BIOLOGY OF REPRODUCTION 84, 957–965 (2011)
Published online before print 5 January 2011.
DOI 10.1095/biolreprod.110.088575
Suppression of Stra8 Expression in the Mouse Gonad by WIN 18,4461
Cathryn A. Hogarth,3 Ryan Evanoff,3 Elizabeth Snyder,3 Travis Kent,3 Debra Mitchell,3 Christopher Small,3
John K. Amory,4 and Michael D. Griswold2,3
School of Molecular Biosciences,3 Washington State University, Pullman, Washington
University of Washington Medical Center,4 University of Washington, Seattle, Washington
Bis-(dichloroacetyl)-diamines (BDADs) are compounds that
inhibit spermatogenesis and function as male contraceptives in
many species; however, their mechanism of action has yet to be
fully investigated. It has been proposed that BDADs may
function via inhibition of testicular retinoic acid (RA) biosynthesis. We employed an organ culture technique and the
expression of a marker for RA activity, Stra8 (stimulated by
retinoic acid gene 8), to investigate if the BDAD WIN 18,446
inhibited the biosynthesis of RA from retinol (ROL) in neonatal
and adult murine testis and in the embryonic murine gonad.
After culturing either whole testes or germ cells isolated from
mice at 2 days postpartum (dpp) with WIN 18,446 or with WIN
18,446 plus ROL, Stra8 expression was suppressed, demonstrating that WIN 18,446 inhibited the conversion of ROL to RA in
both systems. We also utilized a transgenic mouse containing an
RA-responsive LacZ reporter gene to demonstrate limited RA
induction of LacZ expression in 2-dpp testes cultured with WIN
18,446 plus ROL. The expression of Stra8 was downregulated in
adult mouse testis tubules cultured with WIN 18,446 when
compared to tubules cultured with the vehicle control. WIN
18,446 also inhibited the conversion of ROL to RA in embryonic
ovaries and testes cultured for 48 h. These murine results
provide critical insights regarding how the BDADs can inhibit
spermatogenesis by blocking the ability of vitamin A to drive
germ cell development. In addition, these techniques will be
useful for screening novel inhibitors of RA biosynthesis as
potential male contraceptives.
Bis-(dichloroacetyl)-diamines, retinoic acid, spermatogenesis,
Stra8, testis
INTRODUCTION
Vitamin A plays an essential role in many different
physiological functions in mammals. Dietary vitamin A is
usually stored in the liver and transported throughout the serum
1
Supported by Contraceptive Center grant U54 42454 and by HD
10808 from the National Institutes of Health (NIH) to M.D.G. and by
the Eunice Kennedy Shriver National Institute of Child Health and
Human Development, a division of the NIH through cooperative
agreement U01 HD060408 to J.K.A. Data were deposited with the
National Center for Biotechnology Information gene expression and
hybridization array data repository (accession no. GSE25610; http://
www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc¼GSE25610).
2
Correspondence: Michael Griswold, School of Molecular Biosciences,
Biotechnology Lifescience Building, 1770 NE Stadium Way, Washington State University, Pullman, WA 99164. FAX: 509 335 4159;
e-mail: [email protected]
Received: 14 September 2010.
First decision: 1 November 2010.
Accepted: 28 December 2010.
Ó 2011 by the Society for the Study of Reproduction, Inc.
eISSN: 1529-7268 http://www.biolreprod.org
ISSN: 0006-3363
957
Downloaded from www.biolreprod.org.
as retinol (ROL), with metabolism to its active form, retinoic
acid (RA), taking place in target tissues [1]. Vitamin A
metabolism is under the control of two families of enzymes, the
alcohol and aldehyde dehydrogenases (Fig. 1) [1, 2]. Within
cells, RA binds to two families of intracellular receptors,
termed either RA receptors or retinoid X receptors [3, 4]. These
receptors, in the form of homo- and heterodimers, regulate
gene expression by binding to specific elements in the
promoter regions of genes under the control of RA.
Retinoic acid is essential for normal spermatogenesis. When
adult male mice are made vitamin A deficient (VAD), all
differentiated germ cells are lost from the seminiferous
epithelium, and only type A spermatogonia and Sertoli cells
remain [5]. However, this loss of differentiated germ cells is
reversible. An injection of either ROL or RA into a VAD male
mouse will reinitiate spermatogonial differentiation throughout
the testis in these animals [6, 7]. Recent evidence generated
using short-term cultures of neonatal testes and isolated
undifferentiated spermatogonia show that RA treatment of
these tissues and cells upregulates the expression of Stra8 and
Kit [8, 9], which are markers of differentiated spermatogonia,
and increases the number of cells containing nuclei reminiscent
of leptotene and zygotene spermatocytes [9]. Likewise, the
subcutaneous delivery of RA to mice at 2 days postpartum
(dpp) or to adult, vitamin A-sufficient male mice induced Stra8
transcript and/or an RA-responsive LacZ transgene in the testis
[10, 11]. Therefore, the importance of RA to germ cell
development and the reversibility of the spermatogenic block
that results from VAD suggest that the enzymes responsible for
the conversion of ROL to RA could make interesting targets
for the development of male contraceptives.
The bis-(dichloroacetyl)-diamines (BDADs) are a set of
compounds that target the aldehyde dehydrogenases and
prevent the oxidation of retinaldehyde to RA (Fig. 1). Nearly
50 yr ago, one particular BDAD, WIN 18,446, was shown to
completely and reversibly inhibit spermatogenesis in men
when dosed orally [12], and more recent studies with this
compound in mice and rabbits [13, 14] have revealed that WIN
18,446-treated testes in these species resemble those of VAD
mice [13]. The recent study in adult rabbits also demonstrated
that both tissue RA and Stra8 expression were significantly
reduced by treatment with WIN 18,446 [14]. However, the
mechanism of action for WIN 18,446 has yet to be fully
studied in the adult mouse testis, and whether WIN 18,446 is
also active in the neonatal testis or the embryonic gonad is
unknown. In addition to the marked impairment of spermatogenesis, men treated with WIN 18,446 experienced a
‘‘disulfiram reaction’’ when they drank alcohol, because this
compound also inhibited the aldehyde dehydrogenase responsible for breakdown of the toxic acetaldehyde to acetic acid in
the liver. Because of this adverse reaction, further development
of this particular compound for use as a male contraceptive was
abandoned without pursuing an understanding of how the
BDADs effectively and reversibly inhibit spermatogenesis and
ABSTRACT
958
HOGARTH ET AL.
FIG. 1. Vitamin A metabolism. The dietary form of vitamin A in animals, ROL, is metabolized to RA via a two-step enzymatic process. ROL is converted
to retinaldehyde by the alcohol dehydrogenases and the short-chain dehydrogenases in a reversible manner. Retinaldehyde is then modified irreversibly to
form RA by the aldehyde dehydrogenases. WIN 18,446, one of the BDAD compounds, can inhibit the enzymatic activity of at least one of the aldehyde
dehydrogenases and prevent the conversion of ROL to RA [14].
MATERIALS AND METHODS
Animals and Tissues
All animal experiments were approved by the Washington State University
Animal Care and Use Committee and were conducted in accordance with the
guiding principles for the care and use of research animals of the National
Institutes of Health. BL/6–129 and CD1 mouse colonies were maintained in a
temperature- and humidity-controlled environment with food and water
provided ad libitum. The BL/6–129 mice, ranging in age from birth to
adulthood (35–90 dpp), and the CD1 time-mated pregnant female mice used in
the present study were collected from these colonies. The animals were
euthanized by decapitation (fetuses and those aged 0–10 dpp) or using CO2
asphyxiation followed by cervical dissociation (10 dpp to adult), and their testes
or ovaries were dissected. Fetal gonad tissues were collected from CD1 mice
embryos staged by forelimb and hindlimb morphology [15]. Because it is
visually impossible to determine the sex of an embryo at Embryonic Day 11.5,
the tail of each embryo was removed during dissection and sexed using Sry
PCR as previously described [16]. Cultured tissue samples for RNA preparation
were snap-frozen immediately after collection and stored at 808C until use.
Grooved Agar Mold Cultures
Isolated urogenital ridges, 2-dpp mouse testes cut into four pieces, or small
pieces of stage-dissected adult mouse testis tubules were cultured on 1.5% agar
blocks with a groove running down the center [17]. Each agar block was placed
into a well of a 24-well plate containing 300 ll of Dulbecco modified Eagle
medium supplemented with 10% fetal bovine serum and cultured at 378C with
5% CO2. All tissues used for Stra8 expression analysis were cultured with 0.7
lM RA, 0.7 lM ROL, or 1 lM WIN 18,446. Tissues used for betagalactosidase analysis were cultured with 0.7 lM RA, 1.4 lM ROL, or 2 lM
WIN 18,446. The final concentrations of RA and ROL used were previously
published [8, 18], and dose-curve analyses were performed to determine the
optimal WIN 18,446 doses for real-time RT-PCR and colorimetric assay
endpoints (data not shown). For each organ culture, the agar block was
equilibrated in medium plus treatment or vehicle (dimethyl sulfoxide [DMSO])
for 24 h. Then, the medium and treatment were replaced and the tissue placed
inside the groove. The neonatal testes and adult testis tubules were cultured for
24 h and the embryonic gonads for 48 h. During the 48-h culture, the medium
was refreshed at the 24-h time point. For the experiments using whole neonatal
testes, adult mouse testis tubules, or embryonic gonads, each treatment or
vehicle-control incubation was performed at least in triplicate on testis tissue or
embryonic gonad collected from at least three different animals or embryos. A
Student t-test was performed to determine statistically significant differences
between vehicle and treated samples.
Gonocyte Cultures
Testes were collected from 2-dpp mice into Hanks balanced salt solution
and the tunica removed. MicroBeads with mouse CD90 (THY1) antibodies and
MACS Separation columns (Miltenyi Biotec) were used in the isolation
according to manufacturer’s instructions. The culture was performed as
previously described [8]. Isolated cells were cultured under feeder cell-free,
serum-free, and growth factor-free conditions for 24 h with only vehicle
(DMSO), 0.7 lM RA, 0.7 lM ROL, or 1 lM WIN 18,446. Cell viability was
monitored by staining a small aliquot of cells after culture with 0.4% trypan
blue solution, with less than 2% of cells found to be dead in each aliquot (data
not shown). For each cell isolation, 12–15 mice at 2 dpp were used to obtain the
necessary number of gonocytes. Each treatment or vehicle-control incubation
was performed on at least three different cell isolations. A Student t-test was
performed to determine statistically significant differences between vehicle and
treated samples.
Real-Time RT-PCR
A two-step, real-time RT-PCR was used to measure the expression of
candidate genes as previously described [19]. The RNA samples were analyzed
in triplicate with primers specific for the target genes. Stra8 primers amplified a
151-bp product (primers, 5 0 -GTTTCCTGCGTGTTCCACAAG-3 0 and 5 0 CACCCGAGGCTCAAGCTTC-3 0 ); Cidea primers amplified a 150-bp product
(primers, 5 0 -GCAGCCTGCAGGAACTTATC-3 0 and 5 0 -CCAAGATCATGA
AATGCGTG-3 0 ); Fstl3 primers amplified a 118-bp product (primers, 5 0 CTTCCGGCAACATCAACAC-3 0 and 5 0 -ACTCCGTCGCAGGAATCTTT3 0 ); and control Rps2 primers amplified a 112-bp product (primers, 5 0 CTGACTCCCGACCTCTGGAAA-3 0 and 5 0 -GAGCCTGGGTCCTCTGA
ACA-3 0 ). Expression of Stra8 was normalized to expression of Rps2. Student
t-test was used to analyze all results. Data are presented as the mean 6 SEM.
Microarray Analysis
Array output was normalized via the GeneChip robust multiarray method,
and data analysis was conducted using GeneSpring (Version 11.0.2; Agilent
Technologies). Genes were considered to be regulated by BDAD treatment if
they 1) had a raw score of greater than 50 in at least one sample, 2) were
determined to be significantly different versus controls by ANOVA (P ¼ 0.05),
with a 5% false-discovery rate multiple-test correction, and 3) showed a 2-fold
or greater increase or decrease versus controls. Data were deposited with the
National Center for Biotechnology Information gene expression and hybridization array data repository (GEO accession no. GSE25610; http://www.ncbi.
nlm.nih.gov/geo/query/acc.cgi?acc¼GSE25610).
Downloaded from www.biolreprod.org.
whether other BDAD derivatives could act in a testis-specific
manner.
The present study represents, to our knowledge, the first indepth analysis of how WIN 18,446 inhibits spermatogenesis in
the murine gonad. We utilized an agar mold testis culture
technique to examine the effect of WIN 18,446 on spermatogenesis through the suppression of Stra8 expression and the
use of transgenic mouse line that expresses LacZ (official
symbol Tg(RARE-Hspa1b/lacZ)12Jrt, hereafter referred to as
RARE-hsplacZ) in response to RA signaling as markers of RA
activity. In the present study, we show, again to our knowledge
for the first time, that WIN 18,446 can inhibit the conversion of
ROL to RA in the whole neonatal testis, isolated neonatal germ
cells, adult mouse testis tubules, and the embryonic murine
gonad.
SUPPRESSION OF Stra8 BY WIN 18,446 IN THE GONAD
959
Functional annotation cluster analysis was performed using tools in the
Database for Annotation, Visualization, and Integrated Discover (DAVID)
Bioinformatics Resources (http://david.abcc.ncifcrf.gov) [20, 21]. Affymetrix
Gene Chip Probe IDs identified in the microarray analysis as being regulated by
WIN 18,446 were uploaded to DAVID Bioinformatics Resources, and
functional annotation clustering was used to examine biological process gene
ontology (GO) terms defined as being significantly overrepresented. Enrichment scores for each cluster are reported, as are the most biologically relevant
GO term in the cluster and the number of transcripts represented.
Germ Cell Nuclear Antigen Immunofluorescence
Beta-Galactosidase Staining and Counting
Cultured RARE-hsplacZ testes were washed twice in PBS, then fixed in 4%
paraformaldehyde with a 0.25% gluteraldehyde additive in PBS for 1 h at 48C.
Fixed tissue was washed and stained in bromo-chloro-indolyl-galactopyranoside (X-gal) as previously described [22]. Samples were washed in LacZ buffer
and then in PBS three times before soaking in an equal-parts solution of 70%
ethanol and PBS. Testis pieces and embryonic gonads were photographed in
PBS before being dehydrated and embedded in paraffin for sectioning.
The paraffin-embedded, cultured RARE-hsplacZ neonatal testes were
sectioned (thickness, 4 lm), melted onto glass slides, and then stained with
Harris hematoxylin. For beta-galactosidase activity analysis, germ cells were
considered to be beta-galactosidase positive if they contained two or more foci
of beta-galactosidase activity (light blue staining) or definitive, diffuse
cytoplasmic staining. Quantification was performed on at least 50 tubule cross
sections from a minimum of three different pieces of cultured testis for each
sample. The frequency of beta-galactosidase-positive germ cells per tubule
cross section was determined for each treatment. The ratio of this value for each
treatment compared to that of the vehicle controls was then used for all
statistical analysis. A Student t-test was performed to determine statistically
significant differences between vehicle and treated samples. In all cases, a
minimum of three biological replicates and technical duplicates were utilized
for analysis.
RESULTS
WIN 18,446 Inhibited Conversion of ROL to RA in Whole
Neonatal Testes
Previous studies have demonstrated that daily treatment of
adult male mice with WIN 18,446 results in a VAD-like
phenotype in the testis [13], and very recent experiments in
adult male rabbits have shown that the levels of tissue RA and
Stra8 transcript levels are reduced in the testis after daily oral
WIN 18,446 treatment [14]. However, the mechanism of WIN
18,446 action that generates the VAD-like phenotype in mice
has yet to be described, and whether WIN 18,446 is also active
in younger animals is unknown. To examine the effect of WIN
18,446 on RA activity in the neonatal testis, testes were
isolated from 2-dpp, wild-type male mice, with each then cut
into four pieces and incubated in grooved agar molds for 24 h
in the presence of RA, ROL, WIN 18,446, or a combination of
RA or ROL plus WIN 18,446. Total RNA was extracted from
these samples, and Stra8 expression was measured using real-
FIG. 2. Inhibition of Stra8 expression with WIN 18,446 treatment of
neonatal testes. Graph depicts Stra8 expression within 2-dpp neonatal
testis pieces after 24 h in culture with RA, ROL, WIN 18,446, or
combinations of these treatments. Relative real-time RT-PCR expression
level is given on the y-axis with the treatment group outlined on the x-axis.
Asterisks represent statistical difference of a treatment group compared to
control except where black lines indicate statistical significance between
treatment groups. Four independent samples were incubated with each
treatment. Error bars represent SEM. *P , 0.05, **P , 0.001.
time RT-PCR. RA was found to significantly induce Stra8
expression in neonatal testes cultured in grooved agar molds
(Fig. 2). This result matches what was observed previously
when neonatal testis pieces were cultured with RA on floating
filters in our laboratory [8]. ROL significantly induced Stra8
expression in cultured whole testes; however, the incubation of
testes with WIN 18,446 alone or in combination with ROL
significantly reduced Stra8 expression. As a control sample,
testes were also cultured with a combination of RA and WIN
18,446, and an induction in Stra8 expression comparable to
that seen in samples cultured with RA alone was observed.
The effect of WIN 18,446 treatment on gene expression in
the testis is currently unknown. To determine whether global
changes occur in gene expression in the testis in response to
WIN 18,446 treatment, total RNA isolated from the 2-dpp,
wild-type testis cultures was processed and hybridized to
Affymetrix Mouse Genome 430 2.0 microarrays. The arrays
were analyzed to determine which genes were up- or
downregulated in these testis cultures in response to WIN
18,446 treatment. Arrays identified 250 transcripts representing
217 unique genes that were downregulated, and 111 transcripts
representing 108 unique genes that were upregulated, in
response to a 24-h WIN 18,446 treatment. The analysis tool
DAVID (http://david.abcc.ncifcrf.gov/) [20, 21] was used to
determine which ontology clusters were enriched in the lists.
Of the 250 downregulated transcripts, 148 were associated with
significantly overrepresented GO terms in a biological process.
Functional annotation clustering revealed clusters associated
with cytoskeletal protein binding (16 transcripts, enrichment
score of 4.62), cell junction (18 transcripts, enrichment score of
2.4), and reproductive developmental process (12 transcripts,
enrichment score of 2.06). Of the 111 upregulated transcripts,
64 were associated with overrepresented GO terms in a
biological process. Annotation clustering revealed clusters
associated with protein glycosylation (33 transcripts, enrichment score of 2.31) and proteolysis (11 transcripts, enrichment
score of 1.55).
Tables 1 and 2 present the 10 annotated transcripts that
displayed the highest fold-changes, up- or downregulated,
Downloaded from www.biolreprod.org.
Immunofluorescence using germ cell nuclear antigen (GCNA) antibody
(George Enders, Department of Anatomy and Cell Biology, The University of
Kansas Medical Center, Kansas City, KS) was performed on the THY1positive gonocytes isolated from the 2-dpp testes to determine cell purity. The
isolated cells were left to settle onto a glass coverslip for 1 h, fixed in Bouin
solution for 30 sec, and then rinsed three times in PBS. The cells were then
blocked with 10% normal rabbit serum diluted in PBS (blocking solution) for
30 min before an overnight incubation in GCNA antibody (1:50 dilution in
blocking solution) at 48C. The cells underwent four 5-min washes with PBS
before being incubated in rabbit anti-rat immunoglobulin M biotin-conjugated
secondary antibody (1:500 dilution in blocking solution; Cortex Biochem).
After four more 5-min PBS washes, bound antibody was detected using a 1-h
incubation in an Alexa Fluor 488-conjugated streptavidin compound (1:1000
diluted in blocking solution; Invitrogen). Visualization followed mounting
under coverslips using Vectashield with 4 0 ,6 0 -diamidino-2-phenylindole
(Vector Laboratories) and a Leica confocal microscope at the Washington
State University Imaging facility.
960
HOGARTH ET AL.
TABLE 1. Top ten annotated transcripts downregulated in response to
WIN 18,446 treatment.a
Gene
symbol
Cidea
Mybphl
Cytip
Eme2
Myoz2
Fstl3
Diras2
Stfa2l1
Vdr
Cldn23
a
Gene name
Cell death-inducing DNA fragmentation
factor, alpha subunit-like effector A
Myosin binding protein H-like
Cytohesin 1 interacting protein
Essential meiotic endonuclease 1
homolog 2
Myozenin 2
Follistatin-like 3
DIRAS family, GTP-binding RAS-like2
Stefin A2 like 1
Vitamin D receptor
Claudin 23
Fold
change
230.1
144.4
106.1
96.4
48.8
36.7
29.6
28.6
27.2
27.1
Microarray data GEO accession number GSE25610.
TABLE 2. Top ten annotated transcripts upregulated in response to WIN
18,446 treatment.a
Gene
symbol
Spata21
Kcnh1
Lbh
Strn
Mcpt2
Tpsb2
Car12
Rimbp2
Kcne1
Adamts4
a
Gene name
Spermatogenesis associated 21
Potassium voltage-gated channel,
subfamily H (eag-related), member 1
Limb-bud and heart
Striatin, calmodulin binding protein
Mast cell protease 2
Tryptase beta 2
Carbonic anhydrase 12
RIMS binding protein 2
Potassium voltage-gated channel, Iskrelated subfamily, member 1
A disintegrin-like and metallopeptidase
(reprolysin type) with thrombospondin
type 1motif, 4
Microarray data GEO accession number GSE25610.
Fold
change
81.5
69.6
56.63
33.28
29.2
23.5
22.9
20.5
18.2
17.3
FIG. 3. Inhibition of Cidea and Fstl3 expression with WIN 18,446
treatment of neonatal testes. Graphs depict the expression of Cidea (A) and
Fstl3 (B) within 2-dpp neonatal testis pieces after 24 h in culture with RA,
ROL, WIN 18,446, or combinations of these treatments. Relative real-time
RT-PCR expression level is given on the y-axis with the treatment group
outlined on the x-axis. Asterisks represent statistical difference of a
treatment group compared to control except where black lines indicate
statistical significance between treatment groups. Four independent
samples were incubated with each treatment. Error bars represent SEM.
*P , 0.05, **P , 0.001.
upregulated in the RA-treated sample and downregulated in the
samples treated with WIN 18,446 alone and with ROL plus
WIN 18,446 (Fig. 3). We also compared our lists of WIN
18,446-regulated transcripts with transcripts reported to be RAregulated in the testis [8, 9] (Table 3). Of the published RAupregulated transcripts, five (one which being Stra8) were
significantly downregulated by WIN 18,446, and of the RAdownregulated genes, two were significantly upregulated by
WIN 18,446 treatment (Table 3).
To visualize the effect of WIN 18,446 on RA activity, the
RARE-hsplacZ transgenic mouse line was utilized [24]. This
mouse expresses the LacZ gene under the control of the hsp
core promoter and a RA-response element (RARE). In these
animals, beta-galactosidase will be produced in cells containing
RA and the necessary RA-response machinery and can be
visualized using a colorimetric assay to detect enzyme activity.
Testes were isolated from 2-dpp RARE-hsplacZ transgenic
males and cultured for 24 h in the same manner described for
the wild-type mice. The tissue was then stained for betagalactosidase activity. At the level of the whole tissue, betagalactosidase staining appeared to increase in RA- and ROLtreated testes as compared to control (Fig. 4, A–F), whereas
beta-galactosidase staining appeared to be reduced by a
combination of ROL and WIN 18,446 (Fig. 4, G and H).
After whole-tissue examination, cultured RARE-hsplacZ testes
were processed for histology, and the number of beta-
Downloaded from www.biolreprod.org.
when vehicle-control and WIN 18,446-treated samples were
compared. Of the transcripts most highly downregulated by
WIN 18,446 (Table 1), cell death-inducing DNA fragmentation
factor, alpha subunit-like effector a (Cidea), and follistatin-like
3 (Fstl3) have been found to be upregulated by RA treatment of
testis tissue in a previously published microarray dataset [8],
and our laboratory has identified the induced expression of
these two genes in RA-treated testis samples collected from
mice of various ages (Hogarth, Evanoff, Small, and Griswold,
unpublished data). In addition, the expression of essential
meiotic endonuclease 1 homolog 2 (Eme2), a protein thought to
be involved in DNA recombination, was also downregulated
by WIN 18,446 (Table 1). Of the transcripts most highly
upregulated by WIN 18,446 (Table 2), the expression of
potassium voltage-gated channel, Isk-related subfamily, member 1 (Kcne1) has been found to be enriched in undifferentiated
spermatogonia, in normal rat testis tissue, and in human
seminoma samples characterized by the overproliferation of
undifferentiated germ cells [23]. In addition, the expression of
genes with known roles in spermatogenesis, such as reproductive homeobox 4 cluster (Rhox4), nuclear receptor subfamily 0,
group B, member 1 (Nr0b1), also known as Dax1, and
spermatogenesis- and oogenesis-specific basic helix-loop-helix
2 (Sohlh2), were also regulated by WIN 18,446.
To confirm the data generated by the arrays, real time RTPCR for two of the WIN 18,446-regulated genes that have been
shown previously to be regulated by RA [8], Cidea and Fstl3,
was performed. The expression of both Cidea and Fstl3 was
961
SUPPRESSION OF Stra8 BY WIN 18,446 IN THE GONAD
TABLE 3. RA regulated gene expression in response to WIN 18,446 treatment.a
Gene
symbol
Gene name
Fold change
WIN 18,446
Fold change
RA
Stra8
Dhrs3
Stra6
Por
Prdm1
Egr2
Pole
Stimulated by retinoic acid gene 8
Dehydrogenase/reductase (SDR family) member 3
Stimulated by retinoic acid gene 6
P450 (cytochrome) oxidoreductase
PR domain containing 1, with ZNF domain
Early growth response 2
Polymerase (DNA directed), epsilon
26.4
2.56
3.57
1.51
2.15
þ1.68
þ1.76
þ9.00*
þ6.00*
þ5.00*
þ2.00*
þ2.21 2.38 3.13 a
Negative numbers denote downregulated expression and positive numbers denote upregulated expression; WIN
18,446 microarray data GEO accession number GSE25610.
* Fold changes with RA treatment reported by Zhou et al., 2008 [8].
Fold changes with RA treatment reported by Pellegrini et al., 2008 [9].
cells. To determine whether germ cells alone were susceptible
to WIN 18,446 treatment, THY1-positive gonocytes were
isolated using magnetic cell sorting and then cultured for 24 h
with or without RA, ROL, and ROL plus WIN 18,446. After
each cell isolation, GCNA immunofluorescence was performed
to determine the purity of the isolated cells. In each case,
greater than 95% of the isolated cells were GCNA positive,
indicating that a nearly pure preparation of gonocytes had been
obtained (data not shown). After treatment, RNA was isolated
and Stra8 expression analyzed by real-time RT-PCR. Both RA
and ROL treatment alone significantly increased Stra8 levels;
however, the presence of WIN 18,446 decreased Stra8
expression in the isolated germ cells (Fig. 6).
WIN 18,446 Can Act Directly on Undifferentiated Germ
Cells to Suppress Stra8 Expression
WIN 18,446 Only Affects Stra8 Expression in Adult Mouse
Testis Tubules in Cells Associated with Clusters 2 and 3
WIN 18,446 was able to suppress Stra8 expression in
whole-testes cultures that contained both germ and somatic
The cells within an adult mouse testis tubule are arranged
such that specific subpopulations of differentiating germ cells
FIG. 4. WIN 18,446 inhibits the expression of a RA-driven LacZ expression construct. Panels depict vehicle (DMSO)
control-treated testis pieces (A) and a
representative cross section (B) as well as 2dpp RARE-hsplacZ mouse testis pieces and
a representative cross section cut from each
of these tissues treated with RA (C and D),
ROL (E and F), or ROL plus WIN 18,446 (G
and H). The black arrows indicate betagalactosidase-positive germ cells. Bars ¼
500 lm (A, C, E, and G) or 50 lm (B, D, F,
and H).
Downloaded from www.biolreprod.org.
galactosidase-positive germ cells per tubule cross section was
determined. To correct for inherent culture and litter variation,
the frequency of beta-galactosidase-positive germ cells within a
given treatment was compared to that of the vehicle controls. In
the case of both RA and ROL treatment, the frequency of betagalactosidase-positive cells was much greater than that
observed in vehicle controls. Conversely, treatment with both
ROL and WIN 18,446 resulted in a lower frequency of betagalactosidase-positive cells as compared to ROL treatment
alone. Although not significant, all samples exposed to WIN
18,446, either alone or in conjunction with ROL, had reduced
frequencies of beta-galactosidase-positive cells as compared to
vehicle controls (Fig. 5).
962
HOGARTH ET AL.
are in association. Under a dissection microscope, distinct
groups (clusters) of these associations can be distinguished,
because the germ cells will refract light differently depending
on their differentiation state [25]. Using this dissection
technique, adult mouse testis tubules can be physically
separated into four distinct clusters, which represent groups of
germ cell associations (stages) that are present within that piece
of tubule: Cluster 1 comprises stages XII and I; cluster 2, stages
II–VI; cluster 3, stages VII–VIII; and cluster 4, stages IX–XI.
This technique is known as stage dissection [25], and it was
utilized to test whether particular stages in the adult mouse testis
respond to RA, ROL, or WIN 18,446 treatment. The testes were
removed from adult male mice and the tubules dissected into
clusters. Pieces of tubules representing clusters 1–4 were placed
in grooved agar mold cultures for 24 h with RA, ROL, or WIN
18,446 alone or with combinations of either RA or ROL and
WIN 18,446. RNA was extracted from these samples, and
quantitative RT-PCR was performed to determine the expression of Stra8 in each of the treated samples compared to the
vehicle control. No significant differences in Stra8 expression
were seen in clusters 1 and 4 in any of the five treatments (data
not shown). Treatment of cluster 2 with RA induced Stra8
expression 6-fold, whereas treatment with WIN 18,446 alone
downregulated Stra8 (Fig. 7A). Interestingly, neither ROL nor
RA treatment had a significant effect on the expression of Stra8
in cluster 3; however, incubation of this cluster with WIN
18,446, alone or in combination with ROL, did significantly
reduced the amount of detectable Stra8 transcript (Fig. 7B).
Stra8 Expression in Embryonic Ovary Is Susceptible
to Treatment with WIN 18,446
Retinoic acid is also essential for meiosis in the embryonic
ovary, but to our knowledge, no published reports have
examined whether the BDAD compounds could also affect
female fertility. In addition, evidence suggests that RA is
important for the mitotic arrest of gonocytes in the embryonic
testis; therefore, the effect of WIN 18,446 treatment on
embryonic gonads from both sexes was investigated. Murine
embryonic gonads were collected at Embryonic Day 11.5, and
additional tissue was harvested from each embryo for
determination of sex by PCR. The gonads were cultured for
48 h with or without RA, ROL, and ROL with WIN 18,446.
Stra8 real-time RT-PCR analysis was conducted on RNA
isolated from these samples. In the presence of RA, Stra8
significantly increased in the cultured testes but not in the
ovaries. Stra8 levels decreased significantly when the embryonic ovary or testis was treated with WIN 18,446 (Fig. 8).
DISCUSSION
The present study is, to our knowledge, the first in-depth
analysis of how one particular BDAD compound, WIN 18,446,
could inhibit murine spermatogenesis. In the 1960s, WIN
18,446 was shown to safely, effectively, and reversibly inhibit
spermatogenesis in men; however, further research into how
WIN 18,446 exerted this effect on spermatogenesis was
suspended because of the serious side effects observed when
this compound was taken with alcohol [12]. A more recent
study with this compound in mice revealed that murine testes
treated chronically with WIN 18,446 closely resembled a VAD
testis, in which tubules were populated by only Sertoli cells and
undifferentiated spermatogonia [13]. Our report strongly
suggests that the likely cause of this phenotype is the blockage
of the conversion of ROL to RA by WIN 18,446 in germ cells
and that this block is effective not only in the adult testis but
also in the neonatal testis, isolated undifferentiated spermatogonia, and the embryonic gonad.
In each of the age groups and tissues tested, WIN 18,446
was able to block the conversion of ROL to RA. Expression of
both Stra8 and an RA-driven reporter gene in a transgenic
mouseline were downregulated after treating the neonatal testis
with both ROL and WIN 18,446; however, WIN 18,446 was
not able to negate the stimulatory effect of RA on these
markers in culture. Further analysis of the RARE-hsplacZ
reporter gene demonstrated that significantly more gonocytes
Downloaded from www.biolreprod.org.
FIG. 5. WIN 18,446 reduced the number of beta-galactosidase-positive
germ cells in cultured neonatal testes. Graph depicts the numbers of betagalactosidase-positive cells per tubule, relative to the vehicle control, after
24 h in culture with RA, ROL, WIN 18,446, or ROL plus WIN 18,446. The
relative numbers of beta-galactosidase-positive cells per tubule are in
given on the y-axis with the treatment group outlined on the x-axis. Values
represent the ratio of beta-galactosidase-positive germ cells per tubule in
the vehicles compared to beta-galactosidase-positive germ cells per
tubule for each treatment. Asterisks represent statistical difference of a
treatment group compared to control except where black lines indicate
statistical significance between treatment groups. Six independent
samples were incubated with each treatment, and all were used to count
beta-galactosidase-positive germ cells. Error bars represent SEM. *P ,
0.05, **P , 0.01, ***P , 0.001.
FIG. 6. Inhibition of Stra8 expression with WIN 18,446 treatment of
isolated gonocytes. Graph depicts Stra8 expression within isolated
gonocytes after 24 h in culture with RA, ROL, or ROL plus WIN
18,446. Relative real-time RT-PCR expression level is given on the y-axis
with the treatment group outlined on the x-axis. Asterisks represent
statistical difference of a treatment group compared to control except
where black lines indicate statistical significance between treatment
groups. Three independent samples were incubated with each treatment.
Error bars represent SEM. *P , 0.05, **P , 0.001.
SUPPRESSION OF Stra8 BY WIN 18,446 IN THE GONAD
963
FIG. 7. Inhibition of Stra8 expression with WIN 18,446 treatment of
stage-dissected clusters. Graphs depict Stra8 expression within adult
mouse testis tubules staged at cluster 2 (A) and at cluster 3 (B) after 24 h in
culture with RA, ROL, WIN 18,446, or combinations of these treatments.
Relative real-time RT-PCR expression level is given on the y-axis with the
treatment group outlined on the x-axis. Asterisks represent statistical
difference of a treatment group compared to control except where black
lines indicate statistical significance between treatment groups. Three
independent samples were incubated with each treatment. Error bars
represent SEM. *P , 0.05, **P , 0.001.
were positive for beta-galactosidase staining in the RA- and
ROL-treated, 2-dpp testes, whereas the addition of WIN
18,446 significantly reduced the ROL-induced beta-galactosidase staining in germ cells. A microarray analysis of the wildtype, 2-dpp cultures determined that transcripts coding for gene
products participating in a multitude of different cellular
functions were regulated by WIN 18,446. For example, WIN
18,446 altered the expression of known regulators of
spermatogenesis, including Dax1 and Sohlh2, yet this analysis
has also identified transcripts that either have no known
function in the testis or are yet to be described as being
regulated by RA, such as potassium voltage-gated channel,
subfamily H (eag-related), member 1 (Kcnh1). A second gene
encoding a potassium-channel protein that is known to be
enriched in undifferentiated spermatogonia but yet to be
reported as RA regulated, Kcne1 [23], was also identified as
being regulated by WIN 18,446 in the present study,
suggesting that further investigation of how retinoids regulate
the expression of genes involved in forming ion channels in the
testis is required.
The analysis of transcripts published as being either up- or
downregulated by RA demonstrated that WIN 18,446 was also
able to affect the expression of these transcripts, but in the
opposite direction. In other words, transcripts that have been
published as being upregulated by RA were found to be
downregulated by WIN 18,446 in our array analysis, and vice
versa. Further investigation of the transcripts, which are usually
upregulated by RA treatment but were downregulated by WIN
18,446, may provide further insight regarding why WIN
18,446 results in a VAD-like phenotype. Two examples of
such transcripts are Cidea and Fstl3, and this regulation was
confirmed by real-time RT-PCR analysis of the RNA samples
derived from the 2-dpp, whole-testis cultures. Cidea encodes a
cell-death protein that has been linked to regulating insulin
sensitivity and the maintenance of adipose tissue in mice [26],
but a role for this protein in testis development has yet to be
described. FSTL3 appears to also play a role in fat and glucose
homeostasis [27]; however, this protein has been shown to be
present in Leydig cells, spermatogonia, and mature spermatids
in the testis [28]. In addition, the fact that WIN 18,446 was able
to alter the expression of such a range of different transcripts in
the absence of added retinoids is very interesting. This suggests
that while the testes are in culture, the conversion of
endogenous ROL to RA can take place and that the
introduction of WIN 18,446 is able to block this process.
Why and how blocking the synthesis of RA regulates gene
expression in the testis, and whether this regulation has a direct
effect on germ cell development, will be the focus of future
studies in our laboratory.
Gonocytes appear to contain the machinery necessary to
convert ROL to RA. Serum free-cultures of gonocytes isolated
from 2-dpp testes increased Stra8 expression in response to
ROL treatment; however, WIN 18,446 was able to significantly
negate this effect. The ability of isolated gonocytes to
upregulate Stra8 in the presence of ROL has previously been
demonstrated [8], and recent data suggest that WIN 18,446 can
target the testis-specific aldehyde dehydrogenase 1A2 (ALDH1A2) [14]. Aldh1a2 has been shown to be expressed by
spermatogonia in the adult mouse testis [29], but further
investigation of whether the protein is present in germ cells of
the neonatal testis is required to determine whether this is the
Downloaded from www.biolreprod.org.
FIG. 8. Inhibition of Stra8 expression with WIN 18,446 treatment of
embryonic gonads. Graph depicts Stra8 expression within embryonic
mouse ovaries (open bars) and embryonic mouse testes (solid bars) after
24 h in culture with RA, ROL, or ROL plus WIN 18,446. Relative real-time
RT-PCR expression level is given on the y-axis with the treatment group
outlined on the x-axis. The relative expression levels for the testis and
ovary treatment groups have been compared only to the control group of
the same sex. Asterisks represent statistical difference of a treatment group
compared to control except where black lines indicate statistical
significance between treatment groups. Three independent samples were
incubated with each treatment. Error bars represent SEM. *P , 0.05, **P
, 0.001.
964
HOGARTH ET AL.
ACKNOWLEDGMENT
The authors thank Professor George Enders, University of Kansas
Medical Center, for supplying the GCNA antibody.
REFERENCES
1. Theodosiou M, Laudet V, Schubert M. From carrot to clinic: an overview
of the retinoic acid signaling pathway. Cell Mol Life Sci 2010; 67:1423–
1445.
2. Molotkov A, Ghyselinck NB, Chambon P, Duester G. Opposing actions of
cellular retinol-binding protein and alcohol dehydrogenase control the
balance between retinol storage and degradation. Biochem J 2004; 383:
295–302.
3. Livera G, Rouiller-Fabre V, Pairault C, Levacher C, Habert R. Regulation
and perturbation of testicular functions by vitamin A. Reproduction 2002;
124:173–180.
4. Mark M, Ghyselinck NB, Wendling O, Dupe V, Mascrez B, Kastner P,
Chambon P. A genetic dissection of the retinoid signaling pathway in the
mouse. Proc Nutr Soc 1999; 58:609–613.
5. Hogarth CA, Griswold MD. The key role of vitamin A in spermatogenesis.
J Clin Invest 2010; 120:956–962.
6. van Pelt AM, de Rooij DG. Synchronization of the seminiferous
epithelium after vitamin A replacement in vitamin A-deficient mice. Biol
Reprod 1990; 43:363–367.
7. Griswold MD, Bishop PD, Kim KH, Ping R, Siiteri JE, Morales C.
Function of vitamin A in normal and synchronized seminiferous tubules.
Ann N Y Acad Sci 1989; 564:154–172.
8. Zhou Q, Li Y, Nie R, Friel P, Mitchell D, Evanoff RM, Pouchnik D,
Banasik B, McCarrey JR, Small C, Griswold MD. Expression of
stimulated by retinoic acid gene 8 (Stra8) and maturation of murine
gonocytes and spermatogonia induced by retinoic acid in vitro. Biol
Reprod 2008; 78:537–545.
9. Pellegrini M, Filipponi D, Gori M, Barrios F, Lolicato F, Grimaldi P,
Rossi P, Jannini EA, Geremia R, Dolci S. ATRA and KL promote
differentiation toward the meiotic program of male germ cells. Cell Cycle
2008; 7:3878–3888.
10. Snyder EM, Small C, Griswold MD. Retinoic acid availability drives the
asynchronous initiation of spermatogonial differentiation in the mouse.
Biol Reprod 2010; 83:783–790.
11. Zhou Q, Nie R, Li Y, Friel P, Mitchell D, Hess RA, Small C, Griswold
MD. Expression of stimulated by retinoic acid gene 8 (Stra8) in
spermatogenic cells induced by retinoic acid: an in vivo study in vitamin
A-sufficient postnatal murine testes. Biol Reprod 2008; 79:35–42.
12. Heller CG, Moore DJ, Paulsen CA. Suppression of spermatogenesis and
chronic toxicity in men by a new series of bis(dichloroacetyl) diamines.
Toxicol Appl Pharmacol 1961; 3:1–11.
13. Brooks NL, van der Horst G. Short-term effects of N 0 ,N-bis(dichloroacetyl)-1,8-octamethylenediamine (WIN 18,446) on the testes, selected
sperm parameters and fertility of male CBA mice. Lab Anim 2003; 37:
363–373.
14. Amory JK, Muller CH, Shimshoni JA, Isoherranen N, Paik J, Moreb JS,
Amory DW, Evanoff R, Goldstein AS, Griswold MD. Suppression of
spermatogenesis by bisdichloroacetyldiamines is mediated by inhibition of
testicular retinoic acid biosynthesis. J Androl 2010; 32:111–119.
15. Kaufman MH. The Atlas of Mouse Development. San Diego: Academic
Press; 1992.
16. Small CL, Shima JE, Uzumcu M, Skinner MK, Griswold MD. Profiling
gene expression during the differentiation and development of the murine
embryonic gonad. Biol Reprod 2005; 72:492–501.
17. Martineau J, Nordqvist K, Tilmann C, Lovell-Badge R, Capel B. Malespecific cell migration into the developing gonad. Curr Biol 1997; 7:958–
968.
18. Koubova J, Menke DB, Zhou Q, Capel B, Griswold MD, Page DC.
Retinoic acid regulates sex-specific timing of meiotic initiation in mice.
Proc Natl Acad Sci U S A 2006; 103:2474–2479.
19. Zhou Q, Shima JE, Nie R, Friel PJ, Griswold MD. Androgen-regulated
transcripts in the neonatal mouse testis as determined through microarray
analysis. Biol Reprod 2005; 72:1010–1019.
20. Dennis G Jr, Sherman BT, Hosack DA, Yang J, Gao W, Lane HC,
Lempicki RA. DAVID: Database for Annotation, Visualization, and
Integrated Discovery. Genome Biol 2003; 4:P3.
21. Huang da W, Sherman BT, Zheng X, Yang J, Imamichi T, Stephens R,
Lempicki RA. Extracting biological meaning from large gene lists with
DAVID. Curr Protoc Bioinformatics 2009; Chapter 13:Unit 13.11.
22. McLean DJ, Russell LD, Griswold MD. Biological activity and
enrichment of spermatogonial stem cells in vitamin A-deficient and
hyperthermia-exposed testes from mice based on colonization following
germ cell transplantation. Biol Reprod 2002; 66:1374–1379.
23. Tsevi I, Vicente R, Grande M, Lopez-Iglesias C, Figueras A, Capella G,
Condom E, Felipe A. KCNQ1/KCNE1 channels during germ-cell
Downloaded from www.biolreprod.org.
aldehyde dehydrogenase that is being acted upon by WIN
18,446 in gonocytes.
To our knowledge, the present study is also the first to
investigate how the individual clusters of stages of the
seminiferous epithelium can respond to RA. A previous study
from our laboratory demonstrated that STRA8 could be
induced by RA in stages other than stages VII and VIII [11],
and the current study narrows this to stages II–VI. Of the four
adult mouse testis tubule clusters, only the cells present in
cluster 2 significantly upregulated Stra8 expression in response
to RA in culture. The current model for spermatogonial
differentiation in the adult testis proposes that spermatogonial
stem cells are located at random throughout the testis tubules
and that at a single point along any given tubule, one of these
spermatogonial stem cells receives the necessary signals to
differentiate and then commits to undergo meiosis. This
decision occurs at stage VIII within cluster 3 and is most
likely driven by RA, because the testes of VAD mice contain
only Sertoli cells and undifferentiated spermatogonia. Treating
VAD mice with RA stimulates the differentiation of the
undifferentiated spermatogonia throughout the testes of these
animals simultaneously [6]. The data in the present study
suggest that the undifferentiated spermatogonia in cluster 2 are
primed to respond to RA and that the lack of available RA to
these cells is one of the factors required to keep them in an
undifferentiated state. Whether this limitation is imposed by
regulating the delivery of RA to the germ cells, by the precise
expression of the retinoid metabolism enzymes and receptors,
or by a combination of both needs to be the focus of future
research to clearly define how the cycle of the seminiferous
epithelium is regulated.
The stage-dependent response to WIN 18,446 is also
informative regarding where the synthesis of RA may be
taking place across the stages of the cycle. Given that Stra8
expression was not induced after RA treatment of cluster 3 but
WIN 18,446 was able to downregulate Stra8 transcript
production in this cluster suggests that sufficient endogenous
RA production occurs to induce Stra8 expression in cluster 3
but also that this production can be inhibited by WIN 18,446.
The observation that WIN 18,446 alone can inhibit Stra8
expression in cluster 2 implies that the enzymes responsible for
the conversion of ROL to RA are present and that the
availability of ROL limits RA production. Clearly, further
analysis of the localization of the retinoid-metabolizing
enzymes throughout the cycle of the seminiferous epithelium
is required to enhance our understanding of how WIN 18,446
exerts a stage-specific effect in the adult mouse testis.
The present study demonstrates that WIN 18,446 can
effectively inhibit the conversion of ROL to RA in the
postnatal testis and in the embryonic gonad, making it an
excellent tool for continuing research into how RA regulates
germ cell development in both sexes. Until now, the
elimination of RA signaling has been controlled by inhibiting
the retinoid receptors, either through gene knockout studies or
receptor antagonists. However, these methods rely on a
particular receptor being expressed in a specific cell type and
do not eliminate possible functional redundancy that may occur
between receptors. WIN 18,446 will provide researchers with
another means of removing RA signaling from the gonad and
allow experiments in the juvenile and adult testis and in the
ovary to be streamlined through using the same compound.
SUPPRESSION OF Stra8 BY WIN 18,446 IN THE GONAD
differentiation in the rat: expression associated with testis pathologies. J
Cell Physiol 2005; 202:400–410.
24. Rossant J, Zirngibl R, Cado D, Shago M, Giguere V. Expression of a
retinoic acid response element-hsplacZ transgene defines specific domains
of transcriptional activity during mouse embryogenesis. Genes Dev 1991;
5:1333–1344.
25. Kotaja N, Kimmins S, Brancorsini S, Hentsch D, Vonesch JL, Davidson I,
Parvinen M, Sassone-Corsi P. Preparation, isolation and characterization
of stage-specific spermatogenic cells for cellular and molecular analysis.
Nat Methods 2004; 1:249–254.
26. Zhou Z, Yon Toh S, Chen Z, Guo K, Ng CP, Ponniah S, Lin SC, Hong W,
965
Li P. Cidea-deficient mice have lean phenotype and are resistant to
obesity. Nat Genet 2003; 35:49–56.
27. Mukherjee A, Sidis Y, Mahan A, Raher MJ, Xia Y, Rosen ED, Bloch KD,
Thomas MK, Schneyer AL. FSTL3 deletion reveals roles for TGF-beta
family ligands in glucose and fat homeostasis in adults. Proc Natl Acad Sci
U S A 2007; 104:1348–1353.
28. Xia Y, Sidis Y, Schneyer A. Overexpression of follistatin-like 3 in gonads
causes defects in gonadal development and function in transgenic mice.
Mol Endocrinol 2004; 18:979–994.
29. Zhai Y, Sperkova Z, Napoli JL. Cellular expression of retinal
dehydrogenase types 1 and 2: effects of vitamin A status on testis
mRNA. J Cell Physiol 2001; 186:220–232.
Downloaded from www.biolreprod.org.