Dehydroepiandrosterone Sulfotransferase As a Possible Shunt for the Control of Steroid Metabolism in Human Mammary Carcinoma1

[CANCER RESEARCH 37, 278-284, January 1977] Dehydroepiandrosterone Sulfotransferase as a Possible Shunt for the Control of Steroid Metabolism in Human Mammary Carcinoma1

J_ B. Adams2 and D. P. Chandra2 ImperialCancerResearchFund,Lincoln's Inn Fields,London,England(J. B. A.), and the Schoolof Biochemistry,Universityof NewSouth Wales, Kensington,AustraliafJ. B. A., D. P. C.)

SUMMARY noid moiety can occur with the sulfate group intact, but such pathways appear to be of minor significance (21). Human primary mammary tumors were examined to de Oao and Libby (15, 16) have demonstrated that the sulfur termine what factors were of importance in deciding relative ylation of steroids by human mammary carcinoma tissue is rates of sulfurylation of dehydroepiandrostenone and 17$- highly correlated with prognosis and response to hormone estradiol, such rates having been shown to correlate with ablative procedures such as adnenalectomy. In patients un the patient's prognosis and response to adrenalectomy (T. dengoing adnenalectomy for late-stage breast cancer, the L. Oao and P. A. Libby. Enzymic Synthesis of Steroid Sul failure of tumor preparations to sulfurylate steroid hon fate by Mammary Cancer and Its Clinical Implications. NatI. mones in vitro was related to a particularly grave prognosis Cancer Inst. Monographs, 34: 205-210, 1971). The sulfur and a complete lack of response to adrenalectomy. When ylation of dehydroepiandrosterone and 17f3-estradiol was DHEA3 and 17f3-estradiol were compared as substrates for studied in 41 tumors in vitro using tumor cytosol, adenosine tumor sulfotransferases, then if the ratio of DHEA sulfate tniphosphate, [35S]SO42, Mg2@,and added steroid. Six tu formation to 17f3-estnadiol sulfate formation was <1, this mors showed no sulfurylating ability, 9 sulfurylated dehy was also associated with a poor prognosis and response to droepiandrosterone at a rate greater than that for 17$- adrenalectomy. The group with the best response and prog estradiol (ratio, >1), and 26 sulfunylated dehydroepiandnos nosis showed DHEA sulfate:17f3-estnadiol sulfate ratios of tenone at a rate lower than that for 17f3-estradiol (ratio, <1). >1 . In studies on primary tumors examined at the time of Evidence was obtained that low levels of dehydnoepiandros mastectomy, these ratios were also correlated with prog terone sulfotransfenase were responsible for ratios of <1 , in nosis, as witnessed by the presence of metastases or by many instances. Adenosine 3'-phosphate 5'-phosphosul early recurrence (16). fate synthesis and steroid sulfotransfenase activities were Lowered ratios of DHEA sulfate to 17f3-estnadiol sulfate measured in 30 tumors. A significant correlation was found could be due to either decreased DHEA sulfurylation or between synthesis of the former and levels of estrogen increased 17f3-estradiol sulfurylation. In this investigation, sulfotransferase, but this relationship did not hold for dehy human mammary tumors were examined for their overall droepiandrosterone sulfotnansferase, again due to low 1ev ability to form steroid sulfates from S042_and, in an associ els of this enzyme in many tumors. It is suggested that ated study, for their ability to form the intermediate PAPS dehydnoepiandrosterone sulfate formation in the tumors is and catalyze subsequent transfer of the sulfate group from mainly controlled by the sulfotnansfenase, which acts as a PAPS to added steroids. In this manner, the limiting steps in shunt in regulating the level of free dehydroepiandroste steroid sulfunylation were assessed. none, and related compounds, available for metabolism to steroids influencing the growth of mammary epithelial cells. MATERIALS AND METHODS

INTRODUCTION TumorTissue

Human mammary carcinoma tissue can sulfurylate ste Specimens of primary carcinoma tissue from patients roids (3) and convert steroids into physiologically active undergoing mastectomy were received from the operating metabolites (7). Of particular significance is the ability of theater. They were placed in plastic containers packed in some breast tumors to produce estrogens (1, 5, 7, 23). The ice, transported to the laboratory, and stored at â€”20Â°.The significance of the sulfunylation process in steroid hormone time between receiving the specimen and freezing it did not

-@ metabolism remains unclanified . Transformations of the ste exceed 1 hr in any instance. Subsequently, the tissue was allowed to thaw, fat and connective tissue were removed, I Supported by grants from the New South Wales Cancer Council and the Australian National Health and Medical Research Council.

2 Permanent address: School of Biochemistry, University of New South 3 The abbreviations used are: DHEA, dehydroepiandrosterone; PAPS, Wales, P. 0. Box 1, Kensington, N. S. W. 2033, Australia. adenosine 3'-phosphate 5'-phosphosulfate; APS, adenosine 5'-phosphosul Received May 26, 1976; accepted September 1, 1976. fate.

278 CANCER RESEARCH VOL. 37

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1977 American Association for Cancer Research. Steroid Sulfotransferases and Breast Cancer and a weighed sample (0.2 to 1.0 g) was cut into small room. After drying, the strips were scanned in a Nuclear pieces and frozen in the barrel of a stainless steel chamber Chicago Actigraph Ill instrument. The [35S]PAPS and prechilled in dry ice. The frozen tissue was shattered by [35SJAPSpeaks were identified by comparison with authen hammering a closely fitting stainless steel plunger into the ticcompounds. Areasofthesepeakswere measured,and chamber. (Details of this piece of equipment can be ob the activity of the PAPS-synthesizing system was expressed tamed from the authors.) The resulting fine powder was in arbitrary units defined as the sum of the areas of PAPS transferred to a small glass homogenizer, 4 volumes of 0.01 plus APS per mg of cytosol protein used in the incubation. Msodiumphosphatebuffer(pH7.4)in0.9%NaCIsolution Assay of Acid MucopolysaccharideSulfotransferase. were added, and the mixture was homogenized using a The incubation mixture was identical to that for the control Teflon plunger. The cytosol fraction was collected after in Paragraph 1. After 2 hr of incubation at 37Â°,10ml of 0.2 M centnifugation at 100,000 x g for 1 hr. Na2SO4and 1 ml of 1% (w/v) aqueous solution of sodium In all cases histological diagnosis of carcinoma was con chondroitin sulfate (Sigma) were added. After mixing, 1 ml firmed from block sections. of 5% (w/v) aqueous solution of cetyl tnimethylammonium bromide (Hopkin and Williams Ltd., Essex, England) was Chemicals added, and the precipitate was allowed to stand for 1 hr on overnight if floculation was delayed. The precipitate was DHEA, DHEA sulfate, 17f3-estradiol, and APS were ob centrifuged down and washed with 5 ml of 0.1 M Na2SO4. tamed from Sigma Chemical Co., St. Louis, Mo. Cholesterol This was repeated 4 times, followed by 1 washing with was obtained from the British Drug Houses Ltd., London, water. Finally, the precipitate was centrifuged, the superna England. Estrone 3-sulfate was purchased from Ikapharm, tant was discarded, and the tube was drained by inversion Ramat-Gan, Israel. 17f3-Estradiol 3-sulfate was prepared on tissue paper. The precipitate was dissolved in 2 ml of from this by NaBH4 treatment. Cholesteryl 3-sulfate was 10% (w/v) NaCI and sodium salts of acid mucopolysaccha prepared by the method of Joseph et al. (21). [35S]Sulfate, rides precipitated by addition of 1.5 volumes of ethanol. carrier-free, was purchased from the Radiochemical After standing for 2 hr at 0Â°,the material was centrifuged Centre, Amensham, England. and the process of dissolution and precipitation was re peated. The material was dissolved in 1 ml of water, and 0.2 Protein ml was removed for counting in 10 ml of detergent-phos phor (see Paragraph 1). Blanks carried out with bovine This was measured by the method of Lowry et a!. (22). serum albumin in place of tumor cytosol were zero. The identification of the radioactive fraction isolated by this Conditionsof Incubationand AssayProcedures procedure as acid mucopolysacchanide was demonstrated as described previously (2). 1. Assay of Steroid Sulfotransferases.The incubation mixture contained the following ingredients in a total vol Examinationof Ester [35SJSuIfatesFormedfrom Endoge ume of 0.2 ml: 1 @tmoIeATP;1.5 @molesMgCI2;0.01 mmole nousAcceptors Tnis-HCI, pH 7.4; 0.04 @moleNa2SO4;25 pCi carrier-free [35S]SO42; 0.1 ml cytosol fraction of tumor; and 0.01 @moIe Incubations of cytosol fractions of a number of tumors, in steroid added in 0.01 ml propylene glycol. Controls con the absence of added steroid, were carried out as in Para tamed o.oi ml propylene glycol. Incubation was for 2 hr graph 1, but on 3 times the scale and in the presence of a at 37Â°.The steroid [35S]sulfate formed was assayed by total of 80 pCi of [35Sjsulfate. Reaction was stopped by extraction into ethyl acetate (14). Briefly, 2 ml of 2 NNA4OH, addition of 1.5 volumes of ethanol and protein was removed saturated with (NH4)2SO4,were added, and the mixture was by centnifugation. The supennatant was then chromato extracted with ethyl acetate (3 x 3 ml). The combined ex graphed on paper. tracts were adjusted to 10 ml and back-extracted once with 0.5 ml of seven-eighths-saturated (NH4)2S04.An aliquot (3 PaperChromatographySystems ml) was evaporated in a vial, 10 ml of detergent-phosphor were added (3), and samples were counted by liquid scintil System A: ethanol:1 M ammonium acetate, pH 7.5 (7.5:3). lation. Sufficient counts were recorded to allow for an accu System B: 0.4 M sodium phosphate, pH 6.3. System C: racy of 3%. After counts were subtracted in the control (all diisopnopyl ethen:tert-butyl alcohol:concentnated ammo ingredients but no steroid), the net counts were expressed nia:waten (6:4:1:9). Steroid sulfates on chromatograms as pmoles of steroid sulfate per mg protein per 2 hn, from a were cleaved to the free steroids by exposure to HCI vapor knowledge of the specific activity of the [35S]sulfate. for 2 hr. Estrogens were then detected by Turnbull's re 2. Assay of [355]PAPS. The incubation mixture was iden agent, and other steroids were detected by phosphotungs tical to that for the control in Paragraph 1, with the excep tic acid. tion that an incubation time of 30 mm was used. At the end of that time, a 0.01-mI aliquot was added to 0.05 ml of ethanol, and the solution was applied to a strip of Whatman RESULTS No. 1 paper and subjected to electnophonesis for 3 hr in 0.05 Msodiumphosphatebuffer,pH6.0.Apotentialof10V/cm Assessmentof Method Used for SteroidSulfotransfer wasappliedandthe procedurewascarriedout in the cold ase Assay. Extractsfroma numberof tissues,whenincu

JANUARY1977 279

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1977 American Association for Cancer Research. J. B. Adams and D. P. Chandra bated with ATP, Mg2@,and [@SJsuIfate of high specific number of counts in the ethyl acetate extract or alter the activity, catalyze [35Sjsulfate activation to [35S]PAPS and pattern of zones obtained on chromatography (Chart 2A). subsequent transfer of the sulfate group to endogenous Addition of 17/3-estradiol, however, increased the counts in acceptors present in the extracts (28, 29). When exogenous the ethyl acetate and resulted in the appearance of a new acceptors, e.g., steroids, are added to such incubation zone in the position of authentic 17f3-estradiol 3-sulfate on mixtures, then a new zone may appear on autonadiographs the chromatognam. The other zones remained unchanged. of chromatograms, indicating formation of exogenous ac Results obtained with another tumor are shown in Chart 2B. ceptor sulfate. The presence of steroid sulfotnansfenase in Increased counts upon addition of either DHEA or 17/3- human mammary carcinoma tissue was detected by this estradiol were recorded, and these were associated with method (3). new peaks on chromatograms corresponding to the appro Formation of total ester [@S]suIfates, extractable into pniate sulfate esters of the steroids. Again, the size of the ethyl acetate, was linear with time for approximately 1 hr other peaks was unchanged upon addition of steroid (Chart (Chart 1). When the ethyl acetate extracts, obtained in the 2B). normal assay procedure, were chromatographed in System When increasing concentrations of DHEA were added, C and scanned for radioactive components, the results the size of the DHEA [35S]sulfate peak on chromatograms shown in Chart 2 were obtained. In the control, without increased up to a concentration of 0.05 mM DHEA and added steroid, 3 peaks were present. For the 1st tumor decreased at higher concentrations. The size of the other cytosol depicted, addition of DHEA did not increase the peaks remained approximately constant at all concentra tions of DHEA. Measurement of DHEA sulfate formation was compared by 2 methods as shown in Chart 3. With increas ing concentrations of DHEA, net counts, i.e., total counts w less counts in the control, gave a curve similar to that U- obtained by measurement of DHEA sulfate peaks on chro -J C,, matograms. Other tumors were examined similarly using

LU DHEA and 17f3-estradiol as substrates, and it was observed C,,Iâ€” LU that net counts correlated well with the size of the respec C.,, tive steroid sulfate peaks. Thus, for routine estimations, it

0@ 20 was justifiable to use net counts to determine steroid sulfo transfenase activities. By contrast to the lack of effect of DHEA and 17f3-estradiol on the formation of [35S]sulfates from endogenous acceptors, addition of p-nitrophenol ne 60 TIME MIN suIted in p-nitrophenyl [35S]sulfate formation and the disap pearance of ester [35S]sulfates derived from endogenous Chart I . Time course for formation of total ester [3S]sulfates. Incubation was carried out with tumor cytosol fraction in the presence of ATP, acceptors. [@S]SO42_, Mg2@, and DHEA (0.05 mM) as described under â€œMaterials and Nature of the EndogenousSulfateAcceptors.BostrOm Methods,â€•Paragraph 1. (9) demonstrated that some of the ester sulfates present in urineprobablyrepresentedaryland steroidsulfates.This COUNTS was based on their chromatographic properties in 2-dimen INETHYL ACETATE sional systems. Endogenous steroids present in tissue ex ,[\N@r?\COUTROL@[email protected] tracts, which have been characterized as their sulfate es tens, following incubation of the extracts with ATP, Mg2@, and [35S]sulfate, are cholesterol (8) and estrone (4). Four ,fI@[\+D@A181@81OU primary human mammary tumors of varying steroid sulfur ylating ability were selected for experiments aimed at eluci dating the nature of the endogenous sulfate acceptors. Incubation of cytosol fractions was carried out as for the f@J@17s@@ETRAOUTTd5O

â€¢ORIGINâ€¢DHEASULFATE. ORIGINâ€¢DHEASULFATE . 17s-ESTRADIOL-3-SULFATEâ€¢

Chart 2. Chromatography (System C) and scanning of the labeled compo nents present in ethyl acetate extracts of tumor cytosols incubated with ATP, MgaF,and[US]SO42_.Peakspresentinthecontrolsrepresentsulfateesters formed from endogenous acceptors. In A , addition of DHEA did not increase the total counts in the ethyl acetate and the pattern of peaks is unchanged. Addition of 17@-estradiol, however, increased the total counts and a new DHEAMM peak, in the position of authentic 17@-estradiol 3-sulfate, is evident. In B, Chart 3. Effect of increasing concentrations of DHEA on formation of using a separate tumor cytosol, addition of both DHEA and 17p-estradiol has DHEA sulfate as measured by net counts (obtained after subtraction of resulted in increased counts and the appearance of new zones in positions counts in control) converted to pmoles DHEA sulfate based on the specific corresponding to the authentic sulfate esters. Shaded areas, positions of activity of the [@SJSO42_(â€¢)andfrom measured areas under the appropriate authentic marker compounds run in parallel. peaks after chromatography, as in Chart 2 (0).

280 CANCER RESEARCH VOL. 37

routine steroid sulfotransferase assay procedure, but on a exogenous substrate. Estrogen sulfotnansfenase is specific larger scale and with increased amounts of [35S]sulfate. for estrogens and will not sulfurylate simple phenols, al Preliminary chromatography of the protein-free incubation though phenols possessing alkyl side chains are sulfur mixture was carried out in System A and the broad ester ylatedatlow rates(6,27). sulfate band eluted. Part of this eluate was chromato It can be seen from Table 2 that the isolation of choles graphed in System B, designed to separate the ester sul teryl [35Sjsulfate from tumors of Subjects 0. C. and M. M. is fates of the 3 classical estrogens (26). No peaks were pres related to the presence of relatively high DHEA sulfotrans ent in the positions corresponding to the 3 estrogen sulfates fenase activities. This suggests that both DHEA and choles or to DHEA sulfate. All 4 tumors showed 2 radioactive peaks terol are sulfurylated by the same enzyme. Cholesteryl with high AF values in System B. In addition, 2 of the [35S]sulfate was detected only when the specific activity of tumors (Subjects 0. C. and M. M.) showed a peak at the the [35S]sulfate was increased by addition of increased origin. In System C, this particular peak in both instances amounts of carrier-free [US]sulfate. Additional experiments exhibited a high AF and moved to a position identical to carried out with tumor tissue from Subject E. 0. , which that of cholesteryl sulfate, used as a marker. These nadioac possessed high estrogen sulfotransferase activity (Table 2), tive areas corresponding to cholesteryl sulfate represented failed to detect sulfate esters of the 3 classical estrogens about one-half the total peak areas and were shown to be among the labeled ester sulfates derived from endogenous identical to this substance by means of reverse isotope acceptors. dilution(Table1). Steroidand Acid MucopolysaccharideSulfotransferase The other radioactive components, common to all 4 tu Activities in Primary Tumors. A summaryof the results mors, possessed chromatographic properties in System B, obtained with 41 primary carcinomas is shown in Table 3. which showed that they were more polar than were steroid Generally, the overall transfer of sulfate to steroid was sulfates. As mentioned earlier, such components, present higher than that reported by Oao and Libby (15). Undoubt in the normal incubation system used for assaying steroid edly, the higher concentration of S042_ used in the present sulfotransferase activity, were not formed when exogenous procedure is responsible. In addition, the presence of p-nitnophenol was sulfurylated. This indicated that the en higher concentrations of PAPS-degrading enzymes in the dogenous acceptors were probably phenolic, mass action particulate fraction of the 12,000 x g supennatant, as used effects dictating that endogenous phenols would not be by Dao and Libby (15), would limit overall sulfation. sulfurylated at detectable rates in the presence of excess Six ofthe 41 tumors (15%) showed no sulfurylating ability, 9 sulfurylated DHEA at a greater rate than did 17/3-estnadiol (ratio, >1), and 26 sulfunylated DHEA at a lower rate than @ Table 1 did 17f3-estradiol (ratio, <1). Crystallization data for the characterization of cholesteryl Correlationof EnzymeActivities.Whenendogenousac [355]sulfate ceptor sulfurylation was plotted in turn against DHEA, 17/3- Using [35S]SO42_ of high specific activity in incubations with estradiol (Chart 4), and acid mucopolysacchanide sulfuryla tumor cytosols (see â€œMaterialsand Methodsâ€•), a labeled compo nent with chromatographic properties corresponding to those of tion (Chart 5), a correlation was established with the steroid cholesteryl sulfate was formed in Subjects D. C. and M. M. (see sulfotnansfenases but not with acid mucopolysacchanide Table 2). These areas of the chromatogramswere eluted, mixed sulfotnansferase. The 6 tumors that failed to demonstrate with 34 mg of cholesteryl sulfate, and crystallized from 70% overall sulfurylation are not included in the data presented ethanol. inChants4 and 5. Subject D. C. M. M. Relationshipbetween [35S]PAPSSynthesisand Sulfo (cpm/mg)Original12708751 (cpm/mg)Subject transferase Activities. Plots of endogenous acceptor [35S]sulfate formation versus [35S]PAPSformation in 30 tu stcrystallization8977632nd crystallization7606393rd mors showed a significant correlation (Chart 34). A signifi crystallization800670lstmotherliquor8007482nd cant correlation was also obtained when estrogen sulfate formation was plotted against PAPS synthesis (Chart 6B), liquor8086723rdmother but this was not observed for DHEA sulfate (Chart 6C). The mother liquor768710 sum ofAPS plusPAPS was used ineach instance,sinceit

Table 2 Sulfotransferase activities ofendogenousof tumor cytosols used in investigation of the nature acceptorsSulfotransferasesulfate activities (pmoles/mg protein/2hr)Net 17$- Acid mu No ste DHEA Net DHEA 17/3-Estra- estradiol copolysac SubjectroidaddedchanideB.S.4341 sulfate diol added sulfate 43D. 0 63 20 84E.O.3454C.561 73 117 115 59 98M.M59125 20 432 398 66 172 113 116

JANUARY1977 281

was judged that such a sum would be a better indication of 120' A the â€œPAPSpotential.â€•A high percentage of the tumors contained a 3'-nucleotidase that converted PAPS formed in @80, situ, or introduced into the system, into APS. Degradation @ of PAPS, and the subcellulan distribution of the degradative 0.002) enzymes in human mammary tumors, will be the subject of a separate communication. When PAPS values alone were plotted against sulfotnansferase activities, similar correla 160 tions to those found for the sum of APS and PAPS were obtained (see legend to Chart 6). â€˜j.'120

@ an @ DISCUSSION 0.02) Estimations of the overall steroid sul'funylating capacity in @i3 such tissues as human liven were first made by BostrOm and

TabIe3 100 DHEA,1713-estradiolandacid mucopolysaccharide sulfotransferasesin 41 human primary breast carcinomas transferred 60 hr)Mean(pmoles/mg protein/2 r=0.27 NS

@ 40 RangeEndogenousSulfotransferaseSulfate Â±S.D. 2-132DHEA44acceptor53 Â± 37 20 0-3741 Â± 70 0-486Acid7$-Estradiol80 Â±105 100 200 300 iOU 500 600 0.5-122chanidemucopolysac 25 Â± 33 UPS+ PAPS(ARBITRARYUNITS)

Chart 6. Correlations between PAPS synthesis and sulfotransferase activi ties in 30 primary breast carcinoma cytosols. Some of the PAPS formed Is 1501 degraded to APS by a 3'-nucleotidase present in the cytosol. For this reason the sum of APS plus PAPS has been used. Similar correlations to those shown were obtained when PAPS alone was used. These were (A) r = 0.54, p @ @OU1 .. â€¢...... @._â€”_O.52 (P'O.Ol) < 0.002; (B)r = 0.39, p < 0.05; (C)r = 0.1 1,p > 0.25. NS, not significant (p < 0.2). kOUr@n

1@U@i@ 2@0 2'SO 3W4@0 Wengle (10). For example, the amount of DHEA sulfate 17'-ESTRADIOL-3-SULFATEROLES formed using cytosol preparations derived from human fe 150 male liver was 385 Â±35nmoles peng wet weight per hr. The amount of ester [35Sjsulfate formed, in parallel incubations without added acceptor, was 20 to 30 nmoles per g wet @@@ 50 (P'0.05) weight per hr. When human primary breast carcinoma tis sues were assayed in the present study, the amounts of .,â€˜ S steroid sulfates formed were much lower than the figures @O 1&T 150 200â€•TO) iiÃ´o DHEASULFATEPMOLES quoted for human liver, and the amounts of ester Chart 4. DHEA sulfotransferase and 17@9-estradiolsulfotransferase activi [35S]sulfates formed in controls without added steroid rep ties versus endogenous acceptor sulfurylation in 35 primary breast carci resented a significant part of total ester [35S]sulfates pro noma cytosols. duced. 150 The steroid sulfotnansferase activities summarized in Ta LU @-. ble 3 are somewhat higher than the values reported for .-JC,, C primary breast tumors by Dao and Libby (14) and Brauns @ 100 berg et al. (11). This is probably due to the quite low con s-v)Q S S 0- LU centrations of sulfate used by these workers4 and could also LU -J L) explain why endogenous acceptor sulfunylation was not @ 50 55 mentioned in these reports. Unfrozen tissue was also used @ 1â€¢â€¢@@ 5 by these workers, but it has been found that small samples LU â€˜p of fibrous breast tumors defy homogenization by normal @ .@S IS I 1 methods. The technique of powdering the tissue sample in 50 100 150 the supencooled state proved very effective in these cases ACIDMIJCOPOLYSACCHARIDESULFATE

PMOLES 4 A discrepancy appears in the concentration of sulfate (0.01 mM) as Chart 5. Acid mucopolysaccharide sulfotransferase activity versus endog described in the original paper of Dao and Libby (14) and in details given enous acceptor sulfurylation in the 35 primary breast carcinoma cytosols subsequently (16) in which the concentration is quoted as 0.2 mM. The used in Chart 4. concentration of sulfate used in the present work was 0.2 mM.

282 CANCERRESEARCHVOL. 37

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1977 American Association for Cancer Research. Steroid Sulfotransferases and Breast Cancer and was adopted as a general procedure. This, in turn, may nous acceptor sulfate was less than 10 pmoles per mg account for the lower percentage of tumors that failed to protein pen2 hr and no marked change in total ester sulfate show overall sulfunylation in this work, as compared to formation occurred in these tumor extracts in the presence those of Dao and Libby (14) and Braunsberg et a!. (11) (see of added steroids. Blanks carried out using bovine serum below). Also, the cytosol fraction was used, rather than the albumin, in place of cytosol, gave values of <1 pmole. The 12,000 x g supennatant as used by Dao and Libby (14), since fact that acid mucopolysacchanide sulfotnansfenase activi the sulfotransferases and PAPS-synthesizing enzymes are ties were also generally low in these cases eliminated the present in this cell fraction and the higher concentrations of possibility that high levels of this enzyme, by utilization of PAPS-degradative enzymes, present in the microsomes, are available PAPS, may have been responsible for the lack of avoided. Freezing of the cytosol fraction for 2 weeks did not demonstrable steroid sulfotransfenase activity. This cate alter steroid sulfotransferase activities appreciably; thus gory comprised 15% of the tumors studied, a value consid freezing the tissue would be unlikely to lead to a differential erably lower than the 32% of 92 primary tumors reported by loss of one of these enzymes and thus alter the ratio of Dao and Libby (15) on 49% of 41 primary tumors reported in DHEA sulfate to 17f3-estradiol sulfate. This is borne out by the series of Braunsbeng et al. (11). As mentioned previ the fact that the same distribution of activities was found as ously, of those tumors that did show steroid sulfunylating was reported by Dao and Libby (15), namely, DHEA sul ability,DHEAsulfate:17/3-estradiolsulfateratiosof>1 oc fate:17f3-estradiolsulfate,>1inone-thirdand<1 intwo curred in 26% of cases and ratios of <1 occurred in 74% of thirds of the tumors. cases. These figures were in excellent agreement with those The endogenous acceptors, detected in the tissue ex reported by Dao and Libby (15), namely, 24 and 76%, ne tracts under the conditions used in the routine assays, spectively. appeared to be phenolic. They were not substrates for es Very few subjects in this series have shown recurrence, trogen sulfotransferase, since they were still formed in the so that at this stage it is not possible to attempt to relate presence of exogenous estrogen (Chart 2) but were elimi steroid sulfotransferase ratios with prognosis. In Dao and nated upon addition of p-nitnophenol. It was previously Libby's series of 92 primary carcinomas, some 40 cases that shown that an endogenous accepton-[35Slsulfate, formed showed recurrence within 3 years of mastectomy all fell into upon incubation of a breast tumor extract with [35SJPAPSof the group exhibiting either no sulfunylation or that in which high specific activity and having an RF identical to that of the DHEA sulfate:17/3-estradiol sulfate ratio was <1 (16). If cholesteryl sulfate, was specifically eliminated when addi we momentarily leave the question of the possible signifi tion of DHEA was made to the incubations (3). Later, cho cance of a complete lack of steroid sulfurylation as being lestenol was formally identified as an endogenous sulfate associated with a poor prognosis, the other group having a acceptor in breast tumors incubated under similar condi poor prognosis was, as stated, that in which the DHEA tions (8). The identification of cholesteryl [35S]sulfate in 2 sulfate:17/3-estnadiolsulfateratiowas<1 .@Thiscan be due tumors, possessing relatively high levels of DHEA sulfo to either depressed DHEA sulfate formation or elevated 17/3- transfenase (Table 2), would then support the previous data estradiol sulfate formation. Comparison of such activities which suggested that both these steroids are sulfunylated by with a 3rd sulfotnansferase, preferably not a steroid alcohol the same enzyme. Although estrogens were not detected sulfotransferase due to lack of knowledge on their substrate among the endogenous acceptors, even in 1 tumor pos specificities, might then suggest an answer to these alterna sessing a very high estrogen sulfotransferase activity (Table tives. Endogenous acceptor sulfurylation should serve this 2), steroids other than cholesterol have been detected. A purpose. From Chart 4 it can be seen that both 17f3-estra primary breast carcinoma extract, upon incubation with diol and DHEA sulfurylation correlate with the sulfunylation [35S]PAPSof high specific activity, yielded a zone travelling of endogenous acceptor. However, for DHEA sulfotransfen in the area occupied by steroid sulfates in the 2-dimensional ase, the scatter diagram reveals a bunching of points near chromatographic system developed by Vestermark and the ordinate. This is not evident for 17f3-estnadiol sulfotnans BostrOm (29). Although this zone was not in the position ferase; thus the bunching of the points cannot be due to occupied by authentic 17f3-estradiol sulfate on DHEA sul lack of available PAPS. Rather, a diminished level of DHEA fate, it alone among the numerous labeled ester sulfates sulfotnansferase is suggested in these cases. This conclu present on the chromatograms was increased upon addi sion is supported by the data in Chart 6, which show that the tion of either 17/3-estradiol on DHEA to the incubations (3). significant correlation between PAPS synthesis and 17/3- At the time no explanation could be offered for this phe estradiol sulfate formation does not hold in the parallel case nomenon, but it is significant that the [35S]PAPS used had involving DHEA sulfate formation, due evidently to low such a high specific activity that only minute chemical DHEA sulfotransferase activities in many of the tumors. amounts of steroid [35S]sulfates were produced. At these Although the ability to form PAPS is of obvious importance levels the chromatognaphic properties of the steroid in the overall synthesis of 17/3-estradiol sulfate (Chart 6B), [35Sjsulfates were influenced by charged groups on the in this instance also the level of estrogen sulfotnansfenase paper, and they travelled to different positions as compared with authentic unlabeled compounds. Ovenspotting with I One other explanation for the lack of sulfurylation, or altered steroid excess authentic steroid sulfate overcame this discrepancy sulfotransferase activities; namely, one which possibly involved steroid sulfa (4). tases, would appear to be ruled out. These enzymes have been studied In human mammary carcinomas and their levels were found to be unrelated to Tumor specimens were judged as lacking overall steroid ratios of the concentrations of DHEA and 17@-estradioI sulfates formed in sulfotransferase activities when the formation of endoge incubations (13).

JANUARY1977 283

plays a role, as seen from the wider scatter of points when 2. Adams, J. B. The Biosynthesis of Chondroitin Sulphate. Incorporation of Sulphate into Chondroitin Sulphate in Embryonic Chick Cartilage. Bio comparison is made with endogenous acceptor sulfuryla chem. J., 76; 520-533, 1960. tion (Chart 6A). 3. Adams, J. B. Enzymic Synthesis of Steroid Sulphates. II. Presence of These results implicate a lowered DHEA sulfotransferase Steroid Sulphokinase in Human Mammary Carcinoma Extracts. J. Clin. Endocrinol. Metab., 24: 988-996, 1964. as being responsible for depressed ratios of DHEA sulfate to 4. Adams, J. B., and Ellyard, A. K. Endogenous Steroid Sulphation on 17/3-estradiol sulfate in many cases. Such a conclusion is Incubation of Tissue Extracts with [â€œS]Sulphate.Biochim. Biophys. not in agreement with that reached by Dao and Libby (14), Acta, 248: 520â€”523,1971. 5. Adams, J. B., and Li, K. Biosynthesis of 17@3-OestradioIinHuman Breast who favor an increased 17/3-estradiol sulfotransferase to Carcinoma Tissue and a Novel Method for Its Characterization. Brit. J. explain such ratios. It is difficult to reconcile an elevated Cancer, 31: 429-433, 1975.. 6. Adams, J. B., and Low, J. Enzymic Synthesis of Steroid Sulphates. X. sulfotnansferase, namely, 17f3-estradiol sulfotransfenase, on Isolation of Oestrogen Sulphotransferase from Bovine Placenta and one hand and the lack of sulfotransferase activity on the Comparison of Its Properties with Adrenal Oestrogen Sulphotransferase. other as being 2 parameters correlated with a poor prog Biochim. Biophys. Acta, 370: 189â€”196,1974. 7. Adams, J. B., and Wong, M. S. F. Paraendocrine Behaviour of Human nosis. Lowered or immeasurable levels of sulfotransferase, Breast Carcinoma: In Vitro Transformation of Steroids to Physiologically in this instance DHEA sulfotransferase, are reconcilable, Active Hormones. J. Endocrinol., 41: 41-52, 1968. however. 8. Adams, J. B., and Wong, M. S. F. Enzymic Synthesis of Steroid Sul phates. vI. Formation of Cholesteryl@uS@Sulphateon Incubation of Hu As mentioned previously, steroidogenic pathways involv man Breast Carcinoma Extracts with Adenosine-3'-Phosphate-5'-Phos ing steroid sulfates as obligatory intermediates are usually pho-â€•S-Sulphate.Steroids, 11: 313â€”319,1968. 9. BostrOm, H. Studies on Ester Sulphates. II. The Hormonal Influence on of minor significance (21). The sulfurylation step could then the Urinary Ester-Sulphate Pattern in the Rat. Acta Endocrinol., 37: 405â€” serve as a control shunt to limit the availability of free 417,1961. steroids, in particular 3f3-hydnoxy-i@5-stenoids,for conver 10. BostrOm, H., and Wengle, B. Studies on Ester Sulphates. 19. On Sul phate Conjugation in Adult Human Liver Extracts. Acta Soc. Med. Upsa . sion to metabolites influencing growth or homeostasis of lien., 69: 41-63, 1964. mammary tissue. DHEA can serve as a precursor for and ros 11. Braunsberg, H., Killen, E., and Melville, E. Studies on Steroid Sulphation tenedione and testosterone in mammary carcinoma tissue and Binding by Human Malignant Tumours. European J. Cancer, 10: 13- 20, 1974. (7, 19), and these in turn can be converted to estrogens (see 12. Buirchell, B. J. , and HShnel, R. Metabolism of Estradiol-173 in Human â€œIntroductionâ€•)andto 5a-reduction products (5, 24). In the Endometrium during the Menstrual Cycle. J. Steroid Biochem., 6: 1489- 1494, 1975. rat, 5a-dihydrotestostenone is known to inhibit growth of 13. Dao, T. L., Hayes, C., and Libby, P. R. Steroid Sulfatase Activities in mammary tumors but to induce growth of normal mammary Human Breast Tumors. Proc. Soc. Exptl. Biol. Med., 146: 381-384, 1974. epithelium (18). 5a-Andnostane-3a,17f3-diol is an active an 14. Dao, T. L., and Libby, P. A. Conjugation of Steroid Hormones by Normal and Neoplastic Tissues. J. Clin. Endocrinol., Metab. 28: 1431-1439, drogen which is capable of maintaining prostatic weight in 1968. castrated rats (17). 15. Dao, T. L., and Libby, P. A. Enzymic Synthesis of Steroid Sulphate by Estrogen sulfotransferase could also function in the con Mammary Cancer and Its Clinical Implications. NatI. Cancer Inst. Mono graph, 34: 205-210, 1971. trol of the concentration of free, i.e., physiologically active, 16. Dao, T. L., and Libby, P. A. Steroid Sulfate Formation in Human Breast estrogen. It has been shown that estrogen sulfotransferase, Tumors and Hormone Dependency. In: T. L. Dao (ed), Estrogen Target Tissues and Neoplasia, pp. 181-200. Chicago: Chicago University Press, in both porcine (25) and human (12) uteri, is under hor 1972. monal control and serves, in conjunction with 17f3-estradiol 17. Dorfman, A. I., and Shipley, R. A. Androgens, p. 116. New York: John dehydrogenase, to regulate the concentration of free 17/3- Wiley & Sons, Inc., 1956. 18. Huggins, C., and Mainzer, K. Hormonal Influences on Mammary Tu estradiol in the uterus during the estrus and menstrual mours of the Rat II. Retardation of Growth of the Transplanted Fibroad cycles. enoma in Intact Female Rats by Steroids in the Androstane Series. J. An imbalance in the levels of the 2 types of sulfotransfen Exptl. Med., 105: 485-501, 1957. 19. Jones, D., Cameron, E. H. D., Griffiths, K., Gleave, E. N., and Forrest, A. ases in the tumor, along the lines suggested, could disturb P. M. Steroid Metabolism by Human Breast Tumours. Biochem. J., 116: the hormone milieu and increase the growth rate of the 919-921, 1970. 20. Joseph, J. P., Dusza, J. P., and Bernstein, S. The Use of Sulphamic Acid tumor. The extremely poor prognosis associated in late for the Preparation of Steroid Sulphates. Steroids, 7: 577-587, 1966. stage breast cancer with those tumors that exhibit no sul 21. Lebeau, M. C., and Baulieu, E. E. On the Significance of the Metabolism furylating ability (16) is also compatible with an hypothesis of Steroid Hormone Conjugates. In: W. H. Fishman (ed), Metabolic Conjugation and Metabolic Hydrolysis, Vol. 3, pp. 151-187. New York: implicating steroid sulfurylation as a control shunt. Academic Press, Inc., 1973. 22. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. , and Randall, A. J. Protein Measurement with the Folin Phenol Reagent. J. Biol. Chem., 193: 265- ACKNOWLEDGMENTS 275, 1951. 23. Miller, W. A., and Forrest, A. P. M. Oestradiol Synthesis by a Human Dr. Adams is indebted to the Imperial Cancer Research Fund and in Breast Carcinoma. Lancet, 2: 866-868, 1974. particular to Dr. A. D. Bulbrook for the provision of all the facilities of his 24. Miller, W. A., Forrest, A. P. M., and Hamilton, T. Steroid Metabolism by Human Breast and Rat Mammary Carcinomata. Steroids, 23: 379â€”395, laboratory during a sabbatical year. The skilled technical assistance of Rose mary Knowles during this time is acknowledged with thanks. Thanks are also 1974. due to the Royal Marsden and St. Bartholomew's Hospitals and in particular 25. Pack, B. A., and Brooks, S. C. Cyclic Activity of Estrogen Sulfotransfer to John Hayward and the Breast Unit of Guy's Hospital, London, for provision ase in the Gilt Uterus. Endocrinology, 95: 1680-1690, 1974. of surgical specimens and for follow-up data on patients. The provision of 26. Payne, A. H., and Mason, M. The Enzymic Synthesis of the Sulphate surgical specimens by the Prince of Wales Hospital, Sydney, is also gratefully Esters of Estradiol-17f3 and Diethylstilbestrol. Biochim. Biophys. Acta, acknowledged. 71: 719-724,1963. 27. Rozhin, J., Soderstrom, R. L., and Brooks, S. C. Specificity Studies on Bovine Adrenal Estrogen Sulphotransferase. J. Biol. Chem. , 249: 2079- 2087, 1974. REFERENCES 28. Spencer, B. Endogenous Sulphate Acceptors in Rat Liver. Biochem. J., 77: 294-304, 1960. 1. Abul-Hajj, Y. J. Metabolism of Dehydroepiandrosterone by Hormone 29. Vestermark, A., and BostrOm, H. Studies on Ester Sulphates. 1. Applica Dependent and Hormone Independent Human Breast Carcinoma. Ste tion of Two-Dimensional Paper Chromatography in Studies on the Bio roids 26: 488-500, 1975. synthesis of Ester Sulphates. Acta Chem. Scand., 13: 827-830, 1959.

284 CANCER RESEARCH VOL. 37

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 1977 American Association for Cancer Research. Dehydroepiandrosterone Sulfotransferase as a Possible Shunt for the Control of Steroid Metabolism in Human Mammary Carcinoma

J. B. Adams and D. P. Chandra

Cancer Res 1977;37:278-284.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/37/1/278

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/37/1/278. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.