Eur. J. Epidemiol. 0392-2990 July 1992, p. 543-547
EUROPEAN JOURNAL
Vol. 8, No. 4
OF EPIDEMIOLOGY
HETEROGENEITY AMONG HEAT-LABILEENTEROTOXINS PRODUCED BY PORCINE ENTEROTOXIGENICESCHERICHIA COLI T. TSUJI 1, S. YAO and A. MIYAMA Department o f Microbiology - Fujita Health University School o f Medicine - Toyoake Aichi, 470-11 - Japan.
Key words: Heat-labile enterotoxins (LT) - A and B subunits - pI value The heat-labile enterotoxins (LT) produced by various porcine strains (LTp) of enterotoxigenic Escherichia coli were purified to homogeneity and their molecular properties were compared with
each other. By double gel diffusion and rabbit skin permeability test, LTps from WT-1 (LTp-WT-1) and BO-149 (LTp-BO-149) strains were antigenically and biologically identical. By polyacrylamide gel electrophoresis with sodium dodecyl sulfate (SDS), the mobilities of A and B subunits of LTp (BO-149) were identical to those of LTp (WT-1). However, on polyacrylamide gel electrophoresis without SDS, LTp (BO-149) and LTp (WT-1) differed in mobility, suggesting their molecular differences. Though the pI value of the B subunit of LTp (BO-149) was identical to that of LTp (WT-1), the pI value of the A subunit of LTp (BO-149) was higher than that ofLTp (WT-1). These data suggest that there is molecular heterogeneity among LTps produced by the two porcine enterotoxigenic Escherichia coli strains.
INTRODUCTION
Heterogeneity among three LTs produced by human (LTh), porcine (LTp) and chicken (LTc) enterotoxigenic Escherichia coli (ETEC) has already been determined (2, 5, 13, 15). Moreover subtypes of LTh from human ETEC strains have been established (3, 9, 16, 20, 21). LT consist of A and B subunits like cholera toxin (CT). The A subunit (MW = 28,000) increases the level of cyclic AMP in target cells by catalyzing NAD-dependent ADP ribosylation and is responsible for toxic activity. The B subunit (MW = 11,000) recognizes the membrane component, GM1 ganglioside, as a receptor and binds the toxins to the target cells (7). The complete amino acid sequences of A and B subunits of LTh and LTp have already been determined by DNA analysis and amino acid analysis (3, 9, 14, 16, 20, 21). These data suggest that there are 1 Corresponding author.
543
three subtypes of LTh (5, 16, 17) and one type of LTp (20). Moreover, we reported on the charge heterogeneity of LTs (LTh) produced by human ETEC, suggesting that there are other subtypes of LTh (17). However, no papers on other subtypes of LTp produced by porcine ETEC strain have appeared until now. In this study, we purified various LTps produced by various porcine strains and tried to determine heterogeneity among LTp purified from various porcine ETEC strains. MATERIALS AND METHODS
Bacteria
Porcine strain ETEC WT-1 was prepared as described previously (11). Porcine BO-149 strain was isolated from a diarrheal porcine in Brazil and obtained as a gift from Dr Pestana.
Tsuji T. et al.
Eur. J. Epidemiol.
Purification of LT LTs were purified as described previously (1). Detailed procedures of the cell culturing, isolation of crude cell extract and purification of LT by successive column chromatography on Bio-Gel A 5m were described previously (16). Polvacrylamide gel electrophoresis Polyacrylamide gel electrophoresis with or without SDS was performed as described previously (4, 8) in 15% or 7% polyacrylamide gels, respectively. Gels were stained with Coomassie brilliant blue and then destained as described previously (8). Double gel diffusion test The double gel diffusion test was carried out as described previously (10) in 1.25% Agarose L (Difco) in TEAN buffer (50 mM Tris-HCI, 1 mM EDTA, 3 mM sodium azide, 0.2 M NaCI, pH 7.2). Isoelectric focusing Polyacrylamide gel isoelectric focusing was carried out as described previously (19) in 50/0 acrylamide gel. Horse cytochrome c and its derivates were used as standard markers. Gels were stained with 0.001% Coomassie brilliant blue G-250 in 3.5% perchloric acid for 1.5 hrs and 0.4% Coomassie brilliant blue R in 30% methanol containing 10% acetic acid overnight at 37"C with shaking. Destaining was performed as described previously (8).
Figure 1. - Double gel diffusion test of LTp (WT-1) and LTp
(B0-149). Double gel diffusion test was performed as described in Materials and Methods. 1, anti-LTp (WT-1) serum; 2, LTp (BO-149); 3, LTp 0NT-1).
RESULTS
Double gel diffusion test We compared the antigenicity of LTp by the double gel diffusion test. As shown in Figure 1, the precipitin line between LTp (WT-1), LTp (BO-149) and anti-LTp (WT-1) serum fused together. These data suggest that the two LTps are immunologically identical.
50
-
25
-
O
Biological comparison of the two LTps by PF test We compared the rabbit skin permeability activities of nicked and unnicked LTps by a PF test. As shown in Figure 2, the dose response curve of LT (BO-149) was the same as that of LTp (WT-1) with or without nicking of the A subunit. These data suggest that the two LTps are biologically indistinguishable from each other.
Figure 2. - Rabbit skin vascular permeabil#y activities of purified LTp (WT-1) and LTp (B0-149)
Polyacrylamide gel electrophoresis To examine for possible molecular differences, the preparations were subjected to polyacrylamide disk electrophoresis without SDS. As shown in Figure 3, the mobility of LTp (BO-149) was lower than that of LTp (WT-1), indicating the molecular difference between LTp (BO-149) and LTp (WT-1).
The PF test was carried out as described previously (16). The samples with or without digestion by trypsin were diluted with borate-buffer. The curves on the left are bluing dose titration of LTps after digestion with trypsin. The curves on the right are bluing dose titration of LTps without trypsin digestion. Bluing scores were determined as the diameter (mm) of the blue spot x bluing rate of the spot (16). Values are means of four determinations. • LTp (WT-1); © LTp (BO-149).
v
!
-1
I
0
+1
I
+2
Concentration of toxin( log ng/ml)
544
LT from porcine enterotoxigenic E. coil
Vol. 8, 1992
LTp ( B 0 - - 1 4 9 )
LTp(wT--1)
1
2
3
Figure 3. - Po~acrylamide gel disk electrophoresis of LTp (WT-D
and LTp (B0-149) Polyacrylamide gel electrophoresis without SDS was performed as described in Materials and Methods. 1, LTp (BO-149); 2, LTp (WT-1); 3, Mixture ofLTp (WT-1) and LTp (BO-149).
Figure 4. - SDS-polyacrylamide gel electrophoresis of LTp
(WT-1) and LTp (B0-149)
SDS-polvacrylamide gel electrophoresis
SDS polyacrylamide gel etectrophoresis was performed as described in Materials and Methods. 1, LTp (BO-149); 2, LTp (WT-1); 3, Mixture ofLTp (WT-1) and LTp (BO-149).
We examined the difference in molecular weight between the A and B subunits with SDSpolyacrylamide gel electrophoresis. As shown in Figure 4, the mobilities of the A and B subunits of the two LTps were identical to each other and the mixture of the two LTps showed only two bands corresponding to the A and B subunits. These data suggest that the molecular weights of the A and B subunits are identical to each other.
Isoelectric focusing We compared the pI values of LTps from BO-149 and WT-1 strains. As shown in Figure 5, the pI vaues of holo LTp and coligenoid (B subunit oligomer) of the two LTps are indistinguishable from each other. These data suggest that the pI values of holo LTp and coligenoid are the same for the two.
Comparison of the pl values of A and B subunits of the two LTs We compared the pI values of the A and B subunits of the two LTs, which carried the capacity to revert to holo LT toxin and could be abolished by their denaturation, by which all of the charge amino acids of the proteins came out onto the surface of the molecule (18). As shown in Figure 6, the mixture of B subunits of these LTs with or without denaturation showed the same bands, suggesting that the pI values of these LTp-Bs are identical to each other with or without denaturation. However, though the A subunits of these LTps showed to main bands due to their microheterogeneity (12), the mixture of A subunits of the two LTps without denaturation showed more than two bands. Moreover, the mixture
Figure 5. - Comparison of isoelectricJocusing of LTp (WT-D and LTp (B0-149) Isoelectric focusing was carried out as described in Materials and Methods. The gels contained 1% ampholine from 3.0 to 10.0. 1, LTp (BO-149); 2, LTp (WT-1); 3, Mixture ofLTp (WT-1) and LTp (BO-149); 4, markers.
545
Tsuji T. et al.
Eur. J. Epidemiol.
Figure 6. - Comparison of isoelectricfocusing of A and B subunits of LTp (WT-1) and LTp (B0-149) Isoelectric focusing was performed as described in Materials and Methods. The samples were dialyzed against 0.1 M propionic acid containing 6M urea at 0° C for 3 days (treatment 1) and were treated with 2°/0 SDS at 37° C for 30 min for denaturation (treatment 2). The samples were then applied to the gels, which contained 1% ampholine for pH 3.0 to pH 10.0, 6 M urea and 0.01% nonidet p-40. The electrophoresis was performed at a constant voltage of 250V for 24 hrs with cooling at 4° C. 1, LTp (BO-149) prepared by treatment 1;, 2, LTp (WT-1) prepared by treatment 1; 3, mixture of LTp (WT-1) and LTp (BO-149) prepared by treatment 1; 4, mixture of LTp (WT-1)s prepared by treatments 1 and 2; 5, mixture of LTp (BO-149)s prepared by treatments 1 and 2; 6, markers.
of the A subunit from LTp (WT-1) with and without denaturation showed two bands and that of LTp (BO149) was the same. These data suggest that the pI value for the A subunits of these LTps are not identical to each other and that all of the ionic amino acids of the A subunits of the two LTs exist on the surface of the molecule. The difference in the pI values among the A subunits did not affect the pI value ofholo LT, suggesting that some charged amino acids, which exist on the surface of the A subunit and affect the pI value of the A subunit, exist inside the holo LT molecule. DISCUSSION
and established by amino acid sequence analysis (3, 9, 16, 20, 21). Moreover, the analysis of the amino acid sequences shows that there are three subtypes in LTh, that is, LTh-1 (H 10407) (20), LTh-2 (H74-114) (9) and LTh-3 (240-3) (16) and one type ofLTp, LTp-1 (EWD 299) (21). However, there have been no reports about LTp subtypes until now. Therefore, we compared various LTps purified from different porcine ETEC strains. As shown in the above data, among the LTps obtained from porcine strains, LTps from porcine WT-1 and BO-149 ETEC strains were immunologically, biologically and physicochemically compared to each other. As shown in Figures 1 and 2, LTp (BO-149) is immunogically and biologically indistinguishable from that of LTp (WT-1). The molecular weights of the A and B subunits of LTp (BO-149) are also identical to those of LTp (WT-1). However, the mobilities of the LTps were different from each other by polyacrylamide gel electrophoresis without SDS (Fig. 3). We then compared the pI values of the holo LTps and their A and B subunits. Though the pI values of the holo LTps of BO-149 and WT-1 strains and the subunit of the two LTps were identical to each other, the pI values of the A subunits were different from each other. However, after denaturation, the pI values of the A subunits of the two LTps did not change, indicating that all of the charged amino acids of the A subunits of the two LTps exist on the surface of the molecule. These data suggest that LTp (BO-149) is different from LTp (WT1) and that the charge difference observed in the A subunits may be due to charged amino acid exchange occurring in the A subunit of LTp. However, the charge difference of the A subunit did not affect the pI values of the holo LT, suggesting that the amino acids, which affect the pI value difference of the A subunit, exist inside of the holo LT. The protein mobility on polyacrylamide gel electrophoresis without SDS was mainly affected by its molecular weight and ionic charge. However, in LTp (WT-1) and LTp (BO-149), the molecular weights of the A and B subunits and the pI values of the two holo LTps were indistinguishable from the each other. These data suggest that although all of the charged amino acids of the A subunits exist on the surface of the molecule, ionic amino acid differences of the A subunit do not affect the pI values of holo LT, but induce a very small three dimensional difference of holo LTp and results in different mobilities between LTp (WT-1) and LTp (BO-149). Moreover, we'd like to point out that the methods employed in this study, i.e. polyacrylamide electrophoresis, isoelectric focusing and those with and without SDS are very efficient ways to detect molecular differences among LTps produced by ETEC. Acknowledgements
Heterogeneity in the molecular structures of LTh, LTp, LTc and LT-II produced by human, porcine, chicken and cow have been reported (2, 5, 6, 13, 15) 546
This work was supported by Grants-in Aid for Scientific Research from the Minsitry of Education, Science and Culture of Japan.
Vol. 8, 1992
LT from porcine enterotoxigenic E. coli
REFERENCES
1.
2.
12. Takeda Y., Honda T., Taga S. and Miwatani T. (1981):
Clements J.D. and Finkelstein R.A. (1979): Isolation and characterization of homogeneous heat-labile enterotoxins with high specific activity from Escherichia coli cultures - Infect. Irnmun. 24: 760-769. Clements J.D., Flint D.C. and Klipstein F.A. (1982): Immunological and physicochemical characterization of heat-labile enterotoxins isolated from two strains of Escherichia coli - Infect. lmmun. 38: 806-809.
3.
Dallas W.S. and Falkow S. (1980): Amino acid sequence homology between cholera toxin and Escherichia coli heat-labile toxin - Nature 288: 499501.
4.
Davis B.T. (1964): Disk electrophoresis method and application to human serum proteins - Ann. N. Y. Acad. Sci. 121: 404-427.
5.
Geary S.J., Marehlewicz B.A. and Finkelstein R.A. (1982): Comparison of heat-labile enterotoxins from porcine and human strains ofEseheriehia coli -Infect. Immun. 36: 215-220.
6.
Holmes R.K., Twiddy E.M. and Pickett C.L. (1986): Purification and characterization of type II heatlabile enterotoxin ofEseheriehia coli - Infect. Immun. 53: 464-473.
7.
Holmgren or. (1981): Actions of cholera toxin and the prevention and treatment of cholera - Nature 292: 413-417.
8.
Laemmli U.K. (1970): Cleavage of structural proteins during the assembly of the head of Bacteriophage T4 - Nature 227: 680-685.
9.
Leong J., Vinal A.C. and Dallas W.S. (1985): Nucleotide sequence comparison between heatlabile toxin B subunit cistrons from Escheriehia coli of human and porcine origin - Infect. Immun. 48: 7377.
10. Ouehterlony O. (1949): Antigen-Antibody reactions in gels - Acta Path. Microbiol. Scand. 26: 507-515. 11. Soedarmono P., Takeda T., Matsushiro A., Yamamoto K., Takeda Y., Miwatani T. and Yoshikawa M. (1980): Genetic labeling of an Ent plasmid that encodes heat-labile enterotoxin of enterotoxigenic Escherichia eoli with kanamycin resistance transposon - FEMS Microbiol. Lett. 9: 307-310.
547
In vitro formation of hybrid toxins between subunits of Escheriehia coli heat-labile enterotoxin and those of cholera enterotoxin - Infect. Immun. 34: 341-346. 13. Tsuji T, Taga S., Honda T., Takeda Y. and Miwatani T. (1982): Molecular heterogeneity of heat-labile enterotoxins from human and porcine enterotoxigenic Escherichia co# - Infect. Immun. 38: 444-448. 14. Tsuji T., Honda T., Miwatani T., Wakabayashi S. and Matsubara It. (1984): The amino acid sequence of the B subunit: of porcine enterotoxigenic Escheriehia coli enteroxin-Analysis and comparison with literature data - FEMS Microbiol Lett. 25: 243-246. 15. Tsuji T., Joya J.E., Yao S., Honda T. and Miwatani T. (1988): Purification and characterization of heatlabile enterotoxin isolated from chicken enterotoxigenic Escheriehia eoli - FEMS Microbiol. Lett. 52: 79-84. 16.
17.
Tsuji T., Iida T., Honda T., Miwatani T., Nagahama M., Sakurai or., Wada K. and Matsubara 1t. (1987): A unique amino acid sequence of the B subunit of a heat-labile enterotoxin isolated from a human enterotoxigenic Eseherichia eoli - Microbial. Pathog. 2: 381-390. Tsuji T., Yao S., Joya J.E., Honda T. and Miwatani T. (1989): Charge heterogeneity of heat-labile enterotoxins from human enterotoxigenic Escherichia coli - FEMS Microbiol. Lett. 61: 1-6.
18. Tsuji T., Honda T., Miwatani T. and Miyama A. (1990): Comparison of coligenoid formation by B subunits of porcine and human Eseheriehia coli heat-labile enterotoxins - FEMS. Microbiol. Lett. 69: 299-304. 19. Wrigley C.W. (1971): Gel electrofocusing - Method. Enzym. 22: 559-564. 20. Yamamoto T. and Yokota T. (1983): Sequence of heatlabile enterotoxin of Escherichia coli pathogenic for humans - J. Bacteriol. 155: 728-733. 21. Yamamoto T., Tamura T. and Yokota T. (1984): Primary structure of heat-labile enterotoxin produced by Eseherichia eoli pathogenic for humans J. Biol. Chem. 259: 5037-5044.
EUROPEAN JOURNAL
Vol. 8, No. 4
OF EPIDEMIOLOGY
HETEROGENEITY AMONG HEAT-LABILEENTEROTOXINS PRODUCED BY PORCINE ENTEROTOXIGENICESCHERICHIA COLI T. TSUJI 1, S. YAO and A. MIYAMA Department o f Microbiology - Fujita Health University School o f Medicine - Toyoake Aichi, 470-11 - Japan.
Key words: Heat-labile enterotoxins (LT) - A and B subunits - pI value The heat-labile enterotoxins (LT) produced by various porcine strains (LTp) of enterotoxigenic Escherichia coli were purified to homogeneity and their molecular properties were compared with
each other. By double gel diffusion and rabbit skin permeability test, LTps from WT-1 (LTp-WT-1) and BO-149 (LTp-BO-149) strains were antigenically and biologically identical. By polyacrylamide gel electrophoresis with sodium dodecyl sulfate (SDS), the mobilities of A and B subunits of LTp (BO-149) were identical to those of LTp (WT-1). However, on polyacrylamide gel electrophoresis without SDS, LTp (BO-149) and LTp (WT-1) differed in mobility, suggesting their molecular differences. Though the pI value of the B subunit of LTp (BO-149) was identical to that of LTp (WT-1), the pI value of the A subunit of LTp (BO-149) was higher than that ofLTp (WT-1). These data suggest that there is molecular heterogeneity among LTps produced by the two porcine enterotoxigenic Escherichia coli strains.
INTRODUCTION
Heterogeneity among three LTs produced by human (LTh), porcine (LTp) and chicken (LTc) enterotoxigenic Escherichia coli (ETEC) has already been determined (2, 5, 13, 15). Moreover subtypes of LTh from human ETEC strains have been established (3, 9, 16, 20, 21). LT consist of A and B subunits like cholera toxin (CT). The A subunit (MW = 28,000) increases the level of cyclic AMP in target cells by catalyzing NAD-dependent ADP ribosylation and is responsible for toxic activity. The B subunit (MW = 11,000) recognizes the membrane component, GM1 ganglioside, as a receptor and binds the toxins to the target cells (7). The complete amino acid sequences of A and B subunits of LTh and LTp have already been determined by DNA analysis and amino acid analysis (3, 9, 14, 16, 20, 21). These data suggest that there are 1 Corresponding author.
543
three subtypes of LTh (5, 16, 17) and one type of LTp (20). Moreover, we reported on the charge heterogeneity of LTs (LTh) produced by human ETEC, suggesting that there are other subtypes of LTh (17). However, no papers on other subtypes of LTp produced by porcine ETEC strain have appeared until now. In this study, we purified various LTps produced by various porcine strains and tried to determine heterogeneity among LTp purified from various porcine ETEC strains. MATERIALS AND METHODS
Bacteria
Porcine strain ETEC WT-1 was prepared as described previously (11). Porcine BO-149 strain was isolated from a diarrheal porcine in Brazil and obtained as a gift from Dr Pestana.
Tsuji T. et al.
Eur. J. Epidemiol.
Purification of LT LTs were purified as described previously (1). Detailed procedures of the cell culturing, isolation of crude cell extract and purification of LT by successive column chromatography on Bio-Gel A 5m were described previously (16). Polvacrylamide gel electrophoresis Polyacrylamide gel electrophoresis with or without SDS was performed as described previously (4, 8) in 15% or 7% polyacrylamide gels, respectively. Gels were stained with Coomassie brilliant blue and then destained as described previously (8). Double gel diffusion test The double gel diffusion test was carried out as described previously (10) in 1.25% Agarose L (Difco) in TEAN buffer (50 mM Tris-HCI, 1 mM EDTA, 3 mM sodium azide, 0.2 M NaCI, pH 7.2). Isoelectric focusing Polyacrylamide gel isoelectric focusing was carried out as described previously (19) in 50/0 acrylamide gel. Horse cytochrome c and its derivates were used as standard markers. Gels were stained with 0.001% Coomassie brilliant blue G-250 in 3.5% perchloric acid for 1.5 hrs and 0.4% Coomassie brilliant blue R in 30% methanol containing 10% acetic acid overnight at 37"C with shaking. Destaining was performed as described previously (8).
Figure 1. - Double gel diffusion test of LTp (WT-1) and LTp
(B0-149). Double gel diffusion test was performed as described in Materials and Methods. 1, anti-LTp (WT-1) serum; 2, LTp (BO-149); 3, LTp 0NT-1).
RESULTS
Double gel diffusion test We compared the antigenicity of LTp by the double gel diffusion test. As shown in Figure 1, the precipitin line between LTp (WT-1), LTp (BO-149) and anti-LTp (WT-1) serum fused together. These data suggest that the two LTps are immunologically identical.
50
-
25
-
O
Biological comparison of the two LTps by PF test We compared the rabbit skin permeability activities of nicked and unnicked LTps by a PF test. As shown in Figure 2, the dose response curve of LT (BO-149) was the same as that of LTp (WT-1) with or without nicking of the A subunit. These data suggest that the two LTps are biologically indistinguishable from each other.
Figure 2. - Rabbit skin vascular permeabil#y activities of purified LTp (WT-1) and LTp (B0-149)
Polyacrylamide gel electrophoresis To examine for possible molecular differences, the preparations were subjected to polyacrylamide disk electrophoresis without SDS. As shown in Figure 3, the mobility of LTp (BO-149) was lower than that of LTp (WT-1), indicating the molecular difference between LTp (BO-149) and LTp (WT-1).
The PF test was carried out as described previously (16). The samples with or without digestion by trypsin were diluted with borate-buffer. The curves on the left are bluing dose titration of LTps after digestion with trypsin. The curves on the right are bluing dose titration of LTps without trypsin digestion. Bluing scores were determined as the diameter (mm) of the blue spot x bluing rate of the spot (16). Values are means of four determinations. • LTp (WT-1); © LTp (BO-149).
v
!
-1
I
0
+1
I
+2
Concentration of toxin( log ng/ml)
544
LT from porcine enterotoxigenic E. coil
Vol. 8, 1992
LTp ( B 0 - - 1 4 9 )
LTp(wT--1)
1
2
3
Figure 3. - Po~acrylamide gel disk electrophoresis of LTp (WT-D
and LTp (B0-149) Polyacrylamide gel electrophoresis without SDS was performed as described in Materials and Methods. 1, LTp (BO-149); 2, LTp (WT-1); 3, Mixture ofLTp (WT-1) and LTp (BO-149).
Figure 4. - SDS-polyacrylamide gel electrophoresis of LTp
(WT-1) and LTp (B0-149)
SDS-polvacrylamide gel electrophoresis
SDS polyacrylamide gel etectrophoresis was performed as described in Materials and Methods. 1, LTp (BO-149); 2, LTp (WT-1); 3, Mixture ofLTp (WT-1) and LTp (BO-149).
We examined the difference in molecular weight between the A and B subunits with SDSpolyacrylamide gel electrophoresis. As shown in Figure 4, the mobilities of the A and B subunits of the two LTps were identical to each other and the mixture of the two LTps showed only two bands corresponding to the A and B subunits. These data suggest that the molecular weights of the A and B subunits are identical to each other.
Isoelectric focusing We compared the pI values of LTps from BO-149 and WT-1 strains. As shown in Figure 5, the pI vaues of holo LTp and coligenoid (B subunit oligomer) of the two LTps are indistinguishable from each other. These data suggest that the pI values of holo LTp and coligenoid are the same for the two.
Comparison of the pl values of A and B subunits of the two LTs We compared the pI values of the A and B subunits of the two LTs, which carried the capacity to revert to holo LT toxin and could be abolished by their denaturation, by which all of the charge amino acids of the proteins came out onto the surface of the molecule (18). As shown in Figure 6, the mixture of B subunits of these LTs with or without denaturation showed the same bands, suggesting that the pI values of these LTp-Bs are identical to each other with or without denaturation. However, though the A subunits of these LTps showed to main bands due to their microheterogeneity (12), the mixture of A subunits of the two LTps without denaturation showed more than two bands. Moreover, the mixture
Figure 5. - Comparison of isoelectricJocusing of LTp (WT-D and LTp (B0-149) Isoelectric focusing was carried out as described in Materials and Methods. The gels contained 1% ampholine from 3.0 to 10.0. 1, LTp (BO-149); 2, LTp (WT-1); 3, Mixture ofLTp (WT-1) and LTp (BO-149); 4, markers.
545
Tsuji T. et al.
Eur. J. Epidemiol.
Figure 6. - Comparison of isoelectricfocusing of A and B subunits of LTp (WT-1) and LTp (B0-149) Isoelectric focusing was performed as described in Materials and Methods. The samples were dialyzed against 0.1 M propionic acid containing 6M urea at 0° C for 3 days (treatment 1) and were treated with 2°/0 SDS at 37° C for 30 min for denaturation (treatment 2). The samples were then applied to the gels, which contained 1% ampholine for pH 3.0 to pH 10.0, 6 M urea and 0.01% nonidet p-40. The electrophoresis was performed at a constant voltage of 250V for 24 hrs with cooling at 4° C. 1, LTp (BO-149) prepared by treatment 1;, 2, LTp (WT-1) prepared by treatment 1; 3, mixture of LTp (WT-1) and LTp (BO-149) prepared by treatment 1; 4, mixture of LTp (WT-1)s prepared by treatments 1 and 2; 5, mixture of LTp (BO-149)s prepared by treatments 1 and 2; 6, markers.
of the A subunit from LTp (WT-1) with and without denaturation showed two bands and that of LTp (BO149) was the same. These data suggest that the pI value for the A subunits of these LTps are not identical to each other and that all of the ionic amino acids of the A subunits of the two LTs exist on the surface of the molecule. The difference in the pI values among the A subunits did not affect the pI value ofholo LT, suggesting that some charged amino acids, which exist on the surface of the A subunit and affect the pI value of the A subunit, exist inside the holo LT molecule. DISCUSSION
and established by amino acid sequence analysis (3, 9, 16, 20, 21). Moreover, the analysis of the amino acid sequences shows that there are three subtypes in LTh, that is, LTh-1 (H 10407) (20), LTh-2 (H74-114) (9) and LTh-3 (240-3) (16) and one type ofLTp, LTp-1 (EWD 299) (21). However, there have been no reports about LTp subtypes until now. Therefore, we compared various LTps purified from different porcine ETEC strains. As shown in the above data, among the LTps obtained from porcine strains, LTps from porcine WT-1 and BO-149 ETEC strains were immunologically, biologically and physicochemically compared to each other. As shown in Figures 1 and 2, LTp (BO-149) is immunogically and biologically indistinguishable from that of LTp (WT-1). The molecular weights of the A and B subunits of LTp (BO-149) are also identical to those of LTp (WT-1). However, the mobilities of the LTps were different from each other by polyacrylamide gel electrophoresis without SDS (Fig. 3). We then compared the pI values of the holo LTps and their A and B subunits. Though the pI values of the holo LTps of BO-149 and WT-1 strains and the subunit of the two LTps were identical to each other, the pI values of the A subunits were different from each other. However, after denaturation, the pI values of the A subunits of the two LTps did not change, indicating that all of the charged amino acids of the A subunits of the two LTps exist on the surface of the molecule. These data suggest that LTp (BO-149) is different from LTp (WT1) and that the charge difference observed in the A subunits may be due to charged amino acid exchange occurring in the A subunit of LTp. However, the charge difference of the A subunit did not affect the pI values of the holo LT, suggesting that the amino acids, which affect the pI value difference of the A subunit, exist inside of the holo LT. The protein mobility on polyacrylamide gel electrophoresis without SDS was mainly affected by its molecular weight and ionic charge. However, in LTp (WT-1) and LTp (BO-149), the molecular weights of the A and B subunits and the pI values of the two holo LTps were indistinguishable from the each other. These data suggest that although all of the charged amino acids of the A subunits exist on the surface of the molecule, ionic amino acid differences of the A subunit do not affect the pI values of holo LT, but induce a very small three dimensional difference of holo LTp and results in different mobilities between LTp (WT-1) and LTp (BO-149). Moreover, we'd like to point out that the methods employed in this study, i.e. polyacrylamide electrophoresis, isoelectric focusing and those with and without SDS are very efficient ways to detect molecular differences among LTps produced by ETEC. Acknowledgements
Heterogeneity in the molecular structures of LTh, LTp, LTc and LT-II produced by human, porcine, chicken and cow have been reported (2, 5, 6, 13, 15) 546
This work was supported by Grants-in Aid for Scientific Research from the Minsitry of Education, Science and Culture of Japan.
Vol. 8, 1992
LT from porcine enterotoxigenic E. coli
REFERENCES
1.
2.
12. Takeda Y., Honda T., Taga S. and Miwatani T. (1981):
Clements J.D. and Finkelstein R.A. (1979): Isolation and characterization of homogeneous heat-labile enterotoxins with high specific activity from Escherichia coli cultures - Infect. Irnmun. 24: 760-769. Clements J.D., Flint D.C. and Klipstein F.A. (1982): Immunological and physicochemical characterization of heat-labile enterotoxins isolated from two strains of Escherichia coli - Infect. lmmun. 38: 806-809.
3.
Dallas W.S. and Falkow S. (1980): Amino acid sequence homology between cholera toxin and Escherichia coli heat-labile toxin - Nature 288: 499501.
4.
Davis B.T. (1964): Disk electrophoresis method and application to human serum proteins - Ann. N. Y. Acad. Sci. 121: 404-427.
5.
Geary S.J., Marehlewicz B.A. and Finkelstein R.A. (1982): Comparison of heat-labile enterotoxins from porcine and human strains ofEseheriehia coli -Infect. Immun. 36: 215-220.
6.
Holmes R.K., Twiddy E.M. and Pickett C.L. (1986): Purification and characterization of type II heatlabile enterotoxin ofEseheriehia coli - Infect. Immun. 53: 464-473.
7.
Holmgren or. (1981): Actions of cholera toxin and the prevention and treatment of cholera - Nature 292: 413-417.
8.
Laemmli U.K. (1970): Cleavage of structural proteins during the assembly of the head of Bacteriophage T4 - Nature 227: 680-685.
9.
Leong J., Vinal A.C. and Dallas W.S. (1985): Nucleotide sequence comparison between heatlabile toxin B subunit cistrons from Escheriehia coli of human and porcine origin - Infect. Immun. 48: 7377.
10. Ouehterlony O. (1949): Antigen-Antibody reactions in gels - Acta Path. Microbiol. Scand. 26: 507-515. 11. Soedarmono P., Takeda T., Matsushiro A., Yamamoto K., Takeda Y., Miwatani T. and Yoshikawa M. (1980): Genetic labeling of an Ent plasmid that encodes heat-labile enterotoxin of enterotoxigenic Escherichia eoli with kanamycin resistance transposon - FEMS Microbiol. Lett. 9: 307-310.
547
In vitro formation of hybrid toxins between subunits of Escheriehia coli heat-labile enterotoxin and those of cholera enterotoxin - Infect. Immun. 34: 341-346. 13. Tsuji T, Taga S., Honda T., Takeda Y. and Miwatani T. (1982): Molecular heterogeneity of heat-labile enterotoxins from human and porcine enterotoxigenic Escherichia co# - Infect. Immun. 38: 444-448. 14. Tsuji T., Honda T., Miwatani T., Wakabayashi S. and Matsubara It. (1984): The amino acid sequence of the B subunit: of porcine enterotoxigenic Escheriehia coli enteroxin-Analysis and comparison with literature data - FEMS Microbiol Lett. 25: 243-246. 15. Tsuji T., Joya J.E., Yao S., Honda T. and Miwatani T. (1988): Purification and characterization of heatlabile enterotoxin isolated from chicken enterotoxigenic Escheriehia eoli - FEMS Microbiol. Lett. 52: 79-84. 16.
17.
Tsuji T., Iida T., Honda T., Miwatani T., Nagahama M., Sakurai or., Wada K. and Matsubara 1t. (1987): A unique amino acid sequence of the B subunit of a heat-labile enterotoxin isolated from a human enterotoxigenic Eseherichia eoli - Microbial. Pathog. 2: 381-390. Tsuji T., Yao S., Joya J.E., Honda T. and Miwatani T. (1989): Charge heterogeneity of heat-labile enterotoxins from human enterotoxigenic Escherichia coli - FEMS Microbiol. Lett. 61: 1-6.
18. Tsuji T., Honda T., Miwatani T. and Miyama A. (1990): Comparison of coligenoid formation by B subunits of porcine and human Eseheriehia coli heat-labile enterotoxins - FEMS. Microbiol. Lett. 69: 299-304. 19. Wrigley C.W. (1971): Gel electrofocusing - Method. Enzym. 22: 559-564. 20. Yamamoto T. and Yokota T. (1983): Sequence of heatlabile enterotoxin of Escherichia coli pathogenic for humans - J. Bacteriol. 155: 728-733. 21. Yamamoto T., Tamura T. and Yokota T. (1984): Primary structure of heat-labile enterotoxin produced by Eseherichia eoli pathogenic for humans J. Biol. Chem. 259: 5037-5044.
