Enterotoxigenic Escherichia coli and Staphylococcus aureus in fish and seafood from the southern region of Brazil

ELSEVIER

International Journal of Food Microbiology 24 (1994) 171-178

International Journal of Food Microbiology

Enterotoxigenic Escherichia coli and Staphylococcus aureus in fish and seafood from the southern region of Brazil Andres Mane R o m e r o Ayulo a,b,* Ruben Abreu Machado b, Vildes Maria Scussel b a Laboratorio de Microbiologia, Instituto Tecnoldgico Pesquero, AP 10-0360, Callao, Per~ b Departamento de Ciencia e Tecnologia de Alimentos, Centro de Ciencias Agrarias, Universidade Federal de Santa Catarina, C.P. 476, Floriandpolis, SC, Brazil

Abstract

An investigation to evaluate the microbiological condition and safety of fish and seafood commonly harvested at the coast of Santa Catarina State and sold in Florian6polis was undertaken. One hundred and seventy-five samples of fish and fish fillets (Cynoscion leiarchus), shrimp tails (Peneaus paulensis), shellfish-meat (Anomalocardia brasiliensis and Metilus edulis), and crab-meat (Callinectes sapidus) were collected from markets and examined within 4 h of purchase. For isolation and enumeration of Escherichia coli the methods used were those of Speck et al. (1975) (Method 1) and Fishbein et al. (1976) (Method 2); for S. aureus, methods recommended by the U.S. Food and Drug Administration were used including biochemical identification of the strains. E. coli was more frequently detected with Method 1 than Method 2. Of 317 E. coli strains tested for STG and LT II toxins, only one (isolated from shellfish-meat) produced ST and none produced LT II toxin. S. aureus was isolated from 20% of 175 samples examined, including 60% of samples of shellfish-meat. Only nine of 109 S. aureus strains produced enterotoxins, including enterotoxin A (4), D (1) and AB (4). It is concluded that greater care must be taken to reduce contamination of fish and seafood during harvesting and post-harvest handling. Keywords: Escherichia coli; Staphylococcus aureus; Fish; Seafood; Contamination; Enterotoxigenicity

* Corresponding author. 0168-1605/94/$07.00 © 1994 Elsevier Science B.V. All rights reserved SSDI 0168-1605(94)00078-K

172

A.M. Romero .4yulo et al. / International Journal of Food Microbiology 24 (1994) 171-178

1. Introduction

Pathogenic Escherichia coli and Staphylococcus aureus are common food-borne infectious and toxigenic agents in several countries (Cliver, 1987). Enterotoxigenic E. coli (ETEC) produces toxins which cause diarrhoea and often occurs in people arriving to a country for the first time (Sack and Sack, 1975; Jay, 1986). The diarrhoea is characterised mainly by dehydration and acidosis resulting from excessive loss of fluid and electrolytes. Apart from infants, where 50% mortality has been reported, and elderly people, incapacitation is the most serious problem for sick individuals (Sack and Sack, 1975; Boylans et al., 1988). Detection and isolation of E T E C may be difficult because the gene for toxin production is generally carried on a plasmid, which may be lost because of temperature conditions, use of inhibitors in culture media, and subsequent culturing (Mehlman and Romero, 1982). E T E C produces both heat labile (LT) and heat stable (ST) enterotoxins which differ with respect to heat resistance and chemical structure. Two types of ST are produced: STa from human and bovine sources, which is active in neonatal piglets (1 to 3 days old) and in infant mice, and STb, which is active in weaned pigs and in the rabbit ileal loop test (Jay, 1986; Trabulsi and Toledo, 1989). Also two types of LT are produced, LT I of human and porcine origin, coded by a plasmid gene, and LT II of human origin, coded by a chromosomal gene, which unlike LT I is not neutralized by anticholera toxin (Guth et al., 1986). Staphylococci can be present in raw food of animal origin or in those handled by man. S. aureus is the most important pathogen among the staphylococci (Jay, 1986). During growth many strains of this species produce enterotoxins which can cause staphylococcal food-poisoning. Seven different enterotoxins have been identified: A, B, C l, C 2, C 3, D and E. Enterotoxin A (SEA) is the most frequently involved in food-poisoning (Bergdoll, 1990), with as little as 100 to 200 ng causing illness (Evenson et aI., 1988). This type of food-poisoning is characterized by vomiting and diarrhoea that occurs two to six hours after ingestion of food containing one or more of the enterotoxins. The present investigation was undertaken to examine fish and seafood commonly found on the coast of Santa Catarina State, and sold in FIorian6polis, Southern Brazil, for the presence of enterotoxigenic E. coli and S. aureus

2. Materials and methods

Samples. One hundred and seventy-five samples of fish and fish fillet (Cynoscion leiarchus), shrimp tails (Peneaus paulensis), shellfish meat (Anomalocardia brasiliensis and Metilus edulis) 30 samples each and crab-meat (Callinectes sapidus) (25 samples). All are commonly harvested from the coast of Santa Catarina State and sold in the markets of Florian6polis. The samples after purchase were

A.M. Romero Ayulo et al. / International Journal of Food Microbiology 24 (1994) 171-178

173

introduced in sterile bags and transported into insulated thermic boxes with ice to maintain a temperature around 4°C. Samples were transported to the Microbiology Laboratory of the Food Science and Technology Department-UFSC and examined within 4 h of collection. Microbiological analysis. For preparation and dilution of the samples, we followed methods recommended by the U.S. Food and Drug Administration (FDA, 1992). Isolation and enumeration of E. coli were done according to Fishbein et al. (1976) (Method 1) and Speck et al. (1975) (Method 2), using the same diluted sample. Both methods are recommended for isolation of enterotoxigenic strains of E. coli. The same diluted sample (25 g food in 225 ml 0.1% peptone, followed by 10 g to 90 ml dilutions in 0.1% peptone) was used for both tests. The Fishbein method is an MPN techniqe, inoculating the dilutions directly into MacConkey broth, incubating 24-48 h at 37°C and subculturing all lactose-positive tubes into lauryl tryptose. All tubes showing gas after 24-48 h at 37°C were subcultured onto eosin methylene blue (EMB) agar and typical colonies confirmed as E. coli using conventional biochemical tests. The Speck method includes a resuscitation step, overlaying pour-plates of tryptose soy agar after 2 h at 20-25°C with violet red bile lactose agar and incubating a further 18 h at 37°C. Representative colonies are subcultured into brilliant green bile broth and all tubes showing gas are subcultured onto EMB agar and confirmed as E. coli using similar methods to the Fishbein technique. Isolation and enumeration of S. aureus was done by surface plating on Baird Parker agar, as recommended by the FDA (1992); tests on hybrid colonies for catalase, coagulase using rabbit plasma, deoxyribonuclease and thermonuclease (Lachica et al., 1971), and the Voges-Proskauer reactions were used to characterize the strains. Detection orE. coli toxins. E. coli was streaked onto nutrient agar and incubated at 37°C for 24 h; it was subcultured in Evans broth (Evans et al., 1976) and incubated in a shaker incubator (Technal, model T E 120) at 37°C overnight. The culture was centrifuged at 3000 rpm for 30 rain, and the supernatant screened for LT II and STb. The indirect haemagglutination method, described by Ricci and Castro (1986) was used to detect LT II. The sera and anti-sera were obtained from Dr. Lucila C. Ricci (UNICAMP, Campinas, SP, Brazil) and prepared according to the instructions of the manufacturer. The test described by Dean et al. (1972), using infant mice, was employed for detection of STb. Detection of S. aureus enterotoxins. The sac culture method, described by Donnelly et al. (1967) was used with brain heart infusion broth as the growth medium. Cultures were incubated at 37°C for 24 h. The cultures were then centrifuged and supernatants tested for enterotoxins A to D by the optimum sensitivity plate (OSP) methods of Robbins et al. (1974). Reference enterotoxins and specific antibodies were obtained through the courtesy of Dr. Merlin S. Bergdoll, Food Research Institute, University of Wisconsin, Madison, USA.

174

A.M. Romero Ayulo et al. / International Journal of Food Microbiology. 24 (1994) 171-I 78

3. Results

Isolation and enumeration of E. coli. E. coli was detected in 37.7% of samples using Method 1 and 24% of samples using Method 2. Results are given in Table 1. Higher numbers of E. coli were detected in more samples when using Method l (Table 2) as compared with Method 2 (Table 3). Detection of enterotoxigenic E. coli. Only one strain of E. coli isolated from shellfish-meat was found to produce toxin (STb). No strains produced LT II toxin (Table 4).

Table 1 Isolation of E. coli from fish and seafood using the Fishbein and Speck methods Sample

Fish Fillet fish Shrimp tails Crab-meat Shellfish-meat Total

N u m b e r of samples

Positive samples (%)

analysed

Method 1 a

Method 2 b

9 (30.0) 8 (26.7) 12 (40.0) 13 (52.0) 19 (63.0) 5 (16.7)

3 (10.0) 6 (20.0) 10(33.7) 7 (28.0) 12 (40.0) 4 (13.3)

66 (37.7)

42 (24.0)

30 30 30 25 30 c 30 d 175

a Fishbein method. b Speck method. CAnomalocardia brasiliensis. d Metilus edulis.

Table 2 R a n g e of positive samples for Escherichia coli using the Fishbein method Sample

R a n g e of Escherichia coli c f u / g (%) 10°_101

101-102

102-103

Fish Fillet fish Shrimp tails Crab-meat Shellfish-meat

8 (26.7) 4 (13.3) 5 (16.7) 9 (36.0) 13 (43.3) a 3 (10.0) b

1 (3.3) 3 (10.0) 4 (13.3) 2 (8.0) 2 (6.7) _

1 (3.3) 1 (3.3) 1 (4.0) 1 (3.3) _

2 (6.7) 2 (6.7) 1 (3.3)

1 (4.0) 1 (3.3) l (3.3)

Total

42 (24.0)

12 (6.9)

4 (2.3)

5 (2.8)

3 (1.7)

Anomalocardia brasiliensis. b Metilus edulis.

103-104

> 104 -

A.M. Romero Ayulo et al. / International Journal of Food Microbiology 24 (1994) 171 -178

175

Table 3 R a n g e of positive s a m p l e s for Escherichia coli using the S p e c k m e t h o d Sample

Fish Fillet fish S h r i m p tails Crab-meat Shellfish-meat

Total

R a n g e of Escherichia coli c f u / g (%) 10°-101

101-102

102-103

103-104

1 (3.3) 1 (3.3) a

1 (3.3) 2 (6.7) 1 (4.0) _

1 (3.3) 1 (3.3) 1 (4.0) 4 (13.3)

_ b

1 (3.3)

1 (3.3)

-

2 (6.7)

2 (1.1)

5 (2.9)

8 (4.5)

18 (10.3)

9 (5.1)

2 4 6 3 3

> 104

(6.7) (13.3) (20.0) (12.0) (10.0)

1 (3.3) 2 (8.0) 4 (13.3)

aAnomalocardia brasiliensis. b Metilus edulis.

Table 4 F r e q u e n c y of e n t e r o t o x i g e n i c Escherichia coli ( E T E C ) in fish a n d seafood Sample

N u m b e r of

No. strains

Positive strains

samples

analysed

L T I I toxin

S T G toxin

Fish Fillet fish S h r i m p tails Crab-meat Shellfish-meat

9 8 12 13 19 a 5 b

31 38 64 47 110 27

-

1 (0.9%) -

Total

66

317

-

1 (0.3%)

aAnomalocardia brasiliensis. b Metilus edulis.

Isolation and identification of S. aureus. S. aureus was isolated from 20% of the 175 samples examined (Table 5), with 60% of positive samples being shellfish-meat ( Anomalocardia brasiliensis and Metilus edulis ) and 20% crab-meat. Table 5 R a n g e of positive s a m p l e s for Staphylococcus aureus Sample

Positive sample

R a n g e of Staphylococcus aureus c f u / g ( % ) 101-102

102_103

103-104

> 104

Fish Fillet fish S h r i m p tails Crab-meat Shellfish-meat

2 4 1 7 14 a 7 b

1 1 1 1

1 2 1 2

-

(3.3) (6.7)

1 (4.0) 1 (3.3) 2 (6.7)

Total

35

4 (2.3)

6 (3.4)

4 (2.3)

21 (12.0)

aAnomalocardia brasiliensis. b Metilus edulis.

(3.3)

(3.3) (3.3) (3.3)

(3.3) (6.7)

1 (3.3) 1 (3.3) 6 (24.0) 11 (36.7) 2 (6.7)

176

A.M. Romero Ayulo et al. / lnternational Journal of Food Microbiology 24 (1994) 171-178

Table 6 Frequency of enterotoxigenic S. aureus and type of toxin produced Sample

Positive samples

Fish Fillet fish Shrimp tails Crab-meat Shellfish-meat

2 4 1 7 14 ~ 7 b

Total

35

No. strains analysed 4 7 3 30 47 18 109

No. strains producing toxin

Type of toxin (No. strains)

1

D ( I)

6 2

A (2), AB (4) A (2)

9

A (4), AB (4), D (1)

a Anomalocardia brasiliensis. b Metilus edulis.

Detection of enterotoxigenic S. aureus. Of 109 S. aureus strains, only 8.3% produced enterotoxin (Table 6). Four strains from crab-meat produced SEA plus SEB and two strains SEA only; two strains from shellfish-meat (Anomalocardia brasiliensis) produced SEA and one from fish produced SED.

4. Discussion

In our work, the prevalence of E. coli using Method 1 was 30% in fish and 26.7% in fish fillets, and in 10% and 20%, respectively, by Method 2. These numbers were higher than those reported for other countries (Foster et a1.,1977; Cliver, 1987). However, the present results are similar to those obtained by Vieira (1987), also in Brazil, who detected E. coli in 16.9% samples of sardine (Sardiniella brasiliensis) examined. Similarly, Cann (1974) detected low numbers of E. coli in tropical shrimp in the United Kingdom and Foster et al. (1977) reported the organism in only 5.9% of shrimp surveyed in USA, whereas 40 and 33.7% of samples were positive in this study, using Methods 1 and 2, respectively. The high incidence in Florian6polis may be explained by the relatively crude manner in which shrimp is captured and manipulated before reaching the market. As far as shellfish-meat and crab-meat are concerned, the number of positive samples for E. coli was much higher in this study than in that of Foster et al. (1977). This difference may be explained by (a) the manner of harvesting on the sea shore, where, in some cases, city waste was present, and (b) mishandling of fishery products before going to the market. The most probable number method (MPN) used in Method 1 was particularly useful for detecting low levels of the organism, this may be the main reason for most of the extra positive samples. It is possible that in some instances the presence of E. coli was masked in Method 1 by competing organisms; this would explain the extra positive samples detected using Method 2. The differences obtained when comparing Methods 1 and 2, were 66 positive samples by Method 1

A.M. Romero Ayulo et al. /International Journal of Food Microbiology 24 (1994) 171-178

177

and only 42 by Method 2 (Table 1); however, both methods seem to be necessary to obtain the maximum number of positive samples. The low incidence of enterotoxigenic E. coli (0.3% of strains) was not unexpected because E. coli is rarely found in the intestinal tract of cold blooded animals. Previous investigators reported only low numbers of ETEC in fish and seafood (Sack and Sack, 1975; Sack et al., 1977), with the exception of Matches and Abayta (1983) who isolated large numbers of ETEC during a food-poisoning outbreak involving shellfish. Some of the E. coli strains studied here may have produced either or both STa or LT I; however, these toxins were sought in this study. The number of positive samples of S. aureus (20%) in this survey was higher than others. In another study, only 8% of seafood samples was positive for S. aureus (Jay, 1986); this difference could reflect the quality of the water from which the fish were taken and the post-harvest handling procedures. In our work, most of the positive samples contained high numbers of S. aureus, particularly shellfish and crab-meat in which enterotoxigenic strains were detected. The ingestion of these contaminated samples risks of staphylococcal food-poisoning and may be considered as a public health hazard. It is possible that most contamination came from human handling, since strains producing SEA and SEB are most frequently of human origin. Some of the strains may have produced low amounts of enterotoxin that could not be detected by the OSP method (Kokan and Bergdoll, 1987). From the results of this study it can be concluded that greater care must be taken during harvesting and post harvesting of fish and seafood in order to reduce contamination.

Acknowledgement A.M. Romero Ayulo is grateful to Mrs B.C.R. Vieira for her supervision, support and suggestions on the first stage of this work.

References Bergdoll, M.S. (1990) Analytical methods for Staphylococcus aureus. Int. J. Food Microbiol. 10, 91-100. Boylans, M., Coleman, D., Scott, J. and Smyth, C. (1988) Molecular cloning of the plasmid-located determinants for CS1 and CS2 fimbriae of enterotoxigenic Escherichia coli of serotype 06:K15;H16 of human origin. J. Gen. Microbiol. 134, 2189-2199. Cann, D.C. (1974) Bacteriological aspects of tropical shrimps, In: R. Kreuser (Editor) Fishery Products. Fishing New Books, West Byfleet, Surrey, England, pp. 388-440. Cliver, D.O. (1987) Food borne disease in the United States, 1946-1986. Int. J. Food Microbiol. 4, 269-277. Dean, A.G., Ching, Y.G., Willians, R.G. and Harden, L.B. (1972) Test for Escherichia coli enterotoxin using infant mice. Application in a study of diarrhoea in children in Honolulu. J. Infect. Dis. 125, 407-411.

178

A.M. Romero Ayulo et al. / lnternational Journal o f Food Microbiology 24 (1994) 171-178

Donnelly, C.B., Leslie, J.F., Black, A.L. and Lewis, K.N. (1967) Serological identification of enterotoxigenic staphylococci from cheese. Appl. Microbiol. 15, 1382-1387. Evans, D.G., Jr., Evans, D.G., Richardson, J.H. and Gorbach, S.L. (1976) Purification of the polymyxin-release heat-labile enterotoxin of Escherichia coli. J. Infect. Dis. 133, s97-s102. Evenson, M.L., Hinds, M.W., Berntein, R.S. and Bergdoll, M.S. (1988) Estimation of human dose of staphylococcal enterotoxin A from a large outbreak of staphylococcal food poisoning involving chocolate milk. Int. J. Food. Microbiol. 7, 311-316. FDA (Food and Drug Administration) (1992) Bacteriological Analytical Manual, 7th edition. AOAC International, Arlington, VA. Fishbein, M., Mehlman, I.J., Chugg, L. and Olson, J.C., Jr. (1976) Coliforms, fecal coliforms, E. coli. In: M. Speck (Editor) Compendium of methods for the microbiological examination of foods. American Public Health Association, Washington, DC, pp. 277-300. Foster, J., Fowler, J.L. and Dacey, J. (1977) A microbial survey of various fresh and frozen seafood products. J. Food Prot. 40, 300-303. Guth, B.E.C., Pickett, C.L., Twiddy, E.M., Holmes, R.K., Gomes, T.A.T., Lima, A.A.M., Guerrant, R.L., Franco, D.G.M. and Trabulsi, L.R. (1986) Production of type II heat-labile enterotoxin by Escherichia coli isolated from foods and human feces. Infect. Immunol. 54, 587-589. Jay, J.M. (1986) Modern Food Microbiology, 3rd edition. Van Nostrand Reinold, New York, pp. 440-442. Kokan, N.P. and Bergdoll, M.S. (1987) Detection of low enterotoxin producing Staphylococcus aureus strains. Appl. Environ. Microbiol. 53, 2675-2676. Lachica, R.V., Genigeorgis, V.C. and Horprich, P.D. (1971) Metachromatic agar diffusion methods for detection of staphylococcal nuclease activity. Appl. Microbiol. 21, 585-587. Matches, J. R. and Abayta, C. (1983) Indicator organisms in fish and shellfish. Food Techonol. 37, 114-117. Mehlman, J. and Romero, A. (1982) Enteropathogenic Escherichia coli methods for recovery from food. Food Technol. 36, 73-79. Ricci, L.C. and Pestana de Castro, A.F. (1986) Indirect haemagglutination test for the detection of thermolabile (LT) enterotoxin from Escherichia coli. Med. Microbiol. Immunol. 175, 251-260. Robbins, R., Goulds, S. and Bergdoll M.S. (1974) Detecting the enterotoxigenicity of Staphylococcus aureus strains. Appl. Microbiol. 28, 946-950. Sack, D.A. and Sack, R.B. (1975) Test for enterotoxigenic Escherichia coli using Y-I adrenal cells in miniculture. Infect. Immunol. 11, 234-236. Sack, R.B., Sack, D.A., Mehlman, l.J., Orskov, F. and Orskov, I. (1977) Enterotoxigenic Escherichia coli isolated from food. J. Infect. Dis. 135, 313-317. Speck, M.L., Ray, B., and Read, R.D. (1975) Repair and enumeration of injured coliform by plating procedure. Appl. Environ. Microbiol. 23, 545-550. Trabulsi, L.R. and Toledo, M.R.F. (1989) In: Microbiologia. 2da ed. Livraria Atheneu, Rio de Janeiro/Sfio Paulo, pp. 149-155. Vieira, B.C.R. (1987) Detec~o e enumeraqS.o de bacterias de significancia ~ satide pgblica em sardinha (Sardiniella brasiliensis, Staindachner, 1789) in natura consumida na regi~o da grande Floriandpolis, Santa Catarina. M.Sc. thesis, Unicamp, Silo Paulo.