Rapid detection of a new cell-associated Salmonella cytotonic enterotoxin in cell sonicates by Chinese hamster ovary cells

FEMS Microbiology Letters 20 (1983) 201-206 201 Published by Elsevier

Rapid detection of a new cell-associated Salmonella cytotonic enterotoxin in cell sonicates by Chinese hamster ovary cells

(Keywords not submitted)

Suraj B. Baloda, A h m e d Faris , Ka re l K r o v a c e k and To rke l Wads t r6 rn *

Department of Bacteriology and Epizootology, College of Veterinary Medicine, Swedish University of Agricultural Sciences, Biomedicum, Box 583, S-751 23, Uppsala, Sweden

Received 23 June 1983 Accepted 30 June 1983

1. I N T R O D U C T I O N

Salmonella strains belonging to different sero- types have been isolated and implicated as causa- tive agents of diarrhoea [1-7]. Unlike enterotoxi- genic Escherichia coli, the enterotoxins from Salmonella species responsible for initiating di- arrhoeal disease have been difficult to detect by different bioassays partly due to the cytotoxic substances produced by different strains, and partly because salmonellae may produce low levels of enterotoxins when cultured in broth by conven- tional methods [8]. It is not surprising, therefore, to note that reports from different laboratories on the detection of Salmonella enterotoxins do not correlate [8-13], and the importance of enter- otoxins in Salmonella infection is still controver- sial.

The purpose of this study is to define a rapid method designed to detect Salmonella enterotoxins in bacterial sonicates. The observations that Salmonella sonicates contain high level Chinese hamster ovary (CHO) cell elongation factor, coin- cide with recent observations in our laboratory that higher levels of heat-labile enterotoxin (LT)

* To whom correspondence should be addressed.

from E. coli are found in cell lysates or sonicates than in cell supernatants [14,15].

2. MATERIALS A N D M E T H O D S

2.1. Bacterial strains

Strains of S. enteritidis, S. typhimurium and S. richmond were isolated from human patients with diarrhoeal disease, and a guinea pig (Table 1) and were stored either on brain heart infusion agar kept at 4°C or preserved in trypticase soy broth containing glycerol (15 % v /v ) at - 70 °C.

2.2. Growth conditions

Cultures were grown on blood agar, nutrient agar, MacConkey agar, desoxycholate citrate agar, endo agar or lactose agar at 37°C. Casamino acid Yeast Extract (CYE) liquid medium [11] was used for growing the Salmonella cultures in shake flasks (37°C, 20 rev. /min) . The CYE grown cultures were centrifuged (18000 × g, 30 min) and the su- pernatants were sterile filtered (Millipore ~, 0.22 ~m). To determine 'percent ' intracellular and ex- tracellular content of Salmonella enterotoxin, 1 ml of CYE culture was centrifuged and the super-

0378-1097/83/$03.00 © 1983 Federation of European Microbiological Societies

202

natant was tested for activity on CHO cells, whereas the cell pellet was dissolved in 1 ml of phosphate-buffered saline (PBS, 0.01 M sodium phosphate buffer, pH 7.4 in 0.14 M NaC1), soni- cated and tested on CHO cells.

2.3. Preparation of sonicates

Strains grown on different agars were harvested using sterile cotton swabs and suspended in 5 ml of sterile PBS in glass centrifuge tubes. Each sus- pension was sonicated (25 k c / s amplitude) for 30 s and 15, 30, 60, 120 and 300 s using an MSE sonicator (MSE Scientific Instruments, UK). The longer sonication times were used to test the minimum sonication time required for best amounts of toxin release and to test the stability of the toxin(s). After sonication, cell suspensions were centrifuged at 3000 x g for 30 min and the super- natants were sterile filtered (Millipore ~, 0.22/~m). Fresh bacterial suspensions of the strains listed in Table 1 were also sonicated using another model of sonicator, trough-type Elgasonic sonicator (25 kc / s , Elgasonic, S.A., Bienne) in which a large number of tubes can be sonicated at the same time using an ice bath, which saves time by avoiding individual sonication of each tube in case large numbers of samples are used.

2.4. Enterotoxin assay

Salmonella enterotoxins in the cell sonicates as well as in supernatants of CYE cultures were assayed using CHO cells in tissue culture [16].

2.5. Coagglutination test

A loopful of bacteria from the blood agar were suspended in 100 /~l of PBS containing 2 m g / m l of Polymyxin B and incubated in a water bath at 37°C for 1 h. Cell lysates were then prepared by adding Triton X-100 (0.1% final concentration) to the sample and the mixture was further incubated for 10 min until clear lysates were obtained. The lysates were then centrifuged (3000 × g, 20 min). 25/zl of the concentrated clear lysate was added to 25 /~1 of a formalin-killed Staphylococcus aureus strain Cowan 1 coated with E. coli LT antibodies

[15]. The mixture, on a glass slide or a Phadebact cellulose card (Pharmacia Diagnostics, Uppsala, Sweden) was mixed by rocking the slide or the cardboard for 2 rain. Presence of bacterial clumps or aggregation between the clear lysate and the coagglutination reagent indicated the presence of Salmonella toxin homologous to E. coli LT.

2.6. Neutralization experiments

The Salmonella enterotoxin samples, cholera toxin and E. coli LT were preincubated at 37°C for 3 h with standard amounts of high titer cholera and E. coli LT antisera. The antibody preparation used for neutralization could detect as low as 1 ng of cholera toxin and 5 ng of E. coli LT in the coagglutination test when S. aureus strain Cowan 1 was coated with homologous antisera [15]. Neu- tralization tests were then performed on the rabbit skin using the standard rabbit skin permeability assay [ 17].

2.7. Chemicals

Casamino acids and yeast extract were pur- chased from Difco, Detroit, MI, USA. Tissue cul- ture media were from Flow Laboratories, Scot- land. Coagglutination reagents were kindly sup- plied by B. ROnnberg and cholera toxin and corre- sponding antiserum were kindly donated by the Swiss Serum Institute, Basel, Switzerland.

3. RESULTS

3.1. Detection of Salmonella enterotoxin I~v CHO cells in tissue culture

As shown in Table 1 and Fig. l b, sonicated samples contain high levels of Salmonella entero- toxin corresponding to culture supernatants. De- tection and titration of this enterotoxin in the presence of CHO cells indicated that the sonicated material possessed detectable enterotoxic activity even when the sonicates were of high dilutions (1 : 2048). The presence of Salmonella enterotoxin activity in CYE supernatants was lower (approx. 24%) as compared to the sonicates. The CYE

203

T a b l e 1

C o m p a r a t i v e r e sponses o f Salmonella cel l-free cu l t u r e s u p e r n a t a n t s a n d cell son ica tes as de tec t ed b y Ch inese h a m s t e r o v a r y ( C H O )

cells in t issue cu l tu re

S t ra in O r g a n i s m Source E longa t i on of C H O cells by: ~

Cu l tu r e s u p e r n a t a n t s b

( 1 : 32 d i lu t ion)

Cell son ica tes b,c

(1 : 100 d i lu t ion)

15879 S. enteritidis H u m a n + + + + + +

5 3 / 8 0 S. richmond H u m a n - (lysis) d + +

- S. typhimurium H u m a n + + + + +

27657 S. enteritidis H u m a n + + + + +

2 8 7 / 8 0 S. typhimurium H u m a n + + + + + + +

25279 S. enteritidis H u m a n + + + +

24952 S. enteritidis H u m a n + + + +

7 4 / 8 0 S. enteritidis H u m a n - ( l y s i s ) + +

27036 S. enteritidis H u m a n + + + + + + +

30612 S. enteritidis G u i n e a pig - (lysis) + +

27655 S. enteritidis H u m a n + + +

a R e f e r to the m o r p h o l o g i c a l c h a n g e s o f the C H O cells 18 h a f t e r a d d i t i o n of samples . + + + + = 90% cells e longa t ed , + + + =

7 5 - 9 0 % cells show e longa t ion , + + = 2 5 - 7 5 % cells a re e longa t ed a n d + = u p to 25% of the cells are e longa ted .

The two- fo ld d i lu t ions were tes ted f r o m the s amp le s o b t a i n e d f r o m equa l a m o u n t s o f cell cu l tu res (i.e. 1 ml) r e p r e s e n t i n g a n equal

cell mass (see M A T E R I A L S A N D M E T H O D S ) .

c A 30-s pulse was used to p r e p a r e the sonica tes .

d T o t a l lysis o f cells is o b s e r v e d (see Fig. lc).

T h e resul ts a re ave rages o f three exper iments .

5¸¸¸¸

iiii~2~i~i~!ii,i~i!iTiiii~il;~i~iiii~¢~;ill ¸~ ; U~;!,ii~i ~i,~ ̧, ~i ii!ii~i!~i~ii~i~i,ii~ii!Ji!i~ili!i~i~ ~̧I~L;,~ ' ii' ili~!iU ~i~ii~ii~!ii~!i'~i ~ ~ i'

41

L

LJ

Fig. 1. (a) CHO cells in tissue culture: Normal untreated (negative control) cells. (Magnification 1000 × ). (b) Elongated CHO cells after contact with sonicates of strain 287/80 at 1 : 100 dilution titer. (Magnification 1000 × ). (c) Complete lysis of CHO cells observed in the 1:32 dilution of culture supernatants of strain 53/80. (Magnification 1000 × ).

supernatants indicated presence of enterotoxic ac- tivity in two fold dilutions of up to I : 32. CYE supernatants of the strains which produce cyto- toxic factors are weakly enterotoxigenic, and often cause CHO cells to lyse or be deformed in dilu- tions of up to 1 : 100 (Fig. lc) [8]. When tested for percent release of intracellular and extracellular activity (see MATERIALS AND METHODS), sonicates of CYE culture pellets gave CHO cell elongation even at 1 : 2048 dilution (25% of CHO cells show- ing elongation or a " o n e + " activity) and the respective CYE supernatants were " + " up to the dilutions of 1 : 128.

A minimum of 15 s was sufficient to release detectable quantities of enterotoxin. However, when various sonication time periods were used, a time lapse of 30 to 60 seconds gave best toxin release (Table 1 and Fig. lb). The sonicates were potent and toxin was stable even at longer sonica- tion times (300 s).

Sonicates prepared from agar-grown cultures indicated that blood-agar culture sonicates were superior in their Salmonella enterotoxin levels as determined by the extent of CHO cell elongation caused by them as compared to the CYE super- natants. Maximum CHO cell activity was seen in sonicates of cells from blood agar followed by endo agar, desoxycholate citrate agar, MacConkey

Table 2

Comparative efficiency of various solid culture media in reveal- ing S a l m o n e l l a enterotoxin activity in cell sonicates as detected by elongation of Chinese hamster ovary (CHO) cells in culture

Culture medium CHO cell elongation titer a.b

Strain 287/80 Strain 27036

Blood agar 2 048 2 048 MacConkey agar 128 128 Nutrient agar 128 128 Deoxycholate citrate agar 256 512 Endo agar 512 512 Lactose agar 64 64

a S a l m o n e l l a sonicates grown on different agar media were used in two-fold dilutions (1:2, 1:4, 1:8, 1:16 and so on) and tested for activity on CHO cells.

b CHO cell elongation titer represents the last serial titer at which at least 25% of CHO cells are elongated, showing ' + ' activity.

The results are the averages of three experiments.

205

agar, nutrient agar and lactose agar, respectively (Table 2).

A trough-type water-bath sonicator (Elgasonic) was equally effective in facilitating the release of enterotoxin from the culture suspensions. Hence, small volumes and a large number of samples can be sonicated in this type of instrument at the same time without affecting the efficiency of release, and saving the individual sonication of each tube containing the sample. Thus, the sonication pro- cess is independent of the type of instrument used, and is effective in releasing the enterotoxin.

Tests performed for detection of Salmonella enterotoxin with the coagglutination test indicate that Salmonella enterotoxin(s) (strain 287/80 and 27036) does not cross-react immunologically with cholera toxin or E. coli LT. Neutralization tests on the rabbit skin also showed no immunological cross-reaction with either of these high titer anti- sera.

4. DISCUSSION

Bacterial sonicates from enterotoxigenic E. coli have been used as a source of heat-labile enter- otoxin [14,18] which is predominantly cell-bound [ 14,19]. In addition to using bacterial cell sonicates our laboratory has employed detergents in lysing cells for LT release [15].

Sonicates of strains of Salmonella isolated from man and animals with diarrhoea exhibit high levels of enterotoxins which were easily detected by CHO cells in tissue culture. The sonicates did not cause cytotoxic effects on CHO ceils as did the cell supernatants of strains 53/80, 74/80 and 30612 (Table 1). It has earlier been reported that culture filtrates of certain strains of E. coli had cytotoxic effects on Vero cells [20]. This toxin was termed as 'Vero toxin' but it had however, no effect on CHO cells, and was distinct from E. coli LT. Our ob- servations with Salmonella culture supernatants of the three strains indicate that cytotoxic factors are released into the medium and they do cause lysis of CHO cells. Hence the cytotoxic factors in Salmonella are probably not cell-bound (or if pre- sent, they are in very low amounts) unlike Shigella cytotoxin which is obtained after lysis of cells [21].

206

Presence of cytotoxic factors in the culture super- natants of Salmonella may lead to 'masking' of enterotoxic effects of Salmonella enterotoxin if it is produced in low amounts by the organism (Fig. lc), leading to the reporting of false negative re- sults.

Cells harvested from blood agar for sonication, contained higher levels of enterotoxin as compared to supernatants of CYE grown cultures (Table 1). Blood agar was followed by endo agar, des- oxycholate citrate agar, MacConkey agar, nutrient agar and lactose agar, in decreasing order of their efficiency of influencing the production of detec- table amounts of Salmonella enterotoxin.

In conclusion, it was observed that sonicates could be used for detection of Salmonella enter- otoxin activity in blood-agar-grown cultures and approx. 60% of the crude Salmonella enterotoxin activity was cell-bound, which is released by soni- cation. This method can be used in the diagnostic laboratories for general screening of enterotoxi- genie Salmonellae, directly from blood agar plates. The method is sensitive enough to detect low levels of Salmonella enterotoxins from 'weak' enterotoxi- genie strains of Salmonella.

A C K N O W L E D G E M E N T S

We are thankful to Dr. S.N. Saxena, Director, Central Research Institute, Kasauli, H.P., India and to Mr. F.Y. Soboh, Department of Microbiol- ogy, Lady Hardinge Medical College, New Delhi, India, for the supply of Salmonella strains. We are grateful to Mr. B. R6nnberg for kindly supplying us with coagglutination reagents. This study was supported by grants from the Swedish Medical Research Council (16 × 04723) and Swedish Board for Technical Development. Suraj B. Baloda holds a scholarship from The Swedish Institute.

REFERENCES

[1] Barnum, D.A. (Ed.) (1977) International symposium on Salmonella and prospects for control. University of Guelph, Guelph, Ont., Canada.

[2] Kelterborn, E. (1979) Zentralbl. Bakteriol. Parasitenkd. Infektionkr. Hyg. Abt. I Orig. Reihe A 243, 289-307,

[3] Prost, E. and Reimann, H. (1967) Annu. Rev. Microbiol. 21,495-528.

[4] Sanstedt, K., Gunnarsson, A., Hurvell, B., Nordblom, B., Rutquist, L. and S6derlind, O. (1980) Nord. Veterinarmed. 32, 57-74.

[5] Taylor, J. (1968) Symp. Zool. Soc. London 24, 51 73. [6] Todd, E.C.D. (1978) J. Food Protect. 41,559-565. [7] Turnbull, P.C.B. (1979) Clinics Gastr-enterol. 8, 663-714. [8] Baloda, S.B., Faris, A., Krovacek, K. and Wadstrbm, T.

(1983) Toxicon, in press. [9] Koupal, L.R. and Deibel, R.H. (1975) Infect. hnmun. 11,

14-22. [10] Caprioli, A., D/ignolo, G., Falbo, V., Roda, L.G. and

Tomasi, M. (1982) Microbiologica 5, 1-10. [11] K~tyi, I., P~csa, S., Em6dy, L., Vert6nyi, A., Kocsis, B.

and Kuch, B. (1979) Acta. Microbiol. Acad. Sci. Hung. 26, 217-223.

[12] Molina, N.C. and Peterson, J.W. (1980) Infect. lmmun. 30, 224-230.

[13] Peterson, J.W. (1980) Pharmac. Ther. 11,719-724. [14] Krovacek, K., Faris, A. and Wadstri3m, T. (1983) Zentralbl.

Bakteriol. Hyg. 1 Abt. Orig., in press. [15] R6nnberg, B and WadstrtSm, T. (1983) J. Clin. Microbiol.,

in press. [16] Guerrant, R.L., Brunton~ L.L., Schnaitman, T.C., Rebhun,

L.I. and Gilman, A.G. (1974) Infect. Immun. 10, 320-327. [17] Craig, J.P. (1965) Nature 207, 614-616. [18J Olsvik, O. and Berdal, B.P. (1982) Acta Pathol. Microbiol.

Immunol. Scand. Sect. B 90, 319-321. [19] Wadstr~Sm, T., S6derlind, O., Mollby, R. and Wretlind, B.

(1974) IRCS 2, 1321. [20] Konowalchuk, J., Speirs, J.I. and Stavric, S. (1977) Infect.

Immun. 18, 775-779. [21} Gentry, M.K. and Dalrymple, J.M. (1980) J. Clin. Mi-

crobiol. 12, 361-366.