MicroRNA-21 regulates the self-renewal of mouse
spermatogonial stem cells
Zhiyv Niua, Shaun M. Goodyeara, Shilpa Raob, Xin Wua, John W. Tobiasb, Mary R. Avarbocka, and
Ralph L. Brinstera,1
aDepartment of Animal Biology, School of Veterinary Medicine, andbPenn Bioinformatics Core, University of Pennsylvania, Philadelphia, PA 19104
Contributed by Ralph L. Brinster, June 21, 2011 (sent for review May 27, 2011)
MicroRNAs (miRs) play a key role in the control of gene expression
in a wide array of tissue systems, where their functions include the
regulation of self-renewal, cellular differentiation, proliferation,
and apoptosis. However, the functional importance of individual
miRs in controlling spermatogonial stem cell (SSC) homeostasis has
not been investigated. Using high-throughput sequencing, we
profiled the expression of miRs in the Thy1+testis cell population,
which is highly enriched for SSCs, and the Thy1−cell population,
composed primarily of testis somatic cells. In addition, we profiled
the global expression of miRs in cultured germ cells, also enriched
for SSCs. Our results demonstrate that miR-21, along with miR-34c,
-182, -183, and -146a, are preferentially expressed in the Thy1+
SSC-enriched population, compared with Thy1−somatic cells. Im-
portantly, we demonstrate that transient inhibition of miR-21 in
SSC-enriched germ cell cultures increased the number of germ cells
undergoing apoptosis and significantly reduced the number of do-
treated cells in recipient mouse testes, indicating that miR-21 is
important in maintaining the SSC population. Moreover, we show
that in SSC-enriched germ cell cultures, miR-21 is regulated by the
transcription factor ETV5, known to be critical for SSC self-renewal.
male germline stem cells|small RNA
composed of Asingle, Apaired, and Aalignedspermatogonia. It is the
Asinglespermatogonia that are considered to have stem cell po-
tential. SSCs are essential for spermatogenesis, but only consti-
tute about 1 in 3,000 cells in the adult mouse testis. Maintenance
of the SSC depends on its capacity for self-renewal, which
encompasses its ability to proliferate, differentiate, and undergo
apoptosis (1, 2). The process of spermatogenesis is complex and
involves numerous endocrine and paracrine signals to coordinate
SSC self-renewal and differentiation of daughter cells to undergo
mitosis, meiosis, and spermiogenesis to generate spermatozoa (1,
2).Recent studies have added a new layerof molecules associated
with the intricate mechanisms of gene regulation, which include
the expression of RNA-induced silencing complex (RISC) com-
ponents as well as a number of microRNAs (miRs), suggesting
that miRs are functionally important in the process of sper-
matogenesis (3, 4). Notably, the loss of the RISC component
Dicer, in germ cells or Sertoli cells, perturbs germ cell devel-
opment and leads to infertility, and highlights the need for miR
function in regulating spermatogenesis (5, 6).
Several studies have reported the global expression of miRs in
the murine testis, but few have examined miR expression in
specific germ cell populations of the testis, particularly the SSC
population (5, 7–10). Moreover, the functional importance of
individual miRs controlling SSC homeostasis has not been in-
vestigated. In the studies presented here, the expression of miRs
in the Thy1+testis cell population, which is highly enriched for
SSCs, and the Thy1−cell population, composed primarily of testis
somatic cells, is profiled. Our results demonstrate that miR-21,
along with miR-34c, -182, -183, -146a, -465a-3p, -465b-3p, -465c-
3p, and -465c-5p, are preferentially expressed in the Thy1+SSC-
permatogonial stem cells (SSCs) are among the testis germ
cells called undifferentiated type A spermatogonia, which are
enriched population, compared with Thy1−testis somatic cells.
Most important, we demonstrate that miR-21 is functionally im-
portant in maintaining the SSC population in vitro, and that the
transcription factor ETV5, known to be critical for SSC self-
renewal, is a direct regulator of miR-21 expression.
Analysis of Small RNAs in Thy1+and Thy1−Testis-Derived Cell
Populations. Using high-throughput sequencing, we analyzed
the small RNA populations in the testis of day 6 wild-type
C57BL/6 mice. Specifically, small RNA libraries were generated
from Thy1+and Thy1−testis cells (referred to as Thy1+SSC-
enriched 1 and 2, and Thy1−somatic cell-enriched 1 and 2, re-
spectively). A small RNA library was also sequenced from an
established germ cell culture enriched for SSCs originally derived
from Thy1+testis cells (referred to as SSC-enriched germ cell
culture). Sequencing analyses generated high-quality raw reads
for each cell population, and these were processed and aligned to
the miRBase database (release 12.0; http://www.mirbase.org) for
detection of miRs (Fig. S1A). Based on annotations from the
miRBase database, we observed that ∼50% of the raw reads
from all five sequence libraries (Thy1+SSC-enriched germ cell 1
and 2, Thy1−somatic cell-enriched 1 and 2, and SSC-enriched
germ cell culture) were identified as known mature miRs, sug-
gesting that miRs are the predominant small RNA species in the
testis cell population from day 6 mice (Fig. S1A). Within the
Thy1+SSC-enriched libraries, 538 and 539 miRs were detected,
and in the Thy1−somatic cell-enriched libraries, 548 and 532
miRs were identified. Sequencing of cultured germ cells enriched
for SSCs detected 512 miRs (Fig. S1A and Table S1). In this
study, only mature miRs were analyzed, and are therefore rep-
resentative of transcriptionally active miRs in each library. To
profile the overall chromosomal distribution of active miRs
detected in the whole testis population, the read counts for each
miR from both Thy1+SSC-enriched and Thy1−somatic cell-
enriched sequence libraries were consolidated and the cloning
frequency (CF) of individual miRs was determined. The CF is
reflective of individual miR abundance within the entire miR
library and, using this parameter, only miRs with a CF >0.01%
were considered. The miRs identified in the pooled population
were mapped to their respective chromosomes using the miR-
Base database, and the total number of transcriptionally active
miRs potentially encoded within each chromosome was enu-
merated. By comparing the total number of actively expressed
miRs encoded within each chromosome with the total number of
Author contributions: Z.N. and R.L.B. designed research; Z.N., X.W., and M.R.A. performed
research; Z.N., S.M.G., S.R., J.W.T., M.R.A., and R.L.B. analyzed data; and Z.N., S.M.G., and
R.L.B. wrote the paper.
The authors declare no conflict of interest.
Data deposition: MicroRNA profiling data reported in this paper have been deposited in
the Gene Expression Omnibus (GEO) database, www.ncbi.nlm.nih.gov/geo (accession no.
1To whom correspondence should be addressed. E-mail: firstname.lastname@example.org.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
www.pnas.org/cgi/doi/10.1073/pnas.1109987108PNAS Early Edition
| 1 of 6
miRs encoded on each chromosome, a percentage of actively
transcribed miRs per chromosome could be plotted (Fig. S1B).
Approximately 30–80% of the total miRs encoded within in-
dividual chromosomes are actively transcribed, and in the com-
bined Thy1+and Thy1−sequence libraries (i.e., representative of
the whole testis population), mature miRs detected and localized
to chromosomes 1, 8, 9, and 19 constituted greater than 70% of
the total miRs possibly encoded within these respective chro-
mosomes. Conversely, less than 40% of the total miRs encoded
within chromosomes 2, 3, 10, 12, and 18 are actively transcribed.
To better assess the differential expression of miRs between
Thy1+and Thy1−testis populations, the sequencing results for
the miR expression level in Thy1+cells were compared with the
differential expression pattern of miRs between Thy1+SSC-
enriched germ cells and Thy1−somatic cell-enriched populations.
The differential miR expression pattern between Thy1+SSC-
enriched and Thy1−somatic cells was plotted using the normal-
ized miR read count number in the Thy1+SSC-enriched library
and plotted against the fold change of miR expression (Fig. 1A).
This approach identified 139 miRs differentially expressed be-
tween Thy1+SSC-enriched and Thy1−somatic cells, of which 84
miRs (highlighted in red) were considered “up-regulated” and 55
miRs (highlighted in blue) were considered “down-regulated” in
the Thy1+SSC-enriched cell population (padj< 0.05). The dif-
ferential expression of miRs in the Thy1+and Thy1−testis cell
populations are described below, and additional characterization
of miRs expressed in the Thy1+SSC-enriched and Thy1−somatic
cell-enriched libraries can be found in SI Results and Discussion.
Characterization of miR Expression in Testis-Derived Thy1−Somatic
Cells. miRs that are both highly abundant and preferentially
enriched in the Thy1−somatic cell-enriched population are
shown in Table S2. Included in this profile are members of the
let-7 family (i.e., let-7a/b/c/d/e/f), which are broadly expressed,
found in multiple somatic cell types, and associated with differ-
entiation in proliferating ES and cancer cells (11, 12). Highly
abundant miRs displaying a significant degree of differential
expression [i.e., n = 55 miRs with log2fold change (Thy1+/
Thy1−) < −1.68; padj< 0.05] were further mapped according to
their individual chromosomal locations, and the sum of the CFs
for miRs on each chromosome was determined (Fig. 1B). The
CF sum provides a simplified means to examine chromosomes
preferentially encoding for actively transcribed miRs. The
chromosomal localization of expressed miRs in the Thy1−so-
matic cell-enriched population was more diverse in its distribu-
tion compared with the abundantly expressed miRs of the Thy1+
SSC-enriched population (Fig. 1B). In the Thy1−somatic cell
population, highly abundant miRs were predominantly localized
to chromosomes X, 2, 9, 13, 15, and 17. The CF sum for miRs
associated with the Thy1−somatic cell population was greatest
on chromosome 13, which expressed the highly abundant miR-
let-7a (Table S2). Examination of chromosome X, which enco-
ded for 32 of the detected miRs in the Thy1−somatic cell pop-
ulation, indicated that the CF sum was ∼6% of the actively
transcribed miRs in the Thy1−somatic cell population (Fig. 1B).
This is in marked contrast to the CF sum of ∼25% for actively
transcribed miRs associated with the Thy1+SSC-enriched testis
cell population (Fig. 1B). Quantitative (q)RT-PCR profiling of
Sum CF per Chromosome
Sum CF per Chromosome
Log10(Thy1+ Read Count)
Log2Fold Change (Thy1+/ Thy1-)
of miRs in the Thy1+SSC-enriched library. A log2calculation was used to normalize the fold change (i.e., Thy1+/Thy1−) value of the read count for individual
miRs (y axis). These values were plotted against the normalized log10read counts of miRs from the Thy1+SSC-enriched library (x axis). Statistical significance of
miRs differentially expressed between the two libraries was determined using R package software analysis (padj< 0.05, where padjrepresents P values after
adjustment for false discovery rate) (40). Red indicates differentially expressed miRs in the Thy1+SSC-enriched cell population; blue indicates differentially
expressed miRs in the Thy1−somatic cell-enriched population. (B) Chromosomal distribution of actively transcribed miRs in Thy1+SSC-enriched and Thy1−
somatic cell-enriched libraries. miRs identified as significantly expressed in A were mapped according to their respective chromosomal locations, and the sum
of the cloning frequencies for these miRs was calculated. The sum of the CF of miRs localized to each chromosome suggests a profile of the chromosomes
contributing to the active transcription of miRs in the Thy1+SSC-enriched and Thy1−somatic cell-enriched libraries. In the cases where mature miRs were
found to be possibly encoded in more than one chromosome, the CF for that miR was divided across the number of possible chromosomal locations (e.g.,
mature miR-let-7f is located on chromosome X or 13; therefore, the CF for miR-let-7f was divided by 2).
Detection of miRs differentially expressed between Thy1+SSC-enriched and Thy1−somatic cell-enriched testis populations. (A) Differential expression
2 of 6
| www.pnas.org/cgi/doi/10.1073/pnas.1109987108Niu et al.
selected miR expression further validated that the expression
levels of miR-let-7a/c/f, -130, and -202-5p were higher in the
Thy1−somatic cells than that of the Thy1+SSC-enriched pop-
ulation or Thy1+SSC-enriched cultured germ cells (Fig. S2).
miR Signature in the Thy1+SSC-Enriched Population. We hypothe-
sized that miRs with a high cloning frequency and/or those with
a high degree of differential expression (i.e., fold change Thy1+/
Thy1−) in the SSCs are likely to play an important role in reg-
ulating self-renewal and maintaining stem cell homeostasis.
Therefore, we examined the 20 most abundant miRs in the
Thy1+SSC-enriched population that also possessed the highest
degree of fold change (Thy1+/Thy1−) and observed that these
miRs account for ∼26% of the total miR transcripts identified in
the Thy1+testis cell population (Table S3). Interestingly, 12 of
these 20 miRs are predominantly expressed on the X chromo-
some, and miR-465a/-465b/-465c and miR-741/-880/-878/-881
are found as clusters on the X chromosome. The chromosomal
distribution of actively transcribed miRs in the Thy1+SSC-
enriched population was again evaluated by determining the CF
sum per chromosome (Fig. 1B). Abundant miRs showing sig-
nificant differential expression within the Thy1+SSC-enriched
population were preferentially localized to chromosomes X, 6,
and 11 (Fig. 1B). Among those miRs showing the greatest fold
change (Thy1+/Thy1−) in the Thy1+SSC-enriched population
were members of the miR-291 family that have been previously
shown to be important in maintaining an undifferentiated state
in ES cells and may possibly have a similar role in maintaining
the SSC population (Table S4) (13, 14).
Characterization of miR Expression in Thy1+SSC-Enriched Germ Cell
Cultures. To better understand the role miRs have in regulating
SSC self-renewal in vitro, we compared the sequence libraries of
freshly isolated Thy1+SSC-enriched testis cells to that of a germ
cell culture established from Thy1+SSC-enriched testis cells and
observed that of the 512 known miRs identified in the cultured
SSC-enriched germ cell population, 68 miRs were expressed at
highly abundant levels (CF > 0.1%). The expression profile of
miRs in cultured SSCs was very similar to the patterns of the two
Thy1+SSC testis cell libraries, with 62 of these 68 miRs observed
as highly abundant in both the Thy1+SSC-enriched libraries and
SSC-enriched germ cell culture (Fig. 2 A and B and Table S6).
Thy1+SSC-enriched germ cell cultures, with the 10 most abun-
dant miRs expressed in cultured SSCs accounting for 67.6% of all
miR molecules, highlighting the possible importance of these
miRs in maintaining stem cell activity in vitro (Table 1). In
comparison with Thy1+SSC-enriched libraries, we observed
ahigher cloning frequencyfor let-7f,miR-34c,and-21in theSSC-
enriched germ cell cultures (Table 1 and Table S1) which may
reflect differences between in vitro and in vivo growth regulation.
Results generated from the sequence analysis were further vali-
dated using qRT-PCR to assess the relative expression of miRs
between Thy1+SSC-enriched, Thy1−somatic cell-enriched, and
SSC-enriched cultured germ cells (Fig. S2). Inagreement with the
sequencing results, the expression of miR-21, -146a, -378, -880-
182, -183, -465a/b/c-3p, and -465c-5p was higher in Thy1+SSC-
enriched and SSC-enriched cultured germ cells compared with
the Thy1−somatic cell-enriched population (Fig. S2). However,
the relative expression of miR-21, -146, -378, -182, -183, -465a-3p,
-465b-3p, -465c-3p, and -465c-5p was dramatically higher in SSC-
enriched cultured germ cells compared with freshly isolated
Thy1+SSC-enriched testis cells. This difference in miR expres-
sion level can likely be attributed to the special characteristics for
propagation and expansion of germ cells in vitro.
ETV5 Directly Regulates miR-21 Expression in Mouse SSCs. The glial
cell line-derived neurotrophic factor (GDNF) pathway plays
a critical role in regulating the self-renewal and maintenance
of the SSC population, both in vivo and in vitro (15–17).
Important downstream effectors of GDNF signaling include
ETV5, BCL6B, POU3f1, and LHX1, all of which are critical to
maintaining stemness (18, 19). Similarly, ETV5 may control miR
expression as one means of regulating SSC homeostasis. Using
a comparative genomic transcription factor binding site analysis
program [TraFac (20)], we observed more than 260 miRs that
possessed ETS binding sites that are highly conserved between
human and mouse, with 8 of the 10 most abundant miRs
expressed in the SSC-enriched germ cell cultures possessing
multiple ETS (E-twenty six) binding sites (Table 1). This ob-
servation suggests a potential role for GDNF-ETV5 signaling in
regulating these and other miRs. In particular, miR-21 has been
shown to be an important antiapoptotic factor that greatly
enhances tumor progression, and additional studies have shown
that miR-21 can prevent apoptosis in mouse periovulatory
granulosa cells (21, 22). The multiple conserved ETS-binding
motifs identified on the miR-21 promoter region (Fig. 3A and
Table 1), along with the high levels of miR-21 expression ob-
served in SSC-enriched germ cell cultures (Fig. S2), strongly
suggest an important in vitro function. To verify possible ETV5
regulation of miR-21, chromatin immunoprecipitation (ChIP)
using mouse germ cell cultures was performed. PCR amplifica-
tion using primers flanking two of the predicted ETS-binding
motifs each produced a band for DNA coprecipitated with ETV5
and ETV5 antibody but not in the isotype antibody controls (Fig.
3B). These results indicate that ETV5 binds to the miR-21 en-
hancer. To further investigate the putative regulation of miR-21
expression by ETV5, germ cell cultures were infected with
a lentivirus vector expressing mouse ETV5 cDNA under the
control of the EF1α promoter. In transduced germ cells, Etv5
expression was significantly elevated to greater than twofold
compared with controls, and this was accompanied by a signifi-
cant, 1.78-fold, increase in the expression of miR-21 (Fig. 3C).
These results strongly suggest that ETV5 has a role in regulating
miR-21 expression in SSC-enriched germ cells. We also observed
putative STAT3 and POU3f1 binding sites within the miR-21
enhancer region (Fig. 3A). Previous reports have demonstrated
that STAT3 promotes differentiation of SSCs (23), whereas
enriched germ cell cultures
Top 10 most abundantly expressed miRs in SSC-
# ETS binding
sites in 5kb
2 ; 6
2 ; 1
2 ; 5
The cloning frequency was used to rank the most abundantly expressed
mature miRs in the SSC-enriched germ cell culture library. The cloning fre-
quency was calculated as the percent abundance of the read counts for
individual miRs over the sum of all read counts in the SSC-enriched germ
cell culture library. Promoter software (TraFac) analysis found that 13 of the
20 miRs present possessed one or more putative ETS binding sites within the
enhancer region of chromosome encoding for respective miRs. Profiling of
ETS binding sites was performed to incorporate 5kb upstream of the mature
miR sequence. In some cases, miRs are encoded on more than one chromo-
some and a semi-colon (;) is used to separate the number of putative ETS
binding sites on each chromosome.
Niu et al. PNAS Early Edition
| 3 of 6
POU3f1, a GDNF-regulated transcription factor, is shown to be
critical in regulating the SSC population in vitro (24), suggesting
that several transcription factors may be involved in the regula-
tion of miR-21 expression.
Transient Inhibition of miR-21 Increases Germ Cell Apoptosis and
Affects Stem Cell Activity. Because the above findings suggested
that ETV5 in part regulates miR-21, and miR-21 expression
accounts for ∼11% of the miR molecules in SSC-enriched cul-
tured germ cells, the biological function of miR-21 in regulating
SSC homeostasis in vitro was investigated. SSC-enriched germ
cell cultures were transiently transfected with a prevalidated
miR-21 inhibitor (anti–miR-21) oligonucleotide (oligos), and its
effect on germ cell expansion and stem cell activity was evalu-
ated. The doubling time for mouse SSCs is ∼5.6 d (25); for that
reason, following exposure to anti–miR-21 oligos or nontargeting
control oligos, germ cells were plated onto new feeders and
cultured for an additional 7 d posttransfection before being
collected and counted. The transient inhibition of miR-21 sig-
nificantly decreased the number of germ cells in vitro compared
with nontargeting controls (0.98 ± 0.14 × 105cells vs. 1.56 ± 0.21
× 105cells, respectively; Fig. 4A). The role of miR-21 as an
antiapoptotic factor has been reported for several systems (21,
22), and therefore Annexin V staining was used to assess the
level of apoptosis caused by the inhibition of miR-21. At 20 h
posttransfection, the apoptotic index for germ cells treated with
Lipofectamine alone or nontargeting control oligos was 3.87 ±
1.06% and 3.67 ± 0.34%, respectively (Fig. 4B). In comparison,
inhibition of miR-21 significantly increased apoptosis in germ
cell cultures to 7.07 ± 0.57% (Fig. 4B).
The in vitro impact of miR-21 inhibition on SSC survival and
proliferation was evaluated with an in vivo transplantation assay.
Transplantation of donor germ cells into the testes of busulfan-
treated recipient mice is the only functional assay to quantify the
number of SSCs in any cell population, and provides an un-
equivocal means to determine whether a reduction in the num-
ber of SSCs occurred as a result of varying treatments to the
in vitro SSC-enriched germ cell culture (19, 26). Following 7 d in
culture posttransfection with anti–miR-21 oligos or nontargeting
control oligos, germ cells were transplanted into recipient testis.
The transient inhibition of miR-21 decreased the number of
donor colonies formed from 178 ± 20.9 per 105germ cells in the
nontargeting oligo control group to 108 ± 23.2 per 105germ cells
in anti–miR-21-treated germ cell cultures (Fig. 4C; P < 0.05).
These results demonstrate a requirement for miR-21 expression
and function in the in vitro maintenance of the mouse SSC
population, and suggest that one mechanism of miR-21 effect is
through the regulation of apoptosis.
Using high-throughput sequencing, we identified an miR signa-
ture that was common to Thy1+SSC-enriched testis cells and
germ cell cultures, and a high degree of similarity was observed
between the two libraries. Notably, some cloning frequencies of
SSC-associated miRs were dramatically higher in SSC-enriched
germ cell cultures compared with Thy1+testis cells. The higher
miR expression in vitro may reflect a more active role of miRs in
regulating SSC self-renewal and/or early differentiation steps
in vitro. However, an important observation is that in the ab-
sence of the physiological in vivo niche influences, the cultured
SSCs and undifferentiated type A spermatogonia maintain a re-
markably similar miR profile to the Thy1+SSC-enriched testis
cells isolated directly from the testes. These similarities in miR
expression suggest that regulation of SSC self-renewal and the
undifferentiated type A spermatogonia are primarily regulated
by the germ cell program rather than by external cues and may
require mostly a permissive environment, such as the in vivo
niche and an appropriate culture milieu. A similar conclusion
regarding germ cell program dominance versus environmental
cues is suggested by the finding that rat spermatogenesis is
supported in vivo by mouse Sertoli cells at the exact timing and
organizational arrangement of the rat and not mouse (27).
Therefore, the mouse Sertoli cell environment cannot change
the length of the rat cycle (52 d) in the seminiferous epithelium
to that of the mouse cycle [35 d (27, 28)]. These findings support
the use of cultured germ cells, enriched for SSCs, as a powerful
model to examine the expression and function of miRs and genes
in this germ cell population, which contains almost exclusively
undifferentiated spermatogonia that are uniform in morpho-
logical characteristics and a portion of which may convert to
SSCs under certain conditions (29, 30).
The transcription factor ETV5 is one of the downstream tar-
gets of GDNF signaling, and is essential for SSC self-renewal
(18, 31–33). In the testis, ETV5 is expressed in Sertoli cells and
in the SSC-enriched cell populations (18, 31, 34). In Etv5−/−
mice, Sertoli cell function is compromised, resulting in the
complete loss of adult germ cells following the first wave of
spermatogenesis (31). We examined putative ETS binding sites
within the enhancer/promoter regions of highly abundant miRs
and found that 8 of 10 of the most abundant miRs in the Thy1+
SSC-enriched Germ Cell Culture
Cloning Frequency of all highly expressed miRs identified in SSC-enriched germ cell
culture and Thy1+SSC-enriched libraries.
(107 miR >0.1%)
(106 miR >0.1%)
SSC-enriched Germ Cell Culture
(68 miR >0.1%)
cultures and Thy1+SSC-enriched testis
cells show similarity in miR expression.
(A) Comparison of relative miR abun-
dance in sequence libraries derived from
cultured germ cells from Thy1+SSC-
enriched testis cells and freshly isolated
Thy1+SSC-enriched testis cells (red,
Thy1+SSC-enriched library 1; blue, Thy1+
SSC-enriched library 2). The CF for in-
dividual miRs from the two Thy1+SSC-
enriched testis libraries was plotted
against the CF of corresponding miRs in
the germ cell culture library. (B) Venn
diagram illustrating overlap of miR ex-
pression between freshly isolated Thy1+
SSC-enriched germ cells and cultured
germ cells. Only miR expression in which
the CF >0.1% is included.
Thy1+SSC-enriched germ cell
4 of 6
| www.pnas.org/cgi/doi/10.1073/pnas.1109987108Niu et al.
SSC-enriched population contained ETS binding sites in their
enhancer regions, including miR-21. We observed that ETV5 is
capable of regulating miR-21 expression and, because ETV5 is
a known downstream effecter of GDNF signaling, it is likely that
GDNF has a role in the regulation of miRs. This is supported by
a report demonstrating that miR-21 expression is induced by
GDNF in BE(2)-C neuronal cells (35). Importantly, we observe
that inhibition of miR-21 increased apoptosis in SSC-enriched
germ cell cultures, suggesting it has a role in maintaining SSC
survival. Because apoptosis is a major regulator of spermato-
genesis (36), miR-21 may play a pivotal role in the early stages of
spermatogenesis by regulating apoptosis in the undifferentiated
spermatogonial cell population, including SSCs.
The stringency of the profiling used to analyze results from high-
throughput sequencing for both the Thy1+SSC-enriched cell
population and SSC-enriched germ cell cultures strongly suggests
that the in vitro culture system provides an ideal model to func-
tionally examine the role of miRs in regulating SSC fate decisions.
sites within the miR-21 enhancer region. (B) ChIP of ETV5 binding to the miR-21 enhancer. ChIP #1 and ChIP #2 represent replicate experiments for two
enhancer regions located 236 and 330 bp upstream of the miR-21 transcriptional start site. (C) Increased expression of ETV5 from 0.99 ± 0.001 to 2.63 ± 0.36
following transduction of germ cell cultures with a lentiviral (pWPI) construct constitutively expressing ETV5 cDNA. Compared with empty vector controls
(1.00 ± 0.02), the overexpression of ETV5 resulted in a significant increase in miR-21 expression (1.78 ± 0.27). The asterisk denotes significance where P < 0.05
(mean ± SEM, n = 5).
ETV5 directly regulates miR-21 expression in Thy1+SSC-enriched germ cell cultures. (A) ETV5, as well as POU3f1 and STAT3, possess multiple binding
Germ Cell Culture
Cell Number (x 105)
Colony No. / 105cells cultured
In Vivo Colony Formation
transiently transfected with anti–miR-21 oligonucleotides or nontargeting control oligo. After 7 d of being maintained in culture posttransfection, the total
number of germ cells in each treatment was counted. The number of anti–miR-21-treated germ cells was significantly reduced to 0.98 × 105± 0.14 cells
compared with germ cells treated with nontargeting control oligos (1.57 × 105± 0.21 cells). (B) Transient inhibition of miR-21 promotes apoptosis in germ
cell cultures. Following 20 h posttransfection with anti–miR-21 oligos, nontargeting control oligos, or Lipofectamine alone, germ cell cultures were collected,
washed, and incubated with Annexin V antibody and 7-aminoactinomycin D reagent. The apoptosis index was determined by comparing the number of ap-
optotic germ cells with the total number of germ cells. Compared with germ cells treated with Lipofectamine alone (3.87 ± 1.06%) or nontargeting control oligos
(3.67 ± 0.34%), the number of germ cells undergoing apoptosis was significantly increased by treatment with anti–miR-21 oligos (7.07 ± 0.57%). (C) Inhibition of
miR-21 activity decreases the in vivo colony formation ability of treated germ cells. The average number of colonies formed in recipient testes from 105cells
transplanted to recipient testes was determined for germ cell cultures treated with nontargeting control oligos or anti–miR-21 oligos. Treated cells were
maintained for 7 d posttransfection before being transplanted into the testes of recipient mice. Two months after transplantation, the number of donor-derived
colonies was counted. Inhibition of miR-21 activity caused the number of donor-derived colonies to significantly decrease from 178 ± 20.9 colonies for control
nontargeting oligo-treated germ cells to 108 ± 23.2 colonies for anti–miR-21-treated germ cells. All data are representative of three independent replicate
cultures (mean ± SEM), and for transplantation studies this resulted in 16 testes (n = 8 mice) per treatment. The asterisk denotes significant differences between
treatment means using Student’s t test (P < 0.05).
Biological importance of miR-21 in Thy1+SSC-enriched germ cell cultures. (A) At the start of the experiment, 1.0 × 105germ cells in culture were
Niu et al.PNAS Early Edition
| 5 of 6
Importantly, significant advances in understanding the cellular and Download full-text
molecular mechanisms regulating SSC self-renewal have been
made using culture systems that are able to support proliferation of
SSC-enriched germ cells in vitro (25), and a robust transplantation
assay that is capable of determining the biological activity of donor-
derived SSCs in the seminiferous tubules of recipient testis pro-
vides the ultimate test of various treatment effects (37, 38). The
combination of these two systems provides a method of analyzing
the biological activity of SSCs by unequivocally measuring SSC
number, and in this report has provided the foundation to dem-
onstrate the validity of using germ cell cultures to examine the
function of miRs in our profile and the means to demonstrate the
specific effect of miR-21 on SSC self-renewal.
Materials and Methods
Isolation of SSC-Enriched Germ Cells. Testes of day 6 postnatal mice were
harvested from two strains: C57BL/6 mice (Jackson Laboratory stock no.
000664) or Escherichia coli β-galactosidase (LacZ)-expressing mice (B6.129S7-
Gtrosa26, designated ROSA; Jackson Laboratory stock no. 002073). miR
profiling studies were conducted using the testes derived from C57BL/6
mouse pups, whereas the testes of ROSA mouse pups, expressing the LacZ
gene, were used to derive germ cell cultures for transplantation assays (see
below). Digestion of testis and isolation of Thy1+and Thy1−testis cell pop-
ulations were carried out as previously described (26, 39). Thy1+SSC-
enriched germ cells and Thy1−somatic cell-enriched testis cells were further
processed for isolation of small RNAs or germ cell culture. Establishment of
germ cell cultures was carried out using Thy1+SSC-enriched germ cells as
previously described (26, 39). A detailed method of maintaining SSC-
enriched germ cell cultures can be found in SI Materials and Methods. All
animal protocols were approved by the Institutional Animal Care and Use
Committee of the University of Pennsylvania.
Small RNA Library Construction and Sequencing. Small RNA libraries from
Thy1+SSC-enriched and Thy1−somatic cell-enriched testis cell populations or
germ cell cultures were isolated using TRIzol (Invitrogen) and purified using
the Illumina manufacturer protocols “Preparing Samples for Analysis of
Small RNA” and “Preparing Samples for Small RNA Sequencing Using the
Alternative v1.5 Protocol.” The final libraries of small RNAs for each sample
were gel-purified and measured using an Agilent Bioanalyzer. Sequencing
of small RNA libraries was conducted in the Penn Microarray Facility using an
Illumina Genome Analyzer. Bioinformatics analysis and quantitative real-
time PCR validation are described in SI Materials and Methods.
Additional Methods. Detailed descriptions of methods for functional analysis
of miR-21 in germ cell cultures are available in SI Materials and Methods.
ACKNOWLEDGMENTS. We thank Drs. R. Behringer and M. Kotlikoff for
critical evaluation of the manuscript, and C. Freeman and R. Naroznowski for
assistance with animal maintenance. This study was supported by National
Institute of Child Health and Human Development Grant HD 052728 (to R.L.B.)
and the Robert J. Kleberg, Jr. and Helen C. Kleberg Foundation (R.L.B.).
1. Oatley JM, Brinster RL (2008) Regulation of spermatogonial stem cell self-renewal in
mammals. Annu Rev Cell Dev Biol 24:263–286.
2. Brinster RL (2007) Male germline stem cells: From mice to men. Science 316:404–405.
3. González-González E, López-Casas PP, Del Mazo J (2008) Gene silencing by RNAi in
mouse Sertoli cells. Reprod Biol Endocrinol 6:29.
4. Kotaja N, et al. (2006) The chromatoid body of male germ cells: Similarity with pro-
cessing bodies and presence of Dicer and microRNA pathway components. Proc Natl
Acad Sci USA 103:2647–2652.
5. Hayashi K, et al. (2008) MicroRNA biogenesis is required for mouse primordial germ
cell development and spermatogenesis. PLoS One 3:e1738.
6. Papaioannou MD, Nef S (2010) microRNAs in the testis: Building up male fertility.
J Androl 31:26–33.
7. Buchold GM, et al. (2010) Analysis of microRNA expression in the prepubertal testis.
PLoS One 5:e15317.
8. Jung YH, Gupta MK, Shin JY, Uhm SJ, Lee HT (2010) MicroRNA signature in testes-
derived male germ-line stem cells. Mol Hum Reprod 16:804–810.
9. Ro S, Park C, Sanders KM, McCarrey JR, Yan W (2007) Cloning and expression profiling
of testis-expressed microRNAs. Dev Biol 311:592–602.
10. Yan N, et al. (2007) A microarray for microRNA profiling in mouse testis tissues. Re-
11. Büssing I, Slack FJ, Grosshans H (2008) let-7 microRNAs in development, stem cells and
cancer. Trends Mol Med 14:400–409.
12. Roush S, Slack FJ (2008) The let-7 family of microRNAs. Trends Cell Biol 18:505–516.
13. Lichner Z, et al. (2011) The miR-290-295 cluster promotes pluripotency maintenance
by regulating cell cycle phase distribution in mouse embryonic stem cells. Differen-
14. Zovoilis A, et al. (2010) Embryonic stem cell-related miRNAs are involved in differ-
entiation of pluripotent cells originating from the germ line. Mol Hum Reprod 16:
15. Kubota H, Avarbock MR, Brinster RL (2004) Culture conditions and single growth
factors affect fate determination of mouse spermatogonial stem cells. Biol Reprod 71:
16. Meng X, et al. (2000) Regulation of cell fate decision of undifferentiated spermato-
gonia by GDNF. Science 287:1489–1493.
17. Ryu B-Y, Kubota H, Avarbock MR, Brinster RL (2005) Conservation of spermatogonial
stem cell self-renewal signaling between mouse and rat. Proc Natl Acad Sci USA 102:
18. Oatley JM, Avarbock MR, Brinster RL (2007) Glial cell line-derived neurotrophic factor
regulation of genes essential for self-renewal of mouse spermatogonial stem cells is
dependent on Src family kinase signaling. J Biol Chem 282:25842–25851.
19. Oatley JM, Avarbock MR, Telaranta AI, Fearon DT, Brinster RL (2006) Identifying
genes important for spermatogonial stem cell self-renewal and survival. Proc Natl
Acad Sci USA 103:9524–9529.
20. Jegga AG, et al. (2002) Detection and visualization of compositionally similar cis-
regulatory element clusters in orthologous and coordinately controlled genes. Ge-
nome Res 12:1408–1417.
21. Carletti MZ, Fiedler SD, Christenson LK (2010) MicroRNA 21 blocks apoptosis in mouse
periovulatory granulosa cells. Biol Reprod 83:286–295.
22. Krichevsky AM, Gabriely G (2009) miR-21: A small multi-faceted RNA. J Cell Mol Med
23. Oatley JM, Kaucher AV, Avarbock MR, Brinster RL (2010) Regulation of mouse sper-
matogonial stem cell differentiation by STAT3 signaling. Biol Reprod 83:427–433.
24. Wu X, et al. (2010) The POU domain transcription factor POU3F1 is an important
intrinsic regulator of GDNF-induced survival and self-renewal of mouse spermato-
gonial stem cells. Biol Reprod 82:1103–1111.
25. Kubota H, Avarbock MR, Brinster RL (2004) Growth factors essential for self-renewal
and expansion of mouse spermatogonial stem cells. Proc Natl Acad Sci USA 101:
26. Oatley JM, Brinster RL (2006) Spermatogonial stem cells. Methods Enzymol 419:
27. França LR, Ogawa T, Avarbock MR, Brinster RL, Russell LD (1998) Germ cell genotype
controls cell cycle during spermatogenesis in the rat. Biol Reprod 59:1371–1377.
28. Russell LD (1993) in The Sertoli Cell, eds Russell LD, Griswold MD (Cache River,
Clearwater, FL), pp 365–390.
29. Kubota H, Avarbock MR, Brinster RL (2003) Spermatogonial stem cells share some, but
not all, phenotypic and functional characteristics with other stem cells. Proc Natl Acad
Sci USA 100:6487–6492.
30. Nakagawa T, Sharma M, Nabeshima Y-i, Braun RE, Yoshida S (2010) Functional hi-
erarchy and reversibility within the murine spermatogenic stem cell compartment.
31. Chen C, et al. (2005) ERM is required for transcriptional control of the spermatogonial
stem cell niche. Nature 436:1030–1034.
32. Lu BC, et al. (2009) Etv4 and Etv5 are required downstream of GDNF and Ret for
kidney branching morphogenesis. Nat Genet 41:1295–1302.
33. Tyagi G, et al. (2009) Loss of Etv5 decreases proliferation and RET levels in neonatal
mouse testicular germ cells and causes an abnormal first wave of spermatogenesis.
Biol Reprod 81:258–266.
34. Yoon K-A, Chae Y-M, Cho J-Y (2009) FGF2 stimulates SDF-1 expression through the
Erm transcription factor in Sertoli cells. J Cell Physiol 220:245–256.
35. Yoong LF, Wan G, Too H-P (2006) Glial cell-line derived neurotrophic factor and
neurturin regulate the expressions of distinct miRNA precursors through the activa-
tion of GFRα2. J Neurochem 98:1149–1158.
36. Shaha C, Tripathi R, Mishra DP (2010) Male germ cell apoptosis: Regulation and bi-
ology. Philos Trans R Soc Lond B Biol Sci 365:1501–1515.
37. Brinster RL, Avarbock MR (1994) Germline transmission of donor haplotype following
spermatogonial transplantation. Proc Natl Acad Sci USA 91:11303–11307.
38. Brinster RL, Zimmermann JW (1994) Spermatogenesis following male germ-cell
transplantation. Proc Natl Acad Sci USA 91:11298–11302.
39. Kubota H, Brinster RL (2008) Culture of rodent spermatogonial stem cells, male
germline stem cells of the postnatal animal. Methods Cell Biol 86:59–84.
40. Anders S, Huber W (2010) Differential expression analysis for sequence count data.
Genome Biol 11:R106.
6 of 6
| www.pnas.org/cgi/doi/10.1073/pnas.1109987108 Niu et al.