Saccharin and Cyclamate Inhibit Binding of Epidermal Growth Factor (Tumor Promotion/Phorbol Esters/Hormone Response Control Unit/Carcinogenesis) L

Proc. NatL Acad. Sci. USA Vol. 78, No. 2, pp. 1042-1046, February 1981 Cell Biology

Saccharin and cyclamate inhibit binding of epidermal growth factor (tumor promotion/phorbol esters/hormone response control unit/carcinogenesis) L. S. LEE General Electric Company, Corporate Research and Development, P.O. Box 8, Schenectady, New York 12301 Communicated by Charles P. Bean, November 3, 1980 ABSTRACT The binding of '5I-labeled mouse epidermal parently is not due to a direct competition between TPA and growth factor (EGF) to 18 cell lines, including HeLa (human car- EGF for the active site of either the EGF receptor or the TPA cinoma), MDCK (dog kidney cells), HTC (rat hepatoma), K22 (rat liver), HF (human foreskin), GM17 (human skin fibroblasts), XP receptor (7). Recent studies also revealed that EGF binding is (human xeroderma pigmentosum fibroblasts), and 3T3-LI (mouse accompanied by a host of dynamic changes in the cell mem- fibroblasts), was inhibited bysaccharin and cyclamate. Thehuman brane including patching, internalization, and down-regulation cells were more sensitive to inhibition by these sweeteners than (19-23). Although the specific physiological roles of EGF and mouse or rat cells. EGF at doses far above the physiological levels EGF receptors are not well understood, it seems plausible that reversed the inhibition in rodent cells but not in HeLa cells. In the regulation of EGF levels in the body and the regulation of HeLa cells, the doses of saccharin and cyclamate needed for 50% inhibition were 3.5 and 9.3 mg/ml, respectively. Glucose, 2-deox- EGF receptors in the cells could be important events in tumor yglucose, sucrose, and xylitol did not inhibit EGF binding. Pre- promotion. vious studies have shown that phorbol esters, strongly potent tu- The carcinogenicity of saccharin and cyclamate has received mor promoters, also inhibit EGF binding to tissue culture cells. considerable attention because of their widespread use as ar- To explain the EGF binding inhibition by such greatly dissimilar tificial sweeteners (24-29, *). Much work has been done on molecules as phorbol esters, saccharin, and cyclamate, it is sug- their mutagenicity in bacteria (30-31), their tumorigenicity in gested that they operate through the activation of a hormone re- animals (24-28, t), and their epidemiological correlations with sponse control unit. human cancers (32-38). The reported effects of these artificial sweeteners are confusing and contradictory. It seems more Cancer-inducing agents have been classified as initiators or pro- likely that saccharin and cyclamate function as weakly potent moters according to their effects on the carcinogenic process tumor promoters rather than initiators when administered to (1-5). A tumor promoter is a substance which, when repeatedly animals at high doses (29, 39, 40), although this remains to be applied to animals previously exposed to a tumor initiator, confirmed. The molecular action of these two compounds is greatly enhances the production of tumors (1-5). Extensive unknown. Identification of the molecular target in the cell may studies have revealed that initiators interact with cellular DNA, provide new insight in the elucidation of their tumor-promoting but the initial biological events in the action of tumor promoters activity. are less-well understood (2, 3). Current knowledge of tumor Accordingly, it seemed worthwhile to determine whether promotion is largely derived from studies of the biological ef- some of the effects exerted by TPA, such as the inhibition of fects exerted by the potent tumor promoter 12-0-tetradeca- EGF binding, might also be exerted by them. I report here that noylphorbol 13-acetate (TPA) (1-5). The effects exhibited by such an effect occurs within 1 hr of exposure of human, mouse, such promoters at hormone-like levels in cellular systems have hamster, rat, and dog cells in culture to saccharin or cyclamate been used to search for the receptor systems involved in tumor of these at the promotion (3, 6). It recently was shown that phorbol esters in- and that it is not due to the toxicity compounds hibit the binding of epidermal growth factor (EGF) to cellular levels used. receptors and that their potencies in the inhibition of EGF binding correlate well with their effects as tumor promoters on MATERIALS AND METHODS mouse skin (6, 7). Materials. Murine EGF was purchased from Collaborative Mouse EGF is a peptide of molecular weight 6000 (8, 9). It Research, Waltham, MA. Other radioactive compounds were exhibits all the biological activities ascribed to human EGF and from New England Nuclear. TPA was from Consolidated competes specifically for binding to human EGF receptors (10, Chemical, Biewster, NY. Cell culture materials were from 11). It can stimulate proliferation of both ectodermal and me- GIBCO. Saccharin and cyclamate were obtained from Sigma, sodermal cells in culture, increase deoxyglucose transport, in- except for saccharin preparation lot 1022 and the purified sac- duce the enzyme ornithine decarboxylase, and stimulate pros- charin that were originally used and distributed by Toxicology taglandin synthesis (12-15). Rose et aL (16) have shown that Research Division of the Health Protection Branch, Health and EGF (2-5 ,ug/g of body weight) can enhance skin tumor in- Welfare, Canada (25). All other chemical reagents were from duction in mice. Many of the effects induced by EGF are also Sigma. Saccharin and cyclamate stock solutions were prepared exerted by TPA (6, 7). The reason for such parallelism in biological activity is not clear at this time. However, it has been Abbreviations: TPA, 12-0-tetradecanoylphorbol 13-acetate; EGF, ep- shown thatTPA can synergistically enhance the mitogenic effect idermal growth factor; ME medium, Eagle's mimimal essential of EGF (17, 18) and that TPA inhibition of EGF binding ap- medium. * Stavric, B. (1977) in Proceedings, Toxicology Forum on Saccharin (Center for Continuing Education, Univ. of Nebraska, Medical Cen- The publication costs of this articlewere defrayed in part by page charge ter, Omaha, NE, May 9), pp. 73-80. payment. This article must therefore be hereby marked "advertise- t Wisconsin Alumni.Research Foundation (1973) Long Term Saccharin ment" in accordance with 18 U. S. C. §1734 solely to indicate this fact. Feeding in Rats, Final Report (WARF, Madison, WI). 1042 Downloaded by guest on October 2, 2021 Cell Biology: Lee Proc. Natl. Acad. Sci. USA 78 (1981) 1043 in Dulbecco's phosphate-buffered saline. TPA stock solutions obtained from Collaborative Research and had a specific activity were prepared in dimethyl sulfoxide. of 55,000 cpm/ng. Cell Culture. The media used in culturing different cell lines were as follows: A, Eagle's minimal essential medium (ME RESULTS medium) plus 5% fetal calf serum; B, ME medium plus 10% The dose-response relationships for the inhibition of EGF heat-inactivated fetal calf serum; C, ME medium plus 5% heat- binding by addition of saccharin and cyclamate 10 min before inactivated fetal calf serum; D, Dulbecco's modified ME me- 125I-EGF are shown in Fig 1. The doses causing 50% inhibition dium plus 10% calf serum; E, F-12 medium plus 10% heat-in- were about 3.5 mg/ml for saccharin and 9.3 mg/ml for cycla- activated fetal calf serum; F, Dulbecco's ME medium plus 10% mate, when 2 ng of 125I-EGF was used in binding to 3.65 x 10' fetal calf serum; G, Dulbecco's ME medium plus 10% heat-in- HeLa cells. Saccharin inhibition of EGF binding was not due activated fetal calf serum; H, F-12medium plus 10% calf serum; to impurities in the preparations because lot 1022 and purified and I, F-12 medium plus 10% fetal calf serum. All cultures were saccharin [used in the Canadian studies (25)] gave the same re- grown at 370C in humidified 5% C02/95% air. The cells were sults in our assays. Inhibition was negligible when the dose of counted with a Coulter Counter (model Zb). saccharin was less than 1 ,ug/ml or that of cyclamate was less Human carcinoma HeLa cells were grown in medium D. than 10 pug/ml (Fig. 1 Insets). Slight variation of this 50% in- HTG was a thioguanine-resistant mutant of HeLa. The K22 hibition dose may occur with different doses of EGF or cell epithelial cell line was isolated from liver of normal young adult numbers. Similar potencies of saccharin and cyclamate in EGF- Lew/Mai rats (41) and was grown in medium E. The HTC cell binding inhibition were also observed in other human cell types line was originally isolated from Morris hepatoma 7288C in- tested. For instance, in GM17 the 50% inhibition doses were duced in a BUF rat by an aromatic amine carcinogen and was 4.6 and 10.3 mg/ml for saccharin and cyclamate, respectively; grown in medium H. in XPII they were 7.6and 11.6 mg/ml, respectively. When the The IAR27 and IAR20-PCI epithelial cell lines were derived binding assay was done in the presence of unlabeled EGF at from normal 10-week-old BD VI rats (42) and provided by R. levels (200 ng/ml) far above the physiological level, together Montesano. At 24 weeks in culture, they did not form tumors with either saccharin at 3 mg/ml or cyclamate at 10 mg/ml, this but at 60 weeks in culture they did. They were maintained in inhibition was still observed in HeLa cells (Fig. 2). On the other medium I. Abbreviations used: IARI, IAR20PCI-24 week, pas- hand, there may be species differences (see below). sage 14; IAR2, IAR20PCI-40 week, passage 21; IAR3, IAR27- 9 24 week, passage 5; IAR4, IAR27-60 week, passage 5. 1001 100 .- 0 -0 MDCK (dog kidney cells) were grown in medium G. HF 90 (human foreskin fibroblasts) were grown in medium F; passage e 11 was used. 3T3 (mouse fibroblasts) were grown in medium 80 80 C; 3T3-BP was a line transformed from 3T3 by treatment with 0.1 1 10 100 1000 benzo[a]pyrene and was grown in medium A. GM17 was a hu- 0 JAg1MI man skin fibroblast line derived from a man with a translocation 60 between chromosomes 15 and 18 and was grown in medium B. CHO (Chinese hamster ovary) cells were grown in medium F. XP cells were skin fibroblasts of human xeroderma pigmento- 40 [ sum; they were obtained from the American Type Culture Col- lection and were maintained in medium B [XPI, ATCC-CRL 1162 (WoMec), passage 30, sex M, age 27; XPII, ATCC-CRL 20 - 0 1233 (Jay Tim), passage 15, sex F, age 7]. 1-. Binding Assay. The binding assay with l"I-labeled EGF 0 (125I-EGF) was performed as described (7). The binding buffer e-0 2 4 6 8 10 12 14 u consisted of ME medium plus (per liter): 1 g of bovine serum C4. Cyclamate, mg/ml albumin, 0.1 g of KCl, 0.1 g of KH2PO4, 4.0 g of NaCl, and 0 1.06 g of Na2HPO4-7H2O. The binding was carried out with -o a 55-min incubation at 37°C. Nonspecific binding was deter- ~Z4 mined by measurement in the presence of unlabeled EGF (2 ,ug/ml) and this value was subtracted from the value obtained with the labeled ligand alone. All assays were performed twice and the values presented are means. Duplicates agreed within 10%. Iodination of EGF. This was performed by using the chloramine-T method according to Carpenter and Cohen (10) with slight modifications. Briefly, 3 mCi (1 Ci = 3.7 x 1010 becque- rels) of carrier-free Na'25I solution (17 Ci/mg) was adjusted to pH 7.5 and mixed with 0.1 ml of EGF in Dulbecco's phosphate- buffered saline 20 ,ug/ml. Then, 12.5 ,ug of chloramine-T in water was added; after 45 sec at room temperature, the reaction was stopped with 25 ,ug of sodium metabisulfite, 0.1 ml of KI (100 mg/ml) was added, and the reaction vial was washed three 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 times with 0.1 ml of KI solution. The labeled EGF was then Saccharin, mg/ml isolated with a Bio-Gel P-2 column. The elution buffer con- FIG. 1. Effectof cyclamate (Upper) or saccharin (Lower) on binding tained 0.1% bovine serum albumin in 3ulbecco's phosphate- of EGF to HeLa cells. HeLa cells (3.65 x 106) were assayed for EGF buffered saline. Lot A had a specific activity of 13,000 cpm/ binding with 2 ng of lmI-EGF and various amounts of sodium cycla- ng; lot B had a specific activity of 61,739 cpm/ng; lot C was mate or saccharin included in the binding buffer (1.5 ml). Downloaded by guest on October 2, 2021 1044 Cell Biology: Lee Proc. Natl. Acad. Sci. USA 78 (1981)

Table 1. Effects of various reagents on EGF binding to HeLa cells 100 25I-EGF Reagents bound, cpm Xylitol 6237 80 2-Deoxy-D-glucose 5768 L-Glucose 5557 4 0 60 - f-D(+)-Glucose 6283 a-D(+)-Glucose 6541 Sodium saccharin 1253 Sodium cyclamate 2333 TPA 902 Sucrose 5973 Phosphate-buffered saline 6179 Dimethyl sulfoxide 6097

The assay was performed with 1.6 x 10 HeLa cells and 64,680 cpm. of 125I-EGF (13,000 cpm/ng). Amounts of the reagents added simul- taneously with 1"I-EGF to the binding buffer: TPA, 100 ng/ml; di- x methyl sulfoxide, 0.01%; all other reagents, 13.3 mg/ml.

0 0 100 300' 500 Unlabeled EGF, ng assay. TPA inhibited the binding of physiological levels of EGF to various cells in various degrees, confirming previous findings 20 (6, 7). 0.5 The inhibition was seen with physiological levels of EGF and milligram per milliliter levels of saccharin and cyclamate in normal cells (XPI, XPII, 3T3-L1, IAR3, IAR1) and tumor cells (HeLa, IAR2, IAR4), in fibroblasts (GM17) and epithelial cells (MDCK and K22), and in cells derived from liver (K22) and kidney (MDCK) as well as from human skin fibroblasts of both sexes (XPI and XPII). The binding of EGF at 92.86 ng/ml to IAR3 cells was not inhibited by saccharin at 13.3 mg/ml, although the binding of EGF at 3.89 ng/ml to these cells was inhibited by this saccharin dose. In addition, dose-response Free ng ldI-EGF, studies showed that saccharin or cyclamate at less than 6.0 mg/ ml had no effect on the binding of EGF at 3.89 ng/ml to IAR3 FIG. 2. Dependence of EGF-binding to HeLa cells on the concen- cells. Similarly, binding with EGF at 19.6 ng/ml was not sig- trations of EGF. 125I-EGF and various amounts of (Upper) (23.8 ng) nificantly inhibited by saccharin at 5 mg/ml or cyclamate at 6.6 unlabeled EGF were used in the binding to 3.1 x 106 HelLa cells in the presence of saccharin at 3 mg/ml ( x), cyclamate at 10 mg/ml (o), or mg/ml when mouse 3T3 cells were used in the binding assays. solvent alone (e). (Lower) 12,51-EGF (1-5 ng) was used in the binding On the other hand, appreciable inhibition by these compounds to 3.1 x 10' HeLa cells in the presence Lof saccharin at 3 mg/ml (x), was observed in mouse 3T3-L1 cells (Table 3). It thus seems that cyclamate at 10 mg/ml (o), or solvent alone (e). certain mouse fibroblast and rat liver cells are less sensitive to the inhibition of the EGF binding by saccharin and cyclamate than are human cells. in human cell lines such as HeLa When other sweeteners such as xylitol, sucrose, or glucose Also, analogs were tested at a dose of 13.3 mg/ml, no effects on the EGF binding were observed (Table 1). Table 2. Effects of TPA, saccharin, and cyclamate on adhesion and Table 2 shows the effects of saccharin, cyclamate, and TPA growth of human cells on the growth and detachment of human cells HeLa and XPII. Growth None of these compounds showed an effect on the adhesion of Adhesion test,* test,t no. human cells when cultures were exposed to these agents for 1 no. cells x 10-' cells x 10-' hr under the assay conditions. In agreement with previous ob- servations, TPA showed slight inhibition of the growth of HeLa Reagents HeLa XPII HeLa cells (7). On the other hand, growth inhibition was observed Dimethyl sulfoxide after 2 days of incubation when either saccharin or cyclamate (0.01%) 2.07 ± 0.11 0.35 ± 0.04 3.80 ± 0.39 at 1.7 mg/ml was included in the culture medium. I have not None 1.99 ± 0.03 0.33 ± 0.04 3.74 ± 0.06 determined whether growth inhibition resulted from inhibition Saccharin 2.01 ± 0.21 0.34 ± 0.02 3.08 ± 0.65 of EGF binding, from impurities in the preparation, or from Cyclamate 1.91 ± 0.12 0.32 ± 0.01 2.62 ± 0.40 the metabolites of saccharin or cyclamate. TPA 2.17 ± 0.09 0.33 ± 0.04 2.79 ± 0.23 The effects of TPA, saccharin, and cyclamate on EGF-bind- Results are shown as mean ± SD. ing to cells derived from various species and cell types are listed * The adhesion test was performed by determining the cell numbers in Table 3. These results have been confirmed bydose-response in dishes after treating the cells with the indicated reagents in the studies. Because the number -of EGF receptors or the affinity binding buffer (saccharin, 13.3 mg/ml; cyclamate, 13.3 mg/ml; TPA, of mouse EGF for the receptors on different cells may vary 100 ng/ml) at 370C for 60 min in 5% C02/95% humidified air and 12 then washing them three times with 5 ml of cold binding buffer. greatly, various amounts of 1-EGF had to be used in the assay t For the growth test, 1.4 x 106 HeLa cells were plated 24 hr before for assessing EGF binding and its inhibition by tumor pro- the medium was replaced with the same growth medium plus the moters in different cells. Whenever ~possible, however, nearly indicated reagents (saccharin, 3 mg/ml; cyclamate, 3 mg/ml; TPA, physiological levels of EGF (about 1 ng/ml) were used in the 100 ng/ml) and the cell numbers were determined 3 days later. Downloaded by guest on October 2, 2021 Cell Biology: Lee Proc. NatL Acad. Sci. USA 78 (1981) 1045

Table 3. Effects of TPA, saccharin, and cyclamate on EGF binding Conditions Cells, no. 125I-EGF, 1'I-EGF '25I-EGF bound, % of 1'2I-EGF added x 10-6 ng lot Cells Control TPA Saccharin Cyclamate 3.6 2.0 A HeLa 17.7 1.5 4.1 7.0 1.1 12.37 A HF 1.7 0.64 0.50 0.42 0.8 3.0 B GM17 3.7 0.58 0.81 1.22 0.68 4.04 B XPI 2.86 0.85 1.79 1.37 0.9 2.5 B XPII 3.89 0.54 0.91 1.58 3.7 18.68 A HTG 8.0 2.1 2.1 1.22 3.3 2.03 B K22 2.11 0.83 1.39 1.12 2.1 186.76 A HTC 0.047 0.019 0.033 0.026 1.14 4.64 B IAR1 2.18 0.57 1.73 0.97 1.2 0.85 C IAR2 1.11 0.21 0.35 0.18 1.3 3.89 B IAR3 0.99 0.11 0.65 0.36 1.24 12.4 B IAR4 0.67 0.13 0.24 0.13 2.0 0.58 C MDCK 4.58 0.18 1.10 0.99 1.76 1.6 B 3T3 1.88 0.81 1.32 1.06 1.8 1.6 B 3T3-BP 0.19 0.10 0.12 0.07 1.2 2.5 B 1OT1/2 3.06 0.20 3.00 1.61 2.7 249 A CHO 0.037 0.000 0.019 0.007 0.91 3.89 B 3T3-L1 0.61 0.05 0.17 0.14 The binding assay was performed 3 days after the cells had been plated in 5-cm petri dishes. The concentrations of the reagents included in the binding buffer (1.5 ml) were: TPA, 100 ng/ml; saccharin, 13.3 mg/ml; cyclamate, 13.3 mg/ml.

and XPII, saccharin at 13.3 mg/ml was more potent than cy- shown that TPA tumor-promoting doses (100 to 1 nM in tissue clamate at 13.3 mg/ml whereas in other species, such as IAR culture or 0.1-10 /ig on mouse skin) are also in the same range cells (rat) and 10T1/2 (mouse), the converse was true. as the doses required to inhibit EGF binding. It seems that a class of promoting compounds which operate through inter- DISCUSSION action with the EGF receptors may be detected with the EGF binding inhibition assay, although a dose-related condition may These results have shown that the EGF receptor system may be required (see below). be an important molecular target with which saccharin and cy- In this sense, whether saccharin orcyclamate can be accepted clamate interact. The observation that saccharin and cyclamate as tumor promoters based on EGF-binding inhibition or EGF- inhibit EGF binding in vitro raises several important questions. binding assay can be established as a testing system based on (a) Is saccharin or cyclamate a tumor promoter? (b) Is EGF bind- their tumor promoting activity is better approached from re- ing inhibition a useful assay system for the activity of tumor phrasing questions c and d as follows: (e) What is the biological promoters? (c) If the answers to questions a and b are "Yes," property that can account for the 125I-EGF binding inhibition what biological properties can account for the observed corre- by the structurally related phorbol esters (tricyclic diterpene lation between tumor promotion and the EGF binding inhi- esters) and structurally unrelated saccharin (1,2-benzisothia- bition among various structurally related and unrelated sub- zolin-3-one 1,1-dioxide), cyclamate (cyclohexanesulfamic acid), stances such as phorbol esters, saccharin, and EGF? (d) If the and EGF (epidermal growth factor), without referring to their answers to questions a and b are "No," what is the reason for carcinogenic properties? Also, () if such a biological property the observed "apparent" correlation between EGF-binding in- can be identified, under what conditions can this biological hibition by the tumor promoting phorbol esters and by the sus- property be related to tumor promotion and under what con- pected promoters saccharin and cyclamate? ditions will it not be related to tumor promotion? At present, there is no definite answer to question a. Al- In order to answer question e, it should be pointed out that though it is more likely that saccharin and cyclamate are weak phorbol esters bind to a receptor that is distinct from the EGF tumor promoters with very weak initiating activity (29, 39, 40) receptor (7, 43). Cyclamate and saccharin do not bind to the which may not conform to the strict criterion established with TPA receptor (unpublished results), nor do they share any potent carcinogens, this has not been generally accepted structural similarity with EGF. It seems reasonable to expect (24-42). On the other hand, at present there are not enough that all cells contain some sort of a hormone response control data to permit the unequivocal claim that the EGF binding assay unit that regulates the abilities of a certain group of receptors can be used without conditions for the testing of the activity of to participate in their hormone-binding processes and modu- all tumor promoters. To establish such an assay system would lates the expression of the receptor response signals. Such a require testing of saccharin or cyclamate analogs whose carcin- control unit would normally be interacting with the specific ogenicities are established unequivocally, as well as the testing binding products of the hormones or growth factors with their of all the known tumor initiators and promoters. Nevertheless, respective receptors, such as the EGF-receptor complex or the EGF binding is inhibited by tumor-promoting phorbol esters. binding product(s) of TPA and the receptor(s) sensitive to-it. Furthermore, EGF per se is tumor promoting at high doses. Thus, the formation of one receptor complex would exert, It is interesting to point out that the doses of saccharin or cy- through this control unit, an effect on the availability of another clamate required to demonstrate their tumorigenic activity are receptor. The observed inhibition of EGF binding by saccharin very high (0.1 to 0.001 M in tissue culture or5% in diet in animal and cyclamate suggests that this hormone response control unit tests) (26, 39), as are the doses required to demonstrate signif- can also be activated by simple organic molecules at sufficiently icant EGF binding inhibition. Previous work (1-5, 39) has high concentrations. Downloaded by guest on October 2, 2021 1046 Cell Biology: Lee Proc. Natd Acad. Sci. USA 78 (1981) With the above hypothesis to answer question e, we can de- 18. Dicker, P. & Rozengurt, E. (1978) Nature (London) 276, 723-726. rive a reasonable theory to answer questionf: namely, in cells 19. Das, M. & Fox, C. F. (1978) Proc. Natl Acad. Sci. USA 75, and ex- 2644-2648. that have been exposed to a tumor initiator, prolonged 20. Schlessinger, J., Shechter, Y., Cuatrecasas, P., Willingham, M. cess activation of the hormone response control unit can also C. & Pastan, I. (1978) Proc. Natl Acad. Sci. USA 75, 5353-5357. affect the permanent changes in gene expressions that are in- 21. Haigler, H., Ash, J. F., Singer, S. J. & Cohen, S. (1978) Proc. Natl volved in tumor promotion. The classical observation that low Acad. Sci. USA 75, 3317-3321. doses of tumor initiators produce negligible numbers of tumors 22. McKanna, J. A., Haigler, H. T. & Cohen, S. (1979) Proc. Natl in the absence of exogenous tumor promoters (1-3) indicates Acad. Sci. USA 76, 5689-5693. 23. Maxfield, F. R., Davies, P. J. A., Klempner, L., Willingham, M. that endogenous growth factors, such as EGF, do not have tu- C. & Pastan, I. (1979) Proc. Natl Acad. Sci. USA 76, 5731-5735. mor-promoting activities at normal physiological levels even 24. Hicks, R. M., Wakefield, J. St. J. & Chowaniec, J. (1975) Chem- though tumor promotion has been reported at very high levels Biol Interact. 11, 225-233. (16). Thus, the degree of stimulation of the hormone response 25. Arnold, D. L., Moodie, C. A., Grice, N. C., Collins, S. M., control unit required for tumor promotion may require that McGuire, P. F. & Munro, I. C. (1977) Long Term Toxicity of some threshold be exceeded. Orthotoluene-sulfonamide and Sodium Saccharin in the Rat: An Interim Report (Toxicology Research Division, Health Protection Further studies are required to deduce the way the hormone Branch, National Health and Welfare Ministry, Ottawa, Canada). response control unit responds to stimulation by various com- 26. Division of Pathology, Food and Drug Administration (1973) Sub- pounds in different systems under various conditions, which acute and Chronic Toxicity and Carcinogenicity of Various Dose may shed more light on the tumor-promoting activity of sac- Levels of Sodium Saccharin, Final Report (Government Printing charin or cyclamate. Office, Washington, DC), pp. 169-170. 27. Munro, I. C., Moodie, C. A., Krewski, K. & Grice, H. C. (1975) Thanks are due to C. Kania and M. Lynch for technical help. Toxicol Appl Pharmacol 32, 513-526. 28. Friedman, L., Richardson, H. L., Richardson, M. E., Lethco, E. 1. Berenblum, I. (1975) in Cancer: A Comprehensive Treatise, ed. J., Wallace, W. C. & Sauro, F. M. (1972)J. Natl Cancer Inst. 49, Becker, F. F. (Plenum, New York), Vol. 1, pp. 323-344. 751-764. 2. Boutwell, R. K. (1974) CRC Crit. Rev. Toxicol. 2, 419-443. 29. Hicks, R. M., Chowaniec, J. & Wakefield, J. St. J. (1978) in Mech- 3. Slaga, T. J., Sivak, A. & Boutwell, R. K., eds. (1978) Mechanisms anisms of Tumor Promotion and Cocarcinogenesis, Carcinogen- of Tumor Promotion and Cocarcinogenesis, Carcinogenesis: A esis: A Comprehensive Survey, eds. Slaga, T. J., Sivak, A. & Bou- Comprehensive Survey (Raven, New York), Vol. 2. twell, R. K. (Raven, New York), Vol. 2, pp. 475-489. 4. Weinstein, I. B., Yamasaki, H., Wigler, M., Lee, L. S., Fisher, 30. Batzinger, R. P., Ou, S. L. & Bueding, E. (1977) Science 198, P. B., Jeffrey, A. & Grunberger, D. (1979) in Carcinogens: Iden- 944-946. tification and Mechanisms ofAction, eds. Griffin, A. C. & Shaw, 31. Arnold, D. L., Moodie, C. A., Stavic, B., Stoltz, D. R., Grice, C. R. (Raven, New York), pp. 399-418. H. C. & Munro, I. C. (1977) Science 197, 320. 5. Hiatt, H., Watson, J. D. & Winston, J. A. (1977) Origins of Hu- 32. Howe, G. R., Burch, J. D., Miller, A. B., Morrison, B., Gordon, man Cancer (Cold Spring Harbor Laboratory, Cold Spring Har- P., Weldon, L., Chambers, L. W., Fodor, G. & Winsor, G. M. bor, NY), Vol. 4. (1977) Lancet ii, 578-581. 6. Lee, L. S. & Weinstein, I. B. (1978) Science 202, 313-315. 33. Miller, A. B. & Howe, G. R. (1977) Lancet ii, 1221-1222. 7. Lee, L. S. & Weinstein, I. B. (1979) Proc. Natl Acad. Sci. USA 34. Simon, D., Yen, S. & Cole, P. (1975) J. Natl Cancer Inst. 54, 76, 5168-5172. 587-591. 8. Cohen, S. (1965) Dev. Biol 12, 394-407. 35. Clark, J. P. (1978) J. Am. Med. Assoc. 240, 349-355. 9. Savage, C. R., Jr. & Cohen, S. (1972) J. Biol Chem. 247, 36. Morgan, R. W. & Jain, M. G. (1974) Cancer Med. Assoc. J. 111, 7609-7611. 1967. 10. Carpenter, G. & Cohen, S. (1979) Annu. Rev. Biochem. 48, 37. Wynder, E. L. & Stellman, S. D. (1980) Science 207, 1214-1216. 193-216. 38. Kessler, I. I. (1976)J. Urol 115, 143-146. 11. Carpenter, G. & Cohen, S. (1976)J. Cell Biol 71, 159-171. 39. Mondal, S., Brankow, D. W. & Heidelberger, C. (1978) Science 12. Stastny, M. & Cohen, S. (1970) Biochim. Biophys. Acta. 204, 201, 1141-1142. 578-589. 40. Cohen, S. M., Arai, M., Jacobs, J. B. & Friedel, G. M. (1979) 13. Hooker, J. K. & Cohen, S. (1967) Biochim. Biophys. Acta 138, Cancer Res. 39, 1207-1217. 347-356. 41. Weinstein, I. B., Orenstein, J. M., Gebert, R., Kaighn, M. E. 14. Barnes, D. & Colowick, S. P. (1976)J. Cell Physiol 89, 633-640. & Stadler, U. C. (1975) Cancer Res. 35, 253-263. 15. Levine, L. & Hassid, D. F. (1977) Biochim. Biophys. Res. Com- 42. Montesano, R., Drevon, C., Kuroki, T., Saint Vincent, L., Hen- mun. 76, 1181-1187. dleman, S., Sanford, K. K., DeFeo, D. & Weinstein, I. B. (1977) 16. Rose, S., Stahn, R., Passovoy, D. S. & Herschman, H. (1976) J. Natl. Cancer Inst. 59, 1651-1658. Experientia 32, 913-915. 43. Delclos, K. B., Nagle, D. S. & Blumberg, P. M. (1980) Cell 19, 17. Frantz, C. N., Stiles, C. D. & Scher, C. D. (1979) Proc. Am. Assoc. 1025-1032. Cancer Res. 20, 207 (abstr.). Downloaded by guest on October 2, 2021