HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: ps.041164305v1 14/5/1375    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Donald, L. J. Articles by Duckworth, H. W. Search for Related Content PubMed PubMed Citation Articles by Donald, L. J. Articles by Duckworth, H. W. Social Bookmarking                   What's this?

FOR THE RECORD

Multiple equilibria of the Escherichia coli chaperonin GroES revealed by mass spectrometry

¹ Department of Chemistry, ² Department of Entomology, and ³ Department of Physics and Astronomy, University of Manitoba, Winnipeg, Manitoba, R3T 2N2, Canada

Reprint requests to: Lynda J. Donald, Department of Chemistry, 507 Parker Building, University of Manitoba, Winnipeg, Manitoba, R3T 2N2, Canada; e-mail: ldonald{at}cc.umanitoba.ca; fax: (204) 474-7608.

(RECEIVED October 21, 2004; FINAL REVISION December 21, 2004; ACCEPTED January 7, 2005)

	Abstract

TOP Abstract Introduction Results Discussion Materials and methods References

 
Nanospray time-of-flight mass spectrometry has been used to study the assembly of the heptamer of the Escherichia coli cochaperonin protein GroES, a system previously described as a monomer–heptamer equilibrium. In addition to the monomers and heptamers, we have found measurable amounts of dimers and hexamers, the presence of which suggests the following mechanism for heptamer assembly: 2 Monomers Dimer; 3 Dimers Hexamer; Hexamer + Monomer Heptamer. Equilibrium constants for each of these steps, and an overall constant for the Monomer Heptamer equilibrium, have been estimated from the data. These constants imply a standard free-energy change, G⁰, of about 9 kcal/mol for each contact surface formed between GroES subunits, except for the addition of the last subunit, where G⁰ = 6 kcal/mol. This lower value probably reflects the loss of entropy when the heptamer ring is formed. These experiments illustrate the advantages of electrospray mass spectrometry as a method of measuring all components of a multiple equilibrium system.

Keywords: electrospray ionization; time-of-flight mass spectrometry; E. coli chaperone protein; GroES equilibrium

Article published online ahead of print. Article and publication date are at http://www.proteinscience.org/cgi/doi/10.1110/ps.041164305.

	Introduction

TOP Abstract Introduction Results Discussion Materials and methods References

 
Mass spectrometry has proven useful in the study of dynamic protein complexes (Hernandez and Robinson 2001), one of which is the Escherichia coli chaperonin assembly of GroEL (with 14 subunits) and its cochaperonin GroES (with seven subunits) (Xu et al. 1997). In the gas phase, GroEL is a stable 14_mer, which breaks into 7_mers and then monomers as the collision energy is increased (Rostom and Robinson 1999). The complex of GroEL and GroES, as well as GroEL with other cochaperonins, has also been studied by mass spectrometry (Pinske et al. 2003).

The GroES heptamer has been mostly overlooked in these measurements, although one GroES spectrum has been reported (Hernandez and Robinson 2001). However, Zondlo et al. (1995) presented good evidence for a monomer–heptamer equilibrium in GroES by comparison of mutant and wild-type proteins in sedimentation equilibrium and hybridization studies. These studies are probably incomplete, since assembly of monomers into heptamers could go through intermediates not obvious in such biochemical measurements, which depend on averages.

Mass spectrometry is not subject to these limitations. It can show all of the ionizable components in the mixtures under appropriate conditions of buffer and voltage. For biologically active proteins, the nanospray ionization of Wilm and Mann (1996) is a gentle transition into the gas phase, because it requires relatively low interface voltages and is tolerant of high concentrations of volatile buffers. The latter was essential to finding conditions suitable for examination of GroES, and we present evidence here for its stepwise assembly, with reasonable estimates of the association constants.

	Results

TOP Abstract Introduction Results Discussion Materials and methods References

 
The simple model of Zondlo et al. (1995) suggests that there should be only monomer ions at low-protein concentrations, and only heptamer ions at high-protein concentrations. Actually, the situation is much more complicated, as illustrated in Figure 1, which represents the situation in 50 mM NH₄HCO₃ and 0.1 mM DTT buffer. At 0.31 µM protein, the lowest concentration for which useful spectra could be acquired, there were four ions that could be assigned as monomer with charge states 4+ to 7+. In addition, there were two more ions that can only be from a dimer with charge states 7+ and 9+. The ion at m/z 2597.5 corresponds to two species, the 4+ monomer ion and the 8+ dimer ion. An increase in the protein concentration, to 0.92 µM, produced two more ion envelopes, a 20+ to 22+ series for hexamer and a 16+ to 21+ series for the expected heptamer. Again, there was another ion that represents two species—the 21+ ion of the hexamer having the same m/z value as the 7+ ion from dimer. Further increases in the protein concentration change the spectrum to more and more heptamer. However, the ions from the monomer persist, although with a somewhat broader charge distribution extending to lower charge states. We note that increasing the collision energy (V_c = 150V), increasing the back pressure on the nanospray capillary, or using old samples changed the spectrum to one with a second envelope of hexamer ions with lower charge states (12+ to 14+) at m/z 4500, and a marked increase in the abundance of monomer ions. Decreasing the buffer concentration to 5 mM NH₄HCO₃, the most common condition used for electrospray ionization, changed the spectrum completely to a broad envelope of monomer ions only, a spectrum typical of denatured proteins. Deconvolution of the Figure 1 spectra showed four species: a monomer of mass 10,387 Da (Full Width Half Maximum, FWHM = 6.6 Da), a dimer of mass 20,773 Da (FWHM = 7.9 Da), a hexamer of mass 62,322 Da (FWHM = 26.6 Da), and a heptamer of mass 72,709 Da (FWHM = 24.5 Da).

View larger version (25K): [in this window] [in a new window]   Figure 1. Representative nanospray spectra of GroES protein at different dilutions in 50 mM NH4HCO3, 0.1 mM DTT. Ions are labeled by the number of subunits with their charge states shown as superscripts. Note that some peaks could arise from more than one species (e.g., 14+ and 28+; 27+ and 621+) because the mass spectrometer measures the ratio m/z. Unlabeled ions are minor, and only appear at high concentration of protein. Spectra were acquired with 100 V declustering voltage with SF6 as the curtain gas. (A) 0.31 µM protein, 10-min acquisition. Only dimer (27+, 28+, 29+) and monomer (14+, 15+, 16+, 17+) ions are visible. (B) A 0.92-µM protein, 10-min acquisition. Although monomer has the highest ion counts, there are also measurable ions from dimer, hexamer, and heptamer. (C) A 19-µM protein, 2-min acquisition. Heptamer ions are now the most prominent, but monomer, dimer, and hexamer ions persist.

Both the raw spectra and a preliminary analysis of the data show clearly that monomer and heptamer are the major components, together accounting for >90% of the ions when the protein concentration is higher than 5 µM. As expected from an equilibrium system, there is relatively more monomer at low-protein concentration, and more heptamer as the protein concentration is increased. However, it is clear that dimers and hexamers are also present, and can therefore contribute to the overall equilibrium. This concentration-dependent change in distribution of species suggests a pathway for assembly for GroES heptamer (Hp) from monomer (M), involving dimer (D), and hexamer (Hx) as intermediates.

Based on this description, the following set of equilibria should apply, and values for the association constants can be estimated:

(1)

(2)

(3)

Assuming that the equilibria are interdependent, then the overall equation can be written as

(4)

and therefore,

(5)

Assigning

R₁

ions from dimers/ions from monomers

R₂

ions from hexamers/ions from monomers,

R₃

ions from heptamers/ions from monomers,

(6)

then,

(7)

(8)

and

(9)

The graph of data points for Equation 7 is shown in Figure 2. Theoretically, the plot should be a straight line, passing through the origin, with slope K_a. At low-protein concentrations (< 5 µM) the measurements seem consistent with this picture, but at higher concentrations, the data (compiled from 56 spectra) show considerable scatter and appear to reach a plateau at high-protein concentration. Similar scatter and plateau effects were also seen in the solution of Equations 8 and 9.

View larger version (15K): [in this window] [in a new window]   Figure 2. Solution of Equation 7 for calculation of Ka: R1 (1 + 2R1 + 6R2 + 7R3) as a function of GroES concentration (by subunit) up to 35 µM protein. The empirical fit is the solid line, y = –(9.38 ± 0.80) (1 – (0.78 ± 0.04)[Pr]) and the tangent at [Pr] = 0 is the dotted line with slope Ka = 2.3 ± 0.5 µM–1.

Thus, an empirical model was used to fit a line to the data in order to determine whether there was a relationship between the rate of change of y and the protein concentration, [Pr]. If so, then a tangent to that line at [Pr] = 0 would represent the hypothetical relationship if there were no fragmentation, and the slope would be an estimate of the relevant association constant. On theoretical grounds, the tangent was forced through the origin, and doing so did not materially reduce the goodness of fit. The fitted equation was:

(10)

where b represents the asymptote and determines the general nature of the slope and is the shape parameter. The left side of Equations 7, 8, or 9 is represented by y. Fitting was by least squares Gauss-Newton iteration using preliminary parameter estimates derived from a graphical examination of the data (Systat 2000). The slope of the line at any value of [Pr] is

(11)

when [Pr] = 0, the slope of the tangent is

(12)

and this equation was used to estimate the association constants. Errors were determined for b and , and only the latter was used to estimate the range for the value of the equilibrium constant as it was the major factor. The tangent describing K_a is a good fit to the data, but only at low-protein concentration (Fig. 2). The tangents for K_b^1/5 and K_c^1/6 also fit well only at low-protein concentration (data not shown).

The values of the three association constants, determined from the slope of the tangents, and the overall value for the heptamer/monomer equilibrium, are listed in Table 1. The error for K_a could be calculated directly from the fitted equation. However, estimation of errors for K_b and K_c was complicated by the complexity of the equations. We have chosen to present a range of estimates, based on the error in . Although the uncertainties in the values of the three association constants (Table 1) are relatively large, the ranges of protein concentrations over which each of the equilibria occurs are reasonably well defined by our data. Thus, when the association constants are used to calculate the standard free-energy changes for the association process, the uncertainties are small enough for the values to be useful (Table 1). The results imply that the standard free-energy changes accompanying formation of a dimer from two monomers, or a hexamer from three dimers, are roughly the same per new contact surface formed. The change accompanying completion of the GroES heptameric ring is somewhat less per contact surface formed, perhaps because of a greater loss of entropy as the ring is completed. These results fit well with the known GroES structure, an open heptameric ring in which the seven contact surfaces are identical (Hunt et al. 1996).

	Discussion

TOP Abstract Introduction Results Discussion Materials and methods References

We did not anticipate the fragility of the heptamer at high-protein concentration. Using conventional ESI, the considerably larger E. coli citrate synthase hexamer (287,310 Da) was stable up to 100 µM protein in 5 mM NH₄HCO₃ (Ayed et al. 1998), and nanospray experiments with that same protein have not shown any evidence of fragmentation because of high-protein concentration (L.J. Donald, un-publ.). The interface parameters were chosen to give the best resolution of the heptamer at the lowest reasonable voltage. No conditions could be found where only heptamer ions were present. Increase of voltage or protein concentration produced more monomers and a new hexamer envelope, ions expected from collisionally induced dissociation. At lower declustering voltage, the heptamer envelope was not clearly resolved. This means that even gentle nanospray ionization was too harsh for the heptamer, but does not explain the apparent dependance on protein concentration. However, the seventh subunit of GroES might be more easily removed during its transit to the mass spectrometer if it is less firmly held in the heptamer. Hunt et al. (1996) suggested that the GroES heptamer might "splay apart" in order to allow the GroES/GroEL complex to internalize unfolded proteins, and it is tempting to suggest that we are looking at a possible mechanism. However, deterioration of the sample, even over 24 h, could produce the same fragility, as we have noticed with other proteins (Donald et al. 2001).

The GroES differed in another significant way from GroEL, because no extra mass was observed on any of the components. In the GroEL complex, the measured masses of the 14-mer and the 7-mer were both considerably larger than expected (2984 and 1177 Da, respectively), the excess being attributed to counterions or ligands in the central channel (Rostom and Robinson 1999). GroES heptamer has an open cap structure (Hunt et al. 1996), with no place for ligands or water to be trapped. Mass spectrometry measurements are therefore giving information about the actual structure of the proteins under study.

	Materials and methods

TOP Abstract Introduction Results Discussion Materials and methods References

 
Protein purification
Purification of the GroES protein was essentially as described by Kamireddi et al. (1997), except we used DE52 (Whatman) as the anion exchange matrix. Purity of fractions was determined by SDS-PAGE. Selected fractions were pooled and then concentrated using vacuum dialysis. For mass spectrometry, 1 mg of pure protein was transferred to a Centricon 50 (Amicon), containing 2 mL of 50 mM NH₄HCO₃, 0.1 mM DTT, and centrifuged at 6000g for 20 min at 4°C. More buffer was then added, and the centrifugation was repeated at least eight times. Concentration of the recovered protein (by subunit) was determined by measuring spectrophotometrically from the known extinction coefficient of 3440 M^–1cm^–1 at 280 nm (Viitanen et al. 1990). Appropriate dilutions of the protein were made in the same buffer. For the electospray measurements, nanospray capillaries (Protana type S) were cut to size, washed with 2 µL buffer, followed by 2 µL of sample, then loaded with 2 µL of sample.

Mass spectrometry analysis
All experiments used an electrospray ionization time-of-flight instrument constructed at the University of Manitoba (Verentchikov et al. 1994; Krutchinsky et al. 1998). Technical details of the instrument have been described elsewhere (Krutchinsky et al. 2000). For each sample, spectra were acquired in positive mode with declustering voltage from 50–200 V and 1 kV spray voltage, using SF₆ curtain gas at reduced flow. Calibration was done with the singly and doubly charged ions of substance P. TOFMA, an in-house software program, was used to acquire and analyze the spectra. The actual instrument sensitivity was in the order of 100–150 pmol, based on a 10-min acquisition of a 0.31-µM sample.

	Acknowledgments

 
The host strain, MZ1 with plasmid pRE-GroES, was graciously provided by Dr. Prasad Reddy, National Institute of Standards and Technology, Rockville, Maryland. Other members of the TOF laboratory have given much appreciated technical advice and help, especially Sasha Loboda, Vic Spicer, and Jim McNabb. We thank Hélène Perreault for her critical review of the manuscript and Ann MacGregor for advice on calculation of errors. The work was supported by Research Grants to H.W.D. and K.G.S. from the Natural Sciences and Engineering Research Council of Canada, and by grant GM 59240 to K.G.S. from the U.S. NIH.

	References

TOP Abstract Introduction Results Discussion Materials and methods References

 
Ayed, A., Krutchinsky, A.N., Ens, W., Standing, K.G., and Duckworth, H.W. 1998. Quantitative evaluation of protein-protein and ligand-protein equilibria of a large allosteric enzyme by electrospray ionization time-of-flight mass spectrometry. Rapid Commun. Mass Spectrom. 12: 339–344.[CrossRef][Medline]

Donald, L.J., Hosfield, D.J., Cuvelier, S.L., Ens, W., Standing, K.G., and Duckworth, H.W. 2001. Mass spectrometric study of the Escherichia coli repressor proteins, IclR and GclR, and their complexes with DNA. Protein Sci. 10: 1370–1380.[Abstract/Free Full Text]

Hernandez, H. and Robinson, C.V. 2001. Dynamic protein complexes: Insights from mass spectrometry. J. Biol. Chem. 276: 46685–46688.[Free Full Text]

Hunt, J.F., Weaver, A.J., Landry, S.J., Gierasch, L., and Deisenhofer, J. 1996. The crystal structure of the GroES co-chaperonin at 2.8 Å resolution. Nature 379: 37–45.[CrossRef][Medline]

Kamireddi, M., Eisenstein, E., and Reddy, P. 1997. Stable expression and rapid purification of Escherichia coli GroEL and GroES chaperonins. Protein Expr. Purif. 11: 47–52.[CrossRef][Medline]

Krutchinsky, A.N., Chernushevich, I.V., Spicer, V.L., Ens, W., and Standing, K.G. 1998. A collisional damping interface for an electrospray ionization time-of-flight mass spectrometer. J. Am. Soc. Mass Spectrom. 9: 469–579.

Krutchinsky, A.N., Ayed, A., Donald, L.J., Ens, W., Duckworth, H.W., and Standing, K.G. 2000. Studies of noncovalent complexes in an electrospray ionization/time-of-flight mass spectrometer. Meth. Mol. Biol. 146: 239–249.[Medline]

Pinske, M.W.H., van Duijn, E., Maier, C.S., and Heck, A.J.L. 2003. Dissociation and association of supramolecular protein complexes studied by mass spectrometry. In Proceedings of 51st ASMS Conference on Mass Spectrometry and Allied Topics, Montreal, Canada.

Rostom, A.R. and Robinson, C.V. 1999. Detection of the intact GroEL chaperonin assembly by mass spectrometry. J. Am. Chem. Soc. 121: 4718–4719.[CrossRef]

Systat. 2000. Systat 10 Statistics II. SPSS Inc., Chicago, IL.

Verentchikov, A.N., Ens, W., and Standing, K.G. 1994. Reflecting time-of-flight mass spectrometer with an electrospray ion source and orthogonal extraction. Anal. Chem. 66: 126–133.[Medline]

Viitanen, P.V., Lubben, T.H., Reed, J., Goloubinoff, P., O’Keefe, D.P., and Lorimer, G.H. 1990. Chaperonin-facilitated refolding of ribulosebiphosphate carboxylase and ATP hydrolysis by chaperonin 60 (groEL) are K⁺ dependent. Biochemistry 29: 5665–5671.[CrossRef][Medline]

Wilm, M. and Mann, M. 1996. Analytical properties of the nanoelectrospray ion source. Anal. Chem. 68: 1–8.[Medline]

Xu, Z., Horwich, A.L., and Sigler, P.B. 1997. The crystal structure of the asymmetric GroEL-GroES-(ADP)₇ chaperonin complex. Nature 388: 741–750.[CrossRef][Medline]

Zondlo, J., Fisher, K.E., Lin, Z., Ducote, K.R., and Eisenstein, E. 1995. Monomer-heptamer equilibrium of the Escherichia coli chaperonin GroES. Biochemistry 34: 10334–10339.[CrossRef][Medline]

CiteULike    Connotea    Del.icio.us    Digg    Reddit    Technorati    What's this?

This Article Abstract Full Text (PDF) All Versions of this Article: ps.041164305v1 14/5/1375    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Donald, L. J. Articles by Duckworth, H. W. Search for Related Content PubMed PubMed Citation Articles by Donald, L. J. Articles by Duckworth, H. W. Social Bookmarking                   What's this?