The testicular soma of Tsc22d3 knockout mice supports spermatogenesis and germline transmission from spermatogonial stem cell lines upon transplantation

Abstract Spermatogonial stem cells (SSCs) are adult stem cells that are slowly cycling and self‐renewing. The pool of SSCs generates very large numbers of male gametes throughout the life of the individual. SSCs can be cultured in vitro for long periods of time, and established SSC lines can be manipulated genetically. Upon transplantation into the testes of infertile mice, long‐term cultured mouse SSCs can differentiate into fertile spermatozoa, which can give rise to live offspring. Here, we show that the testicular soma of mice with a conditional knockout (conKO) in the X‐linked gene Tsc22d3 supports spermatogenesis and germline transmission from cultured mouse SSCs upon transplantation. Infertile males were produced by crossing homozygous Tsc22d3 floxed females with homozygous ROSA26‐Cre males. We obtained 96 live offspring from six long‐term cultured SSC lines with the aid of intracytoplasmic sperm injection. We advocate the further optimization of Tsc22d3‐conKO males as recipients for testis transplantation of SSC lines.


| INTRODUCTION
Spermatogonial stem cells (SSCs) are adult stem cells that continuously undergo self-renewal to maintain the undifferentiated state, and that differentiate to form mature spermatozoa throughout the lifetime of males (Kanatsu- Shinohara & Shinohara, 2013). Theoretically, a single SSC yields~4,096 haploid gametes by passing through successive mitoses and one meiosis-a process that takes~35 days in mouse.
SSCs reside on the basement membrane of the seminiferous tubules in the testis. A specialized microenvironment that is termed a niche supports the self-renewal and differentation of SSCs (Oatley & Brinster, 2012). The widely used biological assay to assess SSC activity within a population of cells entails the transplantation of SSCs by microinjection into the testes of infertile mice (Brinster & Avarbock, 1994;Brinster & Zimmermann, 1994). Transplanted cells give rise to fertile spermatozoa and then to donor cell-derived offspring either by natural mating or with the aid of an assisted reproductive technology such as intracytoplasmic sperm cell injection (ICSI). Long-term culture of mouse SSCs became possible by adding to the medium selfrenewal factors such as glial cell line-derived neurotrophic factor (GDNF; Kubota, Avarbock, & Brinster, 2004). It is estimated that only a small fraction (<1%) of the cells of an SSC line harbor the stem cell potential (Kanatsu-Shinohara & Shinohara, 2013). Genetic manipulation of the genome of SSC lines has enabled the generation of genetically modified strains of mice (Kanatsu-Shinohara et al., 2006;Sato et al., 2015;Wu et al., 2015) and rats (Chapman et al., 2015), but has not replaced gene targeting in embryonic stem cells.
There are two commonly used types of infertile recipients for SSC transplantation in mice. The first type is the Kit W /Kit Wv mutant mouse, which has smaller testes that contain almost no germ cells and are devoid of spermatogenesis. But because Kit W /Kit W homozygous mice die after birth, compound heterozygous Kit W /Kit Wv males must be generated from crosses of mice carrying the Kit W or Kit Wv alleles in the heterozygous state. The second type is obtained by intraperitoneal injection of busulfan, an alkylating chemotherapeutic agent that preferentially kills spermatogonial stem cells. But the typical dose of 40 mg busulfan/kg body weight results not only in infertility, but also in substantial morbidity and mortality. In the few studies that report mortality rates, these ranged from 30% (Ma, Wang, Gao, & Jia, 2018), 31.6% (Qin et al., 2016), 60% (Ganguli et al., 2016), to as high as 87% (Wang, Zhou, Yuan, & Zheng, 2010) for 40 mg busulfan/kg body weight. In certain studies, busulfan-treated mice received a bone marrow transplant to relieve hematopoietic suppression (Aoshima, Baba, Makino, & Okada, 2013;Kanatsu-Shinohara et al., 2002Ogawa, Dobrinski, Avarbock, & Brinster, 1999). We believe that, in the interest of animal welfare and the 3R principle of humane experimental technique with animals (Russell & Burch, 1959), it is imperative to explore alternative candidates for SSC transplantation recipients as they become available through new discoveries or new technological developments.
A novel candidate recipient that has emerged recently but has not been explored yet is the genetically infertile male mouse that is the centerpiece of the goGermline technology (Koentgen et al., 2016). This technology was originally developed to produce chimeric mice that give 100% germline transmission of the embryonic stem cell-derived genome upon microinjection of embryonic stem cells into blastocysts or eight-cell embryos. Infertile males are produced by crossing two strains that can be maintained in the homozygous and hemizygous state and that are healthy and fertile: females homozygous for a Tsc22d3 conditional (floxed) gene-targeted mutation are crossed with males homozygous for a ROSA26-Cre gene-targeted mutation. There is no mortality at any stage; there is no genotyping of any mouse needed; and 100% of male offspring of the cross are Tsc22d3-conKO and infertile. The goGermline technology appeared to us as promising to fulfill the reduction imperative of the 3R principle as source of an alternative, third-type of recipient for SSC transplantation. Here, we report spermatogenesis and germline transmission of the SSC-derived genome upon transplantation into the testes of infertile Tsc22d3-conKO males.
The cauda epididymis was washed twice with Dulbecco's phosphate buffered saline (DPBS, Gibco, #14190094), then directly put in 1 mL of EmbryoMax Human Tubal Fluid (Merck Millipore, #MR-070-D). Cell suspensions were exposed to ultrasound for 1-2 min. Spermatozoa without tail were picked up into a blunt piezo-driven pipette with a tip of 10-15 μm diameter. A single sperm head was injected into a single oocyte in a droplet of M2 medium (Sigma, #M7167) containing 5 μg/mL cytochalasin B (Sigma, #C6762) using a pipette with a tip of 10-15 μm diameter, and a piezo micromanipulator controller (Japan Prime Tech, #PMAS-CT150). Injected oocytes were maintained in KSOM medium (Merck Millipore, #MR-106-D) at 37 C with 5% CO 2 in air. Two-cell embryos were transferred into the oviducts of pseudopregnant ICR females. Offspring were born on day 19.5 of gestation.

| Testes of Tsc22d3-conKO mice after SSC transplantation
We transplanted seven SSC lines into the testes of Tsc22d3-conKO mice ( Figure 3a, Table 1), 10 mice per SSC line, for a total of 70 transplanted mice. Five SSC lines were derived from GFP+ mice:

| Germline transmission from transplanted SSCs
Males were mated with C57BL/6J females 2 months after transplantation, but no offspring was obtained after 5-14 months. We proceeded to apply the assisted reproduction technique ICSI, injecting sperm without tail into BDF1 (C57BL/6J x DBA/2) oocytes or C57BL/6J oocytes (Figure 5a). Zygotes developed into 2-cell embryos after 12-24 hr culture in KSOM medium (Figure 5b). Following transplantation of 2-cell embryos into oviducts of pseudopregnant ICR females, we obtained a total of 96 live pups: 79 pups using BDF1 oocytes and 17 pups using C57BL/6J oocytes ( Table 2). The birth rate for ICSI using C57BL/6J oocytes is known to be very low (Sakamoto, Kaneko, & Nakagata, 2005). Pups exposed to UV light displayed the intrinsic green fluorescence of GFP in a manner that is consistent with the provenance of the SSC line from a heterozygous mouse

| DISCUSSION
Spermatogonial stem cells (SSCs) continuously undergo self-renewal to maintain the undifferentiated state, and differentiate to produce eventually spermatozoa, which transmit genetic information to the next generation (Oatley & Brinster, 2012 F I G U R E 4 Spermatogenesis in testes of Tsc22d3-conKO mice after SSC transplantation. (a) Immunofluorescence for SCP3 in a cryosection of a Tsc22d3-conKO testis after transplantation of HZ-6 (derived from C57BL/6J). DAPI (white) serves as nuclear stain. (b) Immunofluorescence for SCP3, GFP, and TRA98 in a cryosection of a Tsc22d3-conKO testis after transplantation of HZ-9 (derived from ROSA26-EGFP). DAPI (white) serves as nuclear stain. (c) Immunofluorescence for GFP, TNP1, TRA98, and TRIM36 in cryosections of a Tsc22d3-conKO testis after transplantation of HZ-9 (derived from ROSA26-EGFP). DAPI (white) serves as nuclear stain (d) Immunofluorescence for MVH in a cryosection of a Tsc22d3-conKO testis after transplantation of HZ-7 (derived from C57BL/6J). DAPI (white) serves as nuclear stain.
(e) Immunofluorescence for GFP and MVH in a cryosection of a Tsc22d3-conKO testis after transplantation of HZ-11 (derived from B6-GFP). DAPI (white) serves as nuclear stain. Scale bars, 50 μm in (a-c) and 20 μm in (d, e) be applied to treat male infertility (Ogawa, Dobrinski, Avarbock, & Brinster, 2000); can be used to produce transgenic animals (Chapman et al., 2015;Sato et al., 2015;Wu et al., 2015); and is the most stringent functional assay to assess SSC activity. After transplantation into the seminiferous tubules, SSCs pass through the blood testis barrier that is formed by tight, adherens and gap junctions between adjacent Sertoli cells; a fraction of the transplanted SSCs migrate to the basement membrane of the seminiferous tubules; and some cells complete the process of spermatogenesis (Nagano, Avarbock, & Brinster, 1999).
Recipients for SSC transplantation have been prepared or bred in several ways over the past decades: by testicular irradiation (Withers, Hunter, Barkley, & Reid, 1974;Zhang, Shao, & Meistrich, 2006), by cooling the testes (Ehmcke, Joshi, Hergenrother, & Schlatt, 2007;Young et al., 1988;Zhang et al., 2004), by heat shock treatment (Ma et al., 2011), by experimental cryptorchidism (Mendis-Handagama, Kerr, & de Kretser, 1990), by injection of the chemotherapeutic drug busulfan (Brinster & Avarbock, 1994;Brinster & Zimmermann, 1994;Bucci & Meistrich, 1987), and by breeding Kit W /Kit Wv compound heterozygous mice (Brinster & Avarbock, 1994;Brinster & Zimmermann, 1994). A quarter of a century after the first reports of busulfan-treated mice and Kit W /Kit Wv compound heterozygous mice as recipients for spermatogonial transplantation (Brinster & Avarbock, 1994;Brinster & Zimmermann, 1994), they remain the most widely used types of recipients. But disadvantages include the morbidity and mortality (busulfantreated mice), and the inefficient generation of recipients by breeding (only 25% of male offspring from heterozygous parents is Kit W /Kit Wv compound heterozygous). There is thus opportunity for improvement in identifying and optimizing a novel type of recipient that is devoid of morbidity or mortality and that can be bred efficiently in a single cross and without genotyping.
GILZ was originally discovered as an anti-inflammatory protein that is involved in the immunosuppressive effects of glucocorticoids (D'Adamio et al., 1997). The mouse GILZ protein is encoded by the Tsc22d3 gene, which is located on the X chromosome. Males hemizygous for a Tsc22d3 knockout were unexpectedly found to be infertile F I G U R E 5 Offspring generated by ICSI with sperm from testes of Tsc22d3-conKO mice after SSC transplantation. (a) Experimental strategy. Cultured SSCs were transplanted into the testes of Tsc22d3-conKO mice. Two months later, epididymis spermatozoa were isolated and exposed to ultrasound, and sperm heads were injected into oocytes by ICSI. Following 2-cell embryo transfer, live offspring were obtained. (b) Brightfield image of 2-cell embryos after ICSI into BDF1 oocytes with sperm from a Tsc22d3-conKO male transplanted with HZ-9 (derived from ROSA26-EGFP). Scale bar, 50 μm. (c) Pups exposed to UV light. Left panel, pups obtained with HZ-9, which is hemizygous for the targeted mutation in the ROSA26 locus. Four out of 10 pups display the intrinsic green fluorescence of GFP. Middle panel, pups obtained with HZ-3 (derived from Tg(act-EGFP)), which is homozygous for the transgene. All pups express the intrinsic green fluorescence of GFP. Right panel, pups obtained with HZ-6 (derived from C57BL/6J). No pups express the intrinsic green fluorescence of GFP  (Bruscoli et al., 2012;Romero et al., 2012;Suarez et al., 2012). It is not possible to generate homozygous Tsc22d3-KO females by breeding . Hemizygous Tsc22d3-KO males must be bred by crossing heterozygous females with wild-type males, and 50% of the male offspring of such crosses are hemizygous and infertile. The defect in the germline appears to be intrinsic, as spermatogenesis can be restored in Tsc22d3-KO males by transplantation of freshly prepared wild-type germ cells (Bruscoli et al., 2012). But germline transmission of the donor haplotype, either by natural mating or with the aid of ICSI, remained to be shown for Tsc22d3-KO mice.
Our strategy of the goGermline technology consists of mating homozygous Tsc22d3 floxed females with homozygous ROSA26-Cre males, producing 100% male mice that are Tsc22d3-conKO and infertile (Koentgen et al., 2016). Upon microinjection of embryonic stem cells in blastocysts or eight-cell embryos generated in this cross, chimeras can be generated that yield 100% germline transmission of the embryonic stem-cell derived genome (Koentgen et al., 2016). We have now evaluated Tsc22d3-conKO males as recipients for SSC transplantation. Our newly derived SSC lines and the recipient males are in an inbred C57BL/6J background, and are thus immunologically fully compatible. We were able to generate live offspring carrying the donor haplotype from several SSC lines with the aid of ICSI.
The mechanisms of infertility of Tsc22d3-KO mice have been studied by several groups and in several strains (Bruscoli et al., 2012;La et al., 2018;Ngo, Cheng, et al., 2013;Romero et al., 2012;Suarez et al., 2012). The phenotype involves an arrest midway through the pachytene of meiosis I, massive apoptosis, SSC exhaustion, resulting in a progressive depletion of the germline, and terminating in a Sertoli cell-only phenotype. Our contribution here is to demonstrate that the testicular soma of adult Tsc22d3-conKO mice supports spermatogenesis and germline transmission upon transplantation of established, long-term cultured SSC lines. The proof of principle that we have delivered paves the way for optimization of Tsc22d3-conKO mice as a third type of recipient for SSC transplantation-a type of recipient that is devoid of morbidity and mortality and that can be bred efficiently and without genotyping.
ACKNOWLEDGMENTS P.M. is grateful to the Max Planck Society for generous support.
Ozgene was supported by the R&D tax incentive program of the Australian Taxation Office and AusIndustry. F.K. is an employee of Ozgene Pty Ltd.