The Proliferation of Pre-Pubertal Porcine Spermatogonia in Stirred Suspension Bioreactors Is Partially Mediated by the Wnt/β-Catenin Pathway

Male survivors of childhood cancer are at risk of suffering from infertility in adulthood because of gonadotoxic chemotherapies. For adult men, sperm collection and preservation are routine procedures prior to treatment; however, this is not an option for pre-pubertal children. From young boys, a small biopsy may be taken before chemotherapy, and spermatogonia may be propagated in vitro for future transplantation to restore fertility. A robust system that allows for scalable expansion of spermatogonia within a controlled environment is therefore required. Stirred suspension culture has been applied to different types of stem cells but has so far not been explored for spermatogonia. Here, we report that pre-pubertal porcine spermatogonia proliferate more in bioreactor suspension culture, compared with static culture. Interestingly, oxygen tension provides an avenue to modulate spermatogonia status, with culture under 10% oxygen retaining a more undifferentiated state and reducing proliferation in comparison with the conventional approach of culturing under ambient oxygen levels. Spermatogonia grown in bioreactors upregulate the Wnt/ β-catenin pathway, which, along with enhanced gas and nutrient exchange observed in bioreactor culture, may synergistically account for higher spermatogonia proliferation. Therefore, stirred suspension bioreactors provide novel platforms to culture spermatogonia in a scalable manner and with minimal handling.


Introduction
Advances in cancer therapies have led to increased survival rates of pediatric cancer patients [1,2]. As a result, there is a growing number of childhood cancer survivors who suffer from the late effects of gonadotoxic chemotherapy, which include hypogonadism and infertility due to loss of the spermatogonia pool [3,4]. Although cryopreservation of sperm is a routine fertility preservation technique for adult men undergoing chemotherapy, there is no such option for pre-pubertal boys before the onset of spermatogenesis. This has led to an interest in the isolation and cryopreservation of testicular biopsies, to allow subsequent isolation, expansion, and transplantation of spermatogonia back into the testes to restore spermatogenesis [5].
Although the culture of rodent spermatogonia is well established [6], scalable expansion of non-rodent spermatogonia in culture has had limited success [7][8][9][10]. Most of the research reported so far has employed conventional static culture systems, which are labor

In Suspension Culture, Porcine Spermatogonia Proliferate More in Ambient O2 Tension but Contain Fewer Undifferentiated Spermatogonia
Mouse and cattle spermatogonia cultures at 10% O2 support better expansion in vitro [10,18]. It should be noted that incubators filled with ambient air are widely referred to as containing 20.9% O2; however, dilution with CO2 and water vapor results in actual oxygen content that is somewhat lower. At sea level, this reduction is to ~18.6% O2 in the gas phase; in our laboratory in Calgary, elevation (~1084 m) reduced air pressure and consequently oxygen levels as well (see Supplementary Methods for detailed calculations) [13,14,37,38]. To explore if low O2 tension would be beneficial for porcine spermatogonia culture, cells from 1-week-old pigs (containing 82.2 ± 1.6% UCHL1 +ve cells) and from 8- Figure 1. Stirred suspension bioreactor culture at 120 rpm (3.9 dyne/cm 2 ) supports the highest level of spermatogonia proliferation: (A) immunofluorescence images of spermatogonia after static, 80, 100, and 120 rpm SSB culture. Scale bars 50 µm; (B,D,F) number of spermatogonia harvested after 2 weeks of SSB and static culture at 80 (B) 100 (D), and 120 (F) rpm; (C,E,G) percentage of proliferating spermatogonia after 2 weeks of SSB and static culture 80 (C), 100 (E), and 120 (G) rpm. n = 3, mean ± SD. p > 0.05 (ns), * p ≤ 0.05, ** p ≤ 0.01.

In Suspension Culture, Porcine Spermatogonia Proliferate More in Ambient O 2 Tension but Contain Fewer Undifferentiated Spermatogonia
Mouse and cattle spermatogonia cultures at 10% O 2 support better expansion in vitro [10,18]. It should be noted that incubators filled with ambient air are widely referred to as containing 20.9% O 2 ; however, dilution with CO 2 and water vapor results in actual oxygen content that is somewhat lower. At sea level, this reduction is to~18.6% O 2 in the gas phase; in our laboratory in Calgary, elevation (~1084 m) reduced air pressure and consequently oxygen levels as well (see Supplementary Methods for detailed calculations) [13,14,37,38]. To explore if low O 2 tension would be beneficial for porcine spermatogonia culture, cells from 1-week-old pigs (containing 82.2 ± 1.6% UCHL1 +ve cells) and from 8-week-old pigs (containing 80.6 ± 1.4% UCHL1 +ve cells) were cultured in static plates (control) and in 120 rpm SSBs at 10% and ambient O 2 tensions. The 10% O 2 and the ambient culture conditions translate to 3.45 mM and 6.34 mM of O 2 in the gas phase (Table 1). The O 2 tension and concentration in the liquid phase (media) at 48 h as measured via Oxygen Sensor Spots SP-PSt3-NAU (PreSens Regensburg, Germany) ( Figure S2C, Supplementary Methods) [32] are shown in Table 2. Both 1-and 8-week-old spermatogonia were examined to evaluate the effect of SSB culture on early and late pre-pubertal spermatogonia, as we reported previously that these cells have different metabolic requirements [32]. For 1-week-old spermatogonia, total cell yield after 2 weeks of culture was higher in both 10% and ambient O 2 (n = 3, p < 0.05) tension, compared with static culture ( Figure 2C). However, no significant difference in cell yield was observed between SSB cultures in ambient and 10% O 2 tension ( Figure 2C). Similarly, proliferation levels (% of EdU +ve UCHL +ve cells) were higher for SSB cultures, compared with static controls for both O 2 tensions (n = 3, p < 0.05). In addition, spermatogonia proliferation was higher in ambient O 2 SSB culture, compared with 10% O 2 culture (n = 3, p < 0.05) (Figure 2A,D).
For 8-week-old spermatogonia, similar to 1-week-old cells, cell yield after 2 weeks of culture was higher in SSBs at both 10% and ambient O 2 (n = 3, p < 0.05) tension, compared with the static. Cell yield between SSB cultures in 10% and ambient O 2 tension was not significantly different ( Figure 2C). Spermatogonia proliferation was higher for SSB cultures, compared with static controls at both 10% and ambient (n = 3, p < 0.05) O 2 cultures ( Figure 2B,D). Compared with 10% O 2 tension, spermatogonia proliferation was higher in ambient O 2 tension for SSB cultures (n = 3, p < 0.05) ( Figure 2D).
It has been reported that mouse spermatogonia culture at ambient O 2 results in differentiation into progenitor cells and loss of stem cell potential [6,18,39]. To elucidate if the spermatogonia in SSB culture are undergoing differentiation, the transcription of GDNF family receptor alpha 1 (GFRα1) [40] and promyelocytic leukemia zinc finger (PLZF), as markers of undifferentiated spermatogonia, and deleted in azoospermia-like (DAZL), a marker of differentiating spermatogonia [41], was evaluated. Although no significant difference in transcription levels was observed for PLZF and DAZL ( Figure S2A,B), GFRα1 transcription levels were 3.72-fold higher in 1-week-old cells and 3.18-fold higher in 8-week-old cells (n = 3, p < 0.05) cultured at 10% O 2 tension, compared with cells cultured under ambient O 2 ( Figure 2E) indicating retention of a more undifferentiated state, which is consistent with previous reports [6,18,39]. Subsequent analyses were therefore carried out at 10% O 2 tension.

Shear Forces Generated in SSB Culture Activate the Wnt/β-Catenin Pathway
Fluid shear forces are known to activate the Wnt/β-catenin pathway by nuclear translocation of β-catenin [25,28,[42][43][44], likely mediated by the mechano-sensing properties of E-cadherin [45]. β-catenin binds to the cytoplasmic tail of E-cadherin [46], which, upon exposure to fluid shear in SSBs, is dislodged from the adherens junction and accumulates in the cytoplasm resulting in nuclear translocation and pathway activation [25]. To investigate Wnt/β-catenin pathway activation, 1-and 8-week-old cells cultured in static conditions and SSBs at 120 rpm under 10% O2 tension were stained for β-catenin (1:200, Abcam) and vimentin, a marker for Sertoli and interstitial cells (1:500, Sigma) ( Figure S3A,B), and the number of vimentin -ve cells with nuclear β-catenin were counted to determine the % of spermatogonia displaying nuclear β-catenin localization. For both age groups, the percentages of spermatogonia with nuclear β-catenin were higher in SSB cultures, compared with static controls (n = 3, p < 0.05) ( Figure 3A, Figure S3C-H). Spermatogonia express Ecadherin [47] and grow as clusters in the SSBs ( Figure S3I,J). This clustering is mediated by E-cadherin [48], which likely serves as an upstream mediator for the nuclear localization of β-catenin in SSBs. To further characterize the Wnt/β-catenin pathway activation, qPCR for E-cadherin and Axin2, a downstream target of the Wnt/β-catenin pathway [49],

Shear Forces Generated in SSB Culture Activate the Wnt/β-Catenin Pathway
Fluid shear forces are known to activate the Wnt/β-catenin pathway by nuclear translocation of β-catenin [25,28,[42][43][44], likely mediated by the mechano-sensing properties of E-cadherin [45]. β-catenin binds to the cytoplasmic tail of E-cadherin [46], which, upon exposure to fluid shear in SSBs, is dislodged from the adherens junction and accumulates in the cytoplasm resulting in nuclear translocation and pathway activation [25]. To investigate Wnt/β-catenin pathway activation, 1-and 8-week-old cells cultured in static conditions and SSBs at 120 rpm under 10% O 2 tension were stained for β-catenin (1:200, Abcam) and vimentin, a marker for Sertoli and interstitial cells (1:500, Sigma) ( Figure S3A,B), and the number of vimentin -ve cells with nuclear β-catenin were counted to determine the % of spermatogonia displaying nuclear β-catenin localization. For both age groups, the percentages of spermatogonia with nuclear β-catenin were higher in SSB cultures, compared with static controls (n = 3, p < 0.05) ( Figures 3A and S3C-H). Spermatogonia express E-cadherin [47] and grow as clusters in the SSBs ( Figure S3I,J). This clustering is mediated by E-cadherin [48], which likely serves as an upstream mediator for the nuclear localization of β-catenin in SSBs. To further characterize the Wnt/β-catenin pathway activation, qPCR for E-cadherin and Axin2, a downstream target of the Wnt/β-catenin pathway [49], was performed. Transcription data revealed that both E-cadherin and Axin2 were upregulated in SSB culture, compared with conventional static culture, with E-cadherin increasing 5.3-fold for 1-week-old and 2.7-fold for 8-week-old cells, while Axin2 was increased 4-fold and 3-fold, respectively (n = 3, p < 0.05) ( Figure 3B,C). was performed. Transcription data revealed that both E-cadherin and Axin2 were upregulated in SSB culture, compared with conventional static culture, with E-cadherin increasing 5.3-fold for 1-week-old and 2.7-fold for 8-week-old cells, while Axin2 was increased 4-fold and 3-fold, respectively (n = 3, p < 0.05) ( Figure 3B,C).

Activation of the Wnt/β-Catenin Pathway in Static Culture Does Not Fully Mimic the Effect of Suspension Culture
If Wnt/β-catenin is solely responsible for the enhanced proliferation seen in SSB culture, then activating the pathway in static culture would lead to similar levels of proliferation. To test that hypothesis, spermatogonia from 1-and 8-week-old pigs were cultured in SSBs at 120 rpm and in static cultures at 10% O2 tension with CHIR99021, a Glycogen Synthase Kinase 3β (GSK3β) inhibitor (Wnt activator) [50] for 14 days. Preliminary doseresponse experiments showed that 3 µM CHIR99021 was the highest dose that could be used without negatively affecting cell viability. Upon treatment with 3 µM CHIR99021, transcription of Axin2 was increased 3.2-fold for the treatment group and 4.3-fold for SSB, compared with control (n = 3, p < 0.05) for the 1-week-old spermatogonia ( Figure 4A). For 8-week-old cells, Axin2 transcription level increased 2.3-fold for treatment and 3.4-fold for SSB 120 rpm, compared with the control group (n = 3, p < 0.05) ( Figure 4A). Proliferation levels were found to be higher in the treatment groups than in the control but were still lower than in SSB cultures (n = 3, p < 0.05) ( Figure 4B).

Activation of the Wnt/β-Catenin Pathway in Static Culture Does Not Fully Mimic the Effect of Suspension Culture
If Wnt/β-catenin is solely responsible for the enhanced proliferation seen in SSB culture, then activating the pathway in static culture would lead to similar levels of proliferation. To test that hypothesis, spermatogonia from 1-and 8-week-old pigs were cultured in SSBs at 120 rpm and in static cultures at 10% O 2 tension with CHIR99021, a Glycogen Synthase Kinase 3β (GSK3β) inhibitor (Wnt activator) [50] for 14 days. Preliminary doseresponse experiments showed that 3 µM CHIR99021 was the highest dose that could be used without negatively affecting cell viability. Upon treatment with 3 µM CHIR99021, transcription of Axin2 was increased 3.2-fold for the treatment group and 4.3-fold for SSB, compared with control (n = 3, p < 0.05) for the 1-week-old spermatogonia ( Figure 4A). For 8-week-old cells, Axin2 transcription level increased 2.3-fold for treatment and 3.4-fold for SSB 120 rpm, compared with the control group (n = 3, p < 0.05) ( Figure 4A). Proliferation levels were found to be higher in the treatment groups than in the control but were still lower than in SSB cultures (n = 3, p < 0.05) ( Figure 4B).

Discussion
Stirred suspension bioreactors have proven to be a robust culture system for the expansion of various stem cells [23,33,34,36]. Bioreactors can, in most cases, minimize or eliminate the shortcomings of conventionally employed static culture systems, allowing for easily scalable cell culture with minimal handling [12]. This is of particular importance when cells are grown for cell therapy applications [11]. With the advent of computational fluid dynamics modeling, it has become easier to fine-tune the shear environment allowing for more accurate scale-up strategies from smaller to larger bioreactors [51]. Despite these benefits, the application of a dynamic suspension culture system to spermatogonia remains relatively unexplored. In 2014, Zhang et al. reported the use of a dynamically simulated microgravity rotating cell culture system (RCCS) for culturing mouse spermatogonia for 14 days [52]. The cells were co-cultured with mitotically inactivated Sertoli cell feeders on fibrin scaffolds and grew as aggregates. However, this RCCS system did not allow opportunities for scalability and requires the use of feeder cells. We attempted to overcome these limitations with stirred suspension bioreactors.
In stirred suspension bioreactors, cells grow as aggregates [23,25], and the aggregate size is determined by shear stress. Shear stress can be controlled by increasing or decreasing the rotation speed of the impeller, which correlates to the size of eddies [53]. The size of turbulent eddies is inversely proportional to rotation speed. Slower rotation speed leads to larger eddies, which are unable to break down larger aggregates. This allows the aggregates to grow too large and, as a result, fall out of suspension negating the beneficial effects of suspension culture. Too high of a rotation speed can generate eddies that are too small, leading to cell damage and death. At optimal rotation speed, the eddies are large enough to keep aggregates in suspension but small enough to impart shear stress to break down larger aggregates into smaller ones without damaging the cells [35]. According to

Discussion
Stirred suspension bioreactors have proven to be a robust culture system for the expansion of various stem cells [23,33,34,36]. Bioreactors can, in most cases, minimize or eliminate the shortcomings of conventionally employed static culture systems, allowing for easily scalable cell culture with minimal handling [12]. This is of particular importance when cells are grown for cell therapy applications [11]. With the advent of computational fluid dynamics modeling, it has become easier to fine-tune the shear environment allowing for more accurate scale-up strategies from smaller to larger bioreactors [51]. Despite these benefits, the application of a dynamic suspension culture system to spermatogonia remains relatively unexplored. In 2014, Zhang et al. reported the use of a dynamically simulated microgravity rotating cell culture system (RCCS) for culturing mouse spermatogonia for 14 days [52]. The cells were co-cultured with mitotically inactivated Sertoli cell feeders on fibrin scaffolds and grew as aggregates. However, this RCCS system did not allow opportunities for scalability and requires the use of feeder cells. We attempted to overcome these limitations with stirred suspension bioreactors.
In stirred suspension bioreactors, cells grow as aggregates [23,25], and the aggregate size is determined by shear stress. Shear stress can be controlled by increasing or decreasing the rotation speed of the impeller, which correlates to the size of eddies [53]. The size of turbulent eddies is inversely proportional to rotation speed. Slower rotation speed leads to larger eddies, which are unable to break down larger aggregates. This allows the aggregates to grow too large and, as a result, fall out of suspension negating the beneficial effects of suspension culture. Too high of a rotation speed can generate eddies that are too small, leading to cell damage and death. At optimal rotation speed, the eddies are large enough to keep aggregates in suspension but small enough to impart shear stress to break down larger aggregates into smaller ones without damaging the cells [35]. According to our study, 120 rpm, which generated a maximum shear force of 3.9 dyne/cm 2 , was the optimal rotational speed to maintain spermatogonia in suspension and allow for proliferation. While the rotational speed of 80 rpm or 2.2 dyne/cm 2 of maximum shear force was not detrimental to cell survival, it failed to elicit spermatogonia proliferation. In contrast, both 60 (1.5 dyne/cm 2 ) and 140 rpm (4.8 dyne/cm 2 ) were detrimental to cell survival, likely due to too low and too high of shear stress, respectively.
One key parameter in cell culture systems, which is often not reported but is critical for reproducibility, is the O 2 tension that cells experience in the liquid phase. We measured O 2 concentration in the liquid phase to be 5.2 ± 0.01 mM for static culture, which was lower than 6.3 ± 0.06 mM O 2 concentration seen in bioreactor culture under atmospheric conditions. In 10% O 2 cultures, O 2 concentration in the bioreactor was around 1.3 ± 0.003 mM and in static culture was 1.1 ± 0.07 mM. Calgary is located at approximately 1084 m above sea level, and therefore, the O 2 concentration in the gas phase of a tissue culture incubator operated at 37 • C under atmospheric conditions is 6.34 mM, and under 10% O 2 conditions, it is 3.45 mM [14]. The solubility of O 2 , which is proportional to partial pressure, is rather low in a liquid phase (culture media). In addition, diffusion of O 2 through the media, starting from the gas-liquid interphase to the bottom of the plate (where cells are in adherent culture) for non-stirred cultures is also considerably slow. The partial pressure of O 2 in the ambient and 10% O 2 incubators are, respectively, 16.34 kPa and 8.89 kPa; therefore, reduced O 2 concentration in the static conditions, compared with the gas phase, is not surprising [14]. What is noteworthy here is that the O 2 concentration in the liquid phase of bioreactors is higher than their static counterparts. Since the media in bioreactors is continuously stirred, the distribution of O 2 throughout the liquid phase is accelerated, and as a result, the oxygen tension in bioreactors tends to be higher than in static culture [38].
As there is no established static culture system for pig spermatogonia, we compared mouse spermatogonia culture in bioreactors with the mouse static culture system. To the best of our knowledge, an established static culture system is only available for mouse spermatogonia. However, different from our observations with porcine spermatogonia, the stirred suspension bioreactor culture did not support the survival or proliferation of cultured mouse spermatogonia. One factor that may account for this difference could be that the mouse spermatogonia that we cultured in the bioreactors were not primary cells. To obtain sufficient cell numbers to load SSBs, these cells were expanded initially in static culture on mitotically inactivated STO feeder cells [6]. As a result, they may have become dependent on the presence of a feeder layer rendering them unsuitable for suspension culture. It may be necessary to transition primary mouse testicular cells directly onto a feeder-free suspension system without an intermediate static culture. Another way to address this issue would be to incorporate microcarriers, inert plastic beads that can be treated with different substrates such as gelatin or laminin to promote cell adhesion. Cells that do not actively form aggregates with each other may be adhered initially to such microcarriers and then cultured in bioreactors [33]. If feeder cells are essential, they can also be adhered to microcarrier beads and then mitotically inactivated to be incorporated in spermatogonia cultures.
Spermatogonial stem cells (SSCs), precursors of spermatogonia, originate from bipotent primordial germ cells, which depend on oxidative phosphorylation (OXPHOS) [54][55][56][57], while adult cells primarily rely on high glycolytic flux and low mitochondrial activity for their metabolic needs [18,[58][59][60]. We have shown that the transition to a more mature phenotype is initiated at 8 weeks of age, while 1-week-old pig spermatogonia mainly rely on OXPHOS fuelled by pyruvate consumption [32]. Glycolytic metabolism is required to secure carbon atoms for nucleotide, amino acid, and lipid biosynthesis, all required for rapid cell proliferation. At the same time, differences in metabolic flux cause a change in metabolite abundance within the cell, functioning as co-factors to alternate the epigenome and therefore transcriptional profile and ultimately cell function [61].
Lower oxygen tensions generally cause a relative shift in the contribution of glycolytic flux to the energy production of the cell. In SSCs, glycolysis has been shown to contribute to the enhancement of self-renewing circuits in vivo and in vitro via an increase in AKT activity [18,58]. AKT phosphorylation is suggested to be downstream of GFRα1 activation. Therefore, initiating a shift toward anaerobic metabolism via the decrease in oxygen partial pressure increases SSC maintenance circuits via AKT, similarly to what has been described in mouse models. In mice, an increase in AKT pathway activity has been shown to increase glycolytic flux and expression of GFRa1 [58]. PLZF and DAZL have a much broader expression pattern, and their regulation is most likely mainly regulated via post-transcriptional regulation, which can explain the fact that no change was observed on a transcriptional level. Aside from the physical benefits, bioreactors have also been shown to upregulate and maintain pluripotency markers while downregulating the differentiation potential of pluripotent stem cells [23,24,62] This is mediated by the bioreactors' ability in modulating the Wnt/β-catenin pathway [25,34]. Previous studies have shown that the Wnt/β-catenin pathway is involved in the survival and proliferation of spermatogonia and is required for spermatogenesis [30,63,64]. Chassot et al. reported that constitutive activation of the Wnt pathway led to the proliferation of gonocytes [64]. Deletion of β-catenin in adult mouse testes caused a reduction of PLZF +ve undifferentiated spermatogonia [30]. Similar to pluripotent cells [25], porcine spermatogonia also responded to shear forces in bioreactors by upregulating the Wnt/β-catenin pathway, characterized by nuclear β-catenin translocation and elevated Axin2 transcription level. Although activation of the pathway in static culture with a small molecule activator CHIR 99021 caused increased proliferation, compared with controls, the proliferation levels observed after treatment were lower than in bioreactor culture. It is possible that, under the conditions tested, pathway activation with 3 um CHIR 99021 was incomplete [65]. In SSBs, exposure to fluid shear forces releases β-catenin from adherens junctions and also increases the levels of phosphorylated glycogen synthase kinase 3β, thus inhibiting β-catenin degradation. This inactivation of GSK3β is likely missing in static culture, even with CHIR 99021 treatment, causing both reduced Axin2 transcription and spermatogonia proliferation. In addition, the physical benefitsnamely, higher nutrient and gas exchange within the spermatogonia aggregates due to continuous hydrodynamic modulation, may also play an important role in spermatogonia survival and expansion in suspension culture.
Human pre-pubertal spermatogonia culture is challenging due to the scarcity of pre-pubertal tissue. It is also not clearly understood when metabolic transitions occur in humans. However, we have shown that early pre-pubertal human spermatogonia (1 year of age) have similar mitochondrial ultrastructure as 1-week-old pig spermatogonia [61]; therefore, human spermatogenic metabolic development might be similar to the pre-pubertal development in the pig. As the metabolic requirements of human pre-pubertal spermatogonia are still ambiguous, any culture techniques and especially bioreactor culture are still in their infancy. To summarize, we report a novel culture system for undifferentiated, nonrodent spermatogonia using stirred suspension bioreactors. Stirred suspension bioreactors allowed a scalable expansion of spermatogonia by both enhancing oxygen and mass transfer, as well as by modulating the Wnt/ β-catenin pathway within a controlled environment. Therefore, bioreactors could be an invaluable tool for future clinical applications such as cell therapy for treating infertility.

Spermatogonia Isolation and Enrichment
Testes from 1-and 8-week-old piglets were obtained from Sunterra Farms Ltd (Acme, AB, Canada) and the University of Alberta, Edmonton, AB. All procedures were performed with approval and under the oversight of the Animal Care Committee of the University of Calgary. Spermatogonia were harvested using a two-step enzymatic digestion process as previously described [66,67]. Briefly, testes were decapsulated and minced into~1-2 mm pieces. These tissue pieces were then digested using collagenase IV (Sigma, Oakville, ON, Canada) (2 mg/mL), 0.25% trypsin-EDTA (Sigma, Oakville, ON, Canada), and DNase I (Sigma, Oakville, ON, Canada) (7 mg/mL) to obtain the starting cell population. Differential plating was performed with the starting cell population to enrich for spermatogonia. This enriched spermatogonia population was assigned to different experimental groups. All experiments were repeated using a minimum of three independently prepared cell suspensions.

Spermatogonia Culture
Enriched porcine spermatogonia from 1-and 8-week-old piglets were suspended in spermatogonia culture media: Advanced Minimum Essential medium (Thermo Fisher Scientific, Mississauga, ON, Canada) supplemented with glial cell-derived neurotrophic factor (40 ng/mL), GFRα1 (25 ng/mL), and epidermal growth factor (20 ng/mL) [32] and were seeded into 10 mL SSBs (Corning Style Spinner Flask; NDS Technologies Inc., Vineland, NJ, USA) [25,51] at a concentration of 500 × 10 3 cells per mL (which translated to 5 × 10 6 cells suspended in 10 mL media in each SSB) and into 6-well plates (which translated to 1 × 10 6 cells in 2 mL media in each well, a total of 5 wells were prepared for each sample totaling in 5 × 10 6 cells) as static control. The SSBs were placed onto magnetic stirrers inside a cell culture incubator (37 • C which is 2 • C lower than the average body temperature (39 • C) of a domestic pig), which was preset for ambient O 2 tension. The 1-week-old spermatogonia were stirred at 60, 80, 100, 120, and 140 rpm, while 8-week-old spermatogonia were cultured only at 120 rpm. Additionally, 120 rpm SSB and control static culture were performed in a tri-gas incubator, which was preset for O 2 tension of 10%. The cells in SSBs and static control plates were cultured for a total of 14 days, with media changes at every 48 h. The cells were exposed to EdU (5-ethynyl-2 -deoxyuridine) for the last 12 h of culture, to quantify the proportion actively synthesizing DNA during that period, and were harvested using 0.25% trypsin EDTA. Cell viability was assessed using trypan blue exclusion assay.

Oxygen Measurement
Oxygen measurements were performed by using Oxygen Sensor Spots SP-PSt3-NAU (PreSens, Regensburg, Germany). The sensor was calibrated according to the manufacturer's instructions. Measurements were performed for 1-week-old porcine spermatogonia cultured in static and 120 rpm SSBs at both ambient and 10% O 2 tensions [32] for 48 h.

CHIR Treatment
Porcine spermatogonia from 1-and 8-week-old piglets were seeded into a 6-well plate, and the media was supplemented with 3 µM CHIR 99021 (Bio Techne, Toronto, ON, Canada) dissolved in DMSO (0.03% DMSO in medium). The control group was treated with an equal amount of DMSO. The culture was carried out in a hypoxic incubator. The total duration of culture was 14 days with media changes every day. Each time media was supplemented with 3 µM CHIR 99021. The cells were then harvested and analyzed for proliferation and Axin2 transcription level.

EdU Incorporation Assay
Cells were pulsed with 5 µM EdU (5-ethynyl-2 -deoxyuridine), a nucleoside analog incorporated into DNA during replication, for the last 12 h of all cultures. EdU uptake was visualized using Click-iT EdU Alexa Fluor 647 imaging kit as per manufacturers protocol (Thermo Fisher Scientific, Mississauga, ON, Canada).

Immunofluorescence
Testis tissue was fixed in 4% paraformaldehyde (PFA) at 4 • C overnight, dehydrated with a gradient series of ethanol, and embedded in paraffin wax to prepare sections of 5 µm thickness. Cells were fixed with 2% PFA at room temperature for 15 min and stained for

Quantitative Reverse Transcription Polymerase Chain Reaction (qRT-PCR)
The transcription levels of spermatogonia markers, upstream and downstream targets of Wnt/ β-catenin (Table 3), were assessed via qRT-PCR after fluorescent-activated cell sorting (FACS) sorting of enriched spermatogonia as described in [32]. RNA was extracted from 1 × 10 5 cells using RNeasy Micro Kit (QIAGEN, Toronto, ON, Canada). Reverse transcription was performed using SuperScript™ IV VILO™ Master Mix (Thermo Fisher Scientific, Mississauga, ON, Canada). In addition, the 7.500 Fast Real-Time PCR System (Applied Biosystems, Waltham, MA, USA) was used, with SsoFast Eva Green Supermix with Low ROX (Bio-Rad laboratories, Mississauga, ON, Canada) for qPCR. Relative expression is calculated using 2 −∆∆CT . Statistical analysis was performed on the mean of ∆∆Ct.

Maximum Shear Stress Calculation
Maximum shear stress in a 10 mL bioreactor was calculated using equations from Gareau et al. 2014 [34]. The required dimensions of 10 mL bioreactors are stated in Table 4.

Statistical Analysis
All data reported are from at least three independent experiments performed with separately prepared enriched spermatogonia populations from different animals (n = 3). The data were analyzed by unpaired two-tailed t-tests to compare two groups and one-way ANOVA for more than two groups with the Tukey multiple comparisons test. A value of p < 0.05 was set as the limit of statistical significance. Statistical analyses were performed using the GraphPad Prism 8 software (San Diego, CA, USA).  Institutional Review Board Statement: The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Animal Care Committee of the University of Calgary (protocol code: AC20-0143, date of approval: 3 December 2021).

Data Availability Statement:
The data that support the findings of this study are available on request from the corresponding author.