- Email: [email protected]
Cortisol metabolism in vitro—II. Species difference
J. Steroid Biochem. Molec. Biol. Vol.45, No. 5, pp. 445453, 1993
0960-0760/93 $6.00+ 0.00 Copyright© 1993PergamonPressLtd
Printed in Great Britain. All rightsreserved
CORTISOL
METABOLISM
IN
VITRO--II.
SPECIES
DIFFERENCE S. M. ABEL,D. J. BACK,*J. L. MAGGS and B. K. PARK Department of Pharmacology and Therapeutics, University of Liverpool, P.O. Box 147, Liverpool, L69 3BX, England (Received 31 October 1992; accepted 15 January 1993)
Summary--It has been suggested that cortisol 6fl-hydroxylase activity specifically reflects cytochrome P4503A (CYP3A) levels in the liver. However, we have previously reported that the metabolism of cortisol in human liver fractions in vitro is extremely complex and variable, and therefore complete metabolite analysis must be undertaken if 6fl-hydroxycortisol is to be used as a marker of CYP3A activity. In the present study, the metabolism of [3H]cortisol by hepatic microsomes from various animal species, and by cytosol from male and female rats, has been defined and compared with metabolites formed by human liver microsomes. Metabolites were characterized by co-chromatography with authentic standards, mass spectrometry, and quantified by radiometric HPLC. The results show that all microsomes prepared from animal species studied (male and female rat, male and female guinea-pig, male hamsters and mice) can metabolize cortisol, although the metabolic profiles are both quantitatively and qualitatively different from that obtained with human microsomes. In general the metabolic profiles for animal microsomes are simpler: hamster, mouse and guinea pig show only 6fl-hydroxylase and I lfl-dehydrogenase activity, although male rat shows both of these and 20fl-reductase activity while the female rat possesses all of the above as well as the ability to reduce the A-ring (A4-reductase and 3-oxidoreductase activities). The female rat also produces two metabolites undetected in humans. Incubations with male rat cytosol generated 20fl-dihydrocortisone as the major metabolite, and several unidentified minor polar metabolites, whereas female cytosolic products were identical to those generated by human cytosol, the major metabolite being 3ct,Sfl-tetrahydrocortisol. In conclusion, our studies have shown that hepatic cortisol metabolism is extremely variable amongst the species investigated and that the hamster provides the simplest model with which to explore cortisol 6fl-hydroxylase activity.
INTRODUCTION
would be useful to find a suitable animal model which produces a much simpler metabolic profile (i.e. the effective absence of competitive and potentially variable pathways). In the present study we have examined the in vitro metabolism of cortisol in liver microsomes from male and female rat, male and female guinea-pig, male hamster and male mouse. Cytosolic metabolism in the male and female rat was also investigated as there are known to be sex related differences in the expression of a number of cytosolic enzymes [8, 9]. These differences, as well as the more widely known sex-dependent expression of a number of microsomal P450s, including CYP3A [10], are thought to be due to variations in sex specific hormonal control [11]. Finally induction of the 6fl-hydroxylase enzyme has been explored in both male and female rats using dexamethasone, a synthetic glucocorticoid, which induces and is itself metabolized by CYP3A.
Cortisol metabolism in humans mainly involves A-ring and side-chain reduction, both in vivo [1, 2] and in vitro [3, 4]. 6fl-Hydroxycortisol (6fl-OHF) [1, 5] is a minor metabolite, formed by hepatic oxidation and has been widely used as a simple non-invasive in vivo marker of enzyme induction. There is evidence [6] that the cytochrome P4503A (CYP3A) subfamily is responsible for this 6fl-hydroxylation. In a recent study [7] we identified all the metabolites of cortisol produced by microsomal and cytosolic fractions of human liver and have described highly complex and variable metabolite profiles which bring into question the use of 6fl-OHF as a marker of baseline CYP3A activity in man [6]. In order to investigate the effects of other drugs on 6fl-hydroxylation it *To whom correspondence should be addressed. SBMa 4S~S~l
445
446
s.M. ABELet al.
Following a 5 min equilibration period, the reaction was initiated by addition of the Chemicals NADPH-regenerating system (glucose-6-phos[1,2,6,7 -3H]Cortisol (80 Ci.mmol-~) was phate, NADP ÷, and glucose-6-phosphate obtained from Amersham Int. (Bucks., dehydrogenase). England). Cortisol, 6fl-hydroxycortisone The reaction was terminated by cooling in (6fl-OHE), 20fl-dihydrocortisol (20fl-DHF), crushed ice. Cortisol and its metabolites were 20fl-dihydrocortisone (20fl-DHE), cortisone, extracted in ethyl acetate (2 × 2 ml). The solvent 30t,5fl-tetrahydrocortisol (3~,5fl-THF), 3ct,5fl- was evaporated to dryness under nitrogen and tetrahydrocortisone (3~t,5fl-THE), dexametha- samples were reconstituted in methanol (200/~ 1) sone, glucose-6-phosphate, NADP ÷ and glucose- before analysis by radiometric HPLC. Recovery 6-phosphate dehydrogenase were obtained from of radioactivity from the incubation was shown Sigma Chemical Co. (Poole, Dorset, England). to be >95%. 6~- and 6fl-OHF were synthesized by Dr J. Yeung (Chinese University of Hong Kong). H P L C analysis HPLC solvents were of analytical grade Cortisol and its metabolites were resolved on and were supplied by Fisons, (Loughborough, a Nucleosil 5C8 column (5/~m, 25 cm × 4.6 mm England). Scintillation cocktail (Flo-Scint i.d.) protected by an in-line Cl8 guard column. A) was obtained from Canberra-Packard Elution (50min) was isocratic with a mobile (Pangbourne, Bucks., England). All other phase of ammonium orthophosphate buffer chemicals were from B.D.H. (Poole, Dorset, (0.5% w/v, pH 3.0) and acetonitrile (75:25, v/v). England). SEP-PAK Ct~ cartridges were The flow rate was 0.7 ml.min -~ and absorbance obtained from Waters (Milford, MA, U.S.A.). was monitored at 240nm. Analysis was carried out on an SP8800 ternary pump, with an Animals SP100 variable wavelength detector (SpectraLivers were removed from male and female Physics) linked to an on-line Radiomatic A250 Wistar rats, male and female Dunkin-Hartley FLO-ONE detector (Canberra-Packard). guinea-pigs, male Syrian hamsters and male Metabolites were initially identified by coCBA brown mice following cervical dislocation chromatography with authentic standards and and exsanguination, and the mouse livers (six) these identifications confirmed by mass were pooled. The rats were approx. 250g in spectrometry. weight, guinea pigs 400-500 g, hamsters 80-100 g, and mice 25-30 g. Mass spectrometry Dexamethasone induction Samples of metabolites for mass spectrometry MATERIALS AND METHODS
Two male and two female rats were injected i.p. with dexamethasone (100mg'kg-~; 20mg ml-J in PEG: saline; 75 : 25) on 3 consecutive days. The treated rats together with male and female control rats (given vehicle only) were starved overnight before excision of the liver as described above. Microsomal and cytosolic incubations
Washed microsomes were prepared by the classical differential centrifugation technique [12]. Cytosol from the first 105,000g spin was retained. Protein was assayed by the method of Lowry et al. [13]. Microsomal or cytosolic protein (2 mg) were incubated at 37°C for 1 h with [3H]cortisol (0.1 #Ci; 1-50/~M), MgC12 (10 mM), glucose-6-phosphate (10 mM), NADP ÷ (5 mM), glucose-6-phosphate dehydrogenase (2U) and 0.067M phosphate buffer (pH7.4) to give a final volume of 0.5ml.
were isolated by HPLC from incubations containing 50/~M cortisol to ensure efficient mass and following chromatographic separation they were recovered from the eluate using SEPPAK C~8 cartridges. Following a water wash (2 × 3 ml) to remove any residual phosphate buffer, steroids were eluted from the cartridges with a single volume (3ml) of methanol. Isolated metabolites and authentic standards were analysed via the solids probe of a VG TS 250 mass spectrometer. Electron impact (El) and chemical ionization (CI) mass spectra were acquired over m/z 50-800 and 75-800, respectively, at resolution 800 and a scan time of 1 s. Operating conditions were as follows: accelerating voltage 4 × 1 0 3 V; source temperature 180°C; electron energy, 70 eV (El) or 50 eV (CI); emission current 700#A (El) or 500 ~A (CI). The reagent gas was either ammonia or isobutane at a source pressure of 2 × 10-4mb.
Species difference in cortisol metabolism
447
Table 2, Molecular ions and major fragments of cortisol metabolites isolated from various species during in vitro studies (El)
Western blots Hepatic microsomal proteins (50/zg) were separated on a 10% gel by SDS polyacrylamide gel electrophoresis (SDS-PAGE) according to the method of Laemmli[14]. The separated proteins were transferred to nitrocellulose electrophoretically. Immunoblotting was performed by the method of Towbin and Gordon [15]. Briefly, following blocking of nonspecific binding sites, the nitrocellulose was incubated with antiserum containing polyclonal antibodies to rat cytochrome P4503A (1:2500 dilution, Oxygene; Dallas, TX, U.S.A.). A horseradish peroxidase-labelled second antibody (dilution 1:5000) was used to reveal the immunoreactive polypeptides, the sites of antibody binding being visualized by enhanced chemiluminescence (Amersham). The relative intensities of the bands in the different sets of microsomes were quantified by laser densitometry (LKB Ultroscan XL, Bromma, Sweden), and the integration of the absorbance peak associated with each band. RESULTS
Metabolites isolated from animal liver fractions were rigorously identified by mass spectrometry (Tables 1 and 2). The human metabolites [Fig. l(a)] had been characterized previously [7]. There were considerable species and sex differences (both quantitative and qualitative) in the microsomal metabolic profiles (Table 3).
Metabolite 6.3-OHF (hamster) Cortol (female rat) 20,3-DHF (male rat) 20.3-DHE (male rat) Cortisone (guinea-pig) 5~-DHF (female rat)
Electron ionization Ion (relative intensity) 378(12), 360(7), 348(64), 331(28), 318(100), 312(29), 303(22), 285(38), 267(37) 350(M-18;8), 332(9), 314(11), 289(47), 271(100), 253(58) 364(55), 346(29), 331(11), 315(31), 303(43), 285(100), 267(28), 242(44), 227(46), 215(58) 362(30), 344(13), 326(8), 313(16), 301(100), 283(11), 257(14), 122(86) 360(3), 342(3), 330(12), 313(9), 301(11), 285(10), 122(100) 346(16), 346(9), 331(7), 313(14), 304(18), 297(9), 287(52), 269(36), 244(11)
male rat. Two metabolites, not previously detected in human liver were found with R, 12min (24.6%), and 35.Smin (8.3%). Production of both metabolites was concentration dependent (Fig. 2). They were identified as cortol and 5~-DHF, respectively (Tables 1 and 2). The major biotransformation was reduction of the A-ring to give 3~,5fl-THF (47.7%), although identification of the 5~-DHF suggests that A-ring reduction is not carried out by a coupled enzyme such as that seen in humans [4]. 6fl-Hydroxylase activity was also detected. Metabolism of cortisol by the male rat was unique amongst the species studied in that only more polar metabolites were formed [Fig. l(d)]. In neither the male nor female rat was cortisone detected. The major biotransformation in the males was reduction of the C20 keto group to give either 20fl-DHF (60%) or 20flDHE (6.1%). Small amounts of 6fl-OHF and 6~-OHF were generated.
Rat The female rat gave the most extensive metabolism of cortisol with only 4% remaining unchanged. Its metabolic profile [Fig. 1(c)] was different from that of other species and also the Table 1. Molecular ions and major fragments of cortisol metabolites isolated from various species during in vitro studies (CI) Metabolite 6fl-OHF
(hamster) Cortol (female rat) 20fl-DHF (male rat) 20fl-DHE (male rat) Cortisone (guinea-pig) 5~t-DHF (female rat)
Chemical ionization Ion (relative intensity) 361(2), 349(20). 333(15), 319(78), 301(30), 285(11) 386(M+18;20), 368(M;44), 350(23), 333(100), 324(50), 315(95), 304(44), 297(58), 287(47), 273(74) 365(M+1 ÷;100), 347(7), 303(8) 363(M+ 1+;100), 345(16), 329(8), 315(6), 301(9) 361(M + 1+;6), 343(15), 331(32), 301(100) 382(M + 18;39), 364(M;13), 352(53), 346(12), 336(11), 329(16), 322(100), 305(25), 287(17)
Guinea-pig The major metabolite produced by both male and female guinea-pig microsomes was cortisone [11.9 and 26.5%, respectively, Fig. l(e)]. Minor metabolites were 6~- and 6fl-OHF, accounting for only 3.3% of the recovered radioactivity in the female and 2.9% in the male.
Hamster Although the hamster gave the same three metabolites as the guinea-pig, there were major quantitative differences [Fig. l(b)]. The principal metabolite was 6fl-OHF (30.6%), with some 6~-OHF also generated. However, compared with both the guinea-pig and mouse, the ability of the hamster to reduce cortisol to cortisone was considerably reduced.
448
S.M. ABEL etal.
(o) 13o12o~
(b) Cortisol
110~°°L go-
14oo
Cortisol
30~58-THE 1200] 206:DHE
30/,58 ~ "THF II
68-OHF
~.1ooo1 o0,.
;7: i
~
!;;
ljCortisonetl
,--OH,P1 I/tl
' ::
68-OHF ~ l l
VVt
It :I
._~ ._>
I 1 : 4 0 0
200.
5
10 15 20 25 30 35 40 45 50 RetentionTime(minutes)
0
5
~Cortis. , one
10 15 20 25 30 35 40 45 50 RetentionTime(minutes)
(d)
(c) 16oo-
3500-
30f.5B-THF
208-DHF
30(X) -
14011-
Cortol
~'1200-
~,2500-
Q. 0
O. o
208-DHE
_;2oooi
~1000-
t
>
>
'~._o "0
~
Cortisol
Cortisol
501.DHF
600-
¢: 1000
400- 6a-OHF
68-OHF
200~ 0
. . . . . . . . . . .
0
5
q ....
~ . . . . . . . . . . . . . . . .
0
10 15 20 25 30 35 40 45 50 Retention13rne(minutes)
(e) 4000=
, ....
0
, ....
5
, ....
, . . . . . . .
,
....
,:'~' " ~ " , ' . . . 7 . . . : ,
10 15 20 25 30 35 40 45 50 RetentionTime(minutes)
(f)
Cortisol
600O Co~i~l 5O0O 40O0
O.
~. 20o0~
:z 3000i
Cortisone 68-OHF 60t-OHF,~, ,,,~~J..., 5
10
~ ,
........ ,.~ ....... , 20 25 30 35 40 45 50 Retention~me (minutes)
15
100o1
6~-F
j r ....
0
r ....
5
~ ....
i ....
J ....
i ....
t ....
~ ....
t0 15 20 25 30 35 RelentionTime(rninutee)
i ....
40
Fig. 1. HPLC separation of [3H]cortisol and metabolites produced by hepatic microsomes at a concentration of 1 #M: (a) human (WTI, male, 18 years); (b) male hamster; (c) female rat, (d) male rat; (e) female guinea-pig; and (f) mouse.
[ ....
45
i
50
Species differencein cortisol metabolism
449
Table 3. Species difference in metabolism of [3H]cortisol by liver microsomes in vitro
Metabolite
% Formation
Retention time (rain)
3 Rat
9 Rat
9 Guinea-pig
$ Guinea-pig
$ Hamster
5.4 8 12 17.5 20 26 28 31 35.5 40
2.4 4- 0.4 2.0+0.0 -60.0 +_ 5.1 6.1 4- 2.9 37.5 +_ 2.8 -----
5.6 + 2.5 1.34-0.3 24.6 4- 3.0 2.7 4- 0.4 . 3.6 _+ 0.6 47.7 4- 1.8 -8.3 4- 2.5 6.2 4- 1.0
1.2 4- 0.2 2.24-1.1 -. . 70.1 4- 6.9 . 26.5 4- 6.9 . .
1.2 -+ 0.1 1.7 + 0.1 -. . . 85.2 4- 2.0 . 11.9 4- 1.9 . .
4.4 4- 1,0 30.64-6.4 --
1.0 2.2 2.2
57.7 _+ 7.6
74.6
7.35 4- 0.9 .
20.0
6a-OHF 6fl-OHF Cortol 20fl-DHF 20fl-DHE Cortisol 3,,,5~-THF
Cortisone
5a-DHF 3a,5fl-THE
.
of metabolite (mean 4- SD)
.
. . .
Mouse
Human" 2.5 4- 1.5 10.2+11.0 -6.6 4- 2.7 8.9 4- 2.9 32.4 4- 13.1 7.4 4- 7.3 13.0 4- 7.1
.
. . .
15.2 4- 10.0
n = 4 for all species, except m a n (n = 6), and mouse (microsomes pooled from 6 livers). ' H u m a n data from Abel et aL [7].
Mouse
2,500 (o]
Four metabolites were identified but 74.6% of the cortisol recovered was unchanged. The major biotransformation was production of cortisone [20.0%, Fig. l(f)]; 6Qt-OHF, 6fl-OHF and 20fl-DHF were minor metabolites.
30t,58-THF 2,000 ~1,500
Cortol
1,ooo
4,000
co°,,o,
-oH II
0
5
10
50~DHF
\
15
20
30
25
(b)
35
Cytosolic incubations
40
45
5O
30~ 5B-THF
3,500
3,000 ~ 2,500
Cortisol
"2,000
g
n-
1,500
5OI-DHF
~
1,000 500 5
10
.ooo 4,5001(c)
15
20
30
25
35
Dexamethasone induction
B-THE
40
45
50
30/,5B-THF
4,0001
~~a,ooo I 3,500
~2,5~-
Results for male and female rats are shown in Table 4; and Fig. 3(a and b), respectively. In both sexes the metabolic profiles were much simpler than those produced in microsomal incubations. The major metabolite in the female was 3ct,5fl-THF (84.2%); 3~t,5fl-THE and 20flDHF were minor metabolites. In contrast, the major metabolite in male cytosol was 20fl-DHE (65.5%), with some 20fl-DHF (12.4%), whereas the principal metabolite in microsomes was 20fl-DHF. A number of more polar unidentified metabolites were also produced.
Microsomes prepared from livers of dexamethasone treated rats showed increased 6/thydroxylase activity (Table 5; Fig. 4), consistent with the increased level of CYP3A on a Western blot (Fig. 5). There was also an increase in 6~-OHF.
Cortisol
\
'6 2,0OO !
50/.-DHF
Table 4. Metabolism of cortisol in hepatic cytosol of male and female rats
~ 1,5~-
% Formation of metabolite
1,~i 5OO
~~,5B-THE
60C-OHF
0
. . . .
0
,
5
. . . .
I
R!
. . . .
10
i
. . . .
15
i
. . . .
20
,
. . . .
25
i
. . . .
30
,
. . . .
35
i
. . . .
40
~
. . . .
45
i
50
RetentionTime(minutes) Fig. 2. C o n c e n t r a t i o n d e p e n d e n t m e t a b o l i s m o f [3 H ] c o r t i s o l b y f e m a l e r a t h e p a t i c m i c r o s o m e s (in vitro): (a) 1 # M ; (b) 1 0 # M ; a n d (c) 1 0 0 g M .
Metabolite
(rain)
Female rat
Male rat
6a-OHF 20fl-DHF 20fl-DHE Cortisol 3~t,Sfl-THF 3~t,58-THE
5.4 17.5 20.0 26.0 28.0 40.0
2.17_+0.12 2.2 _+ 2.27 -5.98 _+ 1.96 84.23 +_ 2.10 5.38 -+ 1.56
2.31 _+0.07 12.42 4- 1.85 65.48 4- 4.25 8.93 _+ 1.22 ---
Additional unknown metabolites (1, 2, 3, total 10%) were evident in male rat cytosol at R, 8.6, 10.7, 13.3 min--see Fig. 4.
450
S.M. ABEL et al.
(a) 20B-DHE
4OOO
3500 Unknown~
3000
Unknown2
~2500
2ooo
~
nknowns
o 1,oo rr i 0 0 0 :
\ \
500
~
/X/
Cortisol
0 0
5
10
15
20
25
30
35
40
45
Retention Time (min) (b) 3O/,5B-THF
4OO0
35003000
.> 20o0-
Cortisol
1000 2 500
5
10
15
20
25
30
35
40
45
Retention Time (rain)
Fig. 3. Sex differences in the metabolism of [3H]cortisol (1/aM) by rat cytosol: (a) male cytosol; and (b) female cytosol. DISCUSSION
The present work has revealed considerable interspecies variation, both quantitative and qualitative, in the metabolism of cortisol by liver fractions from man and several laboratory animals. In most cases, 6fl-hydroxylation is a minor pathway with reduction of the corticosteroid as the favoured reaction. However, the site of reduction of the steroid varied considerably between species. The female rat gives an interesting microsomal metabolic profile, with the major Table 5. Sex differences in the metabolism of [3H]cortisol to 6fl-OHF (CYP3A) by liver microsomes from control and dexamethasone-induced rats in vitro % Formation of metabolite
Control Induced
Male rat
Female rat
6.33 10.96
2.26 8.6
n = 2, for both control and induced rats.
metabolite being 3~,5fl-THF. In human liver, A-ring reduction of cortisol involves a coupled enzyme system (A4 reductase, 3~-oxidoreductase [4]), since none of the dihydrocortisol intermediate is detected. However, the presence of a 5~-reduced metabolite in the female rat (5~-DHF), suggests that A-ring reduction may be catalysed by two separate enzymes. No 5flD H F was detected, possibly because the pH of the incubation was suboptimal: optimum pH ranges for the stability of the two enzymes have been shown to be 7.0-7.5 for 5~-reductase [16] and 5.5-6.0 for 5fl-reductase[17]; the present incubations being carried out at pH7.4. However, Forchelli et al. [16] provided evidence that only 5~-reductase is present in the female rat. It is important to note that the intermediate in the A-ring reduction pathway is the 5~-DHF, but the final product is 3~,5fl-DHF, suggesting that there is some conformational inversion at C 5 while bound to the 3-oxidoreductase enzyme. The metabolite of R, 12 min was identified as a cortol (A-ring and side chain reduction). This metabolite does not absorb between 200-300 nm, as there are no unsaturated bonds in the molecule and thus has been identified by mass spectrometry alone. The cortol is only generated at concentrations between 1 and 10/~M (Fig. 2), and an increase in 5~-DHF accompanies the decrease in the amount of cortol produced. This suggests that the 3~-HSD enzyme is easily saturable, and possibly that the product of A-ring reduction may inhibit C20 reduction at higher concentrations. The major biotransformation in male rat microsomes is the stereospecific reduction of the C2o keto group to a fl-hydroxyl moiety. There is a complete absence of this pathway in microsomes from female rats, guinea-pigs, hamsters and mice. Since both 20fl-DHF and 20fl-DHE are formed in the male rat microsomes but no cortisone is detected, it is not clear whether the route of metabolism involves reduction of cortisol at C20 to give 20fl-DHF and then oxidation at C~] to give 20fl-DHE, or whether there is an interconversion between cortisol and cortisone followed by reduction of all the cortisone to 20fl-DHE. 6fl-OHF is a minor metabolite in both male and female rat microsomes. The hamster and guinea-pig microsomes produce qualitatively similar but quantitatively different metabolic profiles. The hamster produces a small amount of cortisone, the preferred pathway being 6fl-hydroxylation. This provides
5
5
10
20
25
30
Retention Time (rain)
15
~, Cortisol
IUnkn°wn
208.DHE
2(~6-DHF
15 20 25 30 Retention Time (rain)
~
35
35
40
40
45
45
oooi
11oo-~ IO0O-~
13oo7 12oo-~
0 0
I00-
200-
~: 4 ~ -
600-
8o7
~900
(dl
100S
200 S
3OO
5
5
10
20
25
~)
Cortisol.
30
15
20
25
30
RetentionTime(rain)
35
35
40
40
3OL58-THE
30t,5B-THF
Retention Time (rain)
Corto(
10
15
Cortot
4oo2 68-OHF ,OHF I
.,~ go0-
.~ 7o0-
~900
I000-~
1200-~ 11001
(c] ~3007
45
45
Fig. 4. Sex-differences in the metabolism of {3H]cortisol (I p M) by liver microsomes from control and dexamethasone-induced rats in vitro: (a) control male; (b) induced male; (c) control female; and (d) induced female.
0
10
.
Cortisol
20t3-DHE
20t3.1 ~HF
6B-OHF
.~200- =,~u=
°300_
40O-
0
O'
4oo2
( b / 500-
o
.~" ZOO1
g. ~ 2
11oo 100o-
12oo-
(a) 130o-
O" O
452
S.M. ABEL et al.
evidence to support Lipman et a/.[18] who suggested that 6//-hydroxylation is a compensatory route when there was an inability to perform A-ring reduction. Further work needs to be undertaken in this area in view of reported sex differences in the hamster [19]. The low levels of 6//-hydroxylase activity detected in guinea-pig microsomes can perhaps be explained by the observation that 6fl-OHF production in vivo is strain dependent and that 6fl-OHF has been shown to undergo variable and extensive metabolism itself in vivo [20]. It has been suggested previously by Burstein and Kimball[21], that there is great interindividual variation in the urinary excretion of corticosteroids by the guinea-pig, and that two distinct phenotypes exist; those with high 6fl-OHF excretion type A, and those with low polar corticosteroid excretion type B[22]. Dunkin-Hartley guinea-pigs are known to display this phenotypic variation, but it is possible our sample size was too small to detect it. The major metabolite seen in mouse microsomes was cortisone. 6//-OHF was formed as a minor metabolite; it has previously been shown that mice possess an enzyme comparable to CYP3AI in rats, which is capable of 6//hydroxylation [23]. Cortisol metabolism in liver cytosol of male and female rats differs both qualitatively and quantitatively. The female gives a metabolite profile which is similar to that produced in human liver cytosol, and contains metabolites which have all been rigorously identified [7]. The major pathway is the formation of 3e,5//-THF. Since no intermediate was detected, the enzyme system responsible is different from that found in female microsomes, which yield 5c~-DHF and is possibly comprised of a coupled oxidoreductase as in humans. The male rat cytosol produces 20fl-DHE as the major metabolite. This contrasts with microsomal incubations in which C20 reduction is also the major biotransformation, but the final product is 20fl-DHF, suggesting much higher l l//-oxidoreductase activity in the cytosol. A number of unidentified metabolites were also generated which when combined account for about 10% of the metabolism. Possible thoughts as to the identity of some of these, include polyhydroxyl metabolites, or even Phase II conjugates. It is known, for example, that glutathione transferase is a cytosolic enzyme [24], which will attack an electrophilic centre.
52 kDa
m
D
Fig. 5. Western Blot showing the induction of CYP3A by dexamethasone in male and female rats.
In agreement with others [25] it was found that dexamethasone is a substantial inducer of CYP3A in rats (Fig. 5). Simultaneous with the increase in 6fl-hydroxylase activity, there was a decrease in A-ring reduction, providing further evidence in support of Lipman et al. [18]. The unknown male rat metabolite, more polar than cortisol but less polar than those having undergone C20 reduction, is likely to be a product of the CYP3A subfamily as it is formed only by induced microsomes. Recent work on species differences in overall digitoxin metabolism [26], showed large variations and the rank order of: hamster > rat > guinea-pig > mouse > human. In all species conversion ofdigitoxin to dt2 is catalysed predominantly by CYP3A, whereas enzymes other than CYP3A catalyse other hydroxylations (12-, 16-, 17-). Further evidence for high CYP3A activity in the hamster was obtained by Halvorson et al. [27] in work on testosterone 6//-hydroxylation. It has been suggested that CYP3A1 (the rat 6fl-hydroxylase enzyme) is specific for the //face of the testosterone molecule[28], with minimal hydroxylation on the c~-face. However, it is evident that 6~-OHF was also formed by microsomes of all species and indeed exceeded 6fl-OHF tbrmation in microsomes
Species difference in cortisol metabolism
from the female rat. Furthermore, an increase in formation of 60t-OHF was noted in the dexamethasone induction experiment (Fig. 4), but at present it is not known whether this is due to induction of 60t-hydroxylase activity, or whether 6~- and 6fl-OHF epimeres are formed from a common intermediate, the production of which increases following induction. In conclusion, we have demonstrated that there are major differences between species in the in vitro routes of metabolism of cortisol. There are also marked sex differences in the rat. The hamster provides the simplest metabolic profile and therefore provides the simplest model with which to explore further factors which regulate cortisol 6fl-hydroxylase activity. Acknowledgements--We would like to acknowledge Mrs C. McLean for her assistance with the Western Blot analysis. The TS250 mass spectrometer was purchased with generous grants from the Wellcome Trust and the University of Liverpool.
REFERENCES 1. Katz F. H., Lipman M. M., Frantz A. G. and Tailor J. W.: The physiological significance of 6/3-OHF in human corticoid metabolism. J. Clin. Endocr. Metab. 22 (1962) 71-77. 2. Zumoff B., Bradlow L., GaUagher T. F. and Hellman L.: Cortisol metabolism in cirrhosis. J. Clin. Invest. 46 (1967) 1735-1743. 3. Meigs R. A. and Engel L. L.: The metabolism of adrenocortical steroids by human tissues. Endocrinology 69 (1961) 152-162. 4. Iyer R. B., Binstock J. M., Schwartz I. S., Gordon G. G., Weinstein B. I. and Southren A. L.: Human hepatic cortisol reductase activities: enzymatic properties and substrate specificites of cytosolic cortisol A4-5/3reductase and dihydrocortisol-3~-oxidoreductase(s). Steroids 55 (1990) 495-500. 5. Park B. K. and Kitteringham N. R.: Relevance and means of assessing induction and inhibition of drug metabolism in man. Prog. Drug. Metab. 11 (1989) 1-59. 6. Ged C., Rouillon J. M., Pichard L., Combalbert J., Bressot N., Bories P., Michel H., Beaune P. and Maurel P.: The increase in urinary excretion of 6/3-hydroxycortisol as a marker of human hepatic cytochrome P-450IIIA induction. Br. J. Clin. Pharmac. 28 (1989) 373 387. 7. Abel S. M., Maggs J. L., Back D. J. and Park B. K. Cortisol metabolism by human liver in vitro. (I). Metabolite identification and interindividual variability. J. Steroid Biochem. Molec. Biol. 43 (1992) 713 719. 8. Mode A. and Rafter I.: The sexually differentiated A4-3-ketosteroid 513-reductase of rat liver. J. Biol. Chem. 260 (1985) 7137-7141. 9. Stenberg A.: Development, diurnal and oestrous cycledependent changes in the activity of liver enzymes. J. Endocr. 68 (1976) 265-272. 10. Smith D. A.: Species differences in metabolism and pharmacokinetics: Are we close to an understanding? Drug Metab. Rev. 23 (1991) 355-373.
453
11. Waxman D. J.: Interactions of hepatic cytochromes P-450 with steroid hormones. Biochem. Pharmac. 37 (1988) 71-84. 12. Purba H. S., Maggs J. L., Orme M. L'E., Back D. J. and Park B. K.: The metabolism of 17~-ethinylestradiol by human liver microsomes: formation of catechol and chemically reactive metabolites. Br. J. Clin. Pharmac. 23 (1987) 447-453. 13. Lowry O. H., Rosebrough N. J., Farr A. L. and Randall R. J.: Protein determination with the Folin phenol reagent. J. Biol. Chem. 193 (1951) 265-275. 14. Laemmli U. K.: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 (1970) 680-685. 15. Towbin H. and Gordon J.: Immunoblotting and immunobinding--current status and outlook. J. Immun. Methods 72 (1984) 313-340. 16. Forchielli E., Brown-Grant K. and Dorfman R. I.: Steroid A4-hydrogenases of rat liver. Proc. Soc. Exp. Biol. Med. 99 (1958) 594-596. 17. Tomkins G. M.: The enzymatic reduction of A4-3-keto steroids. J. Biol. Chem. 225 (1957) 13-24. 18. Lipman M. M., Katz F. H. and Jailer J. W.: An alternate pathway for cortisol metabolism: 6fl-OHF production by human tissue slices. J. Clin. Endocr. Metab. 22 (1960) 268-272. 19. Yates F. E., Herbst A. L. and Urquhart J.: Sexdifferences in rate of A-ring reduction of A4-3-keto steroids in vitro by the rat liver. Biochem. J. 63 (1958) 887 902. 20. Challiner M. R.: The assessment of 6fl-OHF as an /n vivo index of microsomal mixed function oxidase activity. Ph.D. Thesis, Liverpool University, Liverpool (1982). 21. Burstein S. and Kimball H. L.: Quantitative paper chromatography of C2~Os and C2~O6 corticosteroids from guinea pig urine extracts. Analyt. Biochem. 4 (1962) 132. 22. Burstein S., Kimball H. L. and Bhavnani B. R.: Urinary corticosteroid excretion patterns in guinea pigs: 2 main phenotypes. Steroids 2 (1963) 195. 23. Wrighton S. A., Schuetz E. G., Watkins P. B., Maurel P., Barwick J., Bailey B. S., Hortle H. T., Young B. and Guzelian P.: Demonstration in multiple species of inducible hepatic cytochromes P-450 and their mRNAs related to the glucocorticoid-inducible P-450 of the rat. Molec. Pharmac. 28 (1985) 312-321. 24. Vos R. M. E. and Van Bladeren P. J.: Glutathione S-transferases in relation to their role in the biotransformation of xenobiotics. Chem. Biol. lnterac. 75 (1990) 241-265. 25. Heuman D. M., Gallagher E. J., Barwick J. L., Elshourbagy N. A. and Guzelian P. S.: Immunochemical evidence for induction of a common form of hepatic cytochrome P-450 in rats treated with pregnenolone-16~-carbonitrile or other steroidal or non-steroidal agents. Molec. Pharmac. 21 (1981) 753-760. 26. Eberhart D. C., Gemzik B., Halvorson M. and Parkinson A.: Species differences in the toxicity and cytochrome P-450IIIA-dependent metabolism of digitoxin. Molec. Pharmac. 40 (1991) 859-867, 27. Halvorson M. A. and Parkinson S. A.: Species difference in testosterone hydroxylation by liver microsomes: studies with antibody against rat cytochrome P-450p. Toxicologist 9 (1988) 9-14. 28. Namkung M. J., Yang H. L., Hulla J. E. and Juchau M. R.: On the substrate specificity of cytochrome P-450IIIAI. Molec. Pharmac. 34 (1988) 628437.
0960-0760/93 $6.00+ 0.00 Copyright© 1993PergamonPressLtd
Printed in Great Britain. All rightsreserved
CORTISOL
METABOLISM
IN
VITRO--II.
SPECIES
DIFFERENCE S. M. ABEL,D. J. BACK,*J. L. MAGGS and B. K. PARK Department of Pharmacology and Therapeutics, University of Liverpool, P.O. Box 147, Liverpool, L69 3BX, England (Received 31 October 1992; accepted 15 January 1993)
Summary--It has been suggested that cortisol 6fl-hydroxylase activity specifically reflects cytochrome P4503A (CYP3A) levels in the liver. However, we have previously reported that the metabolism of cortisol in human liver fractions in vitro is extremely complex and variable, and therefore complete metabolite analysis must be undertaken if 6fl-hydroxycortisol is to be used as a marker of CYP3A activity. In the present study, the metabolism of [3H]cortisol by hepatic microsomes from various animal species, and by cytosol from male and female rats, has been defined and compared with metabolites formed by human liver microsomes. Metabolites were characterized by co-chromatography with authentic standards, mass spectrometry, and quantified by radiometric HPLC. The results show that all microsomes prepared from animal species studied (male and female rat, male and female guinea-pig, male hamsters and mice) can metabolize cortisol, although the metabolic profiles are both quantitatively and qualitatively different from that obtained with human microsomes. In general the metabolic profiles for animal microsomes are simpler: hamster, mouse and guinea pig show only 6fl-hydroxylase and I lfl-dehydrogenase activity, although male rat shows both of these and 20fl-reductase activity while the female rat possesses all of the above as well as the ability to reduce the A-ring (A4-reductase and 3-oxidoreductase activities). The female rat also produces two metabolites undetected in humans. Incubations with male rat cytosol generated 20fl-dihydrocortisone as the major metabolite, and several unidentified minor polar metabolites, whereas female cytosolic products were identical to those generated by human cytosol, the major metabolite being 3ct,Sfl-tetrahydrocortisol. In conclusion, our studies have shown that hepatic cortisol metabolism is extremely variable amongst the species investigated and that the hamster provides the simplest model with which to explore cortisol 6fl-hydroxylase activity.
INTRODUCTION
would be useful to find a suitable animal model which produces a much simpler metabolic profile (i.e. the effective absence of competitive and potentially variable pathways). In the present study we have examined the in vitro metabolism of cortisol in liver microsomes from male and female rat, male and female guinea-pig, male hamster and male mouse. Cytosolic metabolism in the male and female rat was also investigated as there are known to be sex related differences in the expression of a number of cytosolic enzymes [8, 9]. These differences, as well as the more widely known sex-dependent expression of a number of microsomal P450s, including CYP3A [10], are thought to be due to variations in sex specific hormonal control [11]. Finally induction of the 6fl-hydroxylase enzyme has been explored in both male and female rats using dexamethasone, a synthetic glucocorticoid, which induces and is itself metabolized by CYP3A.
Cortisol metabolism in humans mainly involves A-ring and side-chain reduction, both in vivo [1, 2] and in vitro [3, 4]. 6fl-Hydroxycortisol (6fl-OHF) [1, 5] is a minor metabolite, formed by hepatic oxidation and has been widely used as a simple non-invasive in vivo marker of enzyme induction. There is evidence [6] that the cytochrome P4503A (CYP3A) subfamily is responsible for this 6fl-hydroxylation. In a recent study [7] we identified all the metabolites of cortisol produced by microsomal and cytosolic fractions of human liver and have described highly complex and variable metabolite profiles which bring into question the use of 6fl-OHF as a marker of baseline CYP3A activity in man [6]. In order to investigate the effects of other drugs on 6fl-hydroxylation it *To whom correspondence should be addressed. SBMa 4S~S~l
445
446
s.M. ABELet al.
Following a 5 min equilibration period, the reaction was initiated by addition of the Chemicals NADPH-regenerating system (glucose-6-phos[1,2,6,7 -3H]Cortisol (80 Ci.mmol-~) was phate, NADP ÷, and glucose-6-phosphate obtained from Amersham Int. (Bucks., dehydrogenase). England). Cortisol, 6fl-hydroxycortisone The reaction was terminated by cooling in (6fl-OHE), 20fl-dihydrocortisol (20fl-DHF), crushed ice. Cortisol and its metabolites were 20fl-dihydrocortisone (20fl-DHE), cortisone, extracted in ethyl acetate (2 × 2 ml). The solvent 30t,5fl-tetrahydrocortisol (3~,5fl-THF), 3ct,5fl- was evaporated to dryness under nitrogen and tetrahydrocortisone (3~t,5fl-THE), dexametha- samples were reconstituted in methanol (200/~ 1) sone, glucose-6-phosphate, NADP ÷ and glucose- before analysis by radiometric HPLC. Recovery 6-phosphate dehydrogenase were obtained from of radioactivity from the incubation was shown Sigma Chemical Co. (Poole, Dorset, England). to be >95%. 6~- and 6fl-OHF were synthesized by Dr J. Yeung (Chinese University of Hong Kong). H P L C analysis HPLC solvents were of analytical grade Cortisol and its metabolites were resolved on and were supplied by Fisons, (Loughborough, a Nucleosil 5C8 column (5/~m, 25 cm × 4.6 mm England). Scintillation cocktail (Flo-Scint i.d.) protected by an in-line Cl8 guard column. A) was obtained from Canberra-Packard Elution (50min) was isocratic with a mobile (Pangbourne, Bucks., England). All other phase of ammonium orthophosphate buffer chemicals were from B.D.H. (Poole, Dorset, (0.5% w/v, pH 3.0) and acetonitrile (75:25, v/v). England). SEP-PAK Ct~ cartridges were The flow rate was 0.7 ml.min -~ and absorbance obtained from Waters (Milford, MA, U.S.A.). was monitored at 240nm. Analysis was carried out on an SP8800 ternary pump, with an Animals SP100 variable wavelength detector (SpectraLivers were removed from male and female Physics) linked to an on-line Radiomatic A250 Wistar rats, male and female Dunkin-Hartley FLO-ONE detector (Canberra-Packard). guinea-pigs, male Syrian hamsters and male Metabolites were initially identified by coCBA brown mice following cervical dislocation chromatography with authentic standards and and exsanguination, and the mouse livers (six) these identifications confirmed by mass were pooled. The rats were approx. 250g in spectrometry. weight, guinea pigs 400-500 g, hamsters 80-100 g, and mice 25-30 g. Mass spectrometry Dexamethasone induction Samples of metabolites for mass spectrometry MATERIALS AND METHODS
Two male and two female rats were injected i.p. with dexamethasone (100mg'kg-~; 20mg ml-J in PEG: saline; 75 : 25) on 3 consecutive days. The treated rats together with male and female control rats (given vehicle only) were starved overnight before excision of the liver as described above. Microsomal and cytosolic incubations
Washed microsomes were prepared by the classical differential centrifugation technique [12]. Cytosol from the first 105,000g spin was retained. Protein was assayed by the method of Lowry et al. [13]. Microsomal or cytosolic protein (2 mg) were incubated at 37°C for 1 h with [3H]cortisol (0.1 #Ci; 1-50/~M), MgC12 (10 mM), glucose-6-phosphate (10 mM), NADP ÷ (5 mM), glucose-6-phosphate dehydrogenase (2U) and 0.067M phosphate buffer (pH7.4) to give a final volume of 0.5ml.
were isolated by HPLC from incubations containing 50/~M cortisol to ensure efficient mass and following chromatographic separation they were recovered from the eluate using SEPPAK C~8 cartridges. Following a water wash (2 × 3 ml) to remove any residual phosphate buffer, steroids were eluted from the cartridges with a single volume (3ml) of methanol. Isolated metabolites and authentic standards were analysed via the solids probe of a VG TS 250 mass spectrometer. Electron impact (El) and chemical ionization (CI) mass spectra were acquired over m/z 50-800 and 75-800, respectively, at resolution 800 and a scan time of 1 s. Operating conditions were as follows: accelerating voltage 4 × 1 0 3 V; source temperature 180°C; electron energy, 70 eV (El) or 50 eV (CI); emission current 700#A (El) or 500 ~A (CI). The reagent gas was either ammonia or isobutane at a source pressure of 2 × 10-4mb.
Species difference in cortisol metabolism
447
Table 2, Molecular ions and major fragments of cortisol metabolites isolated from various species during in vitro studies (El)
Western blots Hepatic microsomal proteins (50/zg) were separated on a 10% gel by SDS polyacrylamide gel electrophoresis (SDS-PAGE) according to the method of Laemmli[14]. The separated proteins were transferred to nitrocellulose electrophoretically. Immunoblotting was performed by the method of Towbin and Gordon [15]. Briefly, following blocking of nonspecific binding sites, the nitrocellulose was incubated with antiserum containing polyclonal antibodies to rat cytochrome P4503A (1:2500 dilution, Oxygene; Dallas, TX, U.S.A.). A horseradish peroxidase-labelled second antibody (dilution 1:5000) was used to reveal the immunoreactive polypeptides, the sites of antibody binding being visualized by enhanced chemiluminescence (Amersham). The relative intensities of the bands in the different sets of microsomes were quantified by laser densitometry (LKB Ultroscan XL, Bromma, Sweden), and the integration of the absorbance peak associated with each band. RESULTS
Metabolites isolated from animal liver fractions were rigorously identified by mass spectrometry (Tables 1 and 2). The human metabolites [Fig. l(a)] had been characterized previously [7]. There were considerable species and sex differences (both quantitative and qualitative) in the microsomal metabolic profiles (Table 3).
Metabolite 6.3-OHF (hamster) Cortol (female rat) 20,3-DHF (male rat) 20.3-DHE (male rat) Cortisone (guinea-pig) 5~-DHF (female rat)
Electron ionization Ion (relative intensity) 378(12), 360(7), 348(64), 331(28), 318(100), 312(29), 303(22), 285(38), 267(37) 350(M-18;8), 332(9), 314(11), 289(47), 271(100), 253(58) 364(55), 346(29), 331(11), 315(31), 303(43), 285(100), 267(28), 242(44), 227(46), 215(58) 362(30), 344(13), 326(8), 313(16), 301(100), 283(11), 257(14), 122(86) 360(3), 342(3), 330(12), 313(9), 301(11), 285(10), 122(100) 346(16), 346(9), 331(7), 313(14), 304(18), 297(9), 287(52), 269(36), 244(11)
male rat. Two metabolites, not previously detected in human liver were found with R, 12min (24.6%), and 35.Smin (8.3%). Production of both metabolites was concentration dependent (Fig. 2). They were identified as cortol and 5~-DHF, respectively (Tables 1 and 2). The major biotransformation was reduction of the A-ring to give 3~,5fl-THF (47.7%), although identification of the 5~-DHF suggests that A-ring reduction is not carried out by a coupled enzyme such as that seen in humans [4]. 6fl-Hydroxylase activity was also detected. Metabolism of cortisol by the male rat was unique amongst the species studied in that only more polar metabolites were formed [Fig. l(d)]. In neither the male nor female rat was cortisone detected. The major biotransformation in the males was reduction of the C20 keto group to give either 20fl-DHF (60%) or 20flDHE (6.1%). Small amounts of 6fl-OHF and 6~-OHF were generated.
Rat The female rat gave the most extensive metabolism of cortisol with only 4% remaining unchanged. Its metabolic profile [Fig. 1(c)] was different from that of other species and also the Table 1. Molecular ions and major fragments of cortisol metabolites isolated from various species during in vitro studies (CI) Metabolite 6fl-OHF
(hamster) Cortol (female rat) 20fl-DHF (male rat) 20fl-DHE (male rat) Cortisone (guinea-pig) 5~t-DHF (female rat)
Chemical ionization Ion (relative intensity) 361(2), 349(20). 333(15), 319(78), 301(30), 285(11) 386(M+18;20), 368(M;44), 350(23), 333(100), 324(50), 315(95), 304(44), 297(58), 287(47), 273(74) 365(M+1 ÷;100), 347(7), 303(8) 363(M+ 1+;100), 345(16), 329(8), 315(6), 301(9) 361(M + 1+;6), 343(15), 331(32), 301(100) 382(M + 18;39), 364(M;13), 352(53), 346(12), 336(11), 329(16), 322(100), 305(25), 287(17)
Guinea-pig The major metabolite produced by both male and female guinea-pig microsomes was cortisone [11.9 and 26.5%, respectively, Fig. l(e)]. Minor metabolites were 6~- and 6fl-OHF, accounting for only 3.3% of the recovered radioactivity in the female and 2.9% in the male.
Hamster Although the hamster gave the same three metabolites as the guinea-pig, there were major quantitative differences [Fig. l(b)]. The principal metabolite was 6fl-OHF (30.6%), with some 6~-OHF also generated. However, compared with both the guinea-pig and mouse, the ability of the hamster to reduce cortisol to cortisone was considerably reduced.
448
S.M. ABEL etal.
(o) 13o12o~
(b) Cortisol
110~°°L go-
14oo
Cortisol
30~58-THE 1200] 206:DHE
30/,58 ~ "THF II
68-OHF
~.1ooo1 o0,.
;7: i
~
!;;
ljCortisonetl
,--OH,P1 I/tl
' ::
68-OHF ~ l l
VVt
It :I
._~ ._>
I 1 : 4 0 0
200.
5
10 15 20 25 30 35 40 45 50 RetentionTime(minutes)
0
5
~Cortis. , one
10 15 20 25 30 35 40 45 50 RetentionTime(minutes)
(d)
(c) 16oo-
3500-
30f.5B-THF
208-DHF
30(X) -
14011-
Cortol
~'1200-
~,2500-
Q. 0
O. o
208-DHE
_;2oooi
~1000-
t
>
>
'~._o "0
~
Cortisol
Cortisol
501.DHF
600-
¢: 1000
400- 6a-OHF
68-OHF
200~ 0
. . . . . . . . . . .
0
5
q ....
~ . . . . . . . . . . . . . . . .
0
10 15 20 25 30 35 40 45 50 Retention13rne(minutes)
(e) 4000=
, ....
0
, ....
5
, ....
, . . . . . . .
,
....
,:'~' " ~ " , ' . . . 7 . . . : ,
10 15 20 25 30 35 40 45 50 RetentionTime(minutes)
(f)
Cortisol
600O Co~i~l 5O0O 40O0
O.
~. 20o0~
:z 3000i
Cortisone 68-OHF 60t-OHF,~, ,,,~~J..., 5
10
~ ,
........ ,.~ ....... , 20 25 30 35 40 45 50 Retention~me (minutes)
15
100o1
6~-F
j r ....
0
r ....
5
~ ....
i ....
J ....
i ....
t ....
~ ....
t0 15 20 25 30 35 RelentionTime(rninutee)
i ....
40
Fig. 1. HPLC separation of [3H]cortisol and metabolites produced by hepatic microsomes at a concentration of 1 #M: (a) human (WTI, male, 18 years); (b) male hamster; (c) female rat, (d) male rat; (e) female guinea-pig; and (f) mouse.
[ ....
45
i
50
Species differencein cortisol metabolism
449
Table 3. Species difference in metabolism of [3H]cortisol by liver microsomes in vitro
Metabolite
% Formation
Retention time (rain)
3 Rat
9 Rat
9 Guinea-pig
$ Guinea-pig
$ Hamster
5.4 8 12 17.5 20 26 28 31 35.5 40
2.4 4- 0.4 2.0+0.0 -60.0 +_ 5.1 6.1 4- 2.9 37.5 +_ 2.8 -----
5.6 + 2.5 1.34-0.3 24.6 4- 3.0 2.7 4- 0.4 . 3.6 _+ 0.6 47.7 4- 1.8 -8.3 4- 2.5 6.2 4- 1.0
1.2 4- 0.2 2.24-1.1 -. . 70.1 4- 6.9 . 26.5 4- 6.9 . .
1.2 -+ 0.1 1.7 + 0.1 -. . . 85.2 4- 2.0 . 11.9 4- 1.9 . .
4.4 4- 1,0 30.64-6.4 --
1.0 2.2 2.2
57.7 _+ 7.6
74.6
7.35 4- 0.9 .
20.0
6a-OHF 6fl-OHF Cortol 20fl-DHF 20fl-DHE Cortisol 3,,,5~-THF
Cortisone
5a-DHF 3a,5fl-THE
.
of metabolite (mean 4- SD)
.
. . .
Mouse
Human" 2.5 4- 1.5 10.2+11.0 -6.6 4- 2.7 8.9 4- 2.9 32.4 4- 13.1 7.4 4- 7.3 13.0 4- 7.1
.
. . .
15.2 4- 10.0
n = 4 for all species, except m a n (n = 6), and mouse (microsomes pooled from 6 livers). ' H u m a n data from Abel et aL [7].
Mouse
2,500 (o]
Four metabolites were identified but 74.6% of the cortisol recovered was unchanged. The major biotransformation was production of cortisone [20.0%, Fig. l(f)]; 6Qt-OHF, 6fl-OHF and 20fl-DHF were minor metabolites.
30t,58-THF 2,000 ~1,500
Cortol
1,ooo
4,000
co°,,o,
-oH II
0
5
10
50~DHF
\
15
20
30
25
(b)
35
Cytosolic incubations
40
45
5O
30~ 5B-THF
3,500
3,000 ~ 2,500
Cortisol
"2,000
g
n-
1,500
5OI-DHF
~
1,000 500 5
10
.ooo 4,5001(c)
15
20
30
25
35
Dexamethasone induction
B-THE
40
45
50
30/,5B-THF
4,0001
~~a,ooo I 3,500
~2,5~-
Results for male and female rats are shown in Table 4; and Fig. 3(a and b), respectively. In both sexes the metabolic profiles were much simpler than those produced in microsomal incubations. The major metabolite in the female was 3ct,5fl-THF (84.2%); 3~t,5fl-THE and 20flDHF were minor metabolites. In contrast, the major metabolite in male cytosol was 20fl-DHE (65.5%), with some 20fl-DHF (12.4%), whereas the principal metabolite in microsomes was 20fl-DHF. A number of more polar unidentified metabolites were also produced.
Microsomes prepared from livers of dexamethasone treated rats showed increased 6/thydroxylase activity (Table 5; Fig. 4), consistent with the increased level of CYP3A on a Western blot (Fig. 5). There was also an increase in 6~-OHF.
Cortisol
\
'6 2,0OO !
50/.-DHF
Table 4. Metabolism of cortisol in hepatic cytosol of male and female rats
~ 1,5~-
% Formation of metabolite
1,~i 5OO
~~,5B-THE
60C-OHF
0
. . . .
0
,
5
. . . .
I
R!
. . . .
10
i
. . . .
15
i
. . . .
20
,
. . . .
25
i
. . . .
30
,
. . . .
35
i
. . . .
40
~
. . . .
45
i
50
RetentionTime(minutes) Fig. 2. C o n c e n t r a t i o n d e p e n d e n t m e t a b o l i s m o f [3 H ] c o r t i s o l b y f e m a l e r a t h e p a t i c m i c r o s o m e s (in vitro): (a) 1 # M ; (b) 1 0 # M ; a n d (c) 1 0 0 g M .
Metabolite
(rain)
Female rat
Male rat
6a-OHF 20fl-DHF 20fl-DHE Cortisol 3~t,Sfl-THF 3~t,58-THE
5.4 17.5 20.0 26.0 28.0 40.0
2.17_+0.12 2.2 _+ 2.27 -5.98 _+ 1.96 84.23 +_ 2.10 5.38 -+ 1.56
2.31 _+0.07 12.42 4- 1.85 65.48 4- 4.25 8.93 _+ 1.22 ---
Additional unknown metabolites (1, 2, 3, total 10%) were evident in male rat cytosol at R, 8.6, 10.7, 13.3 min--see Fig. 4.
450
S.M. ABEL et al.
(a) 20B-DHE
4OOO
3500 Unknown~
3000
Unknown2
~2500
2ooo
~
nknowns
o 1,oo rr i 0 0 0 :
\ \
500
~
/X/
Cortisol
0 0
5
10
15
20
25
30
35
40
45
Retention Time (min) (b) 3O/,5B-THF
4OO0
35003000
.> 20o0-
Cortisol
1000 2 500
5
10
15
20
25
30
35
40
45
Retention Time (rain)
Fig. 3. Sex differences in the metabolism of [3H]cortisol (1/aM) by rat cytosol: (a) male cytosol; and (b) female cytosol. DISCUSSION
The present work has revealed considerable interspecies variation, both quantitative and qualitative, in the metabolism of cortisol by liver fractions from man and several laboratory animals. In most cases, 6fl-hydroxylation is a minor pathway with reduction of the corticosteroid as the favoured reaction. However, the site of reduction of the steroid varied considerably between species. The female rat gives an interesting microsomal metabolic profile, with the major Table 5. Sex differences in the metabolism of [3H]cortisol to 6fl-OHF (CYP3A) by liver microsomes from control and dexamethasone-induced rats in vitro % Formation of metabolite
Control Induced
Male rat
Female rat
6.33 10.96
2.26 8.6
n = 2, for both control and induced rats.
metabolite being 3~,5fl-THF. In human liver, A-ring reduction of cortisol involves a coupled enzyme system (A4 reductase, 3~-oxidoreductase [4]), since none of the dihydrocortisol intermediate is detected. However, the presence of a 5~-reduced metabolite in the female rat (5~-DHF), suggests that A-ring reduction may be catalysed by two separate enzymes. No 5flD H F was detected, possibly because the pH of the incubation was suboptimal: optimum pH ranges for the stability of the two enzymes have been shown to be 7.0-7.5 for 5~-reductase [16] and 5.5-6.0 for 5fl-reductase[17]; the present incubations being carried out at pH7.4. However, Forchelli et al. [16] provided evidence that only 5~-reductase is present in the female rat. It is important to note that the intermediate in the A-ring reduction pathway is the 5~-DHF, but the final product is 3~,5fl-DHF, suggesting that there is some conformational inversion at C 5 while bound to the 3-oxidoreductase enzyme. The metabolite of R, 12 min was identified as a cortol (A-ring and side chain reduction). This metabolite does not absorb between 200-300 nm, as there are no unsaturated bonds in the molecule and thus has been identified by mass spectrometry alone. The cortol is only generated at concentrations between 1 and 10/~M (Fig. 2), and an increase in 5~-DHF accompanies the decrease in the amount of cortol produced. This suggests that the 3~-HSD enzyme is easily saturable, and possibly that the product of A-ring reduction may inhibit C20 reduction at higher concentrations. The major biotransformation in male rat microsomes is the stereospecific reduction of the C2o keto group to a fl-hydroxyl moiety. There is a complete absence of this pathway in microsomes from female rats, guinea-pigs, hamsters and mice. Since both 20fl-DHF and 20fl-DHE are formed in the male rat microsomes but no cortisone is detected, it is not clear whether the route of metabolism involves reduction of cortisol at C20 to give 20fl-DHF and then oxidation at C~] to give 20fl-DHE, or whether there is an interconversion between cortisol and cortisone followed by reduction of all the cortisone to 20fl-DHE. 6fl-OHF is a minor metabolite in both male and female rat microsomes. The hamster and guinea-pig microsomes produce qualitatively similar but quantitatively different metabolic profiles. The hamster produces a small amount of cortisone, the preferred pathway being 6fl-hydroxylation. This provides
5
5
10
20
25
30
Retention Time (rain)
15
~, Cortisol
IUnkn°wn
208.DHE
2(~6-DHF
15 20 25 30 Retention Time (rain)
~
35
35
40
40
45
45
oooi
11oo-~ IO0O-~
13oo7 12oo-~
0 0
I00-
200-
~: 4 ~ -
600-
8o7
~900
(dl
100S
200 S
3OO
5
5
10
20
25
~)
Cortisol.
30
15
20
25
30
RetentionTime(rain)
35
35
40
40
3OL58-THE
30t,5B-THF
Retention Time (rain)
Corto(
10
15
Cortot
4oo2 68-OHF ,OHF I
.,~ go0-
.~ 7o0-
~900
I000-~
1200-~ 11001
(c] ~3007
45
45
Fig. 4. Sex-differences in the metabolism of {3H]cortisol (I p M) by liver microsomes from control and dexamethasone-induced rats in vitro: (a) control male; (b) induced male; (c) control female; and (d) induced female.
0
10
.
Cortisol
20t3-DHE
20t3.1 ~HF
6B-OHF
.~200- =,~u=
°300_
40O-
0
O'
4oo2
( b / 500-
o
.~" ZOO1
g. ~ 2
11oo 100o-
12oo-
(a) 130o-
O" O
452
S.M. ABEL et al.
evidence to support Lipman et a/.[18] who suggested that 6//-hydroxylation is a compensatory route when there was an inability to perform A-ring reduction. Further work needs to be undertaken in this area in view of reported sex differences in the hamster [19]. The low levels of 6//-hydroxylase activity detected in guinea-pig microsomes can perhaps be explained by the observation that 6fl-OHF production in vivo is strain dependent and that 6fl-OHF has been shown to undergo variable and extensive metabolism itself in vivo [20]. It has been suggested previously by Burstein and Kimball[21], that there is great interindividual variation in the urinary excretion of corticosteroids by the guinea-pig, and that two distinct phenotypes exist; those with high 6fl-OHF excretion type A, and those with low polar corticosteroid excretion type B[22]. Dunkin-Hartley guinea-pigs are known to display this phenotypic variation, but it is possible our sample size was too small to detect it. The major metabolite seen in mouse microsomes was cortisone. 6//-OHF was formed as a minor metabolite; it has previously been shown that mice possess an enzyme comparable to CYP3AI in rats, which is capable of 6//hydroxylation [23]. Cortisol metabolism in liver cytosol of male and female rats differs both qualitatively and quantitatively. The female gives a metabolite profile which is similar to that produced in human liver cytosol, and contains metabolites which have all been rigorously identified [7]. The major pathway is the formation of 3e,5//-THF. Since no intermediate was detected, the enzyme system responsible is different from that found in female microsomes, which yield 5c~-DHF and is possibly comprised of a coupled oxidoreductase as in humans. The male rat cytosol produces 20fl-DHE as the major metabolite. This contrasts with microsomal incubations in which C20 reduction is also the major biotransformation, but the final product is 20fl-DHF, suggesting much higher l l//-oxidoreductase activity in the cytosol. A number of unidentified metabolites were also generated which when combined account for about 10% of the metabolism. Possible thoughts as to the identity of some of these, include polyhydroxyl metabolites, or even Phase II conjugates. It is known, for example, that glutathione transferase is a cytosolic enzyme [24], which will attack an electrophilic centre.
52 kDa
m
D
Fig. 5. Western Blot showing the induction of CYP3A by dexamethasone in male and female rats.
In agreement with others [25] it was found that dexamethasone is a substantial inducer of CYP3A in rats (Fig. 5). Simultaneous with the increase in 6fl-hydroxylase activity, there was a decrease in A-ring reduction, providing further evidence in support of Lipman et al. [18]. The unknown male rat metabolite, more polar than cortisol but less polar than those having undergone C20 reduction, is likely to be a product of the CYP3A subfamily as it is formed only by induced microsomes. Recent work on species differences in overall digitoxin metabolism [26], showed large variations and the rank order of: hamster > rat > guinea-pig > mouse > human. In all species conversion ofdigitoxin to dt2 is catalysed predominantly by CYP3A, whereas enzymes other than CYP3A catalyse other hydroxylations (12-, 16-, 17-). Further evidence for high CYP3A activity in the hamster was obtained by Halvorson et al. [27] in work on testosterone 6//-hydroxylation. It has been suggested that CYP3A1 (the rat 6fl-hydroxylase enzyme) is specific for the //face of the testosterone molecule[28], with minimal hydroxylation on the c~-face. However, it is evident that 6~-OHF was also formed by microsomes of all species and indeed exceeded 6fl-OHF tbrmation in microsomes
Species difference in cortisol metabolism
from the female rat. Furthermore, an increase in formation of 60t-OHF was noted in the dexamethasone induction experiment (Fig. 4), but at present it is not known whether this is due to induction of 60t-hydroxylase activity, or whether 6~- and 6fl-OHF epimeres are formed from a common intermediate, the production of which increases following induction. In conclusion, we have demonstrated that there are major differences between species in the in vitro routes of metabolism of cortisol. There are also marked sex differences in the rat. The hamster provides the simplest metabolic profile and therefore provides the simplest model with which to explore further factors which regulate cortisol 6fl-hydroxylase activity. Acknowledgements--We would like to acknowledge Mrs C. McLean for her assistance with the Western Blot analysis. The TS250 mass spectrometer was purchased with generous grants from the Wellcome Trust and the University of Liverpool.
REFERENCES 1. Katz F. H., Lipman M. M., Frantz A. G. and Tailor J. W.: The physiological significance of 6/3-OHF in human corticoid metabolism. J. Clin. Endocr. Metab. 22 (1962) 71-77. 2. Zumoff B., Bradlow L., GaUagher T. F. and Hellman L.: Cortisol metabolism in cirrhosis. J. Clin. Invest. 46 (1967) 1735-1743. 3. Meigs R. A. and Engel L. L.: The metabolism of adrenocortical steroids by human tissues. Endocrinology 69 (1961) 152-162. 4. Iyer R. B., Binstock J. M., Schwartz I. S., Gordon G. G., Weinstein B. I. and Southren A. L.: Human hepatic cortisol reductase activities: enzymatic properties and substrate specificites of cytosolic cortisol A4-5/3reductase and dihydrocortisol-3~-oxidoreductase(s). Steroids 55 (1990) 495-500. 5. Park B. K. and Kitteringham N. R.: Relevance and means of assessing induction and inhibition of drug metabolism in man. Prog. Drug. Metab. 11 (1989) 1-59. 6. Ged C., Rouillon J. M., Pichard L., Combalbert J., Bressot N., Bories P., Michel H., Beaune P. and Maurel P.: The increase in urinary excretion of 6/3-hydroxycortisol as a marker of human hepatic cytochrome P-450IIIA induction. Br. J. Clin. Pharmac. 28 (1989) 373 387. 7. Abel S. M., Maggs J. L., Back D. J. and Park B. K. Cortisol metabolism by human liver in vitro. (I). Metabolite identification and interindividual variability. J. Steroid Biochem. Molec. Biol. 43 (1992) 713 719. 8. Mode A. and Rafter I.: The sexually differentiated A4-3-ketosteroid 513-reductase of rat liver. J. Biol. Chem. 260 (1985) 7137-7141. 9. Stenberg A.: Development, diurnal and oestrous cycledependent changes in the activity of liver enzymes. J. Endocr. 68 (1976) 265-272. 10. Smith D. A.: Species differences in metabolism and pharmacokinetics: Are we close to an understanding? Drug Metab. Rev. 23 (1991) 355-373.
453
11. Waxman D. J.: Interactions of hepatic cytochromes P-450 with steroid hormones. Biochem. Pharmac. 37 (1988) 71-84. 12. Purba H. S., Maggs J. L., Orme M. L'E., Back D. J. and Park B. K.: The metabolism of 17~-ethinylestradiol by human liver microsomes: formation of catechol and chemically reactive metabolites. Br. J. Clin. Pharmac. 23 (1987) 447-453. 13. Lowry O. H., Rosebrough N. J., Farr A. L. and Randall R. J.: Protein determination with the Folin phenol reagent. J. Biol. Chem. 193 (1951) 265-275. 14. Laemmli U. K.: Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227 (1970) 680-685. 15. Towbin H. and Gordon J.: Immunoblotting and immunobinding--current status and outlook. J. Immun. Methods 72 (1984) 313-340. 16. Forchielli E., Brown-Grant K. and Dorfman R. I.: Steroid A4-hydrogenases of rat liver. Proc. Soc. Exp. Biol. Med. 99 (1958) 594-596. 17. Tomkins G. M.: The enzymatic reduction of A4-3-keto steroids. J. Biol. Chem. 225 (1957) 13-24. 18. Lipman M. M., Katz F. H. and Jailer J. W.: An alternate pathway for cortisol metabolism: 6fl-OHF production by human tissue slices. J. Clin. Endocr. Metab. 22 (1960) 268-272. 19. Yates F. E., Herbst A. L. and Urquhart J.: Sexdifferences in rate of A-ring reduction of A4-3-keto steroids in vitro by the rat liver. Biochem. J. 63 (1958) 887 902. 20. Challiner M. R.: The assessment of 6fl-OHF as an /n vivo index of microsomal mixed function oxidase activity. Ph.D. Thesis, Liverpool University, Liverpool (1982). 21. Burstein S. and Kimball H. L.: Quantitative paper chromatography of C2~Os and C2~O6 corticosteroids from guinea pig urine extracts. Analyt. Biochem. 4 (1962) 132. 22. Burstein S., Kimball H. L. and Bhavnani B. R.: Urinary corticosteroid excretion patterns in guinea pigs: 2 main phenotypes. Steroids 2 (1963) 195. 23. Wrighton S. A., Schuetz E. G., Watkins P. B., Maurel P., Barwick J., Bailey B. S., Hortle H. T., Young B. and Guzelian P.: Demonstration in multiple species of inducible hepatic cytochromes P-450 and their mRNAs related to the glucocorticoid-inducible P-450 of the rat. Molec. Pharmac. 28 (1985) 312-321. 24. Vos R. M. E. and Van Bladeren P. J.: Glutathione S-transferases in relation to their role in the biotransformation of xenobiotics. Chem. Biol. lnterac. 75 (1990) 241-265. 25. Heuman D. M., Gallagher E. J., Barwick J. L., Elshourbagy N. A. and Guzelian P. S.: Immunochemical evidence for induction of a common form of hepatic cytochrome P-450 in rats treated with pregnenolone-16~-carbonitrile or other steroidal or non-steroidal agents. Molec. Pharmac. 21 (1981) 753-760. 26. Eberhart D. C., Gemzik B., Halvorson M. and Parkinson A.: Species differences in the toxicity and cytochrome P-450IIIA-dependent metabolism of digitoxin. Molec. Pharmac. 40 (1991) 859-867, 27. Halvorson M. A. and Parkinson S. A.: Species difference in testosterone hydroxylation by liver microsomes: studies with antibody against rat cytochrome P-450p. Toxicologist 9 (1988) 9-14. 28. Namkung M. J., Yang H. L., Hulla J. E. and Juchau M. R.: On the substrate specificity of cytochrome P-450IIIAI. Molec. Pharmac. 34 (1988) 628437.