Isolation of a novel transposon which carries the Escherichia coli ...

Vol. 162, No. 2

JOURNAL OF BACTERIOLOGY, May 1985, p. 615-620 0021-9193/851050615-06$02.00/0 Copyright C) 1985, American Society for Microbiology

Isolation of a Novel Transposon Which Carries the Escherichia coli Enterotoxin STII Gene LEE,'* SHIAU-TING HU,1

PAMELA J. SWIATEK,1 STEVE L. MOSELEY,2 STEPHEN D. ALLEN, AND MAGDALENE SO3 Department of Pathology, Indiana University School of Medicine, Indianapolis, Indiana 462231; Division of Injectioils Diseases, Children Orthopedics Hospital, Seattle, Washington 981052; and Department of Molecular Biology, Scripps Clinic and Research Fouindation, La Jolla, California 920373

CHAO-HUNG

Received 30 November 1984/Accepted 10 February 1985

The Escherichia coli heat-stable enterotoxin STII gene in P307 is flanked by inverted repeat sequences, suggesting that the STII gene is part of a transposon. To study the transposability, a DNA fragment containing the putative STII transposon has been cloned. Results of transposition assays indicated that the STII gene can transpose from one plasmid to another. The size of the transposon has been determined to be approximately 9 kilobases. The structure and the location of the STII gene in clinical isolates of Escherichia coli have been investigated by restriction enzyme analyses. The structural genes of STII from different clinical isolates appear to be uniform in size, but the flanking sequences are heterogeneous. This result suggests that the STII genes in different isolates are not on the same transposon as observed in P307.

Enterotoxigenic E. coli produce several types of enterotoxins which are classified as heat-stable toxin (ST) or heat-labile toxin (LT) (11, 20). The heat-labile toxin is structurally and functionally related to the cholera toxin (9, 10). Two types of ST have been identified and are distinguishable by their solubility in methanol and by animal assays (2, 14, 28). STI (also called STa) is methanol soluble and can be assayed in newborn mice or in neonatal piglets but is inactive in pigs older than 7 weeks. In contrast, STII (STb) is insoluble in methanol, and its activity can be assayed in weaned piglets or rabbits but not in infant mice (2). Currently, the activity of STII is assayed in pig ileal or jejunal loops (28). The STI gene has been cloned from human and from porcine strains of Escherichia coli (18, 24). Both human and porcine STI genes are very similar and encode a peptide of 72 amino acids (18, 24). However, the toxin purified from culture medium is composed of only 18 amino acids, which corresponds to the 3' end of the gene (4, 27). Surprisingly, a region within this portion of the STI is also found in several conotoxins that are potent neurotoxins (8). The mechanism by which STI induces electrolyte imbalance leading to clinical illness may be related to its ability to stimulate guanylate cyclase (7). The STII gene we have isolated is located on a naturally occurring plasmid, P307, found in a porcine strain of E. coli (25). This plasmid also contains genes coding for LT. The STII gene has been sequenced and was shown to code for a protein of 71 amino acids (16, 19). However, the size of the native toxin has not been determined. There is no sequence homology between the STI and STII genes, indicating that STI and STII are indeed two different toxins. The mode of action of STII is unknown. The STII gene is located between two inverted repeat DNA sequences of ca. 600 base pairs in length (17), suggesting that it is part of a transposon. In this study, we present evidence that the STII gene is indeed associated with a *

transposable element and can be transposed from one plasmid to another. MATERIALS AND METHODS Chemicals, enzymes, and isotopes. Restriction enzymes, E. coli DNA polymerase I and its large fragment, and T4 DNA ligase were obtained from New England Biolabs, Inc., Beverly, Mass. DNase I was purchased from Boehringer Mannheim Biochemicals, Indianapolis, Ind. Deoxyribonucleotides, ribonucleotides, ampicillin, tetracycline, kanamycin, chloramphenicol, and other general chemicals for making buffers were purchased from Sigma Chemical Co., St. Louis, Mo. 32P-deoxyribonucleotides and [32P]ATP were obtained from Amersham Corp., Arlington Heights, Ill. The concentrations of antibiotics used in this study are 50 ,ug/ml for ampicillin, 10 p.g/ml for tetracycline, 50 pLg/ml for chloramphenicol, 25 p.g/ml for streptomycin, and 30 p.g/ml for kanamycin. Bacterial strains and plasmids. E. coli DH-1 (F- recAl endoAl gyrA96 thi-1 hsdR17 supE) is the host of all the recombinant plasmids constructed in this study. E. coli 711 is a K-12 strain and contains the plasmid encoding the STII and LT genes which was transferred into this strain from P307 (21). All E. coli isolates containing the STII gene were obtained from the National Animal Disease Center, Ames, Iowa. E. coli W3100 is F- hsdR- hsdM' (3). MV12 is recA trp thr leai lacY B1- siupE (13). RSF2001 is a derivative of the F plasmid containing the kanamycin resistance gene from the R plasmid pML21 (12). Plasmid DNA preparation. Plasmid DNA was isolated from E. coli culture by using the alkaline lysis method (1). The cells were digested with lysozyme and lysed by sodium dodecyl sulfate in the presence of 0.2 N NaOH. The plasmid DNA was then neutralized by the addition of sodium acetate (pH 4.8) to a final concentration of 0.3 M. The chromosomal DNA was then pelleted by centrifugation. Plasmid DNA was recovered by ethanol precipitation from the clarified supernatant. Southern blot analysis. The agarose gel containing electrophoretically separated DNA fragments was treated with

Corresponding author. 615

616

LEE ET AL.

~BamHI

J. BACTERIOL.

a

L

I

I

b

Ii %

o

H

1 Kb

Hinfl

a

HindEm

B

STI

PstI

14%

HindM

IIIHinfl Hindfm[iN Pstl JBamH1

gene

Probe

500 bp

FIG. 1. (a) Restriction map of the DNA fragment containing the STII transposon. The boxed areas indicate the location of the terminal repeat sequences. (b) Map of restriction enzyme sites of the 3.2-kb ClaI-BamHI fragment of P307 containing the STII gene. This fragment was cloned into pBR322 between the ClaI and BamHI sites. The resulting plasmid was designated as Cla3. The STII gene was located between the Hinfl and PstI sites. The Hinfl-PstI fragment was used as the probe in this study.

0.2 N HCl, followed by 0.5 N NaOH in 5 x SSC (1 x SSC is 15 mM sodium citrate plus 150 mM NaCI). The gel was then placed in a solution containing 3 M NaCl and 1 M Tris-hydrochloride (pH 8.0). The DNA was then transferred from the gel to nitrocellulose paper according to the procedure of Southern (26). The nitrocellulose paper was incubated at 80°C for 2 h in vacuo and then prehybridized with Denhardt solution to block the excessive DNA binding sites (6). Hybridization was performed in a solution containing the radioactively labeled probe, 10% dextran sulfate, 5x SSC, and 1Ox Denhardt solution at 55°C. The nitrocellulose paper was then washed with 0.1x SSC containing 0.1% sodium dodecyl sulfate at 70°C. RESULTS Isolation of the putative transposable elements associated with the STII gene. The location of the STII gene and the two

terminal repeat sequences on P307 was determined by Picken and co-workers (19) (Fig. 1). The plasmid designated as Cla3 contains the ClaI-BamHI fragment which is ca. 3.2 kilobases (kb) in length (16). This fragment of DNA consists of the left-hand end of the terminal repeat sequence and the STII gene. The other end of the terminal repeat sequence is located on an adjacent 7-kb BamHI fragment of P307. To clone the entire transposon, the 7-kb BamHI fragment was isolated and inserted at the unique BamHI site of the plasmid Cla3. The resulting plasmid, designated as pPS1, is shown in Fig. 2a. This plasmid contains a 10-kb insert containing the putative STII transposon in the same orientation as in P307. The two terminal repeat sequences were first discovered by heteroduplex analysis by using electron microscopy (17). These findings suggest that the two terminal regions have homologous sequences. However, restriction enzyme analysis indicated that XhoI and PvuI sites are present on the

_

FIG. 2. (a) pPS1 is a pBR322 derivative containing a 10-kb insert in which the STII gene is located. (b) pPS5 was derived from pPS1. A chloramphenicol resistance gene was inserted at the unique BglII site within the STII structural gene.

STII TRANSPOSON

VOL. 162, 1985

617

N

to -

~~~N

X z S3 e en cn t X X

C-

0 0 0 X CL af a a a: a af a a af

m.'pPS1 (Form II) pCHL6_ (Form II) pBR322 (Form II) pPS1 (Form I) pCHL6_ (Form 1)

:--pBR322 (Form 1)

FIG. 3. Autoradiogram of pST plasmids hybridized to the STII gene probe. pST plasmids were coelectrophoresed with control plasmids on an 1% agarose gel. pPS1 and pCHL6 were used as the positive controls, whereas pBR322 was the negative control. Southern blot analysis was performed after the electrophoresis by using a 32P-labeled STII structural gene as the probe.

left-hand region but not the right-hand terminal region (Fig. 1). These observations showed that the sequences of these two ends are not identical. Determination of the transposability of the STII gene. To determine whether the putative STII transposon is transposable, transposition assays were performed. To detect the transposition event, a chloramphenicol resistance gene was inserted into the unique BglII site which is located within the structural gene of the STII (see Fig. 1). The 1.3-kb HaeIl fragment which contained the chloramphenicol gene was isolated from pACYC184 (5). The resulting plasmid, which was designated as pPS5 (Fig. 2b), was then transformed into E. coli MV12/RSF2001. This E. coli strain contains an F factor (RSF2001) with a kanamycin resistance gene. Transformants were selected on LB agar plates containing ampicillin, chloramphenicol, and kanamycin. These transformants were then conjugated with E. coli W3110. The exconjugants which are resistant to streptomycin, kanamycin, and chloramphenicol were selected. The donor strain, MV12/RSF2001, is sensitive to streptomycin, whereas the recipient strain, W3110, is resistant to streptomycin. The only organisms which would be resistant to streptomycin, kanamycin, and chloramphenicol are (i) those W3110 cells which obtained the F factor with both kanamycin and chloramphenicol resistance markers and (ii) those W3110 cells containing the F factor and the plasmid pPS5. To distinguish these two possibilities, W3110 organisms with a streptomycin, kanamycin, and chloramphenicol resistance phenotype were tested for ampicillin sensitivity. If the chloramphenicol gene was transposed from pPS5 into the F factor, these organisms should be sensitive to ampicillin. On the other hand, if pPS5 was mobilized into W3110 cells, these colonies should be resistant to ampicillin since pPS5 contains an ampicillin resistance gene. Among the 300 colonies tested, 299 of them were sensitive to ampicillin; only 1 of them was resistant to ampicillin. These results indicate that the chloramphenicol gene located within the structural gene of STII was indeed transposed from pPS5 into the F factor.

To further demonstrate that the STII gene is indeed transposable, the following experiments were performed. W3110 organisms exhibiting a chloramphenicol, kanamycin, and streptomycin resistance and ampicillin sensitivity phenotype were transformed with plasmid pBR322 so that the STII transposon would then transpose into pBR322. If the STII transposon was in the pBR322 plasmid, not only would the STII gene be present in sufficient numbers in the cell to allow easy detection by DNA-DNA hybridization, but it also would render pBR322 resistant to chloramphenicol. Therefore, after the transformation, the transformants were plated on LB agar plates containing chloramphenicol, kanamycin, ampicillin, and streptomycin. A single colony was then picked and inoculated into 100 ml of LB broth containing the same antibiotics. The culture was grown at 30°C with shaking for 18 h. The cells were pelleted, and the plasmid DNA was isolated. This DNA was then used to transform E. coli DH-1. Transformants were selected on LB agar plates containing ampicillin and chloramphenicol. Under this condition, only those transformants containing plasmids with both ampicillin and chloramphenicol resistance genes will grow. The only possibility for pBR322 conferring the chloramphenicol resistance phenotype is by obtaining the chloramphenicol resistance gene from the F factor via DNA transposition mechanism. Approximately 10,000 transformants which were resistant to ampicillin and chloramphenicol were obtained per microgram of DNA used for transformation. Plasmid DNA was then isolated from seven randomly picked colonies. These plasmids were designated pST1 through pST7 (henceforth referred to as pST plasmids). To demonstrate that the STII gene was transposed together with the chloramphenicol gene, pST plasmid DNAs were electrophoresed on an 1% agarose gel. pPS1 and pCHL6 (16) were included to serve as the positive controls, whereas pBR322 was used as the negative control. Southern blot analysis was then performed by using the STII structural gene as the probe. All seven pST plasmids, as well as pPS1 and pCHL6, hybridized to the STII gene probe (Fig. 3). Furthermore, the sizes of the pST plasmids are similar to

618

LEE ET AL. r-

CD

0 cw C.

N

J. BACTERIOL.

Xo

co

r

Lfl

O)

q-

VI,

I-

t%%

1%% IL) CC) CC) l%~ (JD tI-

CY) CD

N

heterogeneous and that the STII gene in these cases is not on the same transposon as observed in P307. The STII transposon in P307 has a unique BgIII site which is located within the STII structural gene. A Southern blot analysis of BglIIdigested plasmid DNA in most cases yields two fragments which hybridize to the probe. However, in P307, 1738, and 263, only one BglII fragment was detected by the STII gene probe (Fig. 5). In the latter cases, it is quite possible that the two BglII fragments have identical sizes, so that they comigrate as one band. The heterogeneity of the BglII fragments in these isolates also indicates that the sequences surrounding the transposon are heterogenous. DISCUSSION

:1.

WNM&.

li.k

4*O':.

FIG. 4. The Hinfl fragment of various clinical E. coli isolates hybridized to the STII gene probe. The plasmid DNAs were digested with Hinfl, electrophoresed, and then hybridized to the STII gene probe.

that of pPS1, suggesting that the size of pST plasmids is ca. 14 kb and that the sequences transposed onto pBR322 are of the same length. To eliminate the possibility that the chloramphenicol gene itself can be transposed, plasmid pACYC184 was likewise transformed into strain MV12/RSF2001. The transformants were conjugated with E. coli W3110. Colonies which had been resistant to chloramphenicol, kanamycin, and streptomycin were then tested for tetracycline sensitivity, since pACYC184 carries a tetracycline resistance gene. All of the 500 colonies tested were resistant to tetracycline. This result indicates that the chloramphenicol gene did not transpose into the F factor. Rather, the entire pACYC184 plasmid was mobilized into W3110. Analysis of the STII gene in STII producing E. coli clinical isolates. Eight E. coli clinical isolates (1726, 1738, 1558, 1697, 1755, 115, 1657, and 263) were identified as STII producers by the pig ileal loop assay and shown to harbor the STII gene by colony hybridization by using as the probe the 32P-labeled STII gene (data not shown). Plasmid DNA was isolated from these strains and used for a series of Southern hybridization studies to determine whether these STII sequences were also part of the same transposon, as observed for the STII gene in P307. As shown in Fig. lb, Hinfl cleaves immediately adjacent to the 5' and 3' ends of the STII structural gene, releasing the STII gene as a 500-base-pair fragment. A Southern blot of the Hinfi-digested plasmid DNA from these isolates with an STII gene probe shows that in all cases the STII gene of these plasmids is encoded by a 500-base-pair Hinfl fragment (Fig. 4). The STII gene can also be released from the transposon as a 1.3-kb HindIll fragment (Fig. lb). The plasmid DNAs of strains 1726, 1738, 1558, 115, and 263 were digested with HindIll and subjected to analysis in a similar manner. Results from this experiment showed that strain 263 contains STII-encoding HindIll fragments of the same size as that of P307 (1.3 kb). The plasmids from the remaining clinical isolates contain HindIll fragments which differ in size as compared with the corresponding fragment in P307 (Fig. 5). Results from the above experiments suggest that the STII flanking sequences on these other plasmids are

We have shown that the STII gene in the plasmid P307 is part of a transposable element. Our assays show that this STII element can transpose from pBR322 to the F plasmid RSF2001 and from RSF2001 back to pBR322. The size of the pST plasmids is ca. 14 kb. Considering that the size of the pBR322 is 4 kb and that the chloramphenicol resistance gene, inserted into the STII gene, is 1 kb, the size of the STII transposon can be deduced as 9 kb, which is roughly the size of the region containing STII flanked by the inverted repeats. Although the structure of the STII transposon has not been characterized, results of our transposition assays

Hind Ill

Bcyl 11 ....

coX0

r

w

0

Nr rl. b

c

r

-

) LO Ln , a)

m

>

(O o

N .

....

Lo

co

Nt-

Lt)'-

L

N

-

-

9.8

U

4 2

a"

--2 7

S

S0

-2.1

13

1. .

FIG. 5. The HindlIl or BglII fragments of various clinical E. coli isolates hybridized to the STII gene. Plasmids isolated from clinical isolates of E. coli were digested with HindIll or BglII. The digested DNAs were electrophoresed on an 1% agarose gel. Southern hybridization was performed by using the STII gene as the probe. The numbers appearing at the top of the figure are the strain numbers of clinical isolates, whereas the numbers on the right side represent the size (in kb) of markers.

STII TRANSPOSON

VOL. 162, 1985

strongly suggest that the STII gene is part of a transposon. The observation that the two inverted repeats of the STII transposon are not identical suggests that the element could be a composite transposon like TnIO (15). Our results suggest that the element transposes predominantly by a simple transposition mechanism, since cointegrates have not been detected. It remains to be determined whether transposition is driven by one or both terminal repeats and which terminal sequence plays the major role in this event. Our analyses of the STII gene and of the flanking sequences in plasmids from several E. coli STII-producing clinical isolates suggest that the transposon is present in similar form in strain 263. The Hindll fragments containing the STII gene in plasmids from the remaining strains are different in size. One possibility for this finding is that additional DNA has been added to the region within the HindIll segment in these other isolates. Another possibility is that the STII genes in strains other than P307 are carried by different transposons or not carried by transposon at all. We are currently investigating these possibilities. The hybridization results of Bglll digestion also reveal that strain 263 may contain the same plasmid as observed in P307. All the other strains show different hybridization patterns. This evidence suggests that the STII genes in different isolates are carried by different plasmids or are located at different places in the same plasmid. Our demonstration on the transposability of the STII gene isolated from P307 indicates that DNA transposition may be one of the possible mechanisms for this heterogeneity. We have previously reported that the gene encoding the STI toxin is part of a transposon (TnI681) flanked by inverted repeats of IS] (22, 23, 24). We have now observed that, at least in some STII producers, the STII gene is also a transposable element. That the same ST gene is observed in plasmids of various sizes in E. coli isolated from different animal sources strongly suggests that transposition is a mechanism used by enterotoxigenic E. coli for dissemination of these virulence factors. The LT gene found in the plasmid 10407 also was flanked by inverted repeat DNA sequences (29). Although there is no evidence to date to suggest this, it is interesting to speculate that the genetic determinant for LT may also be transposable. ACKNOWLEDGMENT This project was supported by Public Health Service Biomedical Research Support grant S S07 RR5371 and grant 1 R01 A121428-01 to C.H.L. from the National Institutes of Health.

1. 2.

3.

4.

5.

LITERATURE CITED Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7:1513-1523. Burgess, N. M., R. J. Bywater, C. M. Cowley, N. A. Mullan, and P. M. Newsome. 1978. Biological evaluation of a methanol-soluble, heat-stable Escherichia coli enterotoxin in infant mice, pigs, rabbits, and calves. Infect. Immun. 21:526-531. Campbell, J. L., C. C. Richardson, and F. W. Studier. 1978. Genetic recombination and complementation between bacteriophage T7 and cloned fragment of T7 DNA. Proc. Natl. Acad. Sci. U.S.A. 75:2276-2280. Chan, S. K., and R. A. Giannella. 1981. Amino acid sequence of heat-stable enterotoxin produced by Escherichia coli pathogenic for man. J. Biol. Chem. 256:7744-7746. Chang, A. C. Y., and S. N. Cohen. 1978. Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the plSA cryptic miniplasmid. J. Bacteriol.

619

134:1141-1156. 6. Denhardt, D. T. 1966. A membrane filter technique for the detection of complementary DNA. Biochem. Biophys. Res. Commun. 23:641-646. 7. Field, M., L. Graf, W. Laird, and P. Smith. 1978. Heat-stable enterotoxin of Escherichia coli: in vitro effects on guanylate cyclase activity, cyclic GMP concentration, and ion transport in small intestine. Proc. Natl. Acad. Sci. U.S.A. 75:2800-2804. 8. Gray, W. R., A. Luque, B. M. Olivera, J. Barrett, and L. J. Cruz. 1981. Peptide toxins from Conus Geographus venom. J.

Biol. Chem. 256:4734-4740. 9. Guerrant, R. L., L. L. Brunton, T. C. Schnaitman, L. I. Rebhun,

and A. G. Gilman. 1974. Cyclic adenosine monophosphate and alteration of Chinese hamster ovary cell morphology: a rapid, sensitive in vitro assay for the enterotoxins of Vibrio cholerae and Escherichia coli. Infect. Immun. 10:320-327. 10. Gyles, C. L. 1971. Heat-labile and heat-stable forms of the enterotoxin from E. coli strains enteropathogenic for pigs. Ann. N.Y. Acad. Sci. 176:314-322. 11. Gyles, C. L. 1974. Immunological study of the heat-labile enterotoxins of Escherichia coli and Vibrio cholerae. Infect. Immun. 9:564-569. 12. Heffron, F., P. Bedinger, J. J. Champoux, and S. Falkow. 1977. Deletions affecting the transposition of an antibiotic resistance gene. Proc. Natl. Acad. Sci. U.S.A. 74:702-706. 13. Hershfield, V., H. W. Boyer, C. Yanofsky, M. A. Lovett, and D. R. Helinski. 1974. Plasmid ColEl as a molecular vehicle for cloning and amplification of DNA. Proc. Natl. Acad. Sci. U.S.A. 71:3455-3459. 14. Kapitany, R. A., G. W. Forsyth, A. Scoot, S. F. McKenzie, and

R. W. Worthington. 1979. Isolation and partial characterization of two different heat-stable enterotoxin produced by bovine and porcine strains of enterotoxigenic Escherichia coli. Infect. Immun. 26:173-177. 15. Kleckner, N. 1981. Transposable elements in prokaryotes. Annu. Rev. Genet. 15:341-404. 16. Lee, C. H., S. L. Moseley, H. W. Moon, S. C. Whipp, C. L. Gyles, and M. So. 1983. Characterization of the gene encoding heat-stable toxin lI and preliminary molecular epidemiological studies of enterotoxigenic Escherichia coli heat-stable toxin Il producers. Infect. Immun. 42:264-268. 17. Mazaitis, A. J., R. Maas, and W. K. Maas. 1981. Structure of a naturally occurring plasmid with genes for enterotoxin production and drug resistance. J. Bacteriol. 145:97-105. 18. Moseley, S. L., J. W. Hardy, M. I. Huq, P. Echeverria, and S. Falkow. 1983. Isolation and nucleotide sequence determination of a gene encoding a heat-stable enterotoxin of Escherichia coli. Infect. Immun. 39:1167-1174. 19. Picken, R. N., A. J. Mazaitis, W. K. Maas, M. Rey, and H. Heyneker. 1983. Nucleotide sequence of the gene for heat-stable enterotoxin II of Escherichia coli. Infect. Immun. 42:269-275. 20. Sack, R. B. 1975. Human diarrheal disease caused by enterotoxigenic Escherichia coli. Annu. Rev. Microbiol. 29:333-353. 21. Santos, D. S., S. Palchaudhuri, and W. K. Maas. 1975. Genetic and physical characteristics of an enterotoxin plasmid. J. Bac-

teriol. 124:1240-1247. 22. So, M., R. Atchison, S. Falkow, S. Moseley, and B. J. McCarthy. 1981. A study of the dissemination of Tn1681: a bacterial transposon encoding a heat-stable toxin among enterotoxigenic Escherichia coli isolates. Cold Spring Harbor Symp. Quant. Biol. 45:53-58. 23. So, M., F. Heifron, ansd B. J. McCarthy. 1979. The E. coli gene encoding heat-stable toxin is a bacterial transposon flanked by inverted repeats of ISl. Nature (London) 277:453-456. 24. So, M., and B. J. McCarthy. 1980. Nucleotide sequence of the bacterial transposon TnI681 encoding a heat-stable (ST) toxin and its identification in enterotoxigenic Escherichia coli strains. Proc. Natl. Acad. Sci. U.S.A. 77:4011-4015. 25. Soijka, W. T., M. K. Lloyd, and E. J. Sweeney. 1960. Escherichia coli serotypes associated with certain pig diseases. Res. Vet. Sci. 1:17. 26. Southern, E. M. 1975. Detection of specific sequences among DNA fragments separated by gel electrophoresis. J. Mol. Biol.

620

LEE ET AL.

98:503-517. 27. Staples, S. J., S. E. Asher, and R. A. Giannelia. 1980. Purification and characterization of heat-stable enterotoxin produced by a strain of E. coli pathogenic for man. J. Biol. Chem. 255:4716-4721. 28. Whipp, S. C., H. W. Moon, and R. A. Argenzio. 1981. Compar-

J. BACTERIOL.

ison of enterotoxic activities of heat-stable enterotoxins from class 1 and class 2 Escherichia coli of swine origin. Infect. Immun. 31:245-251. 29. Yamamoto, T., and T. Yokota. 1981. Escherichia coli heat-labile enterotoxin genes are flanked by repeated deoxyribonucleic acid sequences. J. Bacteriol. 145:850-860.