represented by the entropies that involve the joint probability distribution, P(si, s'j), of occurrence of symbol i at position s and j at position s' of the multiple sequence alignment. The MIC values generated range between 0, indicating independent evolution, and a positive value whose magnitude depends on the amount of covariation. Variable positions included in the alignment and considered in the coevolutionary analyses were those parsimony-informative (i.e. they contain at least two types of amino acids and at least two of them occur with a minimum frequency of two). The significance of the MIC values was assessed by randomization of pairs of sites in the alignment, calculation of their MIC values and comparison of the real values with the distribution of one million randomly sampled values. To correct for multiple non-independent tests we implemented the step-down permutation procedure and corrected the probabilities accordingly . MICK is implemented in the program PECA (Available from the corresponding author on request).
Testing for protein-protein interaction divergence between GroEL copies and protein clients
One of the hypotheses we wanted to test was whether functional divergence between the different GroEL copies also involved a divergence in their coevolutionary patterns with known GroEL protein clients. To test this hypothesis we analysed the coevolution of each GroEL copy with each one of the known GroEL protein clients using the methodology described in the previous section. The strength of the coevolutionary pattern was calculated by classifying significant MIC values into the categories (0.1, 0.15, 0.20, 0.25, 0.30, 0.35, 0.40, 0.45, 0.50, MIC > 0.50). Here 0.1 included all those pairs of amino acid sites with MIC values 0 < MIC ≤ 0.1; 0.15 would include 0.1 < MIC ≤ 0.15, and so on and so forth. This categorization of MIC values allows the direct comparison of the coevolutionary results between different pairs of proteins regardless the set of MIC values obtained in each analysis. To quantify the contribution of each category to the overall MIC value, we first counted the number of pairs of sites showing MIC values within that category. We then calculated the percentage of pairs of sites included in that category by dividing the number of sites in the category by the total number of pairs of sites detected as coevolving significantly. This way, the contribution of each MIC category between pairs of proteins is comparable.
Analysis of selective constraints
The final step in the analysis of functional divergence is the mapping of selective constraints in the protein structure after each duplication event. Here we tested whether functional divergence was the result of the adaptive fixation of amino acid replacements at functional protein regions in GroEL copies. To test this hypothesis we applied two methodologies. First, we applied a sliding-window parsimony-based approach to detect selective constraints in protein-coding genes , implemented in the program SWAPSC version 1.0 . Briefly, the program slides a statistically optimum window size along the sequence alignment to detect selective constraints and estimates the probability of replacements per non-synonymous sites (d N ) and substitutions per synonymous sites (d S ). The window size is optimized by means of using a number of simulated data sets. The standard way to measure the intensity of selection when analysing DNA variability is by comparing d S to d N [43, 44]. The ratio between the two rates () helps to elucidate if the gene has been fixing amino acid replacements neutrally (ω = 1), replacements have been removed by purifying selection (ω < 1), or mutations have been fixed by adaptive evolution (ω > 1). It has been shown, however, that ω is a poor indicator of the action of adaptive evolution due to the fact that signals of adaptive evolution may be swamped in the background of purifying selection under which the protein has evolved most over its evolutionary time .
SWAPSC uses ω to estimate the intensity of selection acting on a protein-coding region at particular branches of the tree. We used 1000 simulated data sets in our analysis obtained using the program Evolver from the PAML package . To perform the simulations we took as initial parameters the average ω value, transition-to-transversion rates and codon table generated under the Goldman and Yang model, using the real sequence alignment as input. The program then slides the window along the real sequence alignment and estimates d N and d S by the Li's method. The program determines significance of these estimates under a Poisson distribution of nucleotide substitutions along the alignment.
In addition we tested adaptive evolution using the maximum-likelihood based approach implemented in the program PAML v3.15 (Yang 1997). We then compared the log-likelihood value of a model (Goldman and Yang model, hereon called G&Y)  that assumes one ω for the whole alignment and phylogenetic tree to a model that estimates an ω value for each branch of the phylogenetic tree (hereon called the free-ratio model FRM). We compared both likelihood values using the Likelihood ratio test (LRT) with the degrees of freedom being the number of branches in the tree minus 1.
This work was supported by Science foundation Ireland, under the program the President of Ireland Young Researcher Award, grant number (04/YI1/M518), to M.A.F. We are most grateful to reviewer 1 for his valuable suggestions to improve the quality of the manuscript. We are also extremely grateful to Prof. David Liberles and to Shruti Rastogi for providing us with their scripts and programs to analyze stability of protein folds.
- Hightower LE: Heat shock, stress proteins, chaperones, and proteotoxicity. Cell. 1991, 66 (2): 191-197. 10.1016/0092-8674(91)90611-2.View ArticlePubMedGoogle Scholar
- Nover L, Hightower L: Heat shock and development. Introduction. Results Probl Cell Differ. 1991, 17: 1-4.View ArticlePubMedGoogle Scholar
- Takenaka IM, Sadis S, Hightower LE: Transforming growth factor-beta regulates basal expression of the hsp70 gene family in cultured chicken embryo cells. Results Probl Cell Differ. 1991, 17: 188-209.View ArticlePubMedGoogle Scholar
- Hartl F, Vlcek A: Bonding Properties of the 1,2-Semiquinone Radical-Anionic Ligand in the [M(CO)(4-n)(L)(n)(DBSQ)] Complexes (M = Re, Mn; DBSQ = 3,5-di-tert-butyl-1,2-benzosemiquinone; n = 0, 1, 2). A Comprehensive Spectroscopic (UV-Vis and IR Absorption, Resonance Raman, EPR) and Electrochemical Study. Inorg Chem. 1996, 35 (5): 1257-1265. 10.1021/ic950018o.View ArticlePubMedGoogle Scholar
- Ellis RJ: Molecular chaperones: avoiding the crowd. Curr Biol. 1997, 7 (9): R531-3. 10.1016/S0960-9822(06)00273-9.View ArticlePubMedGoogle Scholar
- Bukau B, Horwich AL: The Hsp70 and Hsp60 chaperone machines. Cell. 1998, 92 (3): 351-366. 10.1016/S0092-8674(00)80928-9.View ArticlePubMedGoogle Scholar
- Frydman J: Folding of newly translated proteins in vivo: the role of molecular chaperones. Annu Rev Biochem. 2001, 70: 603-647. 10.1146/annurev.biochem.70.1.603.View ArticlePubMedGoogle Scholar
- Hartl FU, Hayer-Hartl M: Molecular chaperones in the cytosol: from nascent chain to folded protein. Science. 2002, 295 (5561): 1852-1858. 10.1126/science.1068408.View ArticlePubMedGoogle Scholar
- Mayhew M, da Silva AC, Martin J, Erdjument-Bromage H, Tempst P, Hartl FU: Protein folding in the central cavity of the GroEL-GroES chaperonin complex. Nature. 1996, 379 (6564): 420-426. 10.1038/379420a0.View ArticlePubMedGoogle Scholar
- Weissman JS, Rye HS, Fenton WA, Beechem JM, Horwich AL: Characterization of the active intermediate of a GroEL-GroES-mediated protein folding reaction. Cell. 1996, 84 (3): 481-490. 10.1016/S0092-8674(00)81293-3.View ArticlePubMedGoogle Scholar
- Fayet O, Ziegelhoffer T, Georgopoulos C: The groES and groEL heat shock gene products of Escherichia coli are essential for bacterial growth at all temperatures. J Bacteriol. 1989, 171 (3): 1379-1385.PubMed CentralPubMedGoogle Scholar
- Vabulas RM, Ahmad-Nejad P, da Costa C, Miethke T, Kirschning CJ, Hacker H, Wagner H: Endocytosed HSP60s use toll-like receptor 2 (TLR2) and TLR4 to activate the toll/interleukin-1 receptor signaling pathway in innate immune cells. J Biol Chem. 2001, 276 (33): 31332-31339. 10.1074/jbc.M103217200.View ArticlePubMedGoogle Scholar
- Brocchieri L, Karlin S: Conservation among HSP60 sequences in relation to structure, function, and evolution. Protein Sci. 2000, 9 (3): 476-486.PubMed CentralView ArticlePubMedGoogle Scholar
- Perschinka H, Mayr M, Millonig G, Mayerl C, van der Zee R, Morrison SG, Morrison RP, Xu Q, Wick G: Cross-reactive B-cell epitopes of microbial and human heat shock protein 60/65 in atherosclerosis. Arterioscler Thromb Vasc Biol. 2003, 23 (6): 1060-1065. 10.1161/01.ATV.0000071701.62486.49.View ArticlePubMedGoogle Scholar
- Karunakaran KP, Noguchi Y, Read TD, Cherkasov A, Kwee J, Shen C, Nelson CC, Brunham RC: Molecular analysis of the multiple GroEL proteins of Chlamydiae. J Bacteriol. 2003, 185 (6): 1958-1966. 10.1128/JB.185.6.1958-1966.2003.PubMed CentralView ArticlePubMedGoogle Scholar
- Lichtenwalner AB, Patton DL, Van Voorhis WC, Sweeney YT, Kuo CC: Heat shock protein 60 is the major antigen which stimulates delayed-type hypersensitivity reaction in the macaque model of Chlamydia trachomatis salpingitis. Infect Immun. 2004, 72 (2): 1159-1161. 10.1128/IAI.72.2.1159-1161.2004.PubMed CentralView ArticlePubMedGoogle Scholar
- Peeling RW, Bailey RL, Conway DJ, Holland MJ, Campbell AE, Jallow O, Whittle HC, Mabey DC: Antibody response to the 60-kDa chlamydial heat-shock protein is associated with scarring trachoma. J Infect Dis. 1998, 177 (1): 256-259.View ArticlePubMedGoogle Scholar
- Sanchez-Campillo M, Bini L, Comanducci M, Raggiaschi R, Marzocchi B, Pallini V, Ratti G: Identification of immunoreactive proteins of Chlamydia trachomatis by Western blot analysis of a two-dimensional electrophoresis map with patient sera. Electrophoresis. 1999, 20 (11): 2269-2279. 10.1002/(SICI)1522-2683(19990801)20:11<2269::AID-ELPS2269>3.0.CO;2-D.View ArticlePubMedGoogle Scholar
- Sasu S, LaVerda D, Qureshi N, Golenbock DT, Beasley D: Chlamydia pneumoniae and chlamydial heat shock protein 60 stimulate proliferation of human vascular smooth muscle cells via toll-like receptor 4 and p44/p42 mitogen-activated protein kinase activation. Circ Res. 2001, 89 (3): 244-250.View ArticlePubMedGoogle Scholar
- Herve C. Gerard JAWH: Differential expression of three Chlamydia trachomatis hsp60-encoding genes in active vs. persistent infections. Microbial Pathogenesis. 2004, 36: 35-39. 10.1016/j.micpath.2003.08.005.View ArticlePubMedGoogle Scholar
- Fenton WA, Kashi Y, Furtak K, Horwich AL: Residues in chaperonin GroEL required for polypeptide binding and release. Nature. 1994, 371 (6498): 614-619. 10.1038/371614a0.View ArticlePubMedGoogle Scholar
- Bates PA, Sternberg MJ: Model building by comparison at CASP3: using expert knowledge and computer automation. Proteins. 1999, Suppl 3: 47-54. 10.1002/(SICI)1097-0134(1999)37:3+<47::AID-PROT7>3.0.CO;2-F.View ArticlePubMedGoogle Scholar
- Bates PA, Kelley LA, MacCallum RM, Sternberg MJ: Enhancement of protein modeling by human intervention in applying the automatic programs 3D-JIGSAW and 3D-PSSM. Proteins. 2001, Suppl 5: 39-46. 10.1002/prot.1168.View ArticlePubMedGoogle Scholar
- Contreras-Moreira B, Bates PA: Domain fishing: a first step in protein comparative modelling. Bioinformatics. 2002, 18 (8): 1141-1142. 10.1093/bioinformatics/18.8.1141.View ArticlePubMedGoogle Scholar
- Rastogi S, Reuter N, Liberles DA: Evaluation of models for the evolution of protein sequences and functions under structural constraint. Biophys Chem. 2006, 124 (2): 134-144. 10.1016/j.bpc.2006.06.008.View ArticlePubMedGoogle Scholar
- Taverna DM, Goldstein RA: Why are proteins so robust to site mutations?. J Mol Biol. 2002, 315 (3): 479-484. 10.1006/jmbi.2001.5226.View ArticlePubMedGoogle Scholar
- Taverna DM, Goldstein RA: Why are proteins marginally stable?. Proteins. 2002, 46 (1): 105-109. 10.1002/prot.10016.View ArticlePubMedGoogle Scholar
- Shakhnovich BE, Deeds E, Delisi C, Shakhnovich E: Protein structure and evolutionary history determine sequence space topology. Genome Res. 2005, 15 (3): 385-392. 10.1101/gr.3133605.PubMed CentralView ArticlePubMedGoogle Scholar
- Codoner FM, Fares MA, Elena SF: Adaptive covariation between the coat and movement proteins of prunus necrotic ringspot virus. J Virol. 2006, 80 (12): 5833-5840. 10.1128/JVI.00122-06.PubMed CentralView ArticlePubMedGoogle Scholar
- Gregory B. Gloor LCM: Mutual information in protein multiple sequence alignments reveals two classes of cevolving positions. Biochemistry. 2005, 44: 4156-7165. 10.1021/bi050293e.Google Scholar
- Kerner MJ, Naylor DJ, Ishihama Y, Maier T, Chang HC, Stines AP, Georgopoulos C, Frishman D, Hayer-Hartl M, Mann M, Hartl FU: Proteome-wide analysis of chaperonin-dependent protein folding in Escherichia coli. Cell. 2005, 122 (2): 209-220. 10.1016/j.cell.2005.05.028.View ArticlePubMedGoogle Scholar
- Lynch M, Conery JS: The origins of genome complexity. Science. 2003, 302 (5649): 1401-1404. 10.1126/science.1089370.View ArticlePubMedGoogle Scholar
- Force A, Lynch M, Pickett FB, Amores A, Yan YL, Postlethwait J: Preservation of duplicate genes by complementary, degenerative mutations. Genetics. 1999, 151 (4): 1531-1545.PubMed CentralPubMedGoogle Scholar
- Stoltzfus A: On the possibility of constructive neutral evolution. J Mol Evol. 1999, 49 (2): 169-181. 10.1007/PL00006540.View ArticlePubMedGoogle Scholar
- Thompson JD, Gibson TJ, Plewniak F, Jeanmougin F, Higgins DG: The CLUSTAL_X windows interface: flexible strategies for multiple sequence alignment aided by quality analysis tools. Nucleic Acids Res. 1997, 25 (24): 4876-4882. 10.1093/nar/25.24.4876.PubMed CentralView ArticlePubMedGoogle Scholar
- Posada D, Crandall KA: MODELTEST: testing the model of DNA substitution. Bioinformatics. 1998, 14 (9): 817-818. 10.1093/bioinformatics/14.9.817.View ArticlePubMedGoogle Scholar
- Gu X: Statistical methods for testing functional divergence after gene duplication. Mol Biol Evol. 1999, 16 (12): 1664-1674.View ArticlePubMedGoogle Scholar
- Wang Y, Gu X: Functional divergence in the caspase gene family and altered functional constraints: statistical analysis and prediction. Genetics. 2001, 158 (3): 1311-1320.PubMed CentralPubMedGoogle Scholar
- Ho Y, Gruhler A, Heilbut A, Bader GD, Moore L, Adams SL, Millar A, Taylor P, Bennett K, Boutilier K, Yang L, Wolting C, Donaldson I, Schandorff S, Shewnarane J, Vo M, Taggart J, Goudreault M, Muskat B, Alfarano C, Dewar D, Lin Z, Michalickova K, Willems AR, Sassi H, Nielsen PA, Rasmussen KJ, Andersen JR, Johansen LE, Hansen LH, Jespersen H, Podtelejnikov A, Nielsen E, Crawford J, Poulsen V, Sorensen BD, Matthiesen J, Hendrickson RC, Gleeson F, Pawson T, Moran MF, Durocher D, Mann M, Hogue CW, Figeys D, Tyers M: Systematic identification of protein complexes in Saccharomyces cerevisiae by mass spectrometry. Nature. 2002, 415 (6868): 180-183. 10.1038/415180a.View ArticlePubMedGoogle Scholar
- Korber BT, Farber RM, Wolpert DH, Lapedes AS: Covariation of mutations in the V3 loop of human immunodeficiency virus type 1 envelope protein: an information theoretic analysis. Proc Natl Acad Sci U S A. 1993, 90 (15): 7176-7180. 10.1073/pnas.90.15.7176.PubMed CentralView ArticlePubMedGoogle Scholar
- Fares MA, Elena SF, Ortiz J, Moya A, Barrio E: A sliding window-based method to detect selective constraints in protein-coding genes and its application to RNA viruses. J Mol Evol. 2002, 55 (5): 509-521. 10.1007/s00239-002-2346-9.View ArticlePubMedGoogle Scholar
- Fares MA: SWAPSC: sliding window analysis procedure to detect selective constraints. Bioinformatics. 2004, 20 (16): 2867-2868. 10.1093/bioinformatics/bth303.View ArticlePubMedGoogle Scholar
- Kimura M: Preponderance of synonymous changes as evidence for the neutral theory of molecular evolution. Nature. 1977, 267 (5608): 275-276. 10.1038/267275a0.View ArticlePubMedGoogle Scholar
- Sharp PM: In search of molecular darwinism. Nature. 1997, 385 (6612): 111-112. 10.1038/385111a0.View ArticlePubMedGoogle Scholar
- Goldman N, Yang Z: A codon-based model of nucleotide substitution for protein-coding DNA sequences. Mol Biol Evol. 1994, 11 (5): 725-736.PubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.