Mineralocorticoid receptors in the epithelial cells of human colon and ileum

J. sreroid Biochem. Vol. 20, No. 1, pp. 311-315, 1984 Printed in Great Britain. All rights reserved

Copyright 0

0022-4731/84 $3.00 + 0.00 1984 Pergamon Press Ltd

MINERALOCORTICOID RECEPTORS IN THE EPITHELIAL CELLS OF HUMAN COLON AND ILEUM M. E.

RAFESTIN-OBLIN, INSERM

M. U36,

LOMBES, J. B. MICHEL,

A.

MICHAUD

and M.

CLAIRE

17, rue du Fer-a-Moulin, 75005 Paris, France

Summary-Specific binding of [3H]aldosterone to the cytosolic fraction of epithehal cells was studied in the human colon and terminal ileum. Analysis of [3H]aldosterone binding to the epithelial cells of ascending colon, caecum and ileum as a function of [3H]aldosterone concentration revealed only one class of specific receptors with an affinity constant of about 2 nmol/l. [sH]aldosterone binding was approximately the same in the sigmoid, descending and transverse colon and in the caecum, but slightly lower in the ascending colon and ileum. The specificity of the [‘Hlaldosterone binding was the same along the

colon. The relative order of potency in inhibiting [‘Hlaldosterone binding was: aldosterone = SC 26304 = dexamethasone>>dihydrotestosterone > estradiol= RU 26988.

INTRODUCTION

An effect of aldosterone on electrolyte transport in the mammalian colon has been inferred from studies that have demonstrated increases of sodium reabsorption, potassium secretion and transmucosal potential difference [l-5]. The action of aldosterone, like that of other steroid hormones, is believed to be initiated by its binding to cytoplasmic receptors [6]. Specific binding of aldosterone has been shown in various target organs including the colonic mucosa of the rat [7-81. However to our knowledge such a study has never been performed in human gut mucosa. The present study was therefore designed to investigate and characterize [3H]aldosterone binding to cytosol prepared from epithelial cells of human colonic mucosa and terminal ileum. In the rat kidney at least two corticoid binding sites have been identified on the basis of their hormonal specificity and proposed as type I (mineralocorticoid) and type II (glucocorticoid) receptors. In this target tissue aldosterone binds to both types of receptors, displaying the highest affinity for type I receptors [6]. The specificity of aldosterone binding to the human colonic mucosa for each class of sites was assessed by competition experiment using various potential type I and/or type II markers. Eventual cross-competition with androgens and oestrogens was also checked. Since the functional properties of the large intestine such as generation of a transepithelial electrical potential difference and net sodium transport exhibit an oral aboral gradient [9-l I], [3H]aldosterone binding was investigated in various colonic segments as well as in the ileum. MATERIALS

AND METHODS

Chemicals

[ 1,2-‘Hlaldosterone (40-50 Ci/mmol) was purchased from the Radiochemical Center, Amersham.

Aldosterone, estradiol and dihydrotestosterone were obtained from Sigma. SC 26304 and RU 26988 were gifts from the Searle and Roussel Laboratories respectively. Dextran T70 and charcoal (Norit A) were purchased from Pharmacia. Scintillation solution for counting was from Packard (Pica-fluor 15) and all other products were of reagent grade from Merck. Tissues

Fresh specimens of sigmoid, descending, transverse and ascending colon and of terminal ileum were obtained during segmental resection for cancer from 8 different patients who included both male and female. Tissue was also obtained from two patients undergoing total colectomy for multiple polyposis (sigmoid, descending and ascending colon and caecum from patient No. 5, and descending and transverse colon, caecum and ileum from patient No. 10). Immediately after surgery each fragment of tissue was removed from the macroscopically healthy portion of the resected segment, rinsed with 0.9% NaCl solution at room temperature, cleared of fatty tissue, everted and rinsed once more with 0.9%NaCl. Preparation

of the cytosolic fraction

The epithelial cells of the various colonic fragments were isolated in the form of crypts at 4°C as described by DuPont et a1.[12]. The buffer used to isolate the crypts was slightly modified and contained 10 mmol/l Tris-HCI, 2.5 mmol/l EDTA, and 0.25 mol/l NaCl adjusted to pH 7.4. As specified by DuPont et a/.[121 crypts were extracted from the sigmoid and descending colon by 4 to 6 manual 10 s shakes. For the transverse and ascending segments and for the caecum the duration of crypt extraction was slightly longer (8 to 10 shakes). Cells were isolated from the ileum by the same technique after 8 to 10 shakes. As described by DuPont et a1.[12] epithelial cells were washed 6 times with 10 mmol/l Tris-HCl pH 7.4 containing 0.15 mol/l NaCl. The duration of the cellisolation procedure was about 90min. Cells were

311

312

M. E.

RAFESTIN-OBLIN et al.

frozen and ground in a mortar in the presence of liquid nitrogen, Three volumes of 20mmol/l Tris-HCl, 1 mmol/l EDTA pH 7.4 containing 1 mmol/l dithiothreitol and 10% glycerol (v/v) were added to the resulting powder. As soon as the mixture was defrozen, it was vigorously vortexed at 0°C and centrifuged at 48,000g for 10 min leading to what will be referred to as the cytosolic fraction. Time course study of [‘H]aldosterone binding to the cytosolic fraction

The cytosolic fraction of the ground epithelial cells from an ascending colon was incubated with 20nmol/l [‘Hlaldosterone at 0°C. The time course study was carried out, at 30,60,90, 120, 180,240, 300 and 360min. At the end of each incubation period, bound and free hormone were separated under the following conditions: 75 ~1 of a suspension of 4% Norit A and 0.4% Dextran T70 in 20mmol/l Tris-HCl, 1 mmol/l EDTA adjusted to pH 7.4 was added to 100 ~1 of the cytosolic fraction. After 10 min incubation at O”C, the mixture was centrifuged for 10 min at 1,OOOgand the supernatant collected for counting. To check the stability of the aldosterone receptor, the cytosolic fraction was divided into 3 aliquots: one with 20 nmol/l incubated immediately was [‘Hlaldosterone for 3 h at 0°C. The other two aliquots were maintained at 0°C in the absence of hormone for 2 h and 4 h respectively, at which time 20 nmol/l [‘Hlaldosterone was added to each aliquot which was then maintained at 0°C until steady state was reached (3 h). Bound and free hormone were then separated as described above. [3H]aldosterone binding characteristics at equilibrium

Increasing concentrations of [3H]aldosterone (0.2 to 200 nmol/l) were added to 100 ~1 aliquots of the cytosolic fraction and maintained at 0°C for 3 h, the time required to reach equilibrium. Total radioactivity was then counted (T), bound and free hormone were separated by the charcoal-dextran technique and the receptor bound hormone was counted (B). Unbound hormone (U) was calculated as the difference between T and B. The evolution of B as a function of U was analyzed by a computer method previously described [13]. In each case increasingly complex models of interaction were tested: one class of specific sites (B = NU/(K, + U)), one class of specific sites + non-specific binding (B = NlJ/(K, + IJ) + /?U) and two classes of specific sites (B = N,U/(K,, + U) + N,U(K,, + U)). Comparison of the least square criteria with tabulated F values for an CI significance level was the basis of model rejection [ 131. Distribution of [3H]aldosterone binding along the gut

Binding assays were performed using either various segments of colon and terminal ileum from different patients or various fragments from the same patient. 100 ~1 aliquots of the cytosolic fraction, prepared as

described for the time course study, were incubated with 20 nmol/l [3H]aldosterone at 0°C for 3 h. Bound aldosterone was measured after charcoal-dextran separation. For each assay, non-specific binding was determined by parallel incubation with a lOOO-fold excess of unlabeled hormone. Competition experiments

100~1 aliquots of the cytosolic fraction of the cytosolic fractions prepared from various colonic segments and ileum were incubated with 20 nmol/l [3H]aldosterone in the presence or in the absence of increasing concentrations of competitors. After 3 h incubation at O”C, bound and free hormone were separated by the charcoal-dextran technique. Nonspecific binding was determined by parallel incubation with a lOOO-fold excess of unlabeled aldosterone. Radioactivity measurements andprotein determination

Aqueous samples were counted in a Tri-Carb 3380 scintillation spectrophotometer after addition of 5 ml of Pica-fluor 15 (Packard). Proteins were determined by the method of Lowry [14]. RESULTS

Aldosterone stability

binding kinetics at 0°C and receptor

The time course study showed that [3H]aldosterone binding to the cytosolic fraction prepared from epithelial cells reached a plateau after 3 h. Thereafter the aldosterone binding remained unchanged up to 6 h. All subsequent incubations of [3H]aldosterone with the cytosolic fraction at equilibrium were thus performed for a 3 h period. After 2 h in the absence of hormone and a 3 h incubation in the presence of [‘Hlaldosterone the binding capacity of the cytosol was the same as that observed when cytosol was immediately incubated with [3H]aldosterone. When the incubation was performed after a 4 h delay a 30% decrease of the binding capacity was observed. [3H]aldosterone binding characteristics at equilibrium

Three experiments were performed with the following segments: ascending colon (patient No. 2), caecum (No. 5) and ileum (No 7). Table 1 indicates the binding parameter values calculated for 3 models of interaction. As indicated by the F test (see Table 1), model 2 (one class of specific sites and nonspecific binding) best described the experimental data in all cases. Figure 1 gives an example of the fitting of the experimental data by the curve simulated from model 2. Distribution of the [3H]aldosterone binding along the gut

As shown in Table 2, specific [3H]aldosterone binding was found in all the segments of the colon

313

Aldosterone receptors in human gut Table 1. [“Hlaldosterone binding parameter values in the epithelial cells of human gut mucosa Origin of cells

N, Model (fmol/mg protein)

Kdl (nmol/l)

NII (fmol/mg protein)

B

Ascending Colon (Patient 2)

Ml M2 M3

110*7 76 f 7 66 f 2

5.89 f 0.61 3.15 + 0.47 3.52 f 0.73

0.004 f 0.0009

Caecum (Patient 5)

Ml M2 M3

15a+4 142*5 142f8

2.26 f 0.40 1.95kO.12 1.94kO.16

0.0022 * o.oOfl3

Terminal Ileum (Patient 7)

Ml M2 M3

134* 10 102 * 10 102* 13

5.15*0.59 3.00 * 0.41 3.00 k 0.52

0.0031 * 0.0005

181 f 316

418 f 5200

Kdll (nmolil)

C

a0

203 i 543

0.388 0.193 0.192

>O.OOl NS

0.119 0.074 0.074

>O.Ol NS

0.640 0.290 0.290

>O.OOl NS

1070 f 14600

36530 f 6769

30040 f 5600

Cytosolic fractions prepared from various gut segments were incubated with increasing concentrations of [‘Hlaldosterone for 3 h at 0°C. At the end of incubation, bound and free hormones were separated by the charcoal technique. Binding parameter values and their confidence limits were calculated from three models: M, , with one class of specific binding sites, M,, with one class of specific sites and non-specific binding, and M, with two classes of specific sites. The minimized criterion (C) was compared with the F value tabulated for an CIsignificance level.

(sigmoid, descending, transverse, ascending and caecum) as well as in the ileum. For each segment, the binding levels were observed to be very closed in different patients. When assays were performed with various segments from the same patient (Nos 5 and 10) binding levels were in a very close range in the sigmoid, descending and transverse segments and in the caecum, but were lower in the ascending segment and ileum (Table 2). Competition experiments

Since only one class of binding sites was detected in the human gut, competition experiments were performed with various competitors which are known to bind preferentially to either type I or type II sites. The results are presented in Fig. 2. SC 26304, which is known to bind preferentially to type I sites [ 151,was as potent as aldosterone to compete in competition for [3H]aldosterone binding sites whereas RU 26988, a highly specific marker of type II sites [ 161displayed a low affinity for the sites labeled with [‘Hlaldosterone. These results are in good agreement with the conclusion of the steady-state analysis

(Table 1). In contrast to RU 26988, dexamethasone was as potent as aldosterone and SC 26304 to compete for [rH]aldosterone binding sites. As shown in Table 3 SC 26304 was as potent as aldosterone to compete for [rH]aldoesterone binding sites in all the gut segments investigated. In contrast, estradiol, dihydrotestosterone and RU 26988 displayed a low affinity for the sites labeled with [‘Hlaldosterone. The order of potency of the various steroids for [3H)aldosterone binding sites was as follows A = SC 26304~E > DHT = RU 26988. When experiments could be performed in different segments from the same patient (Nos 5 and lo), [3H]aldosterone binding specificity was the same for different segments of the gut. DISCUSSION

The results reported in this paper demonstrate the presence of specific aldosterone receptors in the cyto-

Table 2. Specific binding of [rH]aldosterone human gut mucosa Origin of cells Sigmoid colon

I. 0.15

=

‘0

I\ 2

0.10

al

0.05c

.

140 129 180 114

Descending colon

(5) (6) (8) (10)

109 168 122 119

Transverse colon

(9) (10)

168 151

Ascending colon

(2) (5)

65 49

(5) (10)

110 164

. .

1 ’ >w-._. +v

0.5

I

Caecum

B(nM)

Fig. 1. [‘Hlaldosterone binding to epithelial cells isolated from human ascending colon. Cytosolic fractions prepared from epithelial cells of ascending colon were incubated with increasing concentrations of [3H]aldosterone for 3 h at 0°C. Bound (B) and free hormone (U) were separated by the charcoal technique. The curve was simulated from the parameter values listed in Table 1 for model 2.

Specific [‘Hlaldostcrone binding (fmol/mg protein)

(1) (3) (4) (5)

. h

Patient no.

Ileum

to epithelial cells of

(141 f 14)

(123 f 13)

81 (7) 51 (10) Aliquots of the cytosolic fraction prepared from cpithelial cells from various gut segments were incubated with 20nmoljl [‘Hlaldosterone for 3 h at 0°C. At the end of incubation, bound and free hormones were separated by the charcoal technique. Parallel incubations with lOOO-fold excess of unlabeled aldosterone were performed to determine non-specific binding.

M. E. RAFESTIN-OBLIN et al.

314

E

-k 0 4 f

CI

Molar rotlo

I 5 20 100

I

5 20100

SC 26304

I 5 20100 Oexomeihoxne

Ill I 5 20100 RU 26988

Fig. 2. Effect of unlabeled steroid on [‘Hlaldosterone binding to epithelial cells isolated from human descending colon. Cytosolic fractions prepared from epithelial cells of descending colon were incubated at 0°C for 3 H with 5 nmol/l [‘Hlaldosterone in the presence or absence of increasing concentrations of unlabeled aldosterone, dexamethasone and RU 26988 (from l-100 fold the concentration of [‘Hlaldosterone). Bound and free steroids were separated by the charcoal technique. Results are expressed as the percentage of specific binding measured in the absence of unlabeled competitor. They are means of triplicate determination in as single experiment.

solic fraction of epithelial cells isolated from the human colon and ileum. Unlike observations in rat kidney [17], the aldosterone binding sites of human gut were relatively stable, since after 4 h at 0°C in the absence of aldosterone the capacity of the cytosolic fraction to bind this hormone was not less than 70% of the capacity observed at time 0. The ascending colon, caecum and terminal ileum were all shown to contain only one class of specific binding sites. The dissociation constants of aldosterone for these sites were very close in the three segments, and the number of sites ranged from 70 to 140 fmol per mg protein. Estimation of the residual endogenous hormones bound to the corticoid receptors after the cell isolation was not possible. Thus the data presented in this paper concerned free receptors. Nevertheless, it seems likely that a large proportion of the endogenous hormones were already eliminated during the early step of the cell preparation which involved several washes at 20°C. At this temperature the dissociation rate of aldosterone is known to be very

rapid. The detection of only one class of specific sites in the gut epithelial cells was unexpected, inasmuch as two classes of specific receptors for aldosterone were always detected in the cytosol of aldosterone mammalian target tissues [6]: type I sites, with dissociation constant ranging from 0.5 to 3 nmol/l, and type II sites, with dissociation constant of about 50 nmol/l, the maximum number of type II sites being 10 times that of type I sites. In the present work, the dissociation constant of aldosterone for the cytosolic sites of human colon and terminal ileum was found to be in the same range as that estimated for type I sites in other target tissues. This suggests that in human colon, aldosterone binding sites are essentially of the mineralocorticoid type. This hypothesis is in accordance with the results of the present competition studies, in which aldosterone binding was observed to be completely inhibited by SC 26304, a spirolactone with high affinity for aidosterone type I receptors [15], but slightly displaced by RU 26988, a synthetic steroid with high affinity for type II and very low affinity for type I receptors [16].

Table 3. Percentage of specific [‘Hlaldosterone binding to epithelial cells of human gut mucosa in the presence of comoetitors Origin of cells

Patient No.

SC 26304

Estradiol

RU 26988

(4) (5)

7.8 0

36.4 50.7

58.4 53.3

61.9 60.5

Descending

(5) (8) (10)

2.5 0.3 0.5

42.9 30. I 36

57.3 58.7 51.4

47.8 56.1 44.0

Transverse

(10)

0.4

30.4

55.4

52.0

Ascending

(5)

2.8

37

55.3

54.4

Caecum

(5) (10)

3.2 0.4

38.9 28.4

59.7 54.0

47.4 48.5

Ileum

(7) 1101

0 0

47.3 30.5

44.7 35.7

44.4 38.5

Sigmoid

Dihvdrotestosterone

Aliquots of the cytosolic fractions prepared from epithelial cells from various colonic segments and from ileum were incubated with 20 nmol/l [3H]aldosterone for 3 h at 0°C with or without a IOO-fold excess of unlabeled steroid. At the end of incubation, bound and free hormones were separated by the charcoal technique. Results are expressed as the percentage of specific [‘Hlaldosterone binding measured in the absence of unlabeled competitor.

Aldosterone

receptors

Since in human gut dexamethasone appeared to be as potent as aldosterone to inhibit [3H]aldosterone binding it is likely that dexamethasone and aldosterone bind to the same receptors suggesting that in this organ dexamethasone could act as a mineralocorticoid. This is not really surprising since it has already been described in the rat [8] and in the toad

in human

5.

P81. In view of the small inter-patient variations in the amount of aldosterone binding for a same segment two additional observations can be made: (1) the amount of binding is probably the same in the sigmoid, descending and transverse colon and in the caecum, and (2) the level of aldosterone binding seems lower in the ascending colon and terminal ileum than in the other segments explored. This variation in receptor levels along the colonic axis displayed the same trend as the variations in the transmucosal potential difference observed by RaskMadsen, who showed an increase in the potential difference from the caecum to the anus [ 111.Edmonds also demonstrated that the potential difference observed in the sodium-depleted rat was larger in the descending colon and caecum than in the ascending colon [9]. Acknowledgement-The authors wish to thank Professors Alexandre, Huguier and Malafosse from departments of surgery of Broussais, Tenon and Rothschild Hospitals for providing colon specimens.

10.

Il.

12.

13.

14. REFERENCES 1. Edmonds C. J. and Marriott J.: The effect of aldosterone and adrenalectomy on the electrical potential difference of rat colon and on the transport of sodium, potassium, chloride and bicarbonate. J. Endocr. 39 (1967) 517-531. 2. Edmonds C. J. and Marriott J.: Sodium transport and short circuit current in rat colon in oiuo and the effect of aldosterone. J. Physiol. 210 (1970) 1021-1039. 3. Levitan R. and Ingeltinger F. J.: Effect of d-aldosterone on salt and water absorption from the intact human colon. J. clin. Invest. 44 (1965) 801-808. 4. Efstratopoulos A. D., Peart W. S. and Wilson G. A.: The effect of aldosterone on colonic potential difference and

15.

16.

17.

18.

gut

315

renal electrolyte excretion in normal man. Clin. Sci. 46 (1974)489-499. Edmonds C. J. and Godfrey R. C.: Measurement of electrical potentials of the human rectum and pelvic colon in normal and aldosterone-treated patients. Gur 11 (1970) 330-337. Feldman D. J., Funder J. W. and Edelman I. S.: Subcellular mechanisms in the action of adrenal steroids. Am. J. Med. 53, (1972) 545-560. Pressley L. and Funder J. W.: Glucocorticoid and mineralocorticoid receptors in gut mucosa. Endocrinology 97, (1975) 588-596. Marusic E. T., Hayslett J. P. and Binder H. J.: Corticosteroid binding studies in cytosol of colonic mucosa of the rat. Am. J. Physiol. 240 (1981) G417-423. Edmonds C. J.: The gradient of electrical potential difference and of sodium and potassium of the gut contents along the cecum and colon of normal and sodium depleted rats. J. Physiol. 193 (1967) 571-588. Fromm M. and Hegel U.: Segmental heterogeneity of epithehal transport in rat large intestine. Pjlugers Archs 378 (1978) 71-83. Rask-Madsen J.: Simultaneous measurement of electrical polarization and electrolyte transport by the entire normal and inflamed human colon during in viuo perfusion. Stand. J. Gastroent. 8 (1973) 327-336. DuPont C., Laburthe M., Broyart J. P., Bataille D. and Rossehn G.: Cyclic AMP production in isolated colonic eoithelial crypts: a highly sensitive model for the evalu&on of vasoactive intestinal peptide action in human intestine. Eur. J. clin. Invest. 10 (1980) 67-76. Claire M., Rafestin-Oblin M. E.,. Michaud A., Corvol P., Venot A., Roth-Meyer C., Boisvieux J. F. and Mallet A.: Statistical tests of models and computerized parameter estimation for aldosterone binding in rat kidney. FEBS Letters 88 (1978) 285-299. Lowry 0. H., Rosebrough N. J., Farr A. L. and Randall R. J.: Protein measurement with Folin phenol reagent. J. biol. Chem. 193 (1951) 265-275. Marver D.. Stewart J.. Funder J. W., Feldman D. and Edelman L’S: Renal aidosterone receptors: studies with [‘Hlaldosterone and anti-mineralocorticoid the [3H]spirolactone (SC 26304). Proc. natn. Acad. Sci. U.S.A. 71 (1974) 1431-1435. Moguilewsky M. and Raynaud J. P.: Evidence for a specific mineralocorticoid receptor in rat pituitary and brain J. steroid. Biochem. 12 (1980) 309-314. Rafestin-Oblin M. E., Michaud A., Claire M. and Corvol P.: Dramatic protectic effect of ligand against thermal degradation on mineral0 and glucocorticoid receptors. J. steroid Biochem. 8 (1977) 19-23. Rossier B.: Unpublished data.