Content uploaded by Alberto Sulli
Author content
All content in this area was uploaded by Alberto Sulli on Apr 23, 2015
Content may be subject to copyright.
British Journal of Rheumatology 1997;36:433–439
TESTOSTERONE METABOLISM AND CYCLOSPORIN A TREATMENT
IN RHEUMATOID ARTHRITIS
M. CUTOLO, M. GIUSTI,* B. VILLAGGIO, A. BARONE, S. ACCARDO, A. SULLI,
O. GRANATA,† G. CARRUBA† and L. CASTAGNETTA†
Division of Rheumatology, Department of Internal Medicine and *Department of Endocrinology and Metabolism, University of
Genova, Viale Benedetto XV, Genova
16132
and †Institute of Oncology, University of Palermo and Experimental Oncology Unit,
Palermo Branch of Scientific Institute for Cancer of Genova, c/oM.Ascoli Cancer Hospital Centre,
Palermo, Italy
SUMMARY
A constant dose-dependent side-effect in cyclosporin A (CSA)-treated patients is the appearance of hypertrichosis; this occurs
in both sexes and suggests an androgenizing activity. To determine the influence of CSA on peripheral androgen metabolism,
we evaluated in rheumatoid arthritis (RA) patients treated with low-dose CSA (3.5 mg/kg/day), during a period of 12 months,
plasma levels of testosterone (Tes) and of 5a-androstane-3a,17b-diol glucuronide (Adiol-G), an important peripheral Tes
metabolite. Clinical and laboratory parameters of RA were also monitored. Furthermore, the metabolism of physiological
concentrations of Tes (1 × 10
−8
) was evaluated in primary cultures of RA synovial macrophages (Mf) in the presence of CSA
concentrations close to the pharmacological immunosuppressive doses (100–500 ng/ml). At the final time of observation (12
months), a significant increase in the mean plasma Adiol-G level was observed in patients of both sexes. The increase was evident
after 1 month of treatment in male patients (P Q 0.01) and after 3 months in female patients (P Q 0.05). Almost all the patients
experienced the side-effect of a low-degree hypertrichosis after a mean period of 1–2 months. No significant correlations with
the laboratory parameters of the disease were observed. Results from in vitro experiments on Tes metabolism by cultured synovial
Mf showed at 24 and 48 h, in the presence of CSA, a significantly (P Q 0.0001) greater formation of dihydrotestosterone and
increased amounts of other Tes metabolites, including androstenedione, androsterone and epiandrosterone, when compared to
untreated controls. In conclusion, the appearance of a dose-related hypertrichosis and the increase in plasma androgen
metabolites (i.e. Adiol-G) in CSA-treated patients, as well as the hormonal metabolic effects on cultured synovial Mf, should
be regarded as possible markers of the influence of CSA on peripheral androgen metabolism at the level of target cells.
K : Testosterone, Androgens, Rheumatoid arthritis, Synoviocytes, Macrophages, Cyclosporin A, Hypertrichosis.
C A (CSA) is the first non-steroidal drug
having clearly specific immunosuppressive activities
that has been used extensively in the prevention of
allograft rejection, as well as in the treatment of
immunomediated diseases, including rheumatoid
arthritis (RA), psoriasis (Ps) and psoriatic arthritis
(PsA) [1–4].
CSA may have more than one mode of action and
can act on different cell types involved in the immune
response, including T lymphocytes and antigen-
presenting cells (APCs) such as macrophages [5, 6]. A
nearly constant dose-dependent side-effect in CSA-
treated patients is the appearance of hypertrichosis,
suggesting an androgenizing activity [7]. However,
previous studies failed to demonstrate increased plasma
androgens in patients treated with CSA [8]. On the
contrary, decreased plasma testosterone (Tes) in
long-term treatment has been described [9]. Recent
reports have suggested a dose-related influence of CSA
administration on Tes peripheral metabolism in
patients affected by insulin-dependent diabetes (IDD)
and PsA [9, 10]. Tes, and its derived metabolite
dihydrotestosterone (DHT), may exert suppressive
activities on androgen receptor-positive cells involved
in the immune response, including macrophages
[11, 12]. In particular, it is suggested that androgens
might exert immunosuppressive activities in human
and murine synovial macrophages (Mf) that are
androgen receptor-positive cells able to metabolize Tes
and to influence cytokine production [13–19].
Based on these observations, we decided to
investigate the plasma concentrations of Tes and
5a-androstane-3a,17b-diol glucuronide (Adiol-G), an
important Tes metabolite, in patients affected by RA
and treated with low-dose CSA (3.5 mg/kg/day),
during a period of 12 months. Adiol-G is considered a
specific hormonal marker in ‘idiopathic hypertrichosis’
and the increase in plasma Adiol-G was suggested to
be associated with the hypertrichosis in CSA-treated
patients [9, 10]. To verify in vitro the influence exerted
by CSA on androgen metabolism, we evaluated
metabolic pathways of Tes in primary cultures of RA
synovial Mf in the presence of pharmacological
concentrations of CSA.
MATERIALS AND METHODS
Patients
Eleven patients (six men and five women, mean
age 2 ..=4825 yr) fulfilling the American College
of Rheumatology criteria for adult RA were entered
into the study after informed consent was obtained
[20]. All patients were unresponsive or very poorly
Submitted 9 July 1996; revised version accepted 18 October 1996.
Correspondence to: M. Cutolo, Division of Rheumatology,
Department of Internal Medicine, Viale Benedetto XV, Genova,
Italy.
= 1997 British Society for Rheumatology
433
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 4434
responsive to conventional treatment, and the arthritis
was not adequately controlled by non-steroidal
anti-inflammatory drugs. Patients were excluded from
the study with CSA in the presence of one or more of
the following conditions, in agreement with recent
International Consensus Conference indications:
previous treatment with CSA, sulphasalazine,
methotrexate, etretinate, or other immunosuppressive
agents within at least 8 weeks before proposed entry
into the study, serum creatinine q110 mmol/l,
proteinuria q0.5 g/24 h, leucopenia, thrombocyto-
penia or abnormal liver function test, hypertension
(systolic blood pressure q180 mmHg or diastolic
blood pressure q100 mmHg after 5 min in a sitting
position), malignancy or history of malignancy,
immunodeficiency, uncontrolled infections, pregnancy,
breast feeding, impairment of cardiovascular or
cerebral function, IDD, drug or alcohol abuse [21].
None of the patients received other drugs reported to
induce hypertrichosis or hirsutism [22].
Cyclosporin A treatment
CSA, divided into two oral doses, was generally
started at 3.5 mg/kg/day. The dose of CSA had to be
reduced by 0.5 mg/kg/day if there was an increase in
serum creatinine of q30% above basal values,
hypertension, not tolerated hypertrichosis or gingival
hyperplasia. The CSA dose was reduced by 25% if total
bilirubin or liver transaminases increased by q100%
over baseline. Patients were allowed to continue their
non-steroidal anti-inflammatory therapy during the
study.
Assessment of disease activity
Each patient was examined monthly by the same
rheumatologist for the first 3 months, thereafter every
3 months for 12 months. At each visit, the number of
swollen joints and the number of tender joints were
evaluated. Patients were monitored for the appearance
of hypertension and carefully examined for gingival
hyperplasia. Hypertrichosis was evaluated following
the Ferriman and Gallwey score [23]. Laboratory
parameters to assess CSA therapeutic activity and
tolerance included complete blood and differential
count, erythrocyte sedimentation rate (ESR), C-
reactive protein (CRP), serum creatinine, liver trans-
aminases and alkaline phosphatase. The analyses were
performed by standard methods at baseline and every
4–8 weeks.
Plasma androgen assays
After collection (at baseline, every week for the first
month and every 4 weeks thereafter) plasma for
hormonal analysis was frozen at −10°C until assay.
Plasma Adiol-G was measured by radioimmuno-
assay (by evaluating both 3-a and 17-b androstanediol
glucuronide) without previous extraction (Diagnostic
System Laboratories Inc., Webster, TX, USA). Mean
concentrations of Adiol-G established in normal
controls were 3.32 2 1.80 nmol/l in pre-menopausal
women (range 1.06–11.50 nmol/l), 2.19 2 1.19 nmol/l
in post-menopausal women (range 0.21–12.80 nmol/l)
and 6.04 2 3 nmol/l in men (range 4.24–38.86 nmol/l).
The lowest detectable level of Adiol-G was 0.25 nmol/l
at the 95% confidence limit. An inter-assay coefficient
of variation of 7.3% and an intra-assay coefficient of
variation of 6.7% were found. Serum total Tes was
measured by direct immunoassay as previously
described [24]. Mean concentrations of total Tes
established in normal women and men were
0.78 2 0.27 and 20.10 2 4.50 .. nmol/l, respectively.
An inter-assay coefficient of variation of 7% was
found.
Preparation of synovial macrophage culture
Synovial tissue samples were obtained from three
women (mean age 49.6 2 14 yr) and three men (mean
age 53.7 2 15 yr) who fulfilled the American College of
Rheumatology criteria for adult RA and were
undergoing surgical synovectomy of the knee [20, 25].
All the RA patients were in Steinbrocker class III.
At the time of surgery, all of the patients were being
treated exclusively with non-steroidal anti-inflamma-
tory drugs. None had received any oral or intra-artic-
ular corticosteroid therapy or disease-modifying
anti-rheumatic drugs during the 4 months preceding
the investigation, nor was the use of the oral
contraceptive pill, 1,25-dihydroxyvitamin D3 or any
hormone replacement therapy allowed [26–28]. All
subjects had normal liver, renal, prostate and thyroid
functions, and were within 20% of ideal body weight.
The synovial tissue, soon after surgery, was dissected
away from fatty, capsular and cartilaginous
components, cut into 3–5 mm pieces, washed in Hank’s
balanced salt solution and incubated in collagenase
(1 mg/ml; Sigma Chemical Co., St Louis, MO, USA)
for 3 h at 37°C with constant agitation. The digest was
then passed through a wire mesh with a pore size of
200 mm to separate dissociated cells from tissue debris.
The cells were washed twice in collagenase and
counted. The mean cell yield was 4.9 × 10
6
(range
0.53–16.8 × 10
6
) and the mean viability was 88%
(range 69–98%). The cells were then cultured with
RPMI supplemented with 10% fetal calf serum (FCS)
at 37°C in a humidified 5% CO
2
atmosphere in air.
After 3 h, non-adherent cells were removed and the
adherent cells were cultured with RPMI supplemented
with 10% FCS at 37°C in a humidified 5% CO
2
atmosphere in air for 24 h. Every 24 h, cytocentrifuge
preparations (800 g × 3 min) of suspended cells (gentle
scraping) were obtained and stained as previously
reported with specific monoclonal antibodies directed
toward the macrophage-specific antigens (Ber-MAC3;
Dakopatts, Copenhagen, Denmark); they were found
to be q90% Mf [14].
Testosterone metabolism in primary cultures of synovial
macrophages
Methodological approaches and procedures used to
measure metabolic pathways of steroids have been
established and optimized previously [29].
Adherent synoviocytes were harvested by mild
CUTOLO ET AL.: TESTOSTERONE METABOLISM AND CYCLOSPORIN A 435
TABLE I
Variations of plasma levels of testosterone (Tes) and Adiol-G in six male RA patients
Months of treatment with CSA
Parameter 0 1 2 3 6 12
Adiol-G 2 .. (nmol/l) 4.93 2 0.83 8.78 2 1.77* 7.56 2 1.99 8.90 2 2.05** 7.93 2 1.21 9.17 2 0.72***
Tes 2 .. (nmol/l) 19.10 2 4.67 21.77 2 5.20 17.95 2 2.80 20.17 2 5.23 22.98 2 6.76 24.47 2 6.48
*P Q 0.01; **PQ 0.05; ***P Q 0.001.
trypsinization, counted in a haemocytometer and
plated onto 6-well tissue culture plates at a density of
0.5–1 × 10
6
cells/dish. After 24 h, cells were washed
twice with PBS-A and the medium substituted with
2 ml FCS-free, phenol red-free RPMI medium,
containing 10
−8
tritiated Tes as precursor (1,2,6,7-
3
H(N)T; sp. act. 92.1 Ci/mmol; DuPont de Nemours
Italiana Spa, Milan, Italy). The tritiated precursor was
periodically checked using high-performance liquid
chromatography (HPLC) prior to experimental use.
Furthermore, parallel dishes with medium plus
precursor without cells were used to control precursor
degradation during each experiment. Parallel dishes
received two different concentrations of CSA (100 and
500 ng/ml) (Sandoz Ltd, Basel, Switzerland)
throughout the culture period. Following 24–48 h
incubation, the medium was transferred to sterile
plastic tubes (Costar) and stored at −80°C until
analysis. Aliquots (100 ml) of the cell lysates were
therefore used to estimate DNA content, as described
elsewhere [30]. Androgen extraction was carried out on
1 ml of medium with 10 ml of diethyl ether. After
mixing at 4°C for 30 min, the resulting aqueous
phase was freeze–dried in a Speed Vac evaporator–
concentrator (Savant Instruments Inc., Farmingdale,
NY, USA) and then resuspended using 970 mlof
acetate buffer (0.75 , pH 5.0) added with 30 mlof
glusulase enzyme mixture (DuPont Co., Wilmington,
DE, USA) and incubated at 37°C for 18 h to hydrolyse
androgen conjugates (sulphates and glucuronides).
Following incubation, samples were transferred to
glass vials and extracted again using 10 ml diethyl ether
as before. The two ether phases (free and hydrolysed
steroids) were evaporated to dryness under nitrogen
and then stored at −20°C until required for analysis
using reverse-phase high-performance liquid chroma-
tography (RP-HPLC). The HPLC system consisted of
a Model 324 system, equipped with a Model 160 UV
detector set at 280 nm (Beckman Instruments Inc.,
Berkeley, CA, USA) and with an ‘on-line’ Flo-One/
beta (Model 1C) three-channel radiometric detector
(Radiomatic Instruments, High Wycombe). Steroids
were separated under isocratic conditions using a
Spherisorb ODS-2 (Aldrich Chimica, Milan, Italy)
column (250 × 4.6 mm ID) and 45% acetonitrile in
citric acid (0.05 ) as optimized mobile phases for
androgen separation, with a flow rate of 1 ml/min.
Routine data integration was automatically achieved
and computed in net c.p.m. by a Flo-One/beta F1B
program (Radiomatic, Tampa, FL, USA). Limits for
radiometric detection were 050 c.p.m. (corresponding
to 1.1 fmol) using [1,2,6,7-3H]Tes (sp. act. 92.1) as also
reported previously [29]. The Tes metabolites evaluated
included DHT, androstenedione (D4-Ad), androstane-
dione (5d-Ad), 5a-androstane-3a,17b-diol (Adiol),
androsterone (A) and epiandrosterone (Epi-A).
Statistical analysis
Non-parametric procedures were employed to
evaluate the statistical significance between basal and
final observation times for the considered values.
Statistical analysis was carried out using the time series
analysis and the correlation coefficient between the
hormonal parameters. Results on plasma steroid
concentrations were assessed by paired t-test in
individuals. The two-tailed paired Student’s t-test
(95% confidence limits) was used to compare Tes
metabolic rates in controls and CSA-treated synovial
Mf.
RESULTS
Plasma androgen variations and clinical response
In male RA patients, a significant increase in the
mean plasma Adiol-G concentrations (P Q 0.01) was
observed after 1 month of CSA treatment (see Table I).
The increase was confirmed after 3 and 12 months of
treatment (P Q 0.05 and P Q 0.001, respectively).
In female RA patients, an appreciable increase in the
mean plasma Adiol-G concentration was observed
after 1 and 2 months of CSA administration, with a
significant increase after 3, 6 and 12 months (P Q 0.05
and P Q 0.001 at 6 and 12 months, respectively)
(Table II).
No significant variations of Tes concentrations were
TABLE II
Variations of plasma levels of testosterone (Tes) and Adiol-G in five female RA patients
Months of treatment with CSA
Parameter 0 1 2 3 6 12
Adiol-G 2 .. (ng/ml) 2.97 2 0.63 3.50 2 0.73 4.14 2 0.71 4.80 2 0.83** 6.15 2 1.97*** 5.56 2 1.58***
Tes 2 .. (ng/dl) 0.20 2 0.06 0.22 2 0.05 0.20 2 0.04 0.18 2 0.04 0.17 2 0.03 0.18 2 0.04
**P Q 0.05; ***PQ 0.001.
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 4436
TABLE III
Androgen levels in conditioned media after 24 h of incubation with CSA. Mean 2 .. values of three different experiments expressed as fmol/ml
Tes DHT D4-Ad 5a-Ad A Epi-A Total
Control 6822 2 106 60 2 1 345 2 76224221323 7250 2 109
CSA (100 ng/ml) 7276 2 10 84 2 1 370 2 332211222421 7759 2 12
(P) (0.002) (0.0008) (0.005) (0.01) (0.004)
CSA (500 ng/ml) 6113 2 80 68 2 3 333 2 69225211522 6538 2 81
(P) (0.0008) (0.01)
observed during the study in all the patients. Regarding
male RA patients, a slight increase in Tes (not
significant) was present after 12 months (24.47 vs basal
19.10 nmol/l). Almost all the patients experienced a
different degree of hypertrichosis after a mean period
of 6–8 weeks. The Ferriman and Gallwey score was
lower than eight in all patients before treatment. In
particular, in women the mean score changed from
basal 4 2 1to1223 after 7–8 weeks and persisted
until the end of the study. In men, the mean score
started at 6 2 2 and changed to 14 2 2 and 15 2 3 after
6–8 weeks and 12 months, respectively. No liver
damage or neurological deterioration was noted during
the observation period. No significant variations were
observed during the study period concerning the
laboratory parameters investigated (ESR, CRP, Hb),
excluding a significant reduction of the CRP (P Q 0.01)
in female RA patients after 12 months (data not
shown). However, the lack of clear effects exerted by
CSA on acute-phase reactants seems to confirm other
reports [31, 32]. Concerning the clinical parameters, a
significant improvement (P Q 0.05) of the number of
swollen joints and of the number of tender joints was
observed after 12 months in RA patients, in agreement
with recent reports (data not shown) [33].
Testosterone metabolism in primary cultures of synovial
macrophages
Overall results from in vitro experiments on Tes
metabolism by cultured RA synovial Mf are reported
in Tables III and IV. These data indicate that CSA
treatment, even in the short term (24 h), significantly
affected the 17b-hydroxysteroid dehydrogenase
(17bHSD) oxidative activity, which presides over the
conversion of Tes into D4-Ad. At the same time, CSA
also increased the 5a-reductase activity, which presides
over conversion of Tes into DHT.
DHT formation was stimulated significantly by both
100 and 500 ng/ml of CSA after 24 h incubation (30
and 25% above control, respectively); on the other
hand, the conversion of Tes into D
4
-Ad was only
slightly enhanced by CSA addition (nearly 7% above
controls at the 100 ng/ml dose) (see Table III).
Following 48 h incubation, the increase in DHT
production was less pronounced, although still
significant, at 500 ng/ml (13%) of CSA. In contrast, the
Tes conversion to D
4
-Ad was much increased, reaching
a peak stimulation of 22% above control with the
100 ng/ml dose (see Table IV). CSA addition produced
an opposite effect on the 5a-Ad formation, the latter
being inhibited and stimulated at 100 or 500 ng/ml,
respectively; however, these differences from controls
at both 24 and 48 h were not significant (see Table IV).
The CSA-induced stimulation of DHT production
by treated synoviocytes was also corroborated by the
formation of Adiol (DHT metabolite) in some assays,
although inconsistent and in small amounts. The
fact that this DHT derivative is only occasionally
observed in vitro may be ascribed to the experimental
conditions used (i.e. small cell number, physiological
concentrations of labelled precursor, characteristiques
of conditioned media), probably leading to Adiol levels
below the limits of detection.
DISCUSSION
In the present study, we have shown that the
treatment of RA patients with low-dose CSA induces
within a few weeks a significant increase in the mean
plasma Adiol-G concentration, suggesting an increased
peripheral metabolism of Tes. Although we cannot
exclude that, as previously proposed by others, the
increase in Adiol-G is due to elevated metabolism of
circulating DHEAS, at least in women, our data on
primary cultures of RA synovial macrophages support
the concept that CSA mostly increases Tes transform-
ation into DHT [34, 35]. In addition, several studies
have shown significantly lower levels of plasma
(DHEAS) dehydroepiandrosterone sulphate in RA
patients, particularly in women, suggesting a reduced
TABLE IV
Androgen levels in conditioned media after 48 h of incubation with CSA. Mean 2 .. values of three different experiments expressed as fmol/ml
Tes DHT D4-Ad 5a-Ad A Epi-A Total
Control 65012 54 99 2 1 322 2 31 7 2 310211524 6954 2 53
CSA (100 ng/ml) 6933 2 178 111 2 1 392 2 15 6 2 312211924 7474 2 196
(P) (0.0001) (0.02)
CSA (500 ng/ml) 6521 2 16 114 2 1 366 2 18 12 2 49231523 70372 122
(P) (0.00005)
CUTOLO ET AL.: TESTOSTERONE METABOLISM AND CYCLOSPORIN A 437
role for the DHEAS-related activities in these patients
[36, 37]. The results from in vitro studies clearly indicate
that cultured RA synovial Mf from these patients
are endowed with key enzyme activity of steroid
metabolism and that CSA may have a role in the
regulation of androgen metabolic pathways in these
cells. The evidence that the CSA-induced increase in
DHT production is obtained using pharmacological
CSA concentrations and short-term exposure strongly
supports the view that this effect may be biologically
important. This assumption is also further substanti-
ated by the appearance in CSA-treated cells of
Adiol (though in small amounts), which represents a
typical DHT derivative in androgen-responsive tissues,
including the prostate [38, 39].
Both Tes and DHT may exert suppressive activities
on androgen receptor-positive cells involved in the
immune response [40, 41]. Our results may suggest that
the previously reported immunosuppressive activity
exerted by CSA on an earlier stage of the immune
response, i.e. antigen presentation by macrophages and
by other APCs to T lymphocytes, should, at least
partially, be associated with the activation, induced by
the drug, of the peripheral androgen metabolism at the
level of these hormone-responsive cells [16–19, 42]. In
addition, preliminary results from the present study
have shown a dose- and time-dependent reduction (24
and 48 h) of granulocyte-monocyte colony-stimulating
factor (GM-CSF) production in conditioned media of
cultured RA synovial Mf in the presence of both Tes
and CSA (data not shown). Since the GM-CSF is a
macrophage activator and, among other effects, causes
release of tumour necrosis factor-a (TNF-a), induction
of major histocompatibility (MHC) class II molecules
and potentiation of antigen processing, the observed
inhibition of GM-CSF by CSA should further explain,
at least partially, some of the immunosuppressive
mechanisms exerted by the drug [43, 44]. As a matter
of fact, a reduction of TNF-a synthesis in the presence
of CSA has already been reported [45].
Although the major actions of CSA are considered
to be on T lymphocytes, there is evidence for possible
direct effects on other cell types, e.g. B cells,
macrophages, as well as bone and cartilage cells [46].
All these cells produce a range of cytokines and are of
interest in relation to the tissue changes that occur in
inflammatory diseases such as RA. Therefore, the
observed reduction of interleukin (IL)-1 and GM-CSF
production in cultured macrophages in the presence of
CSA could, in turn, influence the reduced IL-2
synthesis by T lymphocytes, already observed during
CSA therapy [46]. In fact, the initial stimulus for the
IL-2 production from T cells is the exposure to IL-1,
derived from macrophages during the process of
antigen processing and presentation.
Furthermore, APCs such as synovial Mf contain
significant levels of androgen receptors, indicating that
they are potentially target cells for androgen action
[12, 47]. As a matter of fact, CSA exerts immunosup-
pressive effects among different cells involved in the
immune response, and in particular both Tes and CSA
have been found to inhibit the APCs [11, 13, 16–19, 48–
53]. Moreover, in recent studies, CSA was shown to
increase the hepatic cytosolic androgen receptor levels
in treated animals, and in another study the presence
of Tes significantly prolonged the survival time of
allogenic skin graft in rats treated with CSA by
enhancing the immunosuppressive activity [54–56].
Since almost all the patients entered in the present
CSA study experienced the ‘side-effect’ of a low-degree
hypertrichosis after a mean period of 5–8 weeks, the
concomitant increase in Adiol-G, in agreement with
previous reports, might be confirmed as a plasma
marker of the CSA activity on peripheral androgen
metabolism (mainly increasing the 5a-reductase
activity) [9]. Adiol-G, in fact, is not secreted by the
adrenals, ovaries or testes.
The observed increase in plasma Adiol-G does not
seem related to the decrease in urinary Adiol-G
clearance that is a possible consequence of a reduction
in the glomerular filtration rate described in CSA
therapy [9]. In fact, the urinary Adiol-G clearance in
CSA-treated patients was found to be greater than that
observed among normal subjects [9]. Furthermore, in
CSA-treated patients, no increase in the main
androgens and a decrease in sex hormone binding
globulin were observed, thus probably eliminating
increased free Tes as the source of the reported raised
Adiol-G levels [9].
Conversely, the 17bHSD activity (oxidative activity)
appears to be less susceptible to CSA stimulation.
Although the rise in 17bHSD activity may also lead, if
prolonged, to a reduction of serum Tes, it seems more
likely to postulate that the CSA-induced increase of
Tes conversion to DHT at target tissue level could be
mainly responsible for the influence exerted by
androgens on the cells. Furthermore, the significant
increase in the Tes conversion to DHT on target cells
(i.e. macrophages), observed in our study, at both 100
and 500 ng/ml of CSA, seems to be in agreement with
previous studies showing a biphasic pattern of
inhibition of dendritic cells (APCs) pulsed with varying
concentrations of the drug (from 0 to 500 ng/ml) [4].
A significant clinical improvement of the articular
symptoms complained of by RA patients was observed
during the CSA treatment. The effects of CSA on
clinical measures in RA have already been demon-
strated in several controlled trials [33, 57–59]. CSA
differs from other disease-modifying anti-rheumatic
drugs regarding its limited efficacy on acute-phase
reactants, as confirmed by our study [33, 41, 42, 58, 59].
The implications of this for predicting impact upon the
radiological progression of disease are unclear and
should be a matter of future research, in relation to the
reported CSA-induced influence on some RA synovial
Mf activities [47].
In conclusion, the CSA-induced hypertrichosis
observed in treated patients is accompanied by an
increase in peripheral androgen metabolism (i.e.
Adiol-G formation). Studies in vitro on Tes metabolism
by primary cultures of CSA-treated RA synovial Mf
confirm a peculiar increase in the 5a-reductase activity,
BRITISH JOURNAL OF RHEUMATOLOGY VOL. 36 NO. 4438
forming DHT and possibly its important metabolite
Adiol.
However, further studies are needed to investigate
the mechanisms of CSA action on APCs such as Mf,
by considering that CSA is generally indicated to
produce immunosuppression through its inhibitory
effect on T-helper/inducer and cytotoxic cell activation
[60–62].
A
The work was partially supported by grants from
Consiglio Nazionale delle Ricerche (C.N.R.), Progetto
bilaterale, Italy–UK (no. 95.01708.CT04 to MC),
Special Project Ageing (no. 95.01017.PF40), the Italian
Ministry of Scientific Research and University (to MC
and LC), and the Italian Association for Cancer
Research (AIRC).
R
1. Borel JF, Gunn H. Cyclosporine as a new approach to
therapy. Ann NY Acad Sci 1986;475:307–19.
2. Picascia DD, Gardern JM, Frenkel RK, Roenigk HH Jr.
Treatment of resistant severe psoriasis with systemic
cyclosporine. J Am Acad Dermatol 1987;17:408–14.
3. Feutren G. The optimal use of cyclosporin A in
autoimmune diseases. J Autoimmun 1992;5(suppl. A):
183–95.
4. Shevac EM. The effects of cyclosporin A on the immune
system. Annu Rev Immunol 1985;3:397–423.
5. Graffenried von B, Krupp P. Side effects of cyclosporine
(Sandimmun) in renal transplant recipient patients and in
patients with autoimmune diseases. Transplant Proc
1986;18:876–83.
6. Schmidt JB, Gebhart W, Kopsa H, Spona J. Does
cyclosporin A influence sex hormone levels? Exp Clin
Endocrinol 1986;88:207–11.
7. Lindholm A, Pousette A, Carlstro¨ m K, Klintmalam R.
Cyclosporin-associated hypertrichosis is not related to
sex hormone levels following renal transplantation.
Nephron 1988;50:199–204.
8. Krueger BA, Trakshel GM, Sluss PM, Maines MD.
Cyclosporin-mediated depression of luteinizing hormone
receptors and heme biosynthesis in rat testes: a possible
mechanism for decrease in serum testosterone. Endocrin-
ology 1991;129:2647–54.
9. Vexiau P, Fiet J, Boudou P, Villette J et al. Increase in
plasma 5a-androstane-3a,17b-diol glucuronide as a
marker of peripheral androgen action in hypertrichosis:
a side-effect induced by cyclosporine A. J Steroid
Biochem 1990;35:133–7.
10. Cutolo M, Sulli A, Giusti M, Barone A, Seriolo B,
Accardo S. Increase in 5a-androstane-3a,17b-diol
glucuronide as a possible marker of the androgen-medi-
ated immunosuppressive activity exerted by Cyclo-
sporine A: preliminary results. Clin Exp Rheumatol
1994;12:350–1.
11. Araneo BA, Dowell T, Diegel M, Daynes RA.
Dihydrotestosterone exerts influence on the production
of interleukin-4 (IL-4), IL-5, and g-interferon (g-IFN),
but not IL-2 activated murine T cells. Blood
1991;78:688–99.
12. Cutolo M, Accardo S, Villaggio B et al. Evidence for the
presence of androgen receptors in the synovial tissue of
rheumatoid arthritis patients and healthy controls.
Arthritis Rheum 1992;35:1007–15.
13. Cutolo M, Villaggio B, Accardo S et al. Cultured
macrophage-like synoviocytes from rheumatoid arthritis
synovium metabolize testosterone and exhibit inhibition
of IL-1 production. Arthritis Rheum 1993; 36:S158.
14. Kirschner MA, Samojlik E, Szmal E. Clinical usefulness
of plasma androstanediol glucuronide measurements in
women with idiopatic hypertrichosis. J Clin Endocrinol
Metab 1987;65:597–601.
15. Choudhry R, Hodgins MB, Van der Kwast TH,
Brinkmann AO, Boersma WJA. Localization of andro-
gen receptors in human skin by immunohistochemistry:
implications for the hormonal regulation of hair growth,
sebaceous glands and sweat glands. J Endocrinol
1992;133:467–75.
16. Cooper MH, Gregory SH, Thomson AW, Fung JJ,
Cohen Z, Lavy GA. Evaluation of the influence of FK
506, Rapamycin and Cyclosporine on processing and
presentation of particulate antigen by macrophages:
assessment of a drug ‘carry-over’ effect. Transplant Proc
1991;23:2957–8.
17. Dupuy P, Bagot M, Michel L, Descourt B, Dubertret L.
Cyclosporin A inhibits the antigen-presenting function of
freshly isolated Langerhans cells in vitro. J Invest
Dermatol 1991;96:408–13.
18. Gilfar A, Etzioni A, Pillar T, Assy B, Eidelman S. Effect
of cyclosporine A on the regulation of Ia antigen
keratinocytes expression. Clin Immunol Immunopathol
1991;60:349–55.
19. Palay DA, Cluff CW, Wentworth PA, Ziegler HK.
Cyclosporine inhibits macrophage-mediated antigen
presentation. J Immunol 1986;136:4348–53.
20. Arnett F, Edworthy SM, Bloch DA et al. The American
Rheumatism Association 1987 revised criteria for the
classification of rheumatoid arthritis. Arthritis Rheum
1988;31:315–24.
21. An International Consensus report: the use of cyclo-
sporin A in rheumatoid arthritis. Br J Rheumatol
1993;32(suppl. 1):1–3.
22. Mihatsch MJ, Wolff K. Report of a meeting: consensus
conference on cyclosporin A for psoriasis, February
1992. Br J Dermatol 1992;126:621–3.
23. Ferriman D, Gallwey JD. Clinical assessment of body
hair growth in women. J Clin Endocrinol Metab
1961;21:1440–7.
31. Fredericksson T, Petterson U. Severe psoriasis oral
therapy with a new retinoid. Dermatologica 1978;157:
238–44.
24. Cutolo M, Balleari E, Giusti M, Intra E, Accardo S.
Androgen replacement therapy in male patients with
rheumatoid arthritis. Arthritis Rheum 1991;34:1–5.
25. Lindblad S, Wredmark T. Traumatic synovitis analysed
by arthroscopy and immunohistopathology. Br J
Rheumatol 1990;29:422–5.
26. Kay A, Wingrave SJ. Oral contraceptives and rheum-
atoid arthritis. Lancet 1983;i:1437.
27. Tsoukas CD, Provvedini DM, Manolangas SC. 1,25
hydroxy vitamin D3: a novel immunoregulatory hor-
mone. Science 1984;224:1438–40.
28. Ettinger B, Genan HK, Cann CE. Long-term estrogen
replacement prevents bone loss and fractures. Ann Intern
Med 1985;102:319–24.
29. Castagnetta L, Granata OM, Lo Casto M, Calabro`M,
Arcuri F, Carruba G. Simple approach to measure
metabolic pathways of steroids in living cells. J
Chromatogr 1991;572:25–39.
30. Leake RE, Habib F. Steroid hormone receptors: assay
and characterization. In: Green R, Leake RE, eds.
CUTOLO ET AL.: TESTOSTERONE METABOLISM AND CYCLOSPORIN A 439
Steroid hormones: a practical approach. Oxford: IRL
Press, 1987.
31. Van Rijthoven AWAM, Dijkmans BAC, The` HSG,
Boers M, Cats R. Comparison of cyclosporine and
D-penicillamine for rheumatoid arthritis: A randomized,
double-blind, multicentre study. J Rheumatol 1991;18:
815–20.
32. Tugwell P, Bombardier C, Gent M et al. Low dose
cyclosporin versus placebo in patients with rheumatoid
arthritis. Lancet 1990;335:1051–5.
33. Pasero GP, Priolo F, Marubini E et al. Slow progression
of joint damage in early rheumatoid arthritis treated with
cyclosporin A. Arthritis Rheum 1996;39:1006–15.
34. Vermeulen A, Giagulli VA. Physiopathology of plasma
androstanediol-glucuronide. J Steroid Biochem Mol Biol
1991;39:829–33.
35. Morimoto I, Edmiston A, Hawks D, Horton R. Studies
on the origin of androstanediol and androstanediol
glucuronide in young and elderly men. J Clin Endocrinol
Metab 1981;52:772–8.
36. Masi AT, Feigenbaum SL, Chatteron RT. Hormonal
and pregnancy relationship to rheumatoid arthritis:
convergent effects with immunologic and microvascular
system. Semin Arthritis Rheum 1995;25:1–27.
37. Hall GM, Perry LA, Spector TP. Depressed levels of
DHEAS in postmenopausal women with rheumatoid
arthritis but no relation to axial bone density. Ann
Rheum Dis 1993;52:211–4.
38. Klein H, Bressel M, Kastendieck H, Voigt KD.
Biochemical endocrinology of prostate cancer. In: Voigt
KD, Knabbe C, eds. Endocrine dependent tumors. New
York: Raven Press, 1991:131–63.
39. Castagnetta L, Granata OM, Polito L, Blasi L, Cannella
S, Carruba G. Different conversion metabolic rates of
testosterone are associated to hormone-sensitive status
and -response of human prostate cancer cells. J Steroid
Biochem Mol Biol 1994;49:351–7.
40. Masi AT, Feigenbaum SL, Chatteron RT, Cutolo M.
Integrated Hormonal-Immunological-Vascular (‘H-I-V’
Triad) system interactions in rheumatic diseases. Clin
Exp Rheumatol 1995;13:203–16.
41. Masi AT, da Silva JAP, Cutolo M. Perturbations of the
hypothalamic pituitary gonadal axis in rheumatoid
arthritis. In: Chikanza IC, ed. Neuroendocrine immune
mechanisms of rheumatic diseases. Baillie` re’s Clin
Rheumatol 1996;10:295–331.
42. Paulson RJ, Serafini PC, Catalino JA, Lobo RA.
Measurements of 3a-17b-androstanediol glucuronide in
serum and urine and the correlation with skin
5a-reductase activity. Fertil Steril 1986;46:222–6.
43. Wing EJ, Magee DM, Whiteside TL, Kaplan SS,
Shadduck RK. Recombinant human granulocyte-macro-
phage colony-stimulating factor enhances monocyte
toxicity and secretion of tumor necrosis factor a and
interferon in cancer patients. Blood 1989;73:643–6.
44. Fischer HG, Frosch S, Reske K, Reske-Kunz AB.
Granulocyte-macrophage colony-stimulating factor
activates macrophages derived from bone marrow
cultures to synthesis of MHC class II molecules and to
augmented antigen presentation function. J Immunol
1988;141:3882–8.
45. Espevic T, Figari IS, Shalaby MR et al. Inhibition of
cytokine production by cyclosporin A and transforming
growth factor-beta. J Exp Med 1987;166:571–6.
46. Russel RGG, Graveley R, Coxon F et al. Cyclosporin A.
Mode of action and effects on bone and joint tissues.
Scand J Rheumatol 1992;21(suppl. 95):9–18.
47. Cutolo M, Accardo S, Villaggio B et al. Androgen and
estrogen receptors are present in primary cultures of
human synovial macrophages. J Clin Endocrinol Metab
1996;81:820–7.
48. Manca F, Fenoglio D, Kunkl A, Caltabellotta M, Celada
F. Effect of cyclosporine on the antigen-presenting
function of human and murine accessory cells. Trans-
plantation 1989;46:40s–43s.
49. Benson A, Ziegler HK. Macrophages as targets for
inhibition by cyclosporine. Transplantation 1989;47:696–
703.
50. Roberts MS, Knight SC. Low-dose immunosuppression
by cyclosporine operating via antigen-presenting
dendritic cells. Transplantation 1990;50:91–5.
51. Teunisen MBM, Jager MHDE, Kapsenburg ML, Bos
JD. Inhibitory effect of cyclosporin A on antigen and
alloantigen presenting capacity of human epidermal
Langerhans cells. Br J Dermatol 1991;125:309–16.
52. Roth J, Goebler M, Erpenstein U, Sorg C. Differential
regulation of the macrophage-specific surface antigen
RM3/1bycyclosporine,azathioprineanddexamethasone.
Transplantation 1994;57:127–33.
53. Matsushima Y, Baba T. The in vivo effect of
cyclosporine A on macrophages. J Exp Pathol 1990;5:39–
48.
54. Kahn D, Gavaler JS, Lai H et al. Effect of cyclosporine
on hepatic cytosolic estrogen and androgen receptor
levels before and after partial hepatectomy. Dig Dis Sci
1990;35:6–11.
55. Hirasawa K, Enosawa S. Sex-associated differences in
organ transplantation: different effects of steroid
hormones, testosterone, oestradiol, progesterone and
prednisolone on the survival time of allogenic skin graft
in rats treated with cyclosporin A. Transplant Proc
1991;23:714–5.
56. Hirasawa K, Enosawa S. Effects of sex steroid hormones
on sex-associated differences in the survival time of
allogenic skin graft in rats. Evidence that testosterone
enhances and oestradiol reverses the immunosuppressive
activity of cyclosporine. Transplantation 1990;50:637–
41.
57. Forre O, Bjerkhoel F, Salvesen CF. An open, controlled,
randomized comparison of cyclosporine and azathio-
prine in the treatment of rheumatoid arthritis. Arthritis
Rheum 1987;30:88–92.
58. Dougados M, Awada H, Amor B. Cyclosporin
in rheumatoid arthritis: A double-blind, placebo
controlled study in 52 patients. Ann Rheum Dis 1988;
47:127–33.
59. Forre O, Norwegian Arthritis Study Group. Radiologic
evidence of disease modification in rheumatoid arthritis
patients treated with cyclosporine. Arthritis Rheum
1994;57:1506–12.
60. Kahan BD. Cyclosporine. N Engl J Med 1989;32:1725–
34.
61. Tokuda M, Kurata N, Mizoguchi A et al. Effect of
low-dose cyclosporin A on systemic lupus erythematosus
disease activity. Arthritis Rheum 1994;37:551–8.
62. Cutolo M, Castagnetta L. Immunomodulatory mechan-
isms mediated by sex hormones in rheumatoid arthritis.
Ann NY Acad Sci 1996;784:534–41.
