Adherence of Clostridium perfringens spores to human intestinal epithelial Caco-2 cells

Adherence of Clostridium perfringens spores to human intestinal epithelial Caco-2 cells Abstract Clostridium perfringens is a gram-positive, spore-forming bacillus, and is a causative agent of foodborne infection, antibiotic-associated diarrhoea and sporadic diarrhoea in humans. In cases of antibiotic-associated and sporadic diarrhoea, C. perfringens colonises the intestine, proliferates and causes disease. However, bacterial colonisation of the intestine is not considered necessary in the pathogenesis of foodborne illness, because such pathogenesis can be explained by anchorage-independent production of diarrhoeic toxin by the bacterium in the intestine. In this study, we used an in vitro adherence assay to examine the adherence of C. perfringens spores to human intestinal Caco-2 cells. Adherence of spores from isolates of foodborne illness and nosocomial infection was observed within 15 min, and plateaued 60 min after inoculation. Electron microscopy revealed a tight association of spores with the surface of Caco-2 cells. The adherence of vegetative cells could not be confirmed by the same method, however. These results suggest that C. perfringens spores may adhere to intestinal epithelial cells in vivo, although its biological significance remains to be determined. Clostridium perfringens, foodborne illness, spore, nosocomial infection, sporadic diarrhoea INTRODUCTION Clostridium perfringens is an anaerobic, gram-positive and spore-forming bacterium, and occurs in ubiquitous habitats such as sewage, soil, plant surfaces, faeces and the normal intestinal microbiota of humans and animals. It is an important pathogen that causes a variety of human and animal diseases, ranging from myonecrotic to enteric infections (Songer 1996; Petit, Gibert and Popoff 1999). It reportedly produces up to 17 toxins; of these, four typing toxins—alpha, beta, epsilon and iota—are used to classify the species into five toxinotypes: A, B, C, D and E (Li et al.2013). Type A strains, which of the four typing toxins produce only the alpha toxin, can be isolated from human clinical specimens of foodborne illness (FBI), antibiotic-associated diarrhoea (AAD) and sporadic and nosocomial diarrhoea (SND) (Tompkins et al.1999; Watanabe, Hitomi and Sawahata 2008; Miyamoto, Li and McClane 2012; Polage, Solnick and Cohen 2012). A common feature of these isolates is the production of the nontyping toxin, C. perfringens enterotoxin, which is an important virulence factor (Li et al.2013). Despite this common feature, the pathogenesis of FBI and two other diarrhoeal diseases (AAD and SND) apparently differs in terms of colonisation behaviour. AAD develops via antibiotic-induced dysbiosis of gut microbiota, when endogenous C. perfringens is allowed to overcome and outgrow its competition (Polage, Solnick and Cohen 2012). This indicates that C. perfringens colonises the intestinal tract, where it persists as a component of the commensal microbiota for many years; in fact, many investigators have reported such bacterial colonisation in healthy human intestines (Carman et al.2008; Nagpal et al.2015). The aetiology of SND is not fully understood, but may include oral infection from spores in the hospital or community. The ingested spores can colonise the intestine, which is followed by bacterial growth and the production of CPE, which causes diarrhoea. In contrast, FBI strains do not seem to colonise or bind directly to the intestinal surface, because the development of the acute disease can be explained by the transformation of vegetative cells into spores in the intestine, without the need for physical binding to the tissue surface. No direct demonstration of the necessity of colonisation by FBI strains has been reported, mainly because there are no animal infection models of the gastroenteritis induced by type A C. perfringens currently available (Uzal and McClane 2012). In this study, we investigated whether human isolates of C. perfringens adhere to human intestinal epithelial cells in vitro. We used the intestinal cell line Caco-2, two different FBI strains and an SND strain in cell adherence assays, and the results indicate that spores from the FBI and SND strains adhere tightly to Caco-2 cells. An electron microscopy investigation supports our conclusion, and provides additional interesting observations of the interaction between C. perfringens spores and intestinal epithelial cells. To the best of our knowledge, this is the first report to demonstrate visually that C. perfringens spores, especially those from an SND strain, adhere to human intestinal cells in vitro. MATERIALS AND METHODS Bacterial strains and spore preparation Clostridium perfringens type A strain SM101 is a derivative of FBI strain NCTC8798. NCTC8239, another FBI strain, was purchased from the National Collection of Type Cultures. W10030 is an isolate from a sporadic gastrointestinal infection that occurred in geriatric hospital wards in 2006, containing an IS1151-associated cpe gene on a 75-kb plasmid (Kobayashi et al.2009). In this hospital, several sporadic cases of diarrhoea were recognised in 1992–1993, 2006 and 2015–2016. Molecular epidemiological analyses revealed that the isolates seemed to have been derived from a single clonal lineage, or were closely related to each other (Monma et al., in preparation), suggesting that the hospital may have been colonised by these related strains for a long time, and that these strains caused the repeated incidents of SND. Based on these characteristics, we used strain W10030 as a typical SND strain. The culture stocks of the bacterial strains were prepared as previously described (Yasugi et al.2015), except that the bacterial culture was heated either at 70°C (SM101), 75°C (NCTC8239) or 60°C (W10030). The bacterial cells were heated, separated into small aliquots and stored at −80°C until use. Spores were prepared as described previously (Sacks and Alderton 1961). Briefly, bacterial cultures re-grown in Fluid Thioglycollate (FTG) medium were inoculated into 40 ml of Duncan-Strong medium (DS medium; Duncan and Strong 1968), followed by incubation for 22–26 h at 37°C. The cultures were centrifuged at 7500× g for 15 min at 25°C, and the supernatant was then removed. Sterilised ultrapure water (10 ml) was added to the precipitate, followed by 3.41 ml of 3 M potassium phosphate buffer (pH 7.1) and 1.118 g of solid polyethylene glycol 4000 (SERVA). After gentle mixing until complete dissolution of the polyethylene glycol, the mixture was left standing for more than 30 min to ensure two-phase partitioning. Bacterial spores in the upper phase were recovered by centrifugation at 15 000× g for 10 min at 4°C, and washed thrice with ultrapure water. The resultant spore suspension was subjected to another round of partitioning to remove contaminated vegetative cells. Colony forming units (CFU) of the final preparations were enumerated after heat treatments at 70°C (SM101), 75°C (NCTC8239) or 60°C (W10030). The dilution of the heated samples was performed in Brain Heart Infusion (BHI) broth with 0.1% (v/v) Triton X-100, and the CFUs were counted on BHI agar incubated anaerobically for 24 h. Adherence assay Human colonic epithelial Caco-2 cells were maintained in Dulbecco's Modified Eagle's Medium (DMEM, Sigma) supplemented with 10% foetal bovine serum (FBS). For the co-culture experiments, cells were plated in 35-mm culture dishes and incubated for 4 days (final density approximately 1.4 × 106 cells/dish). The cells were washed once with phosphate-buffered saline (PBS) and incubated in 2 ml of glucose-deficient DMEM (DMEM(−); Thermo Fisher Scientific, Cat. No. 11966025) supplemented with 0.4% soluble starch and 50 μM deoxycholic acid (Yasugi et al.2015). The bacteria from the culture stock were grown in FTG medium (3 ml) for 11–13 h at 37°C, collected and washed twice with PBS by centrifugation and resuspended in 3 ml of PBS (final concentration, 1.0 × 106 CFU/ml). Aliquots of 200 μl containing vegetative bacterial cells were inoculated into Caco-2 culture dishes and incubated for the indicated period. The total number of bacterial cells in the culture supernatant was determined by enumerating the CFU, as described above. When determining the number of heat-resistant spores, the samples were heat-treated prior to dilution. Detection thresholds under these conditions were 20 CFU/ml. Bacterial vegetative cells and spores that adhered on Caco-2 cells were quantified after ultrasonic homogenisation of washed Caco-2 cells. The washed cells, recovered in PBS by pipetting, were completely disrupted by ultrasonication for 10 s using an Ultrasonic Disruptor (TOMY SEIKO, Co., Ltd) at a low oscillation output level (0.1), with micro-tip TP-040. This weak ultrasonication did not affect the viability of the vegetative bacteria or spores, but resulted in the disruption of the Caco-2 cells into fine fragments (<1 μm; data not shown). To examine the adherence of isolated vegetative bacteria or spores, Caco-2 cells were plated in 35-mm dishes at 3.0 × 106 cells/dish, and incubated in DMEM supplemented with 5% FBS for 20 h. Before exposure to the bacterial cells, the medium was changed to DMEM(−) supplemented with 0.4% soluble starch and 50 μM deoxycholic acid. The Caco-2 dish (2.0 × 106 CFU/dish) was inoculated with the spore suspension and incubated at 37°C for the indicated period. The cells were washed thrice with 3 ml of PBS, and detached by pipetting in 0.1% Triton X-100 in PBS. The number of spores was enumerated by counting the CFU after heat treatment, as described above. The detection threshold of this procedure was 20 CFU/ml. The vegetative bacteria were prepared by culturing in FTG medium for 11–13 h, and were washed twice with 10 ml of PBS. The bacteria suspended in DMEM(−) supplemented with 0.4% soluble starch and 50 μM deoxycholic acid were inoculated into the Caco-2 dishes at a multiplicity of infection (MOI) of 5.4, and incubated at 37°C for 2 h. After incubation, the cells were washed 10 times with 3 ml of PBS, and observed under phase-contrast microscopy. In all adherence assays, incubations were performed under a 5% CO2 atmosphere. Electron microscopy Caco-2 cells were cultured in 35-mm dishes in DMEM supplemented with 5% FBS for 20–24 h (final density 4.5 × 106 cells/dish), and treated with vegetative bacteria or spores for 60 min at 2.7 MOI and 12.9 MOI, respectively. For scanning electron microscopy (SEM), Caco-2 cells were cultured on a (γ-aminopropyl) trimethoxysilane-coated cover glass (4 × 8 mm, Matsunami Glass) placed in the culture dish. The co-cultured cells were doubly fixed with 2% glutaraldehyde and 2% OsO4, both in 0.05 M cacodylate buffer (pH 7.2), and dehydrated with serially graded ethanol. Dehydrated samples were subjected to critical-point drying, coated with platinum at 5-nm thick using an ion sputter (E-1030, Hitachi), and observed under a scanning electron microscope (S-5000, Hitachi). For transmission electron microscopy (TEM), 0.05% ruthenium red was added in both fixatives. The samples were embedded in plain resin (Nissin EM), and ultrathin sections were made and observed under an electron microscope (H-7100, Hitachi). RESULTS We initiated our study to examine the adherence of C. perfringens to Caco-2 cells, using our co-culture system. When Caco-2 cells were co-cultured with the vegetative C. perfringens strain SM101 under spore-forming conditions (Yasugi et al.2015), the total number of bacterial cells (i.e. vegetative bacteria plus spores) in the culture supernatant increased over time (Fig. 1). Heat-resistant spores were also detected in the supernatant at 8 h, and continued to increase until 12 h. Notably, the Caco-2 cell monolayer began to detach from the substratum after 12 h, probably via the action of the enterotoxin produced by C. perfringens during sporulation (Yasugi et al.2015). As a result, we were unable to quantify the cell-adhered bacteria or spores thereafter (data not shown). Under these conditions, bacteria and spores appeared to adhere to Caco-2 cells during a 12-h incubation (Fig. 1, circles); however, adhesion levels were only 1.5 (total bacteria) and 0.1 (spores) CFU/100 Caco-2 cells, i.e. only 15 vegetative bacteria and one spore attached per thousand of Caco-2 cells. Figure 1. View largeDownload slide Detection of C. perfringens associated with Caco-2 cells in co-cultured experiments. Clostridium perfringens strain SM101 was co-cultured with Caco-2 cells for the indicated period. The total number of bacterial cells (dotted lines) or the number of spores (bold lines) in the culture supernatants (squares) or on Caco-2 cells (circles) was determined as described in the Materials and Methods section. Averages and standard deviations of three independent experiments are shown. Figure 1. View largeDownload slide Detection of C. perfringens associated with Caco-2 cells in co-cultured experiments. Clostridium perfringens strain SM101 was co-cultured with Caco-2 cells for the indicated period. The total number of bacterial cells (dotted lines) or the number of spores (bold lines) in the culture supernatants (squares) or on Caco-2 cells (circles) was determined as described in the Materials and Methods section. Averages and standard deviations of three independent experiments are shown. We then attempted to confirm the physical binding of the bacterial cells to Caco-2 cells by visualising the adhered bacteria using phase contrast and electron microscopy. We observed that a certain proportion of the vegetative bacteria remained attached to the Caco-2 cell monolayer in the adherence assay, even after vigorous washing (Fig. 2A). Each of the bacterial cells, however, appeared to be oscillating according to Brownian motion, suggesting that the observed association between the bacteria and Caco-2 cells may have been an artefact, or that the adherence may be mediated by thin fibrous extensions such as pili (data not shown). Similar results were obtained with strain NCTC8239 (data not shown). To further examine the possible adherence, the Caco-2 cells were observed using SEM. We found that very few SM101 cells attached to Caco-2 cell surfaces (Fig. 2B and C). Furthermore, TEM provided no evidence for direct contact between SM101 cells and Caco-2 cell surfaces (Fig. 2D and E), indicating that vegetative SM101 bacteria, and probably also NCTC8239 bacteria, do not adhere tightly to Caco-2 cells under these conditions. Figure 2. View largeDownload slide Assessment of adhesion of vegetative C. perfringens bacteria to Caco-2 cells. (A) A phase contrast micrograph of Caco-2 cells treated with vegetative strain-SM101 C. perfringens bacteria after 2 h of incubation (5.4 MOI). Scale bars, 20 μm. (B). (C) Scanning electron micrographs of Caco-2 cells treated with vegetative strain-SM101 C. perfringens bacteria for 60 min at 2.7 MOI. Bacterial cells are indicated with arrows. Scale bars, 10 μm. (D). (E) Transmission electron micrographs of Caco-2 cells treated with vegetative strain-SM101 C. perfringens bacteria for 60 min at 2.7 MOI. Scale bars, 5 μm (D) and 1 μm (E). Figure 2. View largeDownload slide Assessment of adhesion of vegetative C. perfringens bacteria to Caco-2 cells. (A) A phase contrast micrograph of Caco-2 cells treated with vegetative strain-SM101 C. perfringens bacteria after 2 h of incubation (5.4 MOI). Scale bars, 20 μm. (B). (C) Scanning electron micrographs of Caco-2 cells treated with vegetative strain-SM101 C. perfringens bacteria for 60 min at 2.7 MOI. Bacterial cells are indicated with arrows. Scale bars, 10 μm. (D). (E) Transmission electron micrographs of Caco-2 cells treated with vegetative strain-SM101 C. perfringens bacteria for 60 min at 2.7 MOI. Scale bars, 5 μm (D) and 1 μm (E). Next, we examined the adherence of isolated C. perfringens spores to Caco-2 cells. Heat-resistant spores associated with Caco-2 cells became evident after 15 min of incubation, and had reached close to their maximum abundance by 60 min (Fig. 3). The numbers of adhered spores differed according to the bacterial strain; NCTC8239 spores exhibited the greatest adhesion to Caco-2 cells, followed by SM101, and then W10030. Similarly, SEM revealed a number of spores attached to the Caco-2 cell monolayer when treated with isolated SM101 or W10030 spores for 60 min (Fig. 4A–D). Furthermore, TEM revealed evidence for direct contact between spores from both strains and Caco-2 cell surfaces (Fig. 4E and F). We also found numerous examples of microvilli-like structures on the Caco-2 cells, to which the spore surfaces were attached, in both SEM and TEM (Fig. 4B and D), confirming that SM101 and W10030 spores adhere to Caco-2 cells under these conditions. Figure 3. View largeDownload slide Quantification of C. perfringens spores adhering to Caco-2 cells. Clostridium perfringens spores (isolated spore suspension) from strains (A) SM101, (B) NCTC8239 and (C) W10030 were inoculated on Caco-2 cells at 0.44 MOI, and incubated for the indicated period. The number of heat-resistant spores associated with Caco-2 cells was determined as described in the Materials and Methods section. Averages and standard deviations of three experiments are shown. Figure 3. View largeDownload slide Quantification of C. perfringens spores adhering to Caco-2 cells. Clostridium perfringens spores (isolated spore suspension) from strains (A) SM101, (B) NCTC8239 and (C) W10030 were inoculated on Caco-2 cells at 0.44 MOI, and incubated for the indicated period. The number of heat-resistant spores associated with Caco-2 cells was determined as described in the Materials and Methods section. Averages and standard deviations of three experiments are shown. Figure 4. View largeDownload slide Assessment of adhesion of C. perfringens spores to Caco-2 cells. (A–D) Scanning electron micrographs of Caco-2 cells treated with isolated C. perfringens spores. Spores (5.8 × 107 spores) from strains (A, B) SM101 and (C, D) W10030 were inoculated onto Caco-2 cells (4.5 × 106 cells) and incubated for 60 min at 12.9 MOI. Attachment sites of microvilli-like structures to spores are indicated with arrows. Scale bars, 10 μm (A, C) and 0.5 μm (B, D). (E, F) Transmission electron micrographs of Caco-2 cells treated with isolated C. perfringens spores. Spores from strains (E) SM101 and (F) W10030 were incubated with Caco-2 cells at 12.9 MOI as described above. Attachment sites of microvilli-like structures to spores are indicated with arrows. Scale bars, 0.5 μm. Figure 4. View largeDownload slide Assessment of adhesion of C. perfringens spores to Caco-2 cells. (A–D) Scanning electron micrographs of Caco-2 cells treated with isolated C. perfringens spores. Spores (5.8 × 107 spores) from strains (A, B) SM101 and (C, D) W10030 were inoculated onto Caco-2 cells (4.5 × 106 cells) and incubated for 60 min at 12.9 MOI. Attachment sites of microvilli-like structures to spores are indicated with arrows. Scale bars, 10 μm (A, C) and 0.5 μm (B, D). (E, F) Transmission electron micrographs of Caco-2 cells treated with isolated C. perfringens spores. Spores from strains (E) SM101 and (F) W10030 were incubated with Caco-2 cells at 12.9 MOI as described above. Attachment sites of microvilli-like structures to spores are indicated with arrows. Scale bars, 0.5 μm. DISCUSSION In this study, we evaluated the adherence of two strains of type A C. perfringens to human intestinal Caco-2 cells. First, we examined the ability of vegetative bacterial cells to adhere to Caco-2 cells in co-culture experiments, and observed a certain level of time-dependent adherence (Fig. 1). Several investigators have previously reported the binding of vegetative C. perfringens to host cells or tissues. In particular, the association of SM101 and NCTC8239 with Caco-2 cells has been examined in relation to the bacterial sialidase, NanI (Li and McClane 2014). The authors concluded that both strains show only weak adherence activity, because of their NanI-negative phenotype. However, their study did not provide morphological evidence for adherence, so we tried to confirm it using electron microscopy. We could not find visible evidence of the bacterial adherence, probably because the adherence was so weak that the bacteria could not be kept adhered during canonical fixation and drying treatments. In addition, several investigators have reported the binding of other human or animal C. perfringens strains to intestinal cells or extracellular matrix proteins, such as collagens and fibronectin, and have proposed the importance of the bacterial proteins encoded in the genes cna (Redondo et al.2015; Wade et al.2016) and fbp (Hitsumoto et al.2014) in such binding. However, these genes are not present in the SM101 genome (GenBank accession No. CP000312.1), which could explain why this strain did not show tight adhesion, if any, to Caco-2 cells under our experimental conditions. On the other hand, our electron microscopy confirmed the adherence of C. perfringens spores to Caco-2 cells (Figs 3 and 4). Adherence of SM101 spores to Caco-2 cells has been reported elsewhere (Paredes-Sabja and Sarker 2012), but morphological evidence for this adhesion had not previously been presented. Our observations clearly indicate physical association of the bacterial spores with the surface of Caco-2 cells (Fig. 4). Interestingly, the tips and lateral surfaces of numerous microvilli-like structures on the Caco-2 cells appear to attach to the surface of the spores (Fig. 4B and D–F). Whether these unique morphological features reflect an important biological phenomenon should be investigated in further studies. Binding of C. difficile spores via microvilli expressed on Caco-2 cells has also been reported; this binding appeared to be mediated by hair-like extensions of the exosporium-like layer, and the attachment sites appeared to be adjacent to either the tip or the lateral surface of the microvilli (Paredes-Sabja and Sarker 2012; Mora-Uribe et al.2016). These findings may suggest that the adhesins on the spore surface are different for C. difficile and C. perfringens. Notably, in the co-culture experiments, only 1% of spores adhered to Caco-2 cells (Fig. 1), whereas 7.9%, 45% and 2.7% of the input of isolated spores (SM101, NCTC8239 and W10030, respectively) did so in the adherence assays (Fig. 3). One possible explanation for the different adherence efficiency is the presence of concomitant materials derived from the bacteria. In the co-culture dish, it is likely that a number of viable and dead vegetative cells, bacterial cells debris, immature spores and probably heat-sensitive germinating spores are all present. These bacteria-related substances may have hampered the adherence of the heat-resistant spores to the Caco-2 cells. Additionally, the numbers of Caco-2-associated spores were relatively lower (1–20 spores/100 Caco-2 cells) when enumerated via CFU counting (Fig. 3), even though several dozen spores per cell were visualised on the surface of Caco-2 cells by electron microscopy (Fig. 4A). We consider that differences in MOIs in these experiments (10 times lower in the CFU study) do not directly explain the high level of adherence observed in the SEM study. Thus, another explanation is that many spores become undetectable by CFU assay when associated with Caco-2 cells; association with Caco-2 cells may render the attached spores insensitive or unresponsive to germinants, via an unknown mechanism. A previous finding, showing that C. difficile spores attached to the surface of Caco-2 cells exhibited different germination abilities on different co-cultured cell lines, may support this idea (Paredes-Sabja and Sarker 2011). The mechanism and molecules that mediate the binding and interaction of C. perfringens spores with Caco-2 cells remain unclear at this stage. Hydrophobic interactions between spore surface proteins and Caco-2 surface proteins could be one of the binding mechanisms. Hydrophobic interaction has been proposed as a mechanism governing the interactions of spores from clostridial species with cultured mammalian cells (Andersson, Granum and Rönnera 1998; Tauveron et al.2006; Thwaite et al.2009; Paredes-Sabja and Sarker 2012). Based on TEM, some authors have suggested that the exosporium layer of the outermost surface of the bacterial spores is responsible for these interactions (Russell et al.2007). We also conducted TEM in the present study, and found that none of the strains used in this study possesses an exosporium layer (Fig. 5), which suggests that the mode of adherence in these strains is not exosporium mediated. Indeed, Paredes-Sabja and Sarker (2012) have reported that hydrophobicity is not particularly important for the interaction between spores and Caco-2 cells in the case of some C. perfringens strains, including SM101. The factors governing cell-to-cell interactions, such as the physical interactive force, spore surface adhesins and adhesion receptors, should be further investigated to unveil the biological significance of this phenomenon. Figure 5. View largeDownload slide Structure of spores from strains SM101, NCTC8239 and W10030. Bacterial strains were precultured in FTG medium for 11–13 h and incubated in DS medium for 24 h at 37°C. Total bacterial cells were collected by centrifugation at 7500× g for 15 min, followed by the removal of the supernatant and the addition of 2% glutaraldehyde in 0.05 M Ruthenium Red. The samples were then prepared for TEM as described in the Materials and Methods section. TEM micrographs of (A) SM101, (B) NCTC8239 and (C) W10030 are shown. Three areas representing coat layer, cortex and core are indicated. Scale bar, 500 nm. Figure 5. View largeDownload slide Structure of spores from strains SM101, NCTC8239 and W10030. Bacterial strains were precultured in FTG medium for 11–13 h and incubated in DS medium for 24 h at 37°C. Total bacterial cells were collected by centrifugation at 7500× g for 15 min, followed by the removal of the supernatant and the addition of 2% glutaraldehyde in 0.05 M Ruthenium Red. The samples were then prepared for TEM as described in the Materials and Methods section. TEM micrographs of (A) SM101, (B) NCTC8239 and (C) W10030 are shown. Three areas representing coat layer, cortex and core are indicated. Scale bar, 500 nm. Molecular biological and epidemiological studies, as well as animal model experiments, have demonstrated that spores play a crucial role in the transmission of C. difficile during infection (Abt, McKenney and Pamer 2016). A similar pathological mechanism is suspected in case of AAD associated with C. perfringens infection (Kokai-Kun et al.1994; Carman 1997). It is therefore important to examine the ability of AAD- and SND-associated C. perfringens strains to adhere to human cultured cells or intestinal tissue. Unlike for C. difficile, an animal model for C. perfringens-associated diarrhoea using bacterial cells or spores as an inoculum has not yet been established (Uzal and McClane 2012). Thus, further investigations to establish such a model are now warranted. The FBI caused by C. perfringens is an acute disease that usually lasts for less than 24 h (Li et al.2013). Thus, it seems that the spore colonisation of the intestine is not involved in the development of the disease. Our observation of the adherence of SM101 spores to Caco-2 cells suggests that some spores may remain attached in the patients’ intestine after FBI has developed, causing these patients to become potential reservoirs once they have recovered from the disease. This may also cause AAD following antibiotic treatment, or these patients may become an infection source of SND in a hospital or community. An epidemiological examination using several clinical strains from different types of diseases within a selected community will be needed to investigate this possibility. Acknowledgements We thank Dr Tohru Shimizu (Kanazawa University) for providing bacterial strain SM101. We also thank Yoshihiko Fujioka (Department of Microbiology and Infection Control, Osaka Medical College) for his technical assistance. FUNDING This work was supported by JSPS KAKENHI (15K14858 to MM) and a grant from the Research Project to Improved Food Safety and Animal Health (Ministry of Agriculture, Forestry, and Fisheries, Japan, 2013–2017 FY to MM). Conflict of interest. None declared. REFERENCES Abt MC, McKenney PT, Pamer EG. Clostridium difficile colitis: pathogenesis and host defence. Nat Rev Microbiol  2016; 14: 609– 20. Google Scholar CrossRef Search ADS PubMed  Andersson A, Granum PE, Rönner U. The adhesion of Bacillus cereus spores to epithelial cells might be an additional virulence mechanism. Int J Food Microbiol  1998; 39: 93– 9. Google Scholar CrossRef Search ADS PubMed  Carman RJ. Clostridium perfringens in spontaneous and antibiotic-associated diarrhoea of man and other animals. Rev Med Microbiol  1997; 8: S43– 5. Google Scholar CrossRef Search ADS   Carman RJ, Sayeed S, Li J et al.   Clostridium perfringens toxin genotypes in the feces of healthy North Americans. Anaerobe  2008; 14: 102– 8. Google Scholar CrossRef Search ADS PubMed  Duncan CL, Strong DH. Improved medium for sporulation of Clostridium perfringens. Appl Microbiol  1968; 16: 82– 9. Google Scholar PubMed  Hitsumoto Y, Morita N, Yamazoe R et al.   Adhesive properties of Clostridium perfringens to extracellular matrix proteins collagens and fibronectin. Anaerobe  2014; 25: 67– 71. Google Scholar CrossRef Search ADS PubMed  Kobayashi S, Wada A, Shibasaki S et al.   Spread of a large plasmid carrying the cpe gene and the tcp locus amongst Clostridium perfringens isolates from nosocomial outbreaks and sporadic cases of gastroenteritis in a geriatric hospital. Epidemiol Infect  2009; 137: 108– 13. Google Scholar CrossRef Search ADS PubMed  Kokai-Kun JF, Songer JG, Czeczulin JR et al.   Comparison of Western immunoblots and gene detection assays for identification of potentially enterotoxigenic isolates of Clostridium perfringens. J Clin Microbiol  1994; 32: 2533– 9. Google Scholar PubMed  Li J, Adams V, Bannam TL et al.   Toxin plasmids of Clostridium perfringens. Microbiol Mol Biol Rev  2013; 77: 208– 33. Google Scholar CrossRef Search ADS PubMed  Li J, McClane BA. Contributions of NanI sialidase to Caco-2 cell adherence by Clostridium perfringens Type A and C strains causing human intestinal disease. Infect Immun  2014; 82: 4620– 30. Google Scholar CrossRef Search ADS PubMed  Miyamoto K, Li J, McClane BA. Enterotoxigenic Clostridium perfringens: detection and identification. Microbes Environ  2012; 27: 343– 9. Google Scholar CrossRef Search ADS PubMed  Mora-Uribe P, Miranda-Cárdenas C, Castro-Córdova P et al.   Characterization of the adherence of Clostridium difficile spores: the integrity of the outermost layer affects adherence properties of spores of the epidemic strain R20291 to components of the intestinal mucosa. Front Cell Infect Microbiol  2016; 6: 99. Google Scholar CrossRef Search ADS PubMed  Nagpal R, Ogata K, Tsuji H et al.   Sensitive quantification of Clostridium perfringens in human feces by quantitative real-time PCR targeting alpha-toxin and enterotoxin genes. BMC Microbiol  2015; 15: 219. Google Scholar CrossRef Search ADS PubMed  Paredes-Sabja D, Sarker MR. Germination response of spores of the pathogenic bacterium Clostridium perfringens and Clostridium difficile to cultured human epithelial cells. Anaerobe  2011; 17: 78– 84. Google Scholar CrossRef Search ADS PubMed  Paredes-Sabja D, Sarker MR. Adherence of Clostridium difficile spores to Caco-2 cells in culture. J Med Microbiol  2012; 61: 1208– 18. Google Scholar CrossRef Search ADS PubMed  Petit L, Gibert M, Popoff MR. Clostridium perfringens: toxinotype and genotype. Trends Microbiol  1999; 7: 104– 10. Google Scholar CrossRef Search ADS PubMed  Polage CR, Solnick JV, Cohen SH. Nosocomial diarrhea: evaluation and treatment of causes other than Clostridium difficile. Clin Infect Dis  2012; 57: 982– 9. Google Scholar CrossRef Search ADS   Redondo LM, Carrasco JM, Redondo EA et al. Clostridium perfringens type E virulence traits involved in gut colonization. PLoS One  2015; 10: e0121305. Google Scholar CrossRef Search ADS PubMed  Russell BH, Vasan R, Keene DR et al. Bacillus anthracis internalization by human fibroblasts and epithelial cells. Cell Microbiol  2007; 9: 1262– 74. Google Scholar CrossRef Search ADS PubMed  Sacks LE, Alderton G. Behavior of bacterial spores in aqueous polymer two-phase systems. J Bacteriol  1961; 82: 331– 41. Google Scholar PubMed  Songer JG. Clostridial enteric diseases of domestic animals. Clin Microbiol Rev  1996; 9: 216– 34. Google Scholar PubMed  Tauveron G, Slomianny C, Henry C et al.   Variability among Bacillus cereus strains in spore surface properties and influence on their ability to contaminate food surface equipment. Int J Food Microbiol  2006; 110: 254– 62. Google Scholar CrossRef Search ADS PubMed  Thwaite JE, Laws TR, Atkins TP et al.   Differential cell surface properties of vegetative Bacillus. Lett Appl Microbiol  2009; 48: 373– 8. Google Scholar CrossRef Search ADS PubMed  Tompkins DS, Hudson MJ, Smith HR et al.   A study of infectious intestinal disease in England: microbiological findings in cases and controls. Comm Dis Public Health  1999; 2: 108– 13. Uzal FA, McClane BA. Animal models to study the pathogenesis of enterotoxigenic Clostridium perfringens infections. Microbes Infect  2012; 14: 1009– 16. Google Scholar CrossRef Search ADS PubMed  Wade B, Keyburn AL, Haring V et al.   The adherent abilities of Clostridium perfringens strains are critical for the pathogenesis of avian necrotic enteritis. Vet Microbiol  2016; 197: 53– 61. Google Scholar CrossRef Search ADS PubMed  Watanabe M, Hitomi S, Sawahata T. Nosocomial diarrhea caused by Clostridium perfringens in the Tsukuba-Tsuchiura district, Japan. J Infect Chemother  2008; 14: 228– 31. Google Scholar CrossRef Search ADS PubMed  Yasugi M, Sugahara Y, Hoshi K et al.   In vitro cytotoxicity induced by Clostridium perfringens isolate carrying a chromosomal cpe gene is exclusively dependent on sporulation and enterotoxin production. Microb Pathogenesis  2015; 85: 1– 10. Google Scholar CrossRef Search ADS   © FEMS 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png FEMS Microbiology Letters Oxford University Press

Adherence of Clostridium perfringens spores to human intestinal epithelial Caco-2 cells

Loading next page...
 
/lp/ou_press/adherence-of-clostridium-perfringens-spores-to-human-intestinal-BkovxsXrYf
Publisher
Oxford University Press
Copyright
© FEMS 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
0378-1097
eISSN
1574-6968
D.O.I.
10.1093/femsle/fny016
Publisher site
See Article on Publisher Site

Abstract

Abstract Clostridium perfringens is a gram-positive, spore-forming bacillus, and is a causative agent of foodborne infection, antibiotic-associated diarrhoea and sporadic diarrhoea in humans. In cases of antibiotic-associated and sporadic diarrhoea, C. perfringens colonises the intestine, proliferates and causes disease. However, bacterial colonisation of the intestine is not considered necessary in the pathogenesis of foodborne illness, because such pathogenesis can be explained by anchorage-independent production of diarrhoeic toxin by the bacterium in the intestine. In this study, we used an in vitro adherence assay to examine the adherence of C. perfringens spores to human intestinal Caco-2 cells. Adherence of spores from isolates of foodborne illness and nosocomial infection was observed within 15 min, and plateaued 60 min after inoculation. Electron microscopy revealed a tight association of spores with the surface of Caco-2 cells. The adherence of vegetative cells could not be confirmed by the same method, however. These results suggest that C. perfringens spores may adhere to intestinal epithelial cells in vivo, although its biological significance remains to be determined. Clostridium perfringens, foodborne illness, spore, nosocomial infection, sporadic diarrhoea INTRODUCTION Clostridium perfringens is an anaerobic, gram-positive and spore-forming bacterium, and occurs in ubiquitous habitats such as sewage, soil, plant surfaces, faeces and the normal intestinal microbiota of humans and animals. It is an important pathogen that causes a variety of human and animal diseases, ranging from myonecrotic to enteric infections (Songer 1996; Petit, Gibert and Popoff 1999). It reportedly produces up to 17 toxins; of these, four typing toxins—alpha, beta, epsilon and iota—are used to classify the species into five toxinotypes: A, B, C, D and E (Li et al.2013). Type A strains, which of the four typing toxins produce only the alpha toxin, can be isolated from human clinical specimens of foodborne illness (FBI), antibiotic-associated diarrhoea (AAD) and sporadic and nosocomial diarrhoea (SND) (Tompkins et al.1999; Watanabe, Hitomi and Sawahata 2008; Miyamoto, Li and McClane 2012; Polage, Solnick and Cohen 2012). A common feature of these isolates is the production of the nontyping toxin, C. perfringens enterotoxin, which is an important virulence factor (Li et al.2013). Despite this common feature, the pathogenesis of FBI and two other diarrhoeal diseases (AAD and SND) apparently differs in terms of colonisation behaviour. AAD develops via antibiotic-induced dysbiosis of gut microbiota, when endogenous C. perfringens is allowed to overcome and outgrow its competition (Polage, Solnick and Cohen 2012). This indicates that C. perfringens colonises the intestinal tract, where it persists as a component of the commensal microbiota for many years; in fact, many investigators have reported such bacterial colonisation in healthy human intestines (Carman et al.2008; Nagpal et al.2015). The aetiology of SND is not fully understood, but may include oral infection from spores in the hospital or community. The ingested spores can colonise the intestine, which is followed by bacterial growth and the production of CPE, which causes diarrhoea. In contrast, FBI strains do not seem to colonise or bind directly to the intestinal surface, because the development of the acute disease can be explained by the transformation of vegetative cells into spores in the intestine, without the need for physical binding to the tissue surface. No direct demonstration of the necessity of colonisation by FBI strains has been reported, mainly because there are no animal infection models of the gastroenteritis induced by type A C. perfringens currently available (Uzal and McClane 2012). In this study, we investigated whether human isolates of C. perfringens adhere to human intestinal epithelial cells in vitro. We used the intestinal cell line Caco-2, two different FBI strains and an SND strain in cell adherence assays, and the results indicate that spores from the FBI and SND strains adhere tightly to Caco-2 cells. An electron microscopy investigation supports our conclusion, and provides additional interesting observations of the interaction between C. perfringens spores and intestinal epithelial cells. To the best of our knowledge, this is the first report to demonstrate visually that C. perfringens spores, especially those from an SND strain, adhere to human intestinal cells in vitro. MATERIALS AND METHODS Bacterial strains and spore preparation Clostridium perfringens type A strain SM101 is a derivative of FBI strain NCTC8798. NCTC8239, another FBI strain, was purchased from the National Collection of Type Cultures. W10030 is an isolate from a sporadic gastrointestinal infection that occurred in geriatric hospital wards in 2006, containing an IS1151-associated cpe gene on a 75-kb plasmid (Kobayashi et al.2009). In this hospital, several sporadic cases of diarrhoea were recognised in 1992–1993, 2006 and 2015–2016. Molecular epidemiological analyses revealed that the isolates seemed to have been derived from a single clonal lineage, or were closely related to each other (Monma et al., in preparation), suggesting that the hospital may have been colonised by these related strains for a long time, and that these strains caused the repeated incidents of SND. Based on these characteristics, we used strain W10030 as a typical SND strain. The culture stocks of the bacterial strains were prepared as previously described (Yasugi et al.2015), except that the bacterial culture was heated either at 70°C (SM101), 75°C (NCTC8239) or 60°C (W10030). The bacterial cells were heated, separated into small aliquots and stored at −80°C until use. Spores were prepared as described previously (Sacks and Alderton 1961). Briefly, bacterial cultures re-grown in Fluid Thioglycollate (FTG) medium were inoculated into 40 ml of Duncan-Strong medium (DS medium; Duncan and Strong 1968), followed by incubation for 22–26 h at 37°C. The cultures were centrifuged at 7500× g for 15 min at 25°C, and the supernatant was then removed. Sterilised ultrapure water (10 ml) was added to the precipitate, followed by 3.41 ml of 3 M potassium phosphate buffer (pH 7.1) and 1.118 g of solid polyethylene glycol 4000 (SERVA). After gentle mixing until complete dissolution of the polyethylene glycol, the mixture was left standing for more than 30 min to ensure two-phase partitioning. Bacterial spores in the upper phase were recovered by centrifugation at 15 000× g for 10 min at 4°C, and washed thrice with ultrapure water. The resultant spore suspension was subjected to another round of partitioning to remove contaminated vegetative cells. Colony forming units (CFU) of the final preparations were enumerated after heat treatments at 70°C (SM101), 75°C (NCTC8239) or 60°C (W10030). The dilution of the heated samples was performed in Brain Heart Infusion (BHI) broth with 0.1% (v/v) Triton X-100, and the CFUs were counted on BHI agar incubated anaerobically for 24 h. Adherence assay Human colonic epithelial Caco-2 cells were maintained in Dulbecco's Modified Eagle's Medium (DMEM, Sigma) supplemented with 10% foetal bovine serum (FBS). For the co-culture experiments, cells were plated in 35-mm culture dishes and incubated for 4 days (final density approximately 1.4 × 106 cells/dish). The cells were washed once with phosphate-buffered saline (PBS) and incubated in 2 ml of glucose-deficient DMEM (DMEM(−); Thermo Fisher Scientific, Cat. No. 11966025) supplemented with 0.4% soluble starch and 50 μM deoxycholic acid (Yasugi et al.2015). The bacteria from the culture stock were grown in FTG medium (3 ml) for 11–13 h at 37°C, collected and washed twice with PBS by centrifugation and resuspended in 3 ml of PBS (final concentration, 1.0 × 106 CFU/ml). Aliquots of 200 μl containing vegetative bacterial cells were inoculated into Caco-2 culture dishes and incubated for the indicated period. The total number of bacterial cells in the culture supernatant was determined by enumerating the CFU, as described above. When determining the number of heat-resistant spores, the samples were heat-treated prior to dilution. Detection thresholds under these conditions were 20 CFU/ml. Bacterial vegetative cells and spores that adhered on Caco-2 cells were quantified after ultrasonic homogenisation of washed Caco-2 cells. The washed cells, recovered in PBS by pipetting, were completely disrupted by ultrasonication for 10 s using an Ultrasonic Disruptor (TOMY SEIKO, Co., Ltd) at a low oscillation output level (0.1), with micro-tip TP-040. This weak ultrasonication did not affect the viability of the vegetative bacteria or spores, but resulted in the disruption of the Caco-2 cells into fine fragments (<1 μm; data not shown). To examine the adherence of isolated vegetative bacteria or spores, Caco-2 cells were plated in 35-mm dishes at 3.0 × 106 cells/dish, and incubated in DMEM supplemented with 5% FBS for 20 h. Before exposure to the bacterial cells, the medium was changed to DMEM(−) supplemented with 0.4% soluble starch and 50 μM deoxycholic acid. The Caco-2 dish (2.0 × 106 CFU/dish) was inoculated with the spore suspension and incubated at 37°C for the indicated period. The cells were washed thrice with 3 ml of PBS, and detached by pipetting in 0.1% Triton X-100 in PBS. The number of spores was enumerated by counting the CFU after heat treatment, as described above. The detection threshold of this procedure was 20 CFU/ml. The vegetative bacteria were prepared by culturing in FTG medium for 11–13 h, and were washed twice with 10 ml of PBS. The bacteria suspended in DMEM(−) supplemented with 0.4% soluble starch and 50 μM deoxycholic acid were inoculated into the Caco-2 dishes at a multiplicity of infection (MOI) of 5.4, and incubated at 37°C for 2 h. After incubation, the cells were washed 10 times with 3 ml of PBS, and observed under phase-contrast microscopy. In all adherence assays, incubations were performed under a 5% CO2 atmosphere. Electron microscopy Caco-2 cells were cultured in 35-mm dishes in DMEM supplemented with 5% FBS for 20–24 h (final density 4.5 × 106 cells/dish), and treated with vegetative bacteria or spores for 60 min at 2.7 MOI and 12.9 MOI, respectively. For scanning electron microscopy (SEM), Caco-2 cells were cultured on a (γ-aminopropyl) trimethoxysilane-coated cover glass (4 × 8 mm, Matsunami Glass) placed in the culture dish. The co-cultured cells were doubly fixed with 2% glutaraldehyde and 2% OsO4, both in 0.05 M cacodylate buffer (pH 7.2), and dehydrated with serially graded ethanol. Dehydrated samples were subjected to critical-point drying, coated with platinum at 5-nm thick using an ion sputter (E-1030, Hitachi), and observed under a scanning electron microscope (S-5000, Hitachi). For transmission electron microscopy (TEM), 0.05% ruthenium red was added in both fixatives. The samples were embedded in plain resin (Nissin EM), and ultrathin sections were made and observed under an electron microscope (H-7100, Hitachi). RESULTS We initiated our study to examine the adherence of C. perfringens to Caco-2 cells, using our co-culture system. When Caco-2 cells were co-cultured with the vegetative C. perfringens strain SM101 under spore-forming conditions (Yasugi et al.2015), the total number of bacterial cells (i.e. vegetative bacteria plus spores) in the culture supernatant increased over time (Fig. 1). Heat-resistant spores were also detected in the supernatant at 8 h, and continued to increase until 12 h. Notably, the Caco-2 cell monolayer began to detach from the substratum after 12 h, probably via the action of the enterotoxin produced by C. perfringens during sporulation (Yasugi et al.2015). As a result, we were unable to quantify the cell-adhered bacteria or spores thereafter (data not shown). Under these conditions, bacteria and spores appeared to adhere to Caco-2 cells during a 12-h incubation (Fig. 1, circles); however, adhesion levels were only 1.5 (total bacteria) and 0.1 (spores) CFU/100 Caco-2 cells, i.e. only 15 vegetative bacteria and one spore attached per thousand of Caco-2 cells. Figure 1. View largeDownload slide Detection of C. perfringens associated with Caco-2 cells in co-cultured experiments. Clostridium perfringens strain SM101 was co-cultured with Caco-2 cells for the indicated period. The total number of bacterial cells (dotted lines) or the number of spores (bold lines) in the culture supernatants (squares) or on Caco-2 cells (circles) was determined as described in the Materials and Methods section. Averages and standard deviations of three independent experiments are shown. Figure 1. View largeDownload slide Detection of C. perfringens associated with Caco-2 cells in co-cultured experiments. Clostridium perfringens strain SM101 was co-cultured with Caco-2 cells for the indicated period. The total number of bacterial cells (dotted lines) or the number of spores (bold lines) in the culture supernatants (squares) or on Caco-2 cells (circles) was determined as described in the Materials and Methods section. Averages and standard deviations of three independent experiments are shown. We then attempted to confirm the physical binding of the bacterial cells to Caco-2 cells by visualising the adhered bacteria using phase contrast and electron microscopy. We observed that a certain proportion of the vegetative bacteria remained attached to the Caco-2 cell monolayer in the adherence assay, even after vigorous washing (Fig. 2A). Each of the bacterial cells, however, appeared to be oscillating according to Brownian motion, suggesting that the observed association between the bacteria and Caco-2 cells may have been an artefact, or that the adherence may be mediated by thin fibrous extensions such as pili (data not shown). Similar results were obtained with strain NCTC8239 (data not shown). To further examine the possible adherence, the Caco-2 cells were observed using SEM. We found that very few SM101 cells attached to Caco-2 cell surfaces (Fig. 2B and C). Furthermore, TEM provided no evidence for direct contact between SM101 cells and Caco-2 cell surfaces (Fig. 2D and E), indicating that vegetative SM101 bacteria, and probably also NCTC8239 bacteria, do not adhere tightly to Caco-2 cells under these conditions. Figure 2. View largeDownload slide Assessment of adhesion of vegetative C. perfringens bacteria to Caco-2 cells. (A) A phase contrast micrograph of Caco-2 cells treated with vegetative strain-SM101 C. perfringens bacteria after 2 h of incubation (5.4 MOI). Scale bars, 20 μm. (B). (C) Scanning electron micrographs of Caco-2 cells treated with vegetative strain-SM101 C. perfringens bacteria for 60 min at 2.7 MOI. Bacterial cells are indicated with arrows. Scale bars, 10 μm. (D). (E) Transmission electron micrographs of Caco-2 cells treated with vegetative strain-SM101 C. perfringens bacteria for 60 min at 2.7 MOI. Scale bars, 5 μm (D) and 1 μm (E). Figure 2. View largeDownload slide Assessment of adhesion of vegetative C. perfringens bacteria to Caco-2 cells. (A) A phase contrast micrograph of Caco-2 cells treated with vegetative strain-SM101 C. perfringens bacteria after 2 h of incubation (5.4 MOI). Scale bars, 20 μm. (B). (C) Scanning electron micrographs of Caco-2 cells treated with vegetative strain-SM101 C. perfringens bacteria for 60 min at 2.7 MOI. Bacterial cells are indicated with arrows. Scale bars, 10 μm. (D). (E) Transmission electron micrographs of Caco-2 cells treated with vegetative strain-SM101 C. perfringens bacteria for 60 min at 2.7 MOI. Scale bars, 5 μm (D) and 1 μm (E). Next, we examined the adherence of isolated C. perfringens spores to Caco-2 cells. Heat-resistant spores associated with Caco-2 cells became evident after 15 min of incubation, and had reached close to their maximum abundance by 60 min (Fig. 3). The numbers of adhered spores differed according to the bacterial strain; NCTC8239 spores exhibited the greatest adhesion to Caco-2 cells, followed by SM101, and then W10030. Similarly, SEM revealed a number of spores attached to the Caco-2 cell monolayer when treated with isolated SM101 or W10030 spores for 60 min (Fig. 4A–D). Furthermore, TEM revealed evidence for direct contact between spores from both strains and Caco-2 cell surfaces (Fig. 4E and F). We also found numerous examples of microvilli-like structures on the Caco-2 cells, to which the spore surfaces were attached, in both SEM and TEM (Fig. 4B and D), confirming that SM101 and W10030 spores adhere to Caco-2 cells under these conditions. Figure 3. View largeDownload slide Quantification of C. perfringens spores adhering to Caco-2 cells. Clostridium perfringens spores (isolated spore suspension) from strains (A) SM101, (B) NCTC8239 and (C) W10030 were inoculated on Caco-2 cells at 0.44 MOI, and incubated for the indicated period. The number of heat-resistant spores associated with Caco-2 cells was determined as described in the Materials and Methods section. Averages and standard deviations of three experiments are shown. Figure 3. View largeDownload slide Quantification of C. perfringens spores adhering to Caco-2 cells. Clostridium perfringens spores (isolated spore suspension) from strains (A) SM101, (B) NCTC8239 and (C) W10030 were inoculated on Caco-2 cells at 0.44 MOI, and incubated for the indicated period. The number of heat-resistant spores associated with Caco-2 cells was determined as described in the Materials and Methods section. Averages and standard deviations of three experiments are shown. Figure 4. View largeDownload slide Assessment of adhesion of C. perfringens spores to Caco-2 cells. (A–D) Scanning electron micrographs of Caco-2 cells treated with isolated C. perfringens spores. Spores (5.8 × 107 spores) from strains (A, B) SM101 and (C, D) W10030 were inoculated onto Caco-2 cells (4.5 × 106 cells) and incubated for 60 min at 12.9 MOI. Attachment sites of microvilli-like structures to spores are indicated with arrows. Scale bars, 10 μm (A, C) and 0.5 μm (B, D). (E, F) Transmission electron micrographs of Caco-2 cells treated with isolated C. perfringens spores. Spores from strains (E) SM101 and (F) W10030 were incubated with Caco-2 cells at 12.9 MOI as described above. Attachment sites of microvilli-like structures to spores are indicated with arrows. Scale bars, 0.5 μm. Figure 4. View largeDownload slide Assessment of adhesion of C. perfringens spores to Caco-2 cells. (A–D) Scanning electron micrographs of Caco-2 cells treated with isolated C. perfringens spores. Spores (5.8 × 107 spores) from strains (A, B) SM101 and (C, D) W10030 were inoculated onto Caco-2 cells (4.5 × 106 cells) and incubated for 60 min at 12.9 MOI. Attachment sites of microvilli-like structures to spores are indicated with arrows. Scale bars, 10 μm (A, C) and 0.5 μm (B, D). (E, F) Transmission electron micrographs of Caco-2 cells treated with isolated C. perfringens spores. Spores from strains (E) SM101 and (F) W10030 were incubated with Caco-2 cells at 12.9 MOI as described above. Attachment sites of microvilli-like structures to spores are indicated with arrows. Scale bars, 0.5 μm. DISCUSSION In this study, we evaluated the adherence of two strains of type A C. perfringens to human intestinal Caco-2 cells. First, we examined the ability of vegetative bacterial cells to adhere to Caco-2 cells in co-culture experiments, and observed a certain level of time-dependent adherence (Fig. 1). Several investigators have previously reported the binding of vegetative C. perfringens to host cells or tissues. In particular, the association of SM101 and NCTC8239 with Caco-2 cells has been examined in relation to the bacterial sialidase, NanI (Li and McClane 2014). The authors concluded that both strains show only weak adherence activity, because of their NanI-negative phenotype. However, their study did not provide morphological evidence for adherence, so we tried to confirm it using electron microscopy. We could not find visible evidence of the bacterial adherence, probably because the adherence was so weak that the bacteria could not be kept adhered during canonical fixation and drying treatments. In addition, several investigators have reported the binding of other human or animal C. perfringens strains to intestinal cells or extracellular matrix proteins, such as collagens and fibronectin, and have proposed the importance of the bacterial proteins encoded in the genes cna (Redondo et al.2015; Wade et al.2016) and fbp (Hitsumoto et al.2014) in such binding. However, these genes are not present in the SM101 genome (GenBank accession No. CP000312.1), which could explain why this strain did not show tight adhesion, if any, to Caco-2 cells under our experimental conditions. On the other hand, our electron microscopy confirmed the adherence of C. perfringens spores to Caco-2 cells (Figs 3 and 4). Adherence of SM101 spores to Caco-2 cells has been reported elsewhere (Paredes-Sabja and Sarker 2012), but morphological evidence for this adhesion had not previously been presented. Our observations clearly indicate physical association of the bacterial spores with the surface of Caco-2 cells (Fig. 4). Interestingly, the tips and lateral surfaces of numerous microvilli-like structures on the Caco-2 cells appear to attach to the surface of the spores (Fig. 4B and D–F). Whether these unique morphological features reflect an important biological phenomenon should be investigated in further studies. Binding of C. difficile spores via microvilli expressed on Caco-2 cells has also been reported; this binding appeared to be mediated by hair-like extensions of the exosporium-like layer, and the attachment sites appeared to be adjacent to either the tip or the lateral surface of the microvilli (Paredes-Sabja and Sarker 2012; Mora-Uribe et al.2016). These findings may suggest that the adhesins on the spore surface are different for C. difficile and C. perfringens. Notably, in the co-culture experiments, only 1% of spores adhered to Caco-2 cells (Fig. 1), whereas 7.9%, 45% and 2.7% of the input of isolated spores (SM101, NCTC8239 and W10030, respectively) did so in the adherence assays (Fig. 3). One possible explanation for the different adherence efficiency is the presence of concomitant materials derived from the bacteria. In the co-culture dish, it is likely that a number of viable and dead vegetative cells, bacterial cells debris, immature spores and probably heat-sensitive germinating spores are all present. These bacteria-related substances may have hampered the adherence of the heat-resistant spores to the Caco-2 cells. Additionally, the numbers of Caco-2-associated spores were relatively lower (1–20 spores/100 Caco-2 cells) when enumerated via CFU counting (Fig. 3), even though several dozen spores per cell were visualised on the surface of Caco-2 cells by electron microscopy (Fig. 4A). We consider that differences in MOIs in these experiments (10 times lower in the CFU study) do not directly explain the high level of adherence observed in the SEM study. Thus, another explanation is that many spores become undetectable by CFU assay when associated with Caco-2 cells; association with Caco-2 cells may render the attached spores insensitive or unresponsive to germinants, via an unknown mechanism. A previous finding, showing that C. difficile spores attached to the surface of Caco-2 cells exhibited different germination abilities on different co-cultured cell lines, may support this idea (Paredes-Sabja and Sarker 2011). The mechanism and molecules that mediate the binding and interaction of C. perfringens spores with Caco-2 cells remain unclear at this stage. Hydrophobic interactions between spore surface proteins and Caco-2 surface proteins could be one of the binding mechanisms. Hydrophobic interaction has been proposed as a mechanism governing the interactions of spores from clostridial species with cultured mammalian cells (Andersson, Granum and Rönnera 1998; Tauveron et al.2006; Thwaite et al.2009; Paredes-Sabja and Sarker 2012). Based on TEM, some authors have suggested that the exosporium layer of the outermost surface of the bacterial spores is responsible for these interactions (Russell et al.2007). We also conducted TEM in the present study, and found that none of the strains used in this study possesses an exosporium layer (Fig. 5), which suggests that the mode of adherence in these strains is not exosporium mediated. Indeed, Paredes-Sabja and Sarker (2012) have reported that hydrophobicity is not particularly important for the interaction between spores and Caco-2 cells in the case of some C. perfringens strains, including SM101. The factors governing cell-to-cell interactions, such as the physical interactive force, spore surface adhesins and adhesion receptors, should be further investigated to unveil the biological significance of this phenomenon. Figure 5. View largeDownload slide Structure of spores from strains SM101, NCTC8239 and W10030. Bacterial strains were precultured in FTG medium for 11–13 h and incubated in DS medium for 24 h at 37°C. Total bacterial cells were collected by centrifugation at 7500× g for 15 min, followed by the removal of the supernatant and the addition of 2% glutaraldehyde in 0.05 M Ruthenium Red. The samples were then prepared for TEM as described in the Materials and Methods section. TEM micrographs of (A) SM101, (B) NCTC8239 and (C) W10030 are shown. Three areas representing coat layer, cortex and core are indicated. Scale bar, 500 nm. Figure 5. View largeDownload slide Structure of spores from strains SM101, NCTC8239 and W10030. Bacterial strains were precultured in FTG medium for 11–13 h and incubated in DS medium for 24 h at 37°C. Total bacterial cells were collected by centrifugation at 7500× g for 15 min, followed by the removal of the supernatant and the addition of 2% glutaraldehyde in 0.05 M Ruthenium Red. The samples were then prepared for TEM as described in the Materials and Methods section. TEM micrographs of (A) SM101, (B) NCTC8239 and (C) W10030 are shown. Three areas representing coat layer, cortex and core are indicated. Scale bar, 500 nm. Molecular biological and epidemiological studies, as well as animal model experiments, have demonstrated that spores play a crucial role in the transmission of C. difficile during infection (Abt, McKenney and Pamer 2016). A similar pathological mechanism is suspected in case of AAD associated with C. perfringens infection (Kokai-Kun et al.1994; Carman 1997). It is therefore important to examine the ability of AAD- and SND-associated C. perfringens strains to adhere to human cultured cells or intestinal tissue. Unlike for C. difficile, an animal model for C. perfringens-associated diarrhoea using bacterial cells or spores as an inoculum has not yet been established (Uzal and McClane 2012). Thus, further investigations to establish such a model are now warranted. The FBI caused by C. perfringens is an acute disease that usually lasts for less than 24 h (Li et al.2013). Thus, it seems that the spore colonisation of the intestine is not involved in the development of the disease. Our observation of the adherence of SM101 spores to Caco-2 cells suggests that some spores may remain attached in the patients’ intestine after FBI has developed, causing these patients to become potential reservoirs once they have recovered from the disease. This may also cause AAD following antibiotic treatment, or these patients may become an infection source of SND in a hospital or community. An epidemiological examination using several clinical strains from different types of diseases within a selected community will be needed to investigate this possibility. Acknowledgements We thank Dr Tohru Shimizu (Kanazawa University) for providing bacterial strain SM101. We also thank Yoshihiko Fujioka (Department of Microbiology and Infection Control, Osaka Medical College) for his technical assistance. FUNDING This work was supported by JSPS KAKENHI (15K14858 to MM) and a grant from the Research Project to Improved Food Safety and Animal Health (Ministry of Agriculture, Forestry, and Fisheries, Japan, 2013–2017 FY to MM). Conflict of interest. None declared. REFERENCES Abt MC, McKenney PT, Pamer EG. Clostridium difficile colitis: pathogenesis and host defence. Nat Rev Microbiol  2016; 14: 609– 20. Google Scholar CrossRef Search ADS PubMed  Andersson A, Granum PE, Rönner U. The adhesion of Bacillus cereus spores to epithelial cells might be an additional virulence mechanism. Int J Food Microbiol  1998; 39: 93– 9. Google Scholar CrossRef Search ADS PubMed  Carman RJ. Clostridium perfringens in spontaneous and antibiotic-associated diarrhoea of man and other animals. Rev Med Microbiol  1997; 8: S43– 5. Google Scholar CrossRef Search ADS   Carman RJ, Sayeed S, Li J et al.   Clostridium perfringens toxin genotypes in the feces of healthy North Americans. Anaerobe  2008; 14: 102– 8. Google Scholar CrossRef Search ADS PubMed  Duncan CL, Strong DH. Improved medium for sporulation of Clostridium perfringens. Appl Microbiol  1968; 16: 82– 9. Google Scholar PubMed  Hitsumoto Y, Morita N, Yamazoe R et al.   Adhesive properties of Clostridium perfringens to extracellular matrix proteins collagens and fibronectin. Anaerobe  2014; 25: 67– 71. Google Scholar CrossRef Search ADS PubMed  Kobayashi S, Wada A, Shibasaki S et al.   Spread of a large plasmid carrying the cpe gene and the tcp locus amongst Clostridium perfringens isolates from nosocomial outbreaks and sporadic cases of gastroenteritis in a geriatric hospital. Epidemiol Infect  2009; 137: 108– 13. Google Scholar CrossRef Search ADS PubMed  Kokai-Kun JF, Songer JG, Czeczulin JR et al.   Comparison of Western immunoblots and gene detection assays for identification of potentially enterotoxigenic isolates of Clostridium perfringens. J Clin Microbiol  1994; 32: 2533– 9. Google Scholar PubMed  Li J, Adams V, Bannam TL et al.   Toxin plasmids of Clostridium perfringens. Microbiol Mol Biol Rev  2013; 77: 208– 33. Google Scholar CrossRef Search ADS PubMed  Li J, McClane BA. Contributions of NanI sialidase to Caco-2 cell adherence by Clostridium perfringens Type A and C strains causing human intestinal disease. Infect Immun  2014; 82: 4620– 30. Google Scholar CrossRef Search ADS PubMed  Miyamoto K, Li J, McClane BA. Enterotoxigenic Clostridium perfringens: detection and identification. Microbes Environ  2012; 27: 343– 9. Google Scholar CrossRef Search ADS PubMed  Mora-Uribe P, Miranda-Cárdenas C, Castro-Córdova P et al.   Characterization of the adherence of Clostridium difficile spores: the integrity of the outermost layer affects adherence properties of spores of the epidemic strain R20291 to components of the intestinal mucosa. Front Cell Infect Microbiol  2016; 6: 99. Google Scholar CrossRef Search ADS PubMed  Nagpal R, Ogata K, Tsuji H et al.   Sensitive quantification of Clostridium perfringens in human feces by quantitative real-time PCR targeting alpha-toxin and enterotoxin genes. BMC Microbiol  2015; 15: 219. Google Scholar CrossRef Search ADS PubMed  Paredes-Sabja D, Sarker MR. Germination response of spores of the pathogenic bacterium Clostridium perfringens and Clostridium difficile to cultured human epithelial cells. Anaerobe  2011; 17: 78– 84. Google Scholar CrossRef Search ADS PubMed  Paredes-Sabja D, Sarker MR. Adherence of Clostridium difficile spores to Caco-2 cells in culture. J Med Microbiol  2012; 61: 1208– 18. Google Scholar CrossRef Search ADS PubMed  Petit L, Gibert M, Popoff MR. Clostridium perfringens: toxinotype and genotype. Trends Microbiol  1999; 7: 104– 10. Google Scholar CrossRef Search ADS PubMed  Polage CR, Solnick JV, Cohen SH. Nosocomial diarrhea: evaluation and treatment of causes other than Clostridium difficile. Clin Infect Dis  2012; 57: 982– 9. Google Scholar CrossRef Search ADS   Redondo LM, Carrasco JM, Redondo EA et al. Clostridium perfringens type E virulence traits involved in gut colonization. PLoS One  2015; 10: e0121305. Google Scholar CrossRef Search ADS PubMed  Russell BH, Vasan R, Keene DR et al. Bacillus anthracis internalization by human fibroblasts and epithelial cells. Cell Microbiol  2007; 9: 1262– 74. Google Scholar CrossRef Search ADS PubMed  Sacks LE, Alderton G. Behavior of bacterial spores in aqueous polymer two-phase systems. J Bacteriol  1961; 82: 331– 41. Google Scholar PubMed  Songer JG. Clostridial enteric diseases of domestic animals. Clin Microbiol Rev  1996; 9: 216– 34. Google Scholar PubMed  Tauveron G, Slomianny C, Henry C et al.   Variability among Bacillus cereus strains in spore surface properties and influence on their ability to contaminate food surface equipment. Int J Food Microbiol  2006; 110: 254– 62. Google Scholar CrossRef Search ADS PubMed  Thwaite JE, Laws TR, Atkins TP et al.   Differential cell surface properties of vegetative Bacillus. Lett Appl Microbiol  2009; 48: 373– 8. Google Scholar CrossRef Search ADS PubMed  Tompkins DS, Hudson MJ, Smith HR et al.   A study of infectious intestinal disease in England: microbiological findings in cases and controls. Comm Dis Public Health  1999; 2: 108– 13. Uzal FA, McClane BA. Animal models to study the pathogenesis of enterotoxigenic Clostridium perfringens infections. Microbes Infect  2012; 14: 1009– 16. Google Scholar CrossRef Search ADS PubMed  Wade B, Keyburn AL, Haring V et al.   The adherent abilities of Clostridium perfringens strains are critical for the pathogenesis of avian necrotic enteritis. Vet Microbiol  2016; 197: 53– 61. Google Scholar CrossRef Search ADS PubMed  Watanabe M, Hitomi S, Sawahata T. Nosocomial diarrhea caused by Clostridium perfringens in the Tsukuba-Tsuchiura district, Japan. J Infect Chemother  2008; 14: 228– 31. Google Scholar CrossRef Search ADS PubMed  Yasugi M, Sugahara Y, Hoshi K et al.   In vitro cytotoxicity induced by Clostridium perfringens isolate carrying a chromosomal cpe gene is exclusively dependent on sporulation and enterotoxin production. Microb Pathogenesis  2015; 85: 1– 10. Google Scholar CrossRef Search ADS   © FEMS 2018. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com

Journal

FEMS Microbiology LettersOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off