Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Dec 9.
Published in final edited form as: Mol Cell. 2011 Dec 9;44(5):734–744. doi: 10.1016/j.molcel.2011.09.022

Structural Instability Tuning as a Regulatory Mechanism in Protein-Protein Interactions

Li Chen 1, Vassilia Balabanidou 2,3, David P Remeta 1, Conceição ASA Minetti 1, Athina G Portaliou 2, Anastassios Economou 2,3, Charalampos G Kalodimos 1,4,5,*
PMCID: PMC3240846  NIHMSID: NIHMS337172  PMID: 22152477

SUMMARY

Protein-protein interactions mediate a vast number of cellular processes. Here we present a regulatory mechanism in protein-protein interactions mediated by finely-tuned structural instability coupled with molecular mimicry. We show that a set of type III secretion (TTS) autoinhibited homodimeric chaperones adopt a molten-globule-like state that transiently exposes the substrate binding site as a means to become rapidly poised for binding to their cognate protein substrates. Packing defects at the homodimeric interface stimulate binding whereas correction of these defects results in less labile chaperones that give rise to non-functional biological systems. The protein substrates use structural mimicry to offset the “weak spots” in the chaperones and to counteract their autoinhibitory conformation. This regulatory mechanism of protein activity is evolutionary conserved among several TSS systems and presents a lucid example of functional advantage conferred upon a biological system by finely-tuned structural instability.

INTRODUCTION

Protein-protein interactions mediate a vast number of regulatory pathways and are thus central to cell physiology (Kuriyan and Eisenberg, 2007; Yu et al., 2008). Formation of protein complexes is often under precise regulation as a means to control protein activity and to prevent premature and undesirable interactions among cellular components (Kobe and Kemp, 1999; Schlessinger, 2003). This role is frequently served by molecular chaperones whose cellular functions include assisting with folding and unfolding, biogenesis, regulation of protein conformation and activity, targeting, and assembly and disassembly of large protein complexes (Hartl and Hayer-Hartl, 2009; Haslbeck et al., 2005; Stirling et al., 2006).

Chaperones have particularly prominent and multiple roles in various protein transport and secretion pathways (Cross et al., 2009; Waksman and Hultgren, 2009). Specialized chaperones are important components of type III secretion (TTS) systems wherein they assist with the assembly and operation of the entire machinery (Birtalan et al., 2002; Cornelis, 2006; Feldman and Cornelis, 2003; Galan and Wolf-Watz, 2006; Parsot et al., 2003). The TTS apparatus is an exquisitely engineered molecular machinery that has specifically evolved to deliver bacterial virulence proteins directly into eukaryotic cells (Cornelis, 2006; Galan and Wolf-Watz, 2006). Loss of a TTS chaperone generally results in rapid degradation, aggregation or reduced secretion of its cognate secretion substrate(s) (Feldman and Cornelis, 2003; Parsot et al., 2003).

CesAB is a chaperone for EspA in the enteropathogenic Escherichia coli (EPEC) (Creasey et al., 2003). EPEC is the archetype of a group of pathogens that adhere to host enterocytes via formation of attaching and effacing (A/E) lesions and cause extensive host cell cytoskeletal rearrangements (Dean and Kenny, 2009). Once secreted, EspA undergoes self-polymerization thereby forming a long extracellular filamentous extension that connects the needle to the translocation pore in the eukaryotic plasma membrane and likely acts as a molecular conduit for TTS protein translocation (Knutton et al., 1998). Because of its high tendency to self-oligomerize it is necessary that EspA be captured in its monomeric state in the bacterial cytosol, a role served by the CesAB chaperone (Creasey et al., 2003; Yip et al., 2005).

Here we show that CesAB, in contrast to typical chaperones, exists as a loosely packed, conformationally dynamic homodimer in solution. CesAB adopts an autoinhibited conformation to prevent self-aggregation but undergoes a subunit exchange mechanism to form a stoichiometric complex with EspA. CesAB becomes rapidly poised for EspA binding by transiently exposing part of the binding site in a mechanism facilitated by packing defects at its homodimeric coiled-coil subunit interface. Correction of the naturally-occurring packing defects results in a less labile CesAB that fails to bind to EspA thereby giving rise to a non-functional TTS system in vitro and in vivo. EspA uses structural mimicry to offset the “weak spots” in CesAB thereby inducing folding of both partners and selectively stabilizing the heterodimer. We show that this mechanism is evolutionary conserved among several TSS systems. This regulatory mechanism of protein activity presents a lucid example of functional advantage conferred upon a biological system by finely-tuned structural instability.

RESULTS

CesAB Adopts a Molten-Globule-Like Conformation

Biophysical characterization of CesAB shows that the protein is all α-helical (Figure 1A) and exists in solution as a homodimer (~27 kDa) with a dimer dissociation constant (Kd) of ~0.5 µM (Figures S1A and S1B available online). The backbone NMR spectra of CesAB show, surprisingly, far fewer signals than expected for a natively folded protein (Figure 1B, blue). Moreover, the observed peaks are poorly dispersed and show severe line broadening. Similarly, poor dispersion is also observed for the methyl groups of hydrophobic residues suggesting that CesAB is relatively loosely packed (Figure 1C). This observation is further corroborated by the circular dichroism (CD) 222:208 nm ratio (~0.91; Figure 1A), a value indicative of loose interstrand association in coiled-coil proteins (McNamara et al., 2008) and near-UV data (Figure S1C). In addition, the CD thermal-denaturation profile of CesAB (Figure 1D) features a long transition that is suggestive of non-cooperative unfolding of the protein. Collectively, these data provide strong evidence that CesAB is a loosely packed, conformationally heterogeneous dimeric chaperone with molten-globule-like conformational properties.

Figure 1. The CesAB Chaperone has Molten-Globule-Like Properties.

Figure 1

Open in a new tab

(A) Far-UV CD data of native CesAB (blue) and in the presence of 10% TFE (magenta). The cross-over point at 201 nm is characteristic of coiled coils whereas the 222:208 nm ratio (~0.9) of native CesAB is suggestive of a poorly packed coiled coil (see Experimental Procedures). Addition of 10% TFE increases significantly the CesAB helicity.

(B,C) Overlaid 1H-15N HSQC (B) and 1H-13C HMQC (C) spectra of U-[2H,12C], Ala-, Leu-, Met-, Val-, Ile- δ1-[13CH3] CesAB under native conditions (blue) and in 10% TFE (magenta). Only a fraction of the expected amide signals in native CesAB are present because of severe line broadening indicating the presence of structural fluctuations on the milli-to-microsecond (ms-µs) time scale between conformations with different chemical shifts. The methyl resonances show poor dispersion indicating a relatively loose packing of the hydrophobic regions in the CesAB homodimer. All these features are the hallmark of a poorly packed protein with conformational heterogeneity and dynamic fluctuations among multiple conformational states. Various conditions (pH, temperature, salt) had very little effect on the spectra of CesAB consistent with the protein being in a molten-globule-like conformation (Receveur-Brechot et al., 2006). Addition of TFE (10% v/v) shifts the equilibrium towards the folded conformation of CesAB (magenta). Many of the well-dispersed peaks present in 10% TFE are already present in the native spectrum (representative resonances are plotted at lower contour and shown in the rectangular boxes), although broad, and addition of TFE decreases their line width.

(D) Far-UV CD thermal denaturation of CesAB features a long transition suggesting that the protein unfolds in a non-cooperative manner.

See also Figure S1.

Structure Determination of CesAB

The conformational heterogeneity and dynamic fluctuations of conformations in CesAB give rise to poorly dispersed, severely broadened NMR spectra (Figures 1B and 1C, blue) that precludes direct determination of its structure. To overcome the technical challenges presented by the low quality spectra and the conformational heterogeneity of the protein, we exploited the fact that addition of small amounts of trifluoroethanol (TFE) as cosolvent (up to 10% v/v) shifts the conformational equilibrium towards the folded conformation. This results in marked increase in chemical shift dispersion, narrowing of most of line widths and the appearance of the majority of the expected CesAB resonances (Figures 1B and 1C, magenta). Close comparison of the 1H-15N heteronuclear single quantum coherence (HSQC) NMR spectra of CesAB in the native state (0% TFE) and in the presence of 10% TFE (Figure 1B) reveals that several of the well-dispersed peaks are present in the native spectrum, although broad, and addition of TFE decreases their line width, without significantly affecting their chemical shifts. This observation indicates that TFE (up to 10% TFE) simply stabilizes the regions that are mostly folded already in the native state. This is especially true in the case of the methyl groups that are located at the dimerization interface (e.g. see Ala region in the 1H-13C heteronuclear multiple quantum coherence (HMQC) spectrum in Figure 1C) whose chemical shifts are not affected significantly by the addition of TFE but their linewidths become significantly narrower. The combined analysis of 1H-15N HSQC and 1H-13C HMQC spectra shows that TFE increases the helicity of many regions but is not affecting appreciably the packing. In agreement with the NMR results, far-UV CD data show that increasing TFE concentration stabilizes the α-helical structure of CesAB (Figure 1A). Multi-angle laser light scattering (MALLS) data shows that CesAB remains dimeric in 10% TFE. Addition of such small amounts of TFE has been often used to suppress unfavorable conformational exchange due to unfolding and to determine high-resolution structures by NMR (Buck, 1998; Kim et al., 2000).

The structure of CesAB under native conditions (Figure 2 and Figure S2A) was determined by refining the initial CesAB structure obtained in 10% TFE by including distance and dihedral angle restraints obtained under native conditions (0% TFE) to more reliably represent the secondary structure and packing of the native CesAB (a detailed NMR assignment and structure calculation protocol is provided in the Experimental Procedures). Several mutants were designed and tested on the basis of the determined CesAB structure and the phenotype of every single one of them could be accounted for by the structural data providing strong support of the validity of the structural approach (see below). As an ultimate proof of the robustness of the CesAB structure, a triple CesAB mutant was successfully designed on the basis of the structural data to stabilize the dimeric interface of CesAB and suppress the unfavorable conformational exchange, thereby giving rise to improved spectral properties under native conditions (see below).

Figure 2. CesAB Forms a Partially Folded, Loosely Packed Four-Helical Bundle.

Figure 2

Open in a new tab

(A) Solution structure of the CesAB homodimer colored using a gradient scheme (blue to red, from the N to the C terminus) for each subunit. The conformational ensemble is shown in Figure S2A.

(B) The interface of the helical bundle (delineated region in dashed box in (A)) is dominated by hydrophobic residues and thus CesAB dimerization is primarily mediated by non-polar interactions. Hydrogen bonds are represented by black dashed lines. The two subunits are colored blue and green.

(C) CesAB dimerization buries significant amounts of hydrophobic surface. One subunit is displayed as a solvent-accessible surface with hydrophobic residues colored green, whereas the other subunit is displayed as a blue ribbon.

See also Figure S2.

Structure of CesAB Reveals a Loosely Packed Helical Bundle

The structural data, which it should be noted that are biased towards the folded population of the CesAB native ensemble, show that CesAB adopts a four-helix bundle conformation (Figure 2A). Each of the CesAB subunits adopts a “U” shape with an all-helical conformation, consisting of helices α1 (Arg8-Glu36), α2 (Gln50-Lys54) and α3 (Asp72-Thr84), with helices α1 and α2 connected by a hairpin (Figure S2B). The C-terminal region (Ser86-Val107) is unstructured. Dimer formation is mediated by helical segments in both the N- and C-terminal regions, with helices α1 and α3 from each polypeptide chain packing together via extensive hydrophobic interactions to form a coiled-coil four-helix bundle (Figures 2A and 2B). The hairpins from each subunit pack in an anti-parallel manner with 2-fold symmetry providing an overall topology that resembles a bisecting-U motif (Hill et al., 2000). The radius of gyration calculated based on the structure of CesAB (Rg ~26.5 Å) is in agreement with the value measured experimentally by MALLS (Rg ~26.9 Å).

Most of the hydrophobic residues project between the helices or into the center of the bundle to form a single hydrophobic core (Figures 2B and 2C). A relatively small area (~1,000–1,200 Å2) appears to consistently mediate dimerization in all of the conformers in the structural ensemble (Figure S2A). This area, consisting exclusively of hydrophobic residues (Leu13, Ile17, Ile24, Ile27, Ile28, and Phe31 of helix α1 and Ile78, Leu81, and Leu85 of helix α3), is formed primarily by the interaction of the α1 helices, α1-α1’, and is supported by α1-α3’ and α3-α3’ interactions (Figure 2B). In line with the structural data, substitution of Ile17, Ile28 and Leu81 by Ala weakens the inter-subunit hydrophobic interactions and monomerize CesAB (Figure S2C).

Folding and Dynamic Properties of the CesAB Homodimer

As noted above, the determined structure of CesAB (Figure 2A and Figure S2A) is heavily biased toward the folded conformation and thus provides a limited view of the actual native conformational ensemble of the protein. To gain insight into the dynamic and folding properties of CesAB we determined secondary structure propensity (SPP) (Marsh et al., 2006) values, random coil chemical shift index (RCI) order parameters (S2) (Berjanskii and Wishart, 2005; Meinhold and Wright, 2011) and monitored equilibrium denaturation (McParland et al., 2002; Redfield et al., 1999) on a per residue basis by NMR.

A SSP score at a given residue of 1 reflects fully formed α-helical structure, whereas a score of, for example, 0.5 indicates that 50% of the conformers in the native state ensemble of the protein are helical at that position (Marsh et al., 2006). The SPP data (Figure 3A) indicate that about a third of the residues in native CesAB are found at positions that form fully or near-fully α-helical structure (SSP score 0.8–1), a third of the residues are located in regions that have substantial propensity to form α-helical structure, but with a significant fraction in the unfolded state (SSP score 0.5–0.7), whereas the other third of the residues are located in regions that have low propensity to form α-helical structure or are completely disordered (Figure 3A). RCI-S2 values, which may vary from S2=1, for a very rigid bond vector, to S2=0, for a very flexible bond vector, further support these observations (Figure 3B).

Figure 3. Folding and Dynamic Properties of the CesAB Homodimer.

Figure 3

Open in a new tab

(A) CesAB is colored according to the SSP values, using a gradient coloring scheme. Higher SSP values indicate higher propensity to form α helical structure.

(B) CesAB is colored according to the RCI-S2 values, using a gradient coloring scheme. Higher RCI-S 2 values indicate rigidity whereas lower values indicate flexibility.

(C–E) Residue-specific free energy of unfolding (ΔG0U-F) of CesAB (C), CesAB-E20L (D), and CesAB-D14L/R18D (E), determined from NMR-monitored residue-specific urea denaturation experiments, are mapped by continuous-scale color onto the structure of the CesAB homodimer. Higher ΔG0U-F values indicate regions of higher resistance to urea-induced denaturation and thus of higher stability.

See also Figure S3.

To better understand the conformational and folding properties of native CesAB, we used NMR to measure equilibrium denaturation. This approach is very powerful as it can provide residue-specific information about structural and dynamic transitions in partially folded, such as molten globule, conformational states (McParland et al., 2002). CesAB was titrated with increasing concentrations of urea (in the range of 0 to 9 M) and 1H-15N HSQC spectra were acquired at each urea concentration (increments of 0.5 M; Figure S3A). As the concentration of urea increases the line-broadening effect is greatly reduced and a large number of peaks start gradually appearing (Figure S3A). This is due to the fact that urea shifts the conformational equilibrium towards the unfolded state. By measuring the intensity of each one of the resonances as a function of urea concentration, a denaturation profile for each residue was constructed (Figure S3B and Experimental Procedures).

The NMR-determined urea denaturation profiles are distinct (Figure S3B), especially with regard to the per-residue free energy of unfolding (ΔG0U-F) (Figure 3C) or midpoint of denaturation (Cm) (Figure S3C), further corroborating the NMR data (Figures 1B and 1C) showing that CesAB is noncooperatively stabilized. The group of residues that is most resistant to urea denaturation consists of residues 17, 18, and 23–29, all located in helix α1, followed by residues 78–86 of helix α3. These results are in excellent agreement with the structure of the native CesAB homodimer pinpointing this surface as being primarily responsible for mediating CesAB dimerization (Figure 2). Taken together, the data show that the CesAB homodimer adopts a molten-globule-like structure, characterized by a relatively compact helical bundle conformation with regions of mixed folded and partially-folded secondary structure, but lacking extensive, specific tertiary side-chain packing characteristic of well folded structures.

Coiled-Coil Sequence Irregularities Cause Packing Defects in CesAB

Interactions within the helical bundle of CesAB are mediated by coiled-coil contacts and are thus governed by the properties of the amino acids along the ‘heptad’ sequence repeat in the form ‘abcdefg’ (Grigoryan and Keating, 2008). Analysis of the CesAB structure shows that there are coiled-coil sequence irregularities at several positions (Figures 4A and 4B). A prominent irregularity appears to be the presence of a charged residue (Glu) at position 20, a d position, most favorably occupied by a hydrophobic residue in coiled coils (Grigoryan and Keating, 2008). Thus, the interaction between the two CesAB subunits (α1-α1’ interface) must be unfavorable because it juxtaposes two like-charged residues (Figure 4A). Indeed, sequence optimization by substituting Leu for Glu20 (E20L) increases substantially the α-helical content and confers notable stabilization to the CesAB structure (Figures 4C–E). NMR analysis shows that the E20L substitution stabilizes the folded conformation of CesAB (Figure S4A) and near-UV data show a significant improvement in side-chain packing (Figure S4B). The structural data further suggest that substitution of Glu30 by a hydrophobic residue would strengthen the α1-α1’ interface by optimizing coiled-coil contacts. Indeed, the E30L substitution further stabilizes CesAB (Figure 4E). In fact, the double E20L/E30L substitution confers remarkable stability to CesAB homodimer with the melting temperature (Tm) of CesAB-E20L/E30L increasing by ~32 °C, as compared to wild-type CesAB (Figure 4E).

Figure 4. Coiled-Coil Irregularities Result in Suboptimal Contacts within the CesAB Helical Bundle.

Figure 4

Open in a new tab

(A) Close-up view of the CesAB intersubunit α1−α1’ coiled–coil interface. The residues of the heptad sequence that mediate CesAB dimerization are shown. Sequence irregularities at positions 20 and 30 are highlighted.

(B) Helical-wheel representation of the intersubunit α1−α1’ and α3−α1’ coiled coils highlighting (in yellow) the irregularities at positions 14, 18, 20 and 30 that prevent optimal juxtaposition.

(C) Far-UV CD data of CesAB and variants show that the amino acid substitutions increase helicity.

(D) Far-UV CD thermal denaturation data, monitored at 222 nm as a function of temperature, of CesAB and variants showing that the amino acid substitutions increase CesAB stability.

(E) Effect of amino acid substitutions on the melting temperature (Tm) of CesAB given as the difference between the Tm of substituted CesAB and the Tm of wild-type CesAB (ΔTm). Positive values denote increased stability of the substituted CesAB.

See also Figure S4.

Nevertheless, despite the great stabilization afforded to the CesAB homodimer by the substitutions at the α1-α1’ interface, the CesAB homodimer still appears to unfold in a rather non-cooperative manner, as evidenced by the CD-monitored thermal denaturation (Figure 4D, red profile) and residue-specific ΔG0U-F (Figure 3D) and Cm values (Figure S3C). Moreover, the NMR spectra show that substantial conformational exchange phenomena are still present in CesAB-E20L and CesAB-E20L/E30L (Figure S4A). Taken together, the data indicate that sequence optimization along the α1-α1’ interface of CesAB is not sufficient to yield a well-packed four-helix bundle.

Analysis of the CesAB structure highlighted two positions as possible contributors to the overall poor packing in CesAB: Asp14 and Arg18 (Figure 4B). Indeed, substitution of Asp14 by Leu and of Arg18 by Asp (to form a salt bridge with Ly77’), substitutions that are expected to enhance the inter-subunit interactions between helices α1 and α3’ (Figure 4B), stabilize CesAB in a highly cooperative manner as evidenced by the sigmoidal thermal denaturation profile (Figure 4D, orange), the uniform residue-specific ΔG0U-F (Figure 3E) and Cm values (Figure S3C) and the large free energy of global unfolding (ΔΔGun ~6.3 kcal mol−1). Collectively, the results indicate that CesAB has several “weak spots” at its helical bundle interface resulting in low stability and poor packing. Notably, the CesAB-D14L/R18D/E20L triple mutant, designed to correct the three most prominent sequence irregularities, results in complete suppression of the conformational exchange, and NMR peak dispersion and linewidths that are characteristic of well folded proteins (Figure S4C and S4D).

CesAB Folds upon Binding to its Physiological Substrate, the Translocator EspA

The CesAB homodimer undergoes subunit exchange to interact with its cognate substrate, the translocator EspA, to form a 1:1 heterodimeric complex (Figure 5A and Figure S5A) (Yip et al., 2005). NMR analysis shows that CesAB, which is poorly folded in the homodimer, acquires a well folded structure upon binding to EspA (Figures S5B and S5C), in agreement with the crystal structure of a truncated form of the heterodimer (Yip et al., 2005). NMR analysis of the intact complex shows that CesAB forms three well-folded α helices (α1, residues 3–46; α2, residues 50–60; and, α3, residues 67–85), while EspA forms four α helices and a β sheet (α1, residues 37–58; α2, residues 77–92; α3, residues 132–144; α4, residues 149–188; β1, residues 96–102; β2, residues 123–130) (Figures 5B and 5C and Figure S5D). The CesAB protomer adopts a similar overall fold in the homo- and heterodimer (Figure 5C); however, in the CesAB homodimer all three α helices are much shorter and largely unwound and dynamic, in contrast to the CesAB–EspA complex wherein they are well folded (Figure 5C and Figure S5D). Thermal denaturation of CesAB–EspA shows that the heterodimeric complex is more stable than the CesAB homodimer and unfolds in a largely cooperative fashion (Figure S5E). This system is rather unusual in that the least expected partner, the chaperone, appears to be in a metastable state in the absence of its substrate and becomes folded only when bound to it.

Figure 5. CesAB Homodimer vs. CesAB–EspA Heterodimer: Autoinhibition and Structural Mimicry.

Figure 5

Open in a new tab

(A) EspA binding to CesAB homodimer displaces one CesAB subunit to form a CesAB:EspA 1:1 heterodimeric complex in an apparent subunit exchange mechanism. The coloring scheme matches the coloring of the corresponding protein subunits on the other panels.

(B) Superposition of the CesAB homodimer (subunits are colored green and blue) and the CesAB–EspA heterodimer (CesAB is in orange and EspA in magenta).

(C) CesAB and CesAB–EspA are superimposed as in (B) but the second CesAB subunit in the homodimer and the EspA subunit in the heterodimer are not shown for clarity.

(D) EspA binding to CesAB stabilizes and induces folding of the chaperone in the heterodimeric complex by providing compensatory contacts to CesAB residues that form unfavorable contacts in the homodimer. Hydrogen bonds/salt bridges are represented by black dashed lines.

(E) Superposition of the homodimer and the heterodimer (as in (B)) reveals that CesAB adopts an autoinhibitory conformation. The CesAB helices α1 and α3 in the homodimer overlap structurally with helices α4 and α3, respectively, of EspA in the heterodimer. Colors are as in (B). CesAB is displayed as a solvent-accessible surface.

See also Figure S5.

EspA Uses Structural Mimicry to Selectively Stabilize the Heterodimer

Interestingly, structure analysis shows that EspA uses structural mimicry to induce folding to CesAB and to stabilize the heterodimeric complex. More specifically, whereas residues Glu20 and Glu30 destabilize the CesAB homodimer (Figure 4), EspA offsets these “weak spots” by providing juxtaposed amino acids (Arg174 and Gln181) that can form favorable polar interactions in the heterodimer: EspA Arg174 forms a salt bridge with CesAB Glu20 and EspA Gln181 forms a hydrogen bond with CesAB Glu30 (Figure 5D) (Yip et al., 2005). In addition, while helices α1 and α3’ in the homodimer are not properly packed because of the mismatch caused by Asp14 and Arg18 (Figure 4B), the residues at the corresponding positions in EspA in the heterodimer form optimal coiled-coil contacts with CesAB (Figure 5D): Leu180 forms hydrophobic coiled-coil contacts with the α3 helix of CesAB and Asp176 forms a salt bridge with Lys77 at the CesAB α3 helix. As a result of the improved packing, a considerably larger number of nonpolar and polar contacts are present in the heterodimer (Yip et al., 2005), which buries ~4,400 Å2, whereas the homodimer buries a maximum of ~1,200 Å2 (Figures S5F and S5G). The data show that EspA makes use of structural mimicry to form a complex with CesAB that has a similar fold to the homodimer, but takes advantage of sequence irregularities in CesAB and provides compensatory contacts that selectively stabilize the heterodimer over the homodimer.

The CesAB Homodimer is Autoinhibited

Notably, superposition of the structure of the CesAB homodimer on the structure of the CesAB–EspA heterodimer shows that the EspA-binding site in the CesAB homodimer is totally buried in the dimer interface (Figures 5B and 5E). Specifically, helices α1 and α3 of CesAB in the homodimer have very similar positions, relative to the other CesAB subunit, to the EspA helices α4 and α1, respectively (Figures 5B and 5E). These structural data demonstrate that the binding sites for one of the CesAB protomers and EspA are mutually exclusive. Thus, the CesAB homodimer adopts an autoinhibited conformation, apparently as a means to bury the extended hydrophobic surface presented by each protomer (Figure 2C) and to remain soluble in the absence of EspA. Thus, CesAB–EspA complex formation requires that EspA counteract the autoinhibitory arrangement to compete for one of the CesAB subunits.

CesAB Becomes Poised for EspA Binding by Transient Opening of its Structure Facilitated by Packing Defects

Relief of inhibition in autoregulated systems is typically accomplished by allosteric effector binding or covalent modifications that elicit substantial conformational changes (Pufall and Graves, 2002; Schlessinger, 2003). However, CesAB binds readily to EspA in the absence of such an external stimulus. On the basis of our collective data we hypothesized that the severe conformational exchange experienced by the CesAB homodimer is due to the transient opening and closure of helices α3 with respect to helices α1 (Figures 6A and S6A), a process that takes place on the milli-to-microsecond (ms-µs) time scale since it results in line broadening (Mittermaier and Kay, 2006). Indeed, conformational exchange is drastically suppressed in the NMR spectra of CesAB containing the double D14L/R18D substitution, which enhances cooperative interactions in CesAB by strengthening the α1−α3’ interface (Figures S4C, S4D and S6C). Additional evidence is provided by a dimeric truncated CesAB construct (CesAB1–48) lacking helices α2 and α3 that, in sharp contrast to full-length CesAB, unfolds in a largely cooperative manner and experiences no conformational exchange (Figure S6B). Taken together, the results provide strong evidence that the conformational exchange in CesAB is caused primarily by the opening and closure of the second “arm” of the hairpin, consisting of helices α2 and α3, relative to the first “arm”, consisting of helix α1 (Figure 6A).

Figure 6. CesAB Transient Opening Mechanism and Effect of its Suppression on EspA Binding and Secretion.

Figure 6

Open in a new tab

(A) Mechanism for the relief of autoinhibition in CesAB and subsequent binding to EspA. CesAB is poised for binding to the α1 helix of EspA by transiently detaching and exposing the α3 helices, which is facilitated by the poor packing at the α1−α3’ interface. The first docking event is followed by CesAB subunit dissociation, facilitated by the poor packing of the α1-α1’ interface, and formation of an ultimately stable CesAB–EspA heterodimer (see also Figures S6A–D).

(B) Effect of CesAB and EspA amino acid substitutions on the relative stability between the CesAB homodimer and the CesAB–EspA heterodimer (see Figures S6I and S6J) assessed by measuring the amount of CesAB bound to EspA. Higher values of EspA-bound CesAB indicate that the heterodimer is more stable than the homodimer. In the case of lower EspA-CesAB values, EspA cannot be prevented from forming filaments (Figure S6J).

(C) In vivo secretion of EspA from EPECΔcesAB strains that contained pASK-IBA7 plasmids expressing wild-type or mutated CesAB. The graph reports the total amount of EspA secreted in 120 min (Experimental Procedures).

(D) Infection of HeLa cells by bacterial EPECΔcesAB strains carrying plasmids that express CesAB or CesAB-E20l/E30L. When CesAB-E20L/E30L is present EspA secretion and filament formation is severely compromised and thus actin polymerization seen with wild type CesAB (white arrow) does not occur. b denotes bacterial cells and n denotes HeLa cell nuclei (Experimental Procedures).

See also Figure S6.

The transient opening of helix α3, which constitutes an EspA binding site (Figure 5B), could function as an anchoring structural element for the initial binding of EspA to CesAB homodimer. To test this hypothesis, we created a disulfide cross-linked CesAB variant wherein the α1 helices of the two CesAB subunits in the homodimer have been crosslinked to prevent dissociation (Figures S6D and S6E). Size-exclusion chromatography shows that EspA indeed binds to the CesAB α1-α1’ cpmqq-linked variant (CesABcl) forming a higher-molecular weight complex, despite the fact that the α1 helices, which provide the largest EspA binding surface, are not available for binding (Figure S6F). EspA eventually dissociates from CesABcl since formation of an ultimately stable heterodimeric complex requires CesAB homodimer dissociation. NMR characterization (Figure S6D) of this heterodimeric complex indeed confirms that helices α3 and α3’ of CesAB are bound to EspA, indicating that CesAB helices α3 and α3’ constitute the initial binding regions to EspA (Figure 6A). Thus, the results suggest that initial EspA docking to CesAB does not require prior dissociation of the homodimer; instead, EspA appears to bind to the transiently exposed α3 helices of the CesAB homodimer. This conclusion is further supported by the observation that a CesAB variant wherein the two α3 helices are crosslinked does not bind to EspA (Figure S6H). Collectively, the data suggest that the packing defects at the α1−α3’ (and α1’−α3) interface allows α3 (and α3’) to transiently detach from the helical bundle and become available for EspA binding (Figure 6A).

The Finely-Tuned Folding Properties of CesAB Regulate its Activity and the Biological System in vitro and in vivo

On the basis of our combined data we hypothesized that the intriguing structural and dynamic properties of CesAB modulate the transient opening of its structure and thus its affinity for EspA. To test this hypothesis we assessed the effect of various CesAB mutations on its binding activity for EspA by measuring the relative stability of the heterodimer over the homodimer (Figure 6B and Figures S6I and S6J). As noted above, the transient opening of the α1−α3’ interface provides the pathway for CesAB to relieve the autoinhibition mechanism and become poised for EspA binding (Figure 6A). Indeed, packing optimization between helices α1 and α3’ in the CesAB homodimer by the D14L/R18D substitution, which prevents the transient opening of the α3 helices (Figure S6C), decreases drastically the CesAB affinity for EspA binding (Figure 6B), and as a result almost all of the EspA protein forms filaments (Figure S6J). Therefore, transient opening of the α1−α3’ interface is necessary in order for CesAB to rapidly overcome the autoinhibitory conformation and efficiently capture EspA, which has a strong tendency to self-polymerize.

In addition to the α1−α3’ interface, the folding and dynamic properties of the α1−α1’ interface are also finely tuned. As expected, substitutions that improve the packing at the α1−α1’ interface, such as E20L and E30L, render the homodimer much more stable than the heterodimer (Figure 6B). Combined substitutions at both the α1−α1’ and α1−α3’ interface (e.g. CesAB-D14L/R18D/E20L) have a much stronger effect. Interestingly, in vivo genetic complementation assays show that EspA secretion is severely compromised in cells with cesAB genes carrying these packing-optimized mutants (Figure 6C). In addition, infection of HeLa cells by EPEC strains carrying packing-optimized cesAB mutated genes fail to cause actin polymerization indicating that secretion of TTS effectors is defective (Figures 6D and S6K). Control experiments show that these CesAB amino acid substitutions exert their effect primarily by stabilizing the homodimer and not by destabilizing the heterodimer (Figures S5H and S5I). Collectively, these functional assays provide strong evidence that a less labile CesAB becomes non-functional by causing severe EspA binding and secretion deficiencies.

Interestingly, the regulatory capacity of the α1−α3’ interface appears to be much higher than that of the α1−α1’ interface. For example, ΔG0U-F of CesAB-E20L is ~3 kcal mol−1 higher than that of CesAB-D14L/R18D (Figures 3D and 3E and Figure S3C) and thus it would be expected that the E20L substitution would have a much stronger effect, compared to D14L/R18D, on EspA binding. However, the opposite is observed (Figure 6B). Moreover, although EspA binds very weakly to CesAB-E20/E30L, EspA-R174L/Q181L, designed to optimally juxtapose with CesAB-E20L/E30L (Fig. 5D), forms a stable heterodimeric complex (Figure 6B). Nevertheless, this heterodimeric complex does not form at all when the transient exposure of the α3 helix is suppressed by the double D14L/R18D substitution (Figure 6B).

The CesAB Binding Mechanism is Evolutionary Conserved Among Other TTS Systems

SseA (Ruiz-Albert et al., 2003) and EscC (Zheng et al., 2007) are two chaperones whose function is to prevent oligomerization of their cognate translocators, SseB and EseB, respectively, in the TTS systems of two different pathogenic bacteria (Salmonella sp. and Ed. Tarda, respectively) (Figure S7A). MALS data show that both SseB and EscC chaperones are dimeric but they form a 1:1 heterodimeric complex with their cognate translocators (Figures S7B and S7C). Thus, similarly to CesAB binding to EspA, these chaperones undergo a subunit exchange mechanism upon interacting with their substrates (Figure S7C). The CD data show that both chaperones are α-helical (Figure S7D) and undergo a non-cooperative unfolding transition (Figure S7E). Remarkably, NMR analysis of SseA and EscC homodimers demonstrate that both chaperones adopt a molten-globule-like conformation (Figure S7F). Binding of their cognate translocator induces folding to the chaperones (Figure S7F). These data show that the exchange subunit mechanism underlying the CesAB binding mechanism to its substrate (EspA), a mechanism assisted and stimulated by the dynamic and folding properties of the dimeric chaperone, is evolutionary conserved among other chaperones in different bacterial organisms and TTS systems.

DISCUSSION

Regulation of the vast number of cellular protein-protein interactions may be underpinned by a multitude of mechanisms (Boehr et al., 2009; Kuriyan and Eisenberg, 2007; Smock and Gierasch, 2009; Tokuriki and Tawfik, 2009; Tsai et al., 2009). The discovery of additional mechanisms will allow better understanding of how protein complexes are regulated and will advance our ability to manipulate their formation or disruption. Here we report an evolutionary conserved mechanism that tunes the dynamic and folding properties at protein interfaces as a means to regulate binding.

The cesAB and espA genes are located 25-kilobases apart and are thus not co-transcribed. Because EspA has a strong tendency to quickly self-oligomerize to form filaments, proper function of the TTS system requires that CesAB rapidly capture EspA in the monomeric state (Figure 7). Because of the opposite charge of CesAB and EspA (pI is ~9.2 and 4.3, respectively), electrostatic steering (Sheinerman et al., 2000) could accelerate the association kinetics of complex formation. However, given the fact that CesAB exists in an autoinhibitory conformation, fast and productive formation of the CesAB–EspA heterodimer can occur only if CesAB becomes rapidly poised for EspA binding. CesAB accomplishes this by transiently exposing an EspA-binding region with sub-microsecond kinetics (helices α3; Figure 7). This mechanism bypasses the need for a complete CesAB dissociation, which, because of the intertwined nature of the CesAB structure (Figure 2A), would likely require complete unfolding of the subunits and it would thus be quite slow. The mechanism by which CesAB becomes poised for EspA binding is strongly stimulated by packing defects at the α1−α3’ interface originating in coiled-coil sequence irregularities (Figure 4B). Correction of these sequence irregularities suppresses the binding mechanism (Figure 7) giving rise to a non-functional TTS system (Figure 6). For the initial CesAB–EspA intermediate to collapse to a stable complex (Figures 6A and 7), the CesAB homodimer will have to undergo efficient subunit exchange. Coiled-coil sequence irregularities at the α1−α1’ interface results in a rather unstable dimerization interface in CesAB (Figure 4), enhancing the efficiency of the partner exchange mechanism.

Figure 7. Model for Relief of Autoinhibition in CesAB and EspA Binding Facilitated by Finely-Tuned Instability.

Figure 7

Open in a new tab

Once EspA has been synthesized it should be rapidly captured by CesAB (reaction a) otherwise EspA tends to quickly self-oligomerize to form filaments (reaction b). CesAB homodimer exists predominantly in a closed, autoinhibited conformation (1) but it transiently populates an open state, stimulated by packing defects at the α1−α3’ interface, wherein helix α3 is accessible to EspA for binding. Because the transient opening of helix α3 is fast (it occurs on the sub-millisecond time scale) it can effectively capture EspA in its monomeric state. Optimization of the contacts at the α1−α3’ interface results in suppression of the transient opening of helix α3 (2) and thus abrogation of the binding of EspA by CesAB (reaction a’). See also Figure S7.

It should be noted that the functionality of this system is not simply dependent on the concentration difference between the homodimer and the heterodimer and their relative affinities, but rather on the efficiency of the disordered-stimulated binding mechanism. This is clearly demonstrated by the fact that the D14L/R18D substitution, which suppresses the binding mechanism, causes a drastic decrease in EspA binding irrespectively of the extent of the α1−α1’ dimerization interface optimization (Figure 6B). For example, even though the stability afforded to CesAB dimerization by E20L is greater than the stability afforded by D14L/R18D (Figures 3D and 3E), the effect of the latter on reducing CesAB affinity for EspA is much stronger than the former (Figure 6B). Even the single D14L substitution, which has a negligible effect on the stabilization of the CesAB dimerization (Figure 4E), has a stronger effect than the E20L substitution on preventing EspA secretion (Figure 6C). Thus, if CesAB were well folded it would be incapable of interacting with EspA, even if the dimerization interface were weak enough.

Although the instability and dynamic nature of the α1−α3’ interface is crucial for stimulating the binding mechanism of CesAB, the α1−α1’ interface, which constitutes the main dimerization interface, is also important for the formation of a functional CesAB–EspA complex for two reasons. First, the low stability of the α1−α1’ interface stimulates partner exchange so that all of the CesAB homodimer dissociates to form a complex with EspA. Second, if the intersubunit contacts at the α1−α1’ interface were optimal, e.g. in the case of CesAB-E20L/E30L, a CesAB–EspA complex would form only if EspA had evolved to optimally juxtapose with this form of CesAB, e.g. EspA-R174L/Q181L (Figure 6B). However, the resulting heterodimer is of much higher stability than the wild-type CesAB–EspA complex (ΔTm ~20 °C), and prevents EspA secretion (Figure 6C), presumably because the heterodimer is impossible to dissociate at the injectisome base, thus resulting in a non-functional system. Indeed, TTS chaperone−substrate complexes of excessive stability cannot be dissociated and are not secreted (Akeda and Galan, 2005; Sorg et al., 2005).

Apparently, in order to carry out its function with maximum efficiency, the CesAB chaperone has evolved to adopt a partially folded, molten-globule-like conformation. Such a conformational state affords CesAB a significant regulatory capacity, enabling it to finely tune the structural and dynamic properties of both α1−α1’ and α1−α3’ interface simultaneously. The fact that other chaperones from different TTS systems that encounter the same challenge have also evolved to adopt molten-globule-like conformations (Figure S7) strongly argue for the physiological relevance of this mechanism.

It is of particular interest that the translocator EspA makes use of structural mimicry to interact with the homodimeric CesAB chaperone, displace one of the subunits and form a stable heterodimeric complex (Figure 5B). While structural mimicry is a common strategy used by TTS effector proteins to interact with target eukaryotic proteins (Elde and Malik, 2009; Galan, 2009), our data suggest that structural mimicry may in fact be a widespread mechanism among various TTS components.

Molten globule conformations in proteins emerge as a particularly efficient mechanism to regulate binding/enzymatic activities and allosteric interactions (Demarest et al., 2002; Liu and Nussinov, 2008; Pervushin et al., 2007) and may even be a common feature in TTS (Dawson et al., 2009; Faudry et al., 2007). The CesAB chaperone provides a compelling example of a system that exploits structural instability for function as has been shown previously for a number of biological systems (Dyson and Wright, 2005).

EXPERIMENTAL PROCEDURES

Protein preparation

Procedures for cloning, expression, purification and isotopic labeling of recombinant proteins utilized for NMR studies and in vitro biochemical assays are described in the Supplemental Experimental Procedures.

NMR spectroscopy

Procedures for NMR characterization and resonance assignment are described in the Supplemental Experimental Procedures.

NMR urea denaturation experiments

This approach was used to assess the effect of amino acid substitutions on the stability and folding cooperativity of CesAB. Two CesAB samples with identical protein concentration (0.5 mM) were prepared, one containing protein in 9 M urea and the second containing protein in the absence of urea. The two solutions were then mixed to give separate samples with urea concentrations ranging from 0 to 9 M urea (in 0.5 M increments). Each sample was equilibrated for 2 h at 25 °C before acquisition of the 1H-15N HSQC spectrum. The intensity of every non-degenerate resonance observed in each spectrum during the urea titration was measured, normalized and a denaturation profile for each residue was constructed (Figure S3B). Details for the determination of the apparent midpoint (Cm) of the unfolding transition and the per-residue free energy of (un)folding are described in the Supplemental Experimental Procedures.

Determination of secondary structure propensity (SSP) Values

The chemical shift deviation from random coil values were used to assess the secondary structure propensities along the backbone of CesAB homodimer. Random coil values were extracted from the chemical shifts of CesAB in 9 M urea. In this case chemical shift differences are sequence-independent and so there is no need for referencing correction. The 13Cα, 13C’, and and 13Cβ chemical shifts were used to assess the secondary structure propensity of CesAB homodimer in a residue-specific manner as described (Marsh et al., 2006).

Approach for the structure determination of the CesAB homodimer

Procedures for the determination of NOEs, paramagnetic relaxation enhancement (PRE), and dihedral angle restraints as well as the structure calculation protocol are described in the Supplemental Experimental Procedures. Briefly, A significant number of restraints (Table S1) were collected for CesAB homodimer in 10% TFE, consisting of medium and long-range NOEs (Figure S1D) and PREs and dihedral angle restraints. The final structure of CesAB under native conditions (0% TFE) was obtained by refining the initial structure determined for CesAB in 10% TFE by incorporating restraints obtained for CesAB under native conditions to more reliably represent the secondary and tertiary structure of the native state. Restraints used for the final structure determination of the native CesAB homodimer included NH-NH, NH-methyl (Ala, Ile, Leu, and Val residues) and methyl-methyl NOEs, PREs and dihedral angle restraints (Table S1).

CD spectroscopy and thermal unfolding

Procedures for the collection of CD spectra and the determination of transition temperatures (Tm) are described in the Supplemental Experimental Procedures.

In vivo secretion from EPEC strains and infection of HeLa cells

Procedures for the secretion and infection assays are described in the Supplemental Experimental Procedures. Briefly, in vivo secretion was induced by incubation of derivatives of EPEC strain E2348/69 at 37 °C and the expression of the different genes was induced when OD600 reached ~0.3. Total cells and supernatant fractions were separated by centrifugation, precipitated and analyzed by SDS-PAGE and immunoblotting. Secreted polypeptides were analyzed by 15% SDS-PAGE and immunoblotting using rabbit polyclonal antibody against EspA. All the loaded samples were adjusted so as to represent equal numbers of bacteria. For the in vivo infection assays, bacterial cultures of EPEC strains were grown for 18 h in LB. Sub-confluent lawns of HeLa cells were infected for 2 h with the primed bacterial cultures. Bacterial cell numbers were calculated from the OD600 of each culture and ~3.5 × 107 bacteria were inoculated in each well.

Supplementary Material

01

ACKNOWLEDGEMENTS

We wish to thank: G. Frankel for gifts of antibodies, primers, strains and protocols; V. Crepin-Sevenou for training on in vivo infection assays and gene replacement at Imperial College (London, UK); S. Karamanou for MALLS experiments and guidance on antibody and protein handling; G. Sianidis and M. Koukaki for mutagenesis and cloning; B. Pozidis for MALLS experiments and protein preparations; J. Zheng, J.Kaper and V. Koronakis for gifts of plasmids and genomic DNA. This work was supported by NIH grant AI094623 (C.G.K.) and by the Greek General Secretariat of Research PENED03ED623 (A.E.) and the European Regional Development Fund 01AKMON46 (A.E.).

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCESSION NUMBERS

Structural coordinates for CesAB are deposited in the Protein Data Bank with the accession code 2LHK.

SUPPLEMENTAL INFORMATION

Supplemental Information includes Supplemental Experimental Procedures and seven figures and can be found with this article online at…

REFERENCES

  1. Akeda Y, Galan JE. Chaperone release and unfolding of substrates in type III secretion. Nature. 2005;437:911–915. doi: 10.1038/nature03992. [DOI] [PubMed] [Google Scholar]
  2. Berjanskii MV, Wishart DS. A simple method to predict protein flexibility using secondary chemical shifts. J Am Chem Soc. 2005;127:14970–14971. doi: 10.1021/ja054842f. [DOI] [PubMed] [Google Scholar]
  3. Birtalan SC, Phillips RM, Ghosh P. Three-dimensional secretion signals in chaperone-effector complexes of bacterial pathogens. Mol Cell. 2002;9:971–980. doi: 10.1016/s1097-2765(02)00529-4. [DOI] [PubMed] [Google Scholar]
  4. Boehr DD, Nussinov R, Wright PE. The role of dynamic conformational ensembles in biomolecular recognition. Nat Chem Biol. 2009;5:789–796. doi: 10.1038/nchembio.232. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Buck M. Trifluoroethanol and colleagues: cosolvents come of age. Recent studies with peptides and proteins. Q Rev Biophys. 1998;31:297–355. doi: 10.1017/s003358359800345x. [DOI] [PubMed] [Google Scholar]
  6. Cornelis GR. The type III secretion injectisome. Nat Rev Microbiol. 2006;4:811–825. doi: 10.1038/nrmicro1526. [DOI] [PubMed] [Google Scholar]
  7. Creasey EA, Friedberg D, Shaw RK, Umanski T, Knutton S, Rosenshine I, Frankel G. CesAB is an enteropathogenic Escherichia coli chaperone for the type-III translocator proteins EspA and EspB. Microbiology. 2003;149:3639–3647. doi: 10.1099/mic.0.26735-0. [DOI] [PubMed] [Google Scholar]
  8. Cross BCS, Sinning I, Luirink J, High S. Delivering proteins for export from the cytosol. Nat Rev Mol Cell Biol. 2009;10:255–264. doi: 10.1038/nrm2657. [DOI] [PubMed] [Google Scholar]
  9. Dawson JE, Seckute J, De S, Schueler SA, Oswald AB, Nicholson LK. Elucidation of a pH-folding switch in the Pseudomonas syringae effector protein AvrPto. Proc Natl Acad Sci U S A. 2009;106:8543–8548. doi: 10.1073/pnas.0809138106. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Dean P, Kenny B. The effector repertoire of enteropathogenic E.coli: ganging up on the host cell. Curr Opin Microbiol. 2009;12:101–109. doi: 10.1016/j.mib.2008.11.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Demarest S, Martinez-Yamout M, Chung J, Chen H, Xu W, Dyson H, Evans R, Wright P. Mutual synergistic folding in recruitment of CBP/p300 by p160 nuclear receptor coactivators. Nature. 2002;415:549–553. doi: 10.1038/415549a. [DOI] [PubMed] [Google Scholar]
  12. Dyson HJ, Wright PE. Intrinsically unstructured proteins and their functions. Nat Rev Mol Cell Bio. 2005;6:197–208. doi: 10.1038/nrm1589. [DOI] [PubMed] [Google Scholar]
  13. Elde NC, Malik HS. The evolutionary conundrum of pathogen mimicry. Nat Rev Microbiol. 2009;7:787–797. doi: 10.1038/nrmicro2222. [DOI] [PubMed] [Google Scholar]
  14. Faudry E, Job V, Dessen A, Attree I, Forge V. Type III secretion system translocator has a molten globule conformation both in its free and chaperone-bound forms. Febs J. 2007;274:3601–3610. doi: 10.1111/j.1742-4658.2007.05893.x. [DOI] [PubMed] [Google Scholar]
  15. Feldman MF, Cornelis GR. The multitalented type III chaperones: all you can do with 15 kDa. FEMS Microbiol Lett. 2003;219:151–158. doi: 10.1016/S0378-1097(03)00042-9. [DOI] [PubMed] [Google Scholar]
  16. Galan J. Common themes in the design and function of bacterial effectors. Cell Host Microbe. 2009;5:571–579. doi: 10.1016/j.chom.2009.04.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Galan JE, Wolf-Watz H. Protein delivery into eukaryotic cells by type III secretion machines. Nature. 2006;444:567–573. doi: 10.1038/nature05272. [DOI] [PubMed] [Google Scholar]
  18. Grigoryan G, Keating AE. Structural specificity in coiled-coil interactions. Curr Opin Struct Biol. 2008;18:477–483. doi: 10.1016/j.sbi.2008.04.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Hartl FU, Hayer-Hartl M. Converging concepts of protein folding in vitro and in vivo. Nat Struct Mol Biol. 2009;16:574–581. doi: 10.1038/nsmb.1591. [DOI] [PubMed] [Google Scholar]
  20. Haslbeck M, Franzmann T, Weinfurtner D, Buchner J. Some like it hot: the structure and function of small heat-shock proteins. Nat Struct Mol Biol. 2005;12:842–846. doi: 10.1038/nsmb993. [DOI] [PubMed] [Google Scholar]
  21. Hill RB, Raleigh DP, Lombardi A, Degrado WF. De novo design of helical bundles as models for understanding protein folding and function. Accounts of Chemical Research. 2000;33:745–754. doi: 10.1021/ar970004h. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kim A, Kakalis L, Abdul-Manan N, Liu G, Rosen M. Autoinhibition and activation mechanisms of the Wiskott-Aldrich syndrome protein. Nature. 2000;404:151–158. doi: 10.1038/35004513. [DOI] [PubMed] [Google Scholar]
  23. Knutton S, Rosenshine I, Pallen M, Nisan I, Neves B, Bain C, Wolff C, Dougan G, Frankel G. A novel EspA-associated surface organelle of enteropathogenic Escherichia coli involved in protein translocation into epithelial cells. EMBO J. 1998;17:2166–2176. doi: 10.1093/emboj/17.8.2166. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Kobe B, Kemp BE. Active site-directed protein regulation. Nature. 1999;402:373–376. doi: 10.1038/46478. [DOI] [PubMed] [Google Scholar]
  25. Kuriyan J, Eisenberg D. The origin of protein interactions and allostery in colocalization. Nature. 2007;450:983–990. doi: 10.1038/nature06524. [DOI] [PubMed] [Google Scholar]
  26. Liu J, Nussinov R. Allosteric effects in the marginally stable von Hippel-Lindau tumor suppressor protein and allostery-based rescue mutant design. Proc Natl Acad Sci U S A. 2008;105:901–906. doi: 10.1073/pnas.0707401105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Marsh JA, Singh VK, Jia Z, Forman-Kay JD. Sensitivity of secondary structure propensities to sequence differences between alpha- and gamma-synuclein: implications for fibrillation. Protein Sci. 2006;15:2795–2804. doi: 10.1110/ps.062465306. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. McNamara C, Zinkernagel AS, Macheboeuf P, Cunningham MW, Nizet V, Ghosh P. Coiled-coil irregularities and instabilities in group A Streptococcus M1 are required for virulence. Science. 2008;319:1405–1408. doi: 10.1126/science.1154470. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. McParland VJ, Kalverda AP, Homans SW, Radford SE. Structural properties of an amyloid precursor of beta(2)-microglobulin. Nat Struct Biol. 2002;9:326–331. doi: 10.1038/nsb791. [DOI] [PubMed] [Google Scholar]
  30. Meinhold DW, Wright PE. Measurement of protein unfolding/refolding kinetics and structural characterization of hidden intermediates by NMR relaxation dispersion. Proc Natl Acad Sci U S A. 2011;108:9078–9083. doi: 10.1073/pnas.1105682108. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Mittermaier A, Kay LE. New tools provide new insights in NMR studies of protein dynamics. Science. 2006;312:224–228. doi: 10.1126/science.1124964. [DOI] [PubMed] [Google Scholar]
  32. Parsot C, Hamiaux C, Page AL. The various and varying roles of specific chaperones in type III secretion systems. Curr Opin Microbiol. 2003;6:7–14. doi: 10.1016/s1369-5274(02)00002-4. [DOI] [PubMed] [Google Scholar]
  33. Pervushin K, Vamvaca K, Vogeli B, Hilvert D. Structure and dynamics of a molten globular enzyme. Nat Struct Mol Biol. 2007;14:1202–1206. doi: 10.1038/nsmb1325. [DOI] [PubMed] [Google Scholar]
  34. Pufall M, Graves B. Autoinhibitory domains: modular effectors of cellular regulation. Annu Rev Cell Dev Biol. 2002;18:421–462. doi: 10.1146/annurev.cellbio.18.031502.133614. [DOI] [PubMed] [Google Scholar]
  35. Receveur-Brechot V, Bourhis JM, Uversky VN, Canard B, Longhi S. Assessing protein disorder and induced folding. Proteins. 2006;62:24–45. doi: 10.1002/prot.20750. [DOI] [PubMed] [Google Scholar]
  36. Redfield C, Schulman BA, Milhollen MA, Kim PS, Dobson CM. Alpha-lactalbumin forms a compact molten globule in the absence of disulfide bonds. Nat Struct Biol. 1999;6:948–952. doi: 10.1038/13318. [DOI] [PubMed] [Google Scholar]
  37. Ruiz-Albert J, Mundy R, Yu X, Beuzon C, Holden D. SseA is a chaperone for the SseB and SseD translocon components of the Salmonella pathogenicity-island-2-encoded type III secretion system. Microbiology. 2003;149:1103–1111. doi: 10.1099/mic.0.26190-0. [DOI] [PubMed] [Google Scholar]
  38. Schlessinger J. Signal transduction. Autoinhibition control. Science. 2003;300:750–752. doi: 10.1126/science.1082024. [DOI] [PubMed] [Google Scholar]
  39. Sheinerman F, Norel R, Honig B. Electrostatic aspects of protein-protein interactions. Curr Opin Struct Biol. 2000;10:153–159. doi: 10.1016/s0959-440x(00)00065-8. [DOI] [PubMed] [Google Scholar]
  40. Smock RG, Gierasch LM. Sending signals dynamically. Science. 2009;324:198–203. doi: 10.1126/science.1169377. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Sorg JA, Miller NC, Marketon MM, Schneewind O. Rejection of impassable substrates by Yersinia type III secretion machines. J Bacteriol. 2005;187:7090–7102. doi: 10.1128/JB.187.20.7090-7102.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Stirling P, Bakhoum S, Feigl A, Leroux M. Convergent evolution of clamp-like binding sites in diverse chaperones. Nat Struct Mol Biol. 2006;13:865–870. doi: 10.1038/nsmb1153. [DOI] [PubMed] [Google Scholar]
  43. Tokuriki N, Tawfik DS. Protein dynamism and evolvability. Science. 2009;324:203–207. doi: 10.1126/science.1169375. [DOI] [PubMed] [Google Scholar]
  44. Tsai CJ, Ma B, Nussinov R. Protein-protein interaction networks: how can a hub protein bind so many different partners? Trends Biochem Sci. 2009;34:594–600. doi: 10.1016/j.tibs.2009.07.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Waksman G, Hultgren SJ. Structural biology of the chaperone-usher pathway of pilus biogenesis. Nat Rev Microbiol. 2009;7:765–774. doi: 10.1038/nrmicro2220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Yip CK, Finlay BB, Strynadka NC. Structural characterization of a type III secretion system filament protein in complex with its chaperone. Nat Struct Mol Biol. 2005;12:75–81. doi: 10.1038/nsmb879. [DOI] [PubMed] [Google Scholar]
  47. Yu H, Braun P, Yildirim MA, Lemmens I, Venkatesan K, Sahalie J, Hirozane-Kishikawa T, Gebreab F, Li N, Simonis N, et al. High-quality binary protein interaction map of the yeast interactome network. Science. 2008;322:104–110. doi: 10.1126/science.1158684. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Zheng J, Li N, Tan Y, Sivaraman J, Mok Y, Mo Z, Leung K. EscC is a chaperone for the Edwardsiella tarda type III secretion system putative translocon components EseB and EseD. Microbiology. 2007;153:1953–1962. doi: 10.1099/mic.0.2006/004952-0. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

01
NIHMS337172-supplement-01.doc (2.1MB, doc)

RESOURCES