Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M504263200 on July 26, 2005

J. Biol. Chem., Vol. 280, Issue 39, 33419-33425, September 30, 2005
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow An addition or correction has been published
Right arrow All Versions of this Article:
280/39/33419    most recent
M504263200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kennaway, C. K.
Right arrow Articles by Keep, N. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kennaway, C. K.
Right arrow Articles by Keep, N. H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Dodecameric Structure of the Small Heat Shock Protein Acr1 from Mycobacterium tuberculosis*

Christopher K. Kennaway{ddagger}, Justin L. P. Benesch§, Ulrich Gohlke{ddagger}, Luchun Wang{ddagger}, Carol V. Robinson§, Elena V. Orlova{ddagger}, Helen R. Saibi{ddagger}, and Nicholas H. Keep{ddagger}1

From the {ddagger}School of Crystallography and Institute of Structural Molecular Biology, Birkbeck, University of London, Malet Street, London WC1E 7HX and the §Department of Chemistry, University of Cambridge, Lensfield Road, Cambridge CB2 1EW, United Kingdom

Received for publication, April 19, 2005 , and in revised form, July 26, 2005.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Small heat shock proteins are a ubiquitous and diverse family of stress proteins that have in common an {alpha}-crystallin domain. Mycobacterium tuberculosis has two small heat shock proteins, Acr1 ({alpha}-crystallin-related protein 1, or Hsp16.3/16-kDa antigen) and Acr2 (HrpA), both of which are highly expressed under different stress conditions. Small heat shock proteins form large oligomeric assemblies and are commonly polydisperse. Nanoelectrospray mass spectrometry showed that Acr2 formed a range of oligomers composed of dimers and tetramers, whereas Acr1 was a dodecamer. Electron microscopy of Acr2 showed a variety of particle sizes. Using three-dimensional analysis of negative stain electron microscope images, we have shown that Acr1 forms a tetrahedral assembly with 12 polypeptide chains. The atomic structure of a related {alpha}-crystallin domain dimer was docked into the density to build a molecular structure of the dodecameric Acr1 complex. Along with the differential regulation of these two proteins, the differences in their quaternary structures demonstrated here supports their distinct functional roles.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 
Tuberculosis is one of the major killer infective diseases and represents a threat in both underdeveloped and developed countries due both to increased drug resistance and to the high levels of TB occurrence in human immunodeficiency virus-infected individuals. About one-third of the world's population is infected with the bacterium, which can spend many years in a dormant state inside lung granulomas. Disease is often the result of the bacteria sequestered inside lung macrophages being activated when the immune system of the infected individual is weakened.

The {alpha}-crystallin/small heat shock protein (sHSP)2 family is ubiquitous throughout nature and carries out a general cellular protective role in preventing aggregation of denatured proteins and facilitating subsequent refolding by other chaperones (1, 2). This function is particularly important in Mycobacterium tuberculosis, which must be able to survive an inhospitable environment while sequestered within phagosomes of alveolar macrophages. sHSPs typically form large homo-oligomeric complexes and often exhibit a high degree of dynamic subunit exchange, which is thought to contribute to, or be a consequence of, their chaperone function (1, 3).

Acr13 (also known as Hsp16.3/HspX/16-kDa Antigen/Rv2031c) is a 16.3-kDa protein, one of two members of the sHSP family found in M. tuberculosis. Acr2 (also known as Heat-stress-induced Ribosome-associated Protein A/HrpA/Rv0251c), the second sHSP, has a mass of 17.8 kDa (4, 5). Both proteins, like all sHSPs, share a conserved central domain of ~90 amino acids called the {alpha}-crystallin domain and have divergent N- and C-terminal extensions. The sequence similarity between Acr1 and Acr2 is 43% overall and 55% in the crystallin domain. Free growing Mycobacteria (M. smegmatis and M. marinarium) have homologues of both Acr1 and Acr2 and a third Acr protein, Acr3, which is most similar to the single sHSP in M. leprae (6).

Acr1 is the most abundant protein in M. tuberculosis during its dormant, non-replicative phase but is not present under conditions of logarithmic growth (7). Deletion of the acr gene encoding Acr1 resulted in a strain that showed normal growth in culture but impaired growth in macrophages (8). Expression is controlled by the DosR transcription factor, regulated by histidine sensor kinases (9, 10). The dosR regulon induces Acr1 transcription under conditions of hypoxia and other stresses such as S-nitrosoglutathione and ethanol, but not under heat shock (11–14). Acr1 is an antigen frequently recognized in the sera of tuberculosis patients (5, 15).

The second sHSP present in M. tuberculosis, Acr2, is the gene most strongly up-regulated by heat shock. It is under negative control by the hspR heat shock regulator (16) and is further positively regulated by the alternative sigma factors {sigma}H and {sigma}E, which respond both to heat and oxidative stress (17). Acr2 is also the most up-regulated gene on uptake of M. tuberculosis by macrophages whether or not the macrophages are stimulated by interferon (18). Acr2 deletion mutants have unaffected growth in macrophages (unlike Acr1 mutants) but show reduced pathogenicity in mouse models of infection (6). Recently, Acr2 has been shown to be a major target of humoral and T cell immunity during early phases of infection in humans (19) and as such could be important in developing improved vaccines.

The sHSPs studied so far show a range of quaternary structures and symmetries. In the archaeon Methanococcus jannaschii the sHSP is formed from 24 single chains in an octahedral (4,3,2) point group (20). sHSP 16.9 from wheat (Triticum aestivum) is a 12-mer formed from dimers in a D3 (3,2) point group, where one half of the dimer binds to the next dimer in the 3-fold ring and the other to the next dimer across the 2-fold, causing the C termini to take different paths (21). Yeast is a 24-mer formed from dimers in a tetrahedral (2,3) point group (22). The basic structural unit of sHSPs is a dimer formed by interactions between the conserved {alpha}-crystallin domains. The one exception to this in the literature has been M. tuberculosis Acr1, which has been reported to be a nonamer in solution, formed from a trimer of trimers (23, 24) and as such at variance with the existing pattern of sHSPs based on dimers.

sHSP sequences vary more in length and sequence N- and C-terminal to the {alpha}-crystallin domain. In the two crystal structures of sHSPs (20, 21), the C-terminal residues form elongated chains that bind onto adjacent dimers. The N-terminal regions are in the center of the complexes and are poorly ordered in the M. jannaschii and in half of the subunits in the wheat crystal structure. Deletion studies in a number of species show that removal of the N termini alters the oligomeric assembly and abolishes the chaperone activity (25, 26). Deletion of the N terminus of Acr1 showed a similar change in oligomeric state and loss of chaperone function, whereas deletion of the C-terminal sequence of Acr1 generated dimers but did not remove the chaperone activity (27).

Using mass spectrometry and electron microscopy (EM), we demonstrate a wide range of sizes and morphologies of Acr2. In contrast, these methods show that M. tuberculosis Acr1 is a dodecamer, and we present a 16 Å resolution three-dimensional structure with tetrahedral symmetry determined by negative stain EM and single particle analysis. A model based on the x-ray structure of the wheat {alpha}-crystallin domains was computationally docked into the EM map to give a molecular structure of the Acr1 assembly.


   EXPERIMENTAL PROCEDURES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
RESULTS
 DISCUSSION
 REFERENCES
 
Cloning, Expression, and Purification—The gene encoding M. tuberculosis Acr1, acr (Rv2031c), was PCR amplified from M. tuberculosis genomic DNA using a Pfu proofreading polymerase, cloned into TOPOblunt vector (Invitrogen), and then verified by automated DNA sequencing. The gene insert was then subcloned into a modified pET28a plasmid that lacks the His tag sequence using NdeI and BamHI restrictions sites, which left 3 additional residues, Met-Ser-Ala preceding the N-terminal Met of the acr gene. Soluble protein was expressed in Escherichia coli BL21 (DE3) Star (Novagen) at 37 °C, with a 4-hour induction with 1 mM isopropyl-1-thio-{beta}-D-galactopyranoside. Harvested cells were resuspended in 50 mM Tris-Cl (pH 7) buffer containing protease inhibitors (Merck). Cells were lysed in a French press, centrifuged at 48,000 x g, and the cleared lysate was loaded onto a Q-Sepharose anion exchange column (Amersham Biosciences). Elution was carried out using a linear NaCl gradient with an Åkta prime system (Amersham Biosciences), and fractions were checked for protein content by SDS-PAGE. Following centrifugal concentration (10-kDa cutoff; Amicon), the protein was further purified by gel filtration using a Sephacryl S300 column (Amersham Biosciences).

Acr2 (Rv0251c) was cloned using NdeI and BamH1 into a pET15b plasmid (Novagen) so that Acr2 was preceded by an N-terminal His6 tag. Protein was expressed in E. coli (strain BL21 (DE3) Rosetta) at 37 °C without induction. Purification was by nickel affinity chromatography (Amersham Biosciences) and then gel filtration (as above). Purity was checked by SDS-PAGE. His-tagged Acr2 protein was used in this study as this was our initial construct. In subsequent experiments, untagged protein also showed a similar spread of sizes in size exclusion chromatography and mass spectrometry (data not shown).

Protein Size Analysis—The protein was loaded onto a Superdex 200HR 10/30 size-exclusion column (Amersham Biosciences) and eluted at 0.4 ml/min with 200 mM ammonium acetate, 0.5 mM dithiothreitol, at 8 °C. The column was calibrated with gel filtration protein markers (Sigma). Fractions corresponding to the peak elution volume were pooled and concentrated to 1 mg/ml using a Biomax centrifugal filtration device (Millipore, Watford, UK), ready for mass spectrometry analysis.



View larger version (13K):
[in this window]
[in a new window]
  		FIGURE 1.
Size exclusion chromatography of Acr proteins. A, Acr2 (+His) shows a range of sizes, with the main peak at ~300 kDa and a secondary peak on the left corresponding to the void volume. B, Acr1 elutes as a single peak in size exclusion chromatography that is slightly skewed toward high mass, at a molecular mass of ~210 kDa. Molecular mass markers (in kDa) are indicated by arrows.

 
Nanoelectrospray mass spectrometry experiments were performed on a Q-ToF 2 instrument (Waters Micromass, Manchester, UK) that has been modified for high mass operation (28). Conditions were chosen to allow the ionization and detection of the protein complex without disrupting the non-covalent interactions that maintain the quaternary structure. Nanoelectrospray ionization capillaries were prepared in-house from borosilicate glass tubes of 1-mm outer diameter and 0.78-mm inner diameter (Harvard Apparatus, Holliston, MA) using a Flaming-Brown P-97 micropipette puller (Sutter Instruments, Hercules, CA) and gold-coated using an S.E. sputter-coater (Polaron, Newhaven, UK). The capillary tips were cut under a stereomicroscope to give inner diameters of 1–5 µm, and 1 µl of solution was loaded for sampling. The following experimental parameters were used (positive ion mode): capillary voltage 1.7 kV, cone gas 150 liters/h, sample cone 200 V, extractor cone 10 V, collision energy 10 V, ion transfer stage pressure 7.5 x 10-3 mbar, quadrupole analyser pressure 1.2 x 10-3 mbar, time-of-flight analyser pressure 1.6 x 10-6 mbar, with 3.5 x 10-2 mbar of argon in the collision cell. External calibration of the spectra was achieved using a solution of caesium iodide. Data were acquired and processed with MassLynx software (Waters Micromass, Manchester, UK), and are displayed here without background subtraction and with minimal smoothing.

Dynamic light scattering was done at room temperature with a Dynapro-801 machine (Protein Solutions Inc.) on serial dilutions of Acr1, starting at 1.0 mg/ml protein in 20 mM Tris (pH 7), 200 mM NaCl. The sample was filtered through a 0.2-µm filter before loading. The diffusion coefficient was calculated assuming an aqueous buffer containing 1% salt (viscosity 1.020), and the molecular mass was calculated by standard curve of protein hydrodynamic radii versus molecular weight given in the Dynapro manual (Protein Solutions Inc.). Molecular mass was also calculated by volume hydration, with an average specific volume of 0.7344 and a frictional ratio of 1.3324.



View larger version (18K):
[in this window]
[in a new window]
  		FIGURE 2.
A, Acr2 (+His) nanoelectrospray mass spectrometry spectrum showing a range of different sized complexes. Analysis of the individual peaks by argon collision-induced dissociation reveals homo-oligomers consisting of even numbers of subunits from 12 up to 28, as well as other larger assemblies. B, nanoelectrospray mass spectrometry of Acr1 reveals the protein to exist as a dodecamer. The individual charge states of the peaks are labeled.

 
Electron Microscopy—Acr1 was negatively stained on carbon-coated grids that were glow discharged in air immediately before the application of 5 µl of Acr1 solution (2–10 µg/ml protein in 50 mM Tris, pH 7). Excess protein solution was blotted off after 2 min, immediately followed by two rounds of adding and then blotting off 5-µl aliquots of 2% uranyl acetate.

Cryo-EM samples of Acr2 (+His) protein at 0.25 mg/ml were prepared on holey carbon-coated copper grids made according to the method of Lepault and Dubochet (29) and glow discharged for 30 s before sample application. Samples were vitrified in a humidified atmosphere by plunging into liquid ethane cooled to -160 °C.

Negative stain grids were imaged in a Tecnai T12 electron microscope (FEI, Eindhoven, The Netherlands) at x42,000 magnification and 120 kV. Micrographs were recorded on Kodak SO163 film in low dose mode with a defocus of ~0.4–0.5 µm. 18 micrographs selected by optical diffraction were digitized at 14 µm using a Zeiss Photoscan TD linear array scanner, giving 3.33 Å/pixel.

Cryoelectron microscopy was carried out with a Tecnai T20 microscope equipped with a 200 kV Field Emission Gun and a nitrogen-cooled cryo stage. Low dose micrographs were taken at a magnification of x40000 and a defocus between 1.2 and 2.5 µm. 30 micrographs selected by optical diffraction were digitized at 7 µm, giving 1.75 Å/pixel.



View larger version (109K):
[in this window]
[in a new window]
  		FIGURE 3.
Cryoelectron microscopy of Acr2. A, negative stain micrograph of Acr2, showing similar shapes to the cryo images processed. B, a selection of typical cryo-EM class sum images of Acr2 (~20 images/class). The range of sizes and morphologies present in the Acr2 sample is clearly visible. Particles are arranged in three sizes, small (~95–105 Å diameter, top row), medium (~105–120 Å diameter, middle row), and large (~115–130 Å diameter, bottom row). C, eigenimages (statistical difference images) of the centered Acr2 particles. The rings present in the first two images are indicative of size variations, whereas the third image is typical of poorly centered particles. The presence of 2-, 3-, and 4-fold symmetry is suggested by the last three eigenimages. White 100 Å scale bars are included in all panels

 
Image Processing—CTFFind3 software (30) was used to determine the defocus and check astigmatism. Well separated particles were manually selected using Ximdisp (31) and were cut out into boxes using the MRC program Label (32). The rest of the image processing and three-dimensional reconstruction was done using the IMAGIC suite (33).

For Acr2 17,231 cryo-EM particle images were picked from 30 negatives and then corrected for contrast transfer function by phase flipping. The images were band-pass filtered between (120 Å)-1 and (5 Å)-1. Because of size variation, the images were aligned to six model discs of diameters spanning the range of particle sizes to improve centering (22). Reconstruction using common lines was attempted, primarily on a subset of small particles, but heterogeneity prevented any consistent three-dimensional maps from being produced.

A total of 6,021 negatively stained Acr1 single particles selected from 10 micrographs were boxed. Images were normalized and band-pass filtered between (67 Å)-1 and (4 Å)-1. Particles were then masked by applying a soft-edged circular mask. Centering was done by translational alignment to the rotationally averaged sum of all images, iterated three times. Multivariate statistical analysis was used to classify the images.

The orientations of class averages were determined by angular reconstitution using common lines (34). Three-dimensional reconstructions were calculated by exact filtered back projection (35). Of the various point groups tried, only tetrahedral symmetry (2,3) gave a three-dimensional reconstruction whose reprojections were compatible with the data. Several rounds of refinement with sets of reprojections, alignment, multivariate statistical analysis classification, and then angular reconstitution followed by three-dimensional reconstruction led to an improved map. The final three-dimensional map was generated by 400 of 600 input class averages with an average of 10 images/class. A three-dimensional mask was applied to the map. The resolution was determined using the criterion of 0.5 Fourier shell correlation. Fitting the coordinates of wheat Hsp16.9 (Protein Data Bank code 1GME [PDB] ) into the density was done with URO (36), using one dimer of the Hsp16.9 {alpha}-crystallin domain (residues 43–137 and the C-terminal residues 146–151 of the adjacent subunit bound to them), applying appropriate symmetry operators.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
DISCUSSION
 REFERENCES
 
Molecular Mass Determination of Acr1 and Acr2 Complexes—Expression of recombinant Acr1 and Acr2 yielded ~40 mg of soluble protein/liter of culture. Following purification ~95% pure protein was observed by SDS-PAGE (not shown).



View larger version (121K):
[in this window]
[in a new window]
  		FIGURE 4.
Three-dimensional reconstruction of Acr1 from negative stain electron microscopy. A, raw negative stain micrograph showing individual particles of Acr1 protein. B, input class averages (~10 images/class). C, their matching reprojections of the final three-dimensional map. White 100 Å scale bars are included in all panels.

 



View larger version (31K):
[in this window]
[in a new window]
  		FIGURE 5.
Sequence alignment of five sHSPs: M. tuberculosis Acr1 and Acr2, Hsp16.5 from M. jannaschii (MetJ), human {alpha}A crystallin (Hu {alpha}-C), and Hsp16.9 from T. aestivum (Wheat). White on black indicates fully conserved residues, dark gray highly conserved residues, and light gray partially conserved residues. The secondary structure of the wheat protein (S, {beta}-sheet; H, {alpha}-helix) is marked below. The {alpha}-crystallin domain (ACD) is marked with plus (+) symbols. Hash (#) symbols denote the 6 residues containing the C-terminal IXI motif also included in the fitting. XX above the alignment denotes the 4 fewer residues in Acr1 in the region that protrudes from the EM map using wheat Hsp 16.9 as the model.

 
Size exclusion chromatography of Acr2 shows the protein exists as a range of assemblies, with the main peak at ~300 kDa (Fig. 1A). The left peak corresponds to the void volume of the column, indicating that some of the protein is present as very large (>1 MDa) aggregates. Dynamic light scattering of Acr2 showed the protein was polydisperse (data not shown).

Size exclusion chromatography of Acr1 gave a single elution peak at ~210 kDa (Fig. 1B), in agreement with previous reports (23). A slight asymmetry at the base of the curve indicates the presence of a small amount of larger species. Despite this the main peak is quite sharp, indicating a relatively monodisperse distribution.

Dynamic light scattering showed that Acr1 is monodisperse, with a polydispersity figure of 15.1% (data not shown). The diffusion coefficient of 434 x 109 cm s-2 (corresponding to a hydrodynamic radius of 5.39 nm) gives a molecular mass of 185 kDa, calculated using a standard curve of proteins. Inaccuracies in the calculation of molecular masses are likely because of variations within the standard curve. An alternative method of calculating molecular mass using volume hydration gave a mass of 210 kDa for Acr1. These figures do not agree with the earlier reported mass for Acr1 of 147 kDa (23) but are closer to the expected mass of 197 kDa for a 12-mer.

The mass spectrum of Acr2 is broad and complex (Fig. 2A), showing a wide distribution of homo-oligomeric assemblies. Further analysis of the spectrum by argon collision-induced dissociation showed that all the peaks were formed by oligomers with an even number of subunits. The peaks corresponding to 16-, 20-, 24-, and 28-mers were most intense. Therefore, Acr2 like other sHSPs is built from dimers, and a further assembly into a tetramer is probably a building block of the larger oligomers. In contrast the nanoelectrospray mass spectrum of Acr1 shows distinct peaks in the range of 6,000 to 7,000 m/z (Fig. 2B). Analysis of the peaks reveals a single species that has a mass of 196.628 ± 0.022 kDa, corresponding to a dodecamer of Acr1.

Characterization of Acr2 by Cryoelectron Microscopy—An area of a negative stain EM image of Acr2 is shown in Fig. 3A. Cryo-EM images were similar but of lower contrast and were used for image analysis. Multivariate statistical analysis of a data set of cryoimages of Acr2 gave rise to the eigenimages shown in Fig. 3C. These representations of the principal differences present in the whole set of images show rings (Fig. 3C, numbers 1 and 2) that are typical of images with significant size variations (22). Features relating to particle symmetry can also be seen in some of the other eigenimages. The class averages after size separation shown in Fig. 3B demonstrate the diverse array of particle sizes and morphologies present in the Acr2 sample, with particle diameters ranging from ~95 to ~130 Å. Many particles appear to be hollow. Each group has distinctive features, with hints of 2- or 4-fold symmetry in the smallest particles and 3-fold in the middle group. The quality of the class averages indicates that some of the complexes are relatively well ordered, unlike the polydisperse mammalian {alpha}-crystallins (37). Attempts to generate three-dimensional structures from the Acr2 cryo-EM data (after sorting by size) did not produce any consistent maps, most likely because of the small size of the particles, the lack of well defined symmetries, and the difficulty in assigning class averages to particular oligomeric states.



View larger version (44K):
[in this window]
[in a new window]
  		FIGURE 6.
A, surface representation of the three-dimensional map of Acr1 contoured at 2.2 {sigma}, viewed down a 2-fold axis. The black line beneath is a 100 Å scale bar. B, model of Acr1 using the wheat {alpha}-crystallin dimers. N-terminal residues 1–42 are omitted because of their unknown structure and likely flexibility. C-terminal residues 146–151 containing the IXI motif are shown in magenta, positioned as they are in the wheat crystal structure. These 6 residues occupy bulges in the density along the outer edges, seen most clearly at the top and bottom of this view. C, contacts between dimers are formed by the C-terminal extensions. Residues 146–151 (magenta sticks) can be seen binding to the edges of {beta}-sheets 3 and 7 of an {alpha}-crystallin domain in an adjacent dimer (blue). The surface of the Acr1 EM reconstruction is shown, and the truncated end (residue 137) of the green {alpha}-crystallin domain is shown colored red. The general direction of the path of the omitted residues is shown as a yellow dashed line. Figures were produced with Pymol (www.pymol.org).

 
Three-dimensional Reconstruction of Acr1—Acr1 protein was studied by negative stain electron microscopy because a compact structure of 200 kDa is difficult to analyze by cryo-EM. As can be seen in Fig. 4A, the micrographs show distinct particles with a homogeneous size distribution. The protein complexes are approximately ~100 Å in diameter, with many views appearing triangular in shape. Our data are consistent with images from an earlier EM study of ice-embedded Acr1 (23). Statistical analysis showed strong 2- and 3-fold features in the eigenimages (not shown). The point groups that could be consistent with the data and are compatible with an assembly of 12 subunits are C1 (no symmetry), C2, C3, D2, D3, and 2,3 (tetrahedral) point groups. These were all tried for initial Euler angle assignment, but only the tetrahedral (2,3) point group gave coherent results in which the reprojections matched the input classes. Fig. 4, B and C, shows input class averages and their corresponding reprojections from the final three-dimensional reconstruction. The resolution was determined as 16 Å by the criterion of 0.5 Fourier shell correlation.

Alignment of Acr1 with the wheat and M. jannaschii sHSP sequences shows little difference in the length of the sheets and turns within the ~100-residue {alpha}-crystallin domain (Fig. 5), giving a reliable model to use for fitting. A surface representation of the Acr1 three-dimensional reconstruction is shown in Fig. 6A. The main regions of mass lie along each edge of a tetrahedron, and there is a small amount of diffuse density in the center. Truncated dimers of the wheat sHSP crystal structure (21) containing the {alpha}-crystallin domain were docked into the density. Automated fitting was done using the program URO (36), taking into account the tetrahedral symmetry. The result of the fitting was independent of the starting position of the dimer. The fitted model fills the density along the edges of the tetrahedron, consistent with the size and shape of the {alpha}-crystallin domains, leaving unfilled density at the corners and in the center (Fig. 6B). The only loop of the {alpha}-crystallin domain that is partially outside the density, between strands 4 and 6 (Fig. 6B), is 4 residues shorter in Acr1 (Fig. 5) and so is likely to be accommodated within the EM map. Although the differences between the two possible maps of opposite hand are small, URO gave a better fit (in terms of correlation values (81.5 versus 78.2%) and visual inspection) into the map with the handedness that is shown in Fig. 6A.

As the N-terminal regions and C-terminal tails are divergent in the M. jannaschii and wheat structures, these were omitted from the fitting, except for the conserved IXI motifs in the C-terminal extension. This motif binds to hydrophobic regions at the ends of sheets 3 and 7 and is important in forming interdimer contacts (20, 21). When this motif (residues 146–151 of the wheat protein) is retained in the model, the fitting into the density improved (correlation coefficients in URO 81.5% compared with 78.5% without the C-terminal peptide). This can be seen in Fig. 6B where these extra residues fill two bulges on the outer surface of the dimers.

In the three different monomers from the two crystal structures of sHSPs, the linkers to the C termini extend out in roughly the same direction, that is, to the left when looking at the external face of a monomer. We propose that this is also the case in Acr1, as the gap between residues 137 and 146 in the model is about 30 Å in this direction (clockwise looking down the 3-fold) as opposed to ~45 Å in the other direction. The linker to the C termini may explain at least some of the residual density near the 3-fold, but we cannot state the exact path (Fig. 6C).

On the basis of the crystal structures of sHSPs and experimental data on Acr1 (37) the N-terminal regions are likely to be pointing inwards, and Fu et al. (27) report that they form intersubunit contacts that help stabilize the quaternary structure of Acr1. The diffuse density present in the internal cavity could accommodate these regions, which are likely to be flexible or only partially ordered as seen in the wheat and archaeal structures.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
REFERENCES
 
Nanoelectrospray mass spectrometry was used to reveal the stoichiometries of the two sHSP oligomers present in M. tuberculosis. The contrast between the spectra of Acr1 and Acr2 is a clear reflection of their different molecular organizations, and this is further characterized in the electron microscopy studies.

Acr2 exists as a wide variety of homo-oligomeric complexes in solution. The polydisperse nature of this protein, visible in size exclusion chromatography, dynamic light scattering, mass spectrometry, and the cryo-EM images, thwarted three-dimensional reconstruction despite efforts to sort images into homogeneous subsets. Nevertheless, the evidence for 2-, 3-, and 4-fold symmetries suggests that some ordered assemblies might be present in the population of Acr2 species. The mass spectrometry data obtained on Acr2 show that these oligomers are made up from dimeric and tetrameric building blocks.

Acr1 exhibits quite different behavior from its sister protein in terms of quaternary structure. We find that Acr1 is a dodecameric assembly formed from a tetrahedral arrangement of monomers. This can be thought of as six dimers on the edges of a tetrahedron. The particle diameter and features are similar to those seen in the cryo-EM images from the previous report on Acr1 (23), indicating that we are looking at the same particle and that no significant shrinkage of the protein has occurred in the staining and drying process. Tetrahedral symmetry has precedents in bacterial proteins: two archaeal aminopeptidases have been shown by x-ray crystallography and electron microscopy to be tetrahedra similar in subunit arrangement to our proposed Acr1 structure (38, 39), and the related sHSP from yeast is a tetrahedral 24-mer (22). The fitted assembly shows features that are consistent with the known crystal structures and biochemical properties of sHSPs.

The common basic unit of all sHSPs studied so far is a dimer (3, 40, 41), other than in previous reports on Acr1 (23, 24). The large surface area of the dimer interface and the swapping of loops in the {alpha}-crystallin domains also suggest that the dimer is a relatively stable unit. Our data show Acr1 to be formed from the same dimer building block as other sHSPs.

The C-terminal extension of Acr1 contains the conserved IXI motif important for the binding to the hydrophobic groove on {beta}-sheets 3 and 7 of neighboring subunits. The unfilled density along the external side of each subunit when fitting with the {alpha}-crystallin dimer alone can be accounted for by including the 6 C-terminal residues containing the IXI motif. The intervening residues may explain the density near the 3-fold axis. Fu et al. (27) show that deletion of the C-terminal residues of Acr1 yields dimers that fully retain their chaperone function. This supports the primary role of the C-terminal as making intersubunit contacts between dimers, thereby linking together the larger assemblies, and agrees with our proposed assembly.

We propose that the N-terminal residues not present in our model occupy the central region of the complex and are only partially ordered. Acr1 is able to form a mixed oligomer of wild type and N-terminal-truncated protein, suggesting that this region is disordered or flexible (27). Removal of the first 36 residues of Acr1 ablates both assembly and chaperone function. The role of the N-terminal region in the chaperone mechanism is compatible with our structure because the active dimeric form can be released by dynamic exchange (42). The data are consistent with a model of sHSP function in which the large assemblies act as a storage form, providing a reservoir of chaperones that can be quickly brought into service. The sequestering of the N termini within complexes may also prevent unwanted binding to hydrophobic regions on native proteins under normal conditions.

It is clear from the very different expression patterns of the Acr1 and Acr2 genes that M. tuberculosis has two sHSP genes so that they can be expressed in response to different stimuli. In this regard it is striking that Acr1 and Acr2 show distinct types of oligomeric assembly, corresponding in one case to a fixed stoichiometry and in the other to a variable stoichiometry despite the {alpha}-crystallin domains being very closely related and diverging after the evolution of the mycobacteria (6). Although it is possible that there is little evolutionary selection on the exact type of oligomeric assembly of the sHSPs and that they merely act as a sequestered source of dimers, it seems more likely that the significant difference between the assembly of the two closely related sHSPs relates to a functional advantage, although we have no data to back up this assertion. It is clear from several studies by other groups that Acr1 is an active sHSP against a number of substrates (23, 27, 43). There is less published on Acr2, but activity against a limited number of model substrates has been shown (4). The fact that the two proteins are expressed in complementary, rather than interacting, conditions suggests that they do not act synergistically, in contrast to a recent report on the sHSPs IbpA and B from E. coli (44).

We have shown that the two closely related sHSP proteins from M. tuberculosis have very different oligomeric assemblies. The Acr2 oligomers are heterogeneous and are assembled predominantly from dimers and tetramers, whereas Acr1 forms a single tetrahedral dodecameric assembly assembled from dimers, making it consistent with the known properties of the rest of the sHSP family. The structural information we describe here enables further understanding of these two proteins that play a significant role in mycobacterial survival and the host immune response to infection.


   FOOTNOTES
 
The atomic coordinates and structure factors (code 2BYU) have been deposited in the Protein Data Bank, Research Collaboratory for Structural Bioinformatics, Rutgers University, New Brunswick, NJ (http://www.rcsb.org/).

* This work was supported by a Medical Research Council UK studentship (to C. K. K.) and a component of European Union Grant QLK2-CT-2001-02018 (to N. H. K.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

1 To whom correspondence should be addressed. Tel.: 020-7631-6852; Fax: 020-7631-6803; E-mail: n.keep{at}mail.cryst.bbk.ac.uk.

2 The abbreviations used are: sHSP, small heat shock protein; EM, electron microscopy. Back

3 The wheat sHSP coordinates as fitted to the Acr1 EM map are available through PDB number 2BYU [PDB] . The electron density map for Acr1 is available through the electron microscopy data bank (www.ebi.ac.uk.msd-srv/emsearch/index.html) through EMDB number 1149. Back


   ACKNOWLEDGMENTS
 
We thank Drs. Brian Robertson and Graham Stewart for the gift of plasmids for Acr2 and helpful discussions on tuberculosis biology, Drs. Christine Slingsby and Robin Stamler for helpful discussions about small heat shock proteins, and Drs. David Houldershaw and Richard Westlake for computer support.



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 EXPERIMENTAL PROCEDURES
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Van Montfort, R., Slingsby, C., and Vierling, E. (2001) Adv. Protein Chem. 59, 105-156[Medline] [Order article via Infotrieve]
  2. Narberhaus, F. (2002) Microbiol. Mol. Biol. Rev. 66, 64-93[Abstract/Free Full Text]
  3. Sobott, F., Benesch, J. L., Vierling, E., and Robinson, C. V. (2002) J. Biol. Chem. 277, 38921-38929[Abstract/Free Full Text]
  4. Tabira, Y., Ohara, N., and Yamada, T. (2000) Microb. Pathog. 29, 213-222[CrossRef][Medline] [Order article via Infotrieve]
  5. Verbon, A., Hartskeerl, R. A., Schuitema, A., Kolk, A. H., Young, D. B., and Lathigra, R. (1992) J. Bacteriol. 174, 1352-1359[Abstract/Free Full Text]
  6. Stewart, G. R., Newton, S. M., Wilkinson, K. A., Humphreys, I. R., Murphy, H. N., Robertson, B. D., Wilkinson, R. J., and Young, D. B. (2005) Mol. Microbiol. 55, 1127-1137[CrossRef][Medline] [Order article via Infotrieve]
  7. Yuan, Y., Crane, D. D., and Barry, C. E., III (1996) J. Bacteriol. 178, 4484-4492[Abstract/Free Full Text]
  8. Yuan, Y., Crane, D. D., Simpson, R. M., Zhu, Y. Q., Hickey, M. J., Sherman, D. R., and Barry, C. E. (1998) Proc. Natl Acad. Sci. (U. S. A.) 95, 9578-9583[Abstract/Free Full Text]
  9. Roberts, D. M., Liao, R. P., Wisedchaisri, G., Hol, W. G., and Sherman, D. R. (2004) J. Biol. Chem. 279, 23082-23087[Abstract/Free Full Text]
  10. Saini, D. K., Malhotra, V., Dey, D., Pant, N., Das, T. K., and Tyagi, J. S. (2004) Microbiology (Read.) 150, 865-875[Abstract/Free Full Text]
  11. Kendall, S. L., Movahedzadeh, F., Rison, S. C., Wernisch, L., Parish, T., Duncan, K., Betts, J. C., and Stoker, N. G. (2004) Tuberculosis (Edinb.) 84, 247-255[Medline] [Order article via Infotrieve]
  12. Florczyk, M. A., McCue, L. A., Purkayastha, A., Currenti, E., Wolin, M. J., and McDonough, K. A. (2003) Infect. Immun. 71, 5332-5343[Abstract/Free Full Text]
  13. Park, H. D., Guinn, K. M., Harrell, M. I., Liao, R., Voskuil, M. I., Tompa, M., Schoolnik, G. K., and Sherman, D. R. (2003) Mol. Microbiol. 48, 833-843[CrossRef][Medline] [Order article via Infotrieve]
  14. Sherman, D. R., Voskuil, M., Schnappinger, D., Liao, R., Harrell, M. I., and Schoolnik, G. K. (2001) Proc. Natl. Acad. Sci. (U. S. A.). 98, 7534-7539[Abstract/Free Full Text]
  15. Lee, B.Y., Hefta, S. A., and Brennan, P. J. (1992) Infect. Immun. 60, 2066-2074[Abstract/Free Full Text]
  16. Stewart, G. R., Wernisch, L., Stabler, R., Mangan, J. A., Hinds, J., Laing, K. G., Young, D. B., and Butcher, P. D. (2002) Microbiology 148, 3129-3138[Abstract/Free Full Text]
  17. Manganelli, R., Voskuil, M. I., Schoolnik, G. K., and Smith, I. (2001) Mol. Microbiol. 41, 423-437[CrossRef][Medline] [Order article via Infotrieve]
  18. Schnappinger, D., Ehrt, S., Voskuil, M. I., Liu, Y., Mangan, J. A., Monahan, I. M., Dolganov, G., Efron, B., Butcher, P. D., Nathan, C., and Schoolnik, G. K. (2003) J. Exp. Med. 198, 693-704[Abstract/Free Full Text]
  19. Wilkinson, K. A., Stewart, G. R., Newton, S. M., Vordermeier, M. H., Wain, J. R., Murphy, H. N., Horner, K., Young, D. B., and Wilkinson, R. J. (2005) J. Immunol. 174, 4237-4243[Abstract/Free Full Text]
  20. Kim, K. K., Kim, R., and Kim, S. H. (1998) Nature 394, 595-599[CrossRef][Medline] [Order article via Infotrieve]
  21. van Montfort, R. L., Basha, E., Friedrich, K. L., Slingsby, C., and Vierling, E. (2001) Nat. Struct. Biol. 8, 1025-1030[CrossRef][Medline] [Order article via Infotrieve]
  22. White, H. E., Saibil, H. R., Ignatiou, A., and Orlova, E. V. (2004) J. Mol. Biol. 336, 453-460[CrossRef][Medline] [Order article via Infotrieve]
  23. Chang, Z., Primm, T. P., Jakana, J., Lee, I. H., Serysheva, I., Chiu, W., Gilbert, H. F., and Quiocho, F. A. (1996) J. Biol. Chem. 271, 7218-7223[Abstract/Free Full Text]
  24. Abulimiti, A., Fu, X., Gu, L., Feng, X., and Chang, Z. (2003) J. Mol. Biol. 326, 1013-1023[CrossRef][Medline] [Order article via Infotrieve]
  25. Stromer, T., Fischer, E., Richter, K., Haslbeck, M., and Buchner, J. (2004) J. Biol. Chem. 279, 11222-11228[Abstract/Free Full Text]
  26. Studer, S., Obrist, M., Lentze, N., and Narberhaus, F. (2002) Eur. J. Biochem. 69, 3578-3586[CrossRef]
  27. Fu, X., Zhang, H., Zhang, X., Cao, Y., Jiao, W., Liu, C., Song, Y., Abulimiti, A., and Chang, Z. (2005) J. Biol. Chem. 280, 6337-6348[Abstract/Free Full Text]
  28. Sobott, F., Hernandez, H., McCammon, M. G., Tito, M. A., and Robinson, C. V. (2002) Anal. Chem. 74, 1402-1407[Medline] [Order article via Infotrieve]
  29. Lepault, J., and Dubochet, J. (1986) Methods Enzymol. 127, 719-730[Medline] [Order article via Infotrieve]
  30. Mindell, J. A., and Grigorieff, N. (2003) J. Struct. Biol. 142, 334-347[CrossRef][Medline] [Order article via Infotrieve]
  31. Smith, J. M. (1999) J. Struct. Biol. 125, 223-228[CrossRef][Medline] [Order article via Infotrieve]
  32. Crowther, R. A., Henderson, R., and Smith, J. M. (1996) J. Struct. Biol. 116, 9-16[CrossRef][Medline] [Order article via Infotrieve]
  33. van Heel, M., Harauz, G., Orlova, E. V., Schmidt, R., and Schatz, M. (1996) J. Struct. Biol. 116, 17-24[CrossRef][Medline] [Order article via Infotrieve]
  34. van Heel, M., Gowen, B., Matadeen, R., Orlova, E. V., Finn, R., Pape, T., Cohen, D., Stark, H., Schmidt, R., Schatz, M., and Patwardhan, A. (2000) Q. Rev. Biophys. 33, 307-369[CrossRef][Medline] [Order article via Infotrieve]
  35. Haruaz, G., and van Heel, M. (1986) Optik 73, 146-156
  36. Navaza, J., Lepault, J., Rey, F. A., Alvarez-Rua, C., and Borge, J. (2002) Acta Crystallogr. Sect. D Biol. Crystallogr. 58, 1820-1825[CrossRef][Medline] [Order article via Infotrieve]
  37. Haley, D. A., Bova, M. P., Huang, Q. L., McHaourab, H. S., and Stewart, P. L. (2000) J. Mol. Biol. 298, 261-272[CrossRef][Medline] [Order article via Infotrieve]
  38. Russo, S., and Baumann, U. (2004) J. Biol. Chem. 279, 51275-55128[Abstract/Free Full Text]
  39. Franzetti, B., Schoehn, G., Hernandez, J. F., Jaquinod, M., Ruigrok, R. W., and Zaccai, G. (2002) EMBO J. 21, 2132-2138[CrossRef][Medline] [Order article via Infotrieve]
  40. Wintrode, P. L., Friedrich, K. L., Vierling, E., Smith, J. B., and Smith, D. L. (2003) Biochemistry 42, 10667-10673[CrossRef][Medline] [Order article via Infotrieve]
  41. Haslbeck, M., Walke, S., Stromer, T., Ehrnsperger, M., White, H. E., Chen, S. X., Saibil, H. R., and Buchner, J. (1999) EMBO J. 18, 6744-6751[CrossRef][Medline] [Order article via Infotrieve]
  42. Gu, L., Abulimiti, A., Li, W., and Chang, Z. (2002) J. Mol. Biol. 319, 517-526[CrossRef][Medline] [Order article via Infotrieve]
  43. Valdez, M. M., Clark, J. I., Wu, G. J. S., and Muchowski, P. J. (2002) Eur. J. Biochem.269, 1806-1813[Medline] [Order article via Infotrieve]
  44. Matuszewska, M., Kuczynska-Wisnik, D., Laskowska, E., and Liberek, K. (2005) J. Biol. Chem. 280, 12292-12298[Abstract/Free Full Text]

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


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
V. H. Hayes, G. Devlin, and R. A. Quinlan
Truncation of {alpha}B-Crystallin by the Myopathy-causing Q151X Mutation Significantly Destabilizes the Protein Leading to Aggregate Formation in Transfected Cells
J. Biol. Chem., April 18, 2008; 283(16): 10500 - 10512.
[Abstract] [Full Text] [PDF]

Home page
 Appl. Environ. Microbiol.Home page
M. Ventura, C. Canchaya, Z. Zhang, G. F. Fitzgerald, and D. van Sinderen
Molecular Characterization of hsp20, Encoding a Small Heat Shock Protein of Bifidobacterium breve UCC2003
Appl. Envir. Microbiol., July 15, 2007; 73(14): 4695 - 4703.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow An addition or correction has been published
Right arrow All Versions of this Article:
280/39/33419    most recent
M504263200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Kennaway, C. K.
Right arrow Articles by Keep, N. H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Kennaway, C. K.
Right arrow Articles by Keep, N. H.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2005 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement