| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
,1
,||,
* Department of Biochemistry,
Graduate Institute of Medical Sciences, School of Medicine, and
Center for Reproductive Medicine, Taipei Medical University Hospital, Taipei Medical University, Taipei, Taiwan;
Division of Reproductive Endocrinology and Infertility, Department of Obstetrics and Gynecology, College of Medicine and Hospital, National Taiwan University, Taipei, Taiwan;
|| Graduate Institute of Immunology and
¶ Institute of Pharmacology, College of Medicine, National Taiwan University, Taipei, Taiwan;
# National Laboratory Animal Center, National Applied Research Laboratories, Taipei, Taiwan;
** Stem Cell Program, Genomic Research Center, and
Institute of Cellular and Organismic Biology, Academia Sinica, Taipei, Taiwan; and
Graduate Institute of Life Sciences, National Defense Medical Center, Taipei, Taiwan
1 Correspondence: Y.-H.H., Department of Biochemistry, Graduate Institute of Medical Sciences, School of Medicine, Taipei Medical University, Taipei, Taiwan. E-mail: rita1204{at}tmu.edu.tw; T.-Y.L., Institute of Pharmacology, College of Medicine, National Taiwan University, Taipei, Taiwan. E-mail: tyling{at}ntu.edu.tw
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: germline stem cells • stem cell-microenvironment interaction • Leydig cells • serum-free culture • cell signaling
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. In addition to postfertilization development, the pluripotency of GSCs has been shown by their ability to form teratomas in testis or ovary (2)
. Recent studies (3)
have demonstrated that p53-knockout neonatal SSCs respond to culture conditions and acquire pluripotency. However, the lack of understanding regarding the regulatory mechanisms of pluripotency often results in an inefficient generation of pluripotent stem cells from wild-type neonatal mouse testis (3)
. The use of serum-containing medium for the generation of multipotent adult germline stem cells (maGSCs) from testis of Stra8-GFP transgenic mice also limits the possibility of identifying specific endocrine factors that mediate the pluripotency of SSCs (4)
.
Undefined components in certain batches of serum (5)
may limit the ability of SSCs to acquire pluripotency or may affect the maintenance of stem cell pluripotency. To circumvent the serum factors, there have been several attempts to cultivate mouse SSCs in vitro in serum-free conditions with mouse embryonic fibroblast feeders (6
7
8
9
10)
. However, in these studies, utilizing these nonphysiologically relevant feeders resulted in loss of spermatogonial potency in culture (8)
and selective outgrowth of differentiated sperm cells (A4–A32 alignment; ref. 10
). Therefore, the medium was supplemented with selected exogenous cytokines to support the self-renewal growth of GSCs. With the use of this approach, glial cell line-derived neurotropic factor (GDNF) was the first cytokine demonstrated to be directly involved in regulating the self-renewal of mouse SSCs (8
, 11)
. GDNF was able to activate PI3K/Akt and Src kinase-mediated signaling pathways, which led to the up-regulation of the self-renewal-associated transcription factors Bcl6b, Erm, and Lhx1 (12
, 13)
. However, the genes that regulate pluripotency or self-renewal like Oct-4 and Plzf were not affected by GDNF treatment (13)
. These observations suggest that factors involved in mediating pluripotency of SSCs have yet to be identified.
In the present study, we established a serum-free culture system using testicular stromal cells as feeders to clonally derive pluripotent SSCs from wild-type neonatal mouse testes. These cells showed strong alkaline phosphatase (AP) activity and displayed characteristics of embryonic stem cells (ESCs) in which they were able to differentiate into neuron-like cells, hepatocyte-like cells, and c-kit+ germ cells in vitro and to form embryonic chimeras after injection into blastocysts. AP+GSCs could also contribute to spermatogenesis after transplantation to recipient testes pretreated with bulsufan. Most importantly, in addition to the role of GDNF in SSC self-renewal, we further identified a paracrine factor, insulin-like growth factor-1 (IGF-1), which was secreted by Leydig cells, as a key factor that regulated the self-renewal growth and pluripotency of SSCs. These findings led to exploration of germ cell pluripotency.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1.5 x 105 cells. Total testicular cells were resuspended in the basic culture medium (BM) composed of MCDB-201 supplemented with 1x ITS (insulin, transferrin, selenium) and 10 ng/ml of epidermal growth factor (EGF; Gibco-BRL). Testicular cells were seeded at a density of 8 x 104 cells/cm2 in a laminin-coated culture plate, and cultivated in an incubator at 37°C, 5% CO2 for 7 d.
AP activity assay
Clump GSC colonies in serum-free medium were fixed with 3.7% paraformaldehyde for 30 min at room temperature. The AP activity of these GSC colonies was measured with an AP detection kit according to the manufacturer’s instructions (Chemicon, Chandlers Ford, UK).
RNA isolation and RT-PCR
The total RNA of AP+GSC colonies was extracted with the RNeasy Micro kit (Qiagen, Valencia, CA, USA), according to the manufacturer’s instructions. Three micrograms of total RNA was used to synthesize cDNA with random primer (Invitrogen, Carlsbad, CA, USA). The cDNA synthesis was performed at 50°C for 50 min in a final volume of 20 µl, according to the manufacturer’s instructions for Superscript III reverse transcriptase (Invitrogen). The PCR was conducted with PlatinumTaq polymerase (Invitrogen), and the RT-PCR amplifications were titrated to be within a linear range of amplification. Primer sequences and annealing temperature are listed in Supplemental Table 1. Gapdh mRNA was used as an internal control. PCR products were separated by agarose gel electrophoresis, and the DNA bands were visualized with ethidium bromide under ultraviolet light. The RT-PCR analysis of at least three independent cultures was performed for all experiments.
Immunostaining
For detection of Oct-4 expression in primary culture cells, the cells were fixed in methanol:acetone (1:1) at room temperature for 10 min. For other antigens, the cells were fixed with 4% paraformaldehyde at room temperature for 30 min. After fixation, the cells were rinsed with PBS twice and then treated with PBS containing 0.05% Tween-20 (PBST) at room temperature for 10 min and blocked with BSA (5 mg/ml) in PBST for 1 h at room temperature. The cells were then incubated at 4°C overnight with the following antibodies: anti-Oct-4 (sc-9081; Santa Cruz Biotechnology, Santa Cruz, CA, USA), anti-CD29 (clone 9EG7), anti-CD49f (clone GoH3), anti-CD31 (clone MEC13.3), anti-CD34 (clone RAM34; all from BD Biosciences, San Jose, CA, USA), anti-c-kit (CD117, Clone YB5.B8, BD Biosciences), anti-SSEA-1 (clone MC-480; Chemicon), anti-IR
(sc-710) and anti-IGF-1R (sc-7952) (Santa Cruz Biotechnology), anti--SMA (A-2547, Sigma, for myoid cells), anti-CYP11A1 (sc-18043, Santa Cruz Biotechnology, for Leydig cells), anti-MIS (sc-6886, Santa Cruz Biotechnology, for Sertoli cells), and anti-IGF-1 (sc-9013, Santa Cruz Biotechnology). The nuclei of all cells were counterstained with DAPI (Sigma). All cells were covered with an antifading reagent (Vector Laboratories, Burlingame, CA, USA) and analyzed with a fluorescence microscope (Olympus, Melville, NY, USA). For confocal spectroscopic fluorescence, testicular cells were seeded and grew on the cover glasses for 7 d. The forming AP+GSC colonies were fixed, blocked, and immunoprobed with specific antibodies. All cells were covered with antifading reagent and analyzed with a confocal laser scanning microscope (Leica, Hercules, CA, USA). Neural-lineage cell types, hepatocytes, and germ cell differentiation of AP+GSCs were demonstrated using immunostaining. The differentiated cells were fixed in 4% paraformaldehyde for 30 min at room temperature, and then they were incubated with 5% normal serum for 1 h. After the blocking process, cells were incubated with primary antibody at 4°C overnight. Antibodies used were: anti-MAP2 (for neuron, clone 5F9; UBI, Lake Placid, NY, USA), anti-O4 (for oligodendrocytes, MAB345; Chemicon), anti-glial fibrillary acidic protein (for astrocyte, GFAP, clone G-A-5; Sigma), anti--fetoprotein (anti-AFP, for hepatocytes, A0008; DakoCytomation, Carpinteria, CA, USA), or anti-c-kit (CD117, clone YB5.B8; BD Biosciences). Specific labeling for the primary antibodies was detected with Cy3 and/or FITC-conjugated secondary antibodies (Jackson ImmunoResearch, West Grove, PA, USA). The AP+GSCs were mounted on slides using Vectashield mounting medium (Vector Laboratories), and the staining was examined using epifluorescence microscopes (Olympus).
For immunohistochemical staining, the cyrosection slides of d 18 embryonic chimeras were probed with anti-GFP antibodies (Invitrogen) for microscope analysis or colocalized with CD31 (clone MEC13.3, BD Biosciences), cytokeratin 14 (clone LL002; Novocastra, Newcastle, UK), and AFP (DakoCytomation) by confocal laser scanning microscope (Leica).
Differentiation and proliferation of the AP+GSCs in vitro
For neuron-lineage differentiation, the AP+GSC colonies were transferred to gelatin (1 mg/ml)-coated plates and treated with trans-retinoic acid (Sigma, 5 µM) combined with 10% FBS-containing BM for 2 wk. For hepatocyte induction, the AP+GSC colonies were digested into single cells and treated with acidic fibroblast growth factor (aFGF; 20 ng/ml) and basic fibroblast growth factor (bFGF; 10 ng/ml; both from Peprotech, Rocky Hill, NJ, USA) in 15% FBS-IMDM, 2 mM L-glutamine (Sigma), 300 µM monothioglycerol (Sigma) for 2 d, followed by HGF (10 ng/ml; R&D Systems, Minneapolis, MN, USA) treatment for another 2 d. The cells were then incubated with Oncostanin M (10 ng/ml), dexamethasone (100 nM; both from R&D Systems), and 1x ITS supplement (Invitrogene) for the next 2 d (14)
. For germ cell differentiation, the AP+GSC colonies were transferred to an uncoated culture plate and incubated with BM containing 10% FBS and stem cell factor (10 ng/ml; Peprotech) for 10 d. To test the PGC potential of retinoic acid-stimulated cell proliferation, the AP+GSC colonies were digested by collagenase and reseeded on the BM-conditioned methylcellulose soft agars. The medium was supplemented with trans-retinoic acid in concentrations of 0, 0.5, and 1 µM for 10 d to stimulate the proliferation of AP+GSCs. The colonies were counted for statistical analysis.
Testis transplantation and chimera formation
Functional transplantation of the EGFP-AP+GSCs to recipient testis was performed according to a previous report (15)
. For chimera formation, 5 to 10 EGFP-AP+GSCs were injected into the blastocoels of 3.5 days postcoitum (dpc) blastocysts from C57BL/6 mice using a piezo-driven micromanipulator (Prime Tech, Tsuchiura, Japan). The blastocysts were returned to the uteri of 2.5 dpc pseudo-pregnant CByB6F1 foster mothers on the day of microinjection. The d 18 embryonic chimeras were fixed in 4% paraformaldehyde and frozen in Tissue-Tek OCT compound (Cryochrome; Shandon, Pittsburgh, PA, USA) for cyrosection and anti-GFP immunostaining. Newborn d 4 chimeras were subjected to noninvasive in vivo live fluorescence imaging by image visualization and infrared spectroscopy (IVIS Imaging System, 200 Series; Caliper Life Science, Xenogen, Alameda, CA, USA). Three-week-old chimeras were killed to get organ tissues. The tissues were fixed, embedded, and sectioned for GFP immunostaining.
Effect of exogenous factors on AP+GSC colony formation
Exogenous factors, such as EGF, insulin, IGF-1, selenium, and transferrin, as well as anti-insulin receptor antibodies (IR, sc-710; Santa Cruz Biotechnology), anti-IGF-1 receptor antibodies (IR3; Calbiochem, Darmstadt, Germany), rabbit and mouse IgG (Jackson ImmunoResearch), PPP (cyclolignan picropodophyllin; Calbiochem), and LY294002 (Cell Signaling, Danvers, MA, USA), were added to the medium, and the GSCs were cultivated at 37°C, 5% CO2 for 7 d. After 7 d of cultivation, the AP activity of GSC colonies under different experimental condition was determined. The AP+GSC colonies were counted for statistical analysis.
Cytokine array analysis
The endocrine factors in the testicular-stromal niche coculture system were examined by mouse cytokine antibody array analysis (AAM-CYT-4–4; RayBiotech, Norcross, GA, USA). Briefly, the array membranes were incubated in blocking buffer at room temperature for 30 min and then incubated with the collected MCDB201 and d 3-conditioned medium (D3-CM) at room temperature for 1–2 h. After incubation, the array membrane was washed at room temperature with gentle shaking. The membranes were then incubated with primary biotin-conjugated antibodies at room temperature for 1–2 h, washed several times, and incubated with horseradish perioxidase (HRP)-conjugated streptavidin for 2 h at room temperature. The array membranes were then washed and exposed to X-ray films.
Western blotting
NCCIT cells and AP+GSC colonies under different culture conditions were collected and lysed in reducing 2x Laemmli sample buffer. The cell lysates were subjected to 10% SDS-PAGE and then transferred to a PVDF membrane for Western blot analysis. Monoclonal anti-phosphoAkt (Ser-473, anti-Akt, Cell Signaling, 1:1000) and polyclonal anti-Oct-4 antibodies (Santa Cruz Biotechnology, 1:1000) were used as the primary antibody, and HRP-conjugated anti-mouse/rabbit IgG (1:2000) served as the secondary antibody. The enzyme activity of HRP was detected by the ECL system according to the manufacturer’s instructions (Amersham Pharmacia Biotechnology, Little Chalfont, UK).
Statistical analysis
All experiments were repeated at least 3 times with different individual samples. Data are expressed as mean ± SD. Difference in means was assessed by one-way ANOVA, followed by the Tukey-Kramer multiple comparisons test.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, the GSC colonies were tightly packed in clump morphology, similar to PGCs and embryonic germ cells (Fig. 1Aai
; also see Supplemental Fig. 1 and ref. 16
). Unlike SSCs, which had weak AP activity (3
, 8)
, these GSC colonies showed strong AP activity similar to the PGC activity (Fig. 1Aaii
). We tentatively designated these GSCs as AP+GSCs. RT-PCR demonstrated that the AP+GSC colonies expressed the Oct-4 gene. Oct-4 mRNA was expressed in each of the AP+GSC colonies (Fig. 1Ab
); C-kit was not expressed in the colonies (not shown). The expression of Oct-4 protein in colonies was demonstrated by immunostaining and Western blotting (Fig. 1Ac, d
). Figure 1Ac
shows the specific nuclear Oct-4 expression of the AP+GSCs. Figure 1Aci
shows a DAPI image. The specific nuclear expression of Oct-4 protein in each AP+GSC was further demonstrated by immunostaining and colocalization with DAPI, as shown by confocal microscopy (Fig. 1Acii
). Oct-4 protein expression in the AP+GSCs was also demonstrated by Western blotting. NCCIT cells (seminoma-embryonal carcinoma cells, which dominantly express Oct-4 protein in the nucleus) were used as a positive control (Fig. 1Ad
).
|
Characterization of the AP+ GSC colonies
To characterize the AP+ GSC colonies, the colony cells were immunostained for specific cell surface markers. As shown in Fig. 1B
, these AP+ GSCs were positively immunostained with GSC-related cell surface markers, including CD29 (integrin β1, GSC, and ESC marker); CD49f (integrin 6, GSC, and ESC marker); and notably, embryonic stem cell marker SSEA-1 (stage specific embryonic antigen-1, ESC marker). These marker proteins were expressed in the cytoplasm as well as on the cell surface of each AP+GSC (confocal images in insets). The AP+GSC colonies did not show positive C-kit staining (C-kit panel). Staining for CD31 and CD34 was also negative (not shown). The gene expression profile of these AP+GSC colonies was assessed by RT-PCR (Fig. 1C
). The colonies expressed Oct-4, Nanog, and Sox2 (for ESCs); Blimp1, Fragilis, Stella, and Mvh (for PGCs); and Piwi, Plzf, C-Ret, and Ngn-3 (for SSCs). There was very weak expression of the differentiation germ cell markers Dazl and Tex14 and no expression of Gcnf (Oct-4 suppressor factor) or c-kit (differentiated germ cells; Fig. 1C
). The AP+GSC (AP+Oct-4+c-kit–) colonies accounted for 0.2% of all testicular cells.
ESC/PGC characteristics of the AP+GSC colony cells
Given that the AP+GSC colonies had gene expression patterns similar to those of ESCs and/or PGCs, we next compared the characteristics of the AP+GSC colonies with those of ESCs/PGCs. As shown in Fig. 2A
, the AP activity of the GSCs was similar to that of ESCs (Fig. 2Aa, b
). Further analysis by quantitative real-time PCR showed that the Oct-4 mRNA level of the AP+GSCs was 70% that of ESCs (Fig. 2Ac
). The AP+GSC colonies also showed PGC-like characteristics, such as colony cell migration (Fig. 2Ba, b
) and retinoic acid-stimulated cell proliferation (Fig. 2Bc
) (17)
. The gene expression profile of the ESCs and AP+GSCs was further analyzed by RT-PCR. As shown in Fig. 2C
, the colonies not only expressed SSC-related genes (c-Ret and Ngn3) but also expressed genes of the ESCs (Oct-4, Nanog, Sox2, and Eras) and PGCs (Oct-4, Blimp 1, Fragilis, Stella, and Mvh). These results suggest that AP+GSC colony cells possess pluripotent potential.
|
Differentiation potential of AP+GSC colonies in vitro
Because of the similar characteristics of AP+GSC colonies and ESCs/PGCs, we examined the differentiation potential of the AP+GSC colonies in vitro. As shown in Fig. 3
, the AP+GSC colonies formed neurons (Fig. 3A
, MAP2+ staining), astrocytes (Fig. 3B
, GFAP+ staining), and oligodendrocytes (Fig. 3C
, O4+ staining) under retinoic acid treatments. Furthermore, AP+GSCs have the differentiation plasticity to form hepatocyte-like cells (Fig. 3D
, AFP+ staining, solid arrowhead). These observations suggested AP+GSCs have multipotent differentiation capability in vitro. The positive and negative controls of the antibodies used for immunostaining are shown in Supplemental Fig. 2.
|
Contribution of AP+GSCs to spermatogenesis and normal embryonic development
To address whether AP+GSCs possess germ cell differentiation capability, we first found that treatment of stem cell factor would be able to induce AP+GSCs or EGFP+ AP+GSCs to differentiate into c-kit+ germ cell precursors in vitro (Fig. 3E
, c-kit+ staining; also see Supplemental Fig. 3A). The in vivo germ cell differentiation potential of the AP+GSC colony cells was initially investigated by direct testis transplantation, since the colony cells were germ cell oriented (15)
. The EGFP-AP+GSC colony cells were transplanted into one testis of bulsufan-treated FVB mice (Fig. 4A
). Eight weeks after transplantation, the recipient mice were euthanized to check the testis size and spermatogenic reconstitution. As shown in Fig. 4Aa
, the gross morphology of the transplanted testis (TT) showed a larger size than the untransplanted testis (UT), suggesting the successful reconstitution of the transplanted EGFP-AP+GSCs in TT. Histological staining confirmed that the UT lost the germ cells (Fig. 4Ab
). In some bulsufan-treated seminiferous tubules, spermatogenesis was recoverable (Fig. 4Ac
). However, these cells showed negative immunostaining with anti-GFP antibody (Fig. 4Ad
). TT demonstrated complete spermatogenesis with full mature sperm in the seminiferous tubules (Fig. 4Ae
). That spermatogenesis originated from the EGFP-AP+GSCs was further confirmed by anti-EGFP immunostaining. The significant EGFP-positive immunostaining of cells in the seminiferous tubules of TT is shown in Fig. 4Af, g
. A negative control was produced using a control IgG as the primary antibody (Fig. 4Ah
).
|
Next, the EGFP-AP+GSCs were microinjected into blastocysts to test their pluripotent contribution to chimera offspring. Chimerism was observed in 76% (19 of 25) of the neonatal mice. EGFP-positive donor cells were found in the three germ layers of d 18 embryonic chimeras (Supplemental Fig. 3Ba–l) and colocalized with CD31 (mesoderm), cytokeratin-14 (ectoderm), and AFP proteins (endoderm) (Supplemental Fig. 2Bm–o). The d 4 neonatal chimeras were further subjected to IVIS for in vivo live fluorescence imaging. The GFP fluorescence was predominantly detected on the ventral side of the chimeras (Fig. 4B
). The contribution of the EGFP+AP+GSCs in offspring seems not dominantly locate at skin and hair, although we still observed chimeric phenotype in the tail and dorsal/ventral sides (Fig. 4C
, open arrowheads). However, by analyzing 3-wk-old chimera mice, we found GFP-positive cells were observed in a wide variety of organs, including brain, lung, stomach, small intestine, heart, spleen, and tail. A small amount of the GFP-positive cells could be found in ovary (Fig. 4D
). The teratoma formation ability of the AP+GSCs was further examined by using NOD/SCID mice. The AP+GSCs contributed some of the embryonic germ layers, such as muscle, adipose tissue, salivary-like cells, and CK-14+ epithelial-like cells (Supplemental Fig. 4). Together with these observations it is suggested that AP+GSCs derived from the serum-free testicular stromal coculture system exhibited the pluripotency.
Endocrine effect of IGF-1 on the formation of AP+GSC colonies
The serum-free testicular stroma coculture system seems to provide microenvironmental factors that regulated the self-renewal growth and pluripotent status of AP+GSCs. To uncover the microenvironmental factors that regulated the pluripotency of AP+GSCs, cytokine antibody array was utilized to find the potential niche cytokine. As shown in Fig. 5A
, the AP+GSC formation began on the third culture day and reached maximum with colony size and density at d 7. BM and D3-CM were collected and assayed for cytokine expression, which may activate the AP+GSC formation. Figure 5B
shows that several cytokines increased over time, most notably, IGF-1. To address the role of IGF-1 in AP+GSC formation, increasing concentrations of IGF-1 were added to the medium in place of insulin. As shown in Fig. 5C
, IGF-1 enhanced the colony formation in a dose-dependent manner at concentrations of 0–1 ng/ml. The IGF-1 effect reached a plateau at 1–10 ng/ml (with 80% of colony formation efficiency; P<0.001). This result not only demonstrates the dominant effect of IGF-1 on AP+GSCs but also suggests the cooperative potential of IGF-1/IGF-1R with other receptor-mediated signaling pathways.
|
The important potential of IGF-1/IGF-1R-mediated signaling for the AP+GSC colony formation was further supported by the insulin-containing medium we used in the cultures. In our experiments, by dissecting the medium components, we found insulin (5 µg/ml) significantly increased the percentage of AP+GSC formation while compared with MCDB201 medium (Supplemental Fig. 5A; P<0.001). As insulin is known to activate both the IR and the IGF-1R in a dose-dependent fashion, the dose effect of insulin on AP+GSC colony formation was examined. As shown in Supplemental Fig. 5B, insulin increased the formation of AP+GSCs at concentrations of 50 ng/ml to 5 µg/ml, which is much higher than its ligand affinity. This observation hints at the action of insulin on IGF-1R and strongly supports the premise that IGF-1/IGF-1R-mediated signaling is important for AP+GSC colony formation.
The specific role of IGF-1 and insulin on AP+GSC formation was further examined by using neutralizing antibody against IGF-1R (IR3) and anti-IR (IR). As shown in Fig. 5D
, IR3 specifically decreased the AP+GSC formation (P<0.001). The IR showed a slight suppression effect (20%; P<0.05). The effect of IGF-1/IGF-1R on the AP+GSC formation was further confirmed by using PPP treatment, which is known to selectively inhibit tyrosine-phosphorylation of IGF-1R and hence to suppress downstream signaling (18)
. As shown in Fig. 5E, F
, PPP treatment not only affected the formation of AP+GSC colony but also significantly suppressed the AP activity of these cells (P<0.001). Moreover, treatment of PPP significantly down-regulated the expression of Oct-4, Nanog, and Sox2, suggesting that IGF-1R-mediated signaling is important for maintaining the pluripotency of cultured SSCs. In addition, genes related to PGCs (such as Blimp1, Mvh, and Fragilis) and genes associated with self-renewal regulation (e.g., Plzf and Piwi) were reduced (Fig. 5G
). To further examine the possible signaling pathway responsible for AP+GSC colony formation, the specific PI3K inhibitor LY294002 was used to block the IGF-1/IGF-1R-mediated signal pathway. As shown in Fig. 5H
, LY294002 significantly suppressed GSC colony formation and AP activity of the AP+GSCs. Akt phosphorylation (Ser-473 of the AP+GSCs) was also dramatically suppressed by PPP treatment (Fig. 5I
). Taken together, these findings strongly support the essential role of IGF-1/IGF-1R-mediated signaling is important for maintaining the stemness of SSCs.
Expression of IGF-1 by testicular stromal cells
To further identify the specific type of stromal cells responsible for producing IGF-1 to the culture, antibodies specifically against various types of testicular stromal cells were used including antibodies against myoid cells (anti-SMA), Leydig cells (anti-CYP11A1), and Sertoli cells (anti-MIS). The results indicate laminin-coating apparently selectively adheres myoid cells and Leydig cells, which serve as feeder cells in AP+GSC cultures. Most of the AP+GSC colonies were closely associated with myoid cells (Supplemental Fig. 6), but the Leydig cells were responsible for producing IGF-1 to the culture, as expression of IGF-1 was obtained in Leydig cells, as judged by positive immunostaining results. As shown in Fig. 6Ac
, the expression of IGF-1 can be detected within Leydig cells (Fig. 6Ad
, yellow). Furthermore, it seems that low-level expression of IGF-1 protein was also detected in the AP+GSCs (Fig. 6Ba
, arrows). The expression of Igf-1 mRNA in individual AP+GSC colonies is shown in the inset of Fig. 6Ba
. Myoid cells showed negative immunostaining for IGF-1 expression (Fig. 6Ba, b
; open arrowhead). In support of this observation, IGF-1 was positively detected in Leydig cells (solid arrowhead) and weakly in the seminiferous tubules (open arrowhead) in testis of d 0–2 neonatal mice (Fig. 6Bc
). The positive and negative controls of the antibodies used for immunostaining of Fig. 6
are shown in Supplemental Fig. 7. The expression of IGF-1R and IR in AP+GSCs was also confirmed by RT-PCR and immunocytochemical staining (Supplemental Fig. 8).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
3
4)
. However, little is known about the endocrine regulation of germ cell pluripotency. This lack of understanding leads to low efficiency when generating multipotent mouse GSCs from wild-type testis in vitro (3)
. In this report, we utilized a serum-free culture system and d 0–2 neonatal mice in which the testicular germ cells were in the gonocyte stage, to generate pluripotent stem cells from wild-type mouse testis. With this system, we found that IGF-1 secreted by Leydig cells may mediate IGF-1/IGF-1R through PI3K/Akt signaling to regulate pluripotent transcription factors such as Oct-4 and germ cell pluripotency.
Many researchers have used serum-containing medium to cultivate GSCs and SSCs (3
, 4
, 6
, 7)
; however, the detrimental effects of serum on the expansion of GSCs have also been reported (8
, 19)
. In serum-free medium, MEF or STO cells have been commonly used as feeders to support the maintenance or proliferation of SSCs in vitro (6
, 8
, 19)
. Exogenous factors such as GDNF, bFGF, and/or GFR1 have been regularly added to medium to support the SSC proliferation. In these experiments, some of the SSCs showed weak AP activity (8)
, c-kit+SSEA1– (6)
, or chain formation (20)
. As stem cells are able to change their cell fate by quickly responding to their microenvironment, the complex serum components and nonphysiological feeders (MEF and/or STO cells) may dramatically hinder the pluripotent processes in vitro as well as hinder the identification of key endocrine factors involved in the regulation of GSC stemness.
In our experiment, the stem cell colonies showed strong AP activity (AP+GSCs), in contrast to SSCs (3
, 8)
. The AP+GSC colonies showed clump morphology with no chain formation. This phenomenon is coincident with the very weak expression of Tex 14 (Fig. 1C
) located at the cytoplasmic bridges of Apair or Aaligment4–32 cells of type A spermatogonia (10)
. With serum treatment, the AP+GSC colony cells changed morphology to become flattened and expressed genes of differentiated sperm, including Tex14, Dazl, and c-kit (Supplemental Fig. 9). This result is consistent with a previous study that demonstrated a negative effect of serum on mouse SSC formation (8)
. The AP+GSC colonies expressed SSEA-1 protein (Fig. 1B
) (21)
and shared some characteristics with ESCs/PGCs. For example, these colonies expressed Oct-4 (Pou5f1) (22)
, Nanog (23)
, E-ras (24)
, and Sox2 (25)
in a manner similar to ESCs. Nanog is known to be expressed in ESCs but not SSCs (26)
. The expression of Nanog in AP+GSCs strongly supports their pluripotent potential. The AP+GSCs also showed PGC-like gene expression, including expression of Oct-4, Sox2, Fragilis, and Stella (27)
, Blimp1 (B-lymphocyte-induced maturation protein-1) (28)
, and Mvh (Fig. 1C
). The phenomenon of PGC migration (with AP positive staining) and trans-retinoic acid stimulated cell proliferation in the AP+GSC colonies was also observed (Fig. 2B
). The pluripotency of the AP+GSC colonies was further demonstrated by their in vitro differentiation capacity (Fig. 3
). In addition, the EGFP-AP+GSCs showed the ability of teratoma-like formation by using NOD/SCID mice model and contributed to the three germ layers of the chimeras offspring by in vivo blastocyst injection (Fig. 4
; Supplemental Figs. 3 and 4). Apparently, these primitive AP+GSCs (Oct-4+c-kit– GSCs) in neonatal testis were not limited to germ cell differentiation but also underwent dedifferentiation to become ES-like pluripotent stem cells.
Stromal cells that control the GSC fates in a culture environment were selected by coating material. In contrast to our previous study of lung stem cells, in which collagen I was the preferred substrate (29)
, laminin (at 275 ng/cm2) was the optimal coating material to support AP+GSC colony formation (Supplemental Fig. 1) and PGC-like migration (Fig. 2B
). In support of these results, laminin has been shown to play an important role in PGC migration (30)
. Further examination by immunostaining demonstrated that laminin selected Leydig cells and myoid cells as feeders in our culture system (Supplemental Fig. 6; Fig. 6A
). These cells may provide essential endocrine components, in addition to exogenous factors, that support the formation of pluripotent AP+GSCs. Cytokine antibody arrays identified IGF-1 as an important factor (Fig. 5
). This observation is consistent with our culture condition, which used insulin at a concentration of 5 µg/ml in medium. At this concentration, insulin is able to bind to the IGF-1R as well as the IR. In line with that, our data suggested an effective response on AP+GSC colony formation at low concentrations of IGF-1 (1 ng/ml; Fig. 5C
). The IGF-1 may be secreted from Leydig cells (31)
and AP+GSCs (Fig. 6B
). As the AP+GSCs express both the IR and IGF-1R at the gene and protein levels (Supplemental Fig. 8), the IGF-1 may interact with IGF-1R on the AP+GSCs via a paracrine and/or autocrine manner to support the stemness of GSCs. This hypothesis is strongly supported by the positive immunohistochemical staining of IGF-1 in interstitial spaces and seminiferous tubules (Fig. 6Bc, d
). Myoid cells are known to physiologically locate at the basal membrane and to be in close contact with the SSCs in testis. Leydig cells are located in the interstitial space close to myoid cells. The proximity of Leydig cells and myoid cells to basal membrane SSCs in vivo may govern the paracrine regulation of IGF-1 on SSC stemness. This hypothesis is strongly supported by a recent study by Oatley and Brinster (26)
, which suggests paracrine regulation of testicular interstitial space in germ cell stemness.
The IGF-1/IGF-1R-mediated signals in AP+GSC formation was further verified by an antibody neutralizing assay and PPP treatment (Fig. 5E
). PPP is known to efficiently inhibit the phosphorylation of IGF-1R without interfering with IR activity (18)
. The suppressive effect of PPP on formation of AP+GSC colonies is consistent with the neutralizing effect of IR3 (IGF-1R-neutralizing antibody). Most importantly, the pluripotent genes of the AP+GSCs, such as Oct-4, Blimp1, Sox2, and Nanog, were dramatically suppressed by PPP, suggesting the IGF-1/IGF-1R-mediated signaling pathway in the regulation of the AP+GSC pluripotency. IGF-1/IGF-1R is known to transmit intracellular signaling through PI3K/Akt or Ras/ERK pathway (32)
. In our experiments, the colony formation efficiency and AP activity of the AP+GSCs were also dramatically reduced by LY294002 (Fig. 5H
) but not by PD98059 (data not shown). In addition, the Akt phosphorylation (Ser-473 of AP+GSCs was also dramatically suppressed by PPP (Fig. 5I
). This observation strongly supports the association of IGF-1/IGF-1R-mediated PI3K/Akt signaling in AP+GSC pluripotency. In line with our observations, recent studies (33
, 34)
have shown a close association between the PI3K-Akt pathway and the pluripotency of germ cells. Activation of the PI3K-Akt pathway in PGCs promotes cell proliferation as well as the conversion into teratoma or pluripotent embryonic germ cells. In addition, the PI3K/Akt signal axis has been shown to crosstalk with self-renewal mechanisms in ESCs (35
, 36)
. In ESCs, IGF-1 has also been reported to cooperate with bFGF in the self-renewal process of ESCs (37)
. Similar to this observation, our results by cytokine array analysis demonstrated that the expression of IGF-1 and bFGF were increased in our culture medium for AP+GSC formation (Fig. 5B
). These results strongly support the role of IGF-1/IGF-1R signaling in the regulation of pluripotency of mouse GSCs.
The endocrine factors in the stroma cell microenvironment apparently play important roles in the regulation of germ cell pluripotency. The most important finding of our study was elucidating the role of endocrine factor IGF-1 and IGF-1/IGF-1R-mediated PI3K/Akt signaling in the regulation of mouse SSC stemness. This finding may have important implications in the study of endocrinology, germ cell development, and tumorigenesis.
|
ACKNOWLEDGMENTS |
|---|
Received for publication September 25, 2008. Accepted for publication January 29, 2009.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. Sadri-Ardekani, S. C. Mizrak, S. K. M. van Daalen, C. M. Korver, H. L. Roepers-Gajadien, M. Koruji, S. Hovingh, T. M. de Reijke, J. J. M. C. H. de la Rosette, F. van der Veen, et al. Propagation of Human Spermatogonial Stem Cells In Vitro JAMA, November 18, 2009; 302(19): 2127 - 2134. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. W Kuijk, B. Colenbrander, and B. A J Roelen The effects of growth factors on in vitro-cultured porcine testicular cells Reproduction, October 1, 2009; 138(4): 721 - 731. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
