Suramin, hydrocortisone, and retinoic acid modify inhibitory effects of 1,25-dihydroxyvitamin D3 on prostatic epithelial cells

ELSEVIER

Suramin, Hydrocortisone, and Retinoic Acid Modify Inhibitory Effects of I ,25=Dihydroxyvitamin D, on Prostatic Epithelial Cells Donna M. Peehl, PhD, Stephen T. Wong, BS, Scott D. Cramer, PhD, Coleman Gross, MD, and David Feldman, MD Departments

of Urology

and Medicine,

Stanford

University

The proliferation of prostatic epithelial cells is regulated by the complex interplay of numerous growth-stimulatory and growth-inhibitory factors. I ,25-dihydroxyvitamin D, [I ,25(OH),D,] has recently been identified as a potent inhibitor of the growth of prostatic epithelial cells. Epidemiologic studies indicate that vitamin D deficiency may be a risk factor for the development

of clinical

prostate cancer, possibly due to increased growth and reduced differentiation of prostatic cells in an environment with decreased I ,25(OH),D,. The application of vitamin D or analogs in chemotherapy against prostate and other cancers is being explored by several investigators. In order to use vitamin D most efficaciously in a clinical setting, it may be beneficial to learn more about the interaction of I ,25(OH),D, with other factors that regulate prostatic epithelial cellular growth. In this study, we examined the effect of the proliferative status of cultured cells on their ability to respond to 1,25(OH),D,, and found that minimally proliferative cells were equally as responsive to I ,25(OH),D, as actively dividing cells. We noted no apparent interaction of 1,25(OH),D, with epidermal growth factor, insulin-like growth factor, cholera toxin, or transforming growth factor-p, but we did find synergistic inhibitory effects of I ,ZS(OH),D, with suramin and retinoic acid. Perhaps most noteworthy was the dramatic increase in potency of 1,25(OH),D, that occurred upon deletion of hydrocortisone from the culture medium. Our in vitro studies indicate that combination therapy of vitamin D analogs with suramin, vitamin A analogs, or anti-glucocorticoids might be considered for prostate cancer. (Ural Oncol I 995; I : /88-I 94)

KEY WORDS: Vitamin D, suramin, coids, prostate cancer

vitamin

A, glucocorti-

Address correspondence to: Donna M. Peehl, PhD, Department of Urology, Stanford University School of Medicine, Stanford, California 94305-5118. Supported by NIH Grant DK47551 to DMP, NIH Grant DK42482 to DF, and CaP CURE Awards to DMP and DF. Ural Oncol 1995;1:18&194 0 1995 Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010

School

of Medicine,

Stanford,

California

,25_dihydroxyvitamin D, [ 1,25(OH),D,] is a steroid hormone with diverse targets.’ Vitamin D receptors (VDR) have been detected in prostatic tissueszS3 prostatic cancer cell lines415 and primary cell cultures derived from the epithelium and stroma of the prostate.3 Responses of cultured prostatic cells to 1,25(OH),D, include induction of 24-hydroxylase, inhibition of proliferation, and increased expression of prostate-specific antigen (PSA).3-5 Epidemiologic investigations suggest that vitamin D deficiency may increase the risk of clinical prostate cancer. Hanchette and Schwartz’ found that mortality rates from prostate cancer in the United States are inversely proportional to exposure to ultraviolet irradiation, which is essential for the synthesis of vitamin D. Other factors potentially associated with vitamin D deficiency (age, race, and diet) also correlate with increased risk of mortality from prostate cancer.7 Additional data supporting a relationship between vitamin D status and the risk of developing clinical prostate cancer were reported by Corder et al8 In a prospective study of stored sera, mean levels of 1,25(OH),D, were slightly but significantly lower in sera from prostate cancer patients compared with controls. For men older than 57 years, serum levels of 1,25(OH),D, were an important predictor of risk for the development of clinical prostate cancer. These epidemiologic findings may reflect a propensity for increased growth and decreased differentiation of prostate cancer cells in an environment with insufficient 1,25(OH),D,. The intriguing epidemiologic and cell culture results have inspired efforts to use vitamin D therapeutically. The results of a small phase II trial testing the effects of 1,25dihydroxyvitamin D (calcitriol) on hormone-refractory prostate cancer have been compiled by Osborn et al.’ A maximum oral daily dose of 1.5 mg of calcitriol was given to 13 evaluable patients for a median duration of 13.5 weeks. No objective responses were seen. A current limitation of the therapeutic application of vitamin D is toxicity due to hypercalcemic effects. Analogs with increased antiproliferative activity and decreased hypercalcemic effects

I

1078-1439/95/$15.00 SSDI 10781439(95)00063-l

Ural Oncol

I 995; I: /88-

I94

are being developed and show promise in in vih-o tests against prostate cancer cell lines.‘” In order to use 1,25(OH),D, or analogs most efficaciously, information about factors affecting the activity or potency of 1,25(OH),D, would be valuable. In our previous studies, we noted that approximately 1 nM of 1,25(OH),D, caused half-maximal growth inhibition of primary cultures of prostatic epithelial cells, and 10 nM caused complete inhibition.3 These growth-inhibitory effects of 1,25(OH),Da occurred in serum-free medium containing all of the mitogens and other factors necessary to promote rapid proliferation of prostatic epithelial cells in the absence of 1,25(OH),D,. In vivo, the growth rate of prostate cells, even malignant ones, is often not very rapid. In order to emulate in vivo conditions more closely, in the current study we modified the proliferation rate of prostate cells by reducing the concentrations of individual mitogens in the medium, then compared the effects of 1,25(OH),D, in complete versus deficient media. We also examined the interactive effects of 1,25(OH),D, and other factors that inhibit the growth of prostatic epithelial cells, including transforming growth factor-p (TGFB), retinoic acid, and suramin. Finally, we wanted to determine whether we could find any conditions in which 1,25(OH),D, stimulated, rather than inhibited, the growth of prostatic epithelial cells. Although previous studies indicated that the overall effect of 1,25(OH),D, on prostate cells was growth inhibitory, one situation was reported in which 1,25(OH),D, stimulated growth. When LNCaP cells were cultured in medium supplemented with charcoal-stripped rather than whole serum, 1,25(OH),D, was observed to stimulate cell growth by two independent groups.4,5 The explanation for this seeming anomaly is not known, but the possibility of growth-stimulator-y effects of 1,25(OH),D, is disquieting if 1,25(OH),Da is to be used as a therapeutic agent against cancer. In our current study, we were unable to find any conditions in which 1,25(OH),D, promoted rather than inhibited growth of primary cultures of prostatic epithelial cells. However, we did find that certain factors, specifically hydrocortisone, retinoic acid, and suramin, altered the potency of 1,25(OH),D,. It may be feasible to utilize these findings in the design of future therapeutic protocols.

Materials

and Methods

Cell Culture Six strains of adult human prostatic epithelial cells were used for these studies. Strains E-PZ-40, E-PZ-60, and E-PZ-73 were obtained from normal peripheral zone tissues of radical prostatectomy specimens. Strains E-CA-42 and E-CA-50 were derived from adenocarcinomas of Gleason grade 3 from radical prostatectomy specimens. These cell strains were established in culture by previously described methods.” E-PZ51 cells were cultured from a prostatic needle biopsy of the normal peripheral zone; culture techniques for the needle biopsy were as previously described.” Epithelial cell cultures established by these methods have

Suromin,

Hydrocortisone,

and Retinoic Acid

189

been extensively characterized.13 These cells are mortal and undergo approximately 30 population doublings before senescence; they are androgen-insensitive and express keratins, prostatic acid phosphatase, and PSA.“,‘3 Primary cultures were frozen, then thawed and passaged as described for growth assays and vitamin D receptor binding analyses.

Clonal Growth Assays Primary cultures were thawed to establish secondary cultures (at approximately eight population doublings). Cells from secondary cultures were used for clonal growth assays. Previously described protocols were followed.i4 Cells (200-400 per dish) were inoculated into 60-mm, collagen-coated dishes containing 5 ml of control or experimental medium. After incubation for 10 days in a 37°C humidified, 5% COJ95% air incubator, dishes were fixed with 10% formalin and stained with crystal violet.” The total area covered by cells on each dish was quantitated with an Artek image analyzer (Dynatech, Chantilly, VA). It has been demonstrated that the relative values obtained by this method are directly proportional to cell number.15 Control medium for clonal assays was MCDB 105 (Sigma, St. Louis, MO) supplemented with cholera toxin 10 ng/ml, epidermal growth factor 10 ng/ml, bovine pituitary extract 10 pg/ml, phosphoethanolamine 0.1 nM, hydrocortisone 3 pM, selenous acid 30 nM, gentamycin 100 pg/ml, insulin 4 pg/ml, alpha-tocopherol2.3 pM, and retinoic acid 0.03 nM. The sources and preparation of these supplements have been described.” Individual factors were deleted and added back at the concentrations described in each experiment. Other factors tested in these experiments included insulin-like growth factor-l and TGFBl (R&D Systems, Minneapolis, MN) and suramin (National Cancer Institute, Bethesda, MD). 1,25(OH),D, was obtained from Biomol (Plymouth Meeting, PA). Concentrated stocks of 1,25(OH),Da were prepared in ethanol and stored at -20°C. The concentration of ethanol in media did not exceed 0.01%.

VDR Binding Analyses Tertiary cultures (at approximately 20 population doublings) were grown to near-confluency, rinsed twice with ice-cold phosphate-buffered saline, then harvested by scraping with a rubber policeman. After an initial wash in phosphate-buffered saline, cell pellets were resuspended in KTEDM buffer (0.3 M KCI, 10 mM Tris-Cl, pH 7.4, 1.5 mM EDTA, 1 mM dithiothreitol, and 10 nM sodium molybdate) containing a protease inhibitor cocktail (50 pg/ml soybean trypsin inhibitor, 0.5 pg/ml leupeptin, 1.4 pg/ml pepstatin, and 330 pg/ml benzamidine). Cells were disrupted by sonication on ice. The sonicate was centrifuged at 207,000 x g for 35 minutes at 4°C to obtain a soluble extract used for binding studies. In each singlepoint saturation experiment, 200 p1 of soluble extract (l-2 mg of protein/ml) were incubated with 1 nM of la, 25(OH),[23,24(n)-3H] vitamin D3 (specific activity, 102 Ci/mmol; Amersham, Arlington Heights, IL) for 16-20 hours at 4°C with or without a 250fold excess of radioinert hormone (a gift from Dr. M. Usko-

190

Ural Oncol

Peehl et al.

I 995; I : I 88-I

94

kovic, Hoffman-La Roche, Nutley, NJ). Bound and free hormone were separated by the hydroxylapatite method.16 The protein concentration of soluble extract samples was measured by the method of Bradford.17 Specific binding was calculated by subtracting nonspecific binding obtained in the presence of a 250-fold excess of radioinert 1,25(OH),D, from the total binding measured in the absence of radioinert steroid.

maximal inhibition at approximately 0.25 nM and almost complete inhibition at 2.5 nM. In medium with 0.5% (v/v) charcoal-stripped serum, 1,25(OH),D, was about ten-fold less potent, with half-maximal inhibition occurring at approximately 2.5 nM and complete inhibition at 25 nM. However, at no level of 1,25(OH),D, (0.025-25 nM) was growth stimulated in either serum-free medium or medium supplemented with charcoal-stripped serum.

Results

Relationship of Growth to 1,25(OH),D,

Activity of /,25(OH),D, in Presence of Charcoal-stripped Serum To test whether charcoal-stripped serum would alter the response of prostatic epithelial cell strains to 1,25(OH),D,, we compared the clonal growth response of these cells to 1,25(OH),D, in serum-free medium versus medium supplemented with charcoal-stripped serum. The previously reported experiments in which 1,25(OH),D, stimulated the growth of LNCaP cells were conducted with 5% (v/v) or greater of charcoal-stripped serum added to the medium.4v5 Because the medium we developed for clonal assays of prostatic epithelial cell strains was specifically designed to be optimal without serum, the addition of serum becomes toxic with increasing concentrations. Based on previous experiments, a level of 0.5% (v/v) was chosen as the highest amount of charcoal-stripped serum that could be added to the serum-free medium without overwhelming toxic effects. Figure 1 shows that E-PZ-40 cells responded to 1,25(OH),D, in serum-free medium as previously reported for primary cultures of prostatic epithelial cells, with half-

0

0.025

0.25

2.5

25

1.26(0H),D, (nM) in medium with charcoalFIGURE I. Resoonse to 1.2510HLD, ,stripped serum. E-PZ-40 cells were inoculated into complete, serum-free medium (0 serum) or into complete medium supplemented with 0.5% (v/v) charcoal-stripped serum (0.5% CSS). 1,25(OH),D, was added at the concentrations indicated. After IO days of incubation, growth was quantitated by image analysis. Each point represents the mean of duplicate experiments, with duplicate dishes in each experiment, fSEM. The value obtained for growth in each experimental condition (0 serum or 0.5% CSS) in the absence of I ,25(OH),D3 was standardized as 100%.

Status and Response

The results described in the previous section indicated that the presence of charcoal-stripped serum did not promote a growth-stimulatory effect of 1,25(OH),D, on prostatic epithelial cell strains. In the previously published experiments in which 1,25(OH),D, stimulated the growth of LNCaP cells, the basal level of growth of LNCaP cells in the absence of 1,25(OH),D, was considerably less in medium with charcoal-stripped serum compared with medium with whole serum.4 It is possible that the observed stimulatory effect of 1,25(OH),D, was not directly related to the presence of charcoal-stripped serum, but to the slower growth rate of cells in medium with charcoal-stripped serum. Alternatively, the slower growth of LNCaP cells in medium with charcoal-stripped serum perhaps indicated that essential growth factors were removed during the stripping process, and stimulation by 1,25(OH),D, may have been due to altered interactions with specific growth factors. To evaluate these possibilities, we analyzed the response of prostatic epithelial cells to 1,25(OH)2D, in factordepleted, serum-free medium. Epidermal growth factor and insulin or insulin-like growth factor are required for the optimal proliferation of prostatic epithelial cells.14 Each was individually deleted and added back at reduced levels; the effects of 1,25(OH),D, were then compared in factor-deficient versus complete media. The response of E-PZ51 cells to 1,25(OH),D, in media with 0.1 or 10 ng/ml of epidermal growth factor is depicted in Figure 2. Although overall growth in medium with 0.1 ng/ml of epidermal growth factor was less than in medium with 10 ng/ml of epidermal growth factor, 0.25 and 2.5 nM of 1,25(OH),D, induced the same degree of growth inhibition in both media. No growth stimulation by 1,25(OH),D, occurred in either medium. Similar results were seen when the response of E-PZ-60 cells to 0.25 or 2.5 nM of 1,25(OH),D, in media with 0.1 or 10 ng/ml of insulin-like growth factor 1 was measured (Figure 2). Again, although overall growth was reduced in medium with the lesser amount of insulin-like growth factor, 1,25(OH),D, induced similar growth inhibition in each medium.

2

Interactions of 1,25(OH),D, Toxin and Hydrocortisone

with Cholera

Cholera toxin aid hydrocortisone are two our serum-free medium that alone have growth, but which act synergistically with factors in the medium.18 We analyzed

components of little effect on peptide growth the effects of

Ural Oncol /995;1:/88-194

O0

Suramin, Hydrocortisone, and Retinoic Acid

0.25 1,25(D~),o,

2.5

F

d

120

$

100

0 0”

60

l-

60

2 3

46

L

20 0

(nW

- CT -0,

6

120

F

100

z5 0

-CT + 0,

+ CT

+ CT

- 0,

+ D,

0’ p 120

80

x

100 60

t;

60

0 8

5 w U

40

g

60

%

20

k?

46

i

20

0' 0

191

-I 0.25 ~,26(W,D,

2.5 WV

0

FIGURE 2.

Response to 1,25(OH),D, in factor-deficient media. EPZ-5 I cells (top) or E-PZ-60 cells (bottom) were inoculated into complete medium with 0.1 or IO ng/ml of EGF (top), or 0.1 or IO ng/ml of IGF-I (bottom), and concentrations of I ,25(OH),D, as indicated. After IO days of incubation, growth was quantitated by image analysis. Each point represents the mean of duplicate experiments, with duplicate dishes in each experiment, &EM. The value obtained for growth in each experimental condition (0.1 EGF, IO EGF, 0. I IGF, or IO IGF) in the absence of I .25(OH),D, was standardized as 100%.

1,25(OH),D, in media with either cholera toxin or hydrocortisone individually deleted. Figure 3 demonstrates that, as expected, growth of E-PZ40 cells in medium with cholera toxin was not substantially different from that in medium without cholera toxin in the absence of 1,25(OH)*D,. Growth inhibition induced by 2.5 nM of 1,25(OH),D, was also not significantly different in media with or without cholera toxin. The deletion of hydrocortisone from the medium also had no effect on the growth of E-PZ40 cells in the absence of 1,25(OH),D, (Figure 3). Growth inhibition induced by 1,25(OH),Da, however, was dramatically different depending on whether hydrocortisone was present. In these experiments, 2.5 nM of 1,25(OH),D, was minimally inhibitory in the presence of 3 pM of hydrocortisone, while causing almost complete growth inhibition in the absence of hydrocortisone. The elimination of hydrocortisone therefore greatly enhanced the potency of 1,25(OH),D,. A similar effect of hydrocortisone on responsiveness to 1,25(OH),D, was also seen with two cell strains (E-CA-42 and E-CA-50) derived from prostatic adenocarcinomas. For both of these cell strains, growth inhibition by 1,25(OH),D, was ten-fold greater in medium without hydrocortisone compared with medium with hydrocortisone (not shown).

-HC

-HC

+HC

- 0,

+ 0,

* 0,

+HC + Da

FIGURE 3.

Effects of cholera toxin or hydrocortisone on response to I ,25(OH),D,. E-PZ-40 cells were inoculated into complete medium f IO nglml of cholera toxin (CT; top graph) or f3 uM of hydrocortisone (HC; bottom graph), f2.5 nM of 1,25(OH),D, (Da). After IO days of incubation, growth was quantitated by image analysis. Each bar represents the means of duplicate experiments, with duplicate dishes in each experiment, *SEM. The values obtained for growth in medium with CT or with HC, and without Da, were standardized as I OO%, respectively.

I,ZS(OU),D, and Other Growth-Inhibitory Factors TGFP, retinoic acid, and suramin each inhibit the growth of normal and cancer-derived prostatic epithelial cells.‘g-21 Sub-optimal inhibitory levels of these factors were combined with 1,25(OH),D, to test for synergistic or other effects. TGFf3 and 1,25(OH),D, were not synergistic when tested on E-PZ51 cells (Figure 4). The combinations of retinoic acid and 1,25(OH),Da, however, did result in synergistic growth-inhibitory activity (Figure 4). The concentrations of retinoic acid and 1,25(OH),D, used in these experiments each alone minimally inhibited growth (~20% inhibition), whereas the combination of the two factors caused greater than 80% inhibition. Suramin and 1,25(OH),D, were also synergistic (Figure 4). The concentration of suramin used by itself did not inhibit growth of E-PZ-40 cells, but when added with 1,25(OH),D,, growth inhibition was significantly greater than induced by 1,25(OH),D, alone. The growth of cancer-derived cell strains was also inhibited in a synergistic manner by retinoic acid or suramin in combination with 1,25(OH)*D, (not shown). Cell strains E-CA-42 and E-CA-50 were inhibited on average 60% by retinoic acid (0.3 nM) and 20% by 1,25(OH),D, (0.25 nM), but

Peehl et al.

Ural Oncol I995;1:188-I

loo 80 60

40 20 0 -TGF - D3

- TGF

+ TGF

+ TGF

+ 03

- D3

+ D3

1 100

80

94

titularly noteworthy and stimulated further investigation. One likely possibility explaining this phenomenon would be a decrease in VDR levels in the presence of hydrocortisone, which could decrease cellular responsiveness to 1,25(OH),D,. Two separate attempts with two different cell strains to find a decrease in VDR levels by ligand-binding assays in response to hydrocortisone failed to demonstrate any significant change (Table 1). We concluded that alterations in VDR levels did not explain the decreased potency of 1,25(OH),D, in the presence of hydrocortisone. We also measured VDR levels in cells in response to retinoic acid and suramin (Table 2). Using concentrations of retinoic acid and suramin that acted synergistically with 1,25(OH),D, in growth assays, we did not observe any significant change in VDR levels by binding analysis. Therefore, induction of increased VDR does not appear to be the mechanism by which retinoic acid or suramin increase cellular response to 1,25(OH),D,.

so 40

Discussion

20

The results of these studies provide information relevant to the clinical application of 1,25(OH),D, as a chemotherapeutic agent against prostate cancer. First, under no circumstances did we observe any stimulatory effects of 1,25(OH),D, on prostatic epithelial cell strains. The variety of conditions that we tested, ranging from the addition of charcoal-stripped serum to the depletion of growthstimulatory factors to the addition of growth-inhibitory factors, presumably mimicked the status of LNCaP ceils when they were reportedly stimulated by 1,25(OH),D,4~5 Nevertheless, in all of these conditions, 1,25(OH),D, continued to exert only inhibitory effects on cell strains. We conclude that growth stimulation by 1,25(OH),D, of LNCaP cells cultured in charcoal-stripped serum is seemingly a unique event that is not likely to occur in vivo on the basis of the results we obtained with cells cultured directly from the prostate. Second, our results indicate that conditions exist in which the growth-inhibitory effects of 1,25(OH),D, on prostate cells can be amplified. This finding may provide opportunities to use relatively low, nonhypercalcemic levels of 1,25(OH),D, with increased efficacy. Our results suggest that low levels of 1,25(OH),D, could be combined with low levels of retinoic acid, for example, in therapeutic applications. As for 1,25(OH),D,, therapeutic effects of reti-

0 -RA

-RA

- 4

+ 4

+RA

+RA

- 03

+ D3

- SUR

+ SUR

+ SUR

+ D3

- D3

+ D3

120 100 a0 60 40 20 0 - SUR - D3

FIGURE 4. Interactions of 1,25(OH),D, with other growthinhibitory factors. Top: E-PZ-5 I cells were inoculated into complete medium kO.01 &ml of TGFB and +0.25 nM of 1,25(OH),D, (Da). Middle: E-PZ-51 cells were inoculated into complete medium k3 nM of retinoic acid (RA) and k0.25 nM of 1,25(OH),D, (D,). Bottom: E-PZ-40 cells were inoculated into complete medium +5 uglml of suramin (SUR) and + I I- 2.5 nM of I ,25(OH),D, (Da). After IO days of incubation, growth was quantitated by image analysis. Each bar represents the mean of duplicate (top, middle) or triplicate (bottom) experiments, with duplicate dishes in each experiment, *SEM. The values obtained for growth in medium without TGFB. RA, or SUR, and without Da, were standardized as lOO%, respectively.

TABLE I. VDR LEVELS IN CELLS GROWN WITH OR WITHOUT HYDROCORTISONE VDR fmol/mg protein*

84% by the

inhibited E-CA-50 cells by less than 20%, and 1,25(OH),D, by 42%, but the combination inhibited growth by 62%. combination.

Similarly,

5 pg/ml

of suramin

VDR in Response to Hydrocortisone, Retinoic Acid, and Suramin The dramatically increased inhibitory 1,25(OH),D, in the absence of hydrocortisone

effects of were par-

E-PZ-40** E-PZ-60**

-HC

+HC

15.6 9.5

15.5 9.1

Abbreviations: VDR = vitamin D receptor; HC = hydrocortisone. VDR levels were measured by single-point ligand binding assays as described in Materials and Methods: each cell strain was assayed once. *‘Cells were grown to semiconfluency in complete medium, then changed to medium zHC for 72 hours, at which time VDR levels were measured.

Ural Oncol I995; /:188-l 94

Suramin,

TABLE

2. VDR LEVELS IN CELLS TREATED WITH RETINOIC ACID OR SURAMIN VLIR

Control** +Retinoic acid** +Suramin* *

fmol/mg protein* 11.9 13.7 17.5

Abbreviations: VDR = vitamin D receptor. ?? VDR levels were measured by single-point ligand binding assays as described in Materials and Methods. ““Cells @P-P&73) were grown to semi-confluency in complete medium, then changed to complete medium with nothing added (control), with retinoic acid (3 niV), or with suramin (5 pg/mI). VDR levels were measured 72 hours later.

noic acid or other forms of vitamin A are limited by toxicity, but vitamin A as well as vitamin D analogs are being developed and clinical trials using retinoids to treat prostate cancer are in progress. Similarly, combined treatment with 1,25(OH),D, and suramin might be considered. Suramin has demonstrated considerable potential for treatment of hormoneinsensitive prostate cancers,” and simultaneous administration of 1,25(OH),D, might further improve its efficacy. An element complicating this strategy might be the fact that adrenal insufficiency is often a side-effect of suramin therapy, necessitating hydrocortisone replacement therapy. Our results indicate that 1,25(OH),D, is less potent in the presence of hydrocortisone. An alternate strategy to combination suramin and 1,25(OH),D, therapy might be administration of 1,25(OH),D, in conjunction with treatment to reduce glucocorticoid levels. The basis of the observed interactions of 1,25(OH),D, with other factors is interesting to consider. Certainly the synergism of 1,25(OH),D, with retinoic acid or suramin cannot be simply explained by increased sensitivity of suboptimally growing cells to 1,25(OH),D,, since cells growing suboptimally in response to TGFf3 or because of factordeficient (epidermal growth factor or insulin-like growth factor) media were not more sensitive to 1,25(OH),D,. Synergy between retinoic acid and 1,25(OH),D, has been found from studies with other types of cells,23 and a mechanism involving interaction between retinoic acid and vitamin D signaling pathways has been proposed.24 To our knowledge, synergy of 1,25(OH),D, and suramin has not been previously reported. We have observed that growth inhibition of prostatic epithelial cells by 1,25(OH),D, is irreversible even after two hours of exposure,3 whereas growth inhibition by suramin is reversible even after prolonged periods of exposure.” The mechanism of growth inhibition of prostate cells by suramin is not clear. While some propose a blockage of growth factorreceptor interactions by suramin,25X26 others suggest that suramin inhibits prostate cell growth by inhibition of mitochondrial energy pathways.‘7,28 1,25(OH),Da, on the other hand, may inhibit growth of prostatic cells by inducing differentiation.5 How these proposed mechanisms of action of suramin and 1,25(OH),D, could result in synergistic activity is not obvious. One possible mechanism for the enhancement of

Hydrocortisone,

and Retinoic Acid

193

1,25(OH),D, action by retinoic acid and suramin, and the antagonism of 1,25(OH),D, by hydrocortisone, is regulation of VDR abundance or affinity. Retinoic acid has been shown to increase the amount of VDR in rat osteosarcoma cells” and in primary cultures of mouse osteoblasts,“’ but to decrease VDR in rat osteoblasts.30 The up or down regulation of VDR in mouse or rat osteoblasts was associated with comparable increased or decreased responsiveness to 1,25(OH),D,. Similarly, glucocorticoid regulation of VDR differed by species, showing decreased VDR in mouse and increased VDR in rat bone3’ and intestine.31 Nielson et a1.32 showed that glucocorticoids decreased nuclear uptake of [3H]1,25(OH)2D3 in human monocytes.32 The data from regulation of VDR in mouse intestine and bone and human monocytes would predict glucocorticoid antagonism of 1,25(OH),D, activity by downregulation of VDR. However, in prostatic cells we did not detect a change in the amount of VDR in response to hydrocortisone, retinoic acid, or suramin. Further studies will be required to elucidate the interactions of hydrocortisone, retinoic acid, or suramin with 1,25(OH),D,. References 1. Walters MR. Newly identified actions of the vitamin D endocrine system. Endocr Rev 1992;13:719-64. 2. Berger U, Wilson P, McClelland RA, Colston K, Haussler MR, Pike JW, Coombes RC. Immunocytochemical detection of 1,25dihyroxyvitamin D receptors in normal human tissues. J Clin Endocrin Metab 1988;67:607-13. 3. Peehl DM, Skowronski RJ, Leung GK, Wong ST, Stamey TA, Feldman D. Antiproliferative effects of 1,25_dihydroxyvitamin D, on primary cultures of human prostatic cells. Cancer Res 1994;54:1-6. 4. Miller GJ, Stapleton GE, Ferrara JA, Lucia MS, Pfister S, Hedlund TE, Upadhya P. The human prostatic carcinoma cell line LNCaP expresses biologically active, specific receptors for 1 01, 25dihydroxyvitamin D,. Cancer Res 1992;52:515-20. 5. Skowronski RJ, Peehl DM, Feldman D. Vitamin D and prostate cancer: 125 Dihydroxyvitamin D, receptors and actions in human prostate cancer cell lines. Endocrinology 1993;132:195260. 6. Hanchette CL, Schwartz GG. Geographic patterns of prostate cancer mortality. Cancer 1992;70:2861-9. 7. Schwartz GG, Hulka BS. Is vitamin D deficiency a risk factor for prostate cancer? (Hypothesis). Anticancer Res 1990;10:1307I”

1‘.

8.

Corder EH, Guess HA, Hulka BS, Friedman GD, Sadler M, Vollmer RT, Lobaugh B, Drezner MK, Vogelman JH, Orentreich N. Vitamin D and prostate cancer: A prediagnostic study with stored sera. Cancer Epidemiol Biomarkers & Prev 1993;2:46772. 9. Osborn JL, Bahnson R, Schwartz GG, Smith DC, Trump DL. Phase II trial of oral 1,25_dihydroxyvitamin D (calcitriol) in hormone refractory prostate cancer. Submitted. 10. Skowronski RJ, Peehl DM, Feldman D. Actions of vitamin D, analogs on human prostate cancer cell lines: Comparison with 1,25_dihydroxyvitamin D,. Endocrinology 1995;136:1-7. 11. Peehl DM. Culture of human prostatic epithelial cells. In: Freshney RI (ed). Culture of Epithelial Cells. New York: WileyLiss, Inc, 1992;159-80. 12. Peehl DM, Wong ST, Terris MK, Stamey TA. Culture of prostatic epithelial cells from ultrasound-guided needle biopsies. Prostate 1991;19:141-7. 13 Peehl DM. The male reproductive system: Prostatic cell lines. In: Hay RJ, Park JG, Gazdar A (eds). Atlas of Human Tumor Cell Lines. San Diego, CA: Academic Press, 1994;387-411.

194

14.

Peehl et al.

Peehl DM, Wong ST, Stamey TA. Clonal growth characteristics of adult human prostatic epithelial cells. In Vitro 1988;24: 530-6. 15. Tsao MC, Walthall BJ, Ham PG. Clonal growth of normal human epidermal keratinocytes in a defined medium. J Cell Physiol 1982;110:219-29. 16. Malloy PJ, Hochherg Z, Pike JW, Feldman D. Abnormal binding of vitamin D receptors to deoxyribonucleic acid in a kindred with vitamin D-dependent ricketts, type 11.J Clin Endocrinol Metab 1989;68:263-9. 17. Bradford MM. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein dye binding. Anal Biochem 1976;72:248-54. 18. Peehl DM, Stamey TA. Serum-free growth of adult human prostatic epithelial cells. In Vitro 1986;22:82-90. 19. Peehl DM, Wong ST, Bazinet M, Stamey TA. In vitro studies of human prostatic epithelial cells: Attempts to identify distinguishing features of malignant cells. Growth Factors 1989;l: 23 7-50. 20. Peehl DM, Wong ST, Stamey TA. Cytostatic effects of suramin on prostate cancer cells cultured form primary tumors. J Urol 1991;145:624-30. 21. Peehl DM, Wong ST, Stamey TA. Vitamin A regulates proliferation and differentiation of human prostatic epithelial cells. Prostate 1993;23:69-78. 22. Eisenberger MA, Reyno L, Sinibaldi V, Sridhara R, Carducci M, Egorin M. The experience with suramin in advanced prostate cancer. Cancer 1995;75(suppl):1927-34. 23. Koga M, Sutherland RL. Retinoic acid acts synergistically with 1,25_dihydroxyvitamin D, or antioestrogen to inhibit T-47D human breast cancer cell proliferation. J Steroid Biochem Molec Biol 1991;39:455-60.

Ural Oncol

I995;/:/88-I

94

24. Schrader M, Bendik I, Becker-Andre M, Carlberg C. Interaction between retinoic acid and vitamin D signaling pathways. J Biol Chem 1993;268:17830-6. 25. Kim JH, Sherwood ER, Sutkowski DM, Lee C, Kozlowski JM. Inhibition of prostatic tumor cell proliferation by suramin: Alterations in TGF alpha-mediated autocrine growth regulation and cell cycle distribution. J Ural 1991;146:171-6. 26. Stein CA. Suramin: A novel antineoplastic agent with multiple potential mechanisms of action. Cancer Res 1993;53:2239-48. 27. Rago R, Mitchen J, Cheng A-L, Oberley T, Wilding Cl.Disruption of cellular energy balance by suramin in intact human prostatic carcinoma cells, a likely antiproliferative mechanism. Cancer Res 1991;51:6629-35. 28. Rago RP, Brazy PC, Wilding G. Disruption of mitochondrial function by suramin measured by rhodamine 123 retention and oxygen consumption in intact DU145 prostate carcinoma cells. Cancer Res 1992;52:6953-5. 29. Petkovich PM, Heersche JN, Tinker DO, Jones G. Retinoic acid stimulates 1,25 dihydroxyvitamin D, binding of rat osteosarcoma cells. J Biol Chem 1984;259:8274-80. 30. Chen TL, Feldman D. Retinoic acid modulation of 1,25(OH), vitamin D, receptors and bioresponse in bone cells: Species differences between rat and mouse. Biochem Biophys Res Comm 1985;132:74-80. 31. Hirst M, Feldman D. Glucocorticoid regulation of 1,25(OH), vitamin D, receptors: Divergent effects on mouse and rat intestine. Endocrinology 1982;111:1400-2. 32. Nielsen HK, Eriksen EF, Storm T, Mosekilde L. The effects of short-term, high-dose treatment with prednisone on the nuclear uptake of 1,25dihydroxyvitamin D, in monocytes from normal human subjects. Metabolism 1988;37:109-14.