Ultrastructural Variability of the Exosporium Layer of Clostridium difficile Spores

Abstract

The anaerobic sporeformer Clostridium difficile is the leading cause of nosocomial antibiotic-associated diarrhea in developed and developing countries. The metabolically dormant spore form is considered the transmission, infectious, and persistent morphotype, and the outermost exosporium layer is likely to play a major role in spore-host interactions during the first contact of C. difficile spores with the host and for spore persistence during recurrent episodes of infection. Although some studies on the biology of the exosporium have been conducted (J. Barra-Carrasco et al., J Bacteriol 195:3863–3875, 2013, http://dx.doi.org/10.1128/JB.00369-13; J. Phetcharaburanin et al., Mol Microbiol 92:1025–1038, 2014, http://dx.doi.org/10.1111/mmi.12611), there is a lack of information on the ultrastructural variability and stability of this layer. In this work, using transmission electron micrographs, we analyzed the variability of the spore's outermost layers in various strains and found distinctive variability in the ultrastructural morphotype of the exosporium within and between strains. Through transmission electron micrographs, we observed that although this layer was stable during spore purification, it was partially lost after 6 months of storage at room temperature. These observations were confirmed by indirect immunofluorescence microscopy, where a significant decrease in the levels of two exosporium markers, the N-terminal domain of BclA1 and CdeC, was observed. It is also noteworthy that the presence of the exosporium marker CdeC on spores obtained from C. difficile biofilms depended on the biofilm culture conditions and the strain used. Collectively, these results provide information on the heterogeneity and stability of the exosporium surface of C. difficile spores. These findings have direct implications and should be considered in the development of novel methods to diagnose and/or remove C. difficile spores by using exosporium proteins as targets.

INTRODUCTION

Clostridium difficile is a leading cause of nosocomial antibiotic-associated diarrhea in developed and developing countries (1, 2, 5). Antibiotic treatment results in disturbance of the natural gut microflora, allowing germination and outgrowth of C. difficile spores and proliferation and colonization of live cells (3). During the course of the infection, C. difficile initiates a sporulation process that culminates with the production of metabolically dormant spores (3, 4), which are essential for (i) the persistence of C. difficile in the host and (ii) transmission of the disease between hosts (4).

C. difficile spore morphology resembles that of other spore-forming bacteria; however, the protein composition of the outer layers (i.e., the spore coat and exosporium layer) differs greatly from that of other Gram-positive bacteria (13, 14). The composition of the spore coat has been described recently (7); fewer than 25% of the spore proteins and several spore coat proteins have been functionally characterized (8, 9). Transmission electron microscopy (TEM) reveals the presence of typical laminations (i.e., lamellae) (6, 10–12) in the spore coat of other spore-forming bacteria (13). As reported for Bacillus subtilis (14), the spore coat layer of C. difficile spores is resistant to enzymatic digestion with proteases (i.e., trypsin and proteinase K) (11). The outermost exosporium layer of C. difficile spores is an electron-dense layer with several unique features that differentiate this layer, both morphologically and functionally, from the spore coat (6, 10, 11). (i) Unlike the outermost layer of members of the B. cereus group, which exhibits hairlike projections and does not interact directly with the spore coat surface (13), the C. difficile exosporium exhibits hairlike projections in most strains (14) and interacts directly with the spore coat (10, 14). (ii) In contrast to the spore coat, the exosporium is easily digested with proteases (i.e., trypsin and proteinase K) (11). (iii) The exosporium contributes to the spore's hydrophobicity (11). (iv) Its assembly seems to depend, at least partly, on the morphogenetic protein CdeC, which is uniquely located in the exosporium of C. difficile spores (6, 10) and is essential for the assembly of the exosporium and for correct spore coat assembly (10).

Recent studies have implicated the exosporium layer in the pathogenesis of C. difficile infections (CDI) and the persistence of C. difficile spores on stainless steel (15, 16). Recently, Phetcharaburanin et al. demonstrated that the exosporium collagen-like BclA1 glycoprotein is required for animal susceptibility to colonization and infection by C. difficile laboratory strain 630. Moreover, recent reports have demonstrated that strain 630 spores have a hairless exosporium (6, 11), which also differs in general ultrastructural features from what has been observed in epidemically relevant strains (10, 14, 15). Furthermore, the stability of this layer is also a matter of controversy; several reports suggest that it is a fragile and easily lost layer (8, 9), while others provide evidence that it is robust and remains attached to the spore (6, 10, 11, 14, 15). The stability of the exosporium layer seems to be relevant, as the exosporium layer also contributes to spore adherence to inert surfaces (i.e., stainless steel) and consequently to the persistence of spores on hospital surfaces (15). Consequently, the aims of this study were to determine the heterogeneity of the exosporium ultrastructure of C. difficile spores in various isolates of clinical relevance, determine the stability of this layer during long-term storage at room temperature, and evaluate whether the exosporium layer is present in spores formed during biofilm development.

MATERIALS AND METHODS

Bacterial strains and plasmids.

The C. difficile strains used in this study were the laboratory strain 630 ribotype 012 (RT012); epidemic strain R20291 (RT027); and strains M120 (RT078-126), TL176 (RT014-020), and TL178 (RT02). C. difficile was routinely grown under anaerobic conditions in a Bactron III-2 anaerobic chamber (Shellab) in 3.7% brain heart infusion broth supplemented with 0.5% yeast extract (BHIS) or on BHIS agar plates.

Spore suspensions.

Spore suspensions were prepared by plating a 1:500 dilution of an overnight culture onto a 3% Trypticase soy (TS)–0.5% yeast extract (TY) agar plate and incubating it for 5 days at 37°C under anaerobic conditions. Spores were harvested with ice-cold sterile distilled water, centrifuge washed (5,000 rpm, 5 min) once, and gently resuspended in sterile distilled water. Spores were loaded onto a 50% Nycodenz (Sigma-Aldrich) solution, centrifuged (10,000 rpm, 40 min), and centrifuge washed (5,000 rpm, 5 min) twice with ice-cold sterile distilled water to remove Nycodenz remnants.

TEM.

Spores (∼2 × 10⁸) of C. difficile strains 630, R20291, M120, TL176, and TL178 were fixed with 3% glutaraldehyde in 0.1 M cacodylate buffer (pH 7.2) overnight at 4°C and stained for 30 min with 1% tannic acid. Samples were further processed and embedded in Spurr resin as previously described (12). Thin sections of 90 nm obtained with a microtome were placed on glow discharge carbon-coated grids and double lead stained with 2% uranyl acetate and lead citrate. Grids were analyzed with a Philips Tecnai 12 Bio Twin at the electron microscopy facility of the Pontificia Universidad Católica de Chile.

Room temperature storage.

Suspensions of 5 × 10⁹ spores/ml were diluted in phosphate-buffered saline (PBS) to a final concentration of 1.7 × 10⁸ spores/ml. Aliquots (50 μl) of spore suspensions in sterile distilled water were stored in Eppendorf tubes for up to 6 months at room temperature (23°C). Three independent batches of spores were used, and aliquots were stored in triplicate.

Immunofluorescence microscopy.

C. difficile R20291 and M120 spores were fixed in 3% paraformaldehyde (pH 7.4) for 20 min on poly-l-lysine-coated glass cover slides. Fixed spores were rinsed three times with PBS, blocked with 1% bovine serum albumin (BSA) for 30 min, and further incubated overnight at 4°C with 1:50-diluted rabbit antiserum raised against the N-terminal domain of BclA1 (α-NTD BclA1) (17) and rat serum raised against the exosporium marker CdeC (anti-CdeC) (10). Next, coverslips were incubated for 1 h at room temperature with 1:200 anti-rabbit or -rat IgG Alexa Fluor 488 conjugate (Invitrogen) in PBS with 1% BSA and washed three times with PBS and once with distilled water. Samples were dried at room temperature for 30 min, and then the coverslips were mounted using Dako Fluorescence Mounting medium (Dako North America) and sealed with nail polish. Samples were analyzed with an Olympus BX53 fluorescence microscope. A total of 150 spores were analyzed. Control experiments with serum obtained prior to immunization or in the absence of primary antibodies yielded no fluorescence signal in R20291 and M120 spores (data not shown). Fluorescence intensity was quantified with ImageJ software (http://rsbweb.nih.gov/ij/).

Biofilm formation.

Biofilm was evaluated by using previously described protocols (18, 19). Briefly, cultures grown overnight in BHIS were diluted 1:100 in fresh BHIS or TS broth and added to 24-well polystyrene plates (Greiner Bio-One, Stuttgart, Germany). Plates were incubated at 37°C under anaerobic conditions for 5 days. Some wells were used to measure biofilm biomass with crystal violet (CV) as previously described (19). Briefly, biofilm was stained with 1 ml of filter-sterilized 0.2% CV and incubated for 30 min at 37°C under anaerobic conditions. CV was removed from the wells, and then the wells were washed twice with sterile PBS. The dye was extracted by adding 1 ml of methanol to each well, and then the plates were incubated for 30 min at room temperature. The methanol-extracted dye was serially diluted, and optical density at 570 nm (OD₅₇₀) was measured with a spectrophotometer (ELx800; BioTek). Other wells were used to remove nonbiofilm material, wells were rinsed three times with sterile PBS, and the biofilm mass was removed by mechanical scraping.

The presence of the exosporium layer on spores formed during biofilm development was analyzed by immunofluorescence as described above, with some modifications. Briefly, biofilms of strains R20291 and M120 were fixed with paraformaldehyde and blocked with 1% BSA for 1 h as described above. Biofilms were subsequently immunostained at room temperature for 1 h with 1:100-diluted rat antiserum raised against CdeC, rinsed, incubated for 1 h at room temperature with 1:400-diluted anti-rat Alexa 488 conjugate (LifeScience Technologies) in PBS–1% BSA, washed three times, and rinsed with sterile distilled water. Samples were mounted with Dako fluorescence mounting medium and analyzed with an Olympus BX53 fluorescence microscope. Control experiments included the use of rat serum obtained prior to immunization, which yielded no fluorescence signal in R20291 and M120 spores (data not shown). Control experiments also included purified spores obtained from TY agar plates.

Statistical analysis.

Student's t test was used to detect statistically significant differences between groups.

RESULTS

Two distinctive exosporium morphotypes within C. difficile spore populations.

To gain insight into the variability of the outermost exosporium, we evaluated spores of five different strains by TEM; these included a laboratory strain (630), an epidemic strain (R20291), and three additional isolates belonging to different ribotypes (i.e., M120, TL176, and TL178). TEM confirmed previously described differences (10, 12, 14) in the exosporium layer between C. difficile 630 and R20291 spores. That is, 630 spores had a smooth, electron-dense exosporium layer lacking the classical hairlike extensions observed in exosporium structures of Bacillus anthracis spores (13) (Fig. 1A). In contrast, R20291 spores had an exosporium layer with hairlike projections, which were also observed by TEM of M120, TL176, and TL178 spores (Fig. 1A), and depending on the strain, these hairlike projections varied from 36 to 65 nm in length (Fig. 1B).

FIG 1.

Ultrastructural variability of the outer layers of C. difficile spores. (A) C. difficile spores of strains 630 (ribotype 012), R20291 (ribotype 027), M120 (ribotype 078-126), TL176 (ribotype 014-020), and TL178 (ribotype 002) were analyzed by TEM as described in Materials and Methods. Spore populations (n = 100) were analyzed, and two distinct ultrastructural phenotypes became evident in spore samples of all of the strains tested, i.e., spores with a thick exosporium layer and spores with a thin exosporium layer. The percentages of spores with a thick exosporium (black bar) and those a thin exosporium (gray bar) are depicted below the micrographs. Results are representative of one experiment. Experiments were repeated with two independent spore preparations with essentially similar results. (B) The thickness of the exosporium and the outer and inner coat layers of C. difficile spores with a thick (black bars) or a thin (gray bars) exosporium layer were estimated by TEM of at least 10 individual spores. Scale bars are 100 and 200 nm in micrographs with whole spores and selected spore sections, respectively. Ex, exosporium; EC, external coat; IC, inner coat. Error bars denote standard errors of the means. Asterisks denote statistically significant differences (n.s., not significant; *, P < 0.05; ***, P < 0.001). ND, not detected.

During the ultrastructural analysis, it became evident that all of the strains produced spores with two distinctive thicknesses of the exosporium, that is, spores with a thick exosporium and spores with a thin exosporium (Fig. 1A). For example, nearly 50% of the strain 630 and TL176 spores had a thick exosporium, while 75% of the M120 and TL176 spores had a thick exosporium. Notably, only 25% of the R20291 spores had a thick exosporium (Fig. 1A). On average, C. difficile spores with a thick exosporium had an exosporium nearly 2-fold thicker than that of spores with a thin exosporium (Fig. 1B). Although significant differences in the length of the hairlike extensions were observed between thick- and thin-exosporium spores, the magnitude of this change was small (i.e., a 5- to 20-nm difference, depending on the strain) (Fig. 1B). These differences were not observed between M120 spores with a thick exosporium and those with a thin exosporium (Fig. 1B). However, significant differences in the thickness of the spore inner and external coat layers were detectable between these two exosporium morphotypes, among all of the strains analyzed (Fig. 1B). Another noteworthy observation was that only those spores with a thick exosporium that had hairlike projections (i.e., strains R20291, M120, TL176, and TL178) exhibited geometrically distributed bumps in the exosporium layer (Fig. 1A). These bumps were not observed in spores of strains R20291, M120, TL176, and TL178 with a thin-exosporium morphotype or in either thick- or thin-exosporium strain 630 spores (Fig. 1A). Notably, these bumps seem to be made of the same electron-dense material as the exosporium of strain 630 spores. Besides these differences, the underlying layers (i.e., spore outer and inner coats) had similar ultrastructural features among the strains analyzed (Fig. 1A); all of the strains had spore coats with lamellae, a conserved feature of Bacillus and Clostridium spores (Fig. 1A). Collectively, these results demonstrate the presence of two exosporium phenotypes (i.e., a thick exosporium and a thin exosporium) in the same population of C. difficile spores.

Long-term storage decreases the abundance of exosporium proteins of C. difficile spores.

Given the plausible roles of the exosporium in early C. difficile spore-host interactions (16) and spore transmission (15), it is of interest to determine the stability of this outer layer. To evaluate the stability of the exosporium, spore suspensions of strains R20291 and M120 were incubated for 6 months at room temperature in PBS and the abundance of two previously described markers of the exosporium (i.e., NTD-BclA1 and CdeC) (6, 10, 17) were measured by indirect immunofluorescence assay. Significant fluorescence signals of NTD-BclA1 and CdeC were readily detectable in freshly prepared spores of strains R20291 and M120 (Fig. 2A to D). However, after 6 months of storage at ambient temperature, there was a nearly 2-fold decrease in the fluorescence intensity of both markers in spores of both strains (Fig. 2A to D), suggesting that some exosporium protein markers are lost during long-term storage.

FIG 2.

Quantitative analysis of loss of exosporium proteins after 6 month of storage. (A, B) Microscopy analysis of purified C. difficile spores of strains R20291 and M120 by phase-contrast microscopy and immunofluorescence assay. Spores from fresh crops or from crops stored at room temperature for 6 months were treated with exosporium protein-specific antibodies. The panels are representative of the fluorescence intensity observed for fresh and 6-month-old spores of each strain. (C, D) Quantitative analysis of the fluorescence (Fl.) intensity in fresh and 6-month-old spores of strains R20291 and M120 for each exosporium protein-specific antibody was conducted as described in Materials and Methods. The values beside the symbols in the graphs represent the average ± the standard deviation of the fluorescence intensity under each spore condition (n = 150 spores analyzed for each age/strain condition). The data shown are from one experiment that is representative of three independent experiments.

Effect of long-term storage on the ultrastructure of C. difficile spores.

To gain insight into how long-term storage could affect the ultrastructure of the thick- and thin-exosporium morphotypes, new and 6-month-old spores were analyzed by TEM. Representative micrographs of new (i.e., 0-month-old) and 6-month-old spores of strains R20291 and M120 are shown in Fig. 3. As expected, new spores of R20291 and M120 exhibit a mixed population of thick- and thin-exosporium spores (Fig. 3), which is consistent with the aforementioned results (Fig. 1A and B). Analysis of 6-month-old spores of strains R20291 and M120 reveals that although thick-exosporium spores were detectable in proportions similar to those seen before (Fig. 1A and B), the integrity of their exosporium was considerably affected compared to that of freshly prepared spores (Fig. 3). Six-month-old R20291 spores of both exosporium morphotypes had hairlike extensions, but these were less abundant than in newly prepared spores (Fig. 3). Ultrastructural analysis of 6-month-old thick-exosporium R20291 spores revealed that much of the electron-dense material had been peeled off during storage (Fig. 3). Similarly, 6-month-old M120 spores that appeared to be of the thick-exosporium morphotype had remnants of the electron-dense material of the exosporium layer (Fig. 3). Remarkably, after 6 months of storage at room temperature, thin-exosporium spores of both strains lacked the hairlike projections of freshly prepared spores (Fig. 3). Collectively, these results demonstrate that incubation of C. difficile spores of both ultrastructural morphotypes for 6 months affects the exosporium's integrity.

FIG 3.

TEM of C. difficile spores incubated at room temperature for 6 months. C. difficile spores of strains R20291 and M120 were incubated at room temperature (23°C) for 6 months in PBS. Six-month-old spores and newly prepared spore suspensions were prepared and analyzed by TEM as described in Materials and Methods. Representative spores are shown. Scale bar, 200 nm.

Presence of exosporium layer in C. difficile spores from biofilms.

Several studies have demonstrated that during biofilm development (18–20), C. difficile spores are also formed, yet whether all spores form an exosporium layer under these conditions remains unclear. In this context, to gain more insight into the variability of the exosporium of C. difficile spores formed during biofilm development, we used CdeC as an exosporium marker (6) to detect the presence/absence of the exosporium layer in C. difficile strain R20291 and M120 spores during in vitro biofilm formation. First, as a control, we performed immunofluorescence analysis of unwashed sporulating cultures and purified spores. R20291 and M120 spores that did not immunoreact with anti-CdeC antibodies were classified as CdeCb negative (Fig. 4A and B), suggesting that they might have either a thin-exosporium phenotype or completely lack the exosporium layer, while those that gave an immunofluorescence signal above the background level were classified as CdeC positive and presumably have a thick-exosporium phenotype (Fig. 4A and B). To evaluate if these immunofluorescence phenotypes (i.e., CdeC positive and CdeC negative) were also detectable in spores formed during biofilm formation, 5-day-old biofilm biomasses with OD₅₇₀s of 6 to 8 developed in BHIS and TS medium were analyzed by immunofluorescence assay (Fig. 4A and B). No biofilm was detectable after 24 h of incubation (data not shown). Immunofluorescence intensity was quantitative for unwashed, pure spores as well as for spores formed during biofilm development. Immunoreactive CdeC-positive spores had a fluorescence intensity against anti-CdeC antibodies 3- to 5-fold higher than that of nonimmunoreactive, CdeC-negative spores (Fig. 4C and D). These observations indicate that, regardless of the method of preparation (i.e., agar plates or broth in culture wells), C. difficile spores form two spore populations (i.e., CdeC negative and CdeC positive) that can be quantitatively identified.

FIG 4.

Detection of the exosporium marker CdeC in spores from biofilm. (A, B) C. difficile spores of strains R20291 and M120 from 5-day-old BHIS and TS biofilm cultures were analyzed by immunofluorescence assay with anti-CdeC antibodies as described in Materials and Methods. Unwashed and purified spore cultures from TY agar plates were included as a control. (C, D) Quantitative analysis of the fluorescence (Fl.) intensity in C. difficile spores with and without an immunofluorescence signal against anti-CdeC antibodies. The values beside the symbols in the graphs represent the average ± the standard error of the fluorescence intensity for each type of spore. The data shown are from one experiment that is representative of three independent experiments. Asterisks denote statistically significant differences (**, P < 0.001) between CdeC (+) and CdeC (−) spores under each spore condition, as indicated by horizontal brackets. (E, F) The percentages of spores with (gray bars) and without (black bars) a detectable fluorescence signal were calculated for strains R20291 (E) and M120 (F). Experiments were performed two independent times, and more than 300 spores were analyzed under each experimental condition. Values are representative of three experimental conditions, and asterisks denote statistically significant differences (*, P < 0.01).

Next, using the aforementioned approach, we proceeded to determine the fractions of spores under each culture condition that were CdeC positive and CdeC negative. Immunofluorescence analysis demonstrated that while similar levels (i.e., 80%) of CdeC-positive M120 spores were detected in unwashed and purified spores, the levels of CdeC-positive R20291 spores increased from 44 to 87% in washed and purified spores, respectively (Fig. 4E and F). Similarly, high levels of CdeC-positive spores were observed in M120 spores obtained from BHIS and TS biofilms (i.e., 75 and 86%, respectively) (Fig. 4F). Notably, while a high percentage (i.e., 87%) of R20291 spores obtained from TS biofilms were CdeC positive, only 59% of R20291 spores obtained from BHIS biofilms were CdeC positive (Fig. 4E), contrasting with the high levels of M120 CdeC-positive spores obtained under similar conditions (Fig. 4F). These results suggest that (i) most of the C. difficile spores formed under biofilm conditions are likely to have an exosporium and (ii) the proportion of exosporium-positive spores depends on the culture medium used to induce biofilm formation, as well as on the strain tested.

DISCUSSION

C. difficile spores have proven to be essential for the initiation, persistence, and dissemination of CDI (3, 4, 14). In this context, the interactions of C. difficile spores with the host and environmental surfaces might be essential for spore persistence and transmission, respectively. The composition and mechanism of assembly of the external surface of C. difficile spores have recently started to be elucidated (6, 10). In this work, we have addressed several unexplored features of the outermost exosporium layer of C. difficile spores, specifically, those related to (i) the ultrastructural variability of the layer between strains and within spores of the same strain, (ii) the stability of the exosporium during long-term storage, and (iii) the presence of the exosporium layer in C. difficile spores formed during biofilm development.

A first conclusion from this work is that two ultrastructural morphotypes of the exosporium layer (i.e., a thick exosporium and a thin exosporium) were detectable in a spore population of C. difficile of a single strain. Notably, the remaining spore layers varied in thickness between spores of these two exosporium morphotypes, suggesting that slightly different regulatory mechanisms might underlie the formation of each morphotype. These observations raise the obvious question of whether these different morphotypes appear randomly or are attributable to the differential expression of a protein(s) involved in the assembly of this layer. It is also unclear whether these two morphotypes might have (i) differential virulence phenotypes that may affect the outcome of infection or (ii) differential adherence to inert surfaces or hand skin that might be associated with hospital transmission and/or persistence.

Another contribution of this work is that C. difficile spores of most of the strains analyzed have an exosporium with hairlike projections. Most of the previous work on the C. difficile exosporium has been conducted with a laboratory strain (i.e., 630) that lacks hairlike projections (6–9, 12, 16, 17, 21, 22). Indeed, 630 spores are substantially different from spores of epidemiologically relevant strains reported here and elsewhere (10, 14). In B. anthracis, these hairlike projections are attributed to the presence of the collagen-like glycoprotein BclA, which is homogeneously distributed on the spore surface (23, 24). Notably, although all three C. difficile BclA homologues (i.e., BclA1, BclA2, and BclA3) are expressed during sporulation and localize to the surface of 630 spores (17), TEM of 630 spores indicated the absence of hairlike projections, which are present in epidemiologically relevant strains (i.e., R20291). These observations suggest that the assembly mechanism of the collagen-like BclA glycoproteins in the exosporium of epidemiologically relevant spores might be substantially different from that in 630 spores. These main ultrastructural differences in the exosporium layer might also have in vivo implications (i.e., persistence in the host or on hospital surfaces). For example, a recent study (16) demonstrated that BclA1 affects the susceptibility and colonization of mice with C. difficile strain 630. However, it is noteworthy that unlike the genome of strain 630, those of most C. difficile isolates that have been sequenced have a bclA1 pseudogene because of the appearance of an early stop codon in the N-terminal domain (17), raising the hypothesis that BclA2 and BclA3 might have roles in infections with epidemiologically relevant strains.

Previously, we have shown that spores stored for 6 months at room temperature were viable and exhibited a germination efficiency similar to or higher than that of freshly prepared spores (25). Here we demonstrate that 6-month-old spores also retain most of their exosporium material, yet the apparent loss of the hairlike extensions in the 6-month-old spores analyzed might have implications for spore adherence to host mucosal surfaces, inert surfaces, and hand skin that remain to be elucidated.

During in vivo infection, C. difficile initiates a sporulation cycle that contributes to C. difficile spore persistence and dissemination (4). Recently, Semenyuk et al. demonstrated that C. difficile forms biofilm communities in vivo in a mouse model of infection (26). In vitro studies have demonstrated that C. difficile forms spores during biofilm development (20); yet, as demonstrated in that work, the spores formed in these biofilms lack the exosporium layer. Here, we provide evidence that most biofilm-formed spores of strains R20291 and M120 are positive for the exosporium marker CdeC and therefore are likely to have an exosporium layer and that the proportion of the presence/absence of anti-CdeC immunofluorescence signal was dependent on the strain and biofilm culture condition; therefore, it is conceivable that the lack of an exosporium observed by Semenyuk et al. in spores formed during biofilm development could be attributable to specific experimental conditions that affected the formation of the exosporium layer of that particular strain (20). Considering these observations, it is likely that the composition of the spore outer layer (i.e., exosporium) might differ markedly, depending on whether spores are formed in the planktonic state or inside a biofilm.

In summary, these observations highlight the need to perform further studies to address whether the variability in the exosporium layer in terms of the presence/absence of hairlike extensions and thin- and thick-exosporium morphotypes affects virulence, persistence in the host, and transmission and persistence in the hospital environment. Spore removal strategies based on exosporium proteins as targets need to consider the variability in the exosporium layer within and between strains.