In vitro influence of plasma steroid-binding proteins on androgen metabolism in human leukocytes

I'~UTTERWQRTH I~E I N E M A N N

In vitro influence of plasma steroid-binding proteins on androgen metabolism in human leukocytes Henri D6chaud,* Raymond Goujon,* Francine Claustrat,* Michel Boucherat,* and Michel Pugeatt'$ Hospices Civils de Lyon: *Laboratoire Central de Biochimie and fLaboratoire de la Clinique Endocrinologique, HOpital de l'Antiquaille, Lyon; $1NSERM U329, H @ i t a l Debrousse, Lyon, France

The purpose of this study was to investigate the influence of plasma steroid-binding proteins on androgen metabolism in intact leukocytes prepared from normal male and female blood samples. Leukocyte preparations were incubated for 24 h at 37°C with either labeled or unlabeled testosterone (T), 5et-dihydrotestosterone (5ct-DHT), and androstenedione (A). After extraction, the formed labeled metabolites were first identified by high performance liquid chromatography, then, using unlabeled substrates, metabolite concentrations were measured by specific radioimmunoassays. The conversion ratios of substrate to metabolite were calculated for each preparation using either labeled or unlabeled substrates. In the absence of steroid-binding proteins, the mean conversion ratios ofT to A, A to T, T to 5ct-DHT, and 5a-DHT to 3ct-androstanediol (3et-D) were, in males and females, respectively, 5.6% and 6.•% (n = 11), 5.6% and5.6% (n = 5), 2.8% and2.2% (n = 11), 43.1% and 40.0% (n = 5), these sex differences being non-significant. The presence of increasing amounts of plasma, purified albumin or sex hormone binding-globulin (SHBG ) in the incubation media reduced metabolite formation dose-dependently. However, a lO00-fold greater concentration of albumin than of SHBG was necessary for 50% inhibition of androgen metabolism by leukocytes, showing SHBG to have the main protective effect. Moreover, in the presence of various concentrations of T or 5et-DHT, and of albumin or SHBG, metabolite formation was positively and highly significantly correlated with the concentration of protein-unbound (free) substrate measured by equilibrium dialysis (r = 0.964 for conversion of T to A and r = 0.998 for conversion of 5ct-DHT to 3ct-D) but weakly correlated with total substrate concentration (r = 0.375 for total T and r = 0.669for total 5et-DHT). Additionally, leukocyte metabolism of 5ct-DHT was significantly enhanced when free 5ct-DHT levels increased in the presence of l7~-estradiol (which displaced 5ct-DHT from the SHBG-binding sites). We conclude that the metabolism of androgens by human leukocytes is mainly regulated by SHBG level, in both men and women. Although the rate of androgen metabolism by leukocytes was not determined in this study, reduced serum SHBG levels with unchanged albumin levels or drugs capable of displacing androgens from serum SHBG can be expected to increase androgen metabolism in human leukocytes. (Steroids 60:226-233, 1995)

Keywords: androgenmetabolism;steroid-bindingproteins; sex hormone binding-globulin (SHBG); 5a-reductase; leukocytes

Introduction The ability of human leukocytes to metabolize steroid androgens has been reported by some authors~'2 who have This work was partly presented at the Ninth International Congress of Endocrinology, Nice, France, 1992, Abstract P 13.03.008. Address reprint requests to Henri l~ehand, Laboratoire Central de Biochimie, H6pitalde l'Antiquaille, 1 rue de I'Antiquaille, 69321 LyonCedex 05, France. Received June 12, 1994; accepted August 19, 1994 Steroids 60:226-233, 1995 © 1995 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010

identified 1713-hydroxysteroid o x i d o r e d u c t a s e (1713HSOR), 5ot-reductase (5or-R) and 3ct- and 313-hydroxysteroid oxidoreductase (3or- and 313-HSOR) enzyme activities. Androgen metabolism was also investigated in others human cells including alveolar m a c r o p h a g e s , 3 pubic skin, 4'5 and cancer cell lines derived from human endometrium and prostate. 6 Labeled substrates in combination with various chromatographic techniques and/or crystallization have been employed to characterize the metabolites. 1-3 Most studies have measured the metabolism of androgens within a syn-

0039-128X/95/$10.00 SSDI 0039-128X(94)001M2-B

In vitro influence of plasma steroid-binding proteins on androgen: Ddchaud et aL thetic medium, i.e., in absence of steroid-binding proteins. However, the binding proteins should have a major influence on the metabolism of androgens since it is generally admitted that only the protein unbound (free) fraction of androgens is available for the ceUs. 7-11 The purpose of the present study was to quantify the metabolism o f androgens by intact leukocytes, and to investigate the influence o f albumin and sex hormone binding-globulin (SHBG) on this metabolism. Since 5ot-R activity has been reported to be sex-linked, 2 we also compared the androgen metabolism of leukocytes obtained from normal males and females.

Experimental

Chemicals [1,2,6,7-3H]Androstenedione ([3H]A, specific activity [SA]: 81.4 Ci/mmol), [1,2-3H]5et-androstan-3ct,1713-diol ([3H]3ct-D, SA 46.5 Ci/mmol) and [1,2-3H]5ct-androstan-3[3,1713-diol ([3H]313D, SA 60.0 Ci/mmol) were purchased from New England Nuclear Corp. (Boston, MA, USA). [1,2,6,7-3H]Testosterone ([3H]T, SA 87.0 Ci/mmol) and [1,2-3H]5ct-dihydrotestosterone ([3H]5aDHT, SA 47.0 Ci/mmol) were obtained from Amersham France SA (Les Ulis, France) and purified by HPLC in our laboratory. All reference steroids (androstenedione, testosterone, 5ctdihydrotestosterone, androsterone, and isoandrosterone) were purchased from Sigma Chemical Co. (Saint Louis, MO, USA). The anti-testosterone and anti-androstenedione rabbit antibodies used for immunoassays were raised in our laboratory. The 5a-DHT antiserum was purchased from Radioassay Systems Laboratories (Carson, CA, USA), and the anti-5ct-androstan-3ot, 1713-diol from Biosys (Compi6gne, France). Human albumin solution (200 g/L) was purchased from the Centre de Transfusion Sanguine de Lyon (Beynost, France). The concentration of SHBG (<0.03 nmol/L) in this solution was estimated by an immunoradiometric assay (~25ISBP-Coat RIA kit, bioMerieux, Marcy l'Etoile, France). Immunopurified SHBG was kindly provided by Catherine Grenot of INSERM U329 (Lyon, France). 12 The control of the purity was performed by a native polyacrylamide gel electrophoresis (PAGE) and sodium dodecylsulfate-PAGE. SHBG-enriched plasma was obtained by the ammonium sulfate precipitation method previously described. ~3 The concentration of albumin in this fraction is less than 1.0 g/L.

Preparation of leukocytes Blood samples were obtained from normal healthy men and women, between 20 and 40 years old, with normal androgenic profiles; the women were non-menopausal and withdrawn from testing during the follicular phase. Venous blood (5-20 mL) was collected in glass vacutainer tubes containing EDTA (Beckton Dickinson Co., Pont de Claix, France) and centrifuged at 1500 × g for 5 min. The leukocyte layer was collected, washed twice with PBS buffer, and suspended in 5 mL of a lysing solution containing NH4C1 (8.29 g/L), Na2HPO4 (1.78 g/L), NaH2PO 4 (0.312 g/L), and EDTA (37.2 mg/L) (pH = 7.2). After twenty minutes, the residual erythrocyte lysis was achieved and the leukocytes were centrifuged for 10 min at 500 × g, washed twice with PBS buffer, then resuspended in RPMI 1640 (Gibco, Eraghy, France) and counted on a Coulter Counter (Coultronics, Margency, France). Assays were conducted so as to give a final concentration of 5 × 106 or 107 leukocytes/mL.

Incubation procedure For the characterization of formed labeled metabolites, leukocyte suspensions in RPMI (1 mL) were incubated in glass tubes and

placed in a shaking bath at 37°(2 for 24 h with labeled substrates [3H]A, [3H]T, [3H]5ot-DHT (300, 000 dpm ~ 3 pmol). At the end of the incubation period, reactions were stopped by storage at - 30°C. Tubes with boiled leukocytes were assayed as blank controis. In the same conditions but using unlabeled substrates from 0-3000 pmol/mL, metabolites were quantified and the relationship between the amounts of formed metabolite and substrate concentration was studied. Because of the linearity of this relationship (see Results section), a substrate concentration of 700 pmol/mL was fixed for the comparative studies of conversion ratios for healthy men and women. The influence of steroid-binding proteins on androgen metabolism was studied by addition either of increasing volumes of normal plasma or of increasing amounts of purified albumin or SHBG-enriched plasma. The influence of protein-unbound substrates on androgen metabolism was studied by addition of various amounts of T or 5etDHT and of SHBG or albumin so as to produce different concentrations of free-T and free-5ct-DHT, which were checked by equilibrium dialysis.14 The effects of displacing 5et-DHT from SHBG binding sites were studied by incubating whole blood (200 p,L) containing approximately 106 leukocytes, 109 erythrocytes, 6 pmol of SHBG, and 120 nmol of albumin with [3H]5et-DHT (200,000 dpm --~ 2 pmol) and increasing amounts of 1713-estradiol. In this study, both leukocytes and erythrocytes are responsible for the 5ot-DHT metabolism observed, since erythrocytes also have 3ct- and 3[~-HSOR activities (data not shown).

Androgen metabolite analysis Method with labeled steroids. After addition of 5 mL of diethylether to extract the steroids, the tubes were shaken for 5 rain and the organic phase was transferred and evaporated to dryness under a stream of nitrogen. The residue obtained was dissolved in 200 p,L of mobile phase and metabolites were separated by high perforrnance liquid chromatography (HPLC) using a C18-5 Ixm reversed phase column. The mobile phase consisted of tetrahydrofuran:methanol:water (22:10:68 vol/vol/vol) flowing isocratically at 0.8 mL/min. Fractions containing radioactivity were collected every 15 seconds and the activity was measured in a Packard liquid scintillation spectrometer. Metabolite formation is indicated by the presence of radiopeaks coinciding with corresponding reference steroid retention times.

Method with unlabeled steroids. To confirm and quantify metabolite formation, unlabeled steroids were incubated with leukocytes. After incubation, 3H-steroids, corresponding to the formed metabolites, were added to measure the recovery rates. Specific radioimmunoassays of T, A, 5et-DHT, and 3a-D were then performed in duplicate after diethylether extraction and appropriate separation (celite column chromatography for T and A,~3 HPLC for 5a-DHT, and HPLC plus celite column chromatography for 3a-D). The following antiserum cross-reactivities were found: for A-antiserum with 5ct-androstanedione, 13.5%; with T, 0.93%; with androsterone, 0.22%; with DHEA 0.07%; with 5a-DHT, 0.04%; and with 3a-D, 0.006%; for T-antiserum with 5ct-DHT, 58%; with A, 0.2%; with 3a-D, 0.16%; with 5a-androstanedione, 0.08%; with androsterone, 0.002%; and with DHEA, 0.002%; for 5c~-DHT-antiserum with T, 22.7%; with 5a-androstanedione, 17.1%; with A, 2.4%;with 3ct-D, 0.06%; with androsterone, 0.04%; and with DHEA, <0.01%; for 3a-D-antiserum with androsterone, 10.6%; with 5a-DHT, 0.25%; with T, 0.12%; with 313-D, 0.10%; with A, 0.10%; and with DHEA, <0.10%. Bound/ free separation was performed with dextran-coated charcoal. The bound fraction was analysed in a scintillation counter. Interassay

Steroids, 1995, vol. 60, February

227

Papers variation coefficients for T, A, 5a-DHT, and 3et-D were less than 15%.

Results

Characterization and quantification of androgen formed metabolites in leukocytes

Expression of results The metabolite amounts formed during the incubation period were

After incubation of human leukocytes with [3H]A or [3H]T, radiochromatograms C 1 and C2 (Figure 1) indicated radioactivity peaks identified respectively as T and A. Such peaks were not found with boiled leukocytes (control incubation). The amounts of formed A or T were proportional to those of T or A from a concentration of 0 - 3 0 0 0 pmol/mL, yielding the following equations: y = 0.0869x - 0.101 (r = 0.998) for T, and y = 0.0374x + 1.308 (r = 0.999) for A. After incubation with 700 pmol, the values for T for-

expressed in pmol/incubation; conversion ratios, i.e., ratios of formed metabolites to substrate, were expressed as percentages (CR%) of labeled and unlabeled substrates.

Statistical analysis Results are given as mean -+ SD. For statistical analysis of data the Student's t-test was used; a P value <0.05 was considered significant. ,20000

CI

[SH] A : ASSAY

C2

120000

[SH] T : ASSAY

,2oo00

C3

[3H] 5a-DHT : ASSAY

T A

100000

100(300

locooo

Z =o LM

i

80000

80000

80000

60000

60000

60000

t-,,~ Y I'--.

E

-8"

"140000

T

40000

A

40000

5a-DHT

6

8

10

12

14

8

12

16

20

24

~,

I 1

20000

20000

20000

6

'= 03

13

28

15

17

19

21

23

25

Retention time (min.)

120000

C' I

[3H] A : CONTROL

C' 2

f200go

[3H] T : CONTROL

1~0000 C'3

[3H] 5a-DHT : CONTROL

] ii i

100000

100000 1

100000

!

i' 80000

r 80~o 1 !

80000

60000

60000

60000

,~CO0

40000 t

¢

.2 E

-g-

40000

. i

20000

20000

6

8

10

12

14

I

li i

E ,

2~00

O.L 4

8

12

-

16

2D

,= 24

GI 28

It

13

t5

7

19

21

23

"25

Reim~on time (min.) Figure 1 Radiochromatograms of the metabolites for A, T, and 5(~-DHT labeled substrates. Leukocyte suspensions (107/mL) were incubated in RPMI at 37°C with labeled substrates (300,000 dpm ~ 3 pmol). After 24 h incubation, the metabolites were extracted and separated by HPLC. Radiochromatograms obtained from [all]A, [3H]T, [3H]5(x-DHT were respectively C1, C2, C3 with intact leukocytes, C'1, C'2, and C'3 with boiled leukocytes.

228

Steroids, 1995, vol. 60, February

In vitro influence of plasma steroid-binding proteins on androgen: Ddchaud et aL

Influence of steroid-binding proteins on androgen leukocyte metabolism

marion from A were 38.9 --- 9.4 pmol/incubation (CR% = 5.6 - 1.3) in healthy men (n = 5) and 39.3 ± 4.6 pmol/ incubation (CR% = 5.6 --- 0.7) in healthy women (n = 5). The formation rate of A from T in healthy men (n = 11) and women (n = 11) was respectively 39.4 --- 8.0 pmol/ incubation (CR% = 5.6 ± 1.1) and 42.6 ± 9.1 pmol/ incubation (CR% = 6.1 ± 1.3). Thus, the mean conversion ratios obtained for unlabeled or labeled substrates were in good agreement (Table 1). Incubation with [3H]T showed also radioactivity peaks (radiochromatogram C2 in Figure 1) identified respectively as 5ot-DHT and 3ot-D. The amounts of formed 5ot-DHT were proportional to the amount of T incubated in each assay, for the range of 0-3000 pmol (y = 0.0165x 0.528; r = 0.998). 5a-DHT formation, measured after incubation of 700 pmol of unlabeled T, was 15.3 ± 5.2 pmol/incubation (CR% = 2.2 --- 0.7) in normal females (n = 11) and 19.6 ± 6.1 pmol/incubation (CR% = 2.8 ± 0.9) in normal males (n = 11), the difference being nonsignificant. The formation of 3ot-D from T as substrate was 9.9 --- 3.7 and 9.5 ± 2.7 pmol/incubation in females and males respectively. The conversion ratio of T to 50~-DHT was higher with [3H]T (7.6%) than with unlabeled T (2.5%; Table 1). This difference may have resulted from the presence of a nonidentified metabolite with a retention time identical to that of 5a-DHT. Thus, the addition of a large excess of unlabeled progesterone (10 p,g) during the incubation suppressed the conversion of T to 5ot-DHT and to 3or-D, and showed a residual peak after HPLC, suggesting the formation of a non 5or-reduced compound. The conversion ratio after subtraction of this residual peak was 3.0%, which fits with the 2.5% obtained for incubation with unlabeled T. Incubation with [3H]5et-DHT showed radioactivity peaks (radiochromatogram C3 in Figure 1) identified respectively as 3ot-D and 313-D. The formation of 3ot-D was linear from 0 to 3000 pmol/mL of 5ot-DHT (y = 0.506x - 2.742; r = 0.999). Since a 313-D radioimmunoassay was not available and the formation of 5ot-androstanedione from 5ot-DHT being low ( < 1.5%; data not shown), the formation of 3oc-D + 313-D, evaluated by measuring the concentration of nonmetabolized 5et-DHT, was linear from 0 to 3000 pmol/mL of 5ot-DHT (y = 0.6969x - 5.659; r = 0.999). The 3a-D formation after incubation of 700 pmol of 5a-DHT was 280.0 --- 19.7 pmol/incubation (CR% = 40.0 ± 2.8) and 301.5 ± 28.2 pmol/incubation (CR% = 43.1 - 4.0) respectively in healthy females (n = 10) and males (n = 10). The mean conversion ratios obtained with unlabeled and labeled substrates were not significantly different (Table 1).

Plasma. The addition of increasing volumes of plasma from a normal subject to the incubation of leukocytes with [3H]5ot-DHT decreased metabolite formation as a function of increasing albumin and SHBG concentrations in the incubation media (Figure 2).

SHBG in plasma. The addition of a constant volume of plasma (30 p,L) with increasing concentrations of SHBG (9, 47, 102, 156, and 234 nmol/L) but similar albumin concentrations (39 to 41 g/L) to the leukocyte preparation incubated with [3H]5ot-DHT decreased the metabolite formation in a dose-dependent relation to SHBG (Figure 3). Purified albumin and SHBG-enriehed plasma. The addition of increasing amounts of purified albumin or SHBGenriched plasma to leukocytes which were incubated with unlabeled T decreased A formation dose-dependently. However, the concentration of albumin necessary for 50% inhibition of the metabolite formation of T was approximately 1000-fold greater than that for SHBG (Figure 4).

Importance of protein unbound substrates With 5 × 106 leukocytes and in absence of steroid-binding proteins, the amount of formed A was proportional to total T (y = 0.035x - 0.043, r = 0.964, P < 0.001). Conversely, in presence of SHBG-enriched plasma or purified albumin, the A formation was independent of total T concentration (y = 0.006x + 1.674; r = 0.375, not significant), but correlated to the concentration of proteinunbound (free) T as measured by equilibrium dialysis (y = 0.024x + 0.817, r = 0.996 with SHBG enriched plasma; y = 0.019x - 0.815, r = 0.937 with albumin; (Figure 5a). In presence of immunopurified SHBG, 3ot-D formation with 5ot-DHT as substrate was weakly related to the total concentration of 5ot-DHT incubated with leukocytes, but significantly related to the concentration of protein-unbound 5cx-DHT (r = 0.669 versus r = 0.998; Figure 5b).

Metabolic effect of displacing 5a-DHT from SHBG by 17f3-estradiol in blood In whole blood incubated with 5 a - D H T , increasing amounts of 1713-estradiol increased the concentration of protein-unbound 5o~-DHT dose-dependently and resulted in increased 3or- and 313-D metabolite formation (Figure 6).

Table 1 Mean conversion ratios of androstenedione (A), testosterone (T), and 5~-dihydrotestosterone (5~-DHT) with unlabeled and labeled substrates Mean conversion ratios (%)

A--*T Unlabeled substrates Labeled substrates

5.6 (n 6.4 (n

± = ± =

1.0 10) 0.9 4)

T--*A 5.9 (n 7.6 (n

± = ± =

1.2 22) 0.8 4)

T--->5~-DHT 5ct-DHT-*3~-DIOL 2.5 (n 7.6 (n

± = ± =

0.8 22) 2.0 4)

41.5 (n 42.3 (n

Steroids,

± = ± =

3.4 10) 7.1 4)

1995, vol. 60, February

229

Papers ~" 5.55

100°

°O O

80.

t

~

.S

o

0

.EE~3.70

' 60-

C O .=O

O

~ 4

.|

0

E O

~

O

" 1.85 ._o

20

-'a o e-

o 0

0

50

2OO

500

950

Volume of added plasma.(pl}

"0 c

< 10~

0

1.75

7.0

17.5

10 2

103

104

I 0~

I 06

SHBG l-e-) or Alb. (-C~)concentrations (nmol/I)

33.2

[SHBGI (pmol/ml) 0

2.86 10 .5

1.14 '~0~

2.86 1 0 '

5.43 10 "~

[AIb] (mmol/ml) Figure 2 Influence of increasing volumes of plasma in the incubation medium on the [3H]5~-DHT metabolism by leukocytes. Leukocytes (107) were incubated in RPMI, for 24 h at 37°C with [3H]5~-DHT (300,000 dpm = 3 pmol) and increasing volumes of normal plasma. Final volume of the leukocyte suspension: 1 mL. [3H]3Q-D + [3H]3~-D formed metabolites were measured after diethylether extraction and HPLC. 100 .,lI,,-

d 8o y = - 0 . 2 6 x + 77.6

"6 6o

r = 0.994

._o

~ 40 ._o r,

~ 2o

0 U

0

o

5'o

]bo

l~o

2bo

2io

[SHBG] (nmol/l)

Figure 3 Influence of a constant volume of plasma with increasing concentrations of sex hormone binding-globulin (SHBG) and normal albumin concentrations (39-41 g/L) on the [SH]5~-DHT metabolism by human leukocytes. Leukocytes (107) were incubated in RPMI, for 24 h at 37°C with [3H]5~-DHT (300,000 dpm = 3 pmol) and 30 I~L of plasma with increasing concentrations of SHBG but similar albumin concentration. Final volume of the leukocyte suspension: 1 mL. [3H]3~-D + [SH]313-D formed metabolites were measured after diethylether extraction and HPLC.

Discussion

This paper describes a reliable method for assessing the metabolism of androgens in intact leukocytes. It also reports that the bioavailability of androgens for leukocyte metabolism is modulated by the binding of androgen substrate to albumin and to SHBG. Our methodology is based on the use of labeled steroids for identification of radiopeaks and

230

Steroids, 1995, vol. 60, February

Figure 4 Influence of purified albumin or SHBG-enriched plasma in the leukocyte incubation medium on A formation from T. Leukocytes (5 × 106) were incubated in RPMI, for 24 h at 37°C with 250 pmol of T and increasing amounts of purified albumin or SHBG-enriched plasma. Final volumes of leukocyte suspension = 1 mL. Formed A was measured by radioimmunoassay after diethylether extraction and celite column chromatography.

unlabeled steroids for specific radioimmunoassays of formed metabolites. A 24 h incubation period was used to study androgen metabolism. At the end of this time, there remain significant amounts of all the substrates used and enough metabolites formed to obtain accurate and reproducible measurements. In these conditions, comparative studies of androgen metabolism between men and women and the study of the influence of binding proteins on this metabolism can be undertaken. The significant rates of transformation of T to A, T to 5e-DHT, and 5et-DHT to 3et-D and 313-D suggest the presence in human leukocytes of 1713HSOR, 5et-R, 3e- and 313-HSOR activities. The data on metabolism are in agreement with a report by Milewich et al. 1 who studied androgen metabolism in human lymphocytes. On the other hand, in the absence of plasma, we cannot confirm the sex difference in the formation of 5eDHT from T reported by Clair et al. 2 Our results suggest that 5e-R activity in leukocytes may be the expression of a single gene, unlike the two 5et-R genes reported in the human prostate, 15 the leukocyte copy being the one not under the regulation of androgens. The study of the messenger RNAs is being undertaken in our laboratory to explore this hypothesis. Comparison of the conversion ratios of androgens in leukocytes ranked them in the following order of increasing magnitude: T to 5tx-DHT < A to T < T to A < 5e-DHT to 3oL-D and 313-D. These results suggest that leukocytes are more active in androgen catabolism than in peripheral conversion of androgenic hormones. Androgen metabolism is highly dependent on androgen availability. ~6-18 The data show that purified albumin or SHBG reduces T and 5et-DHT metabolism dose-dependently. However, the concentration of albumin necessary for a 50% decrease in T metaholite formation is 1000-fold greater than that for SHBG. This difference corresponds to the ratio of the affinity constants of SHBG and albumin for T binding. ~9 Moreover, in plasma with a constant concen-

In vitro influence of plasma steroid-binding proteins on androgen: Ddchaud et al.

80

20.

e-

T

r =, 0 . 3 7 5

=o

NS

r=0.669

P=0.003

0 .O

J J

= 60 u .E

1S =

I

0

E

i

10 I= m

0

I

0 m

•

o

<

IE zo

|

m= 0

0 200

o

400

600

800

Total T (pmol/ml)

2(30

0

600

800

Total 5=-OHT (pmol/ml)

.=!

- MEDIUM WITH ALBUMIN OR SHBG.

400

- MEDIUM WITH SHBG. -

50,

20.

r = 0.996

0

p < 0.001

40

0

._=

30

0 10 ¸ ZO

C 0 0

o

=o < m

0 0

10 - ~ , ~ =

0 ~00

300

200

400

500

600

0

200

4OO

6OO

80O

free 5a-DHT (pmol/ml)

free T (pmoJ/ml)

- MEDIUM WITH SHBG. -

- MEDIUM WITH SHBG. -

5b Z0.

r=0.937

0

p=0.002

Cl .Q 1S m

C 0

'

o < 0

,oo

zoo

300

400

soo

600

700

free T (pmol/ml) - MEDIUM WITH ALBUMIN. -

5a Figure 5

Importance of protein u n b o u n d substrates. Leukocytes (5 x 10e) were incubated with various concentrations of T or 5c=-DHT and a l b u m i n or SHBG. Final v o l u m e of leukocyte suspension: 1 mL. Free T and free 5e=-DHT concentrations w e r e evaluated by equilibrium dialysis. A and 3c~-D were measured by r a d i o i m m u n o a s s a y after diethylether extraction and c h r o m a t o g r a p h i c separation. Figure 5a: Relationships b e t w e e n A f o r m a t i o n and total or free T in presence of a l b u m i n or SHBG-enriched plasma in the incubation media. Figure 5b: Relationships between 3~-D f o r m a t i o n and total or free 5~-DHT in presence of i m m u n o p u r i f i e d SHBG.

Steroids, 1995, vol. 60, February

231

Papers 300

bumin and SHBG, leukocyte metabolism of androgens can be expected to be limited. This metabolism depends on the precursor availability, which is mainly regulated by the plasma concentration of SHBG, and, furthermore, can be modified by drugs which displace androgens from SHBG.

0

.=_

£

Acknowledgments

200-

.o

E

We are particularly indebted to Frank Z. Stanczyk and Jean Andr6 for their excellent critical review, and to lain McGill for revising the manuscript.

m,

References

0 0

~1= 1001.

÷

a m

2.

3.

o

0

Ii,

4 I0 -2

4 101

4

4 101

4 102

4 103

4.

pmol of estradiol added. Figure 6 Effect of increasing amounts of 1713-estradiol in the incubation medium on 5~-DHT metabolism in human blood. Whole blood (200 I~L), containing approximately 10e leukocytes, 109 erythrocytes, SHBG (6 pmol), and albumin (120 nmol), was incubated in RPMI, for 24 h at 37°C with [3H]5~-DHT (200,000 dpm = 2 pmol) and increasing amounts of 17~-estradiol (0-4.0 103 3pmol). Final volume of leukocyte suspension: 1 mL. [SH]3~-D + [ H]3~-D formed metabolites were measured after diethylether extraction and HPLC. Results were expressed in percentage increase compared to 3a-D + 313-Dformation in absence of 171~-estradiol.

5.

6.

7.

8. 9.

tration of albumin, the availability of 5ot-DHT for leukocyte metabolism is clearly related to the SHBG concentration as shown in Figure 3. This is in agreement with the relationship reported in normal men between non-SHBG bound T concentration and androstanediol-glucuronide, a peripheral marker of androgen metabolism, 2° and further evidences the predominant effect of SHBG on precursor availability to circulating leukocytes. Our experiments support the hypothesis that free hormone is the bioavailable fraction. Indeed, in presence of various concentrations of androgens and binding proteins in the leukocyte incubation media, we were able to show clearly that leukocyte metabolism of plasma androgens is proportional to the concentration of androgen unbound to SHBG and/or albumin, in agreement with several studies and theoretical models. 7,17,21 This finding was further supported by the increase in 5ot-DHT metabolism found when 5ct-DHT was displaced from SHBG sites by increasing concentrations of 1713-estradiol. In this particular experiment, the conversion ratios of 5ot-DHT to 3~t and 313 metabolites increased proportionally to the free fraction of 5ct-DHT (Figure 6). Moreover, the interaction of drugs with SHBG binding sites, in concentrations sufficient to displace T from SHBG, as a few studies have shown, 22'23 may also increase T metabolism in leukocytes. In conclusion, human leukocytes have enzymatic activities for metabolizing androgens. Because of binding to al-

232

Steroids, 1995, vol. 60, February

10. 11.

12.

13.

14.

15.

16.

17. 18.

Milewich L, Whisenant MG, Sawyer MK (1982). Androstenedione metabolism by human lymphocytes. J Steroid Biochem 16:81-85. Clair P, Patricot MC, Mathian B, Revol A (1984). Androgen metabolism in vitro by human leukocytes. Variations with sex and age. J Steroid Biochem 20:377-381. Milewich L, Kaimal V, Toews GB (1985). Androstenedione metabolism in human alveolar macrophages. J Clin Endocrinol Metab 56:920-924. Serafini P, Ablan FL, Lobo RA (1985). 5et-reductase activity in the genital skin of hirsute women. J Clin Endocrinol Metab 60:349355. Kutten F, Mauvais-Jarvis P (1975). Testosterone 5a-reduction in the skin of normal subjects and of patients with abnormal sex development. Acta Endocrinol 79:164-176. Castagnetta LA, Granata OM, Lo Casto ML, Calabro M, Arcuri F, Carruba G (1991). Simple approach to measure metabolic pathways of steroids in living cells. J Chromato 572:25-39. Lasnitzki I, Fanklin HR (1972). The influence of serum on uptake, conversion and action of testosterone in rat prostate glands in organ culture. J Endocrinol 54:333-342. Beaulieu EE (1975). Some aspects of the mechanism of action of steroid hormones. Mol Cell Biochem 7:157-174. De Ryck LMH, Ross JA, P6tra PH, Gurpide E (1985). Estradiol entry into endometrial cells in suspension. J Steroid Biochem 23: 145-152. Tait JF, Tait SAS (1991). The effect of plasma protein binding on the metabolism of steroid hormones. J Endocrinol 131:339-357. P6traPH (1991). The plasma sex steroid binding protein (SBP or SHBG). A critical review of recent developments on the structure, molecular biology and function. J Steroid Biochem 40:735-753. Grenot C, De Montard A, Blach6re T, De Ravel MR, Mappus E, Cuilleron CY (1992). Characterization of Met-139 as the photolabeled amino acid residue in the steroid binding site of sex hormone binding globulin using A6 derivatives of either testosterone or estradiol as unsubstituted photoaffinity labeling reagents. Biochemistry 31:7609-7621. Dechaud H, Lejeune H, Garoscio-Cholet M. Mallein M, Pugeat M (1989). Radioimmunoassay of testosterone not bound to sex-steroid binding protein in plasma. Clin Chem 35:1609-1614. Forest MG, Bertrand J (1972). Studies of the protein binding of dihydrotestosterone (17fl-hydroxy-5et-androstane-3-one) in human plasma (in different conditions and effect of medroxyprogesterone). Steroids 19:197-214. Andersson S, Berman DM, Jenkins EP, Russel DW (1991). Deletion of steroid 5ot-reductase 2 gene in male pseudohermaphroditism. Nature 354:159-161. Pardridge WM (1988). Selective delivery of sex steroid hormones to tissues in vivo by albumin and sex hormone-binding globulin. Ann NY Acad Sci $38:173-192. Ekins R (1992). The free hormone hypothesis and measurement of free hormones. Clin Chem 38:1289-1293. Mendel CM (1992). The free hormone hypothesis. Distinction from the free hormone transport hypothesis. J Androl 13:107-116.

In vitro influence of plasma steroid-binding proteins on androgen: Ddchaud et aL 19.

20.

21.

Dunn JF, Nisula BC, Rodbard D (1981). Transport of steroid hormones: binding of 21 endogenous steroids to both testosteronebinding globulin and corticostemid binding globulin in human plasma. J Clin Endocrinol Metab 53:58-68. LookingbiU DP, Egan N, Santen RJ, Dcmers LM (1988). Correlation of serum 3~-androstanediol glucuronide with acne and chest h~r in men. J Clin Endocrinol Metab 67:986-991. Damassa DA, Lin TM, Sonnenschein C, Soto AM (1991). Biological effects of sex hormone-binding globulin on androgen-induced

22.

23.

proliferation and androgen metabolism in LNCaP prostate cells. Endocrinology 129:75-84. Pugeat M, Dunn JF, Nisula BC (1981). Transport of steroid hormones: interaction of 70 drugs with testosterone-binding globulin and corticosteroid-binding globulin in human plasma. J Clin Endocrinol Metab 53:69-75. Pugeat M, Lejeune H, I~chand H, Emptoz-Bonneton A, Fleury MC, CharriE A, Tourniaire J, Forest MG (1987). Effects of drug administration on gonadotropins, sex steroid hormones and binding proteins in humans. Hormone Res 28:261-273.

Steroids, 1995, vol. 60, February

233