THE JOURNAL OF INFECTIOUS DISEASES • VOL. 133, SUPPLEMENT © 1976 by the University of Chicago. All rights reserved.
•
MARCH 1976
Interactions of Choleragenoid and G M 1 Ganglioside with Enterotoxins of Vibrio cholerae and Escherichia coli in Cultured Adrenal Cells Sam T. Donta
From the Division of Infectious Diseases, Department of Medicine, University of Iowa, and Veterans Administration Hospitals, Iowa City, Iowa
The heat-labile enterotoxins of Vibrio cholerae and Escherichia coli induce morphologic changes and steroidogenesis in clonal lines of adrenal tumor cells in monolayer tissue culture [I, 2]. The results of several studies of the effects of the toxins on these adrenal cells suggested that the two enterotoxins may act by mechanisms that are similar to each other but dissimilar from the mechanism of the natural hormonal inducer, adrenocorticotropic hormone [3-5]. It has been proposed that the ganglioside GM! (galactosyl-N-acetylgalactosaminyl [sialosyl] lactosyl ceramide; GGnSLC) is the natural receptor in the intestine and perhaps in nonintestinal tissue for the enterotoxin of V. cholerae but probably is not the receptor for the heat-labile enteroThis study was supported by grants from the Veterans Administration Research and Education Service and by grants no. AI 11416 and AI 11794 from the National Institute of Allergy and Infectious Diseases. Dr. Donta is a clinical investigator of the Veterans Administration. I thank Ms. Shelley Kreiter for her excellent technical assistance. Please address requests for reprints to Dr. Sam T. Donta, Department of Medicine, Veterans Administration Hospital, Iowa City, Iowa 52240.
Sl1S
toxin of E. coli. This hypothesis is based primarily on the relative degrees of inactivation of the two toxins by G M ! ganglioside [6-9]. Similar results were not obtained, however, when the effects of the ganglioside on the activities of the toxins in adrenal cells were studied [10]. These studies demonstrated not only that the G M 1 ganglioside was capable of equivalent degrees of inhibition of the morphologic and steroidogenic activities of equipotent concentrations of the two toxins, but also that, in competition experiments, either toxin could interfere with the ability of the ganglioside to inactivate the other toxin. Since crude preparations of E. coli toxin were used in all of these studies, it may well be that impurities were present in concentrations capable of interfering with the inactivation of the E. coli toxin by the ganglioside in the nonadrenal systems; however, since the adrenal cells are more sensitive to the effects of the toxin than are the other systems studied, much smaller concentrations of the crude toxin preparation, and perhaps therefore of the impurities, were present and did not interfere with the binding of the E. coli toxin to the ganglioside. The results of these previous studies with adrenal cells suggested that, if the G m ganglioside
Downloaded from http://jid.oxfordjournals.org/ at University of New South Wales on August 21, 2015
The heat-labile enterotoxins of Vibrio cholerae and Escherichia coli induce morphologic changes and steroidogenesis in clonal lines of adrenal tumor cells in tissue culture; these effects are preventable by prior incubation of either toxin with GM ! ganglioside (galactosyl-N-acetylgalactosaminyl [sialosyl] lactosyl ceramide; GGnSLC) but are not preventable by prior incubation of adrenal cells with choleragenoid. Choleragenoid, however, is capable of interfering with the ability of G M ! ganglioside to neutralize the effects of either toxin. These results suggest that the G m ganglioside may not be the true receptor for the toxins on adrenal cells, but that it is acting as a pseudoreceptor. The ability of a subunit (A) of the cholera enterotoxin molecule to induce in adrenal cells morphologic changes and steroidogenesis similar to those effects inducible by the whole toxin indicates the possibility that separate receptor sites on these adrenal cells may exist for the binding and active fragments of the molecule.
Donta
8116
Materials and Methods
Toxins and toxoid. Purified preparations of cholera enterotoxin and choleragenoid were obtained from Dr. R. A. Finkelstein [I 1]. Dialyzed, lyophilized culture filtrates of E. coli 078: H 12 (strain 408-3) were obtained from Dr. R. B. Sack [12]. Ganglioside (GMt) was obtained from Dr. W. E. van Heyningen [9]. The toxins and gangliosides were diluted in a 0.05 M Tris-O.OOlM EDTA buffer system (pH 7.5). Tissue culture methods. Monolayer cultures of Yl adrenal cells were propagated and maintained in minimal essential medium supplemented with 10% fetal calf serum at 37 C in a humidified atmosphere of 95% air and 5% CO 2 • More specific details of the tissue culture experimental methods and of the spectrophotometric assay for ~4,3-ketosteroids extractable from the tissue culture medium have been described [5]. The results are expressed as nanomoles of steroid per plate of cells per incubation period. Some of the variation in the amount of steroid produced between experiments is secondary to either the total incubation time or the density of cells per plate (variation between the number of cells per plate in any given experiment never exceeded 10%). In addition, some subclonal lines of Yl cells have lower basal rates of steroidogenesis than do others. For experiments in which the interactions among GMt ganglioside, choleragenoid, and enterotoxins were studied, Tris buffer or choleragenoid was incubated with ganglioside at 37 C for 20 min; toxin was added to the mixture and the incubation continued for another 20 min. An aliquot was then removed from the mixture and was added to the tissue culture medium surround-
ing Yl cells. Controls with and without ganglioside, choleragenoid, and toxin were similarly incubated with and without buffer. Results
When two different lots of choleragenoid were tested on Yl cells, morphological and steroidogenic changes induced by each of the preparations could be observed (table I); these changes represented 0.001%-0.1% of the activity of the native toxin and presumably were due to small amounts of residual whole toxin in the purified preparations of choleragenoid. This presumption was supported by the observation of similar heat labilities and dose responsiveness of the activities present in the choleragenoid and native toxin preparations. When amounts of choleragenoid 50- to 500-fold greater than the submaximal steroidogenic concentrations of cholera toxin were used and the choleragenoid was incubated with adrenal cells prior to addition of toxin, no inhibition of the effects of the toxin could be demonstrated (table 2), regardless of whether cells were incubated with choleragenoid for as little as 0 hr or as long as 8 hr before the addition of toxin. The amounts (50-100 ng) of choleragenoid used in these experiments were sufficient to provide 10 6 molecules per cell. Similarly, no inhibition of the Table 1. Steroidogenic effects of choleragenoid (CO) on Yl adrenal cells in culture. Test system
nmol/plate*
Control CT,t 10 ng/rnl
3.2 ± 0.1 17.2 ± 0.7
CG, lot no. GOl72 1 ug/rnl 100 ng/ml 50 ng/rnl 10 ng/ml 5 ng/rnl 1 ng/ml
16.7 13.5 10.0 4.0 5.7 4.3
± ± ± ± ±
CG, lot no. G0673 1 ug/rnl 100 ng/ml 50 ng/ml 10 ng/rnl 5 ng/rnl 1 ng/rnl
11.4 3.3 2.8 3.4 3.4 2.8
± ± ± ± ± ±
2.6 0.1 1.0 1.0 0.8 ± 0.5 4.5 1.4 0.1 0.1 0.6 1.2
* Nanomoles (± so) of steroid per plate of cells per 22-hr incubation period. t CT cholera enterotoxin.
=
Downloaded from http://jid.oxfordjournals.org/ at University of New South Wales on August 21, 2015
was the receptor for cholera enterotoxin, it very likely was the receptor for the E. coli toxin as well. This conclusion becomes difficult to support, however, if the results of studies delineating the interactions of choleragenoid (the naturally occurring toxoid and binding fragment of the cholera enterotoxin molecule) with the two enterotoxins and with GMt ganglioside are taken into account. The implication of these findings, reported here, is that, at least for adrenal tumor cells in tissue culture, G~fl ganglioside may not be the true receptor for either toxin.
Choleragenoid, G M 1 , and Enterotoxins
S117
Table 2. Lack of effect of choleragenoid (CG), lot no. G0673, on cholera enterotoxin (CT)-induced steroidogenesis in Yl adrenal cells in culture. Test system nmol/plate* Control 5.1 ± 1.0 4.4 ± 0.1 CG (50 ng/ml) 11.0 ± 0.8 CT (1.0 ng/rnl) CG,t then CT 11.5 ± 1.3 6.4 ± 0.3 7.4 ± 0.2
* Nanomoles (± SD) of steroid per plate of cells per 24-hr incubation period. t Added to cultures 4 hr prior to addition of toxin. steroidogenic effects of E. coli toxin by either choleragenoid preparation was observed (table 3). In contrast to the results of the above studies, which demonstrated a lack of effect of choleragenoid on toxin-induced steroidogenesis, choleragenoid was found to interfere with the ability of GMt ganglioside to neutralize the effects of the toxins. This interference was demonstrable only within a very narrow range (approximately fiveto 1O-fold) of ganglioside concentrations (table 4). Use of a concentration of GMt ganglioside within this range showed that choleragenoid produced a dose-related interference effect (table 5). Some interference was noticeable with equal concentrations of choleragenoid and toxin, and maximal interference was observed at concentrations of choleragenoid 10-20 times that of toxin. Choleragenoid interfered with the ability of GMt ganglioside to neutralize the steroidogenic effects of E. coli toxin (table 6). These findings suggest that the interference by choleragenoid was not specific to cholera toxin. Table 3. Lack of effect of choleragenoid (CG) on steroidogenesis induced by the enterotoxins of Vibrio cholerae (CT) and Escherichia coli (ECT) in cultured Yl adrenal cells. CG, lot CG, lot no. G0172 no. G0673 Test system (10 ng/ml) (100 ng/ml) Control Control CT (2 ng/ml) ECT (5 ug/rnl )
4.4 ± 1.5
4.4 ± 0.7
4.7 ± 2.3
9.0± 0.8
11.7 ± 2.4*
11.1 ± 2.2*
6.9 ± 1.5
9.5 ± 1.0* 10.2 ± 2.5* NOTE. Results are expressed as nanomoles (± SD) of steroid per plate of cells per 23-hr incubation period. * CG was added 4 hr prior to CT or ECT.
o 6.67
o o
6.67 6.67 1.67 1.67
o o 33.3
o o
o o o 1.67
2.5 ± 3.3 ± 4.1 ± 17.3 ±
1.4 1.0 0.2 3.2
1.67 1.67 1.67 1.67
8.3 ± 2.1 6.7 ± 1.4 o 8.7 ± 3.4 33.3 19.5 ± 0.7 0.33 o 11.6 ± 0.8 1.67 0.33 33.3 18.7 ± 2.2 1.67 NOTE. GMt ganglioside was incubated at 37 C for 20 min with or without CG or buffer, then CT was added to the mixture for another 20 min of incubation prior to addition of an aliquot of this mixture to the tissue culture medium (see Materials and Methods). Concentrations shown are final concentrations in the medium. * GMt == galactosyl-Neacetylgalactosaminyl [sialosyl] lactosyl ceramide; GGnSLC. t Nanomoles (± SD) of steroid per plate of cells after a total incubation time of 23 hr in minimal essential medium. 33.3
Discussion The results of several investigations have suggested that the ganglioside GMt is or closely resembles the receptor site for cholera enterotoxin [6-9]. The evidence for this hypothesis is based primarily on the ability of the GMt ganglioside to bind, mole for mole, the cholera toxin and prevent Table 5. Interference by choleragenoid (CG) with GMt ganglioside inhibition of cholera enterotoxin (CT)-induced steroidogenesis in cultured Yl adrenal cells. CG CT G MI nmol/plate* (ng/rnl ) (ng/ml) (ng/rnl ) 6.0 ± 0.7 o o o 5.6±0.1 o 1.67 o 5.2 ± 0.6 o o 33.3 20.0 ± 1.1 1.67 o o 12.7 ± 0.3 1.67 1.67 o 22.2 ± 1.3 33.3 1.67 1.67 16.7 19.4 ± 1.7 1.67 1.67 14.6 ± 2.0 3.33 1.67 1.67 1.67 13.9 ± 2.0 1.67 1.67 0.33 12.1 ± 0.5 1.67 1.67 * Nanomoles (± SD) of steroid per plate of cells per 22-hr incubation period.
Downloaded from http://jid.oxfordjournals.org/ at University of New South Wales on August 21, 2015
CT (0.1 ng/ml) CG,t then CT
Table 4. Interference with G Ml ganglioside inhibition of cholera enterotoxin (CT)-induced steroidogenesis by choleragenoid (CG). CG CT (ng/ml) (ng/ml) nmollplatet
Donta
S118
Table 6. Interference by choleragenoid (CG) with G 1YIl ganglioside inhibition of steroidogenesis induced by enterotoxins of Vibrio cholerae (CT) and Escherichia coli (BCT) in cultured Yl adrenal cells. GM l (ng/ml)
o 0.33
o o
o 0.33 0.33
CT (ng/ml)
ECT (f.l.g/ml)
nmollplate*
0 0 66.6
0 0 0
0 0 0
5.5 ± 0.1 6.0 ± 0.2 5.6 ± 0.7
0 0 66.6
1.67 1.67 1.67
0 0 0
24.7 ± 1.7 15.9 ± 2.7 23.2 ± 2.1
0 0 66.6
0 0 0
6.67 6.67 6.67
15.8 ± 3.1 10.4 ± 0.9 15.7 ± 1.8
* Nanomoles (± SD) of steroid per plate of cells per 22-hr incubation period.
its subsequent effects on intestinal and nonintestinal tissues. This evidence, in combination with the findings that choleragenoid, the presumed binding portion of the enterotoxin molecule, can both prevent the effects of the toxin on several target tissues and interfere with the ability of GMt ganglioside to inactivate the toxin, has added considerable support to the hypothesis [6, 7]. Not many model systems have been used, however, to study these various interactions among choleragenoid, G.\fl ganglioside, and the heat-labile enterotoxins of Vibrio cholerae and Escherichia coli. With at least one system, adrenal tumor cells in tissue culture, results similar to those found in previous studies have not been obtained [3, 10]. The results of the studies reported here demonstrate that choleragenoid does not prevent the morphological or steroidogenic effects of cholera toxin on adrenal cells, even if the choleragenoid is incubated with cells for as long as 8 hr before the addition of toxin and if a concentration of choleragenoid equivalent to 106 molecules per cell (an amount more than 3,000-fold greater than the number of toxin molecules per cell required to produce a half-maximal steroidogenic response [3] and 50-100 times more than that needed for a maximal response) is used. Neither of the two lots of choleragenoid produced any toxin-inhibitory effects when tested with a variety of submaximal and maximal steroidogenic concentrations of toxin and with 50- to 500-fold greater concentrations of choleragenoid. In contrast to these findings, choleragenoid, in
Downloaded from http://jid.oxfordjournals.org/ at University of New South Wales on August 21, 2015
0.33 0.33
CG (ng/ml)
concentrations one to 10 times that of toxin, effectively interfered with the ability of the G M l ganglioside to inactivate cholera toxin. Therefore, these results, which demonstrate on the one hand that choleragenoid does not interfere with toxinmediated steroidogenesis in adrenal cells, and on the other that it does interfere with neutralization of the toxin's effects by G M l ganglioside, support the hypothesis that the G~1l ganglioside is not the receptor site for cholera toxin on adrenal cells, and that it represents a nonspecific pseudoreceptor. Further support for this hypothesis stems from the findings that (1) on a molar basis, 100-200 times as much GMt ganglioside as toxin is required to inhibit the toxin's steroidogenic effects fully [10]; (2) GMt ganglioside inactivates the steroidogenic effects of the heat-labile E. coli toxin as effectively as it inactivates those of cholera toxin [10]; (3) either toxin can effectively interfere with the ability of G M 1 ganglioside to neutralize the effects of the other toxin [10]; and (4) choleragenoid can interfere with the OMl ganglioside neutralization not only of the cholera toxin, but of the E. coli toxin as well. Furthermore, the inability of choleragenoid to inhibit the effects of the toxins on adrenal cells suggests either that separate receptor sites exist for choleragenoid (i.e., the binding portion of the toxin molecule) and for the active fragment or that the binding of the intact toxin is much stronger than the binding of the toxoid for that receptor on adrenal cells. (Studies by others using nonadrenal systems have not yielded results showing any differences between the affinities of toxin and toxoid for target sites, e.g., G M 1 ganglioside [6, 8]). That the prior binding of the toxoid fragment is not a prerequisite for toxin activity has been demonstrated by the findings that the "active" subunit of the toxin, fragment A, is capable of morphologic and steroidogenic effects similar to those of the intact toxin, albeit at concentrations 1,000-10,000 times greater than that of toxin [10, 13]. The effects of the active fragment were not neutralizable by GMt ganglioside or by antiserum to choleragenoid when submaximal steroidogenic concentrations of subunit and amounts of ganglioside and antibody capable of complete neutralization of the effects of the whole toxin were used. Antiserum to the intact toxin was slightly inhibitory to the active fragment under
Choleragenoid, G M 1 , and Enterotoxins
References 1. Donta, S. T., King, M., Sloper, K. Induction of
steroidogenesis in tissue culture by cholera enterotoxin. Nature (New BioI.) 243:246-247,1973. 1 S. T. Donta, S. R. Kreiter, and G. WendelschaferCrabb, "Morphological and Steroidogenic Changes in Cultured Adrenal Tumor Cells Induced by a Subunit of Cholera Enterotoxin," manuscript in preparation.
2. Donta, S. T., Moon, H. W., Shipp, S. C. Detection of heat-labile Escherichia coli enterotoxin with the use of adrenal cells in tissue culture. Science 183 :334-336, 1974. 3. Donta, S. T. Comparison of the steroidogenic effects of cholera enterotoxin and ACTH. Am. J. PhysioI. 227:109-113, 1974. 4. Donta, S. T. Differentiation between the steroidogenic effects of cholera enterotoxin and adrenocorticotropin through use of a mutant adrenal cell line. J. Infect. Dis. 129:728-731, 1974. 5. Donta, S. T., Smith, D. M. Stimulation of steroidogenesis in tissue culture by enterotoxigenic Escherichia coli and its neutralization by specific antiserum. Infec. Immun. 9:500-505, 1974. 6. Holmgren, 1. Comparison of the tissue receptors for Vibrio cholerae and Escherichia coli enterotoxins by means of gangliosides and natural cholera toxoid. Infec. Immun. 8:851-859, 1973. 7. Pierce, N. F. Differential inhibitory effects of cholera toxoids and ganglioside on the enterotoxins of Vibrio cholerae and Escherichia coli. J. Exp. Med. 137:1009-1023, 1973. 8. Cuatrecasas, P. Gangliosides and membrane receptors for cholera toxin. Biochemistry 12: 35583566, 1973. 9. King, C. A., van Heyningen, W. E. Deactivation of cholera toxin by a sialidase-resistant monosialosylganglioside. J. Infect. Dis. 127:639-647, 1973. 10. Donta, S. T., Viner, 1. P. Inhibition of the steroidogenic effects of cholera and heat-labile Escherichia coli enterotoxins by G M 1 ganglioside: evidence for a similar receptor site for the two toxins. lnfec. Immun. 11 :982-985, 1975. 11. Finkelstein, R. A., LoSpalluto, J. J. Production of highly purified choleragen and choleragenoid. J. Infect. Dis. 12l(SuppI.) :S63-S72, 1970. 12. Smith, N. W., Sack, R. B. Immunologic cross-reactions of enterotoxin from Escherichia coli and Vibrio cholerae. J. Infect. Dis. 127:164-170, 1973.
Downloaded from http://jid.oxfordjournals.org/ at University of New South Wales on August 21, 2015
the same conditions." Although the results of these studies with subunit A demonstrate that the effects of the toxin can be elicited without the necessity of prior binding of the choleragenoid portion of the molecule to Yl adrenal cells, they do not preclude an important role for the binding fragment, perhaps by facilitation of the binding of the active subunit to its own receptor or by assistance in the localization of the subunit A near its active site. In summary, it would appear that, at least for Yl adrenal cells, the receptor site(s) for the V. cholerae and E. coli enterotoxins is similar to but is probably not Gl>l1 ganglioside (although it may resemble or be analogous to Gr.!!) and may be composed of specific and separate sites for the active and binding subunits, with the sites for the latter perhaps being of various affinities. Further experiments are necessary to clarify the important issues and problems raised by the results of the current studies and to determine the significance of the results achieved with adrenal cells and the relationship thereof to the in vivo effects of the enterotoxins of V. cholerae and E. coli.
S119
•
MARCH 1976
Interactions of Choleragenoid and G M 1 Ganglioside with Enterotoxins of Vibrio cholerae and Escherichia coli in Cultured Adrenal Cells Sam T. Donta
From the Division of Infectious Diseases, Department of Medicine, University of Iowa, and Veterans Administration Hospitals, Iowa City, Iowa
The heat-labile enterotoxins of Vibrio cholerae and Escherichia coli induce morphologic changes and steroidogenesis in clonal lines of adrenal tumor cells in monolayer tissue culture [I, 2]. The results of several studies of the effects of the toxins on these adrenal cells suggested that the two enterotoxins may act by mechanisms that are similar to each other but dissimilar from the mechanism of the natural hormonal inducer, adrenocorticotropic hormone [3-5]. It has been proposed that the ganglioside GM! (galactosyl-N-acetylgalactosaminyl [sialosyl] lactosyl ceramide; GGnSLC) is the natural receptor in the intestine and perhaps in nonintestinal tissue for the enterotoxin of V. cholerae but probably is not the receptor for the heat-labile enteroThis study was supported by grants from the Veterans Administration Research and Education Service and by grants no. AI 11416 and AI 11794 from the National Institute of Allergy and Infectious Diseases. Dr. Donta is a clinical investigator of the Veterans Administration. I thank Ms. Shelley Kreiter for her excellent technical assistance. Please address requests for reprints to Dr. Sam T. Donta, Department of Medicine, Veterans Administration Hospital, Iowa City, Iowa 52240.
Sl1S
toxin of E. coli. This hypothesis is based primarily on the relative degrees of inactivation of the two toxins by G M ! ganglioside [6-9]. Similar results were not obtained, however, when the effects of the ganglioside on the activities of the toxins in adrenal cells were studied [10]. These studies demonstrated not only that the G M 1 ganglioside was capable of equivalent degrees of inhibition of the morphologic and steroidogenic activities of equipotent concentrations of the two toxins, but also that, in competition experiments, either toxin could interfere with the ability of the ganglioside to inactivate the other toxin. Since crude preparations of E. coli toxin were used in all of these studies, it may well be that impurities were present in concentrations capable of interfering with the inactivation of the E. coli toxin by the ganglioside in the nonadrenal systems; however, since the adrenal cells are more sensitive to the effects of the toxin than are the other systems studied, much smaller concentrations of the crude toxin preparation, and perhaps therefore of the impurities, were present and did not interfere with the binding of the E. coli toxin to the ganglioside. The results of these previous studies with adrenal cells suggested that, if the G m ganglioside
Downloaded from http://jid.oxfordjournals.org/ at University of New South Wales on August 21, 2015
The heat-labile enterotoxins of Vibrio cholerae and Escherichia coli induce morphologic changes and steroidogenesis in clonal lines of adrenal tumor cells in tissue culture; these effects are preventable by prior incubation of either toxin with GM ! ganglioside (galactosyl-N-acetylgalactosaminyl [sialosyl] lactosyl ceramide; GGnSLC) but are not preventable by prior incubation of adrenal cells with choleragenoid. Choleragenoid, however, is capable of interfering with the ability of G M ! ganglioside to neutralize the effects of either toxin. These results suggest that the G m ganglioside may not be the true receptor for the toxins on adrenal cells, but that it is acting as a pseudoreceptor. The ability of a subunit (A) of the cholera enterotoxin molecule to induce in adrenal cells morphologic changes and steroidogenesis similar to those effects inducible by the whole toxin indicates the possibility that separate receptor sites on these adrenal cells may exist for the binding and active fragments of the molecule.
Donta
8116
Materials and Methods
Toxins and toxoid. Purified preparations of cholera enterotoxin and choleragenoid were obtained from Dr. R. A. Finkelstein [I 1]. Dialyzed, lyophilized culture filtrates of E. coli 078: H 12 (strain 408-3) were obtained from Dr. R. B. Sack [12]. Ganglioside (GMt) was obtained from Dr. W. E. van Heyningen [9]. The toxins and gangliosides were diluted in a 0.05 M Tris-O.OOlM EDTA buffer system (pH 7.5). Tissue culture methods. Monolayer cultures of Yl adrenal cells were propagated and maintained in minimal essential medium supplemented with 10% fetal calf serum at 37 C in a humidified atmosphere of 95% air and 5% CO 2 • More specific details of the tissue culture experimental methods and of the spectrophotometric assay for ~4,3-ketosteroids extractable from the tissue culture medium have been described [5]. The results are expressed as nanomoles of steroid per plate of cells per incubation period. Some of the variation in the amount of steroid produced between experiments is secondary to either the total incubation time or the density of cells per plate (variation between the number of cells per plate in any given experiment never exceeded 10%). In addition, some subclonal lines of Yl cells have lower basal rates of steroidogenesis than do others. For experiments in which the interactions among GMt ganglioside, choleragenoid, and enterotoxins were studied, Tris buffer or choleragenoid was incubated with ganglioside at 37 C for 20 min; toxin was added to the mixture and the incubation continued for another 20 min. An aliquot was then removed from the mixture and was added to the tissue culture medium surround-
ing Yl cells. Controls with and without ganglioside, choleragenoid, and toxin were similarly incubated with and without buffer. Results
When two different lots of choleragenoid were tested on Yl cells, morphological and steroidogenic changes induced by each of the preparations could be observed (table I); these changes represented 0.001%-0.1% of the activity of the native toxin and presumably were due to small amounts of residual whole toxin in the purified preparations of choleragenoid. This presumption was supported by the observation of similar heat labilities and dose responsiveness of the activities present in the choleragenoid and native toxin preparations. When amounts of choleragenoid 50- to 500-fold greater than the submaximal steroidogenic concentrations of cholera toxin were used and the choleragenoid was incubated with adrenal cells prior to addition of toxin, no inhibition of the effects of the toxin could be demonstrated (table 2), regardless of whether cells were incubated with choleragenoid for as little as 0 hr or as long as 8 hr before the addition of toxin. The amounts (50-100 ng) of choleragenoid used in these experiments were sufficient to provide 10 6 molecules per cell. Similarly, no inhibition of the Table 1. Steroidogenic effects of choleragenoid (CO) on Yl adrenal cells in culture. Test system
nmol/plate*
Control CT,t 10 ng/rnl
3.2 ± 0.1 17.2 ± 0.7
CG, lot no. GOl72 1 ug/rnl 100 ng/ml 50 ng/rnl 10 ng/ml 5 ng/rnl 1 ng/ml
16.7 13.5 10.0 4.0 5.7 4.3
± ± ± ± ±
CG, lot no. G0673 1 ug/rnl 100 ng/ml 50 ng/ml 10 ng/rnl 5 ng/rnl 1 ng/rnl
11.4 3.3 2.8 3.4 3.4 2.8
± ± ± ± ± ±
2.6 0.1 1.0 1.0 0.8 ± 0.5 4.5 1.4 0.1 0.1 0.6 1.2
* Nanomoles (± so) of steroid per plate of cells per 22-hr incubation period. t CT cholera enterotoxin.
=
Downloaded from http://jid.oxfordjournals.org/ at University of New South Wales on August 21, 2015
was the receptor for cholera enterotoxin, it very likely was the receptor for the E. coli toxin as well. This conclusion becomes difficult to support, however, if the results of studies delineating the interactions of choleragenoid (the naturally occurring toxoid and binding fragment of the cholera enterotoxin molecule) with the two enterotoxins and with GMt ganglioside are taken into account. The implication of these findings, reported here, is that, at least for adrenal tumor cells in tissue culture, G~fl ganglioside may not be the true receptor for either toxin.
Choleragenoid, G M 1 , and Enterotoxins
S117
Table 2. Lack of effect of choleragenoid (CG), lot no. G0673, on cholera enterotoxin (CT)-induced steroidogenesis in Yl adrenal cells in culture. Test system nmol/plate* Control 5.1 ± 1.0 4.4 ± 0.1 CG (50 ng/ml) 11.0 ± 0.8 CT (1.0 ng/rnl) CG,t then CT 11.5 ± 1.3 6.4 ± 0.3 7.4 ± 0.2
* Nanomoles (± SD) of steroid per plate of cells per 24-hr incubation period. t Added to cultures 4 hr prior to addition of toxin. steroidogenic effects of E. coli toxin by either choleragenoid preparation was observed (table 3). In contrast to the results of the above studies, which demonstrated a lack of effect of choleragenoid on toxin-induced steroidogenesis, choleragenoid was found to interfere with the ability of GMt ganglioside to neutralize the effects of the toxins. This interference was demonstrable only within a very narrow range (approximately fiveto 1O-fold) of ganglioside concentrations (table 4). Use of a concentration of GMt ganglioside within this range showed that choleragenoid produced a dose-related interference effect (table 5). Some interference was noticeable with equal concentrations of choleragenoid and toxin, and maximal interference was observed at concentrations of choleragenoid 10-20 times that of toxin. Choleragenoid interfered with the ability of GMt ganglioside to neutralize the steroidogenic effects of E. coli toxin (table 6). These findings suggest that the interference by choleragenoid was not specific to cholera toxin. Table 3. Lack of effect of choleragenoid (CG) on steroidogenesis induced by the enterotoxins of Vibrio cholerae (CT) and Escherichia coli (ECT) in cultured Yl adrenal cells. CG, lot CG, lot no. G0172 no. G0673 Test system (10 ng/ml) (100 ng/ml) Control Control CT (2 ng/ml) ECT (5 ug/rnl )
4.4 ± 1.5
4.4 ± 0.7
4.7 ± 2.3
9.0± 0.8
11.7 ± 2.4*
11.1 ± 2.2*
6.9 ± 1.5
9.5 ± 1.0* 10.2 ± 2.5* NOTE. Results are expressed as nanomoles (± SD) of steroid per plate of cells per 23-hr incubation period. * CG was added 4 hr prior to CT or ECT.
o 6.67
o o
6.67 6.67 1.67 1.67
o o 33.3
o o
o o o 1.67
2.5 ± 3.3 ± 4.1 ± 17.3 ±
1.4 1.0 0.2 3.2
1.67 1.67 1.67 1.67
8.3 ± 2.1 6.7 ± 1.4 o 8.7 ± 3.4 33.3 19.5 ± 0.7 0.33 o 11.6 ± 0.8 1.67 0.33 33.3 18.7 ± 2.2 1.67 NOTE. GMt ganglioside was incubated at 37 C for 20 min with or without CG or buffer, then CT was added to the mixture for another 20 min of incubation prior to addition of an aliquot of this mixture to the tissue culture medium (see Materials and Methods). Concentrations shown are final concentrations in the medium. * GMt == galactosyl-Neacetylgalactosaminyl [sialosyl] lactosyl ceramide; GGnSLC. t Nanomoles (± SD) of steroid per plate of cells after a total incubation time of 23 hr in minimal essential medium. 33.3
Discussion The results of several investigations have suggested that the ganglioside GMt is or closely resembles the receptor site for cholera enterotoxin [6-9]. The evidence for this hypothesis is based primarily on the ability of the GMt ganglioside to bind, mole for mole, the cholera toxin and prevent Table 5. Interference by choleragenoid (CG) with GMt ganglioside inhibition of cholera enterotoxin (CT)-induced steroidogenesis in cultured Yl adrenal cells. CG CT G MI nmol/plate* (ng/rnl ) (ng/ml) (ng/rnl ) 6.0 ± 0.7 o o o 5.6±0.1 o 1.67 o 5.2 ± 0.6 o o 33.3 20.0 ± 1.1 1.67 o o 12.7 ± 0.3 1.67 1.67 o 22.2 ± 1.3 33.3 1.67 1.67 16.7 19.4 ± 1.7 1.67 1.67 14.6 ± 2.0 3.33 1.67 1.67 1.67 13.9 ± 2.0 1.67 1.67 0.33 12.1 ± 0.5 1.67 1.67 * Nanomoles (± SD) of steroid per plate of cells per 22-hr incubation period.
Downloaded from http://jid.oxfordjournals.org/ at University of New South Wales on August 21, 2015
CT (0.1 ng/ml) CG,t then CT
Table 4. Interference with G Ml ganglioside inhibition of cholera enterotoxin (CT)-induced steroidogenesis by choleragenoid (CG). CG CT (ng/ml) (ng/ml) nmollplatet
Donta
S118
Table 6. Interference by choleragenoid (CG) with G 1YIl ganglioside inhibition of steroidogenesis induced by enterotoxins of Vibrio cholerae (CT) and Escherichia coli (BCT) in cultured Yl adrenal cells. GM l (ng/ml)
o 0.33
o o
o 0.33 0.33
CT (ng/ml)
ECT (f.l.g/ml)
nmollplate*
0 0 66.6
0 0 0
0 0 0
5.5 ± 0.1 6.0 ± 0.2 5.6 ± 0.7
0 0 66.6
1.67 1.67 1.67
0 0 0
24.7 ± 1.7 15.9 ± 2.7 23.2 ± 2.1
0 0 66.6
0 0 0
6.67 6.67 6.67
15.8 ± 3.1 10.4 ± 0.9 15.7 ± 1.8
* Nanomoles (± SD) of steroid per plate of cells per 22-hr incubation period.
its subsequent effects on intestinal and nonintestinal tissues. This evidence, in combination with the findings that choleragenoid, the presumed binding portion of the enterotoxin molecule, can both prevent the effects of the toxin on several target tissues and interfere with the ability of GMt ganglioside to inactivate the toxin, has added considerable support to the hypothesis [6, 7]. Not many model systems have been used, however, to study these various interactions among choleragenoid, G.\fl ganglioside, and the heat-labile enterotoxins of Vibrio cholerae and Escherichia coli. With at least one system, adrenal tumor cells in tissue culture, results similar to those found in previous studies have not been obtained [3, 10]. The results of the studies reported here demonstrate that choleragenoid does not prevent the morphological or steroidogenic effects of cholera toxin on adrenal cells, even if the choleragenoid is incubated with cells for as long as 8 hr before the addition of toxin and if a concentration of choleragenoid equivalent to 106 molecules per cell (an amount more than 3,000-fold greater than the number of toxin molecules per cell required to produce a half-maximal steroidogenic response [3] and 50-100 times more than that needed for a maximal response) is used. Neither of the two lots of choleragenoid produced any toxin-inhibitory effects when tested with a variety of submaximal and maximal steroidogenic concentrations of toxin and with 50- to 500-fold greater concentrations of choleragenoid. In contrast to these findings, choleragenoid, in
Downloaded from http://jid.oxfordjournals.org/ at University of New South Wales on August 21, 2015
0.33 0.33
CG (ng/ml)
concentrations one to 10 times that of toxin, effectively interfered with the ability of the G M l ganglioside to inactivate cholera toxin. Therefore, these results, which demonstrate on the one hand that choleragenoid does not interfere with toxinmediated steroidogenesis in adrenal cells, and on the other that it does interfere with neutralization of the toxin's effects by G M l ganglioside, support the hypothesis that the G~1l ganglioside is not the receptor site for cholera toxin on adrenal cells, and that it represents a nonspecific pseudoreceptor. Further support for this hypothesis stems from the findings that (1) on a molar basis, 100-200 times as much GMt ganglioside as toxin is required to inhibit the toxin's steroidogenic effects fully [10]; (2) GMt ganglioside inactivates the steroidogenic effects of the heat-labile E. coli toxin as effectively as it inactivates those of cholera toxin [10]; (3) either toxin can effectively interfere with the ability of G M 1 ganglioside to neutralize the effects of the other toxin [10]; and (4) choleragenoid can interfere with the OMl ganglioside neutralization not only of the cholera toxin, but of the E. coli toxin as well. Furthermore, the inability of choleragenoid to inhibit the effects of the toxins on adrenal cells suggests either that separate receptor sites exist for choleragenoid (i.e., the binding portion of the toxin molecule) and for the active fragment or that the binding of the intact toxin is much stronger than the binding of the toxoid for that receptor on adrenal cells. (Studies by others using nonadrenal systems have not yielded results showing any differences between the affinities of toxin and toxoid for target sites, e.g., G M 1 ganglioside [6, 8]). That the prior binding of the toxoid fragment is not a prerequisite for toxin activity has been demonstrated by the findings that the "active" subunit of the toxin, fragment A, is capable of morphologic and steroidogenic effects similar to those of the intact toxin, albeit at concentrations 1,000-10,000 times greater than that of toxin [10, 13]. The effects of the active fragment were not neutralizable by GMt ganglioside or by antiserum to choleragenoid when submaximal steroidogenic concentrations of subunit and amounts of ganglioside and antibody capable of complete neutralization of the effects of the whole toxin were used. Antiserum to the intact toxin was slightly inhibitory to the active fragment under
Choleragenoid, G M 1 , and Enterotoxins
References 1. Donta, S. T., King, M., Sloper, K. Induction of
steroidogenesis in tissue culture by cholera enterotoxin. Nature (New BioI.) 243:246-247,1973. 1 S. T. Donta, S. R. Kreiter, and G. WendelschaferCrabb, "Morphological and Steroidogenic Changes in Cultured Adrenal Tumor Cells Induced by a Subunit of Cholera Enterotoxin," manuscript in preparation.
2. Donta, S. T., Moon, H. W., Shipp, S. C. Detection of heat-labile Escherichia coli enterotoxin with the use of adrenal cells in tissue culture. Science 183 :334-336, 1974. 3. Donta, S. T. Comparison of the steroidogenic effects of cholera enterotoxin and ACTH. Am. J. PhysioI. 227:109-113, 1974. 4. Donta, S. T. Differentiation between the steroidogenic effects of cholera enterotoxin and adrenocorticotropin through use of a mutant adrenal cell line. J. Infect. Dis. 129:728-731, 1974. 5. Donta, S. T., Smith, D. M. Stimulation of steroidogenesis in tissue culture by enterotoxigenic Escherichia coli and its neutralization by specific antiserum. Infec. Immun. 9:500-505, 1974. 6. Holmgren, 1. Comparison of the tissue receptors for Vibrio cholerae and Escherichia coli enterotoxins by means of gangliosides and natural cholera toxoid. Infec. Immun. 8:851-859, 1973. 7. Pierce, N. F. Differential inhibitory effects of cholera toxoids and ganglioside on the enterotoxins of Vibrio cholerae and Escherichia coli. J. Exp. Med. 137:1009-1023, 1973. 8. Cuatrecasas, P. Gangliosides and membrane receptors for cholera toxin. Biochemistry 12: 35583566, 1973. 9. King, C. A., van Heyningen, W. E. Deactivation of cholera toxin by a sialidase-resistant monosialosylganglioside. J. Infect. Dis. 127:639-647, 1973. 10. Donta, S. T., Viner, 1. P. Inhibition of the steroidogenic effects of cholera and heat-labile Escherichia coli enterotoxins by G M 1 ganglioside: evidence for a similar receptor site for the two toxins. lnfec. Immun. 11 :982-985, 1975. 11. Finkelstein, R. A., LoSpalluto, J. J. Production of highly purified choleragen and choleragenoid. J. Infect. Dis. 12l(SuppI.) :S63-S72, 1970. 12. Smith, N. W., Sack, R. B. Immunologic cross-reactions of enterotoxin from Escherichia coli and Vibrio cholerae. J. Infect. Dis. 127:164-170, 1973.
Downloaded from http://jid.oxfordjournals.org/ at University of New South Wales on August 21, 2015
the same conditions." Although the results of these studies with subunit A demonstrate that the effects of the toxin can be elicited without the necessity of prior binding of the choleragenoid portion of the molecule to Yl adrenal cells, they do not preclude an important role for the binding fragment, perhaps by facilitation of the binding of the active subunit to its own receptor or by assistance in the localization of the subunit A near its active site. In summary, it would appear that, at least for Yl adrenal cells, the receptor site(s) for the V. cholerae and E. coli enterotoxins is similar to but is probably not Gl>l1 ganglioside (although it may resemble or be analogous to Gr.!!) and may be composed of specific and separate sites for the active and binding subunits, with the sites for the latter perhaps being of various affinities. Further experiments are necessary to clarify the important issues and problems raised by the results of the current studies and to determine the significance of the results achieved with adrenal cells and the relationship thereof to the in vivo effects of the enterotoxins of V. cholerae and E. coli.
S119
