Mycoplasma pneumoniae Cytoskeletal Protein HMW2 and the ...

6 downloads 116 Views 3MB Size Report
Oct 22, 2008 - Department of Microbiology, University of Georgia, Athens, Georgia 30602. Received 22 October ..... Scale bar, 250 nm. VOL. 191, 2009 .... part in C1 was present at lower steady-state levels than the. HMW2 protein in I-2 or ...
JOURNAL OF BACTERIOLOGY, Nov. 2009, p. 6741–6748 0021-9193/09/$12.00 doi:10.1128/JB.01486-08 Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Vol. 191, No. 21

Mycoplasma pneumoniae Cytoskeletal Protein HMW2 and the Architecture of the Terminal Organelle䌤 Stephanie R. Bose,† Mitchell F. Balish,‡ and Duncan C. Krause* Department of Microbiology, University of Georgia, Athens, Georgia 30602 Received 22 October 2008/Accepted 24 August 2009

The terminal organelle of Mycoplasma pneumoniae mediates cytadherence and gliding motility and functions in cell division. The defining feature of this complex membrane-bound cell extension is an electron-dense core of two segmented rods oriented longitudinally and enlarging to form a bulb at the distal end. While the components of the core have not been comprehensively identified, previous evidence suggested that the cytoskeletal protein HMW2 forms parallel bundles oriented lengthwise to yield the major rod of the core. In the present study, we tested predictions emerging from that model by ultrastructural and immunoelectron microscopy analyses of cores from wild-type M. pneumoniae and mutants producing HMW2 derivatives. Antibodies specific for the N or C terminus of HMW2 labeled primarily peripheral to the core along its entire length. Furthermore, truncation of HMW2 did not correlate specifically with core length. However, mutant analysis correlated specific HMW2 domains with core assembly, and examination of core-enriched preparations confirmed that HMW2 was a major component of these fractions. Taken together, these findings yielded a revised model for HMW2 in terminal organelle architecture. due to a frameshift in the corresponding MPN310 open reading frame, which also encodes protein P28 at its 3⬘ end, in the same reading frame encoding HMW2 (6). Mutants C1 and H9 are similar to mutant I-2 but result from Tn4001 disruption of MPN310 (15, 23) (Fig. 1). The loss of HMW2 and the inability of these mutants to assemble a core are accompanied by an abnormal morphology, reduced levels of terminal organelle proteins HMW1, HMW3, P24, P28, P41, and P65, failure to localize the major adhesin P1 to the terminal organelle, and the loss of cytadherence (6, 20, 22, 31). Imprecise transposon excision from mutant C1 yielded excision revertant C1R1, having an in-frame deletion in MPN310 that truncates HMW2 and eliminates P28 (6) (Fig. 1). Analysis using immunofluorescence (40) or fluorescent protein fusions (3, 19) localizes HMW2 generally to the terminal organelle. Based on its localization, its requirement for core formation, and its deduced length relative to that of the core, we proposed previously that HMW2 is a major component of the electron-dense core and, with P28, may form bundles oriented longitudinally to yield the large rod of the core (3). In the present study, we explored further the role of HMW2 in core formation relative to the current model, by which (i) HMW2 is predicted to orient with its N- and C-terminal domains at the ends of the large rod of the core and (ii) mutants producing shorter HMW2 proteins are expected to have correspondingly shorter cores. We report here the successful localization of HMW2 by immunoelectron microscopy (immunoEM) and the ultrastructural analysis of electron-dense cores in several HMW2 truncation mutants, allowing us to correlate specific regions of HMW2 with normal core formation. Finally, we evaluated core enrichment following detergent and salt extractions, demonstrating that HMW2 was a major component of a core-enriched fraction (CEF). Alternative models for HMW2 in core architecture are considered, based on our observations.

Mycoplasma pneumoniae is a cell wall-less pathogen of the human respiratory tract causing community-acquired tracheobronchitis and atypical, or “walking,” pneumonia (38). Colonization of the respiratory mucosa is mediated in large part by the terminal organelle, a polar, tapered extension of the mycoplasma cell having a high density of receptor-binding proteins (4, 22, 28). The terminal organelle also constitutes the motor in gliding motility (5, 11), and its duplication precedes cell division (5, 12, 32). Ultrastructurally, the terminal organelle is defined by a characteristic electron-dense core consisting of a thick rod and a thin rod oriented longitudinally in parallel and capped by a terminal button at the distal end (4, 16, 17, 39, 41). The core and terminal button are elements of the mycoplasma cytoskeleton (triton shell), a complex network of proteins resistant to extraction with Triton X-100 (TX) (1, 7, 24), much like the cytoskeletal fraction of eukaryotic cells (18, 33, 34). The composition of the M. pneumoniae triton shell has been examined by using antibody probes (22) and by mass spectrometry (29), but the identities of proteins specific to the electron-dense core are largely unknown, although cores fail to assemble in the absence of cytoskeletal proteins HMW1 and HMW2, both of which localize to the terminal organelle (3, 31, 36). HMW2 is a large protein (1,818 residues) predicted to have a globular N terminus followed by 10 dimeric or trimeric coiled-coil domains interspersed with leucine zipper motifs (23) (Fig. 1). Spontaneously arising mutant I-2 lacks HMW2

* Corresponding author. Mailing address: Department of Microbiology, 019 Riverbend Research South, 220 Riverbend Road, University of Georgia, Athens, GA 30602. Phone: (706) 542-2671. Fax: (706) 542-3804. E-mail: [email protected]. † Present address: Department of Biology, University of Nebraska at Omaha, Omaha, NE. ‡ Present address: Department of Microbiology, Miami University, Oxford, OH. 䌤 Published ahead of print on 28 August 2009. 6741

6742

BOSE ET AL.

FIG. 1. Structural features of the indicated wild-type (WT), engineered, and mutant HMW2 proteins. White boxes, predicted dimeric coiled coils; black boxes, predicted trimeric coiled coils; dark gray boxes, leucine zipper motifs; arrows, predicted N terminus of P28; black triangles, cysteine residues. The numbers above each diagram correspond to the coiled-coil regions, while the letters below correspond to the leucine zipper motifs.

MATERIALS AND METHODS Strains and culture conditions. M. pneumoniae strain M129 (wild type) and mutants I-2, H9, C1, C1R1, and I-2/HMW2⌬mid were described elsewhere (2, 15, 23). We previously established that recombinant wild-type HMW2 restores a normal phenotype to mutant I-2 (6). The mutants C1/HMW2⌬mid and C1R1/ HMW2⌬mid, expressing the partially functional HMW2 derivative HMW2⌬mid, which has an internal, in-frame deletion of 80% of HMW2 (2), were engineered as follows. Plasmid pKV185 containing the HMW2⌬mid construct was amplified by PCR using the upstream primer CAAGCTTGCATGCCTGGATCCAGAG ATGCG and the downstream primer CCCCGGATCCTTATTTAGCTGCTTT TTGGGC to create BamHI sites (underlined) for cloning into the corresponding site in Tn4001cat in pKV304 (8), forming pKV316, which was sequenced for orientation and introduced into C1 or C1R1 by electroporation (8, 15, 23). Gentamicin (18 ␮g/ml), chloramphenicol (24 ␮g/ml), or both were included as appropriate for transformant cultures (8, 15). Mycoplasmas were grown to midlog phase in Hayflick medium (14) and harvested (36); protein concentrations were determined by using a bicinchoninic acid assay kit (Pierce, Rockford, IL). Antibody preparation and analysis. Peptides corresponding to the HMW2 N terminus (1-MNDTDKKFPLQPVYDTGF) or a region near the C terminus (1783-PAFLATQQSISKQQIAQ; numbers indicate initial amino acid positions) were synthesized and coupled separately to keyhole limpet hemocyanin by using glutaraldehyde (9). For each rabbit immunization (carried out by Covance, Denver, PA), Freund’s complete adjuvant was included in the initial dose and Freund’s incomplete adjuvant was included in three subsequent booster doses. Antibodies were affinity purified using HMW2 protein bands following polyac rylamide gel electrophoresis (PAGE) and transfer onto nitrocellulose and were assessed by Western immunoblotting (37). EM. For immunogold labeling of thin sections, mycoplasmas were fixed in 2% formaldehyde–2% glutaraldehyde in 0.1 M phosphate buffer, pH 7.3, washed in phosphate-buffered saline (PBS), enrobed in 3% Noble agar (Difco Laboratories, Detroit, MI) at 58 to 60°C, postfixed in 1% OsO4 for 1 h, and dehydrated and embedded in Epon-Araldite (25). Thin sections (ca. 65 nm) were placed onto

J. BACTERIOL. hexagonal-mesh 200-mesh Cu Veco grids (Electron Microscopy Sciences, Fort Washington, PA), poststained with 5% methanolic uranyl acetate and lead citrate (30), and viewed using a JEM-1210 transmission EM (TEM; JEOL USA, Inc., Peabody, MA). Triton shells were prepared for TEM by using a protocol modified from that described by Regula et al. (29). Mycoplasma stocks were passed through a needle to disperse clumps, incubated in Hayflick medium in 24-well tissue culture dishes containing copper grids coated with 0.01% poly-L-lysine and 2% Parlodion at 37°C for 2 to 6 h, washed three times by replacement of the growth medium with PBS at room temperature, and incubated in TN buffer (20 mM Tris-HCl, pH 7.5, 150 mM NaCl) with 2% TX for 30 min at 37°C. Grids were washed three times with 10 mM PBS (pH 7.2) at room temperature, fixed for 10 s with 0.2% glutaraldehyde and 0.2% (para)formaldehyde in PBS at room temperature, immediately poststained with 3% aqueous phosphotungstic acid (pH 6.8 to 7.0) for 45 s, and air dried. Grids were viewed as described above, and core lengths were measured using ImagePro Plus software (MediaCybernetics, Silver Spring, MD). Mycoplasmas were fixed for immuno-EM as described for thin sections except that postfixation with OsO4 was omitted. For immunogold labeling, thin sections were incubated for 10 min in wash buffer (0.2 M Tris-buffered saline [TBS], pH 8.2, with 1% crystallized bovine serum albumin [Sigma-Aldrich, St. Louis, MO]) and then in TBS with 3% bovine serum albumin for 30 min. Grids were incubated in wash buffer for 1 min and then in affinity-purified HMW2-specific antibody (1:15 dilution) in wash buffer overnight at 4°C. Grids were washed for 1 min in wash buffer and at least 1 min in 0.2 M TBS, incubated 30 min with goat anti-rabbit antibody conjugated to 10-nm colloidal gold particles (Amersham Biosciences, Piscataway, NJ) diluted 1:20 in wash buffer, washed, and air dried before being viewed as described for thin sections. TX fractions were prepared for immunogold labeling as described above except that samples were probed with antibodies prior to poststaining. Gold distribution was assessed with ImagePro Plus software by measuring the distance from each gold particle to the edge of the core. Assuming 20 nm as the maximum distance between an antigen and a gold particle (10 nm each for the primary and secondary antibodies [Amersham Biosciences, Piscataway, NJ]) (10), measurements of ⱕ20 nm from the core were considered to correspond to the core itself. The mean width ⫾ standard deviation of the attachment organelle at the base was 110 ⫾ 13 nm, and the width of the core at the base was 46 ⫾ 8 nm; by subtracting these two widths, the electron-lucent region between the core and membrane was calculated to be 64 nm, or 32 nm per side. Based on this measurement and the estimated sizes of immune complexes, gold particles 20 to 52 nm from the core were considered to correspond to this peripheral region. Particles 52 to 150 nm from the core were considered to be outside the range of the terminal organelle, while particles greater than our arbitrary maximum distance from the core (150 nm) were excluded, as were rare particles found with membrane material not associated with the core. Core enrichment. Mycoplasma cells were subjected to treatments that included 0.5 to 4% Na deoxycholate, 0.5 to 4% CHAPS {3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate}, 2 to 4 M KSCN, 0.2 to 1.8 M KCl, 0.6 to 1.8 M KI, 0.1 to 2 M guanidine-HCl, 1 M Na ClO4, 2% TX, 1% digitonin with 4 mM MgCl2, 20% sucrose with guanidine-HCl and/or KI, and 2% TX–DNase I (50 ␮g/ml) in various combinations and sequences. Protein profiles of rifampin (rifampicin; 5-␮g/ml)-treated cells were also analyzed. The following extraction protocol was chosen based on the removal of material not directly associated with the core without disruption of core integrity, as determined by TEM. Cells were extracted sequentially with 2% TX–0.6 M KCl in TN buffer, 0.6 M KI–20% sucrose in TN, and 1 M guanidine-HCl in TN, with the insoluble fraction collected by centrifugation following each step. The resulting insoluble fraction was suspended in TN, urea was added to a concentration of 8 M, and samples were incubated for 10 min at 37°C and then in 3⫻ sample buffer (0.3 M Tris-HCl [pH 6.8]–6% sodium dodecyl sulfate [SDS]–60% glycerol–13% ␤-mercaptoethanol–0.06% bromophenol blue) for 10 min at 37°C. Fractions were analyzed by SDS-PAGE using 4 to 8%, 8 to 12%, and 4 to 12% gradient gels, and protein profiles were examined by silver staining (36). Extractions for EM were performed similarly except that mycoplasma cells were extracted directly on grids rather than in suspension.

RESULTS HMW2 immuno-EM. Analysis by using fluorescent protein fusions and immunofluorescence places HMW2 generally in a distal position relative to the base of the terminal organelle (3, 19, 40), but more precise localization was not possible due to

VOL. 191, 2009

HMW2 AND MYCOPLASMA PNEUMONIAE CORE ARCHITECTURE

FIG. 2. Western immunoblotting analyses with HMW2 N-terminus- and C-terminus-specific antibodies. (A) Results for HMW2 Nterminus-specific antibodies; (B and C) results for affinity-purified HMW2 N-terminus and C-terminus-specific antibodies, respectively. WT, wild type; black arrowheads, HMW2; white arrowheads, truncated HMW2; arrow, P28. Protein size standards are indicated in kilodaltons.

resolution limitations. Affinity-purified HMW2 N-terminusspecific antibodies reacted with HMW2 from wild-type M. pneumoniae and with a less intense, 117-kDa band corresponding to unstable, truncated HMW2 from mutant I-2 (Fig. 2B), as expected (6). Likewise, C-terminus-specific antibodies reacted

6743

with HMW2 and P28 in wild-type profiles and showed reduced levels of P28 in mutant I-2 (Fig. 2C), also as expected (6). Both antibody preparations typically yielded very limited labeling of wild-type cells in thin sections, which appeared to require a specific cut angle for the exposure of HMW2 epitopes; results from multiple attempts with various antibody concentrations and smaller gold particles were likewise inconclusive (data not shown). In contrast, consistent HMW2 labeling was observed with wild-type triton shells, primarily in a peripheral region along the electron-dense core (Fig. 3). With N-terminus-specific antibodies, ⬎64% of the gold particles were along the periphery of the core, 33% were associated with the core itself, and 3% were ⬎52 nm beyond the core (n ⫽ 39 cores) (Table 1). Labeling of the core peripheral region with HMW2 Cterminus-specific antibodies was even more pronounced (⬎83% of the particles were located along the periphery; n ⫽ 42 cores). A similar pattern was observed for C1R1, with ⬎82% of gold particles located in the core peripheral region after labeling with N-terminus-specific antibodies (n ⫽ 34 cores) and ⬎90% of gold particles localized in this region after labeling with HMW2 C-terminus-specific antibodies (n ⫽ 38 cores) (Fig. 3 and Table 1). Thus, neither the shorter length of HMW2 nor the absence of P28 in C1R1 affected labeling distribution. Significantly, no labeling pattern relative to the long axis of wild-type or C1R1 cores was evident, with gold

FIG. 3. Immunolocalization of HMW2 on triton shells of wild-type (WT) M. pneumoniae and mutant strains C1 and C1R1 by using affinity-purified HMW2 N-terminus-specific (1:5) and C-terminus-specific (1:15) antibodies. Scale bar, 250 nm.

6744

BOSE ET AL.

J. BACTERIOL.

TABLE 1. Quantitation of the HMW2 immunogold localization patterns for triton shells from wild-type and mutant C1R1 M. pneumoniae strainsa M. pneumoniae strain

Antibody

No. of cores

No. of gold particles

Avg no. of gold particles per core

% on core, including on edge

% within 52 nm from core

% over 52 nm from core

Wild type C1R1 Wild type C1R1

N-terminal HMW2 antibody N-terminal HMW2 antibody C-terminal HMW2 antibody C-terminal HMW2 antibody

39 34 43 38

206 165 71 71

5.3 4.9 1.7 1.9

32.9 12.1 14.1 5.6

64.3 82.4 83.1 90.1

2.9 5.5 2.8 4.2

a

Due to rounding, the sums of the numbers in the last three columns for each row may not equal 100%.

particles being distributed along the length of the core and not, for example, at one or both ends. We also examined mutant C1, which has low levels of a truncated HMW2 protein as determined by Western immunoblotting (23) (Fig. 2A) and for which C-terminus-specific antibodies yielded only a few gold particles, providing a background control (Fig. 3). Labeling by N-terminus-specific antibodies was evident in the C1 mutant but not evaluated further, as clear core boundaries could not be distinguished. Terminal organelle ultrastructure in mutant strains. Mutants I-2, C1, and H9 (15, 20, 23) produce truncated HMW2 derivatives that differ in size and relative stability (Fig. 2C), while mutant C1R1 has an internal, in-frame deletion in MPN310 that results in a shorter HMW2 protein and the loss of P28 (23). Mutant I-2 produces no electron-dense core, but mutants C1, H9, and C1R1 have not been examined previously. We analyzed cores primarily in TX-extracted preparations (Fig. 4) because characterization of thin sections can be affected by the cut angle, limiting the ability to ascertain core dimensions. The core consists of a larger rod and a smaller rod in parallel, with what has been described previously as a wheelor bowl-like structure at the base (16, 17), but in the present study, only occasionally were both rods or the wheel-like structure clearly visible. The lengths of electron-dense cores in wild-type triton shells averaged 271.8 ⫾ 21.6 nm (n ⫽ 30), and the cores exhibited characteristic periodicity, as described previously (4, 13, 27). As expected, no cores were apparent in mutant H9 samples, either thin sections or triton shells (data

not shown), but higher-ordered structures that may correspond to partially assembled cores were seen in mutant C1 triton shells (Fig. 3 and 4A). C1R1 cores averaged 217.8 ⫾ 26.6 nm (n ⫽ 26), or about 80% of the length of wild-type cores (Fig. 4B) (P ⫽ 8.72 ⫻ 10⫺11). The in-frame deletion in C1R1 results in the loss of P28, which was restored with recombinant HMW2⌬mid (2), leading to an increase in core length to nearwild-type levels (250.2 ⫾ 41.1 nm [n ⫽ 27] for a representative transformant), suggesting that a shorter core in C1R1 was due to the absence of P28 rather than truncation of HMW2. HMW2⌬mid did not restore wild-type cores in mutants I-2 and C1 (although an occasional core was observed in the former, due probably to a revertant population [21]), reaffirming that core assembly requires near-full-length, contiguous HMW2, including the C-terminal domain. Core enrichment. Previous studies (7, 37) established that the treatment of triton shells with 0.6 M KCl or 0.6 M KI removes most material peripherally associated with the core; the more chaotropic KI also partially destabilizes the core (7, 37). We tested various detergents and salts (Table 2) for their abilities to enrich further the electron-dense core and better correlate HMW2 localization and core composition. Briefly, extraction with 2% TX–0.6 M KCl followed by 0.6 M KI–20% sucrose substantially decreased the amount of insoluble material not specifically core associated (Fig. 5C). Sucrose was included in order to reduce the destabilizing effects of 0.6 M KI alone. Structures consistent in dimensions with both the larger and smaller rods that form the core were visible in some im-

FIG. 4. (A) EM analyses of triton shells from wild-type (WT) M. pneumoniae and the indicated HMW2 mutant strains. Scale bar, 250 nm. (B) Comparison of core lengths in wild-type M. pneumoniae and HMW2 mutant C1R1. Scale bar, 250 nm.

VOL. 191, 2009

HMW2 AND MYCOPLASMA PNEUMONIAE CORE ARCHITECTURE

6745

TABLE 2. Summary of sequential treatments for M. pneumoniae core enrichmenta Treatment 1

Treatment 2

2% TX 2% TX 2% TX–DNase I (50 ␮g/ml) 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% Na deoxycholate–1 M KCl 1% CHAPS 500 mM Gdn-HCl 1 M Gdn-HCl

Treatment 3

0.6 M KI 1.8 M KI 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–0.6 M KCl 2% TX–1.8 M KCl 0.6 M KI with 20% sucr 0.6 M KI with 20% sucr 0.6 M KI with 20% sucr 0.6 M KI with 20% sucr 0.6 M KI with 20% sucr 2% Na deoxycholate 2% Na deoxycholate 2% Na deoxycholate 2% Na deoxycholate 2 M Gdn-HCl 2 M Gdn-HCl–20% sucr 2 M KSCN 1 M Gdn-HCl

2% TX–0.6 M KCl 2% TX–1.8 M KCl 1 M Gdn-HCl 1.2 M Gdn-HCl 1.5 M Gdn-HCl 2 M Gdn-HCl NA 1 M Gdn-HCl 1.5 M Gdn-HCl 2 M Gdn-HCl

Presence of coresb

Purityb

Yes No Yes Yes Yes Yes Yes Yes Yes Yes Partial No No Yes Partial Partial No No Partial No Partial Partial Partial No

⫹⫹⫹ NA ⫺ ⫹⫹ ⫹ ⫹ ⫹ ⫹⫹ ⫹⫹⫹ ⫹⫹⫹⫹ ⫹⫹ NA NA ⫹⫹⫹ ⫹⫹ ⫹ NA NA ⫹ NA ⫹ ⫹ ⫹ NA

a

Gdn-HCl, guanidine-HCl; sucr, sucrose; NA, not applicable. The presence/absence of cores and the purity thereof were determined by EM analyses following the indicated extraction treatments. Amounts of residual material associated with the cores following the last treatment are expressed as ⫹⫹⫹⫹, ⫹⫹⫹, ⫹⫹, ⫹, and ⫺, from largest to smallest. b

ages of these preparations (Fig. 5B), due perhaps to the removal of proteins that normally surround the core. Occasionally, large pieces of debris shorter than the average core were apparent, and they may reflect core substructures; whether these correspond to other structural elements revealed by electron cryotomography (17) is unclear. Isolated cores were likewise abundant in preparations extracted with TX–0.6 M KCl– 0.6 M KI–20% sucrose and then 1 M guanidine-HCl, with the distinct rods of the core often visible, as were fragments that may represent core substructures (Fig. 5D). Higher concentrations of guanidine-HCl destabilized the cores completely (data not shown); thus, 1 M guanidine-HCl appeared to be the threshold for extraction. To assess the extent of membrane removal by this regimen, we monitored the fate of [3H]palmitate as a membrane marker, with radioactivity measurements indicating the removal of ⬎99.9% of the [3H]palmitic acid (data not shown). SDS-PAGE analysis of the CEF revealed substantially reduced levels of some protein bands and selective enrichment of others (Fig. 6A), with HMW2 present in abundance (Fig. 6B). DISCUSSION HMW2 is a cytoskeletal component (35, 37) of the terminal organelle of M. pneumoniae (3, 19, 40) and is required to stabilize several other terminal organelle proteins in the assembly of the electron-dense core (22, 26, 31), but its specific role in the architecture of this structure is poorly defined. The lack of a core in the absence of HMW2, when considered in the context of the deduced structural features of HMW2 (23), suggested a model in which parallel bundles of HMW2 and P28 orient lengthwise to constitute the major rod-like compo-

nent of the core (3). According to that model, the N- and C-terminal domains of HMW2 were predicted to localize specifically at the ends of the rod, but immunolocalization data obtained in the present study did not support that prediction. Rather, HMW2 N- and C-terminus-specific antibodies both labeled proximal to the terminal button, along the length and primarily peripheral to the major rod. This pattern differed dramatically from that seen previously for HMW3, where labeling was observed almost exclusively directly on the major rod and terminal button (37). It is noteworthy that findings from both the previous and present studies provide only a two-dimensional view of protein distribution and do not distinguish labeling outward from the plane of the core. Thus, a limitation of immunolocalization with TX fractions is the collapsing of cell constituents from three to two dimensions. In the hopes of overcoming that limitation, we also fixed mycoplasmas before TX extraction, but that approach severely reduced immunogold labeling (data not shown). The previous model (3) also predicts that truncated HMW2 derivatives should yield correspondingly shorter electrondense cores, but no clear correlation was observed here. Truncation of HMW2 at its C terminus resulted in failure to form a core in mutants I-2 (31) and H9 (data not shown), while the shorter cores in mutant C1R1, having a slightly truncated HMW2 and no P28, correlated with the absence of P28 rather than the truncation of HMW2. Moreover, HMW2⌬mid, which has an internal, in-frame deletion of 80% of HMW2 yet is partially functional (2), did not yield short but otherwise intact cores in mutant C1 as predicted. Thus, the ability of HMW2⌬mid to restore HMW3 and P65 stability and an intermediate hemadsorption phenotype in mutant I-2 (2) is sepa-

6746

BOSE ET AL.

J. BACTERIOL.

FIG. 6. Triton shells and CEFs from wild-type (WT) M. pneumoniae and mutant C1. (A) Silver staining profile obtained by SDSPAGE with a 4 to 12% gradient gel; protein standards in kilodaltons are indicated to the left. White arrowheads indicate protein bands from the triton shell in the wild-type profile that were decreased in intensity in the CEF; white arrows indicate protein bands enriched in the wild-type CEF compared to those in the triton shell. (B) Western immunoblotting analysis with HMW2 N-terminus-specific antiserum. FIG. 5. EM analyses of wild-type cells following core enrichment treatments. Samples were treated with 2% TX (A), 2% TX–0.6 M KCl (B), 2% TX–0.6 M KCl–0.6 M KI with 20% sucrose (C), and 2% TX–0.6 M KCl–0.6 M KI with 20% sucrose–1 M guanidine-HCl (D). Black arrowheads, residual material; white and black arrows, small and large rods of the core, respectively; scale bars, 250 nm.

rable from core formation and appears to be linked specifically to the C-terminal domain of HMW2. Likewise, the failure of HMW2⌬mid to stabilize HMW1 (2) demonstrates the importance of the central domain of HMW2, the overall length of HMW2, or both in HMW1 function in core formation. Finally, an unexpected disjunction between HMW2 stability and the formation of higher-ordered structures in mutants I-2, H9, and C1 was noted. The HMW2 derivative in mutant H9 was considerably more stable than that in I-2 (Fig. 2C), and yet neither

yielded a core. In contrast, the slightly longer HMW2 counterpart in C1 was present at lower steady-state levels than the HMW2 protein in I-2 or H9 yet contributed to the formation of higher-ordered structures that we speculate are partially assembled cores (Fig. 3 and 4). A comparison of HMW2 deletion derivatives established correlations between specific regions of HMW2 and function in the context of core architecture. Thus, the C-terminal domain and/or trimeric coiled coil 10 (Fig. 1) of HMW2, but not P28, was required for what appears to be normal core assembly. Likewise, the region containing dimeric coiled coils 5 and 6, leucine zipper motifs C, D, and E, and Cys1217 in HMW2 from mutant C1 but not in HMW2 derivatives from mutants I-2 and H9 (Fig. 1) appears to be essential for core assembly. Although present in mutant C1, leucine zipper motif E was not

VOL. 191, 2009

HMW2 AND MYCOPLASMA PNEUMONIAE CORE ARCHITECTURE

6747

FIG. 7. Two models (A and B) for HMW2 orientation in core architecture. The pale gray box represents the core, as indicated. N and C, HMW2 N and C termini, respectively; black lines, central portion of HMW2. Black, white, and dark gray boxes represent predicted dimeric coiled-coil, trimeric coiled-coil, and leucine zipper domains, respectively, numbered or lettered as in Fig. 1.

considered to be essential for core formation given its absence in C1R1, which forms a functional but slightly shorter core. The program DISULFIND (Department of Systems and Computer Science, Universita` di Firenze, Italy) predicted that Cys1217 forms an interchain disulfide (data not shown), perhaps accounting for HMW2 disulfide-linked dimers (35). Wildtype cores treated with ␤-mercaptoethanol remained intact (data not shown), and therefore, disulfide reduction does not appear to disrupt the integrity of assembled cores, but we cannot rule out the possibility that interchain disulfide formation involving Cys1217 contributes to protein alignment for subsequent coiled-coil interaction during the process of core assembly. This possibility is particularly intriguing given (i) the presence of Cys1217 in mutant C1, which forms higher-ordered structures, and its absence in mutants H9 and I-2, which do not, and (ii) its conservation in the HMW2 homolog in the closely related Mycoplasma genitalium (data not shown). At least two revised models for HMW2 and core architecture encompass our observations in the present study (Fig. 7). In both models, the N and C termini of HMW2 align along the lateral edges of the core, consistent with immunolocalization data. The central domain of HMW2 is oriented outward in one model, perhaps corresponding to the spoke structures radiating from the core (16). This orientation would likely position a portion of the central domain of HMW2 in close proximity to HMW1 at the mycoplasma membrane and, in that respect, would accommodate the requirement for that region of HMW2 in particular for HMW1 stabilization and subsequent core formation (2; this study). Alternatively, the central domain of HMW2 may fold inward as part of the major core itself, with protein interactions involving coiled-coil and leucine zipper motifs potentially contributing to the observed periodicity of the core (4). Loss of the C-terminal domain and, in particular, the predicted trimeric coiled-coil regions therein is probably the most important factor in the failure of mutants I-2 and H9 to assemble a core. Both models reflect a role for

P28 in completing trimeric coiled coils, particularly at the ends of the rod. The C1R1 mutant lacks P28 and assembles a core approximately 20% shorter than the wild-type core. Thus, the loss of P28 may have an impact on the trimeric coiled-coil interaction with the C terminus of HMW2, thereby causing a reduction in length but still allowing an otherwise apparently normal core to form. While HMW2 is an integral component of the major rod of the electron-dense core in both models, neither accounts for the consistent lengths of the cores among individual M. pneumoniae cells (13). Finally, analysis of the CEF by EM indicates substantial removal of TX-insoluble material not specifically associated with the electron-dense core, a conclusion supported by SDS-PAGE profiles that reveal loss of or reductions in some protein bands and enrichment of others. HMW2 is an abundant component of the CEF, and preliminary analysis by immuno-EM revealed labeling similar to that of nonenriched cores pictured in Fig. 3 (data not shown), consistent with the conclusion that HMW2 is a component of the core and not a core-associated substructure. We believe that more definitive analysis will require additional core enrichment, for example, by using differential centrifugation. ACKNOWLEDGMENTS We thank M. Ard, K. Lee, and K. Cureton for technical assistance. This work was supported by Public Health Service grant AI23362 from the National Institute of Allergy and Infectious Diseases to D.C.K. REFERENCES 1. Balish, M. F., and D. C. Krause. 2002. Cytadherence and the cytoskeleton, p. 491–518. In S. Razin and R. Herrmann (ed.), Molecular biology and pathogenicity of mycoplasmas. Kluwer Academic/Plenum Publishers, New York, NY. 2. Balish, M. F., S. M. Ross, M. Fisseha, and D. C. Krause. 2003. Deletion analysis identifies key functional domains of the cytadherence-associated protein HMW2 of Mycoplasma pneumoniae. Mol. Microbiol. 50:1507–1516. 3. Balish, M. F., R. T. Santurri, A. M. Ricci, K. K. Lee, and D. C. Krause. 2003. Localization of Mycoplasma pneumoniae cytadherence-associated protein

6748

4. 5. 6.

7.

8.

9. 10.

11.

12.

13.

14. 15.

16. 17.

18.

19.

20.

21.

22.

BOSE ET AL.

HMW2 by fusion with green fluorescent protein: implications for attachment organelle structure. Mol. Microbiol. 47:49–60. Biberfeld, G., and P. Biberfeld. 1970. Ultrastructural features of Mycoplasma pneumoniae. J. Bacteriol. 102:855–861. Bredt, W. 1968. Motility and multiplication of Mycoplasma pneumoniae. A phase contrast study. Pathol. Microbiol. 32:321–326. Fisseha, M., H. W. Gohlmann, R. Herrmann, and D. C. Krause. 1999. Identification and complementation of frameshift mutations associated with loss of cytadherence in Mycoplasma pneumoniae. J. Bacteriol. 181:4404– 4410. Go ¨bel, U., V. Speth, and W. Bredt. 1981. Filamentous structures in adherent Mycoplasma pneumoniae cells treated with nonionic detergents. J. Cell Biol. 91:537–543. Hahn, T.-W., E. A. Mothershed, R. H. Waldo III, and D. C. Krause. 1999. Construction and analysis of a modified Tn4001 conferring chloramphenicol resistance in Mycoplasma pneumoniae. Plasmid 41:120–124. Harlow, E., and D. Lane. 1988. Antibodies: a laboratory manual, p. 73. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Harrison, D. M., J. Skidmore, J. P. Armitage, and J. R. Maddock. 1999. Localization and environmental regulation of MCP-like proteins in Rhodobacter sphaeroides. Mol. Microbiol. 31:885–892. Hasselbring, B. M., and D. C. Krause. 2007. Cytoskeletal protein P41 is required to anchor the terminal organelle of the wall-less prokaryote Mycoplasma pneumoniae. Mol. Microbiol. 63:44–53. Hasselbring, B. M., J. L. Jordan, R. W. Krause, and D. C. Krause. 2006. Terminal organelle development in the cell wall-less bacterium Mycoplasma pneumoniae. Proc. Natl. Acad. Sci. USA 103:16478–16483. Hatchel, J. M., and M. F. Balish. 2008. Attachment organelle ultrastructure correlates with phylogeny, not gliding motility properties, in Mycoplasma pneumoniae relatives. Microbiology 154:286–295. Hayflick, L. 1965. Tissue cultures and mycoplasmas. Tex. Rep. Biol. Med. 23(Suppl. 1):285–303. Hedreyda, C. T., and D. C. Krause. 1995. Idnetification of a possible cytadherence regulatory locus in Mycoplasma pneumoniae. Infect. Immun. 63: 3479–3483. Hegermann, J., R. Herrmann, and F. Mayer. 2002. Cytoskeletal elements in the bacterium Mycoplasma pneumoniae. Naturwissenschaften 89:453–458. Henderson, G. P., and G. J. Jensen. 2006. Three-dimensional structure of Mycoplasma pneumoniae’s attachment organelle and a model for its role in gliding motility. Mol. Microbiol. 60:376–385. Herrmann, H., and G. Wiche. 1983. Specific in situ phosphorylation of plectin in detergent-resistant cytoskeletons from cultured Chinese hamster ovary cells. J. Biol. Chem. 258:14610–14618. Kenri, T., S. Seto, A. Horino, Y. Sasaki, T. Sasaki, and M. Miyata. 2004. Use of fluorescent-protein tagging to determine the subcellular localization of Mycoplasma pneumoniae proteins encoded by the cytadherence regulatory locus. J. Bacteriol. 186:6944–6955. Krause, D. C., D. K. Leith, R. M. Wilson, and J. B. Baseman. 1982. Identification of Mycoplasma pneumoniae proteins associated with hemadsorption and virulence. Infect. Immun. 35:809–817. Krause, D. C., D. K. Leith, and J. B. Baseman. 1983. Reacquisition of specific proteins confers virulence in Mycoplasma pneumoniae. Infect. Immun. 39:830–836. Krause, D. C., and M. F. Balish. 2004. Cellular engineering in a minimal microbe: structure and assembly of the terminal organelle of Mycoplasma pneumoniae. Mol. Microbiol. 51:917–924.

J. BACTERIOL. 23. Krause, D. C., T. Proft, C. T. Hedreyda, H. Hilbert, H. Plagens, and R. Herrmann. 1997. Transposon mutagenesis reinforces the correlation between Mycoplasma pneumoniae cytoskeletal protein HMW2 and cytadherence. J. Bacteriol. 179:2668–2677. 24. Meng, K. E., and R. M. Pfister. 1980. Intracellular structures of Mycoplasma pneumoniae revealed after membrane removal. J. Bacteriol. 144:390–399. 25. Mollenhauer, H. H., and W. G. Whaley. 1963. An observation on the functioning of the Golgi apparatus. J. Cell Biol. 17:222–225. 26. Popham, P. L., T.-W. Hahn, K. A. Krebes, and D. C. Krause. 1997. Loss of HMW1 and HMW3 in noncytadhering mutants of Mycoplasma pneumoniae occurs post-translationally. Proc. Natl. Acad. Sci. USA 94:13979–13984. 27. Powell, D. A., P. C. Hu, M. Wilson, A. M. Collier, and J. B. Baseman. 1976. Attachment of Mycoplasma pneumoniae to respiratory epithelium. Infect. Immun. 13:959–966. 28. Razin, S., and E. Jacobs. 1992. Mycoplasma adhesion. J. Gen. Microbiol. 138:407–422. 29. Regula, J. T., G. Boguth, A. Gorg, J. Hegermann, F. Mayer, R. Frank, and R. Herrmann. 2001. Defining the mycoplasma “cytoskeleton”: the protein composition of the Triton X-100 insoluble fraction of the bacterium Mycoplasma pneumoniae determined by 2-D gel electrophoresis and mass spectrometry. Microbiology 147:1045–1057. 30. Reynolds, E. S. 1963. The use of lead citrate at high pH as an electronopaque stain in electron microscopy. J. Cell Biol. 17:208–212. 31. Seto, S., and M. Miyata. 2003. Attachment organelle formation represented by localization of cytadherence proteins and formation of the electron-dense core in wild-type and mutant strains of Mycoplasma pneumoniae. J. Bacteriol. 185:1082–1091. 32. Seto, S., G. Layh-Schmitt, T. Kenri, and M. Miyata. 2001. Visualization of the attachment organelle and cytadherence proteins of Mycoplasma pneumoniae by immunofluorescence microscopy. J. Bacteriol. 183:1621–1630. 33. Starger, J. M., and R. D. Goldman. 1977. Isolation and preliminary characterization of 10-nm filaments from baby hamster kidney (BHK-21) cells. Proc. Natl. Acad. Sci. USA 74:2422–2426. 34. Steinert, P. M., S. B. Zimmerman, J. M. Starger, and R. D. Goldman. 1978. Ten-nanometer filaments of hamster BHK-21 cells and epidermal keratin filaments have similar structures. Proc. Natl. Acad. Sci. USA 75:6098–6101. 35. Stevens, M. K., and D. C. Krause. 1990. Disulfide-linked protein associated with Mycoplasma pneumoniae cytadherence phase variation. Infect. Immun. 58:3430–3433. 36. Stevens, M. K., and D. C. Krause. 1991. Localization of the Mycoplasma pneumoniae cytadherence-accessory proteins HMW1 and HMW4 in the cytoskeletonlike Triton shell. J. Bacteriol. 173:1041–1050. 37. Stevens, M. K., and D. C. Krause. 1992. Mycoplasma pneumoniae cytadherence phase-variable protein HMW3 is a component of the attachment organelle. J. Bacteriol. 174:4265–4274. 38. Waites, K. B., and D. F. Talkington. 2004. Mycoplasma pneumoniae and its role as a human pathogen. Clin. Microbiol. Rev. 17:697–728. 39. Wall, F., R. M. Pfister, and N. L. Somerson. 1983. Freeze-fracture confirmation of the presence of a core in the specialized tip structure of Mycoplasma pneumoniae. J. Bacteriol. 154:924–929. 40. Willby, M. J., M. F. Balish, S. M. Ross, K. K. Lee, J. L. Jordan, and D. C. Krause. 2004. HMW1 is required for stability and localization of HMW2 to the attachment organelle of Mycoplasma pneumoniae J. Bacteriol. 186:8221– 8228. 41. Wilson, M. H., and A. M. Collier. 1976. Ultrastructural study of Mycoplasma pneumoniae in organ culture. J. Bacteriol. 125:332–339.