- Email: [email protected]
Interaction of Escherichia coli O157:H7 E318 cells with the mucus of harvested channel catfish (Ictalurus punctatus)
ARTICLE IN PRESS
LWT 40 (2007) 1266–1270 www.elsevier.com/locate/lwt
Interaction of Escherichia coli O157:H7 E318 cells with the mucus of harvested channel catfish (Ictalurus punctatus) Rico R. Suhalima, Yao-Wen Huanga,, Jinru Chenb a
Department of Food Science and Technology, University of Georgia, Athens, GA 30602, USA Department of Food Science and Technology, University of Georgia, Griffin, GA 30223 USA
b
Received 6 December 2005; received in revised form 5 August 2006; accepted 8 August 2006
Abstract Channel catfish skin with or without mucus (0.5 cm in diameter) were immersed into a suspension containing 109 CFU/ml of Escherichia coli O157:H7 E318 cells at 22 1C for 20 min. The inhibitory effect of skin mucus was determined by placing the mucus-side down on tryptic soy agar inoculated with 104–105 CFU of E. coli O157:H7 E318. The inhibition zones of fish mucus had a diameter of approximately 0.7 cm and were only visible for the first 12 h of the incubation. Bacterial cells were observed at 15 mm into the mucus layer under confocal scanning laser microscopy (CSLM). Plate counts and CSLM revealed 0.5- and 1-log less cells, respectively, attached to skin without mucus than to skin with mucus. Results suggest that E. coli O157:H7 E318 could attach to and penetrate through the mucus of channel catfish and may become a source of contamination during catfish processing. r 2006 Swiss Society of Food Science and Technology. Published by Elsevier Ltd. All rights reserved. Keywords: Channel catfish; Escherichia coli O157:H7 e318; Confocal scanning laser microscopy; Green fluorescent protein; Catfish skin
1. Introduction Outbreaks of Escherichia coli O157:H7 infections have been linked to water and a variety of foods (Doyle & Schonei, 1987; Swerdlow et al., 1992). Although fishery products have not been implicated in the outbreaks, they are no doubt potential vehicles for transmitting E. coli O157:H7 (Andrews, Wilson, & Poelma, 1977; Cotton & Marshall, 1998; Leung, Huang, & Pancorbo, 1992; Meyer & Bullock, 1973). Channel catfish (Ictalurus punctatus) grow in a light water environment. They can be contaminated by pathogenic bacteria like E. coli O157:H7 in the growing environment as a result of storm runoff from adjacent agricultural land or droppings from domestic and wild farm animals. Contamination of catfish by pathogenic microorganisms may also occur during processing. The skin and gut of catfish are believed to be the critical sources of microbial contamination (Kim & Marshall, 2002). The microorganisms from these two locations could spread, Corresponding author. Tel.: +1 706 542 1092; fax: +1 706 542 1050.
E-mail address: [email protected] (Y.-W. Huang).
contaminating the entire catfish processing environment (Bal’a, Podolak, & Marshall, 1999). As part of a natural defense system, fish secrete mucus to their surfaces. The mucus contains specific immunoglobins and lysozyme, which can inhibit microorganisms on fish skin (Speare & Mirasalimi, 1992). In addition to immunoglobins and lysozyme, the mucus of channel catfish may contain histone-like proteins (Robinette et al., 1998), which are effective against the cells of Aeromonas hydrophila and Salmonella parathyphi (Ourth, 1980). The mucus also has the ability to entrap microorganisms, preventing them from colonizing on the surface of fish by continuous sloughing (Roberts, 2001). Although the skin mucus of live catfish has the aforementioned defensive mechanisms, it is not known whether the natural function of the mucus remains after the catfish is harvested. This study was undertaken to evaluate the antibacterial effects of harvested catfish skin mucus on the cells of E. coli O157:H7, and to examine the penetration of E. coli O157:H7 cells through the mucus layer of catfish with the aid of confocal scanning laser microscopy (CSLM).
0023-6438/$30.00 r 2006 Swiss Society of Food Science and Technology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.lwt.2006.08.017
ARTICLE IN PRESS R.R. Suhalim et al. / LWT 40 (2007) 1266–1270
CSLM has been used to observe food structures (Hassan, Frank, & Corredig, 2002) and attachment of bacterial cells to food (Prachaiyo & McLandsborough, 2000; Seo & Frank, 1999; Takeuchi & Frank, 2000). The major advantage of CSLM is its ability to focus within a specific depth by optically sectioning the specimen and constructing a three-dimensional (3D) image (Amos & Fordham, 1997; Blonk & van Aalst, 1993). Different from scanning electronic microscopy, CSLM allows observation of the specimen in a fully hydrated state (Blonk & van Aalst, 1993). When used to study bacterial penetration, CSLM reveals the location and distribution of bacterial cells in addition to its ability to quantify the cells without extensive sample preparation and fixation. 2. Materials and methods 2.1. E. coli O157:H7E318 inoculum E. coli O157:H7 E318 containing a jellyfish green fluorescent protein (GFP) plasmid was used in this study. The GFP has an excitation wavelength of 488 nm, and an emission wavelength of 543 nm. The cells carrying an ampiciline resistance gene on the GEP plasmid and a nalidixic acid resistance gene on the chromosome, were grown for 18 h at 37 1C in brain heart infusion broth (BHI, Difco Laboratories, Detroit, MI, USA) supplemented with 100 mg/ml of ampicillin (Sigma, St. Loius, MO, USA) and nalidixic acid (Aldrich Chem Co., Milwaukee, WI, USA). The cells in the BHI broth were centrifuged at 5000g for 20 min, and the resulting cell pellet was washed with 0.1% buffered peptone water (Difco Laboratories). The washed cells were suspended in the same buffer, and diluted into desired concentrations. The cells were plated on BHI agar plates supplemented with 100 mg/ml of ampicillin and nalidixic acid, and grown for 18 h at 37 1C. The GFP colonies appeared on the plates were enumerated under a long-wave UV light. 2.2. Channel catfish Live channel catfish were purchased from the farmers market in Atlanta, GA and were kept in container with pumped air and transported to UGA laboratory in Athens, GA. The live fish were maintained in 40 l of aquaria at 20 1C until use. On the day of the experiment, the fish were sacrificed and fish skin was removed with a sterile surgical blade. Mucus of the fish was removed by scraping the skin surface with the surgical blade to obtain mucus-free skin samples. 2.3. Penetration of E. coli O157:H7 E318 cells through fish mucus Catfish skin samples (0.5 cm in diameter) with or without mucus were inoculated by immersion into a suspension (30 ml) of 109 CFU/ml of E. coli O157:H7 E318 cells at
1267
22 1C for 20 min. The length of time was decided according to results of the preliminary study. The contaminated fish skin samples were rinsed twice with 30 ml of sterile distilled water, each for 3 min at 22 1C. Following the rinsing, the fish skin samples were flooded for 5 min at 22 1C with 7.5 mm of propidium iodide (Molecular Probes, Inc., Eugene, OR, USA) which enabled the distinction of dead (red stained) from live (GFP) E. coli O157:H7 cells. After the treatment with propidium iodide, the fish skin samples were rinsed twice with 30 ml of sterile distilled water, each for 3 min at 22 1C to remove excessive stain. The skin samples were placed on cover slips, and the latter were inverted onto a hanging drop slide. The specimen was viewed under a CSLM Leica DM RBE (Leica Microsystems Inc., Exton, PA, USA) with 100 objective lens. The GFP cells were imaged with an argon laser at 488 nm excitation/543 nm emission, and propidium iodide was scanned with green neon at 543 nm. The reflectance mode was used to visualize mucus and skin and to aid in depth determination. Leica confocal software Version 2.0 Build 0770 (Exton, PA, USA) was used for data acquisition and analysis. E. coli O157:H7 E318 cells were enumerated at various depths along the z-axis until no GFP cells were observed. Optical slices were taken at approximately every 1 mm to avoid repeated counting of the same cells. The bacterial counts obtained from the CSLM observation was multiplied by two in order to give a similar counting effect as the two-sided fish skin surface estimated by a plate count assay. When the plate count assay was performed, each piece of the fish skin sample described above was placed into a sterile stomacher bag containing 20 ml of 0.1 ml/100 ml buffered peptone water. The samples were stomached for 2 min at high speed (Tekmar, Cincinnati, OH, USA). The resulting liquid samples were serially diluted in 0.1 ml/ 100 ml buffered peptone water prior to being surface plated in duplicate on BHI agar containing 100 mg (10 mg/100 ml) ampicillin and nalidixic acid. The plates were incubated for 24 h at 37 1C and the growing colonies were enumerated under a long-wave UV light. The colonies appeared on the plates were confirmed using a latex agglutination test (Oxoid Limited, Hampshire, England).
2.4. Inhibition of E. coli O157:H7 cells by fish mucus Approximately 104–105 CFU of E. coli O157:H7 E318 in 0.1 ml 0.1 g/100 g peptone water was surface plated with a sterile glass rod on tryptic soy agar (TSA, Difco Laboratories) plates. Thirty minutes after the inoculation, catfish skin pieces with or without the mucus were placed mucus-side down on the surface of the inoculated TSA plates. The plates were then incubated for 24 h at 37 1C. The presence of clearing zones around skin samples on the TSA agar plate was the indication of antimicrobial activity of fish mucus towards the cells of E. coli O157:H7 E318.
ARTICLE IN PRESS 1268
R.R. Suhalim et al. / LWT 40 (2007) 1266–1270
Sterile 0.5 cm filter paper discs impregnated with 5 g/100 g acetic acid were used as a control. 2.5. Statistical analysis All data counts were transformed into Log10 CFU/cm2 before comparison of the means. Analysis of the data was accomplished using the Duncan’s multiple range test of the Statistical Analysis Software (SAS; windows version 5.1.2600, 2000, SAS institute, Inc., Cary, NC, USA) based on a 95% confidence level (Po 0.05). 3. Results and discussion In the present study, the inhibition zones of fish mucus had a diameter of approximately 0.7 cm. However, they were only visible for the first 12 h of the incubation. After 18 h at 37 1C, the zones of inhibition faded due to the overgrowth of E. coli O157:H7 E318 cells surrounding the fish skin samples. This may have been due to the less susceptible nature of E. coli O157:H7 E318 cells to the antimicrobial components of the mucus or by the insufficient level of the active compounds in the mucus samples. The mode of inhibition and potency of fish mucus against bacterial cells have not been thoroughly investigated (Ourth, 1980; Robinette et al., 1998). While aquatic bacteria have varying susceptibility levels to fish mucus, the minimal inhibitory level of fish mucus against the cells of E. coli O157:H7 E318 is currently unknown. The determination of the antibacterial activity of fish mucus often implicates extensive extraction and purification of mucus components, which may have limited application in food safety. Intact skin mucus was used in the present study in order to observe the antibacterial activity on fish skin as fish is exposed to bacterial contaminants during processing. The potential of mucus as an agent against bacteria may assist in the reduction of contamination of pathogenic organisms on fish. However, mucus may lose its antibacterial activity soon after the fish is slaughtered. Certain inhibitors in blood serum greatly reduced the antibacterial activity of fish mucus (Hjelmeland, Christie, & Raa, 1983; Smith & Ramos, 1976) as fish skin could easily be contaminated with blood during processing. Proteases present in mucus could also inactivate the antibacterial compounds in fish mucus (Robinette et al., 1998). Other factors such as temperature abuse, bacterial load, and health condition of the fish may also affect the effectiveness of fish mucus in killing bacterial cells. Previous studies have shown that bacterial cells can attach and colonize on the mucus surface of harvested fish (Freter, Allweis, O’Brien, Halstead, & Macsai, 1981; Laux, Sweegan, & Cohen, 1984). Gilt-head sea brim skin mucus was found to be an optimal substrate for accumulation of aquatic bacteria such as Vibrio anguillarum and A. hydrophyla on fish (Balebona et al., 1995; Krovacek, Faris, Ahne, & Masson, 1987). Bacterial attachment to fish
mucus may be supported by the hydrophilic nature of fish skin (Hassan et al., 2002) and the nonhydrophobic surface of E. coli O157:H7 cells (Dewanti & Wong, 1995). The polysaccharide polymers of both fish mucus and bacteria were believed to aid in the adherence of bacterial cells on fish surface (Fletcher & Floodgate, 1973; Marshall, Stout, & Mitchell, 1971; Ourth, 1980; Takashima & Hibiya, 1995). In live fish however, the continual sloughing of the mucus aids in the removal of bacteria and inhibits the colonization of bacteria on fish skin. In the present study, E. coli O157:H7 E318 cells attached primarily to the area where mucus was present (Fig. 1A and Table 1), while the skin without mucus had less numbers of attached E. coli cells (Fig. 1B and Table 1). After being stained with propidium iodide, viable cells appeared green while nonviable cells stained red to yellow under CSLM (Fig. 1). Approximately 18% and 16% of the cells observed under CSLM appeared dead on skin with or without mucus, respectively. E. coli O157:H7 E318 cells were observed at areas as deep as 15 mm into the mucus layer and were decreasing in number at a deeper layer, with the majority of cells found within the first 5 mm. Free moving E. coli cells were observed in greater numbers on skin without mucus than were observed on skin with mucus. The results of the present study indicated that attachment of E. coli O157:H7 E318 cells did not occur randomly but rather in the area where mucus was abundant. E. coli O157:H7 E318 cells were able to penetrate into the mucus layer of catfish. In contrast, fish skin without mucus had fewer attached bacterial cells. The plate count results however, showed that skin without mucus contained a high number of bacterial cells. The high counts on mucus-free skin surface could have been attributed to the bacterial cells that adhered to the film layer of water on the skin after rinsing. A large number of free-moving bacteria on skin without mucus visualized under CSLM may have contributed to the high plate counts. Additionally, while the mucus had been significantly removed, a small amount of mucus might still have been present, allowing bacterial cells to attach. In contrast, the majority of E. coli O157:H7 E318 cells on the skin with mucus attached, and penetrated to a deeper layer, as visualized through the scanned images, rather than moving freely over the skin surface. In both skin types, plate counts were higher than direct counts under CSLM. The high values may have been due to the stomaching of the fish samples when plate counts were determined. Stomaching may have dislodged additional E. coli O157:H7 E318 cells that attached to the edges of the skin and to the muscle side of the skin which tends to be irregular in thickness and surface area. In addition, the disruption of cells aggregating on the skin surface during stomaching may have also contributed to the higher numbers of E. coli O157:H7 E318 cells enumerated by plate counting. Cell enumeration by direct observation was restrained to counting the cells that were not in clusters. The higher plate counts may also have been due to the ability of injured cells to grow on agar media, while the
ARTICLE IN PRESS R.R. Suhalim et al. / LWT 40 (2007) 1266–1270
1269
Table 1 Enumeration (mean7SD) by plate and direct microscopic count of E. coli O157:H7 E318 inoculated on catfish skin with and without mucus Treatment
Plate count (cells/cm2) CSLM (cells/cm2)
Skin with mucus
Skin without mucus
7.870.2aa 5.570.2a
7.370.1b 4.570.2b
a Means on the same row followed by the same letter are not significantly different.
is slaughtered. Cross contamination with pathogens, such as E. coli O157:H7, from skin to muscle is possible if the pathogen has previously been introduced during production or post-harvest processing. Proper skinning of catfish may reduce the cross contamination and/or growth of potential foodborne bacterial pathogens on mucus. Acknowledgments The authors are grateful to the University of Georgia College of Agricultural and Environmental Sciences College Station for support and to Dr. Ashraf Hassan, Department of Food Science and Technology and Dr. John Shields, the Center for Ultrastructural Research at the University of Georgia, for their valuable technical assistance with CSLM application. References
Fig. 1. Attachment and visualization, using of GFP E. coli O157:H7 E318, labeled with green fluorescent protein, on channel catfish skin (A) with mucus and (B) without mucus stained with PI (red dead cell stain) visualized under CSLM. Images were taken at 10 mm (A) and 5 mm (B). Green and yellow arrow heads indicate live and dead cells, respectively.
same cells may have appeared dead under CSLM due to penetration of propidium iodide, which has been shown to have the ability to gain access to leaky cell membranes. 4. Conclusion Mucus is a nonspecific protective barrier of fish, and is also a suitable media for bacterial attachment once the fish
Amos, W. B., & Fordham, M. (1997). An evaluation of confocal versus conventional imaging of biological structures by fluorescence light microscopy. Journal of Cell Biology, 105, 41–48. Andrews, W. H., Wilson, C. R., & Poelma, P. L. (1977). Bacteriological survey of the channel catfish (Ictalurus punctatus) at the retail level. Journal of Food Science, 42, 359–363. Bal’a, M. F., Podolak, A. R., & Marshall, D. L. (1999). Steam treatment reduces the skin microflora population on de-headed and eviscerated whole catfish. Food Microbiology, 16, 495–501. Balebona, M. C., Morinigo, M. A., Faris, A., Krovacek, Manson, K. I., Bordas, M. A., & Borrego, J. J. (1995). Influence of salinity and pH on the adhesion of pathogenic Vibrio strains to Sparus aurata skin mucus. Aquaculture, 132, 112–120. Blonk, J. C. G., & van Aalst, H. (1993). Confocal scanning light microscopy in food research. Food Research International, 26, 297–311. Cotton, L. N., & Marshall, D. L. (1998). Predominant microflora on catfish processing equipment. Dairy Food Environmental Sanitation, 18, 650–654. Dewanti, R., & Wong, A. C. L. (1995). Influence of culture condition s on biofilm formation by Escherichia coli O157:H7. International Journal of Food Microbiology, 26, 147–164. Doyle, M. P., & Schonei, J. L. (1987). Isolation of Escherichia coli O157:H7 from retail fresh market and poultry. Applied Environmental Microbiology, 53, 2394–2396. Fletcher, M., & Floodgate, M. D. (1973). An electron-microscopic demonstration of an acidic polysaccharide involved in the adhesion of a marine bacterium to solid surfaces. Journal of Genetic Microbiology, 74, 325–334. Freter, R., Allweis, B., O’Brien, P. C. M., Halstead, S. A., & Macsai, M. S. (1981). Role of chemotaxis in the association of motile bacteria with intestinal mucosa: in vitro studies. Infectious Immunology, 34, 241–249.
ARTICLE IN PRESS 1270
R.R. Suhalim et al. / LWT 40 (2007) 1266–1270
Hassan, A. N., Frank, J. F., & Corredig, M. (2002). Microstructure of feta cheese made using different cultures as determined by confocal scanning laser microscopy. Journal of Food Science, 67, 2750–2753. Hjelmeland, K., Christie, M., & Raa, J. (1983). Skin mucus protease from rainbow trout, and its biological significance. Journal of Fish Biology, 23, 13–22. Kim, J., & Marshall, D. L. (2002). Influence of catfish skin mucus on trisodium phosphate inactivation of attached Salmonella Typhymurium, Edwardsiella tarda and Listeria monocytogenes. Journal of Food Protection, 65, 1146–1151. Krovacek, K., Faris, A., Ahne, W., & Masson, I. (1987). Adhesion of Aeromonas hydrophyla and Vibrio anguilarum to mucus-coated glass slides. Federation of European Microbiological Societies Microbiology Letters, 42, 85–89. Laux, D. C., Sweegan, E. F., & Cohen, P. S. (1984). A new test based on ‘salting out’ to measure relative surface hydrophobicity of bacterial cells. Biochimica et Biophysica Acta, 677, 471–476. Leung, C. K., Huang, Y. W., & Pancorbo, O. C. (1992). Bacterial pathogens and indicators in catfish and pond environments. Journal of Food Protection, 55, 424–427. Marshall, K. C., Stout, R., & Mitchell, R. (1971). Mechanism of the initial events in the sorption of marine bacteria to surfaces. Journal of General Microbiology, 68, 337–348. Meyer, F. P., & Bullock, G. L. (1973). Edwarsiella tarda, a new pathogen of channel catfish. Applied Microbiology, 25, 155–156. Ourth, D. D. (1980). Secretory IgM, lysozyme and lymphocytes in the skin mucus of the channel catfish, Ictalurus punctatus. Developmental and Comparative Immunology, 4, 65–74.
Prachaiyo, P., & McLandsborough, L. A. (2000). A microscopic method to visualize Escherichia coli interaction with beef muscle. Journal of Food Protection, 63, 427–433. Roberts, R. J. (2001). Fish pathology. London, UK: W.B. Saunders. Robinette, D., Wada, S., Arroll, T., Levy, M. G., Miller, W. L., & Noga, E. J. (1998). Antimicrobial activity in the skin of the channel catfish: characterization of broad spectrum histone-like antimicrobial proteins. Cellular and Molecular Life Sciences, 54, 468–475. Seo, K. H., & Frank, J. F. (1999). Attachment of Escherichia coli to lettuce leaf surface and bacterial viability in response to chlorine treatment as demonstrated by using confocal scanning laser microscopy. Journal of Food Protection, 62, 3–9. Smith, A. C., & Ramos, F. (1976). Occult haemoglobin in fish skin mucus as an indicator of early stress. Journal of Fish Biology, 9, 537–541. Speare, D. J., & Mirasalimi, S. M. (1992). Pathology of the mucous coat of trout skin during an erosive bacterial dermatitis. Journal of Comparative Pathology, 106, 210–211. Swerdlow, D., Woodruff, B. A., Brady, R. C., Griffin, P. M., Tippen, S., Donnell, H. D., Jr., et al. (1992). A waterborne outbreak in Missouri of Escherichia coli O157:H7 associated with bloody diarrhea and death. Annals of Internal Medicine, 117, 812–819. Takashima, F., & Hibiya, T. (1995). An atlas of fish histology (second ed). Tokyo, Japan: Kodansha Ltd. Takeuchi, K., & Frank, J. F. (2000). Penetration of Escherichia coli O157:H7 into lettuce tissues as affected by inoculum size and temperature and the effect of chlorine treatment on cell viability. Journal of Food Protection, 63, 434–440.
LWT 40 (2007) 1266–1270 www.elsevier.com/locate/lwt
Interaction of Escherichia coli O157:H7 E318 cells with the mucus of harvested channel catfish (Ictalurus punctatus) Rico R. Suhalima, Yao-Wen Huanga,, Jinru Chenb a
Department of Food Science and Technology, University of Georgia, Athens, GA 30602, USA Department of Food Science and Technology, University of Georgia, Griffin, GA 30223 USA
b
Received 6 December 2005; received in revised form 5 August 2006; accepted 8 August 2006
Abstract Channel catfish skin with or without mucus (0.5 cm in diameter) were immersed into a suspension containing 109 CFU/ml of Escherichia coli O157:H7 E318 cells at 22 1C for 20 min. The inhibitory effect of skin mucus was determined by placing the mucus-side down on tryptic soy agar inoculated with 104–105 CFU of E. coli O157:H7 E318. The inhibition zones of fish mucus had a diameter of approximately 0.7 cm and were only visible for the first 12 h of the incubation. Bacterial cells were observed at 15 mm into the mucus layer under confocal scanning laser microscopy (CSLM). Plate counts and CSLM revealed 0.5- and 1-log less cells, respectively, attached to skin without mucus than to skin with mucus. Results suggest that E. coli O157:H7 E318 could attach to and penetrate through the mucus of channel catfish and may become a source of contamination during catfish processing. r 2006 Swiss Society of Food Science and Technology. Published by Elsevier Ltd. All rights reserved. Keywords: Channel catfish; Escherichia coli O157:H7 e318; Confocal scanning laser microscopy; Green fluorescent protein; Catfish skin
1. Introduction Outbreaks of Escherichia coli O157:H7 infections have been linked to water and a variety of foods (Doyle & Schonei, 1987; Swerdlow et al., 1992). Although fishery products have not been implicated in the outbreaks, they are no doubt potential vehicles for transmitting E. coli O157:H7 (Andrews, Wilson, & Poelma, 1977; Cotton & Marshall, 1998; Leung, Huang, & Pancorbo, 1992; Meyer & Bullock, 1973). Channel catfish (Ictalurus punctatus) grow in a light water environment. They can be contaminated by pathogenic bacteria like E. coli O157:H7 in the growing environment as a result of storm runoff from adjacent agricultural land or droppings from domestic and wild farm animals. Contamination of catfish by pathogenic microorganisms may also occur during processing. The skin and gut of catfish are believed to be the critical sources of microbial contamination (Kim & Marshall, 2002). The microorganisms from these two locations could spread, Corresponding author. Tel.: +1 706 542 1092; fax: +1 706 542 1050.
E-mail address: [email protected] (Y.-W. Huang).
contaminating the entire catfish processing environment (Bal’a, Podolak, & Marshall, 1999). As part of a natural defense system, fish secrete mucus to their surfaces. The mucus contains specific immunoglobins and lysozyme, which can inhibit microorganisms on fish skin (Speare & Mirasalimi, 1992). In addition to immunoglobins and lysozyme, the mucus of channel catfish may contain histone-like proteins (Robinette et al., 1998), which are effective against the cells of Aeromonas hydrophila and Salmonella parathyphi (Ourth, 1980). The mucus also has the ability to entrap microorganisms, preventing them from colonizing on the surface of fish by continuous sloughing (Roberts, 2001). Although the skin mucus of live catfish has the aforementioned defensive mechanisms, it is not known whether the natural function of the mucus remains after the catfish is harvested. This study was undertaken to evaluate the antibacterial effects of harvested catfish skin mucus on the cells of E. coli O157:H7, and to examine the penetration of E. coli O157:H7 cells through the mucus layer of catfish with the aid of confocal scanning laser microscopy (CSLM).
0023-6438/$30.00 r 2006 Swiss Society of Food Science and Technology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.lwt.2006.08.017
ARTICLE IN PRESS R.R. Suhalim et al. / LWT 40 (2007) 1266–1270
CSLM has been used to observe food structures (Hassan, Frank, & Corredig, 2002) and attachment of bacterial cells to food (Prachaiyo & McLandsborough, 2000; Seo & Frank, 1999; Takeuchi & Frank, 2000). The major advantage of CSLM is its ability to focus within a specific depth by optically sectioning the specimen and constructing a three-dimensional (3D) image (Amos & Fordham, 1997; Blonk & van Aalst, 1993). Different from scanning electronic microscopy, CSLM allows observation of the specimen in a fully hydrated state (Blonk & van Aalst, 1993). When used to study bacterial penetration, CSLM reveals the location and distribution of bacterial cells in addition to its ability to quantify the cells without extensive sample preparation and fixation. 2. Materials and methods 2.1. E. coli O157:H7E318 inoculum E. coli O157:H7 E318 containing a jellyfish green fluorescent protein (GFP) plasmid was used in this study. The GFP has an excitation wavelength of 488 nm, and an emission wavelength of 543 nm. The cells carrying an ampiciline resistance gene on the GEP plasmid and a nalidixic acid resistance gene on the chromosome, were grown for 18 h at 37 1C in brain heart infusion broth (BHI, Difco Laboratories, Detroit, MI, USA) supplemented with 100 mg/ml of ampicillin (Sigma, St. Loius, MO, USA) and nalidixic acid (Aldrich Chem Co., Milwaukee, WI, USA). The cells in the BHI broth were centrifuged at 5000g for 20 min, and the resulting cell pellet was washed with 0.1% buffered peptone water (Difco Laboratories). The washed cells were suspended in the same buffer, and diluted into desired concentrations. The cells were plated on BHI agar plates supplemented with 100 mg/ml of ampicillin and nalidixic acid, and grown for 18 h at 37 1C. The GFP colonies appeared on the plates were enumerated under a long-wave UV light. 2.2. Channel catfish Live channel catfish were purchased from the farmers market in Atlanta, GA and were kept in container with pumped air and transported to UGA laboratory in Athens, GA. The live fish were maintained in 40 l of aquaria at 20 1C until use. On the day of the experiment, the fish were sacrificed and fish skin was removed with a sterile surgical blade. Mucus of the fish was removed by scraping the skin surface with the surgical blade to obtain mucus-free skin samples. 2.3. Penetration of E. coli O157:H7 E318 cells through fish mucus Catfish skin samples (0.5 cm in diameter) with or without mucus were inoculated by immersion into a suspension (30 ml) of 109 CFU/ml of E. coli O157:H7 E318 cells at
1267
22 1C for 20 min. The length of time was decided according to results of the preliminary study. The contaminated fish skin samples were rinsed twice with 30 ml of sterile distilled water, each for 3 min at 22 1C. Following the rinsing, the fish skin samples were flooded for 5 min at 22 1C with 7.5 mm of propidium iodide (Molecular Probes, Inc., Eugene, OR, USA) which enabled the distinction of dead (red stained) from live (GFP) E. coli O157:H7 cells. After the treatment with propidium iodide, the fish skin samples were rinsed twice with 30 ml of sterile distilled water, each for 3 min at 22 1C to remove excessive stain. The skin samples were placed on cover slips, and the latter were inverted onto a hanging drop slide. The specimen was viewed under a CSLM Leica DM RBE (Leica Microsystems Inc., Exton, PA, USA) with 100 objective lens. The GFP cells were imaged with an argon laser at 488 nm excitation/543 nm emission, and propidium iodide was scanned with green neon at 543 nm. The reflectance mode was used to visualize mucus and skin and to aid in depth determination. Leica confocal software Version 2.0 Build 0770 (Exton, PA, USA) was used for data acquisition and analysis. E. coli O157:H7 E318 cells were enumerated at various depths along the z-axis until no GFP cells were observed. Optical slices were taken at approximately every 1 mm to avoid repeated counting of the same cells. The bacterial counts obtained from the CSLM observation was multiplied by two in order to give a similar counting effect as the two-sided fish skin surface estimated by a plate count assay. When the plate count assay was performed, each piece of the fish skin sample described above was placed into a sterile stomacher bag containing 20 ml of 0.1 ml/100 ml buffered peptone water. The samples were stomached for 2 min at high speed (Tekmar, Cincinnati, OH, USA). The resulting liquid samples were serially diluted in 0.1 ml/ 100 ml buffered peptone water prior to being surface plated in duplicate on BHI agar containing 100 mg (10 mg/100 ml) ampicillin and nalidixic acid. The plates were incubated for 24 h at 37 1C and the growing colonies were enumerated under a long-wave UV light. The colonies appeared on the plates were confirmed using a latex agglutination test (Oxoid Limited, Hampshire, England).
2.4. Inhibition of E. coli O157:H7 cells by fish mucus Approximately 104–105 CFU of E. coli O157:H7 E318 in 0.1 ml 0.1 g/100 g peptone water was surface plated with a sterile glass rod on tryptic soy agar (TSA, Difco Laboratories) plates. Thirty minutes after the inoculation, catfish skin pieces with or without the mucus were placed mucus-side down on the surface of the inoculated TSA plates. The plates were then incubated for 24 h at 37 1C. The presence of clearing zones around skin samples on the TSA agar plate was the indication of antimicrobial activity of fish mucus towards the cells of E. coli O157:H7 E318.
ARTICLE IN PRESS 1268
R.R. Suhalim et al. / LWT 40 (2007) 1266–1270
Sterile 0.5 cm filter paper discs impregnated with 5 g/100 g acetic acid were used as a control. 2.5. Statistical analysis All data counts were transformed into Log10 CFU/cm2 before comparison of the means. Analysis of the data was accomplished using the Duncan’s multiple range test of the Statistical Analysis Software (SAS; windows version 5.1.2600, 2000, SAS institute, Inc., Cary, NC, USA) based on a 95% confidence level (Po 0.05). 3. Results and discussion In the present study, the inhibition zones of fish mucus had a diameter of approximately 0.7 cm. However, they were only visible for the first 12 h of the incubation. After 18 h at 37 1C, the zones of inhibition faded due to the overgrowth of E. coli O157:H7 E318 cells surrounding the fish skin samples. This may have been due to the less susceptible nature of E. coli O157:H7 E318 cells to the antimicrobial components of the mucus or by the insufficient level of the active compounds in the mucus samples. The mode of inhibition and potency of fish mucus against bacterial cells have not been thoroughly investigated (Ourth, 1980; Robinette et al., 1998). While aquatic bacteria have varying susceptibility levels to fish mucus, the minimal inhibitory level of fish mucus against the cells of E. coli O157:H7 E318 is currently unknown. The determination of the antibacterial activity of fish mucus often implicates extensive extraction and purification of mucus components, which may have limited application in food safety. Intact skin mucus was used in the present study in order to observe the antibacterial activity on fish skin as fish is exposed to bacterial contaminants during processing. The potential of mucus as an agent against bacteria may assist in the reduction of contamination of pathogenic organisms on fish. However, mucus may lose its antibacterial activity soon after the fish is slaughtered. Certain inhibitors in blood serum greatly reduced the antibacterial activity of fish mucus (Hjelmeland, Christie, & Raa, 1983; Smith & Ramos, 1976) as fish skin could easily be contaminated with blood during processing. Proteases present in mucus could also inactivate the antibacterial compounds in fish mucus (Robinette et al., 1998). Other factors such as temperature abuse, bacterial load, and health condition of the fish may also affect the effectiveness of fish mucus in killing bacterial cells. Previous studies have shown that bacterial cells can attach and colonize on the mucus surface of harvested fish (Freter, Allweis, O’Brien, Halstead, & Macsai, 1981; Laux, Sweegan, & Cohen, 1984). Gilt-head sea brim skin mucus was found to be an optimal substrate for accumulation of aquatic bacteria such as Vibrio anguillarum and A. hydrophyla on fish (Balebona et al., 1995; Krovacek, Faris, Ahne, & Masson, 1987). Bacterial attachment to fish
mucus may be supported by the hydrophilic nature of fish skin (Hassan et al., 2002) and the nonhydrophobic surface of E. coli O157:H7 cells (Dewanti & Wong, 1995). The polysaccharide polymers of both fish mucus and bacteria were believed to aid in the adherence of bacterial cells on fish surface (Fletcher & Floodgate, 1973; Marshall, Stout, & Mitchell, 1971; Ourth, 1980; Takashima & Hibiya, 1995). In live fish however, the continual sloughing of the mucus aids in the removal of bacteria and inhibits the colonization of bacteria on fish skin. In the present study, E. coli O157:H7 E318 cells attached primarily to the area where mucus was present (Fig. 1A and Table 1), while the skin without mucus had less numbers of attached E. coli cells (Fig. 1B and Table 1). After being stained with propidium iodide, viable cells appeared green while nonviable cells stained red to yellow under CSLM (Fig. 1). Approximately 18% and 16% of the cells observed under CSLM appeared dead on skin with or without mucus, respectively. E. coli O157:H7 E318 cells were observed at areas as deep as 15 mm into the mucus layer and were decreasing in number at a deeper layer, with the majority of cells found within the first 5 mm. Free moving E. coli cells were observed in greater numbers on skin without mucus than were observed on skin with mucus. The results of the present study indicated that attachment of E. coli O157:H7 E318 cells did not occur randomly but rather in the area where mucus was abundant. E. coli O157:H7 E318 cells were able to penetrate into the mucus layer of catfish. In contrast, fish skin without mucus had fewer attached bacterial cells. The plate count results however, showed that skin without mucus contained a high number of bacterial cells. The high counts on mucus-free skin surface could have been attributed to the bacterial cells that adhered to the film layer of water on the skin after rinsing. A large number of free-moving bacteria on skin without mucus visualized under CSLM may have contributed to the high plate counts. Additionally, while the mucus had been significantly removed, a small amount of mucus might still have been present, allowing bacterial cells to attach. In contrast, the majority of E. coli O157:H7 E318 cells on the skin with mucus attached, and penetrated to a deeper layer, as visualized through the scanned images, rather than moving freely over the skin surface. In both skin types, plate counts were higher than direct counts under CSLM. The high values may have been due to the stomaching of the fish samples when plate counts were determined. Stomaching may have dislodged additional E. coli O157:H7 E318 cells that attached to the edges of the skin and to the muscle side of the skin which tends to be irregular in thickness and surface area. In addition, the disruption of cells aggregating on the skin surface during stomaching may have also contributed to the higher numbers of E. coli O157:H7 E318 cells enumerated by plate counting. Cell enumeration by direct observation was restrained to counting the cells that were not in clusters. The higher plate counts may also have been due to the ability of injured cells to grow on agar media, while the
ARTICLE IN PRESS R.R. Suhalim et al. / LWT 40 (2007) 1266–1270
1269
Table 1 Enumeration (mean7SD) by plate and direct microscopic count of E. coli O157:H7 E318 inoculated on catfish skin with and without mucus Treatment
Plate count (cells/cm2) CSLM (cells/cm2)
Skin with mucus
Skin without mucus
7.870.2aa 5.570.2a
7.370.1b 4.570.2b
a Means on the same row followed by the same letter are not significantly different.
is slaughtered. Cross contamination with pathogens, such as E. coli O157:H7, from skin to muscle is possible if the pathogen has previously been introduced during production or post-harvest processing. Proper skinning of catfish may reduce the cross contamination and/or growth of potential foodborne bacterial pathogens on mucus. Acknowledgments The authors are grateful to the University of Georgia College of Agricultural and Environmental Sciences College Station for support and to Dr. Ashraf Hassan, Department of Food Science and Technology and Dr. John Shields, the Center for Ultrastructural Research at the University of Georgia, for their valuable technical assistance with CSLM application. References
Fig. 1. Attachment and visualization, using of GFP E. coli O157:H7 E318, labeled with green fluorescent protein, on channel catfish skin (A) with mucus and (B) without mucus stained with PI (red dead cell stain) visualized under CSLM. Images were taken at 10 mm (A) and 5 mm (B). Green and yellow arrow heads indicate live and dead cells, respectively.
same cells may have appeared dead under CSLM due to penetration of propidium iodide, which has been shown to have the ability to gain access to leaky cell membranes. 4. Conclusion Mucus is a nonspecific protective barrier of fish, and is also a suitable media for bacterial attachment once the fish
Amos, W. B., & Fordham, M. (1997). An evaluation of confocal versus conventional imaging of biological structures by fluorescence light microscopy. Journal of Cell Biology, 105, 41–48. Andrews, W. H., Wilson, C. R., & Poelma, P. L. (1977). Bacteriological survey of the channel catfish (Ictalurus punctatus) at the retail level. Journal of Food Science, 42, 359–363. Bal’a, M. F., Podolak, A. R., & Marshall, D. L. (1999). Steam treatment reduces the skin microflora population on de-headed and eviscerated whole catfish. Food Microbiology, 16, 495–501. Balebona, M. C., Morinigo, M. A., Faris, A., Krovacek, Manson, K. I., Bordas, M. A., & Borrego, J. J. (1995). Influence of salinity and pH on the adhesion of pathogenic Vibrio strains to Sparus aurata skin mucus. Aquaculture, 132, 112–120. Blonk, J. C. G., & van Aalst, H. (1993). Confocal scanning light microscopy in food research. Food Research International, 26, 297–311. Cotton, L. N., & Marshall, D. L. (1998). Predominant microflora on catfish processing equipment. Dairy Food Environmental Sanitation, 18, 650–654. Dewanti, R., & Wong, A. C. L. (1995). Influence of culture condition s on biofilm formation by Escherichia coli O157:H7. International Journal of Food Microbiology, 26, 147–164. Doyle, M. P., & Schonei, J. L. (1987). Isolation of Escherichia coli O157:H7 from retail fresh market and poultry. Applied Environmental Microbiology, 53, 2394–2396. Fletcher, M., & Floodgate, M. D. (1973). An electron-microscopic demonstration of an acidic polysaccharide involved in the adhesion of a marine bacterium to solid surfaces. Journal of Genetic Microbiology, 74, 325–334. Freter, R., Allweis, B., O’Brien, P. C. M., Halstead, S. A., & Macsai, M. S. (1981). Role of chemotaxis in the association of motile bacteria with intestinal mucosa: in vitro studies. Infectious Immunology, 34, 241–249.
ARTICLE IN PRESS 1270
R.R. Suhalim et al. / LWT 40 (2007) 1266–1270
Hassan, A. N., Frank, J. F., & Corredig, M. (2002). Microstructure of feta cheese made using different cultures as determined by confocal scanning laser microscopy. Journal of Food Science, 67, 2750–2753. Hjelmeland, K., Christie, M., & Raa, J. (1983). Skin mucus protease from rainbow trout, and its biological significance. Journal of Fish Biology, 23, 13–22. Kim, J., & Marshall, D. L. (2002). Influence of catfish skin mucus on trisodium phosphate inactivation of attached Salmonella Typhymurium, Edwardsiella tarda and Listeria monocytogenes. Journal of Food Protection, 65, 1146–1151. Krovacek, K., Faris, A., Ahne, W., & Masson, I. (1987). Adhesion of Aeromonas hydrophyla and Vibrio anguilarum to mucus-coated glass slides. Federation of European Microbiological Societies Microbiology Letters, 42, 85–89. Laux, D. C., Sweegan, E. F., & Cohen, P. S. (1984). A new test based on ‘salting out’ to measure relative surface hydrophobicity of bacterial cells. Biochimica et Biophysica Acta, 677, 471–476. Leung, C. K., Huang, Y. W., & Pancorbo, O. C. (1992). Bacterial pathogens and indicators in catfish and pond environments. Journal of Food Protection, 55, 424–427. Marshall, K. C., Stout, R., & Mitchell, R. (1971). Mechanism of the initial events in the sorption of marine bacteria to surfaces. Journal of General Microbiology, 68, 337–348. Meyer, F. P., & Bullock, G. L. (1973). Edwarsiella tarda, a new pathogen of channel catfish. Applied Microbiology, 25, 155–156. Ourth, D. D. (1980). Secretory IgM, lysozyme and lymphocytes in the skin mucus of the channel catfish, Ictalurus punctatus. Developmental and Comparative Immunology, 4, 65–74.
Prachaiyo, P., & McLandsborough, L. A. (2000). A microscopic method to visualize Escherichia coli interaction with beef muscle. Journal of Food Protection, 63, 427–433. Roberts, R. J. (2001). Fish pathology. London, UK: W.B. Saunders. Robinette, D., Wada, S., Arroll, T., Levy, M. G., Miller, W. L., & Noga, E. J. (1998). Antimicrobial activity in the skin of the channel catfish: characterization of broad spectrum histone-like antimicrobial proteins. Cellular and Molecular Life Sciences, 54, 468–475. Seo, K. H., & Frank, J. F. (1999). Attachment of Escherichia coli to lettuce leaf surface and bacterial viability in response to chlorine treatment as demonstrated by using confocal scanning laser microscopy. Journal of Food Protection, 62, 3–9. Smith, A. C., & Ramos, F. (1976). Occult haemoglobin in fish skin mucus as an indicator of early stress. Journal of Fish Biology, 9, 537–541. Speare, D. J., & Mirasalimi, S. M. (1992). Pathology of the mucous coat of trout skin during an erosive bacterial dermatitis. Journal of Comparative Pathology, 106, 210–211. Swerdlow, D., Woodruff, B. A., Brady, R. C., Griffin, P. M., Tippen, S., Donnell, H. D., Jr., et al. (1992). A waterborne outbreak in Missouri of Escherichia coli O157:H7 associated with bloody diarrhea and death. Annals of Internal Medicine, 117, 812–819. Takashima, F., & Hibiya, T. (1995). An atlas of fish histology (second ed). Tokyo, Japan: Kodansha Ltd. Takeuchi, K., & Frank, J. F. (2000). Penetration of Escherichia coli O157:H7 into lettuce tissues as affected by inoculum size and temperature and the effect of chlorine treatment on cell viability. Journal of Food Protection, 63, 434–440.