The Emerging Pathophysiology of Age-related Testicular Degeneration with a Focus on the Stallion and an Update on Potential Therapies

Authors


Author’s address (for correspondence): Regina M Turner, Department of Clinical Studies, New Bolton Center, 382 West Street Rd., Kennett Square, PA 19348, USA. E-mail: rmturner@vet.upenn.edu

Contents

Studies in laboratory rodents are shedding light on the pathophysiology of testicular ageing and now suggest a complicated basis for age-related declines in testicular function. A highly significant contributor to infertility may involve failure of specific and complex testicular microenvironments (niches) comprised of a variety of cellular and molecular components. Our laboratory has applied testis tissue xenografting to the study of testicular ageing in the stallion. Using this technique, we have confirmed that the disease is tissue autologous. As would be expected from a tissue autologous disease, hormonal and non-hormonal therapies designed to drive the function of the diseased testis are ineffective. However, we have some evidence that contact with young, normal testicular tissue may improve the condition of aged, degenerate testes. Perhaps, paracrine factors from young testicular cells may partially restore a young microenvironment and allow for the maintenance of testicular function. These findings form the basis for future studies designed to determine whether cells, genes or proteins from a normal testis can aid the function of a degenerate testis.

Introduction

Age-related declines in testicular function, including declines in both androgen and gamete production, can adversely affect male fertility, behaviour and even quality of life. In men, the effects of age-related decreases in testosterone are particularly well described and include changes in body composition and muscle strength, decreased energy, reduced sexual function, depressed mood and decreased cognitive function (Harman et al. 2001; Matsumoto 2002). The effects of ageing on human male fertility are more controversial, although the majority of evidence suggests that increasing paternal age has negative effects on fertility and also may increase genetic risks for offspring (Stewart and Kim 2011). Compared to humans, the effects of ageing on the fertility of domestic animal males are very evident, possibly because many of these males are expected to continue to breed large numbers of females efficiently well into old age, thus making even subtle declines in reproductive function apparent (Kumi-Diaka and Dennis 1978; Kumi-Diaka et al. 1981; Johnson 1989; Blanchard and Varner 1993; Watson et al. 1994).

Among domestic animal species, age-related testicular degeneration is particularly relevant in the horse, because genetically valuable stallions often must continue breeding throughout their relatively long lives. In stallions, age-related testicular degeneration most often affects middle-aged or older animals, but can be seen in much younger animals as well (Gehlen et al. 2001). Regardless of the age of onset, the degeneration is typically progressive and results in a steady decline in reproductive efficiency, sometimes ending in sterility. Economic losses resulting from this disease in the equine breeding industry are substantial, stemming from the losses of breeding fees, increased management costs and loss of valuable male genetics.

Although it has not been shown that all cases labelled as ‘age-related testicular degeneration’ are, in fact, affected by the same disease process (e.g. is the apparently spontaneous degeneration observed infrequently in young animals the same problem as the degeneration observed so commonly in ageing animals?), the progression of the disorder is similar. For the lack of better terminology, in this article, we will use the term ‘age-related testicular degeneration’ or simply ‘degeneration’ to refer to those cases in which testicular function and physical characteristics decline from previously normal levels, without an apparent inciting cause and regardless of age of onset.

The pathophysiology of testicular ageing has historically been poorly understood because the mechanisms involved with biological ageing are exceedingly complex. As a result, in stallions as in many domestic animal species, age-related testicular degeneration remains clinically characterized almost exclusively by the appearance of a common set of non-specific, progressive clinical signs including, but not limited to, an increase in palpable softness and a decrease in size of the testicular parenchyma and a decline in semen quality. Our work focuses on improving our understanding of the pathophysiology of testicular ageing in the stallion, with a long-term goal of developing methods to prevent and treat age-related testicular degeneration in the horse. This article will briefly review the broader literature on age-related declines in testicular function and will present some of our specific findings on testicular ageing in the horse.

Studying Testicular Ageing in Laboratory Model Species

The stem cell niche microenvironment

The testis is a complex organ divided roughly into two regions: the interstitial compartment and the seminiferous tubules. The interstitial compartment is comprised of Leydig cells (the primary androgen-producing cells of the testis), peritubular myoid cells, macrophages and endothelial cells. The seminiferous tubules consist of Sertoli cells and germ cells. Spermatogonial stem cells (SSCs) reside along the basement membranes of the tubules. The SSCs undergo both self-renewal and differentiation, and they are responsible for maintaining spermatogenesis, the continuous, highly organized process of germ cell differentiation that results in the production of spermatozoa. The Sertoli cells support the SSCs and all later stages of male germ cells through a complex network of direct cell–cell interactions, paracrine and endocrine factors. Spermatogenesis also relies on input from somatic cells in the interstitial compartment (Russell et al. 1990).

Recent work in rodents suggests a complicated basis for age-related declines in male fertility. Whereas there is now evidence for some direct effects of ageing on the SSC itself (Zhang et al. 2006; Schmidt et al. 2011), a highly significant contributor to infertility in old male mice was shown to result from the failure of the SSC niche, a specific and complex testicular microenvironment comprised of a variety of cellular and molecular components that externally contribute to the regulation of SSC function. In vivo cell retransplantation experiments were performed in which fluorescently labelled SSCs from donor mice were transplanted by efferent duct injection into young recipient mice that were devoid of endogenous spermatogenesis. Approximately 3 months after transplantation, recipient mouse testes were collected and enzymatically digested and the resultant cell suspension, containing daughters of the fluorescently labelled SSCs from the original donor mouse, was retransplanted into a new, young recipient. Retransplantation was performed serially over a period of years, always into young recipients. This study showed that if SSCs were maintained in a ‘young’ somatic niche environment, their function (including self-renewal and potentially differentiation through spermatogenesis) could be maintained well beyond what would be defined as old age in the donor (Ryu et al. 2006). These studies also showed clear evidence of a loss of SSC number in association with in vivo ageing, although those SSCs that did remain in old males remained functionally robust (Ryu et al. 2006). Thus, to understand ageing, we must study not only individual cell function, but also the broader niche environment.

In rodents, there is evidence that the SSC niche is influenced by factors from the interstitial compartment. Both Leydig cells and possibly peritubular myoid cells have been implicated in the ageing process and are believed to contribute to the degeneration of the SSC niche (Chen et al. 2009, 2010; Oatley et al. 2009). Sertoli cell function also is likely to affect the niche, but the role of Sertoli cells in the niche may be more indirect in that it is influenced by input from the interstitial compartment. Thus, it appears that numerous cell types and cell interactions are required to create a ‘young’ testicular environment. To better understand ageing in the testes of domestic animals, we will need to design experiments that study not only individual cell function and how it changes with age, but also cell–cell interactions, and the young somatic cell population as a whole.

Effects of testicular ageing on individual cell types

Focusing first on individual cell types, there are emerging data describing the effects of ageing on Leydig cell function. In Brown Norway rats, one of the best laboratory models for testicular ageing, it has been determined that Leydig cells from aged rats produce less testosterone than do Leydig cells from young rats (Liao et al. 1993; Zirkin et al. 1993; Chen et al. 1994). This results in an age-related decline in both serum and testicular testosterone concentrations. The decline in testosterone is not because of a decline in Leydig cell number, but rather is the result of a loss of steroidogenic function in the old Leydig cells (Chen et al. 1994). Researchers have identified several differences in aged Leydig cells that could adversely affect the function of the steroidogenic pathway, including differences in LH-receptor number, cAMP production, PK-A activity, cholesterol transport and activities of steroidogenic enzyme (Lin et al. 1983; Liao et al. 1993; Luo et al. 1996, 2005; Chen et al. 2002; Culty et al. 2002).

Molecules potentially involved in the ageing process also have been identified in other testicular cell types. A factor originating from both Leydig and peritubular myoid cells [colony-stimulating factor 1 (Csf1)] has been implicated as a stimulator of SSC self-renewal in mice (Oatley et al. 2009). Thus, changes in the function of even peritubular myoid cells could adversely affect the SSC niche and be involved in age-related declines in spermatogenesis. Additionally, an inability of the Sertoli cells to respond to signals (e.g. changes in FSH responsiveness) or to produce factors required to regulate SSC self-renewal and differentiation [e.g. glial cell-derived neurotrophic factor (GDNF)] has been implicated as part of the ageing process (Oatley et al. 2009). Disruption of Sertoli cell/germ cell communication also has been identified and is postulated to contribute to germ cell loss during ageing (Syed and Hecht 2001).

Other factors

On an interesting side note, it has been shown that the presence of female mice delays reproductive ageing in male mice. Male mice housed with females remained fertile longer and maintained testis weight longer than did males housed in isolation. Histological defects in spermatogenesis also were observed earlier in isolated males compared to those housed with females (Schmidt et al. 2009). These results have significant implications for the maintenance of male fertility in all species and raise the possibility of the existence of a host of external factors, such as pheromones, that could influence testicular ageing.

Restoring function to ageing testes

The growing body of information on the ageing process is now allowing researchers to study possible ways to prevent or reverse ageing in the testis. For example, there is good evidence that stem Leydig cells persist in adult testes (Lo et al. 2004; Chen et al. 2010) and it is interesting to speculate that repopulation of an aged testis with a new generation of ‘young’ Leydig cells derived from Leydig stem cells might improve the function of the SSC niche and enhance spermatogenesis. In this regard, it has been shown that putative stem Leydig cells in aged rat testes are able to repopulate the aged testes with adult Leydig cells with steroidogenic function equivalent to that of Leydig cells from a young adult (Chen et al. 1996). In other words, it is possible that stem Leydig cells do not age, or at least age more slowly than do adult Leydig cells, and thus have the potential to repopulate the testis with ‘young’, fully functional, adult Leydig cells at any point in an animal’s lifetime. Although the idea of introducing ‘young’ Leydig cells into an old testis as a cure for ageing is appealing, biology is seldom that simple and it is likely that, like the SSC niche, a complex and probably overlapping Leydig cell niche also exists [reviewed in (Chen et al. 2010)] and that both the entire SSC niche and the entire Leydig cell niche would need to be recreated to allow for full restoration of ‘young’ testicular function.

On a side note of interest, both bone marrow stem cells and even adipose tissue-derived mesenchymal stem cells can be transformed into steroid-producing cells very similar to Leydig cells (Gondo et al. 2008; Yazawa et al. 2009). This provides for the possibility of a stem cell-based therapeutic approach to androgen insufficiency in ageing males that relies on readily accessibly tissue (bone marrow or fat) rather than the much more elusive stem Leydig cell.

Studying Testicular Ageing in the Stallion

Background

Studies on ageing in the stallion testis have been limited. A decline in intratesticular inhibin was the first detectable endocrinologic change reported in a group of ageing, subfertile stallions (many of which were likely affected by testicular degeneration) (Stewart and Roser 1998). Decreases in testosterone, estrogens and germ cell numbers occurred later in the progression of the disease. Because inhibin is a product of the Sertoli cells, it was therefore suggested that the primary defect in equine age-related testicular degeneration resides within the Sertoli cell. Similar endocrinologic findings have been reported in elderly men (Tenover et al. 1988). Some researchers also have reported a decline in absolute numbers of Sertoli cells with age in the stallion, while others have found no difference in number with age (Jones and Berndtson 1986; Johnson et al. 1991).

Although studies focusing on individual cell types (such as Sertoli cells) remain highly valuable, spermatogenesis involves coordinated input from a variety of cell populations including germ cells and somatic supporting and interstitial cells (Russell et al. 1990). Unfortunately, reproducing the complexity of spermatogenesis in an in vitro system has so far remained largely elusive (Lee et al. 2001). Thus, more difficult experiments involving live animals are required to study the broader microenvironments seen in vivo. Even in mice, these experiments are exceedingly difficult and costly and typically are performed well only by some laboratories. But at least this work is facilitated by the many tools and techniques that are available to those working in the mouse system. Researchers wishing to ask similar questions about domestic animals find themselves very limited in comparison and often are forced to do what they can in a poorly controlled and very costly system.

Testis tissue xenografting has been developed as an alternate approach for the study of spermatogenesis and testicular function in mammals. Unlike in vitro systems, xenografting maintains the complex interactions of germ cells and their somatic supporting cells. Thus, xenografting allows the investigator access to the spermatogenic system while also providing the opportunity to perform well-controlled studies in testicular tissue that supports normal spermatogenesis, all without the need for extensive experimentation in live animals that is often logistically or ethically unacceptable.

Xenografting involves grafting small pieces of testicular tissue from a donor animal under the back skin of castrated, immunocompromised mice. It was initially shown that xenografting of testis tissue from newborn pigs and goats results in the production of normal, functional donor sperm in the mouse host (Honaramooz et al. 2002). Histologically, spermatogenesis within the xenografts was indistinguishable from spermatogenesis from donor testes that remained in situ. Subsequently, it was shown that xenografting of testicular tissue could be applied to a variety of species. However, the efficiency of spermatogenesis in the xenograft was species-dependant, ranging from normal, full spermatogenesis, to more limited spermatogenesis that developed in only a percentage of seminiferous tubules (Schlatt et al. 2002; Oatley et al. 2004, 2005; Snedaker et al. 2004).

We chose to use xenografting as a means of studying the complex testicular environment in the horse. We first reported on the success of xenografting of normal prepubertal and young adult horse testicular tissue to castrated mice (Rathi et al. 2006; Turner et al. 2006, 2010a). The efficiency of spermatogenesis in equine testicular xenografts varied among donors. Equine spermatogenesis, typically including meiotic cells and occasionally including the development of haploid cells, did occur on the mouse hosts; however, full spermatogenesis, including the development of testicular sperm, was not observed in any equine grafts. Nonetheless, these experiments did ‘set the bar’ for what to expect in grafts of normal equine testicular tissue.

Establishing the utility of xenografting for the study of testicular pathology in the stallion

We elected to use the stallion as a model to study the application of xenografting to the study of testicular pathophysiology. Specifically, our aim was to compare the effects of testis tissue xenografting on two different conditions that result in an absence of spermatogenesis: cryptorchidism and severe age-related testicular degeneration.

Bilaterally cryptorchid individuals are sterile because of an absence of spermatogenesis in the abdominal testes. The adverse effects of cryptorchidism on spermatogenesis are hypothesized to be due to the exposure of the testis to core body temperature, which is several degrees higher than scrotal temperature. The testis itself is thought to be otherwise normal, at least initially. We demonstrated that xenografts of small fragments of testicular tissue from abdominally cryptorchid colts and stallions reconstitute equine spermatogenesis up to the development of pachytene spermatocytes, similar to the results observed within grafts of normal, prepubertal equine testicular tissue. Thus, placing these testes into the permissive environment of the mouse host corrected for an extratesticular problem (high abdominal temperature) and resulted in the restoration of testicular function (Turner et al. 2006, 2010a). The cryptorchid model serves as an example of complete spermatogenic arrest because of an extratesticular defect and demonstrates that, when the tissue itself is normal, and when spermatogenesis is absent at the time of grafting, restoration of a normal extratesticular environment through xenografting can result in a rescue of spermatogenesis in adult equine testicular tissue to a degree comparable to that seen in prepubertal equine xenografts (Rathi et al. 2006). This finding is similar to results reported for adult murine testicular tissue in which spermatogenesis has been suppressed (Arregui et al. 2008; Rodriguez Sosa and Dobrinski 2009). This validated testis tissue xenografting as an assay for the functional potential of the grafted tissue.

Up to this point, all of our normal prepubertal testicular samples and most of our cryptorchid testicular samples had been obtained from relatively young animals (all under the age of 6 years and most under the age of 2). In contrast, most horses affected by age-related testicular degeneration are significantly older. This created a potential complicating factor when we extended our studies to include xenografts from aged stallions with testicular degeneration. We therefore designed a study to determine whether or not graft survival and spermatogenic differentiation were dependant on the age of the donor tissue at the time of grafting. Our results showed that cryptorchid testicular tissue from adult stallions supporting little or no spermatogenesis at the time of xenografting, survived grafting and reconstituted spermatogenesis more efficiently than did age-matched, normal descended testicular tissue that supported full spermatogenesis at the time of grafting (Turner et al. 2010a). These observations are consistent with work in mice showing that graft survival and spermatogenic differentiation are dependant not on the age of the tissue, but on the degree of spermatogenesis present in the tissue at the time of grafting (Arregui et al. 2008; Rodriguez Sosa and Dobrinski 2009). Thus, given age-matched adult samples, tissue with less spermatogenesis at the time of grafting should perform better as xenografts than tissue with full spermatogenesis. These findings set the stage for the application of xenografting to the study of age-related testicular degeneration in adult stallions.

Applying xenografting to the study of age-related testicular degeneration in the stallion

Earlier endocrinologic studies suggested that age-related testicular degeneration in the stallion arises from a primary testicular defect (Roser 1995, 1997; Motton and Roser 1997; Stewart and Roser 1998). We proceeded to apply xenografting to the study of age-related testicular degeneration with the hypothesis that abnormal spermatogenesis in testicular tissue affected by age-related testicular degeneration is tissue autologous. If this hypothesis was correct, we expected that using xenografting to transfer affected testicular tissue to the permissive environment of the mouse host would not benefit the condition of the degenerate tissue.

At the time of grafting, degenerate tissue samples supported only abnormal and inefficient spermatogenesis. In this regard, tissue affected with age-related testicular degeneration was most similar to cryptorchid and prepubertal tissue prior to grafting in that all of these tissue types contained very low levels of or no spermatogenesis. Given this, we would predict that, if the tissue is otherwise normal, xenografts of degenerate testicular tissue should perform better than xenografts from age-matched normal adult testes. However, while both cryptorchid and prepubertal tissue responded positively to xenografting by reconstituting some degree of spermatogenesis, all grafted degenerate tissue consistently degenerated further and did not reconstitute spermatogenesis to any degree over the 7-month period of grafting (Turner et al. 2010a). Thus, unlike cryptorchid and prepubertal stallions, the absence of spermatogenesis in testes from stallions with end-stage age-related testicular degeneration is tissue autologous and so spermatogenesis cannot be rescued by returning the diseased tissue to the permissive extratesticular environment of the mouse host. These findings supported our hypothesis, and the work of Stewart and Roser, by providing direct evidence that a primary testicular defect is the cause of age-related testicular degeneration in the stallion. However, our data did not address which cell type, if any, is the first to ‘fail’ in ageing equine testes.

Xenografting experiments fail to identify a beneficial effect of gonadotropins on age-related testicular degeneration

Our inability to reconstitute spermatogenesis in grafted equine testicular tissue affected with age-related testicular degeneration prompted us to search for methods that might improve the outcome. Our earlier studies indicated that the progression of spermatogenesis in prepubertal equine testis xenografts was improved by treatment with subcutaneous injections of exogenous eCG and hCG (10 I.U. each twice weekly) (Rathi et al. 2006). In this regard, several hormone-based therapies have been applied clinically as treatments for ageing stallions with age-related testicular degeneration, largely with little or no scientific evidence as to their efficacy. Herein lies one of the greatest advantages of xenografting. The technique allows us to perform well-controlled studies on horse testicular tissue without the need for performing experiments on the donor species. Thus, for relatively minimal expense, we were able to test the efficacy of a variety of treatments on testicular disease in stallions. We began by supplying some mice carrying degenerate xenografts with exogenous gonadotropins to determine whether the there might be a beneficial effect that could be developed and applied clinically to affected stallions. Unlike the prepubertal grafts, we found no effect of exogenous gonadotropins on these degenerate tissue grafts.

The small size of the seminal vesicles of mice carrying xenografts of degenerate tissue indicated that these grafts were not producing significant amounts of bioactive testosterone. Thus, we also considered the possibility that the stallion tissue might not be able to set-up a normal reproductive regulatory axis on the mouse host. In contrast, it has been shown that porcine spermatogenesis is fully reconstituted in testicular xenografts from prepubertal boars. Additionally, these grafts produce bioactive testosterone as measured by significant increases in size of the seminal vesicles of the host mice (Honaramooz et al. 2002). We therefore investigated whether co-grafting of porcine testicular tissue with degenerate equine testicular tissue would improve the condition of the degenerate xenografts. Again, we found no evidence that the endogenous testosterone supplied by the fully developed and endocrinologically active porcine grafts affected the outcome of the degenerate grafts (Turner et al. 2010a).

These findings impact the prognosis for age-related testicular degeneration and suggest that hormone-based treatments designed to drive the function of the diseased testis are unlikely to have a positive impact on the progression of age-related testicular degeneration in stallions. There are several possible reasons for the lack of effect of exogenous gonadotropins on spermatogenesis in degenerate xenografts. The first is that murine/equine gonadotropin differences may not be the primary cause of the observed spermatogenic inefficiency. This would be consistent with the hypothesis that the defect in age-related testicular degeneration lies within the testis and not with a primary endocrine deficiency. However, our study explored only one treatment regimen that had previously been effective in supporting spermatogenesis in equine and primate xenografts (Honaramooz et al. 2004; Rathi et al. 2006). Therefore, we cannot rule out that different dosage of exogenous hormones, different durations of treatment or different agents may have an effect on degenerate testis grafts. However, given the severe and rapid degeneration that was observed in the degenerate xenografts, it seems unlikely that changes in formulation, dosage or dosing frequency of gonadotropins alone would be enough to rescue this tissue and restore spermatogenesis.

Equine chorionic gonadotropin and hCG were chosen for our horse studies because these hormones had been used with some success in enhancing spermatogenesis in xenografts of normal horse testicular tissue (Rathi et al. 2006). Human chorionic gonadotropin is known to have LH-like effects in the horse and is therefore likely to be a good choice for use in our equine studies. In contrast, although eCG is known to have FSH-like effects in other species, it appears to have little FSH activity in the horse, and therefore, other agents may be preferable for treating mice carrying equine xenografts. Since that time, we have repeated these experiments using hCG and equine FSH to treat the recipient mice. We saw no improvement in the degenerate grafts (our unpublished data).

It is worth noting that our studies to this point focused only on tissue from stallions with severe, end-stage, age-related testicular degeneration. In the future, it will be important to determine at what point in the disease process the testis becomes unsalvageable. Is the testicular defect the primary cause of the degeneration, or does the testicular defect arise secondarily during the progression of the disease? As long as SSCs persist, there remains at least the possibility that the testis could be repopulated with germ cells. However, given the intricate testicular microenvironments that are required for normal SSC function, it is likely to be a difficult question to address.

In this regard, one study of 208 men with mild, moderate or severe idiopathic oligospermia concluded that those individuals with more severe oligospermia were more likely to progress to complete azoospermia, while those with mild or moderate oligospermia were more likely to remain stable or even improve over time (Won Bak et al. 2010). However, the authors made no attempt to determine the cause of these patients’ oligospermia and it was not clear how many, if any, were affected by testicular degeneration. Thus, the question of whether this same pattern would apply to individuals specifically affected with testicular degeneration remains unanswered. We have begun xenografting experiments with tissue from animals with more mild degenerative changes in an attempt to address this question in stallions. Perhaps, treatment with exogenous gonadotropins would be more effective if applied to earlier stages of the disease.

Xenografting experiments fail to identify a beneficial effect of non-hormonal therapies on age-related testicular degeneration

Many non-hormonal therapeutic strategies have been put forth as potential treatments for stallion subfertility often associated with age-related testicular degeneration, although for some of these, there is limited support from the scientific literature. Well-controlled studies on the effects of these treatments are fraught with difficulty and typically are quite expensive. Testis tissue xenografting now provides a means to test these compounds in a carefully controlled setting for relatively little expense and without the need for experiments on the donor stallions. We thus applied xenografting to study several agents that are widely used clinically to treat subfertile stallions, including animals affected by age-related testicular degeneration: long-chain Omega-3 polyunsaturated fatty acids [e.g. Docosahexaenoic acid (DHA)], antioxidants and polyamines. We hypothesized that if age-related testicular degeneration is tissue autonomous, then these treatments would be ineffective, particularly during the later stages of the disease.

Docosahexaenoic acid is one of the better-researched compounds, and a beneficial effect of DHA on sperm motility has been described in some stallions. In a 2005 study, Brinsko et al. showed that dietary supplementation of DHA resulted in some improvements in motility parameters following cooled semen storage and cryopreservation in a group of fertile stallions (Brinsko et al. 2005). Similar findings were subsequently reported in bulls (Gholami et al. 2010). These improvements were more notable in the subgroup of stallions whose sperm typically did not fare well under cooled storage conditions. DHA’s benefits most likely occur by incorporation of the compound into the sperm plasma membrane (Harris et al. 2005) and, although fertility trials were not performed in the study by Brinsko et al., the data held promise that feeding a DHA neutraceutical could be beneficial to fertility in a specific subgroup of breeding stallions. However, it seems unlikely that this would be a panacea for the treatment for stallion subfertility, particularly (as is the case for age-related testicular degeneration) in those cases where the subfertility is related to complex abnormalities of spermatogenesis. Nonetheless, it has become widespread practice to feed many subfertile stallions a DHA supplement, regardless of the cause of their subfertility.

There is evidence in a variety of species that oxidative stress may lead to sperm damage. As a result, antioxidant therapy has been applied to the treatment for male infertility, although in most cases, data confirming a beneficial effect on fertility are lacking (Lombardo et al. 2011). Data in rats suggested that senile testicular degeneration may result from free radicals produced as a side effect of hormone production within the testis itself (Chen and Zirkin 1999). Thus, one could surmise that provision of antioxidants might slow or reverse the problems associated with testicular degeneration in the stallion. Antioxidants are now commonly used as feed supplements to support fertility in subfertile stallions.

Polyamines are known to be essential to some reproductive processes and their absence can lead to infertility. In males, there is some evidence that polyamines benefit sperm motility (Lefevre et al. 2011) and thus these compounds also have been applied to the treatment for infertility in stallions.

We used xenografting to test the effects of three commercially marketed equine feed additives on age-related testicular degeneration in stallions by supplementing the diets of recipient mice carrying degenerate xenografts. The first additive was a commercial mixture of DHA and antioxidants including alpha lipoic acid, vitamin E and vitamin C [Equine Platinum Potency® (EPP), Platinum Performance, Buellton, CA, USA]. The second additive was a similar commercial preparation containing a patented blend of marine-based long-chain polyunsaturated Omega-3 fatty acids [including DHA and Eicosapentaenoic acid (EPA)], antioxidants and vitamins [Magnitude™ (MG); Breeder’s Choice LLC, Aubrey, TX, USA]. The third additive was a liquid formula of polyamines, including both spermine and spermidine [SpermAid (SA); Hagyard Pharmacy, Lexington, KY, USA].

Testicular tissue samples were obtained following castration of two aged stallions: one 30-year-old stallion with advanced age-related testicular degeneration (6-year history of declining fertility in association with significant decreases in testicular size and semen quality together with histologic evidence of advanced degeneration) and one 23-year-old stallion with early signs of the disease (moderate decreases in testicular size and semen quality together with histologic evidence of mild degeneration). Small fragments of testis tissue (1 mm3) from each stallion were grafted under the back skin of anesthetized male scid mice through small incisions as previously described (Honaramooz et al. 2002). Twelve recipient mice were used per donor, and six tissue pieces were grafted onto each mouse. The mice were castrated at the time of grafting. Comparable tissue pieces also were fixed in Bouin’s solution at the time of grafting to serve as a reference points for graft development.

For each donor stallion, three recipient mice carrying a total of 18 grafts were kept as untreated controls, three recipient mice carrying a total of 18 grafts were fed the compounded diet containing Equine Platinum Potency®, three recipient mice carrying 18 grafts were fed the compounded diet containing Magnitude™, and three recipient mice carrying 18 grafts were supplemented with SpermAid in the drinking water (Fig. 1). We extrapolated the murine dose for each agent from the equine daily dosage, while correcting for differences in metabolism between mice and horses. For most adult mammals, one can correct for differences in metabolic rate by raising body weight to the 0.75 power (Kleiber 1947). For example, the standard daily equine dose for PP is 28 000 mg for a 500 kg animal. Metabolic body weight (MBW) for a 455 kg horse is 4550.75 = 105.7 kg.

Figure 1.

 Twelve mice were used for each of two donor stallions. Each mouse received six grafts of degenerate tissue. Three mice per stallion were kept as untreated controls. The remaining nine mice were treated as shown

  • 28 000 mg/105.7 kg = 0.265 mg/g equine MBW

  • MBW for an average 0.035 kg mouse is 0.0350.75 = 81 g

  • 0.265 mg/g × 81 g = 21.5 mg daily

Because an average mouse consumes approximately 3 g of pellets per day, our PP diet was formulated to contain 7.2 g PP per kg of base pellets (Teklad Diets, Harlan Laboratories, Madison, WI, USA). Similar calculations were made for MG. The dose of SA was calculated similarly and added to drinking water at an appropriate volume. Treated mice were housed individually, and all treated mice consumed at least the minimum daily dose of their respective compound based on daily measurement of pellet or water intake.

One mouse for each treatment group and one control mouse were killed at 4, 6 and 8 months post-grafting. The skin tissue containing the grafts was dissected, and grafts were fixed in Bouin’s solution before being processed for histology using haematoxylin and eosin staining. Each graft was analysed microscopically for the presence of seminiferous tubules. A tubule was considered ‘healthy’ if it was expanded, had an intact basement membrane around its circumference and contained distinct cell types. If the graft contained even one healthy seminiferous tubule, it was classified as ‘healthy’. If a graft contained no healthy seminiferous tubules, it was classified as ‘degenerate’. For healthy grafts, all healthy seminiferous tubule cross-sections were counted and examined for the evidence of spermatogenesis. Each tubule was classified based on the most advanced cell type present within the tubule (Sertoli cell only, spermatogonia, spermatocyte, round/condensing spermatid, elongating spermatid or testicular sperm). The data obtained from grafts from a single mouse were pooled, and the average of the results from all the mice in an experimental group was analysed. Seminal vesicles from all recipient mice were weighed as an indication of secretion of bioactive testosterone by the xenografts.

The number of degenerate vs healthy grafts recovered was analysed by comparing data from each paired control and treated mouse. Similarly, the number of healthy tubules was compared between control and treated mice. The Wilcoxon signed rank test was used for all statistical analyses, and a p-value of <0.5 was considered significant.

Regardless of treatment, at the time of graft harvest, 100% of the grafts had degenerated further compared to their condition at the time of grafting. All xenografts shrank in size over time such that, at the time of graft harvest, some were no longer grossly identifiable and all were <25% of their original size. These findings are similar to those we have reported previously for xenografts of degenerate tissue in either untreated or gonadotropin-treated mice (Turner et al. 2010a). Most of the recovered grafts were classified as degenerate based on the histologic absence of seminiferous tubules. Even the few grafts that were classified as healthy contained very few healthy tubules (mean 2.8 ± 1.2 healthy tubules per healthy graft) and none contained cell types other than Sertoli cells. There was no significant difference in the number of healthy or degenerate grafts recovered from control or treated mice, nor was there a difference in the number of healthy seminiferous tubules in control grafts compared to grafts from any of the treatment groups from either of the donor stallions. The lack of significance was maintained whether data were analysed individually for each stallion or pooled for both stallions. Weights of the seminal vesicles in all host mice were <50 mg, indicating that bioactive testosterone was not being produced by the grafts (SV weight: 22.6 mg ± 8.2; mean ± SD; range, 7–43 mg, n = 24 mice) (Schlatt et al. 2003). This is as would be expected from a tissue autologous disease, because treatments designed to drive testicular function are unlikely to be efficacious.

There are certainly limitations to applying xenografting to the study of the effects of pharmacologic agents on the grafted tissue. For example, we cannot be certain of the appropriate murine dose for each tested compound, because metabolic studies on these compounds have not been conducted in mice. Thus, it is difficult to interpret the significance of our negative results. Nonetheless, within the acknowledged limits of this system, we identified no beneficial effect of PP, MG or SA on the condition of testicular tissue affected either moderately or severely by age-related degeneration.

Xenografting experiments do provide evidence for beneficial effects of young testicular tissue on old testicular tissue

Of great interest, we found that if healthy, prepubertal tissue is co-grafted in physical contact with degenerate testicular tissue, we observe an improvement in the degenerate grafts. Specifically, these grafts degenerate only partially and maintain an abundance of healthy seminiferous tubules throughout most of the tissue. In some instances, early-stage germ cells are maintained and there is evidence of limited spermatogenesis (Turner et al. 2010b). These findings suggest that there is a beneficial effect of the young tissue on the old tissue. In other words, as might be expected from a tissue autologous disease, improvement in the disease is seen only when the ‘treatment’ originates from within a normal, young testis. It is possible that paracrine factors from young testicular cells may partially restore a young microenvironment and allow for the maintenance of germ cells. These findings form the basis for future studies designed to determine whether cells, genes or proteins from a normal testis can aid the function of a degenerate testis.

Conclusion

Work in laboratory rodents has provided us with exciting information on the biology of ageing in the mammalian testis. We now can build on this information and apply modern experimental techniques to the study of testicular function and ageing in domestic animals. In stallions, there is good evidence that age-related testicular degeneration is tissue autologous. And, as would be expected for a tissue autologous disease, there is little to no evidence of a beneficial effect of a variety of hormonal or non-hormonal agents on the condition. However, xenografting experiments now raise the possibility that cells or factors from young testicular tissue might be used to benefit ageing, degenerate testicular tissue in the horse. Our future work will focus on identifying what cell type(s) in the young testis might be responsible for this beneficial effect and to determine what factors within these healthy cells promote the function of the aged tissue. It is our hope that we may be able to build on this information to develop cell- or even protein-based therapies for age-related testicular degeneration in the horse.

Acknowledgements

Some work described in this manuscript was supported by National Research Initiative Competitive Grant no. 2007-35203-18213 from the USDA Cooperative State Research, Education, and Extension Service, The Grayson-Jockey Club Research Foundation, Inc., the Raymond Firestone Research Fund at New Bolton Center, and the Robeson Research Endowment through the Department of Clinical Studies at New Bolton Center. Our thanks to Dr. Ina Dobrinski for invaluable support. Additional thanks to Mark Modelski and Yun Li for their technical expertise and to Terry Jordan for help with animal care.

Conflicts of interest

The authors declare that they have no financial or personal relationships with other people or organizations that might inappropriately bias or influence this work.

Author contributions

R Turner designed the study, completed the statistics, reviewed the literature and wrote the manuscript. W Zeng performed the xenografting and was responsible for Animal care and data collection.

Ancillary