Primary culture of germ cells that portray stem cell characteristics and recipient preparation for autologous transplantation in the rhesus monkey

Abstract Fertility preservation for prepubertal cancer patients prior to oncologic treatment is an emerging issue, and non‐human primates are considered to constitute suitable models due to the limited availability of human testicular tissues. However, the feasibility of spermatogonial stem cell (SSC) propagation in vitro and autologous testicular germ cell transplantation in vivo requires further exploration in monkeys. Herein, we characterized germ cells in macaque testes at 6 months (M), 18 M and 60 M of age, and effectively isolated the spermatogenic cells (including the spermatogonia) from macaque testes with high purity (over 80%) using combined approaches of STA‐PUT separation, Percoll gradients and differential plating. We also generated recipient monkey testes with ablated endogenous spermatogenesis using the alkylating agent busulfan in six macaques, and successfully mimicked autologous cell transplantation in the testes under ultrasonographic guidance. The use of trypan blue led to successful intratubular injection in 4 of 4 testes. Although SSCs in culture showed no significant propagation, we were able to maintain monkey testicular germ cells with stem cell characteristics for up to 3 weeks. Collectively, these data provided meaningful information for future fertility preservation and SSC studies on both non‐human primates and humans.

preservation strategy is the cryopreservation of immature testicular biopsy, since the biopsy pieces containing spermatogonial stem cells (SSCs) could restore fertility in the future through tissue grafting or in vitro spermatogenesis. 3 Following the successful transplantation of SSCs and subsequent fertility restoration in mice, 4,5 many similar attempts have been made in other large animal species over recent decades. [6][7][8] However, novel methods are now under investigation for establishing a culture system that expands the human SSC population in vitro, and that allows transplantation to the testes to be more feasible and efficient following cancer treatment and cure. [9][10][11] SSCs possess the properties of self-renewal and differentiation that allow them to maintain the stem cell pool while continuously producing sperm throughout adult life. 12 The development of SSCs into spermatozoa (termed spermatogenesis) is a strictly regulated process that involves the mitosis of spermatogonia, meiotic division of spermatocytes, and spermiogenesis. 13 Studies of SSC biology in rodents have recently significantly improved our understanding of the mechanisms underlying spermatogenesis 14,15 owing to the achievements of in vivo and in vitro methods and techniques of SSC manipulation; these include the establishment of long-term SSC culture systems. 16 and SSC transplantation technology. 4 However, these methods and techniques in large animals-including human and primates-remain relatively limited. Spermatogenesis is, however, highly conserved in mammals.
It has been shown that gene activity in mouse gonocytes (the precursor cells of SSCs) and prepubertal human spermatogonia shares great similarity. 15 There are also many disparities in the spermatogenic lineages between primates and rodents, as the two species diverged phylogenetically at least 75 million years ago. 17 including the classification of germ cell subtypes and the kinetics of spermatogenesis. 18 Since non-human primates (NHPs)-in particular the rhesus monkey-are highly comparable to humans in terms of testicular physiology, NHPs have been applied as the most suitable preclinical models for human germ-cell transplantation. 19 Several approaches have been developed for the characterization of germ cell populations-including magnetic-activated cell sorting (MACS), 20 fluorescence-activated cell sorting (FACS) with fluorescent dyes, 21,22 and STA-PUT velocity sedimentation, 23 and these represent currently popular methodologies to separate various testicular cells. Cell sorting using FACS has been successfully performed for the isolation of highly purified germ cells in multiple species. 24 The STA-PUT method-while unsuitable for the enrichment of spermatocytes and spermatids due to the lack of established surface biomarkers in primates-holds several advantages over FACS, including larger yields of cells per testis and higher cell viability, thus making the isolated cells obtained from STA-PUT more feasible for culture. 25 Using STA-PUT, research groups have achieved the successful isolation of spermatogenic cells with high purity and viability in mice, 23 cattle 26 and humans. 27 And these cells can be used for numerous analyses-including changes in gene expression, 28 nucleosomal dynamics, 29 chromatin remodeling, 30 and other dynamic aspects of germ cells during spermatogenesis such as non-coding RNAs and the use of protein profiling. [31][32][33] In the present study, we reported our approaches to purifying testicular germ cells in rhesus monkeys by STA-PUT, and also demonstrated the primary culture of germ cells with stem cell characteristics-as well as recipient preparation and attempts at testicular transplantation of germ cells using the rhesus monkey model.

| Ultrasound-guided injections of the rete testis
All ultrasonographic measurements were performed with a 12.0 MHz linear superficial probe. Each monkey was anaesthetized and placed on an operating table, lying on its back, and ultrasound transmission gel was applied to its scrotum. The rete testis showed high echogenicity under ultrasonographic examination and could therefore be visualized from systematic longitudinal and transverse scans. The injection was performed from the lower part of the testis under continuous ultrasonographic monitoring, requiring approximately 30 min. Finally, the injected testes (n = 4) were collected and further bisected to evaluate the effect of transplantation.

| Testis histology and immunofluorescence
For histology, testicular tissues were surgically removed and fixed with Hartman's Fixative (Cat#H0290-500ML, Sigma, USA) for 24 h at room temperature. Tissues were dehydrated, embedded in paraffin and sectioned at 5 μm. Sections were then de-paraffinized and stained with haematoxylin (Cat#HHS16, Sigma) and eosin (Cat# E607321-0100, Sangon Biotech) (H&E). For tissue immunofluorescence, frozen testis sections were prepared and washed three times with PBS, and blocked with 1% BSA and 0.1% Triton X-100 for 1 h at room temperature. Slides were incubated with primary antibod- For cellular immunofluorescence, cells were fixed with 2% PFA (Cat#P6148, Sigma) at 4°C overnight, and after washing 3 times with PBS, slides were blocked with 10% goat serum containing Triton X-100 for 1 h at room temperature. Slides were then washed with PBS and incubated with primary antibodies for 2 h and secondary antibodies for 1 h in the dark. Images were collected through a Zeiss microscope as described above.

| Preparation of testicular cell suspensions
Testis tissues from each rhesus monkey were minced and digested

| STA-PUT
STA-PUT was performed as described previously with some modifications. 25 Testes from 2 adult (60-M-old monkey, 60 M) macaques were used for separation of pachytene spermatocytes, round spermatids and elongated spermatids, and testes from 5 neonatal macaques (6-M-old monkey, 6 M) were used for separation of spermatogonia. We added 800 ml (in the case of the 60-M-old rhesus monkey) of DMEM containing 4% BSA into chamber a, and 800 ml (60-M-old monkey) or 300 ml (6-M-old monkey) of DMEM containing 2% BSA into chamber b ( Figure 1A). Germ cells (in total of 5 × 10 8 cells for 800 ml and 1 × 10 8 cells for 300 ml) were resuspended in 25 ml of wash buffer and allowed to slowly flow into the sedimentation chamber after addition to the cell chamber. When the cells were all in the sedimentation chamber, we opened the clamps to let 2% and 4% BSA flow into the sedimentation chamber at a rate of 45 ml min −1 . Once the BSA was in the sedimentation chamber, we allowed the cells to sediment for 3 h (60-M-old monkey) or 1.5 h (6-M monkey). We collected cells at a rate of 10 ml/48 s, and each collection tube was filled with 10 ml of cell suspension.

| Purification of spermatogonia
The spermatogonial population that we collected from STA-PUT was resuspended in 2 ml of DMEM. A gradient of Percoll was then prepared in a 15-ml tube by consecutively adding 0.5 ml of 45% Percoll, 2 ml of 35% Percoll, and 2 ml of 20% Percoll. We slowly added the cell suspension along the wall of the tube to the Percoll gradient, and centrifuged it at 600 g for 20 min at 4°C. We subsequently collected the phase with enriched spermatogonia and repeated the Percoll purification.

| Establishment of transplantation platform
Recipient preparation was performed as described previously. 6 Briefly, adult rhesus monkeys (n = 6) were administered granulocyte colony-stimulating factor (G-CSF, Amgen) at a dosage of 20 μg/kg/ day for 6 days by subcutaneous injection. Twenty-four hours after the collection of peripheral blood stem cells (PBSCs), rhesus monkeys were given 12 mg kg −1 of busulfan intravenously (Busulfex IV, PDL BioPharma), and 18 h later PBSCs were administered subcutaneously. Two days later, G-CSF (300 μg kg −1 ) was injected intravenously into the rhesus monkeys. Trypan blue transplantation was performed 10-12 weeks after busulfan treatment, and 1 ml of contrast medium and 0.7 ml of trypan blue solution were injected into the tubules of the monkey testes.

| qPCR analysis
The cultured cells originally from one of the neonatal (6 M) monkey testis isolation were used for qPCR analysis. Total RNA was extracted from cell culture wells. The reverse transcription was performed using a PrimeScript RT reagent kit (Cat#DRR036A, Takara). qPCR was then performed using SYBR Premix Ex Taq II (Cat#DRR820A, Takara). Gene primers are shown in Table 1.

| Characterization of germ cells in monkey testes
To discern this, we selected monkeys at three ages: neonatal spermatogonia in multiple species. 35,36 As shown in Figure 2B, UCHL1 marked spermatogonia along the basement membrane of the seminiferous tubules in testes from monkeys at all three ages. We demonstrated that the phosphorylated form of H2AX (γH2AX)-a marker for DNA double-strand breaks that is generally accepted in the identification of pachytene spermatocytes in the testes 37 -was expressed in the seminiferous tubules of 60-M-old monkey testes ( Figure 2C), but that it was not observed in either neonatal or juvenile monkey testes; this suggested to us that the middle or late stages of spermatogenesis (including meiosis) were not initiated by 18 months after birth. Since PNA (peanut agglutinin) exclusively binds to the outer acrosomal membrane of spermatids as a marker for spermatids, 38 we next demonstrated notable expression of PNA only around the lumen of the seminiferous tubules in the adult testes ( Figure 2D). Control stainings for immunofluorescence analysis were shown in Figure S1. We also characterized the germ cells at the developmental stages of testes following chronologic growth in monkeys; specifically, that germ cells in testes from monkeys at 6 M and 18 M of age had not yet entered meiosis, confirming the presence of undifferentiated or differentiating spermatogonia. In contrast, the seminiferous tubules of adult testes exhibited complete spermatogenesis with spermatogenic cells at all stages. These data provided the information needed for proper timing to determine which chronologic age would be feasible for the separation of spermatogenic cells from the testes.  1N). First, testicular single-cell suspensions were generated from a two-step enzymatic digestion and subjected to the STA-PUT separator (see Methods for details, Figure 1A). The viability of all cell types obtained was above 95% as evaluated by trypan blue exclusion. Next, we examined the isolated cells by differential interference contrast (DIC) microscopy ( Figure 1F) and immunostaining ( Figure 1G) for verification of purity.

TA B L E 1 Primer sequences of genes used for RT-PCR
Gross cellular morphology appeared to be identical by DIC microscopy, and we further confirmed the purity using marker proteins as shown in Figure 1G. SPG was defined by the expression of UCHL1, and PAC was identified by γH2AX. Both RS and ES were confirmed by PNA staining on the acrosomes ( Figure 1G). In general, the purity of harvested SPG was 81.6+/-1.8%, PAC was 86.6+/-1.0%, RS was 88.6+/-2.9%, and ES was 86.7%+-4.0% using our present approaches Figure 1H.

| Culture of monkey germ cells with stem cell characteristics
Long-term SSC propagation in vitro has been successfully achieved in rodents; however, reproducible culture systems in large animals are still under investigation. We thus explored primary cultures of monkey germ cells following the separation of spermatogonia in testes from neonatal monkeys. Testes from 5 neonatal (6 M) macaques were used for the cell culture study. We first generated a testis fibroblast feeder layer by differential plating with medium containing 7% fetal bovine serum. The identity of fibroblast cells was confirmed by cellular morphology and staining with the intermediate filament protein vimentin, a marker for testicular somatic cells ( Figure 3A).
Control staining for immunofluorescence analysis was shown in Figure S1. When cell growth reached approximately 90% confluency in wells of the plates, the fibroblasts were treated with mitomycin C to inactivate cell division before applying them to germ cell culture as the feeder layer ( Figure 3A). Testis germ cells were harvested from neonatal monkey testes (representing UCHL1-positive spermatogonia in vivo) and plated on 0.1% gelatin-coated culture wells for 24 h to remove possibly adherent cells. Cells were then subjected to further culture on the fibroblast feeder layer with a defined serum-free SSC culture medium supplemented with a combination of the growth factors GDNF, GFR alpha1, and FGF2 in the medium, as previously applied to rodent SSC culture. 14,39 Identical to initial cultures of mouse or rat SSCs, germ cells on the plates formed the typical grape-like cell clumps in the wells, and these clumps could be successfully maintained for 14 days after initial seeding ( Figure 3B).
However, the clumps gradually detached from the feeders and were often lost by medium changes at intervals of 2-3 days, and both the size and number of clumps were significantly decreased following 21 days in culture medium ( Figure 3B), even when a few cell clumps were still found after 3 weeks of culture.
To characterize the cultured cells, we collected an aliquot of cells (an undifferentiated-spermatogonia marker gene) on day 9, we found that they were comparable to the levels on day 1. In contrast, Fgfr3 (fibroblast growth factor receptor 3), a marker gene that reflects the stem cell status of spermatogonia in humans, 40,41 was significantly reduced, and Utf1 (undifferentiated embryonic cell transcription factor 1), a marker gene that reflects SSC differentiation status in humans, 42 was significantly upregulated by our 9-day culture ( Figure 3E).

| Ablation of endogenous spermatogenesis in testes of recipient monkeys
Most previous large-animal models of transplantation have entailed the use of radiation to deplete endogenous spermatogenesis, [43][44][45] as considerable injury to the haematopoietic system is provoked by the toxicity of alkylating agents; however, alkylsulfonate busulfan is favourable to the establishment of testicular recipient models for cellular transplantation in mice 46 and monkeys. 47 We therefore  Figure 4B). This evidence suggested that a majority of spermatogonia-including SSCs-were successfully depleted by busulfan treatment in monkey testes.

| Ultrasound-guided transplantation into recipient monkey testes
In contrast to mouse SSC transplantation in which surgery is typical, we employed the ultrasound-guided rete testis-injection method pioneered by Schlatt and colleagues, which generates less physical damage (with reduced invasiveness) and greater efficiency in cell injection. 45 To pave a platform for the SSC transplantation to rhesus monkey recipients, we explored the feasibility of an injection of trypan-blue saline solution into the seminiferous tubules by following the protocols reported by Hermann et al. 6 Under the guidance of ultrasonographic imaging, the injection needle was observed to puncture through the scrotal skin and enter the rete testis

| DISCUSS ION
In recent decades, much progress has been made in the study of to humans in genome as well as testicular structure and physiology. 18 Therefore, it is necessary to establish the methods for the isolation of spermatogenic cells of multiple stages from monkey testes, and developing such methods should allow us to answer questions regarding primate spermatogenesis; for example, dissecting the chromatin dynamics and molecular reprogramming of chromatin structure from diploid to haploid cells. 30 The STA-PUT method has been applied to the isolation of mouse testis cells 31 and has also been implemented in the isolation of human testicular cells. 48 While we showed that the yield from adult monkey testes of pachytene spermatocytes and other late-stage spermatids was of high purity, we did not achieve the same result with spermatogonia because of their similarity in size to somatic cells and other uncharacterized germ cells. Magnetic separation (MACS) was reported to be one way to isolate spermatogonial populations (including SSCs) in mice and humans via conjugation to highly specific antibodies against a particular surface antigen such as CD90.2 49 or GPR125. 50 Unfortunately, neither marker was applicable to spermatogonial isolation from rhesus monkey testes in our previous attempts; nonetheless, a combination of STA-PUT, Percoll-gradient separation and differential plating significantly improved the purity of monkey spermatogonia from the testes of a 6-M-old prepubertal male ( Figure 1H).
The recovery of SSCs from prepubertal human testes via clinical biopsy samples alone is unlikely to prove satisfactory for the restoration of fertility following autologous transplantation since the number of Ad and Ap spermatogonia with stem cell potential in this population is likely to be extremely low. 18 SSCs, 10 while protocol feasibility has been rarely established, and existing primate SSC culture methods under various conditions were maintained for a short time (less than 2 weeks) in other reports. 55,56 In the present study, we also explored the possibility of in vitro culture of germ cells with stem cell identities. We were able to maintain these clump-forming germ cells for 21 days, and the cultured germ cells were positive for the undifferentiated spermatogonial stem cell and germ cell markers FGFR3, UTF1, DAZL and ZBTB16, which was identical to the data reported by Sharma and colleagues. 57 However, significant cellular proliferation did not occur following culture ( Figure 4). We strongly suggest that the prerequisites of mouse SSC  43 The same laboratory also showed successful autologous SSC transplantation that produced spermatogenesis by busulfan-treated macaques. 47 These advances suggested that improvements in transplantation methods and well-established recipient preparation for autologous transplantation using primate models are required for future preclinical trials of human germ cell transplantation and fertility rescue. We followed the protocols used by Orwig and associates and evaluated germ cell deletion in the recipient testes of 60-M-old macaques treated with busulfan, and found that endogenous spermatogenesis was successfully depleted, as demonstrated by degenerated seminiferous tubules; however, we also (rarely) observed UCHL1-positive spermatogonia in the seminiferous tubules of monkey testes 12 weeks after treatment, which demonstrated reliable recipient preparation for cell transplantation. In addition, the successful intra-testicular cellular transplantation in monkeys needs to be completed under the guidance of ultrasonography, which is slightly different from mouse testicular cell-transplantation technology. Nevertheless, our approach has been proven feasible and reliable using testes from adult monkeys treated for 12 weeks with busulfan.
To mimic the clinic steps for fertility recovery, two prerequisites are not well achieved. Firstly, genetic modification is not yet feasible, while the second concern is the surgical procedure. In the present study, we have provided information on the developmental characterization of germ cells in the testes, the culture of spermatogonia, recipient preparation, and transplantation trials in the rhesus monkey. Collectively, these are constructive data that will allow the further study of SSCs in primates.

ACK N OWLED G EM ENTS
We

CO N FLI C T O F I NTE R E S T
The authors declared no potential conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request.