Spermatogenesis can occur in testis tissue from immature bulls ectopically grafted into mouse hosts; however, efficiency of sperm production is lower than in other donor species. To elucidate a possible mechanism for the impaired spermatogenesis in bovine testis xenografts, germ cell fate and xenograft development were investigated at different time points and compared with testis tissue from age-matched calves as controls. Histologically, an initial decrease in germ cell number was noticed in xenografts recovered up to 2 months post-grafting without an increase in germ cell apoptosis. From 2 months onward, the number of germ cells increased. In contrast, a continuous increase in germ cell number was seen in control tissue. Pachytene spermatocytes were observed in some grafts before 4 months, whereas in the control tissue they were not present until 5 months of age. Beyond 4 months post-grafting spermatogenesis appeared to be arrested at the pachytene spermatocyte stage in most grafts. Elongated spermatids were observed between 6 and 8 months post-grafting, similar to the controls, albeit in much lower numbers. Lumen formation started earlier in grafts compared with controls and by 6 months post-grafting tubules with extensively dilated lumen were observed. A donor effect on efficiency of spermatogenesis was also observed. These results indicate that the low efficiency of sperm production in bovine xenografts is due to an initial deficit of germ cells and impaired meiotic and post-meiotic differentiation. The characterization of spermatogenic efficiency will provide the basis to understand the control of spermatogenesis in testis grafts.
Mammalian testis tissue xenografted into mice has resulted in production of viable spermatozoa (Snedaker et al. 2004) that are capable of fertilization (Honaramooz et al. 2002, 2004, Shinohara et al. 2002, Schlatt et al. 2003). Spermatogenesis in bovine testis tissue xenografts has not been very efficient, with only 5–10% of seminiferous tubules producing elongated spermatids (Oatley et al. 2004, 2005). Mammalian testis tissue xenografted into mice could serve as an excellent in vivo model not only for studying the complete process of mammalian spermatogenesis in different species, but also to genetically modify or preserve the germ line (Honaramooz et al. 2002). Availability of an efficient in vivo model will facilitate the study of spermatogenesis in important livestock species like cattle by allowing replication of treatments within donors and minimizing the use of large experimental animals.
Oately et al.(2004, 2005) grafted testicular tissue from bovine calves and reported germ cell differentiation up to elongated spermatids in tissues recovered 24 weeks post-grafting (cumulative age >24 weeks) in only 5–10% of tubules. This percentage of tubules with complete spermatogenesis is not adequate to reliably study spermatogenesis or to preserve or alter the germ line. The underlying reasons for this low spermatogenic efficiency in bovine testis xenografts are not clear. It is possible that the endocrine environment of the host mice is not optimal for the bovine testicular tissue development; however, it appears to support spermatogenesis in xenografts from pig and goat testis with an efficiency similar to that in the donor species (Honaramooz et al. 2002).
Growth hormone (GH) is thought to affect testicular growth and development, including steroidogenesis and gametogenesis directly or via local insulin-like growth factor-I (IGF-I) production by testicular cells (Hull & Harvey 2000). Around puberty GH is elevated and puberty is delayed in GH receptor-knockout mice (Keene et al. 2002) and in men with GH deficiency (Fujisawa et al. 2002). It was therefore hypothesized that the supplementation of exogenous bovine GH to the host mice might improve maturation and subsequent sperm production in the bovine testicular xenografts.
In the present study we characterized the development of xenografted bovine testis tissue by detailed analysis over time in comparison with the development in the testis tissue from age-matched bulls of the same breed. In addition, the effect of exogenous bovine GH on bovine xenograft development was assessed. The objective of the study was to gain a better understanding of the underlying processes responsible for the low efficiency of sperm production in bovine testis xenografts in order to facilitate the design of treatments to potentially overcome this inefficiency.
Materials and Methods
Donor testis tissue
Testes from sexually immature 1–2-week-old male Holstein calves (n = 6) were used as donor tissue for xenografting. Histological sections of Holstein bull testicular tissue from different ages were used as in situ controls.
Recipient animals and procedures for xenografting
Donor tissue was assigned randomly to recipient mice. Xenografting was performed as described previously (Honaramooz et al. 2002). Briefly, after removal of the capsule and overt connective tissue, donor testes were cut into small fragments (about 1–2 mm in diameter). Testis fragments were kept in Dulbecco’s modified Eagle’s medium (Gibco Lab, Grand Island, NY, USA) on ice until grafting. Recipient immunodeficient (ICR-scid or NCR-nude; Taconic, Germantown, NY, USA) male mice (6 weeks old) were anesthetized and castrated, and during the same surgery eight pieces of donor testis tissue fragments were grafted under the back skin of each mouse (n = 28 mice). Animals were handled and treated in accordance with the University of Pennsylvania Institutional Animal Care and Use Committee.
Testis tissue from three additional 1-week-old bull calf donors was grafted into 21 NCR-nude mice (7 mice per donor) to study the effect of GH on the development of bovine testis tissue xenografts in mice. Starting 16 weeks post-grafting 12 mice (4 per donor) were subcutaneously injected with recombinant bovine somatotropin (rbST; Posilac, Monsanto Company, St Louis, MO, USA) in sesame oil (Sigma, St Louis, MO, USA) for 12 weeks at 1000 μg (n = 4), 100 μg (n = 4), or 25 μg (n = 4) per mouse per week. The remaining 9 untreated mice (3 from each donor) served as controls.
Recovery and analysis of xenografts
The host mice were killed by CO2 inhalation at regular time intervals between 1.5 and 18 months post-grafting to fall within specific age groups (four mice per age group). Grafts were recovered and fixed overnight in Bouin’s solution followed by three changes of 70% ethanol before being processed for histology. In each graft, all seminiferous tubule cross-sections were examined for the status of testicular maturation (seminiferous tubule and lumen diameter) and spermatogenesis (most advanced germ cell type). The number of spermatogonia/gonocytes per tubular cross-section and per 100 Sertoli cells were counted during the pre-meiotic stage of spermatogenesis. Germ cells were identified as gonocytes or spermatogonia based on their morphology and location in the seminiferous tubules; however, analysis was based on total germ cell number. The mouse was considered as the experimental unit. The data obtained from grafts from a single mouse was pooled and the average of the results from all the mice in an age group was analyzed. Seminal vesicles from all recipient mice were weighed as an indication of secretion of bioactive testosterone by the xenografts (Honaramooz et al. 2002, 2004, Schlatt et al. 2003)
Apoptosis and seminiferous tubule analysis
Germ cell/somatic cell apoptosis was analyzed for grafts and donor tissue from the 1–2-month age group using the terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) assay (ApopTag; Chemicon International, Temecula, CA, USA). Point-count based stereology (Russ & Dehoff 1999) was used to assess the relative area occupied by seminiferous tubules in the grafts and the area occupied by lumen in the seminiferous tubules. Images from histology samples from grafts were taken at 100 times magnification using a CoolSnap camera (Photometrics, Tucson, AZ, USA) mounted on a Leica microscope (Leica Microsystems AG, Wetzlar, Germany) with Image-Pro Plus image-capturing software (MediaCybernetics, Silverspring, MD, USA). Thereafter, each image was opened individually in Adobe PhotoShop Elements 2 (Adobe Systems, San Jose, CA, USA) and overlayed by a custom-designed image with 49 evenly spaced points on a transparent background. The points lying over various regions of the sample (interstitial tissue, seminiferous tubule epithelium, seminiferous tubule lumen) were counted. Summation of the points falling on the seminiferous tubule epithelium and in the lumen gave the relative area occupied by the tubule and those falling only on the lumen gave the relative area occupied by lumen. The percentage of points falling on the lumen compared with the whole tubule was calculated and an average per microscopic field was used to evaluate the tubular and luminal dimensions. The entire graft cross-section was analyzed (depending on the size of the graft cross-section a minimum of 10 microscopic fields (range 90–10) per sample were analyzed).
Immunohistochemistry of protein gene product 9.5 (PGP 9.5)
PGP 9.5 immunohistochemistry was used to identify gonocytes and spermatogonia (Wrobel et al. 1995) in the tissue samples. The protocol was modified from Wrobel et al.(1995). Briefly, slides were de-paraffinized and tissue samples were treated with 3% H2O2 in distilled water for 10 min to block the endogenous peroxidase activity followed by two washes in PBS for 5 min each. Subsequently, the sections were incubated at room temperature in 1% BSA in PBS for 1 h to block the non-specific antibody binding. Primary antibody (PGP 9.5; Biogenesis, Kingston NH, USA) diluted to 1:1000 in the blocking buffer was added to samples and incubated overnight at 4 °C in a humidified chamber. The samples were rinsed twice in PBS for 5 min each time and treated with peroxidase-conjugated donkey anti-rabbit IgG (Jackson ImmunoResearch Laboratories, West Grove, PA, USA) as the secondary antibody diluted to 1:500 in PBS with 1% BSA. The samples were washed twice in PBS for 5 min each and peroxidase activity was detected with VIP (VIP substrate kit for peroxidase; Vector Laboratories, Burlingame, CA, USA). Hematoxylin (Vector Laboratories) was used to counterstain the sections. The sections were analyzed under a light microscope at 400 × magnification.
The data from NCR-nude mice and ICR-scid mice were pooled since no difference in the graft development and maturation was observed between the two strains. Student’s t test was performed to compare two groups and a one-way ANOVA followed by a Student–Newman–Keul’s multiple-comparison test was performed to compare the data in more than two age groups. Data were analyzed using SigmaStat 3.0 (SPSS, Chicago, IL, USA). Data were expressed as means ± s.e.m. and P < 0.05 was considered significant.
Recovery of grafted tissue and seminal vesicle weights
Eighty percent of the testis xenografts (313/392) were recovered from all host mice (n = 28 + 21) with a marked increase in the size of the recovered grafts (40-fold by 4 months and 70-fold by 10 months). The seminal vesicle weight from recipient mice was 323.5 ± 34 mg (n = 32).
Histology of the donor tissue at the time of grafting and that of recovered grafts is represented in Fig. 1. Figure 2 illustrates the percentage of tubules containing the most advanced germ cell stage present in grafted and control tissues recovered at different time points. At the time of grafting, 83.3% (range, 94–65%) of the donor testis tissue seminiferous cords had gonocytes as the most advanced germ cell type (Fig. 2A) with 1.6 ± 0.3 gonocytes per cord cross-section and 10.9 ± 0.7 gonocytes per 100 Sertoli cells (n = 4; Fig. 3A). The number of germ cells had decreased significantly in the grafts recovered between 1 and 2 months post-grafting, with 0.5 ± 0.1 germ cells (gonocytes + spermatogonia) per tubule and 5.3 ± 1.3 per 100 Sertoli cells (n = 4; P < 0.05). Beyond 2 months, the grafts showed an increase in the number of germ cells; however, this number was still significantly lower than in the in situ controls, where a continuous increase in number of germ cells (gonocytes + spermatogonia) was observed (Fig. 3). Moreover, no significant difference in the number of apoptotic germ cells per tubule was observed between the xenografts (3.3 ± 0.8 apoptotic germ cells per tubule) and the in situ controls (3.4 ± 0.4 apoptotic germ cells per tubule) within the 1–2- month age group (n = 4).
Seminiferous tubules with pachytene spermatocytes were first noticed in the grafts harvested between 3 and 4 months, while in the controls pachytene spermatocytes were not noticed until 5 months of age (Fig. 2). The number of tubules with pachytene spermatocytes as the most advanced germ cell stage increased to 24% (range, 37.4–4.6%) by 6 months and did not change significantly with time (41% in grafts recovered beyond 10 months; range, 57.4–12.4%; Fig. 2A). In contrast, in the control tissue the number of tubules with pachytene spermatocytes as the most advanced germ cell stage increased until 8 months whereas after 8 months the majority of tubules contained post-meiotic cells (Fig. 2). Elongated spermatids were seen in grafts at around 6 months, similar to that in the control tissue. However, at any given time point, the number of tubules with elongated spermatids in grafts was significantly lower than in controls (7.2 ± 6.5 versus >80% at over 10 months of age, n = 4).
Seminiferous tubule development
During the pre-meiotic stage (up to 4 months) the percentage of points falling over seminiferous tubules per microscopic field in xenograft cross-sections was significantly higher than in the controls (59.9 ± 4.3 versus 47.6 ± 3.1%; P < 0.05). Thereafter, the relative tubular size in xenografts was variable with a range of 83.9–33.1% points (mean, 60.0 ± 5.6%) over seminiferous tubules per microscopic field compared with that in controls, where tubule size was more homogeneous (68.0 ± 4.0%; range, 75.1–60.8%). The presence of lumen in the seminiferous tubules in xenografts was noticed at a much earlier time point (2 months) compared with the controls (4 months). Tubules with extensively dilated lumen (60% or more points falling on lumen) started to appear at around 6 months post-grafting (~15% microscopic fields per sample; range, 40–2%) and increased further with time (~36% microscopic fields per sample between 8 and 10 months; range, 56–20%) in the testes xenografts. However, this increase varied between and within grafts, with some tubules in a graft showing no development at all while others showed normal development or abnormal dilation. In contrast, no abnormal dilation of tubules was noticed and the tubule development was very homogeneous in the control tissue.
Effect of GH treatment on xenografts
No significant effect of the GH treatment on the maturation of bovine xenografts was observed. Irrespective of the treatment the grafts showed similar stages of maturation based on the percentage of tubules showing the most advanced germ cell stage (Table 1). The percentage of seminiferous tubules with various germ cell stages was similar to that observed in the first experiment (Fig. 2A). Notably however, an effect of donor was clearly evident, as complete spermatogenesis was observed in all grafts recovered from mice grafted with tissue from one donor whereas in xenografts obtained from the remaining two donors tubules contained only round spermatids and spermatocytes, respectively, as the most advanced germ cell stage.
Testis tissue collected from immature males of several mammalian species can mature and undergo complete spermatogenesis when grafted ectopically into an immunodeficient mouse host (Honaramooz et al. 2002, 2004, Shinohara et al. 2002, Schlatt et al. 2003, Snedaker et al. 2004). Whereas sperm production in porcine testis xenografts can reach efficiency similar to the donor species on a per gram of tissue basis (Honaramooz et al. 2002), bovine testis tissue grafted into mice results in inefficient spermatogenesis producing elongated spermatids in not more than 5–10% of the tubules (Oatley et al. 2004, 2005, this study). To investigate possible reasons for this low efficiency, this study aimed to evaluate in detail the development of ectopically grafted bovine tissue in comparison with testis development in the donor species (in situ). Understanding the mechanisms controlling spermatogenic efficiency in bovine xenografts could aid in designing interventions directed at improving spermatogenesis and could also serve as model for spermatogenic defects in vivo.
An increase in the size of the recovered grafts was observed by 1 month after grafting signifying growth in seminiferous tubules size (Curtis & Amann 1981), which in turn is attributed to proliferation and differentiation of germ cells and to Sertoli cells (Curtis & Amann 1981, Sinowatz & Amselgruber 1986). Pachytene spermatocytes were noticed in the grafts 1–2 months earlier than in control tissue. However, elongated spermatids were first observed between 6 and 8 months post-grafting, similar to the controls, albeit in much lower numbers. These results indicate that meiotic differentiation is initiated earlier in the grafts than in controls but meiotic and post-meiotic differentiation did not occur in the grafts as efficiently as in control as meiotic arrest of spermatogenesis was seen in the majority of tubules after 4 months post-grafting. The initial accelerated maturation could be due to the exposure of the juvenile tissue to elevated gonadotropin levels in the castrated adult mouse host early on, indicating that the onset of testicular maturation and spermatogenesis is not prevented by immaturity of the testicular tissue but rather by a lack of endocrine support (Honaramooz et al. 2002, Schlatt et al. 2003).
Despite the acceleration in pre-meiotic germ cell differentiation an initial decrease in number of germ cells was noticed in testicular tissue grafts. In grafts recovered between 1 and 2 months post-grafting, the number of germ cells per tubule and per 100 Sertoli cells was lower compared with that in the donor tissue and the age-matched in situ controls. As no difference in germ cell apoptosis was noticed in the grafts compared with in situ, the low germ cell number could be a result of low germ cell division rate and/or relatively higher Sertoli cell division. The transplanted tissue pieces initially lack complete blood supply which could be detrimental for germ cell proliferation. Once the vascular supply from the host mice is established proliferation and differentiation of the surviving germ cells is supported. An increase in number of germ cells is seen with time; however, germ cell numbers remain lower than in situ where they continue to increase at a higher rate. In grafted bovine tissue the germ cells never appear to recover from the initial deficit and this could likely be one of the major causes leading to low sperm production in bovine grafts. The seminal vesicle weight from the recipient mice was similar to that in intact male mice (Schlatt et al. 2003), suggesting secretion of at least normal levels of bioactive testosterone by the grafted tissue (Honaramooz et al. 2002).
Lumen development in xenograft seminiferous tubules started earlier than in situ. Seminiferous tubule fluid (STF) secreted by the Sertoli cells is responsible for lumen formation. Between 4–6 months an extensively dilated lumen was seen in a few tubules of some grafts which could be attributed to the accumulation of STF as a result of imbalance in production and resorption of the STF. In the normal testis tissue the STF is reabsorbed in the rete testis. The absence of rete tissue in the grafts likely impairs absorption of STF and it accumulates in the tubules leading to tubular distension. Dilated tubules are also noticed in xenografts from other species (Honaramooz et al. 2002, 2004), but the number of such tubules is lower than in bovine xenografts, possibly due to secretion of STF in lower quantity or higher rate of fluid absorption in other species compared with bovine tissue.
Exogenous GH and follicle-stimulating hormone have been shown to influence testicular maturation (Swanlund et al. 1995). Graft development appeared to be sufficiently supported by endogenous mouse gonadotropins. GH is elevated around puberty and is associated with compensatory testicular hypertrophy and early onset of spermatogenesis after neonatal hemicastration (Kosco et al. 1987, al Haboby et al. 1988, Swanlund et al. 1995). To test whether hormonal supplementation of recipient mice with bovine GH could improve spermatogenic efficiency, recipient mice were treated with exogenous bovine GH. There was no significant difference in graft maturation and spermatogenic differentiation in mice that were treated with GH at the dose and time course tested compared with untreated controls. Nevertheless, the experiment served as an example for testing a hypothesis that changing the recipient endocrine environment could affect donor-derived spermatogenesis. Perhaps more importantly though, a significant effect of donor was noticed in this experiment with mice carrying tissue from one donor showing significantly better graft development and germ cell differentiation than those carrying tissue from the other donors. It is unclear what causes this effect of donor. However, the possibility of a donor effect on xenograft development and sperm production has to be taken into account when comparing results obtained by grafting of tissue from different donors. Wherever possible, treatments should be replicated within donor tissue to obtain meaningful results. Conversely, the ability to control for donor effects represents one of the major advantages of the approach of testis tissue xenografting.
In summary, characterization of bovine testis tissue xenograft development indicates that an initial deficit in the number of germ cells occurs in testis xenografts from 1-week old bull calves due either to germ cells not being able to re-populate themselves early on or to a loss in germ cells soon after the tissue is grafted. However, the low efficiency of sperm production cannot be attributed entirely to this initial deficit of germ cells. Rather, spermatogenesis appears to become arrested at meiosis in the majority of tubules, associated with a progressive dilation of seminiferous tubules, possibly due to an imbalance in fluid production and resorption. Supplementation of recipient mice with recombinant bovine GH did not have a significant effect on graft development, but the experiment demonstrated that comparisons of treatment effects have to be made within the donor to account for effects of donor tissue on graft development. The characterization of spermatogenic inefficiency in bovine testis xenografts will provide the basis to test hypotheses aimed at understanding the underlying mechanisms governing control of spermatogenesis in testis grafts and may also provide a platform to study effects on male fertility in vivo.
Percentage of seminiferous tubles with the most advanced germ cell stage present in bovine xenografts recovered 12 weeks after initiation of GH treatment of mice carrying the grafts.
|Most advanced germ cell stage (% seminiferous tubules)|
|GH||Donor 1||Donor 2||Donor 3|
|ND, no differentiation; PS, pachytene supermatocytes; ESp, elongated spermatids; RSp: round spermatids.|
|0 μg||ESp (3.4%)||PS (23.1%)||PS (15.4%)|
|25 μg||ND||PS (27.7)||ND|
|100 μg||ESp (3.3%)||RSp (8.2%)||PS (27.3%)|
|1000 μg||RSp (7.2%)||PS (18.3%)||PS (32.1%)|
(A Honaramooz is now at Western College of Veterinary Medicine, University of Saskatchewan, 52 Campus Drive, Saskatoon, SK, Canada S7N 5B4)
Grant support was provided by USDA/CSREES/NRICGP 2003-35203-13486. We thank Dr Jon Hill for providing the control tissue, James Hayden, RBP, for figure preparation and Terry Jordan for help with animal care. The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.
al Haboby AH, Loseth KJ, Wheaton JE & Crabo BG1988 Neonatal hemiorchidectomy of bulls alters plasma growth hormone levels and advances onset of pubertal testosterone secretion. Domestic Animal Endocrinology 5 61–69.
Curtis SK & Amann RP1981 Testicular development and establishment of spermatogenesis in Holstein bulls. Journal of Animal Science 53 1645–1657.
Fujisawa M, Yamanaka K, Okada H, Arakawa S & Kamidono S2002 Growth hormone releasing hormone test for infertile men with spermatogenetic maturation arrest. Journal of Urology 168 2083–2085.
Honaramooz A, Snedaker A, Boiani M, Scholer H, Dobrinski I & Schlatt S2002 Sperm from neonatal mammalian testes grafted in mice. Nature 418 778–781.
Honaramooz A, Li MW, Penedo MCT, Meyers S & Dobrinski I2004 Accelerated maturation of primate testis by xenografting into mice. Biology of Reproduction 70 1500–1503.
Keene DE, Suescun MO, Bostwick MG, Chandrashekar V, Bartke A & Kopchick JJ2002 Puberty is delayed in male growth hormone receptor gene-disrupted mice. Journal of Andrology 23 661–668.
Kosco MS, Bolt DJ, Wheaton JE, Loseth KJ & Crabo BG1987 Endocrine responses in relation to compensatory testicular growth after neonatal hemicastration in boars. Biology of Reproduction 36 1177–1185.
Oatley JM, Reeves JJ & Mclean DJ2004 Spermatogenesis and germ cell transgene expression in xenografted bovine testicular tissue. Biology of Reproduction 71 494–501.
Oatley JM, Reeves JJ & Mclean DJ2005 Establishment of spermatogenesis in neonatal bovine testicular tissue following ectopic xenografting varies with donor age. Biology of Reproduction 72 358–364.
Schlatt S, Honaramooz A, Boiani M, Scholer HR & Dobrinski I2003 Progeny from sperm obtained after ectopic grafting of neonatal mouse testes. Biology of Reproduction 68 2331–2335.
Shinohara T, Inoue K, Ogonuki N, Kanatsu-Shinohara M, Miki H & Nakata Ket al.2002 Birth of offspring following transplantation of cryopreserved immature testicular pieces and in-vitro microinsemination. Human Reproduction 17 3039–3045.
Snedaker AK, Honaramooz A & Dobrinski I2004 A game of cat and mouse: Xenografting of testis tissue from domestic kittens results in complete cat spermatogenesis in a mouse host. Journal of Andrology 25 926–930.
Swanlund DJ, Ndiaye MR, Loseth KJ, Pryor JL & Crabo BG1995 Diverse testicular responses to exogenous growth-hormone and follicle-stimulating-hormone in prepubertal boars. Biology of Reproduction 53 749–757.
Wrobel KH, Bickel D, Kujat R & Schimmel M1995 Configuration and Distribution of Bovine Spermatogonia. Cell and Tissue Research 279 277–289.