CURRENTMICROBIOLOGYVol. 23 (f991), pp. 307-3t3

CuM,~nt

Microh|o|ogy © Springer-Verlag New York Inc. 1991

Catabolite Repression of the Colonization Factor Antigen I (CFA/I) Operon of Escherichia coli Tuomo K. Karjalainen, 1 Dolores G. Evans, z Doyle J. Evans, Jr.,: David Y. Graham, 2 and Chao-Hung Lee 1 ~Department of Pathology, Indiana University School of Medicine, Indianapolis, Indiana; and 2Department of Medicine, Baylor College of Medicine and Veterans Administration Medical Center, Houston, Texas, USA

Abstract. Experiments were performed to study whether the synthesis of the fimbrial colonization factor antigen I (CFA/I) of enterotoxigenic Escherichia coli is affected by glucose. The CFA/I-producing strain H-I0407 (078 : H11 : CFA/I) was grown in CFA medium containing various concentrations of glucose. Addition of 1% glucose into the medium resulted in a pronounced decrease in CFA/I production by H-10407 as assessed by ELISA, hemagglutination, and electron microscopy. The repressive effect of glucose was reversed by the addition of 10 mM cAMP to the medium. Examination of the promoter sequence of the cfaA gene of the CFA/ I operon revealed a consensus binding site for the catabolite activator protein-cAMP complex. With a reporter plasmid containing a fusion of the cfaA promoter, a portion of the cfaA gene, and the lacZ gene, it was shown that the activity of this promoter was influenced by glucose. In a wild-type E. coIi strain, addition of 0.5% glucose to the growth medium diminished the promoter activity more than 70%. The cfaA promoter also exhibited a lower level of activity in cya (adenyl cyclase) and crp (cAMP receptor protein) mutants than in the wild-type strain. The addition of l 0 mM cAMP resulted in a marked increase in the expression from the cfaA promoter in the cya but not in the crp mutant. These results suggest that the suppressive effect of glucose in the CFA/I system is mediated via the mechanism of catabolite repression through the cfaA promoter of the CFA/I operon.

Enterotoxigenic Escherichia coli (ETEC) is a causative agent of travelers' and infantile diarrhea [16, 17]. It possesses two essential virulence factors, namely, colonization or adherence factors and enterotoxins. The proteinaceous colonization factors are usually assembled on the cell surface as fimbriae, which mediate the attachment of ETEC to the intestinal cell receptors. The attached bacteria multiply and produce enterotoxins, leading to secretory diarrhea [9]. Four types of fimbriat colonization factors of human ETEC have been identifed and characterized: CFA/I [9, 11], CFA/II [4, 8, 32], CFA/III [19], and CFA/IV [33, 34]. CFA/I is composed of a single peptide of 147 amino acids [18, 21]. CFA/II encompasses three distinct antigens: CS1, CS2, and CS3; however, CFA/II-positive ETEC usually expresses only two of the three antigens [4, 32]. Similar to CFA/II, CFA/IV encompasses three antigens,

namely CS4, CS5, and CS6 [34]. ETEC isolated from domestic animals also produces distinct colonization factors such as K88 [20], K99 [31], 987P [26], and F41 [6]. Many studies have demonstrated that expression of virulence factors including fimbriae is complex and tightly regulated. Environmental factors, such as temperature, iron, and carbon source, have been shown to influence the expression of a number of virulence factors [9, 10, 15, 30]. Several virulence factor genes have been described to be subject to catabolite repression, including pap and S fimbriae ofuropathogenic E. coli [15, 30], STI [24], and LT enterotoxins [14] of ETEC, and ETEC CFA/II fimbriae [13]. Catabolite repression is due to inactivation ofadenyl cyclase when glucose is transported into cells [I], resulting in the lack of production of cAMP that is required to produce the cAMP-CAP (catabolite activator protein) complex. This complex is an activator

Address reprint requests to: Dr. Chao-Hung Lee, Department of Pathology, Indiana University School of Medicine, 1120 South Drive, FH 419, Indianapolis, IN 46202, USA.

308

CURRENT MICROBIOLOGY WoI. 23 (1991)

Table 1. Effect of glucose and cAMP on the production of CFA/I by ETEC H-10407 grown in CFA medium Antigen production Type of medium CFA CFA CFA CFA CFA CFA

agar agar agar agar agar agar

alone + 0.2% glucose + 1% glucose + 1% glucose + 10 mM cAMP + 2% glycerol + 0.5% Na acetate

ELISA inhibition" 3.040 1.756 0.263 2.887 4.244 3.833

+-+-0.209 + 0A01 -+ 0.197 +- 0.225 +- 0.255 + 0.189

MRHA h 4+ 3+ 1+ 4+ 4+ 4+

" The concentration of antigen is given in mg/101° CFU. Values are means of four different experiments. b A 4+ reaction indicates a complete and instantaneous hemagglutination (HA) involving alt erythrocytes. Slower or less complete HA was graded 3 + , 2 + , I + ,or 0 (10). Values are means of four different experiments.

of various genes [ 1,5, 7]. It binds to specific upstream regulatory sites of promoters, thereby influencing initiation of transcription [5, 15, 27]. We have focused our study on CFA/I and have previously isolated the gene encoding the subunit of the CFA/I fimbriae [21, 22]. This gene is now referred to as cfaB [18]. Analysis of the nucleotide sequence upstream of craB revealed the presence of another gene referred to as cfaA; its function remains to be determined [18]. The cfaA promoter does not contain a typical " - 3 5 " promoter sequence, suggesting that transactivating factors are required for its expression [36]. A transactivator gene of cfaA has indeed been isolated [29]. The production of CFA/I fimbriae is also affected by environmental factors [9, 11]. In this communication, we report the suppressive effect of glucose on the expression of CFA/I fimbriae. We present evidence that this suppression is via the mechanism of catabolite repression and that the cfaA promoter is the target for this regulation. Materials and Methods Bacterial strains and plasmids. ETEC strain H-I0407 [9] was used to study CFA/I fimbriat production. E. coli K12 strain HB101 [3] was used as the host for cloning [28] and MH 1000 [37] as the wildtype host for/3-galactosidase assays. The cya mutant strain M* 7141 is F - , lacZam(AP2246), lacPr(UV5), relA, spcR, lysA(29), cyaA(854), argE: : Tnl0, ilv : : Tn5. The crp mutant strain XE65.2 is AlacX(74), Acq)(39), thi, strA. Construction of pcfaA-lacZ containing a cfaA-lacZ fusion was as follows. The cfaA promoter and the first 88 codons of cfaA were liberated from pBR322/C [21] as a 526-base IlinclI-PstI fragment and cloned in frame with the llth codon of lacZ located in the promoter-less pORF1 vector [37]. Thus, in the resulting plasmid pcfaA-lacZ, the cfaA promoter drives the expression of lacZ.

Growth media. The CFA/I-positive strain H10407 was grown in CFA medium, which is a modified CYE medium [I0]. It is corn-

posed of 1% casamino acids and 0.15% yeast extract plus 0.005% MgSO 4 and 0.0005% MnCl 2 (pH 7.4). For fl-galactosidase assays, bacterial strains were grown in CFA medium with addition of glucose or cAMP as indicated.

Enzyme-linked immunosorbent inhibition assay. For determination of the amount of CFA/I antigen produced by H-10407, 10/zl of an overnight broth culture was spread on the surface of an agar plate, and the plates were incubated at 37°C overnight. The bacteria were then scraped off.the plate and suspended in PBS (137 mM NaC1, 6.4 mM Na2HPO 4, 0.88 mM KH2PO4, 2.7 m M KC1). The number of colony-forming units per ml of the suspension was determined by viable plate counts, and the concentration of the CFA/I antigen was determined by the ELISA inhibition method as described previously [12]. Hemagglutination assays. The presence of functional CFA/I timbriae on bacteria was assayed by the mannose-resistant hemagglutination assay by use of bovine erythrocytes (MRHA). HA was graded as 4 +, 3 + , 2 + , 1 + , or 0, as described previously [103. Electron microscopy. Samples were placed on carbon-coated colloidin grids, air dried, negatively stained with 1% phosphotungstic acid (pH 6.8), and then examined with a transmission microscope (JEOL model JEM 100B) for fimbrial production [9]. /3-Galactosidase assays. Overnight cultures of bacteria were reinocutated into CFA broth at 1 : 40 dilution and grown with vigorous aeration at 37°C in the presence or absence of glucose for 2 h. When indicated, 10 mM cAMP was added to the log-phase cultures, and the incubation was continued for additional 2 h. Cultures were then harvested and resuspended in Z buffer [25]. The density of the bacterial suspensions was adjusted to ODr00 = 0.6, and fl-galactosidase assay was performed as described by Miller [251.

Results Effect of glucose and cAMP on the production of CFA/I fimbriae. In our earlier attempts to isolate large amounts of the CFA/I fimbrial subunit, we first

T.K. Karjalainen et al.: Catabolite Repression of CFA/I

309

Fig. 1. Effect of glucose on CFA/I production as assessed by electron microscopy. Electron micrographs of negatively stained cells of E. coil H-10407 grown on CFA agar in the absence (A) and in the presence (B) of 1% glucose are shown. When glucose is absent, fine filaments (fimbriae) which extend from the periphery of the majority of cells are observed (A). The cells are essentially devoid of fimbriae when glucose is added to the growth medium (B). Original magnification, x 20,000.

grew the CFA/I-positive H-10407 strain in synthetic C F A medium without glucose. Since bacterial yields were rather low in this medium, various concentrations of glucose (0.2, 0.5, 1.0, and 2.0%) were added to improve the growth of bacteria. Although the more glucose was added the better was the bacterial

yield, the amount of C F A / I antigen p r o d u c e d was found to be inversely proportional to the amount o f glucose added to the medium (Table 1). This result was obtained with E L I S A inhibition assays [12], which were used to determine the amount of C F A / I antigen produced by the H-10407 strain grown at

310

Kpnl

I

CURRENT

-265 Hincll Clal

I

Pstl

Clal

I

I

t

EcoRI BamHl

I

I

Hinctl Clal

I

I

CIaB

CfaA

I

I CfaC

"///////////////A

Pstl

Beta-gal

Clal cAMP-CAP CATCGATAC,CA/~TAACACAGGFrACACAATGATTTGTTTGC_AI-fGTCTIL{t I I I I I i ATCTCATTTT

-10

SD

lilt iGiit iGAITI'ATFATI"GAT~GC.-~GGANkT

ATG CAT MET HIS

~ LYS

FrA LEU

TTC TAT PHE TYR

MICROBIOLOGY Vol. 23 (1991)

Fig. 2. Restriction map of the cfaA-cfaB region of the CFA/I operon. Relevant restriction sites are indicated as are the positions of the first three open reading frames of the CFA/I operon, CfaA, CfaB, and CfaC (only portion of CfaC is shown). The number above HincII restriction site indicates the relative position of this restriction site to the CfaA translational initiation codon ATG. The. darkened bar indicates the cfaA promoter sequence cloned in front of lacZ in pORF1 vector (see Materials and Methods). Below the map is given the location of a potential cAMP-CAP binding site in the cfaA promoter. Only the promoter region and the first six codons of CfaA are shown [18]. The consensus motif is underlined, and the two well-conserved pentamer motifs are marked in bold face. The most likely " - 10" promoter consensus sequence is indicated. SD = ShineDalgarno sequence.

different glucose concentrations. MRHA was then performed to assay the production of functional fimbriae, and the results parallelled those of ELISA. MRHA became negative when 1% glucose was present in the medium (Table 1). When H-10407 was grown in CFA medium containing carbon sources other than glucose, e.g., 2% glycerol or 0.5% sodium acetate, no suppressive effect on CFA/I production by the two carbon sources was observed, as determined by ELISA inhibition and MRHA (Table 1). This result suggested that the suppressive effect by glucose on CFA/I production may be due to catabolite repression. To prove this possibility, cAMP was added to the CFA medium containing 1% glucose at a final concentration of 10 mM. As seen in Table 1, addition of 10 mM cAMP relieved glucose suppression on CFA/I production. Finally, electron microscopy was performed to examine whether fimbriae were present on the surface of cells grown on CFA agar under different conditions. As shown in Fig. 1A, the majority of bacteria were fimbriated when grown in the absence of glucose but were devoid of fimbriae when glucose was added to a concentration of 1% (Fig. 1B). Addition of 10 mM cAMP in the latter medium reversed the suppressive effect of glucose, so that fimbriae were produced and could be observed by electron microscopy (data not shown). These results further confirmed the suppressive effect of glucose on CFA/ I fimbrial production.

cfaA promoter of the CFA/I operon is responsive to a transcriptional regulator (CfaR) of the operon [29]. To determine whether the cfaA promoter is the target for catabolite repression in the CFA/I system, its activity was examined in the presence and absence of glucose and cAMP. For this purpose, the promoter region of cfaA was fused to the structural gene for/3-galactosidase (Fig. 2). Initially, pcfaAlacZ was introduced into E. coli K12 strain MH1000 (crp +, cya +) and grown in CFA medium in the presence and absence of glucose and assayed for/3galactosidase activity. As shown in Fig. 3, column A, the cfaA promoter directs a/3-galactosidase activity of 1370 U when MH1000 cells containing pcfaALacZ were grown in CFA medium without glucose. This activity was reduced to 410 units when glucose was added at a concentration of 0.5% (Fig. 3, column B). A clear difference was seen when cfaA promoter activity was examined in crp mutant (XE65.2) and in cya mutant (M-7141) strains. In both strains, the basal level of cfaA promoter activity was much lower than in the wild-type MH1000 strain, whether glucose was present or absent (Fig. 3, columns C, D, F, and G). If 10 mM cAMP was added to the M7141 strain grown in the presence of glucose (Fig. 3, column H), a marked increase in the/3-galactosidase activity was observed.

Effect of glucose and cAMP on the CFA/I operon promoter. It has previously been shown that the

In this study, we have shown that glucose inhibits production of the CFA/I fimbriae and that addition

Discussion

T.K. Karjalainen et al.: Catabolite Repression

311

of CFA/I

Beta-gal activity (x 100) t2

10-"//

S/! // // // //

8--

// // // // // // // // //

6 --

4 --

2 -

;~

// // // // // //

,'//

r-/.

A

B

C

D

E

F

of cAMP relieves glucose suppression of CFA/I fimbrial production. A target of this regulation appears to be the cfaA promoter of the CFA/I operon. The cfaA promoter activity was found to be very low in the presence of glucose and can be reverted to normal by addition of cAMP to the growth medium in a cya mutant of E. coli. These results indicate that the cfaA promoter is regulated by the mechanism of catabolite repression. The observation that the cfaA promoter was also inactive in the crp mutant, which does not produce CAP, and could not be activated by addition of cAMP further strengthens the hypothesis that the cfaA promoter is regulated by catabolite repression. The effect of catabolite repression in general involves cAMP as a mediator, although in some cases factors other than cAMP are involved [23]. Careful analysis of the cfaA promoter region revealed the presence of a typical cAMP-CAP binding site [2, 5] in the area where a " - 35" consensus sequence is normally located (Fig. 2). This putative cAMP-CAP binding site is located approximately 30 basepairs upstream from the " - 10" consensus motif; this distance has been shown to be optimal for the activation of transcription from a cAMPCAP-activated promoter [35]. The putative cAMPCAP binding site of cfaA contains two pentamer motifs that are well conserved in all cAMP-CAP-

G

v

Fig. 3. Effect of glucose and cAMP on the cfaA promoter of the CFA/I operon. The bars represent fl-galactosidase activities from E. coti MH1000 (A and B), XE65.2 (C, D, and E), and M7141 (F, G, and H) containing the reporter plasmid pcfaA-lacZ grown under indicated conditions. A: pcfaA-lacZ in MH1000 grown in the absence of glucose; B: pcfaA-lacZ in MH1000 grown in the presence of glucose; C: pcfaA-lacZ in XE65.2 grown in the absence of glucose; D: pcfaA-lacZ in XE65.2 grown in the presence of glucose; E: pcfaA-lacZ in XE65.2 grown in the presence of glucose and cAMP; F: pcfaA-lacZ in M-7141 grown in the absence of glucose; G: pcfaA-lacZ in M-7141 grown in the presence of glucose; H: pcfaA-lacZ in M-7141 grown in the presence of glucose and cAMP.

regulated promoters. Whether these two sequence motifs of the cfaA promoter are the binding sites for the CAP-cAMP complex remain to be determined. In the P fimbrial system, the binding site for CAPcAMP has been shown to be located in the main promoter for the fimbrial operon [15], whereas in most other virulence operons known to be regulated by catabolite repression, the site for CAP-cAMP binding remains to be elucidated. The cfaA promoter has also been shown to be the target of a transcriptional activator (CfaR or CfaD) that activates the expression of the CFA/I operon [29]. It will be interesting to study the interaction between the two regulators occurring at the cfaA promoter. It is not unprecedented that a prokaryotic promoter is regulated by both the CAP and another positive regulator; for example, AraC and CAP-cAMP regulate the araBAD operon synergestically [5]. It is also possible that the two regulators are responsive to different environmental stimuli, or they may constitute an intertwined regulatory network commonly seen in bacterial systems. In summary, the cfaA promoter appears to be an important target for regulation in the CFA/I operon by at least two different regulators, although the function of CfaA is unknown at present. Likewise, it is unknown whether the expression of other genes or promoters of the CFA/I operon involved in the

312

regulation of fimbria production, the processing of the fimbrial subunit, or the assembly of the fimbria is also influenced by glucose. These possibilities remain to be explored.

CURRENT MICROBIOLOGY VOI. 23 (1991)

14.

15. ACKNOWLEDGMENTS This study was supported by the U.S. Department of Agriculture grant 87-CRCR-1-2456 awarded to C.-H.L. and the U.S. Public Health Service grant DK 35369 awarded to D.G.E. We thank Dr. J. Beckwith for kindly providing us the cya and crp mutant strains.

16.

17.

Literature Cited 1. Adhya S, Garges S (1982) How cyclic AMP and its receptor protein act in Escherichia coti. Cell 29:287-289 2. Berg OG, von Hippel PH (1988) Selection of DNA binding sites by regulatory proteins. II. The binding specificity of cyclic AMP receptor protein to recognition sites. J Mol Biol 200:709-723 3. Boyer HW, Routland-Dussoix D (1969) A complementation analysis of the restriction and modification of DNA in Escherichia coli. J Mol Biol 41:459-472 4. Cravioto A, Scotland SM, Rowe B (1982) Hemagglutination activity and colonization factor antigens I and II in enterotoxigenic and non-enterotoxigenic strains ofEscherichia coli isolated from humans. Infect Immun 36:189-197 5. de Crombrugghe B, Busby S, Buc H (1984) Cyclic AMP receptor protein: role in transcription activation. Science 224:831-838 6. de Graaf FK, Roorda I (1982) Production, purification, and characterization of the fimbrial adhesive antigen F41 isolated from calf enteropathogenic Escherichia coli strain B41M. Infect Immun 36:751-758 7. Epstein W, Rothman-Denes LB, Hesse J (1975) Adenosine 3' : 5'-cyclic monophosphate as mediator ofcatabolite repression in Escherichia coli. Proc Natl Acad Sci USA 72:2300-2304 8. Evans DG, Evans Jr DJ (1978) New surface-associated heatlabile colonization factor antigen (CFA/II) produced by enterotoxigenic Escherichia coli of serogroups 06 and 08. Infect Immun 21:638-647 9. Evans DG, Silver RP, Evans Jr DJ, Chase DG, Gorbach SL (1975) Plasmid-controlled colonization factor associated with virulence in Escherichia coli enterotoxigenic for humans. Infect Immun 12:656-667 10. Evans DG, Evans Jr DJ, Tjoa W (1977) Hemagglutination of human group A erythrocytes by enterotoxigenic Escherichia coli isolated from adults with diarrhea: correlation with colonization factor. Infect Immun 18:330-337 11. Evans DG, Evans Jr DJ, Clegg S, Pauley JA (1979) Purification and characterization of the CFA/I antigen of enterotoxigenic Escherichia coli. Infect Immun 25:738-748 12. Evans DG, Evans Jr DJ, Clegg S (1980) Detection of enterotoxigenic Escherichia coli colonization factor antigen I in stool specimens by an enzyme-linked immunosorbent assay. J Clin Microbiol 12:738-743 13. Evans DG, Evans DJ Jr, Karjalainen T, Lee CH (1991) Production of Colonization Factor Antigen II of enterotoxi-

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

genic Escherichia coli is subject to catabolite repression. in press Gibert I, Villegas V, Barb6 J (1990) Expression of heat-labile enterotoxin genes is under cyclic AMP control in Escherichia coli. Curt Microbiol 20:83-90 Grransson M, Forsman K, Nilsson P, Uhlin BE (1989) Upstream activating sequences that are shared by two divergently transcribed operons mediate cAMP-CRP regulation of pilus-adhesin in Escherichia coli. Mol Microbiol 3:1557-1565 Gorbach SL, Khurana CM (1972) Toxigenic Escherichia coti: a cause of infantile diarrhea in Chicago. N Engl J Med 287:791-795 Gorbach SL, Kean BH, Evans DG, Evans Jr DJ, Bessudo D (1975) Travelers' diarrhea and toxigenic Escherichia coli. N Engl J Med 292:933-936 Hamers AM, Pel HJ, Willshaw GA, Kusters JG, van der Zeijst BAM, Gaastra W (1989) The nucleotide sequence of the first two genes of the CFA/I fimbrial operon of human enterotoxigenic Escherichia coli. Microb Pathogen 6:297-309 Honda T, Arita M, Miwatani T (1984) Characterization of new hydrophobic pili of human enterotoxigenic Escherichia coli: a possible new colonization factor. Infect Immun 43:959-965 Jones GW, Rutter JM (1972) Role of the K88 antigen in the pathogenesis of neonatal diarrhea caused by Escherichia coli in piglets. Infect Immun 6:918-927 Karjalainen TK, Evans DG, So M, Lee CH (1989) Molecular cloning and nucleotide sequence of the colonization factor antigen I gene of Escherichia coli. Infect Immun 57:1126-1130 Karjalainen TK, Evans DG, Evans Jr DJ, Lee CH (1991) A transcriptional regulator of the CFA/I s ubunit gene of enterotoxigenic E. coli is homologous to a portion of Tn3 transposase. Submitted. Laoide BM, Chambliss GH, McConnell DJ (1989) Bacillus licheniforrais a-amylase gene, amyL, is subject to promoterindependent catabolite repression in Bacillus subtilis. J Bacteriol 171:2435-2442 Martinez-Cadena MG, Guzman-Verduzco LM, Stieglitz H, Kupersztoch-Portnoy YM. (1981) Catabolite repression of Escherichia coli heat-stable enterotoxin activity. J Bacteriol 145:722-728 Miller, JH (1972) Experiments in molecular genetics. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory, pp 352-355 Nagy B, Moon HW, Isaacson RE (1976) Colonization of the porcine small intestine by Escherichia coli: ileal colonization and adhesion by pig enteropathogens that lack K88 antigen and by some acapsular mutants. Infect Immun 13:1214-1220 Salles B, and Weinstock GM (1989) Interaction of the CRPcAMP complex with the cea regulatory region. Mol Gen Genet 215:537-542 Sambrook J, Fritsch EF, Maniatis T (1989) Molecular cloning: a laboratory manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory. Savelkoul PHM, Willshaw GA, McConnell MM, Smith HR, Hamers AM, van der Zeijst BAM, Gaastra W (1990) Expression of CFA/I fimbriae is positively regulated. Microb Pathogen 8:91-99

T.K. Karjalainen et al.: Catabolite Repression of CFA/I

30. SchmoU T, Ott M, Oudega B, Hacker J (1990) Use of a wildtype gene fusion to determine the influence of environmental conditions on expression of the S fimbrial adhesion in an Escherichia coli pathogen. J Bacteriol 172:5103-5111 31. Smith HW, Linggood MA (1972) Further observations on Escherichia coli enterotoxins with particular regard to those produced by atypical piglet strains and by calf and lamb strains: the transmissible nature of these enterotoxins and ofa K antigen possessed by calf and lamb strains. J Med Microbiol 5:243-250 32. Smyth CJ (t982) Two mannose-resistant haemagglutinins on enterotoxigenic Escherichia coli of serotype 0 6 : K15 : HI6 or H - isolated from travellers' and infantile diarrhoea. J Gen Microbiol 128:2081-20% 33. Thomas LV, Cravioto A, Scotland SM, Rowe B (1982) New fimbrial antigenic type (E8775) that may represent a colonization factor in enterotoxigenic Escherichia coli in humans. Infect Immun 35:1119-1124

313 34. Thomas LV, McConnell MM, Rowe B, Field AM (1985). The possession of three novel coli surface antigens by enterotoxigenic Escherichia coti strains positive for the putative colonization factor PCF8775. J Gen Microbiol 131:2319-2326 35. Valentin-Hansen P, Holst B, SCgaard-Andersen L, Martinussen J, Nesvera J, and Douthwaite SR (1991) Design of cAMPCRP-activated promoters in Escherichia coli. Mol Microbiol 5:433-437 36. Watson JD, Hopkins NH, Roberts JW, Steitz JA, Weiner AM (1987) Molecular biology of the gene. Menlo Park, CA: The Benjamin/Cummings Publishing Co., pp 478-485 37. Weinstock GM, apRhys C, Berman ML, Hampar B, Jackson D, Silhavy TJ, Weiseman J, Zweig M (1983) Open reading frame expression vectors: a general method for antigen production in Escherichia coli using protein fusions to /3-galactosidase. Proc Natl Acad Sci USA 80:44324436