The Prostate 29:9I- I00 (I 996)

Expression and Regulation of Metallothionein

mRNA Levels in the Prostates of Noble Rats:
Lack of Expression in the Ventral Prostate and
Regulation by Sex Hormones in the
Dorsolateral Prostate
Shibnath Ghatak, Paulo Oliveria, Paula Kaplan, and Shuk-mei Ho
Department of Biology, Tufts University, Medford, Massachusetts

ABSTRACT: The rat prostate is composed of two anatomically well-defined regions

designated as the ventral prostate (VP) and the dorsolateral prostate (DLP). W and DLP are
known to exhibit marked cytological, biochemical, and functional differences including dif-
ferential susceptibilities to carcinogens. While the VP is uniquely susceptible to cadmium
carcinogenicity [1,2], the DLP is sensitive to sex hormone-induced cancer [3,4]. The role
played by the heavy metal binding protein, metallothionein (MT), in the prostate is largely
unknown. It is still controversial as to whether MT is expressed in the rat gland. The aim of
the present study is to examine the expression pattern of MT mRNA in the rat gland and its
probable regulation by heavy metal ions and sex hormones, in order to gain insight into the
biological function of MT in the prostate. Northern hybridization and reverse transcriptase-
polymerase chain reaction analyses revealed constitutive expression of MT mRNA in the
DLP and a lack of expression of the transcript in the W. In situ hybridization localized the
transcript to the epithelium of the DLP, with the lateral prostate epithelium exhibiting
the highest level of expression. Administration of cadmium and zinc failed to induce MT
transcription in the VP, nor were these treatments effective in elevating levels of MT mRNA
in the DLP. A 60% reduction in MT message levels was observed in the DLP following
orchiectomy. MT transcript levels in the DLPs of castrates were restored by readministration
of androgen to the animals. Long-term treatments (16 weeks) of rats with estradiol-17P (EJ
or testosterone 0 plus E, induced a 2.8-fold and a 5-fold increase in MT message content
in the DLP, respectively. In sum, MT mRNA was shown to be absent in the VP and was not
inducible by heavy metal ions or hormones in this prostatic lobe. These findings substantiate
the belief that MT plays a role in heavy metal detoxificationand deficiency in its expression
may contribute to the unique susceptibility of the VP to cadmium carcinogenicity. By con-
trast, constitutive expression of MT was demonstrated in the DLP, which was shown to be
regulated by androgen and not by exogenously administered heavy metal ions. These re-
sults suggest a participatory role of MT in the normal functioning of the DLP. The fact that
high levels of MT mRNA were induced in the DLP following long-term estrogenic or con-
joint androgenic-estrogenic action alludes to the possibility that MT may serve as an intra-
cellular antioxidant in DLP cells. 8 1996 Wiley-Liss, Inc.

KEY WORDS: metallothionein, sex hormone regulation, ventral prostate, dorsolateral

prostate, rat

INTRODUCTION
Received for publication March 27, 1995; accepted July 17, 1995.
Metallothioneins ( W S ) are a family of low-molec- Address reprint requests to Dr. Shuk-mei Ho, Department of Bi-
ular-weight cysteine-rich ( = 30%),aromatic residue- ology, Tufts University, Medford, MA 02155.

0 1996 Wiley-Liss, Inc.

92 Ghatak et al.
free, trace-metal ion binding proteins distributed
widely in animal and plant tissues [5-91. In many
species, MTs exist in several isoforms encoded by
different genes. Two major isoforms of MT, desig-
nated MT-I and MT-11, as characterized according to
their chromatographic properties, have been re-
ported in mammalian species. In the mouse, the
genes encoding these isoforms are located on chro-
mosome 8 [lo], while in humans they are on chro-
mosome 16 [ll]. In the rat, sequences of the MT
structural gene, three pseudogenes, and a single
transcript (approximately 550 basepairs) have been
reported [12,13]. A comparison of rat and mouse
MT-1 sequences reveals extensive homology in the
translated (95% residue homology) and untranslated
regions, while a much greater divergence exists be-
tween rat and human MT coding sequences (85%)
[QI-
MT genes are ubiquitously expressed in many tis-
sues at basal levels and are inducible by a great vari-
ety of endogenous and exogenous factors, which in-
clude glucocorticoids, second messengers, growth
factors and cytokines, inflammatory agents, inter-
feron, and class II-B metals (e.g., zinc and cadmium)
[5-91. Yet the function of MTs is only poorly under-
stood. They have been detected in the cytoplasm as
well as in the cell nucleus [14,15]. Suggested roles of
MTs include regulation of intracellular zinc (Zn), cop-
per and sulfur homeostases, detoxification of heavy
metals such as cadmium (Cd), protection against ion-
izing radiation and oxygen radicals, and participation
in differentiation and proliferation controls [16-181.
Since the most widely suspected role of MTs is heavy
metal detoxification, the majority of studies of MTs,
and their genes have focused on liver and kidney, the
primary sites of heavy metal accumulation [5]. Infor-
mation regarding the function(s) of MTs and their
regulation in other tissues such as the prostate are
sparsely available.
It is known that both human and rat prostates
sequester Zn [19-211. Additionally, the glands are
susceptible to Cd-induced carcinogenesis [1,2,22].
Nevertheless, the relationships, if any, between MT
and the metabolism of these metal ions in the pros-
tate remain unknown. Conflicting data exist in the
literature regarding whether MT is expressed in the
prostate. Although MT-like molecules have been
documented in rodent and human prostates by im-
munohistological methods [15,23,24], chemical anal-
yses of the proteins isolated from the rat prostate
indicate that they have amino acid compositions dis-
tinct from those of classical MTs [2]. Furthermore,
prostatic MT immunostaining intensity and pattern
were not altered in rats treated with Cd, thus sug-
gesting a lack of regulation of prostatic MT expres-
sion by heavy metal ions. This finding is in marked
contrast to results obtained from a study on rat
liver MT regulation, in which enhanced levels of
MT transcripts were observed in the tissue after
cadmium (Cd) exposure [lo]. To the best of our
knowledge, studies documenting the presence of MT
transcripts in prostatic tissues have not been re-
ported. Thus it remains an open question as to
whether the MT gene is transcribed in rat prostate
and if MT transcription is under the regulation of
classical inducers.
To address these issues, as they pertain to the
functional role of MT in the prostate, the present in-
vestigation analyzed the levels and the localization of
MT mRNA in the two major lobes of the rat gland: the
dorsolateral lobe (DLP) and the ventral lobe (VP).
These two anatomically well-defined regions of the
rat prostate are known to exhibit marked cytological,
biochemical, and functional differences [25-27]. For
example, high levels of zinc are sequestered in the
DLP [20,21], while the VP is uniquely susceptible to
Cd-induced cancer [1,2]. It is therefore conceivable
that the expression of MT in VP and DLP is differen-
tially regulated. Finally, since prostatic functions are
extensively regulated by testicular steroids (andro-
gens and estrogens), we seek to discern the effects of
these steroids on MT expression in short- and long-
term experiments.
MATERIALS AND METHODS
The male Noble rats were purchased from Charles
River Co. (Wilmington, MA). The RNAzol B was pur-
chased from Tel-Test (Friendswood, TX), radiochem-
icals dCTP-ol-32P(6,000 Ci/mole) and ATP-y3'P (6,000
Ci/mole) from DuPont New England Nuclear (Bos-
ton, MA), polynucleotide kinase and restriction en-
donucleases from New England Biolabs (Beverly,
MA), agarose from Bethesda Research Laboratory
(Bethesda, MD), Nytran membranes from Schleicher
and Schuell (Keene, NH), testosterone (T) and estra-
diol-17p (E2) from Sigma Chemical Co. (St. Louis,
MO), Silastic tubing for the preparation of capsules
from Dow Corning (Midland, MI), and formamide
from Kodak (Rochester, NY). Synthetic oligonucle-
otide was purchased from Genosys Biotechnology
(The Woodlands, TX). All other reagents were of mo-
lecular biology grade or reagent grade purchased ei-
ther from Sigma Chemical Co. or Fisher (Pittsburgh,
PA).
Extraction of Total RNA
Total RNA was extracted from frozen tissues by a
modified one-step procedure (28) using RNAzol B so-

lution. Briefly, the frozen tissue was homogenized in
RNAzol B (3 ml RNAzol B per 100 mg tissue) using a
Tissuemizer (Tekmer, Cincinnati, OH) at room tem-
perature. The homogenate was centrifuged at 12,OOOg
for 15 min at 4°C to remove the high-molecular-
weight sheared DNA and cellular debris. The clear
supernatant was extracted with l/lOth vol of chloro-
form and centrifuged as before. The aqueous super-
natant was separated and extracted with an equal vol-
ume of chloroform. Total RNA was precipitated from
the clear supernatant at 4°C by adding an equal vol-
ume of isopropanol. The RNA was pelleted by cen-
trifuging at 12,OOOg for 15 min at 4°C. The pellet was
washed twice with 75% ethanol, briefly air dried, and
finally dissolved in neutral formamide [29]. The RNA
in formamide solution was quantitated by absorbance
at 260 nm.
Agarose Gel Electrophoresisand Membrane
Transfer of RNA Samples
Total RNA (10 pg/sample) was mixed with an
equal volume of denaturing buffer containing 16%
formaldehyde, 10%glyorol, 5%borate buffer (1x bo-
rate buffer contains 5 mM sodium borate, 50 mM bo-
ric acid, 10 mM sodium sulfate, 1mM EDTA Na2, pH
8.0) and heated at 65°C for 10 min. RNA was size
fractionated electrophoretically (50V, 3-4 hr) on a
1.0% agarose gel containing 2.2 M formaldehyde in
1x borate buffer with 0.5 pg/ml ethidium bromide.
After electrophoresis, the agarose gel was destained
overnight in diethylpyrocarbonate (DEPC)-treated
water. RNA was transferred to Nytran membranes by
downward alkaline transfer [30] in the presence of 3.3
M NaC1, 8 mM NaOH, and 2 mM sodium sarcosyl.
After the membrane was washed in 50 mM sodium
phosphate buffer, pH 7.0, for 5 min, the transferred
RNA was crosslinked by UV to the membrane in a
Stratalinker (Stratagen, La Jolla, CA). Blots were
stored in sealed bags at -20°C until use.
Northern Analyses
Membranes were prehybridized for 4 hr at 55°C in
a solution containing 0.5M NaH,PO, (pH 7.0), 7%
sodium dodecyl sulfate (SDS), 1%bovine serum al-
bumin (BSA), l mM EDTA, pH 7.4 [31]. A mouse
MT-1 cDNA containing the entire coding region,
cloned in SP6 vector [32], was provided by Dr. Glen
K. Andrews, University of Kansas Medical Center,
Kansas City, MO. The 364-bp insert was isolated and
labeled with dCTI'-y-[32P] by random priming using
the Boehringer-Mannheh Random Primed Labeling
Kit (Indianapolis, IN) to specific activities of 1-3 x
lo9 d p d p g DNA. The labeled probe after purifica-
tion was denatured by 0.1 M NaOH at 37°C for 5 min
Metallothionein mfWA in Rat Prostate 93
and added to the hybridization solution (same as pre-
hybridization solution) at a final concentration of 1-3
x lo6 d p d m l . Hybridization was carried out over-
night at 55°C. Membranes were then washed sequen-
tially in 0.5% SDS solutions containing different con-
centrations of phosphate: 0.25 M phosphate at room
temperature for 15 min, 0.25 M phosphate at 42°C for
30 min, 0.125 M phosphate at 42°C for 30 min, and
0.125 M phosphate at 55°C for 10 min. Finally, all
blots were washed in 100 mM phosphate buffer (pH
7.0) alone for 10 min at room temperature and
wrapped in seal wrap. Fuji RX-GCU film (Fischer)
was exposed to the membrane with intensifymg
screens at -80°C. After hybridization with the MT-1
cDNA probe, blots were stripped and rehybridized
with a 32P-labeled 30-mer oligonucleotide comple-
mentary to 18s rRNA (5'd(CGGCATGTATTAGC-
TCTAGAA'ITACCACAG)3' [33]). Thirty-five pmole
of the oligomer was end-labeled by incubating with
100 pCi ATP-y-[32P](6,000 Ci/mole)in the presence of
20 units of polynucleotide kinase at 37°C for 2-3 hr.
The labeled probe was isolated by 1ml Sephadex G50
spin column and used for the hybridization of the
stripped Northern blots at a final probe concentration
of 1-3 x lo6 d p d m l hybridization buffer. Washing
steps for the 18s rRNA hybridization were as de-
scribed previously in the Northern blot washing pro-
cedure. Autoradiograms were prepared following re-
hybridization. Hybridization signals for MT mRNA
and 18s rRNA were quantitated by densitometric
scanning. The MT hybridization signals were normal-
ized with respect to the corresponding 18s rRNA sig-
nal to correct for loading variations. Fold induction of
the MT gene in any treatment groups was calculated
as a ratio of normalized MT signal in the treated sam-
ple and that of a control sample (see figure legends
for definition of the control sample). Using this for-
mula of calculation, fold induction of the control sam-
ple is arbitrarily set as 1.
Reverse! Tranxriptas4.Polymerase Chain
Reaction (RT-PCR)
The primers used for PCR were derived from rat
liver MT cDNA sequence [12]. The sequences of for-
ward and reverse primers were as follows:
Forward: 5'd (CGTTGC TCC AGA TTC ACC AGA
TC)3'
Reverse: 5'd (TCA CAT GCT CGG TAG AAA ACG
G)3'
Expected size of the PCR product is 312 bp in
length, representing over the entire coding sequence
of rat MT-1. One hundred ng of total RNA samples
isolated from VP and DLP of NBL rats were reverse
94 Ghatak et al.
transcribed with M-MLV reverse transcriptase (Pro-
mega, Madison, WI) using 0.4 pg oligo-dT,,-,, (Phar-
macia, Piscataway, NJ) as primer in presence of 1.875
mh4 DTT (Stratagen, Menasha, WI), 8 pM dNTl3. The
reaction was carried out at 37°C for 1 hr. The cDNA
was mixed with Taq buffer and magnesium chloride
(Promega, Madison, WI), 0.1 pM of each of forward
reverse primers. The mixture was heated to 94°C for
10 min, and the reaction was initiated by adding 2.5
units Taq polymerase (Promega). One polymerization
cycle consisted of denaturation at 95"Ul min, anneal-
ing at 55"C/1 min, and extension at 72"C/3 min. After
30 cycles of reaction, the extension reaction was con-
tinued for another 10 min. The product was directly
analyzed in a 2% agarose gel containing ethidium bro-
mide, using a 100-bp ladder as the molecular size
marker (GIBCO-BRL, Gaithersburg, MD).
In situ hybridization. A modified procedure [34]
adapted from Hofler and colleagues [35] was used. In
brief, prostates were fixed for 2 hr at 4°C in 4% para-
formaldehyde, infiltrated with RNase-free sucrose at
4°C for 18 hr, and embedded in OCT embedding
compounds (Miles Laboratories, Napenrille, FL). The
tissue blocks were stored in seal wrap at -80°C. Cry-
ostat sections (6 pm thick) were mounted on Fisher
SuperfrostPlus slides. Significant prehybridization
steps included (1) 0.2 M HC1 at room temperature for
20 min, (2) proteinase K (1 ~g/ml)at 37°C for 15 min,
(3) 4% paraformaldehyde at room temperature for 20
min, and (4) 0.25% acetic anhydride in triethanola-
mine buffer for 10 min at room temperature. An-
tisense =S-labeled riboprobe was prepared from lin-
earized pSP-MT-1 using a Riboprobe R System Kit
from Promega following manufacturer's protocol.
The riboprobe was diluted in hybridization buffer to
deliver 5 x lo5 cpd60-pYslide. Similarly applied
were unrelated probes that served as one set of neg-
ative controls. Sections pretreated with RNase A (10
pg/ml) before hybridization with the MT probe
served as the second set of negative controls. After
coverslipping with siliconized cover slips and sealing
with petroleum ether-rubber cement (1:4), the sec-
tions were hybridized for 16hr in a humid chamber at
50°C overnight. Signhcant posthybridization steps
included (1)washes with 4 x SSC 4 times for 15 min
at 50°C, (2) RNase A digestion (10 pg/ml) 2 x for 15
min at 37°C and (3) washes in 2 x SSC and 0.1 x SSC
each for 30 min at 37°C. Dehydration was carried out
at 70%, 95%, and 100% ethanol with 0.3 M ammo-
nium acetate. Slides were dipped in Kodak NTB-2
emulsions diluted 50:50 with 0.6 M ammonium ace-
tate. The length of exposure at 4°C ranged from 5 to
7 days. Development of autoradiographs was carried
out in Kodak D-15 developer. After drying, sections
were counterstained with toluidine blue.
RESULTS
Effects of Cd and Z n on MT-I mRNA in
Rat Prostate
Class IIB metals, Cd and Zn, are potent inducers of
MT gene expression in liver, kidneys, and many
other organs [5]. In this study, the effects of these
metal ions on MT mRNA levels in the two major pro-
static lobes, VP and DLP, were investigated. Intact
NBL rats were injected with 50 pmol kg-' body
weight of Zn or Cd subcutaneously. Based on litera-
ture reports [12], these doses of Cd and Zn were
shown to cause induced levels of expression in rat
liver. Northern blot hybridization (Fig. lA, DLP) re-
vealed a single 0.55- to 0.6-kb transcript in the DLP of
untreated intact rats (data not shown in Fig. 1) and
rats injected with saline (Fig. lA, DLP, lane 1 from
the left, Sal), thus suggesting constitutive expression
of MT transcription in this prostatic lobe. Two days
after Cd or Zn administration, MT transcript levels
remained unchanged or might even be slightly sup-
pressed when compared with saline-treated control
values (Fig. lA, DLP, lanes 2 and 3, High Cd and
High Zn). Contrary to the DLP, MT transcripts were
not detected by Northern hybridization in the VPs of
untreated, and saline- or metal ion-treated rats (Fig.
lB, VP, lanes 1,2, and 3). These findings suggest that
mRNA is not detectable by Northern hybridization
and MT expression in this prostatic lobe is not induc-
ible by conventional inducers such as Zn and Cd. The
absence of MT mRNA in rat VP and the inability of
Zn and Cd to induce de nova VP mRNA synthesis
were confirmed by RT-PCR analyses, demonstrating
lack of amplification products from all VP RNA sam-
ples.
Effects of Castration and Short-Term Replacement
Therapies With DHT or E, on DLP MT
mRNA Levels
Prostatic functions are largely regulated by andro-
gens and estrogens produced by the testis [25-271.
Thus, orchiectomy usually leads to down-regulation
of androgen- and/or estrogen-dependent genes, while
hormone replacement therapy restores the transcrip-
tional activities of these genes. In this study, NBL rats
were orchiectomized and 9 days later implanted with
Silastic capsules filled with a nonaromatizable andro-
gen (DHT) or E2 or no hormone. Rats were sacrificed
6 days after Silastic capsuleimplantation. Levels of MT
transcripts in VPs and DLPs of the treated animals
were assessed by Northern hybridization and com-
pared to values obtained from the prostatic samples of

Sal High High
Cd Zn
A. DLP
MT
18s rRNA
B. VP
MT
18s rRNA
Fig. 1. Left: Northern blot analysis of MT mRNA in the DLPs
(A) and VPs (B) of rats treated with saline (a), 50 Fmolekg body
weight of CaCI, (High Cd), or 50 pmolekg body weight of ZnCI,
(High Zn). Injections were administered subcutaneously, animals
sacrificed 2 days after injection, and tissues excised for total RNA
preparation. Aliquots of I0 Fg of RNA were fractionated on I.O%
denaturing agar- gel, transferred to nylon membrane, hybridized
to [32P]-mouseMT- I DNA probe, washed and autoradiographed.
Blots were finally re-probed with a 30-mer end-labeled probe for
18s rRNA [ 331. Loading of the RNAs in each lane of the blots was
fairly similar, as judged by the intensity of the I8SrRNAsignals. The
untreated intact rats. Arbitrarily, levels found in the
untreated intact controls (Fig. 2, lane 1 from the left,
8 ) were assigned a value of onefold. Two weeks after
orchiectomy a 60% reduction in the levels of MT tran-
script was evident in DLP (Fig. 2, lane 2, 8). Replace-
ment therapy with DHT for 6 days was able to restore
MT mRNA levels to values found in intact rats (Fig. 2,
lane 3, 8 + DHT) while an equal duration of E, re-
placement failed to return the transcript levels to val-
ues of noncastrates (Fig. 2, lane 4, 8 + E2).This find-
ing indicates that basal levels of MT mRNA expression
in rat DLP are maintained by testicular androgen, and
not by EZ.Northern blot and RT-PCR analyses again
MetallothioneinmRNA in Rat Prostate 95
Relative MT mRNA Levels
ma mu
hybridization signals for MT mRNA were quantitated by densito-
metric scanning and normalized with respect to the corresponding
18s rRNA signal. No MT mRNA signals were detected in any of
the VP preparations. Right: Relative MT mRNA levels in DLP
samples were expressed as fold induction (Histograms). Fold in-
duction in DLP preparationsfrom saline-treated rats was arbitrarily
assigned a value of I. Fold induction of MT mRNA in the metal
ion-treated groups was calculated as a ratio of normalized MT
signal in the treated sample and that of the saline-treated sample.
Histograms are mean relative MT mRNA contents of three sepa-
rate experiments.
demonstratednondetedable levels of MT mRNA in all
the VP samples derived from these animals (data not
shown).
Effects of Long-Term T, E,, and T +
E, Treatment
on Prostatic MT-I Transcript
This experiment was designed to ascertain the
long-term effects of androgen and estrogen on DLP
MT expression. Four groups of rats were treated for
16 weeks with the following regimens: (1)implanted
with T-filled capsules, (2) implanted with E,-filled
capsules, (3) implanted simultaneously with T- and
96 Ghatak et al.
+ +
DHT E2
MT
18s rRNA
Fig. 2. Lek Northern blot analysis of M T mRNA and 18s rRNA
in the DLPs of intact rats ( 6 or intact), rats orchiectomized for 9
days ( 8 o r castrated), orchiectomized rats treated with DHT for 6
days ( 8 + DHT or castrate + DHT), or orchiectomized rats
+ +
treated with E, for 6 days (8 E,, castrate EJ. Ten pg of total
RNA was loaded in each lane. Right: Relative MT mRNA levels
were expressed as fold induction (see Fig. I legend for details). MT
E,-filled capsules, or (4) implanted with empty cap-
sules (untreated controls). The T-filled capsule im-
plants were shown to simply maintain a physiological
level of T in treated animals, while the E,-filled cap-
sule implants were able to elevate plasma E, by three-
to fourfold [%]. Levels of MT mRNA in the DLP of
untreated controls were assigned an arbitrary value
of 1(Fig. 3, lane 1 from the left), and levels in treated
groups were compared to these values. Sixteen
weeks of T treatment did not alter MT mRNA level in
the DLP (Fig. 3, lane 2, T) while E, treatment induced
a mild increase in the transcript’s level to 2.8-fold of
untreated control value (Fig. 3, lane 3, E,). Finally, T
+ E, treatment caused a drastic increase in DLP MT
mRNA transcript level to fivefold of that observed in
the untreated controls (Fig. 3, lane 4, T + E,). MT
mRNA was shown to be absent in VP samples of
treated and untreated animals by Northern blot and
RT-PCR analyses (data not shown).
In Situ Hybridization of MT Transcripts
In situ hybridization localized occurrence of MT
transcripts to the epithelia of the lateral and dorsal
prostates but not to the VP epithelium of untreated
intact rats (data not shown). The signal intensity in
the epithelial cells of the dorsal lobe was less than
Relative MT mRNA Levels
mRNA hybridization signals were normalized with respect t o the
corresponding 18s rRNA signal. Normalized signals in samples
from intact untreated rats were assigned values of I.O. Normalized
signals from all other groups were compared to those of the un-
treated intact rats. Histograms are mean relative M T mRNA con-
tents of three separate experiments.
that observed in cells of the lateral lobe. No signal
was detected in sections of the VP. A similar pattern
of MT transcript distribution was observed in pros-
tatic lobes of 16-week T + E,-treated rats (Fig. 4). The
combined hormone treatment caused marked in-
crease in signal intensity in lateral (LP) and dorsal
(DP) prostatic sections with expression in the lateral
prostatic sections consistently higher than that found
in the dorsal prostatic sections. Sections form VPs of
T + E,-treated rats exhibited no signal (Fig. 4, VP).
DISCUSSION
The rat prostate is composed of three anatomically
distinct lobes designated the ventral (VP), dorsal, and
lateral lobes, owing to their relationships to the ure-
thra [25-271. However, because of the close associa-
tion of the paired lateral lobes with the dorsal lobe,
these lobes are often dissected out as a single struc-
tural unit known as the dorsolateral complex or the
DLP. Rat VP and DLP have separate embryologic or-
igins [25] and are morphologically and functionally
distinct from each other [26,27]. The presence of MT
in the rat prostate has been a controversial question.
Immunocytochemicalstudies revealed strong positive
staining for MT-like molecules in the glandular epi-
thelia of the DLP, while most of the epithelial cells in

T EZT+EZ
MT
18s rRNA
Fig. 3. Left: Northern blot analysis of MT mRNA and I85 rRNA
in the DLPs of untreated rats (Ist lane from the left), rats treated
for I6 weeks with T (T), E, (Ed or T + +
E, (T E,). Ten pg of
total RNA was loaded in each lane. Right: Relative MT mRNA
levels were expressed as fold induction (see Fig. I legend for
details). MT mRNA hybridization signals were normalized with re-
LP
Fig. 4. Localization of MT message to the epithelia of lateral
prostate (LP) and dorsal prostate (DP) of rats treated 16 weeks
with T + E, by in situ hybridization. The grain density over the LP
the VP were negatively stained [15,23,24]. Biochemical
studies, however, demonstrated the presence of Cd-
binding proteins in both VP and DLP, but these mol-
ecules were different from classical MTs in terms of
amino acid composition [2,37]. Results from these past
MetallothioneinmRNA in Rat Prostate 97
Relative MTmRNA Levels
spect to the corresponding 18s rRNA signal. Normalized signals in
samples from intact untreated rats were assigned values of I.O.
Normalized signals from all other groups were compared to that of
the untreated intact rats. Histogramsare mean relative MT mRNA
contents of three separate experiments.
DP VP
epithelium is greater than that over the DP epithelium. By contrast,
positive signals were absent in the VP section. Light micrographs,
X 750.
studies raised the question of whether "true" MT is
synthesized in the rat prostate. Data presented in this
study help clanfy the controversy. MT mRNA was
demonstrated for the first time in the rat prostate.
Northern blot and in situ hybridization analyses re-
98 Ghatak et al.
vealed constitutive expression of MT mRNA in the
glandular epithelium of the DLP and a lack of expres-
sion of the transcript in the VP. RT-PCR analyses
further confirmed the absence of MT mRNA in rat VP.
Furthermore, both topological distribution and cellu-
lar localization of MT transcripts were found to be
similar to those reported for immunoreactive MT mol-
ecules [15,23,24]. Taken together, these findings dem-
onstrated a clear presence of MT expression in the
dorsolateral lobe and a lack of expression of the gene
in the ventral lobe of the rat gland.
Although the precise functions of MT are un-
known, it has been postulated that the protein par-
ticipates in the detoxification of heavy metals. This
view is supported by observations that, for most
cells, high levels of MT are synthesized during Cd
challenge and the protein binds Cd and other heavy
metal ions with high affinities [35]. Additional evi-
dence in support of a protective role of MT against
heavy metal toxicity were provided for from in vitro
and in vivo studies. Cultured cells selected for Cd
resistance were shown to synthesize large amount of
MT due to gene amplification [38,39], while genetic
inactivation of MT genes in the mouse led to in-
creased susceptibility of the liver to Cd poisoning
[a]. Furthermore, a growing body of evidence has
emerged to implicate a possible role of MT in Cd car-
cinogenesis. Tissue susceptibility to Cd carcinogenic-
ity is linked to deficiency in MT [2]. In this regard, it
is of great interest to reveal, in this study, that MT is
neither expressed nor inducible by Cd in the VP,
while the DLP exhibits a basal level of expression.
Coincidentally, systemic exposure of Waster rats to
Cd induced development of prostatic adenomas (31%
incidence) exclusively in the VPs and not in the DLPs
of the treated animals [l].Thus, the apparent “si-
lence” of the MT gene in rat VP appears to correlate
with this tissue’s unique susceptibility to Cd carcino-
genesis [1,2]. Other target organs of Cd carcinogen-
esis include rodent and monkey testes and hamster
ovary, all of which have been shown to produce little
or no MT [41].
Another often suggested function of MT concerns
intracellular Zn homeostasis. MT is believed to play a
crucial role in regulating intracellular Zn storage
andor in modulating the availability of zinc ions to
proteins in which Zn has a functional role [5].Zinc
ions are cofactors of many enzymatic reactions, and
they also serve as structural elements such as “zinc
fingers” in many DNA-binding proteins [42]. This
metal ion is present in high concentrations in human
and rodent prostates [20,21]. Although the biological
functions of zinc in the prostate remain unclear, de-
ficiency in this essential nutrient could lead to male
infertility [43]. In the rat prostate, Zn distribution is
~
mainly restricted to the lateral and dorsal lobes, with
the lateral lobe having the highest levels among all
body organs [21]. Apparently, prostatic content is un-
der androgen regulation. DLP Zn contents increase
with sexual maturity during the period of 3-6 weeks
of age (44,451,showing a strong positive correlation
with plasma androgen levels, which rise dramatically
during this developmental stage [44,45]. In this pa-
per, we demonstrated constitutive expression of MT
transcripts in the glandular epithelia of the lateral and
dorsal prostate of intact rats. The highest level of MT
mRNA expression was consistently demonstrated in
the lateral lobe, with the dorsal lobe trailing in mes-
sage abundance and the VP not expressing the tran-
script. This pattern of intraprostatic MT transcript ex-
pression showed remarkable resemblance to the
distribution of zinc ions among the various lobes of
the rat gland (lateral > dorsal > ventral [21]). Addi-
tionally, the levels of MT transcripts expressed in the
DLP exhibited an apparent dependency on testicular
androgens. Castration brought about drastic reduc-
tion of MT mRNA levels in the DLP while re-admin-
istration of an androgen effectively restored tran-
script levels to those found in intact animals. Taken
together, these results argue in favor of a possible
link between MT expression and prostatic Zn homeo-
stasis. They also raise the possibility that MT expres-
sion in the DLP may be regulated either directly by
androgens or indirectly via the hormone’s ability to
modulate Zn contents in the tissue.
It is a well-accepted concept that higher levels of MT
synthesis can be induced in a great variety of tissues
and cell lines by Zn or Cd challenge [5]. However, in
the present paper, we were unable to demonstrate the
efficacy of either metal ions, when administered at the
aforementioned doses, to induce a higher level of MT
expression in the DLP. These findings may be inter-
preted as indicative of a lack of response of the MT
gene in the DLP to ZdCd stimulation. An alternative,
and perhaps more likely, explanation may be formu-
lated, however, if one takes into consideration of the
high concentrations of zinc found in the DLP. It is
possible that MT expression in the DLP is maximally
stimulated by the high content of Zn within DLP cells
and is unable to be further augmented by exogenous
metal ions.
Lastly, the long-term effects of androgens and es-
trogens on prostatic MT transcript levels have also
been investigated in the present study. We are par-
ticularly interested in the long-term effects of sex ste-
roids on prostatic functions because we have previ-
ously demonstrated that long-term treatment of
Noble (NBL) rats with a combination of T and E, (T +
E,) induces a preneoplastic lesion, termed dysplasia,
selectively in the DLP of all treated animals [36,46].

Extension of the hormonal treatment period through-
out the animal’s life span causes high incidence of
adenocarcinoma development in the DLP [3,4]. We
here observed a fivefold increase in MT mRNA levels
+
in the dysplastic DLPs of the 16 week T E,-treated
rats and a two- to threefold increase in the DLPs of
E,-treated animals while no increase in transcript lev-
els was observed in the glands of T-treated animals.
These results clearly demonstrated induction of
higher levels of MT expression in the DLP by long-
term estrogenic or conjoint estrogenic-androgenicac-
tion. No previous study has reported a regulatory
effect of estrogen on MT expression. Furthermore,
when E, was re-administered to castrated rats, it did
not restore the low levels of h4T mRNA found in cas-
trates to values observed in intact animals. Jointly,
these results suggested that the long-term estrogenic
effect on prostatic MT expression may be indirect.
One possible mediator could be serum prolactin
which has been shown to rise following prolonged
estrogen exposure [47. Prolactin may directly influ-
ence DLP MT expression or exert its effect via eleva-
tion of zinc content (48). Alternatively, since long-
term T + E,-treatment of rats has been shown to
induce lipid peroxidation fluorescence product accu-
mulation in rat DLP [48],it is also possible that DLP
MT mRNA levels are elevated in response to oxida-
tive stress signals. The latter possibility is in accord
with an emerging view that MT plays the role of an
intracellular antioxidant by scavenging free radicals
[18,49,50].
In sum, this paper was first to demonstrate the
presence of MT mRNA in the DLP and the absence of
this transcript in the VP of the rat. Its results also
suggest either direct or indirect regulation of DLP MT
expression by sex hormones. In addition, it reveals
elevated levels of MT message in DLPs with dyspla-
sia, a presumed preneoplastic lesion. Taken together,
these findings thus suggest several regulatory roles
played by MT in normal and aberrant prostatic tis-
sues.
ACKNOWLEDGMENTS
We thank Dr. Glen K. Andrews of the University
of Kansas Medical Center at Kansas City for the
Mouse MT-1 probe, and Dr. Robert Matusik of the
University of Manitoba at Winnipeg, Manitoba, Can-
ada for his helpful discussion. We are grateful to Dr.
Irwin Leav and Mr. Gerry Parker at Tufts Medical
School at Boston for their expert photographic assis-
tance. We acknowledge the Center for Reproductive
Research at Tufts for its assistance in in situ hybrid-
ization analysis. This work was supported in part by
grants CA-60923 and CA-15776 awarded by the Na-
tional Cancer Institute.
MetallothioneinmRNA in Rat Prostate 99
REFERENCES
1. Waalkes MP, Rehm S, Riggs CW, Bare RM,Devor KE,
Poirier LA, Wenk ML, Henneman JR, Balaschak MS:
Cadmium carcinogenesis in male Wistar [Crl:
(WI)BR]rats: Dose-response analysis of tumor induc-
tion in the prostate, testes, and at the injection site.
Cancer Res 48:4656-4663, 1988.
2. Waalkes MP, Perantoni A, Rehm S: Tissue susceptibil-
ity factors in cadmium carcinogenesis. Correlation be-
tween cadmium-induction of prostatic tumors in rats
and an apparent deficiency of metallothionein. Biol
Trace Element Res 21:483-490, 1989.
3. Noble RL: Prostatic carcinoma of the Nb rat in relation
to hormones. Int Rev Exp Pathol23:113-159, 1982.
4. Drago J R The induction of Nb rat prostatic carcinomas.
Anticancer Res 4255-256, 1984.
5. Hamer DH: Metallothionein. Annu Rev Biochem 55:
913-951, 1986.
6. Palmiter RD: Molecular biology of metallothionein gene
expression. Experientia Suppl52:63-80, 1987.
7. Kagi JHR, Schaffer A: Biochemistry of metallothionein.
Biochemistry 278509-8515, 1988.
8. Thiele DJ: Metal-regulated transcription in eukaryotes.
Nucleic Acids Res 20:1183-1191, 1992.
9. Robinson NJ, Tommey AM, Kuske C, Jackson PJ: Plant
metallothioneins. Biochem J 295:l-10, 1993.
10. Searle PF, Davision BL, Stuart GW, Wilkie TM, Nor-
stedt G, Palmiter RD Regulation, linkage, and se-
quence of mouse metallothionein I and I1 genes. Mol
Cell Biol4:1221-1230, 1984.
11. West AK, Stallings R, Hildebrand CE, Chiu R, Karin M,
Richards RI: Human metallothionein genes: Structure
of the functional locus at 16q13. Genomics 8:513-518,
1990.
12. Andersen RD, Birren BW, Ganz T, Piletz JE, Hersch-
man HR: Molecular cloning of the rat metallothionein 1
(MT-1) mFNA sequence. DNA 2:15-22, 1983.
13. Andersen RD, Birren BW, Taplitz SJ, and Herschman
HR: Rat metallothionein-1 structural gene and three
pseudogenes, one of which contains 5’-regulatory se-
quences. Mol Cell Biol6:302-314, 1986.
14. Banerjee D, Onosake S, Cherian MG: Immunohisto-
chemical localization of metallothionein in cell nucleus
and cytoplasm of rat liver and kidney. Toxicology 24:
95-105, 1982.
15. Bataineh ZM, Heidger PM, Thompson SA, Timms BG:
Immunocytochemical localization of metallothionein in
the rat prostate gland. Prostate 9:397-410, 1986.
16. Thornalley PJ, Vasak M: Possible role for metallothio-
nein in protection against radiation-induced oxidative
stress. Kinetics and mechanism of its reaction with su-
peroxide and hydroxyl radicals. Biochim Biophys Acta
827:36-44, 1985.
17. Kagi JHR Overview of metallothionein. Methods En-
zymol205:613-626, 1991.
18. Bauman JW, Liu J, Liu YP, Klaassen CD: Increase in
metallothionein produced by chemicals that induce ox-
idative stress. Toxic01 Appl Pharmacol 11097-354,
1991.
19. Mawson CA, Fisher MI: The occurrence of zinc in the
human prostate gland. Can J Med Sci 30:336-339,1952.
20. Gunn SA, Gould TC: Differences between dorsal and
lateral components of dorsolateral prostate of rat by
Zn- uptake. Proc SOCExp Biol Med 92:17-20, 1956.
21. Gunn SA, Gould TC, Anderson WAD: Comparison of
I00 Ghatak e t al.
protein and zinc electrophoretic patterns of lobes of rat
prostates. Proc SOCFxp Biol Med 110:32-36, 1962.
22. Ogunlewe JO, Osegbe DN: Zinc and cadmium concen-
trations in indigenous blacks with normal hypertrophic
and malignant prostate. Cancer 63:1388-1392, 1989.
23. Umeyama T, Saruki K, Imai K, Yamanaka H, Suzuki K,
Ikei N, Kodaira T, Nakajima K, Saifoh H, Kimura M:
Immunohistochemical demonstration of metallothio-
nein in the rat prostate. Prostate 10:257-264, 1987.
24. Suzuki T, Umeyama T, Ohma C, Yamanaka H, Suzuki
K, Nakajima K, Kimura M: Immunohistochemical
study of metallothionein in normal and benign prostate
hyperplasia of human prostate. Prostate 19:35-42,
1991.
25. Price D: (1963). Comparative aspects of development
and structure in the prostate. In Vollmer EP (ed): “Bi-
ology of the Prostate and Related Tissues.” National
Cancer Institute Monograph 12. Bethesda, MD: U.S.
Department of Health, Education and Welfare, Public
Health Service. National Cancer Institute, pp 1-27.
26. Jesik CJ, Holland JM, Lee C: An anatomic and histo-
logic study of the rat prostate. Prostate 3:81-97, 1982.
27. Cunha G, Donjacour A, Cooke P, Mee S, Bigsby R,
Higgins S, Sugimura Y: The endocrinology and devel-
opmental biology of the prostate. Endocr Rev 8:338-
362, 1987.
28. Chomcyzynski P, Sacchi N: Single-step method of RNA
isolation by acid quanidinium thiocyanate-phenol-chlo-
roform extraction. Anal Biochem 162:156-159, 1987.
29. Chomcyzynski P: Solubilization in formamide protects
RNA from degradation. Nucleic Acids Res 20:3791-
3792, 1992.
30. Chomcyzynski P: One-hour downward alkaline capil-
lary transfer for blotting of DNA and RNA. Anal Bio-
chem 201:134-139, 1992.
31. Church GM, Gilbert W: Genomic sequencing. Proc Natl
Acad Sci USA 81:1991-1995, 1984.
32. De SK, McMaster MT, Dey SK, Andrews GK: Cell spe-
cific metallothionein gene expression in mouse decidua
and placentae. Development 107:611-621, 1989.
33. Clements JA, Matheson BA, Wines DR, Brady JA, Mac-
Donald RJ, Funder FW:Androgen dependence of spe-
cific kallikrein gene family members expressed in rat
prostate. J Biol Chem 263:16132-16137, 1988.
34. Yu M, Leav BA, Leav I, Merk FB, Wolfe HJ, Ho S-M:
Early alterations in ras protooncogene mRNA expres-
sion in testosterone and estradiol-17P induced prostatic
dysplasia of Noble rats. Lab Invest 68:33-44, 1993.
35. Hofler H, Childers H, Montminy M, Lechan R, Good-
man R, Wolfe H: In situ hybridization methods for the
detection of somatostatin mRNA in tissue sections us-
ing antisense RNA probes. Histochem J 18:597-604,
1986.
36. Leav I, Ho S-M, Ofner P, Merk FB, Kwan PW-L, Dam-
assa D: Biochemical alterations in sex hormone-in-
duced hyperplasia and dysplasia of the dorsolateral
prostate of Noble rats. J Natl Cancer Inst 80:1045-1053,
1988.
37. Waalkes MP, Perantoni A: Apparent deficiency of met-
allothionein in the Wistar rat prostate. Toxicol Appl
Pharmacol 101:83-94, 1989.
38. Beach LR, Palmiter RD: Amplification of the metal-
lothionein-1 gene in cadmium resistant mouse cells.
Proc Natl Acad Sci USA 78:2110-2114, 1981.
39. Durnam DM, Palmiter RD: Analysis of the detoxifica-
tion of heavy metal ions by mouse metallothionein. Ex-
perientia Suppl52:457-463, 1987.
40. Masters BA, Kelly EJ, Quaife CJ, Brinster RL, Palmiter
RD: Targeted disruption of metallothionein I and I1
genes increases sensitivity to cadmium. Proc Natl Acad
Sci USA 91~584-588, 1994.
41. Waalkes MP, Goering PL: Metallothionein and other
cadmium binding proteins: Recent developments.
Chem Res Toxicol 3:281-288, 1990.
42. Vallee BL, Falchuk KM: The biochemical basis of zinc
physiology. Physiol Rev 73:79-118, 1993.
43. Takihara H, Consentino MJ, Cockett AT: Zinc sulfate
therapy for infertile male male with or without varico-
celectomy. Urology 29:638-643, 1987.
44. Timms BG, Chandler JA: Ultrastructural and analytical
studies on the prostate of castrated rats. Prostate 4:37-
55, 1983.
45. Timms BG, Chandler JA: The effect of androgen on the
ultrastructure and subcellular-zinc distribution with
prostatic epithelia of castrated rat. Prostate 4:57-72,
1983.
46. Leav I, Merk FB, Kwan PW-L, Ho S-M: Androgen-sup-
ported estrogen-enhanced epithelial proliferation in the
prostates of intact Noble rats. Prostate 15:23-40, 1989.
47. Oldereid NB, Thomassen Y, Rui H, Gordeladze JO,
Haug E, Purvis K: Effects of oestrogen-induced hyper-
prolactinemia on the rat prostate. Andrologia 18:516-
522, 1986.
48. Ho S-M, Roy D: Sex hormone-induced nuclear DNA
damage and lipid peroxidation in the dorsolateral pros-
tates of Noble rats. Cancer Lett 84:155-162, 1994.
49. Bakka A, Johnsen AS, Endresen L, Rugstad HE: Ra-
dioresistance in cells with high content of metallothio-
nein. Experientia 38:381-383, 1982.
50. Chubatsu LS, Meneghini R: Metallothionein protects
DNA from oxidative damage. Biochem J 291:193-198,
1993.