The architecture of the testes is characterized by two structurally distinct compartments (Figure 1), the seminiferous tubule and the interstitial tissue (27). Within the seminiferous tubule, Sertoli cells form a tight blood-testis barrier to divide the seminiferous epithelium into basal and luminal compartments (28). Developing spermatogonia reside on the basal membrane and are further defined by three types of cells: undifferentiated spermatogonia (quiescent SSCs), differentiating spermatogonia (SSCs that undergo active mitosis), and differentiated spermatogonia (29, 30). The Sertoli cells are the supporting cells for the germ cell population in the testes and are essential for maintaining normal spermatogenesis by providing the cellular matrix and by secreting specific growth factors (31). The surrounding interstitial space consists of various cell types that include the Leydig cells, mesenchymal cells, and immune cells, in addition to lymph vessels, nerve fibers, and connective tissues (27). Leydig cells produce the hormone testosterone and cytokines that may function both directly and indirectly to regulate self-renewal of SSCs (32).

The fine-tuned balance between self-renewal and differentiation of SSCs is regulated by the interplay of extrinsic and intrinsic factors. GDNF, a growth factor produced by the somatic niche cells, is critical for the maintenance of SSCs both in vivo and in vitro (33). It regulates several essential downstream genes, including the germ cell specific and ubiquitously-expressed genes Nanos2, Etv5, Lhx1, T, Bcl6b, Id1, and Cxcr4, to promote SSC self-renewal and inhibit differentiation (34–39). CXCL12/CXCR4 (39), FGFs (33, 40), and VEGF-A (41) act in synergy with GDNF to maintain SSC stem cell status. In contrast, retinoic acid (RA), a hormone secreted primarily by Sertoli cells, plays an indispensable role in inducing differentiation of SSCs by downregulation of GDNF expression and activation of differentiation-promoting factors, such as BMP and SCF (42–45). Genetic ablation studies in mice indicate that several transcription factors are involved in regulating SSC maintenance and recruitment to spermatogenesis. The PLZF transcription factor is expressed by SSCs and interacts with GDNF signaling as one of the master regulators to promote the self-renewal of SSCs (46, 47). Loss of PLZF results in progressive germ cell loss, testicular hypoplasia, and infertility (46–48). One of the downstream targets of PLZF is the SALL4 protein, which is required for the self-renewal of SSCs and maintenance of ability to enter spermatogenic differentiation (49). A potential upstream regulator of PLZF is PRMT5. Disruption of the PRMT5 gene results in a dramatic reduction of PLZF gene expression, and subsequent progressive loss of SSCs leading to male infertility (50). Another transcription factor important for maintenance of SSC self-renewal is FOXO1, which regulates a number of genes that are preferably expressed in SSCs (51). Deletion of the FOXO1 gene results in defects in SSC maintenance and ultimately spermatogenic failure. In addition, recent research has identified numerous microRNAs as critical regulators in spermatogenesis. Some microRNAs regulate the self-renewal of SSCs. For example, miR-202 plays a crucial role in the maintenance of SSC stemness or self-renewal of the stem cell population (52). Other microRNAs, such as miR-1908-3p (53), miRNA-122-5p (54), and miRNA-31-5p (55), enhance the proliferation and inhibit the early apoptosis of human SSCs via targeting key downstream pathways. Conversely, several microRNAs facilitate differentiation via regulation of the expression of genes associated with SSC differentiation. MiR-34c promotes SSC differentiation by inhibiting the function of the NANOS2 gene, leading to the up-regulation of meiotic-related proteins, STRA8, in mice (56). Similarly, miR-486-5p secreted by Sertoli cells stimulates differentiation of SSCs in mice by up-regulating the expression of STRA8 and SYCP3 (57). Further, impaired spermatogenesis is observed in mice carrying a deficiency in miR-17-92 or a gene deletion of miR-17-92 (58, 59). miR-202 similarly regulates spermatogenesis via orchestration meiotic initiation by preventing precocious differentiation of mouse SSCs (52). Taken together, numerous genes act to balance self-renewal and differentiation of SSCs.

Autotransplantation of testicular tissues has achieved success in multiple animal models, which results in live offspring (61–65). However, the approach has the risk of re-introducing malignancy is a concern (66). Studies of xenotransplantation, which transplants immature testicular tissue under the back skin of immune-deficient animals, have been used to examine potential complications including malignancy. In 2002, Nagano and colleagues, for example, transplanted human SSCs into immunodeficient mice for the first time (67). Human SSCs survived in mouse testes for at least six months and proliferated during the first month after transplantation.

To avoid potential complications of malignancy, isolation of SSCs from cryopreserved testicular tissues followed by transplantation has been proposed as the leading alternative stratgey. To separate SSCs from somatic cells, antibodies that recognize human SSC-specific proteins are used for FACS (fluorescent-activated cell sorting) or MACS (magnetic-activated cell sorting) for sorting SSCs from other cell types. Antibodies that have been shown to be useful for sorting SSCs include GFRα (68), GPR125, ID4 (69), ITGA6 (70), SSEA4 (71), PLPPR3 (72), and OCT4 (73). An alternative to cell sorting is to take advantage of different physical properties between SSCs and somatic cells such as velocity sedimentation and differential affinity to extracellular matrices on the culture plate (74–78). Once isolated, SSCs are cultured with growth factors shown to be optimal or essential for SSC maintenance [GDNF, BFGF, EGF, and LIF (79–81)].

A major limitation of SSC transplantation in vivo, for fertility restoration in clinical practice, is the scarcity of SSCs within the testicular tissue. This has necessitated exploration of alternatives including the establishment of a robust in vitro culture system to maintain and expand human SSCs. Extensive effort has been focused on optimization of culture conditions for long-term maintenance and propagation of human SSCs. Multiple culture substrates, including hydrogel, matrigel, and laminin, have been shown to promote the propagation of human SSCs under feeder-free conditions (82). Currently, several markers are used for the verification of human SSCs. However, many of these markers are also expressed in testicular somatic cells. For example, UCHL1, which was used to identify SSCs from humans, is also expressed in Leydig cells and nerve fibers (83). The most stringent assay to assess the function of SSCs is to generate offspring after homologous transplantation. However, despite success in animal models, including non-human primates, no studies are reporting the generation of human functional spermatozoa following autotransplantation or xenotransplantation of testicular tissue or isolated human SSCs for fertility restoration.

Since 1959, a gas-liquid interface was used to culture testicular fragments of the adult rats (84). In this culture system, the differentiation of SSCs was limited up to pachytene spermatocytes (85). In 2003, round spermatids were observed after two weeks of culture in a gas-liquid interface culture system (86). Several other organotypic culture systems have been developed to recapitulate the entire process of spermatogenesis in vitro. One of the breakthroughs in the research was reported in 2011 with the demonstration of live offspring that were generated from in vitro-produced haploid germ cells (87). In this study, testicular tissue fragments from neonatal mice were cultured on an agarose gel-based organ culture system. Subsequently, microfluidic technology was adopted for organ culture, with the goal of providing a better culture environment for SSCs by facilitating the exchange of gases, nutrients, and waste products (88). Recently, successful recapitulation of human testicular organogenesis from fetal gonads was achieved, and in vitro-derived haploid spermatids were shown to undergo meiotic recombination (89).

2D culture systems with testis cell suspensions have been widely used for SSC proliferation and differentiation with two primary types of 2D culture systems most common: (1) SSCs cultured on mitotically-inactivated feeder cells, (2) SSCs co-cultured with somatic cells (90). Using the support of 2D culture sytems, numerous studies have reported that haploid male germ cells could be induced (91–95), and offspring can be produced from these in vitro derived haploid male germ cells in rodent (96). However, the 2D culture system has not been optimized for human germ cells. This may be due to the lack of spatial structure of seminiferous tubules and proper interactions between germ cells and somatic cells.

To better mimic the testicular niche, various 3D culture systems have been developed. In 2006, testicular cells isolated from rats were cultured on collagen gels to mimic the composition of the basal membrane of seminiferous tubules (97). Later, the soft-agar culture system (SACS) was developed (98), and mice haploid germ cells from undifferentiated germ cells were generated in this system in 2012 (99). The SACS system also supports the differentiiation of SSCs of non-human primates. The most commonly used alternate material in 3D culture system is methylcellulose. The methylcellulose culture system (MCS) also supports the differentiation of immature germ cells.

In order to artificially reproduce the in vivo form and function of the seminiferous epithelium, a 3D engineered blood-testis barrier (eBTB) system was designed in 2010 (100). Testicular peritubular myoid cells were first cultured on the underside of culture inserts, and then germ cells and Sertoli cells were added on top of the inserts. The testicular cells from neonatal mice form the aggregate by culturing on a V-shaped plate. The aggregate plated on the top of agarose gel blocks, and the haploid male germ cells were obtained after 30-51 days of incubation (101).

The 3D decellularized testicular scaffold with hyaluronic acid and chitosan provides the condition for the differentiation and proliferation of mice SSCs (102). The proliferation and self-renewal of mice SSCs was stimulated by culturing on the 3D scaffold consisting of alginate hydrogel with Sertoli cells (103). The mice germ cells were cultured in 3D printed one-layer scaffolds at the air-medium interface simulating the tubule-like structure. This culture system provided the condition for long-term survival and differentiation (104).

Soft agar and agarose gel are the most common material used to establish the 3D culture system for human SSCs. A soft agar culture system has been shown to support the proliferation and differentiation of human SSCs (105). Another material that has been used in 3D culture systems for human SSCs is a polycaprolactone (PCL) nanofiber matrix (106). This material may mimic the physical form of collagen fibers in the natural extracellular matrix (107).