Importance of the intracrine metabolism of adrenal androgens in androgen-dependent prostate cancer.

Abstract

The metabolic pathways of androgens and processes by which androgens induce re-growth after androgen deprivation therapy in prostate cancer have not been fully elucidated. In this study, finasteride decreased PSA secretion in medium containing testosterone, androstenedione, androstenediol and dehydroepiandrosterone, whereas dihydrotestosterone (DHT)- and hydroxy-flutamide-induced PSA production was not inhibited by finasteride in LNCaP-FGC cells. The present data show that adrenal androgen precursors do not directly interact with androgen receptors (ARs) but are converted to DHT via the intraprostatic metabolic pathways, resulting in the induction of LNCaP activity. This is the first report confirming this mechanism experimentally and also suggest the use of combined therapies that target ARs and prevent the formation of DHT within prostate cancer cells to achieve optimal therapeutic efficacy.

Full text

Page 1

ORIGINAL ARTICLE
Importance of the intracrine metabolism of adrenal androgens
in androgen-dependent prostate cancer
K Suzuki1, T Nishiyama1, N Hara1, K Yamana1, K Takahashi1and F Labrie2
1Division of Urology, Department of Regenerative and Transplant Medicine, Course of Biological Function and Medicine Control,
Niigata University Graduate School of Medical and Dental Science, Niigata, Japan and2Laboratory of Molecular Endocrinology
and Oncology, Laval University Hospital Research Center (CRCHUL) and Laval University, Quebec City, Quebec, Canada
The metabolic pathways of androgens and processes by which androgens induce re-growth after
androgen deprivation therapy in prostate cancer have not been fully elucidated. In this study,
finasteride decreased PSA secretion in medium containing testosterone, androstenedione,
androstenediol and dehydroepiandrosterone, whereas dihydrotestosterone (DHT)- and hydroxy-
flutamide-induced PSA production was not inhibited by finasteride in LNCaP-FGC cells. The
present data show that adrenal androgen precursors do not directly interact with androgen
receptors (ARs) but are converted to DHT via the intraprostatic metabolic pathways, resulting in the
induction of LNCaP activity. This is the first report confirming this mechanism experimentally and
also suggest the use of combined therapies that target ARs and prevent the formation of DHT
within prostate cancer cells to achieve optimal therapeutic efficacy.
Prostate Cancer and Prostatic Diseases advance online publication, 27 March 2007; doi:10.1038/sj.pcan.4500956
Keywords: prostate cancer; adrenal androgens; metabolic pathway; LNCaP; androgen independence
Introduction
As Huggins and Hodges1observed that disseminated
prostate cancer reacts favorably to castration or to the
administrationof estrogens,
therapy has been used to impair the production or
activity of androgens, or both. At present, androgen
deprivation therapy (ADT), consisting of testicular
androgen blockade, the use of antiandrogens, or the
combination of both (combined androgen blockade) is
the standard therapy for metastatic prostate cancer. In
such cases, the response rate of ADT is up to 80% with
monotherapy and at more than 90% with combination
therapy. However, at the metastatic stage, prostate cancer
becomes resistant to ADT. If we are to improve
significantly survival, new therapeutic strategies de-
signed to avoid the emergence of resistant phenotypes
must be developed.
The mechanisms of development of resistance to
treatment or androgen independence have been the
subject of many investigations. The AR signaling path-
way might still be active during the early stage of
transition to ‘androgen independence’ of prostate cancer
cells. In fact, cancer cells adapt themselves to grow in a
first-linehormonal
low androgen environment by sensitizing the AR
signaling pathway. Prostate cells can amplify and over-
express the AR gene, amplify and overexpress AR
coactivators, mutate AR to become activated directly by
steroids other than dihydrotestosterone (DHT) such as
precursor adrenal androgens, and activate AR by growth
factors and cytokines.2–5Phosphorylation of AR or AR-
associated proteins occurs as observed for the estrogen
receptor.6,7
As mentioned above, many investigators have paid
attention to the AR hypersensitivity-related mechanisms
of development of androgen independence. On the other
hand, steroid metabolic enzymes are still expressed in
prostate cancer cells, although at the androgen-indepen-
dent stage, the enzyme activity has been reported to be
dramatically changed,8thus suggesting that steroido-
genic processes of androgen metabolism are also
important for the development of androgen indepen-
dence. These two processes, namely AR hypersensitivity-
related development mechanisms and changes in the
androgen steroidogenic processes are not mutually
exclusive and could simultaneously operate in prostatic
cells. However, which process is the predominant path-
way in the development of androgen independence
remains to be determined.
The aim of the present study was to clarify the
mechanism of the early stage transition to a mixed
androgen-independent and androgen-sensitive status of
prostate cancer cells. We first elucidated the intracellular
metabolism and action of androgens, especially adrenal
androgens, using LNCaP-FGC cells as model.
Received 1 November 2006; revised and accepted 25 January 2007
Correspondence: Dr K Suzuki, Division of Urology, Department of
Regenerative and Transplant Medicine, Course of Biological Function
and Medicine Control, Niigata University Graduate School of Medical
and Dental Science, Asahimachi 1-757, Niigata 951-8510, Japan.
E-mail: kazuyas@med.niigata-u.ac.jp
Prostate Cancer and Prostatic Diseases (2007), 1–6
& 2007 Nature Publishing Group
www.nature.com/pcan
All rights reserved1365-7852/07 $30.00

Page 2

Materials and methods
Human prostate tumor LNCaP-FGC (American Type
Culture Collection (ATCC), Manassas, VA, USA), PC3
(ATCC) and DU145 (ATCC) cells were grown in Roswell
Park Memorial Institute-1640 medium (Gibco BRL,
Grand Island, NY, USA) containing 10% fetal bovine
serum (CSL Ltd, Victoria, Australia), 1% MEM non-
essential amino acid (Gibco), 1% MEM sodium pyruvate
solution 100mM (Gibco) and 90mg/ml kanamycin. The
prostate epithelial cell line, PrEC (Clonetics, Biowhit-
taker, San Diego, CA, USA) was grown in PrEBM
(Clonetics). Cells were maintained in an atmosphere of
5% CO2and 95% air at 371C. LNCaP-FGC cells, on the
other hand, were plated at a concentration of 5?105cells
per 25cm2flasks (Falcon Labware, Lincoln Park, NJ,
USA) and cultured in phenol-red free medium with 5%
charcoal–dextran stripped of FBS 24h before the experi-
ment. Cells were treated with DHT, testosterone,
androstenedione, androstenediol, dehydroepiandroster-
one (DHEA), hydroxy-flutamide (all purchased from
Sigma Chemical, St Louis, MO, USA) in the presence or
absence of 5a-reductase (5-R) inhibitor finasteride (Merck
Research Laboratory, Rahway, NJ, USA). Cells were also
treated with finasteride alone. The cells were kept for 7
days after culture in the experimental media. The media
were collected to determine PSA levels. The PSA levels in
the culture media were determined by commercially
availablesolid-phaseradioimmunoassay
Tandem PSA Test (Beckman Coulter Inc., San Diego,
CA, USA) according to the manufacturer’s instructions.
Total cellular RNA from the cell lines was isolated
using an ISOGEN RNA isolation kit (Nippongene,
Tokyo, Japan) according to the manufacturer’s instruc-
tions. RNA was subjected to reverse transcription
polymerase chain reaction (RT-PCR) analysis as de-
scribed earlier.9To synthesize first-strand cDNA, 1mg
of the total RNA was reverse-transcribed using a first-
strand cDNA Synthesis Kit (Roche Diagnostics, Tokyo,
Japan). The first-strand DNAwas then amplified by PCR
using the primers shown in Table 1. PCR was performed
in a 50ml reaction volume-containing buffer (50mM KCl;
10mM Tris–HCl, pH 8.4; 2.5mM MgCl2and 200mg/ml
gelatin), with 200mM of each dNTP, 400nM of each outer
primer and 2.5 units of Amplitac Gold (Roche Diagnos-
tics, Tokyo, Japan). PCR consisted of 35 cycles of
denaturation (941C, 30s), annealing (50, 60, 60, 58, 55
and 601C for 5-R type 1, 2, 17b-hydroxysteroid dehy-
using the
drogenase (17-HSD) type 2, 3, 5 and 3b-hydroxysteroid
dehydrogenase (3-HSD) type 1 respectively, 45s) and
extension (721C, 60s) followed by a final 10min exten-
sion at 721C. The PCR products were size fractionated
by 3% agarose gel electrophoresis. DNA bands were
visualized with an ultraviolet transilluminator (Specto-
line, Funakoshi, Tokyo, Japan).
Statistical comparisons were made using the Krusu-
kal–Wallis test with SPSS 10.0 J software for Windows.
Po0.05 was considered significant. The results are
expressed as mean7s.e.m.
Results
We first studied the effect of DHT, testosterone,
androstenedione, androstenediol, DHEA and hydroxy-
flutamide on PSA secretion in LNCaP cells. At 10?10,
10?9and 10–8M, DHT increased PSA secretion 7 days
after treatment by 1.28-, 2.2-, 3.58-fold, respectively
(P¼0.016) (Figure 1a). The same doses of testosterone
(Figure 1a), androstenedione, androstenediol, and DHEA
(Figure 1b) increased PSA production by 1.6-, 2.95-, 3.37-
fold (P¼0.016), 1.15-, 2.05-, 3.09-fold (P¼0.016), 1.69-,
3.12-, 4.82-fold (P¼0.016) and 1.19-, 1.51-, 2.71-fold
(P¼0.019) in a dose-dependent manner, respectively.
Treatment with the antiandrogen hydroxy-flutamide at
10?10, 10?8, 10-6M resulted in a 1.46-, 2.86-, 3.78-fold
dose-dependent increase in PSA production in condi-
tionedmedia,respectively
Although these androgens and hydroxy-flutamide had
some effect on cell growth, no statistically significant
difference was found (data not shown). PSA production
with conditioned medium was normalized to the cell
numbers.
The 5-R inhibitor finasteride was first used to block the
conversion of testosterone into DHT. Treatment with
10?8, 5?10?8and 10?7M finasteride alone had no
significant effect on PSA secretion in the media after
7 days, of incubation (Figure 2a). On the other hand,
treatment with 10?8, 5?10?8and 10?7M finasteride
resulted in 95.5, 89.4 and 80.7% (P¼0.034) dose-
dependent inhibition of 10?8M testosterone-induced
PSA secretion, respectively. However, the same doses of
finasteride treatment had no significant effect on DHT-
induced PSA secretion (Figure 2b). These results suggest
that testosterone-induced PSA production up to a large
(P¼0.019,Figure 1c).
Table 1
Different sets of primers used in the PCRs
RNA transcriptPrimer 5’–3’ Position in cDNAPCR fragment size (bp)
5-R1TGC GAG GAG GAA AGC CTA TG (F)
GCC ACA CCA CTC CAT GAT TTC (R)
CAT ACG GTT TAG CTT GGG TGT (F)
GCT TTC CGA GAT TTG GGG TAG (R)
AGT TGC TTC CAT CCA ACC TGG A (F)
TTC CAT TGC CTA GGT GGC CTT T (R)
CGA GCA TAT TAA AGA AAA ACT TGC AGG CTT (F)
AGA TAC TTT GTC ATT GCA GTC GAG A (R)
CCA GGT GAG GAA CTT TCA CCA A (F)
TGG CCA ATC CTG CAT CCT T (R)
GCA CCC TGT ACA CTT GTG CCT T (F)
GGT GAG GCG TGT CAT CTG AGAT (R)
347–366
654–634
456–476
770–750
851–872
1256–1235
387–416
764–740
381–402
492–474
664–685
918–897
308
5-R2315
17-HSD2 405
17-HSD3 378
17-HSD5112
3-HSD1 255
Abbreviations: F, forward; R, reverse; 5-R, 5a-reductase; 17-HSD, 17b-hydroxysteroid dehydrogenase; 3-HSD, 3b-hydroxysteroid dehydrogenase.
Importance of androgen metabolism in prostate cancer
K Suzuki et al
2
Prostate Cancer and Prostatic Diseases

Page 3

extent, is mediated by conversion to DHT in LNCaP cells
instead of a direct interaction of testosterone with AR.
Treatment with 10?8, 5?10?8and 10?7M finasteride
added to a fixed concentration of 10?6M hydroxy-
flutamide had no significant effect on PSA secretion
(Figure 3a). As shown in Figure 3b, treatment with 10?8,
5?10?8
concentration of 10?8M androstenedione, androstenediol
or DHEA resulted in 98.1, 93.1, 83.9 (P¼0.031), 93.3, 88.5
and 82.3% (P¼0.029) and 83.1, 69.5 and 49.3% (P¼0.016)
dose-dependent inhibitions of PSA secretion in the
and 10?7M finasteride added to a fixed
media, respectively. These results indicate that andros-
tenedione and androstenediol but mostly DHEA induce
PSA secretion in LNCaP cells after transformation into
DHT.
5-R type 1 and 2, 17-HSD types 2, 3, and 5, and 3-HSD
type 1 expressions were detected in LNCaP cells. As
shown in Figure 4, mRNA expression for 5-R types 1, 2,
17-HSD types 2, 3, 5 and 3-HSD type 1 was detected as
specific single bands (308, 315, 405, 378, 112 and 255bp,
respectively) in LNCaP-FGC, PC-3, DU145 and PrEC
cells.
Discussion
The present data show that precursor adrenal androgens
exert a stimulatory effect on PSA secretion in LNCaP
cells following enzymatic conversion into DHT with
subsequent binding to the AR. This is the first report that
describes the major importance of the steroidogenic
process from DHEA over the direct binding of potential
androgens to hypersensitive AR in the action of
precursor adrenal androgens in prostate cancer cells.
This steroidogenic process should be taken into account
in addition to the AR hypersensitivity-related develop-
ment mechanisms leading to resistance to antiandrogen
treatment in prostate cancer.
PSA has been used as parameter to study androgen
action in prostate cancer. LNCaP cells are the most
frequently studied AR-positive prostate cancer cell line
a
b
c
Figure 1
anti-androgen hydroxy-flutamide on PSA production in LNCaP
cells: 24h after 5?105LNCaP cells were seeded, cells were treated
with the indicated concentration of DHT, testosterone (a), andros-
tenedione, androstenediol, DHEA (b), or hydroxy-flutamide (c),
and cultured for 7 days. PSA levels in the treated media were then
measured by solid-phase RIA methods. PSA production with
conditioned medium was normalized to the cell numbers. The
results are expressed as means7s.e.m.
Effect of androgens and their precursors as well as of the
0
20
40
60
80
100
120
0 1050100
DHT
T
0
20
40
60
80
100
120
0 1050 100
Finasteride concentration (nM)
a
b
*P=0.034
*
PSA production
relative rate (%)
PSA production
relative rate (%)
Finasteride concentration (nM)
Figure 2
LNCaP cell medium by finasteride: LNCaP cells were incubated
with 10?8M DHT or testosterone in the absence or presence of
finasteride for 7 days. DHT-induced PSA production was not
inhibited by finasteride.
Inhibition of testosterone-induced PSA accumulation in
Importance of androgen metabolism in prostate cancer
K Suzuki et al
3
Prostate Cancer and Prostatic Diseases

Page 4

with a Thr877Ala point mutation in the AR gene. These
cells were established from a tumor that progressed after
castration and flutamide therapy. These cells have a
mutated AR that responds not only to androgens but also
to flutamide with increased cellular proliferation and
elevated expression of PSA.10LNCaP-FGC cells con-
tinuously grown in steroid-free medium did not die and
the cells remained responsive to androgen.11These
findings indicate that LNCaP-FGC cells are in an early
period of mixed androgen independence and androgen-
sensitive state. In order to evaluate the effect of
androgens, anti-androgen and 5-R inhibitor on the
activity of LNCaP cells, the present study investigated
the levels of PSA production in conditioned media.
Prostate cells contain a variety of steroid-metabolizing
enzymes required for the local formation of active
androgens from precursor steroids provided by the
adrenals. The main enzymes involved in local steroid
metabolism are 17-HSDs, 3-HSDs and 5-Rs. At present,
nine different 17-HSD isoenzymes, namely types 1, 3, 5, 7
(reductive enzymes), 2, 4, 8, 10 and 11 (oxidative
enzymes) have been characterized in humans.12In the
prostate, androstenedione is converted to testosterone by
17-HSD type 5 enzyme.12,13In prostate cancer specimens,
a reduced expression of 17-HSD type 2 mRNA has been
detected14and the expression levels of the 17-HSD type 3
gene were significantly higher than those in nonmalig-
nant tissues.15For 5-R, it is now clear that not only 5-R
type 1 but also the type 2 isozymes are expressed in the
human prostate and the inhibition of both is more
effective in lowering DHT than the inhibition of a single
isozyme.16
In prostate cancer, 5-R type 1 protein levels have been
reported to be increased17whereas type 2 immunostain-
ing has been found to be decreased.18Previous reports
have shown that LNCaP cells contain only 5-R type 1.19
In the present study, we detected 17-HSD types 2, 3 and
5, 3-HSD type 1 and both 5-R types 1 and 2 mRNA
expressions in LNCaP-FGC cells. Such data indicate that
enzymes of androgen metabolism are well expressed in
LNCaP-FGC cells.
Although finasteride mainly inhibits 5-R type 2, a high
dose of finasteride can inhibit the 5-R type 1 enzyme
in LNCaP cells.20In fact, Finasteride has a Ki of
3.25?10?7M for the type 1 isozyme and 1.2?10?8M
toward type 2 5-R.20In the present study, concentrations
of 10?7–10?8M finasteride alone (without androgens) did
not have an inhibitory effect on PSA production of
LNCaP cells. PSA secretion induced by a fixed concen-
tration of testosterone was decreased by finasteride in
a dose-dependent manner although secretion induced
by the same concentration of DHT did not have an
inhibitory effect. These data also show that concentra-
tions of 10?7–10?8M finasteride block type 2 5-R activity
in LNCaP cells.
Previously, we indicated that serum levels of adrenal
androgens in prostate cancer patients remained at about
60% following castration and flutamide, as opposed to
testosterone levels that fell to about 2.7%.21Plasma levels
of the sulfate form of DHEA in adult men are about 100–
500 times higher than those of testosterone in the
circulation in adult men.22It is believed that these large
amounts of precursor adrenal androgens participate in
prostate cancer recurrence after ADT. The transition of
prostate cancer to androgen insensitivity can result from
many different adaptive survival mechanisms. Point
mutation in the steroid-binding domain of the AR gene,
such as Thr877Ala, decreases binding specificity and
0
20
40
60
80
100
120
010 50100
A-dione
A-diol
DHEA

PSA production
relative rate (%)

PSA production
relative rate (%)
a
b
**P=0.029
***P=0.016
**
0
20
40
60
80
100
120
0 1050 100
FLU
*
*P=0.031
***
Finasteride concentration (nM)
Finasteride concentration (nM)
Figure 3
accumulation in LNCaP cell medium by finasteride: LNCaP cells
were incubated with 10?8M androstenedione, androstenediol,
DHEA, or 10?6M hydroxy-flutamide in the absence or presence of
finasteride for 7 days. PSA production induced by hydroxy-
flutamide was not inhibited by finasteride, but was inhibited with
androstenedione, androstenediol and DHEA used as stimulators.
Inhibition of precursor adrenal androgen-induced PSA
5R1
(308bp)
3HSD1
(255bp)
17HSD5
(112bp)
17HSD3
(378bp)
17HSD2
(405bp)
5R2
(315bp)
beta-actin
(500bp)
M LNCaP PC3 DU145 PrEC
Figure 4
3-HSD mRNA in prostate cancer cell lines and prostate epithelial
cell lines: reverse transcription and amplification were performed as
described in Materials and methods. A 500bp fragment of the
human b-actin gene was amplified as a positive control.
RT-PCR detection of 5 a-reductases (5-R), 17-HSD and
Importance of androgen metabolism in prostate cancer
K Suzuki et al
4
Prostate Cancer and Prostatic Diseases

Page 5

allows activation of the AR-regulated gene by multiple
steroid hormones including inactive precursor adrenal
androgens.23AR amplification may also sensitize AR to
activation by low levels of precursor adrenal andro-
gens.24Moreover, in tumor cells overexpressing AR
coactivators SRC-1 and TIF2, higher the AR activities
were measured after treatment with precursor adrenal
androgens.25Such data suggest that precursor adrenal
androgens participate in the transition to androgen
independency via mechanisms that relate to AR hyper-
sensitivity. However, the intracellular actions of adrenal
androgens and the identity of the androgens involved in
AR binding are important in addition to these mechan-
isms potentially involved in the transition to androgen
independence.
In the present study, the dose-dependent production of
PSA induced by the antiandrogen hydroxy-flutamide in
LNCaP cells was not inhibited by finasteride. Such data
indicate that hydroxy-flutamide regulates PSA produc-
tion via binding directly to the mutated AR with no
influence of 5-R. The effect of testosterone on PSA
regulation in LNCaP cells, however, was mediated via
conversion to DHT, and this pathway might be more
predominant than the direct binding of testosterone to a
mutant receptor. Our data also show that the adrenal
precursor androgens androstenedione, androstenediol
and DHEA-induced PSA production in LNCaP cells via
action of 5-R. These data indicate that the adrenal
precursor androgens at least partly exert an effect not
by directly interacting with AR but by undergoing
enzymatic conversion to DHT, with consequent binding
to the AR (Figure 5).
We previously found that DHT levels in prostatic
tissue after castration and flutamide therapy remained at
approximately 25% of the amount measured before ADT
in prostate cancer patients.21These data suggest that the
source of DHT in the prostatic tissue after ADT involved
intracrine production within the prostate, converting
adrenal androgens into DHT.21,26
mutation characterized in the LNCaP cell line is present
in a substantial number of patients with androgen-
independent prostate cancer, and whether DHEA and
androstenedione could induce PSA stimulatory effects
by undergoing enzymatic conversion to more potent
androgens in vivo found in the present in vitro study,
remains to be clarified although the present data are
highly suggestive of such an intracrine mechanism. If the
results obtained with this human prostate cell line
correspond to a biologic model closer to the hormone-
sensitive early stage of recurrent prostate cancer, we have
shown the importance of the intracrine metabolism of
adrenal androgens in many of prostate cancer patients at
such stage. Such data emphasize that DHT made from
adrenal precursors might play an important role in
prostate cancer recurrence after ADT.
In fact, studies have established that therapy combin-
ing orchiectomy with antiandrogens to inhibit the action
prostate cancer patients.27–30Furthermore, an estab-
lished 5-R inhibitor such as finasteride can reduce
intraprostatic levels of DHT.31Finasteride prevents or
delays the appearance of prostate cancer,32although it
can increase the detection of high-grade cancer. Leibow-
its and Tucker have reported that triple androgen
blockade, ADT followed by finasteride maintenance
appears to be a promising alternative for the manage-
ment of patients with clinically localized and locally
advanced prostate cancer.33According to these findings
and our studies, therapies that target the AR and prevent
conversion from adrenal androgens to DHT within
prostate cancer cells, such as treatment with a combina-
tion of 5-R inhibitors and anti-androgens, might offer the
most effective ADT to prolong prostate cancer patient
survival.
Whether the AR
References
1 Huggins C, Hodges CV. Studies on prostate cancer. Effect of
castration, estrogen and androgen injection on serum phospha-
tases in metastatic carcinoma of the prostate. Cancer Res 1941; 1:
293–297.
2 Navarro D, Luzardo OP, Fernandez L, Chesa N, Diaz-Chico BN.
Transition to androgen-independence in prostate cancer. J Steroid
Biochem Mol Biol 2002; 81: 191–201.
3 Gregory CW, Johnson Jr RT, Mohler JL, French FS, Wilson EM.
Androgen receptor stabilization in recurrent prostate cancer is
associated with hypersensitivity to low androgen. Cancer Res
2001; 61: 2892–2898.
4 Rau KM, Kang HY, Cha TL, Miller SA, Hung MC. The
mechanisms and managements of hormone-therapy resistance
in breast and prostate cancers. Endocr Relat Cancer 2005; 12:
511–532.
5 Gioeli D et al. Androgen receptor phosphorylation. Regulation
and identification of the phosphorylation sites. J Biol Chem 2002;
277: 29304–29314.
6 Tremblay A, Tremblay GB, Labrie F, Gigue `re V. Ligand-
independent recruitment of SRC-1 to estrogen receptor b
through phosphorylation of activation function AF-1. Mol Cell
1999; 3: 513–519.
7 Kato S et al. Activation of the estrogen receptor through
phosphorylation by mitogen-activated protein kinase. Science
1995; 270: 1491–1494.
8 Vihko P et al. Control of cell proliferation by steroids: the role of
17HSDs. Mol Cell Endocrinol 2006; 241: 141–148.
AR
AREs
Androgen-responsive gene
PSA production
prostate
androstenedione
testosterone
androstenediol
17βHSD1,5
17βHSD2,4
3βHSD1
5α-R1,2
17βHSD3,5
17βHSD2
3βHSD1
DHT
adrenal gland
DHEA
Figure 5
induced LNCaP cell activity in cells having a mutated AR
stimulated by compounds other than testosterone and DHT: arrow
width indicates relative level of activity. The dotted arrow indicates
relatively low activity (DHT, dihydrotestosterone; DHEA, dehydro-
epiandrosterone; ARE, androgen response element).
Proposed schematic pathway of adrenal androgen-
Importance of androgen metabolism in prostate cancer
K Suzuki et al
5
Prostate Cancer and Prostatic Diseases