|Home | About | Journals | Submit | Contact Us | Français|
Spermatogonial stem cells (SSCs) maintain spermatogenesis throughout a man’s life and may have application for treating some cases of male infertility, including those caused by chemotherapy before puberty. We performed autologous and allogeneic SSC transplantations into the testes of 18 adult and 5 prepubertal recipient macaques that were rendered infertile with alkylating chemotherapy. After autologous transplant, the donor genotype from lentivirus-marked SSCs was evident in the ejaculated sperm of 9/12 adult and 3/5 prepubertal recipients after they reached maturity. Allogeneic transplant led to donor-recipient chimerism in sperm from 2/6 adult recipients. Ejaculated sperm from one recipient transplanted with allogeneic donor SSCs were injected into 85 rhesus oocytes via intracytoplasmic sperm injection. Eighty-one oocytes were fertilized, producing embryos ranging from 4-cell to blastocyst with donor paternal origin confirmed in 7/81 embryos. This demonstration of functional donor spermatogenesis following SSC transplantation in primates is an important milestone for informed clinical translation.
In 1994, Ralph Brinster and colleagues transplanted mouse spermatogonial stem cells (SSCs) into the seminiferous tubules of infertile recipient mice and observed donor-derived spermatogenesis that was competent to produce viable progeny (Brinster and Avarbock, 1994; Brinster and Zimmermann, 1994). SSC transplantation has since become the gold standard bioassay for experimental assessment of SSC activity (Phillips et al., 2010) and may also have application in the human fertility clinic. One potential clinical application of SSC transplantation is to preserve and restore the fertility of male cancer survivors (Kubota and Brinster, 2006; Geens et al., 2008; Schlatt et al., 2009; Wyns et al., 2010; Hermann and Orwig, 2011).
Chemotherapy and radiation treatments for cancer or other conditions can permanently damage fertility (Mitchell et al., 2009). Adult male patients have the option to preserve their future fertility by cryopreserving sperm. Unfortunately, there are no standard-of-care options to preserve the fertility of prepubertal boys who are not yet producing mature sperm. For these patients, it may be possible to isolate and freeze SSCs obtained via testicular biopsy prior to gonadotoxic therapy and have these cells reintroduced into their testes after cure (Brinster, 2007; Clark et al., 2011). If results in animal models translate to the clinic, this autologous transplantation paradigm may permanently restore natural fertility. The feasibility of this approach is supported by observations in lower animal models that SSCs from donors of all ages, newborn to adult, can regenerate spermatogenesis (Shinohara et al., 2001; Ryu et al., 2003) and that SSCs can be cryopreserved and retain spermatogenic function upon thawing and transplantation (Dobrinski et al., 1999; Dobrinski et al., 2000; Brinster, 2002).
Large animal models are critical for examining the safety and feasibility of experimental therapies before they are translated to the clinic. SSC transplantation has been reported in seven previous large animal studies (Table S1). All of those studies, except one in the boar (Mikkola et al., 2006), employed irradiation to destroy spermatogenesis and cause infertility. There is a dearth of information on the efficacy of SSC transplantation in chemotherapy-treated large animals, probably due to the significant challenges associated with clinical management of animals treated systemically with high-dose chemotherapies that cause severe hematopoietic deficits (Hermann et al., 2007). However, the importance of this experimental paradigm should not be overlooked because high-dose alkylating chemotherapies are used routinely for conditioning prior to hematopoietic stem cell (HSC) transplantation and are associated with high risk of infertility (Wallace et al., 2005; Lee et al., 2006; Mitchell et al., 2009; Green et al., 2010).
The first large animal SSC transplants were performed in monkeys by Schlatt and colleagues (see Table S1), who described autologous transplants into irradiated monkey recipients in 2002 and again in 2011 (Schlatt et al., 2002; Jahnukainen et al., 2011). Each study reported that the transplanted right testis was larger than the un-transplanted left testis in one animal, but presence or function of donor sperm was not evaluated. Thus, the question of whether SSC transplant can be translated to the primate system and produce functional sperm still remains. The translational significance of this question is high because clinics around the world (Keros et al., 2007; Wyns et al., 2008; Ginsberg et al., 2010; Sadri-Ardekani et al., 2011; Oktay, 2011; Orwig et al., 2011; Schlatt and Kliesch, 2012) are already cryopreserving testicular tissue for boys in anticipation that SSCs in that tissue can be used to restore fertility via autologous SSC transplantation, autologous tissue grafting, xenografting or in vitro germ cell differentiation (Brinster, 2007; Rodriguez-Sosa and Dobrinski, 2009; Sato et al., 2011; Clark et al., 2011). Establishing the feasibility of SSC transplantation in the primate model will have important implications for how testicular tissue should be processed and for educating patients and physicians about the potential downstream applications.
We previously described a nonhuman primate model of cancer survivorship in rhesus macaques where infertility was caused by alkylating chemotherapy (busulfan) (Hermann et al., 2007). We employed that model in the current study to examine the feasibility of SSC transplantation in prepubertal and adult rhesus macaques, which have testis biology, endocrine regulation and immune function that is similar to humans (Plant and Marshall, 2001; Hermann et al., 2010; Messaoudi et al., 2011). Prophylactic autologous peripheral blood stem cell (PBSC) transplant (Donahue et al., 2005; Kang et al., 2006) was used to counteract the hematopoietic deficits in all animals. This complex experimental design involving HSC and SSC transplantation models the clinical scenario of hematopoietic stem cell (bone marrow or PBSC) transplant patients who are at high risk for infertility (Wyns et al., 2010). Our results indicate that transplanted SSCs can regenerate spermatogenesis in busulfan-treated primates and produce functional sperm capable of fertilizing oocytes and leading to preimplantation embryo development.
Schlatt and co-workers pioneered ultrasound-guided rete testis injection into monkey testes in 1999 (Schlatt et al., 1999) and this technique has now been applied to introduce testis cell suspensions into the seminiferous tubules of several large animal species (Schlatt et al., 1999; Schlatt et al., 2002; Honaramooz et al., 2003; Izadyar et al., 2003; Mikkola et al., 2006; Kim et al., 2008; Herrid et al., 2009). In contrast to a typical rodent SSC transplant where the testis efferent ducts and/or rete testes are accessed surgically through an abdominal incision (Ogawa, 2001), ultrasound-guided rete testis injection does not require surgery. Briefly, ultrasound is used to visualize the rete testis and guide the injection needle through the scrotal skin and into the rete testis space, which is contiguous with all seminiferous tubules (Figure 1 and Movie S1). With this approach, we introduced an average of 1041 ± 82 μl of cell suspension into the rete testis and seminiferous tubules of adult recipients and 222 ± 26 μl into juvenile recipients. Cell concentrations ranged from 58–232 × 106 viable cells/ml; an average of 88 × 106 viable cells were injected per adult testis and 45.8 × 106 viable cells were injected per juvenile testis (Table S2).
To assess the regenerative capacity of primate SSCs, we performed a series of autologous transplant experiments in busulfan-treated macaques (Hermann et al., 2007). Because the doses of busulfan required to deplete endogenous spermatogenesis are also myelosuppressive, all animals received autologous PBSC transplants to support rapid hematopoietic recovery (Figure 2). Testis cells were obtained via hemicastration or biopsy of one testis and cryopreserved prior to busulfan chemotherapy.
In order to distinguish transplanted SSCs and their progeny from endogenous cells we treated donor cells with lentiviral vectors containing Ubiquitin-C (UBC)-eGFP, elongation factor 1α (EF1α)-GFP or EF1α-mCherry transgene inserts (Table S2) prior to transplant. This approach permanently marks donor cells and allows detection of the labeled donor cells in tissue or ejaculated sperm by their genotype (e.g., a specific lentiviral DNA sequence).
Approximately 10–12 weeks after busulfan treatment (corresponding to the time when sperm counts reach 0 in adults), cells were thawed, treated with lentivirus and transplanted back into the other testis of the same animal (Figure 2). Lentivirus-treated autologous SSCs were transplanted into the seminiferous tubules of 12 adult and 5 prepubertal recipient macaques by ultrasound-guided rete testis injection. Polymerase chain reaction (PCR) was used to detect sperm produced from lentivirus-marked SSCs in the ejaculates of recipient animals. Overall, spermatogenesis was evident in 11/12 adult and 5/5 prepubertal (after puberty) recipients after transplant (Figure 3A and Tables S2–S4).
The duration of spermatogenesis, from SSC to sperm is roughly 42–44 days, followed by 10.5 days of epididymal transport time (Amann et al., 1976; Clermont and Antar, 1973; Hermann et al., 2010). Recovery of spermatogenesis to normal levels (≥ 15×106) were observed in adult autologous recipients an average of 40.1 ± 4.9 weeks after busulfan treatment (11 of 12 adults; ranged from 15 to 63 weeks; Table S3). In our previous study, recovery of spermatogenesis from endogenous SSCs occurred by 24 weeks after a low dose of busulfan (4 mg/kg) that did not eliminate endogenous SSCs; spermatogenic recovery was not observed in animals treated with the higher busulfan doses (8 and 12 mg/kg) employed in this study (Hermann et al., 2007). The time to spermatogenic recovery in this study can likely be attributed to the substantial depletion of the endogenous SSC pool, which is not completely replenished by transplanted SSCs. Thus, spermatogenesis originates from sporadic foci of individual endogenous and/or transplanted SSCs that must expand laterally to repopulate the seminiferous tubules as well as differentiate to produce sperm. These factors apparently prolong the time required to reach a steady state threshold sufficient to produce normal sperm counts in the ejaculate.
PCR genotyping for the lentiviral backbone indicated stable donor signal in the ejaculates of 9/12 adult and 3/5 prepubertal autologous recipients (Figure 3B, Tables S2–S4). Donor signal was considered stable when lentiviral genotype was observed in at least four separate semen samples collected over the course of at least three months. Results from autologous recipient M037 are shown in Figure 3 where sperm re-appeared in the ejaculate between 20 and 30 weeks after transplant (Figure 3A). Donor lentiviral sequence was detected by PCR coincident with the appearance of sperm (Figure 3B). Overall, PCR signal from lentivirus marked SSCs decayed over time (see Tables S3–S4), suggesting a low efficiency of virus-marked SSC engraftment. Histological comparison of the testis and cauda epididymis from M037 (Figure 3C) clearly demonstrates more spermatogenic recovery (60% of seminiferous tubule cross-sections contained spermatogenesis) compared with a transplant recipient which failed to exhibit sperm in the ejaculates after transplant (M214; Figure 3D) and had spermatogenesis in only 24% of tubule cross-sections. For reference, all seminiferous tubules were devoid of germ cells in a busulfan-treated animal that received no transplant 26 weeks after busulfan treatment (M104; Figure 3E). We were unable to observe fluorescent segments of seminiferous tubules (i.e., marked by the lentivirus and regenerated spermatogenesis) after systematic evaluation of each autologous recipient testis at necropsy by epifluorescence microscopy. The failure to observe lentiviral reporter expression may result from epigenetic silencing of the transgene, insufficient expression for this detection mode, or both.
The best way to demonstrate that transplanted SSCs produce functional sperm is to demonstrate their ability to fertilize oocytes. Unfortunately, our autologous transplant approach was not amenable to fertilization studies because the efficiency of marking SSCs was very low (data not shown). In addition, we were not able to distinguish fluorescence from lentivirus marked sperm from autofluorescence that was observed in most ejaculates (data not shown). Fertilizing oocytes from a random population of sperm of which only a very small percentage were genetically marked was not practical. Therefore, we performed additional experiments using an allogeneic recipient approach in which all donor sperm had unique DNA microsatellite allele profiles that could be distinguished from endogenous recipient sperm.
We utilized an allogeneic transplant paradigm where donor testis cells from unrelated individual animals were transplanted into recipient testes. While some previous reports demonstrated that transplanted allogeneic testis cells were tolerated in large animal models allowing engraftment of unrelated donor SSCs (Honaramooz et al., 2002; Honaramooz et al., 2003; Kim et al., 2008), SSCs from unrelated donors failed to regenerate spermatogenesis in bull testes (Izadyar et al., 2003) (Table S1). Thus, the potential for immune effects on SSC engraftment are unclear. Therefore, donor and recipient pairs were matched based on low recipient T-cell reactivity to donor antigens using multiple lymphocyte reaction (MLR; data not shown) analysis (Ezzelarab et al., 2008). In addition, 5 of 6 allogeneic recipients were treated with an immune suppression regimen (anti-CD154; Table S2) (Kirk et al., 1999). We discriminated sperm originating from donor and recipient SSCs using microsatellite fingerprinting, as described previously to detect donor sperm production in SSC-transplanted dogs (Kim et al., 2008).
All six allogeneic recipients exhibited low levels of spermatogenic recovery in the post-transplant period of evaluation (Figure 4A and Tables S2–S3). Microsatellite DNA fingerprinting revealed donor/recipient chimerism in sperm from two of the six allogeneic recipients (M212 and M027, Table S2–S3). Sperm retrieved from the left cauda epididymis at necropsy from animal M212 exhibited a minor peak of donor signal (donor M214; 236bp and 244bp alleles at locus DS11S2002; 194bp and 244bp alleles at locus D12S67) amidst a background of recovering endogenous spermatogenesis (Figure 4B–D). Donor signal was not detectable in ejaculated sperm from M212 likely due to inflammation in the left epididymis evident in the post-transplant period and confirmed at necropsy, preventing donor sperm transit to the ejaculate (data not shown). A second allogeneic recipient, M027, produced ejaculated sperm exhibiting donor signal (donor M092; 187bp allele at locus D3S1768; 263bp allele at locus D17S1300) for more than 17 months (50 total samples) with analysis ongoing at the time of manuscript submission (Figure 4E–G). Thus, these data show that transplanted allogeneic SSCs produce sperm in recipient testes. To quantify the degree of donor sperm production as a function of time, we identified single nucleotide polymorphisms (SNPs) that distinguish donor sperm from recipient sperm essentially as described previously (Alizadeh et al., 2002; Kim et al., 2008). We screened 23 SNPs reported in the Monkey SNP database at Oregon National Primate Research Center (monkeysnp.ohsu.edu), but none distinguished donor and recipient and/or were suitable for qPCR. However, while screening one reported SNP (rs4543622) within the class II major histocompatibility complex transactivator (CIITA) locus, we identified a previously unreported SNP for which the recipient (M027) was homozygous for one allele (G) and the donor (M092) heterozygous both alleles (A/G). Standard curves for the relative abundance of each allele as previously described were used to determine the percent of donor chimerism in DNA isolated from recipient M027 following monthly semen samples collected between 3 and 17 months after transplant (Fig. 4H). We observed a consistent level of donor (M092) chimerism (ranging from 1.7 to 17.2%) in M027 sperm samples for the duration of the 14 months analyzed (Fig. 4H).
To assess function of donor (M092) sperm, ejaculated sperm from recipient M027 (collected 30 weeks after transplant) were used to fertilize rhesus oocytes by intracytoplasmic sperm injection (ICSI) (Hewitson et al., 1999; Mitalipov et al., 2006). Of 85 oocytes injected, 81 (95%) were fertilized (formed male and female pronuclei) and subsequently cleaved (Figure 5, Table S5). Upon in vitro culture, 23% of embryos reached the blastocyst stage with normal morphology (Table S5). To determine sire by microsatellite DNA fingerprinting, all blastocysts and arrested embryos were individually harvested and used for whole genome DNA amplification. Genotyping was done for the gender marker AME to determine sex of embryo and for 8 microsatellite loci, two of which (DXS2506 and D15S823) definitively discriminate the genotype of the SSC donor (M092) from the transplant recipient (M027), as well as oocyte donors (Figure 5 and Table S6). In this genotyping paradigm, the 286bp allele at the X-linked locus DXS2506 and the 337bp allele at locus D15S823 were both unique to M092 and their presence in an embryo could only arise from M092 paternal contribution (Figure 5 and Table S6). Of the 81 embryos genotyped, 7 exhibited definitive donor (M092) sire, three of which advanced to the morula stage of preimplantation development (Figure 5 and Table S6). Since DXS2506 is an X-linked marker, male (XY) embryos displayed only the maternal allele at this locus, including 3 XY M092-sired embryos (embryos 1, 8 and 63, Figure 5L and 5O, Table S6). M092 donor paternal contribution in these embryos was confirmed by the presence of the 337bp allele at locus D15S823. These results indicate that sperm generated from transplanted primate SSCs are competent for fertilization and preimplantation embryo development.
Adult tissue stem cell transplantation for homologous tissue regeneration was first described for primates in the 1950s when bone marrow stem cells were used to reconstitute the hematopoietic systems of monkeys and humans treated with chemotherapy or radiation (Crouch and Overman, 1957; Thomas et al., 1957). Large animals, primarily the dog and monkey, were instrumental for establishing the safety, feasibility and range of applications for bone marrow transplantation. Today, approximately 50,000 bone marrow or hematopoietic stem cell (HSC) transplant procedures are performed world-wide each year for diseases ranging from cancer to thalassemia, sickle cell anemia, autoimmune and immune-deficiency disorders (Appelbaum F.R., 2007; Powell et al., 2009).
Like hematopoiesis, spermatogenesis is a highly productive stem cell-based system that produces millions of sperm per gram of tissue each day (Sharpe, 1994). This productivity is possible because a relatively small stem cell pool generates progeny that undergo several rounds of transit-amplifying divisions before producing the terminally differentiated sperm (Potten, 1992). Two sequelae of highly productive stem cell-based systems are 1) that they can become targets of chemotherapy or radiation treatments that damage rapidly dividing cells (Potten, 1995; Meistrich, 1993; Mauch et al., 1995) and 2) that transplantation of a small number of stem cells is adequate to functionally reconstitute the dependent systems (e.g., hematopoiesis and spermatogenesis) (Potten et al., 1979; Potten, 1992; Osawa et al., 1996; Ogawa et al., 2000; Shinohara et al., 2001; Copelan, 2006). Here we demonstrate the feasibility of SSC transplantation in a nonhuman primate model that is infertile due to alkylating chemotherapy (busulfan) and suggest that this technique has application for restoring the fertility of cancer survivors or bone marrow transplant recipients.
SSC transplantation has now been reported in mice, rats, monkeys, goats, bulls, pigs, sheep and dogs (Brinster and Avarbock, 1994; Brinster and Zimmermann, 1994; Ogawa et al., 1999; Schlatt et al., 2002; Honaramooz et al., 2003; Izadyar et al., 2003; Mikkola et al., 2006; Kim et al., 2008; Herrid et al., 2009). Among the seven other large animal SSC transplant studies reviewed in Table S1, four reported evidence of donor sperm in the ejaculate (goat, boar, dog, sheep) and two reported functional sperm (goat and sheep) that produced donor-derived progeny. Although the first large animal SSC transplants were performed in monkeys in 2002 (Schlatt et al., 2002), evidence of donor sperm from transplanted SSCs was lacking until the present study. It is important to demonstrate that transplanted SSCs can produce sperm in higher primate models that have the greatest relevance to human testis anatomy and physiology. It is equally important to demonstrate in primates that the testicular environment is competent to support spermatogenesis from transplanted SSCs following chemotherapy or radiation. Schlatt and colleagues previously reported SSC transplant in nonhuman primates that were rendered infertile by testicular irradiation (Schlatt et al., 2002; Jahnukainen et al., 2011). To date, SSC transplantation into a chemotherapy-treated large animal recipient has been reported only in the pig (Mikkola et al., 2006). Our results indicated that SSCs from prepubertal or adult rhesus macaques could engraft chemotherapy-treated recipient testes and generate spermatogenesis, including the production of donor sperm that were competent to fertilize rhesus oocytes resulting in pre-implantation embryo development.
We found evidence of donor spermatogenesis from both autologous and allogeneic transplant recipients and donor sperm function was evaluated in one allogeneic recipient (M027, the recipient of transplanted SSCs from M092). Donor spermatogenesis in autologous recipients was generally transient in recipient semen samples, appearing several times during post-transplant follow-up and sometimes in a cyclic manner. This result could be linked to a low efficiency of engraftment from virus-marked donor SSCs. Allogeneic recipient M027, on the other hand, demonstrated steady donor spermatogenesis that did not decline over time. The function of donor (M092) sperm in the ejaculates of recipient M027 (which contained a mixture of M092 and M027 sperm) was assessed by intracytoplasmic sperm injection (ICSI) of rhesus oocytes and was conducted in the Assisted Reproductive Technology/Embyronic Stem Cell Support Core of the Oregon National Primate Research Center (K.M, C.R and S.M). In vitro fertilization (IVF) is an alternative approach to test sperm function that would also assess the ability of donor sperm to penetrate the zona pellucida. The efficiency of IVF is similar to ICSI when using sperm from proven male donors. However, since the males used in this study were not proven breeders, ICSI was selected as the approach most likely to produce a definitive outcome with donor-derived embryos. The ICSI approach also eliminated the potential for contamination of genotyping results with a mixture of donor and endogenous recipient sperm. The ability of M092 SSC-derived donor sperm to fertilize rhesus oocytes by ICSI and stimulate early embryo development suggests that they were functionally normal.
In future studies it will be important to demonstrate that donor-derived embryos can be transferred to surrogate females for the production of viable donor-derived offspring. This was considered premature in the current study because only 7/81 embryos (8.6%, Tables S5–S6) had the donor genotype. Embryo biopsy to select only donor type embryos for transfer was not considered feasible (S.M.) and pregnancy rates after transfer are about 25% (Bavister et al., 1984; Wolf et al., 1989; Chan et al., 2001; Wolf et al., 2004). Therefore, the chances of achieving donor type progeny would be about 2.15% (8.6% donor embryos × 25% pregnancy rate). Besides the prohibitive cost, there were an insufficient number of recipient females available to reasonably expect donor offspring in this study. These challenges were less onerous in herd animal species where a single SSC transplant recipient could be used to fertilize a herd of females by natural breeding (Honaramooz et al., 2003) or artificial insemination (Herrid et al., 2009). Improvements in recipient preparation to more completely eliminate endogenous spermatogenesis, combined with development of donor SSC enrichment strategies (Hermann et al., 2009; Hermann et al., 2011) should substantially increase the proportion of donor sperm and enhance the opportunity to produce donor offspring in future nonhuman primate studies.
Due to concerns about immune rejection of cells from unrelated animals, five out of six allogeneic transplant recipients in this study were treated with antibodies against CD154 (Kirk et al., 1999), which blocks the T-cell co-stimulatory pathway. Donor spermatogenesis was observed in 2/5 immune suppressed recipients, but not in the one non-suppressed recipient (Table S2). Beginning in meiosis, spermatocytes and their progeny express novel autoantigens that are tolerated by the immune system, allowing production of genetically divergent gametes. Multiple mechanisms regulate immune privilege in the testis including the blood-testis barrier that limits access of immune components to the differentiated germ cells via Sertoli cell tight junctions, and somatic cell production of soluble factors (e.g., FAS ligand) that suppress the rejection of immunologically disparate cells (Fijak and Meinhardt, 2006). Testicular immune privilege has been used to explain the success of allogeneic SSC transplants between unrelated, immune competent individuals that were previously reported in several large-animal species (Honaramooz et al., 2002; Honaramooz et al., 2003; Mikkola et al., 2006; Kim et al., 2008). Although animal numbers in this study were not sufficient to demonstrate that immune suppression was required, our data clearly indicated that cells from unrelated donor animals were tolerated in immune-suppressed nonhuman primates.
Several promising techniques are in the research pipeline (i.e., SSC transplantation, testicular tissue grafting or xenografting, in vitro development of gametes) that may allow patients receiving gonadotoxic therapies to preserve their future fertility (Brinster, 2007; Rodriguez-Sosa and Dobrinski, 2009; Sato et al., 2011). SSC transplantation has the unique potential to regenerate spermatogenesis in the autologous environment of the seminiferous tubules, enabling the recipient male to father his own genetic children, possibly through normal coitus. As with hematopoiesis, large animal models that are relevant to human anatomy and physiology will be important for translating the SSC transplantation technique to the human fertility clinic. Considering the successful regeneration of spermatogenesis in the nonhuman primate model reported here and the fact that patients are already preserving testicular tissue and/or cells, clinical translation of the SSC transplantation technique appears imminent. Responsible development of the technology in a clinically relevant nonhuman primate system will help to address issues of safety and feasibility. As with hematopoiesis, the clinical significance and breadth of applications for SSC transplantation will ultimately be established in human patients.
All experiments utilizing animals were approved by the responsible Institutional Animal Care and Use Committees of Magee-Womens Research Institute and the University of Pittsburgh (Assurance #A3654-01) and the Oregon National Primate Research Center, Oregon Health & Sciences University (Assurance #A3304-01) and performed in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.
Testis tissue was collected from rhesus macaques by hemicastration or subcapsular biopsy. For biopsies, less than 30% of the testicular parenchyma was removed (3.8g-8.7g) through a transverse incision in the tunica albuginea on the lateral side of the right testis. In one case (M036), the biopsied testis was later removed by hemicastration due to formation of an abscess. Cells were recovered from testicular parenchyma usinga two-step enzymatic digestion procedure, cryopreserved and stored in liquid nitrogen, as described (Hermann et al., 2009; Hermann et al., 2007).
Recipient animals were treated with the alkylating chemotherapeutic agent busulfan (Busulfex IV; PDL BioPharma, Fremont, CA), at doses of 8, 10, 11, or 12 mg/kg (Table S2). Busulfex was diluted in physiological saline and administered intravenously at 0.6 mg/ml over 10–20 minutes.
Autologous transplants of peripheral blood stem cells (PBSCs) were employed to restore the hematopoietic system after busulfan treatment. Briefly, PBSCs were mobilized with six, daily subcutaneous injections with the cytokines G-CSF (10μg/kg/day, Neupogen; Amgen; Thousand Oaks, CA) and SCF (200μg/kg/day; Amgen) or G-CSF alone (20μg/kg/day), essentially as described (Figure 2) (Donahue et al., 2005). PBSC collections were performed by apheresis using either a Spectra or Spectra Optia apheresis device (Caridian BCT; Lakewood, CO). Twenty-four hours after apheresis, animals were treated with busulfan and 18 hours later animals received autologous PBSC transfusions (Figure 2). Two days later, animals received one subcutaneous injection of long-acting G-CSF (300μg/kg; Neulasta, Amgen). Additional details are available in Supplemental Methods.
Portions of testicular parenchyma and epididymis collected above and at necropsy were fixed with Bouin’s solution (Accustain; Sigma-Aldrich, St. Louis, MO), paraffin embedded, sectioned (5μm), and stained with hematoxylin and eosin.
Spermatogonial stem cell transplants were performed 9–15 weeks after busulfan treatment (autologous – unilateral; allogeneic - bilateral). In biopsied animals, autologous transplants were performed into the contralateral testis. Cryopreserved donor cells were recovered for transplant from storage in liquid nitrogen, as described (Hermann et al., 2009; Hermann et al., 2007). In some cases, donor cells were enriched for spermatogonia, including SSCs, on a 24% Percoll cushion (GE Healthcare Life Sciences, Piscataway, NJ) prior to transplant (see Figure S1 and Table S2). Cells were then suspended at approximately 100 × 106 cells/ml in MEMalpha medium (Invitrogen) containing 10% FBS, 20% trypan blue, 20% Optison (ultrasound contrast agent; GE Healthcare, Waukesha, WI) and 0.7mg/ml DNase I in a total volume of ≤ 1ml, depending on recipient testis size and available cells. SSC transplants were performed using ultrasound-guided rete testis injections (Figure 1 and Movie S1). For this purpose, a 13MHz linear superficial probe was used to visualize the rete testis space on a MicroMaxx ultrasound machine (Sonosite, Bothell, WA) and guide a 25Ga 2′ spinal needle into the rete testis. Cells were injected under slow constant pressure and chased with saline.
For autologous transplants, donor cells were treated with lentiviral vectors modified from the FUGW construct originally described by Lois and coworkers (Lois et al., 2002). Details of virus constructs and viral treatments are available in Supplemental Methods.
Five of six allogeneic transplant recipients were treated with human/mouse chimeric anti-CD154 IgG 5C8 (NIH Nonhuman Primate Reagent Resource, Beth Israel Deaconess Medical Center, Boston, MA) at 20mg/kg on d-1, d0, d3, d10, d18, d28 and monthly thereafter to block the T-cell co-stimulatory pathway and prevent T-cell mediated rejection of the grafted cells (Kirk et al., 1999).
Semen samples were collected from experimental animals at weekly intervals before and after busulfan treatment, as described (Gould and Mann, 1988). Total sperm count per ejaculate was determined by hemocytometer. Genomic DNA was extracted from sperm samples and assessed for donor genotype by PCR for lentivirus sequence or by microsatellite DNA fingerprinting (see Supplemental Methods).
Genomic DNA isolated from sperm of allogeneic transplants or amplified from embryos was used for microsatellite repeat fingerprinting (details in the Supplemental Methods). Primer sequences and primer concentrations in multiplexed PCR are described elsewhere (Larsen et al., 2010).
TaqMan probe-based allelic discrimination qPCR was used to detect a novel SNP in the rhesus class II major histocompatibility complex transactivator (CIITA) locus that distinguishes genomic DNA from the donor (M092) and recipient (M027) of an allogeneic SSC transplant essentially as described (Alizadeh et al., 2002; Kim et al., 2008). See Supplemental Methods for additional details.
Controlled ovarian stimulation was performed on two female rhesus macaques, as previously described (Byrne et al., 2007). Oocytes were collected and fertilized with M027 sperm by ICSI, and resulting embryos cultured as described (Hewitson et al., 1999; Mitalipov et al., 2006). Following ICSI and in vitro development, individual embryos were placed into 8ul nuclease-free water. Genomic DNA from each embryo was amplified using the WGA4 GenomePlex® Single Cell Whole Genome Amplification kit according to manufacturer recommendations (Sigma-Aldrich) and used for microsatellite DNA fingerprinting. Additional detail is available in Supplemental Methods.
We would like to acknowledge the outstanding work of Tony Battelli, Pam Wintruba and Joe Hrach, the lab animal staff at Magee-Womens Research institute, who were critical to the conduct of these experiments. We are grateful to Drs. Robert Donahue and Cynthia Dunbar of the National Heart, Lung and Blood Institute, NIH, who provided critical advice for PBSC transplantation in nonhuman primates. Drs. Tony Plant and Judy Cameron provided advice about central line catheter placement in nonhuman primates. We also thank Drs. Regina Norris, Danielle Sweeney, David Rodeberg, and Francis Schneck from the Departments of Urology and General Surgery at the Children’s Hospital of Pittsburgh of UPMC for their assistance with subcapsular testis biopsies. The FUGW lentiviral backbone was provided by Dr. Carlos Lois, University of Massachusetts. The anti-CD154 reagent used in these studies was provided by the NIH Nonhuman Primate Reagent Resource (R24 RR016001, N01 AI040101). The work was supported by Magee-Womens Research Institute and Foundation, NIH grants R01 HD055475, R21 HD061289 to KEO, U54 HD008610 to Tony M. Plant and KEO, P01 HD047675 to GPS and K99/R00 HD062687 to BPH.