Animal models of physiologic markers of male reproduction: genetically defined infertile mice.

The present report focuses on novel animal models of male infertility: genetically defined mice bearing single-gene mutations that induce infertility. The primary goal of our investigations was to identify the reproductive defects in these mutant mice. The phenotypic effects of the gene mutations were deciphered by comparing the mutant mice to their normal siblings. Initially testicular steroidogenesis and spermatogenesis were investigated. The physiologic markers for testicular steroidogenesis were steroid secretion by testes perifused in vitro, seminal vesicle weight, and Leydig cell histology. Spermatogenesis was evaluated by the enumeration of homogenization-resistant sperm/spermatids in testes and by morphometric analyses of germ cells in the seminiferous epithelium. If testicular function appeared normal, we investigated the sexual behavior of the mice. The parameters of male sexual behavior that were quantified included mount patency, mount frequency, intromission latency, thrusts per intromission, ejaculation latency, and ejaculation duration. Females of pairs breeding under normal circumstances were monitored for the presence of vaginal plugs and pregnancies. The patency of the ejaculatory process was determined by quantifying sperm in the female reproductive tract after sexual behavior tests. Sperm function was studied by quantitatively determining sperm motility during videomicroscopic observation. Also, the ability of epididymal sperm to function within the uterine environment was analyzed by determining sperm capacity to initiate pregnancy after artificial insemination. Together, the experimental results permitted the grouping of the gene mutations into three general categories. We propose that the same biological markers used in the reported studies can be implemented in the assessment of the impact that environmental toxins may have on male reproduction.


Introduction
The end product of successful male reproduction is the deposition of spermatozoa capable of fertilization into the female reproductive tract. To achieve this end, the synchronization of a cascade of events is required. These events include the release of gonadotropin from the pituitary, testicular testosterone biosynthesis, production of functional spermatozoa, and the complex behavior of ejaculation (1)(2)(3)(4)(5). The aim of our research program is to elucidate the genetic control of the individual events that constitute male reproduction.
Genetically defined mice are a major aspect of our experimental approach. Mice were chosen as the animal model because they are mammals that exhibit reproductive characteristics approximating those of men (6) and because of the voluminous knowledge of mouse genetics (7)(8)(9). The onrginal working hypothesis was that mutations of single genes controlling male reproduction would be expressed phenotypically as either infertility *Department of Cell Biology and Anatomy, University of Texas Health Science Center at Dallas, Dallas, TX 75235. or reduced fertility. An exhaustive search of the literature revealed that mutations at 40 genetic loci had been reported to induce male infertility. The paucity of information about the mice bearing the gene mutations was striking. The goal of the described studies was to elucidate the phenotypic effects of the single-gene mutations that induce male infertility in mice. This goal was accomplished by implementing a spectrum of biological markers of male reproduction. In this report, we present an overview of our experimental results while emphasizing the biological markers. We propose that the same biological markers can be used to assess the impact that environmental toxins have on male reproduction.

Experimental Approach
The power of the experimental approach selected for the studies lies in the fact that the phenotypic effects of each gene mutation could be defined by comparing sibling mice: one sibling being normal or an unaffected heterozygote (+ / -) and the other sibling being the mu-tant. The primary difference between the two mice was the single gene mutation. This paradigm permitted the differences between the two mice to be assigned to the gene mutation. At present, we cannot state in every case if the effect is direct or a result of spurious pleiotropy (10).

Genetically Defined Mice
The review of the literature identified the gene mutations reported to cause infertility in male mice. A list of these gene mutations is presented in Table 1. Instead of supplying unabridged references in the table, I have cited primary literature sources that refer to the gene mutation's effect on male fertility. The following reviews provide comprehensive information about the gene mutations: (7,(16)(17)(18)(19)(20)(21)(22)(23)(24). Although I did not include chromosomal variants (25), the T locus and t-haplotypes (26,27), or the mutations at the c and p loci (7), all of these have been reported to influence male reproduction. Gene mutations that have been reported recently in the Mouse News Letter are not included in Table 1 because they may not be available to all investigators. Finally, the Mouse News Letter (8) is the most valuable publication for current information about mouse genetics (including recent linkage maps) and for identifying the sources of mutant mice (see Mouse News Letter number 76 for the 1986 mouse gene list). During our studies, the mutant and control mice were either obtained directly or derived from breeding pairs generously provided by individual scientists or the Jackson Laboratory.

Biological Markers of Male Reproduction
Biological markers were selected on the basis of their potential to assess the physiological function of the individual events that constitute male reproduction. A brief discussion of biological markers for each event follows. The reader will be referred to publications containing more details about the experimental procedures and alerted to possible difficulties.

Hypothalamus-Pituitary Function
The function of the hypothalamus-pituitary axis was assessed directly by measuring plasma concentrations of three hormones: luteinizing hormone (LH), folliclestimulating hormone (FSH), and prolactin (PRL). LH, FSH, and PRL are the primary secretions of the anterior pituitary that modulate the function of Leydig and Sertoli cells (18). In our studies, one plasma sample from each animal was analyzed in order to detect significant differences in hormone concentration; however, multiple blood samples from each animal would be necessary to obtain accurate measurements because of the episodic secretion of the hormones (28). Assay details can be located in the following references: LH (29), FSH (30), and PRL (31).

Testicular Steroidogenesis
Testosterone is the predominant steroid secreted by mammalian testes and is considered the primary androgen (32). Two conventional methods were used to estimate in vivo testosterone secretion: radioimmunometric determination of plasma testosterone levels (33) and quantitation of the weight of the androgen-dependent seminal vesicles. Possible problems with these markers include the pulsatile release of testosterone (34) and the variable response of organs to androgen stimulation (35).
The steroidogenic capacity of testes was quantified by perifusing the testes in vitro with a defined perifusion medium (33). Each testis was maximally stimulated with LH. Initially, the venous effluent from the in vitro perifused testes was analyzed for testosterone with a specific antibody (33). If testosterone secretion by mutant mouse testes was decreased, we constructed a secretion profile of testosterone biosynthetic intermediates by using a gas chromatograph equipped with capillary columns (36). Alterations in specific steroidogenic reaction activities could be detected by comparing the steroid secretion profiles of the mutant and normal mouse testes (37).
In vitro perifusion provides several important features for assessing the steroidogenic potential of testes. For example, the complex interrelationship between seminiferous tubules and Leydig cells is maintained, secretory products are removed via the blood vasculature thereby minimizing product inhibition, steroid secretion is continuous and can be maximally stimulated, steroidogenesis results only from endogenous cofactors and substrates, and the controlled conditions approximate those occurring in vivo.
The biochemical analysis of Leydig cell function was complemented with microanatomical studies. In these studies, testes were perifusion-fixed, embedded in methacrylate, and stained with toluidine blue (38). Morphometric methods (38) were used to determine the Leydig cell volume and mass. Concomitantly, Leydig cell morphology was examined for gross alterations.

Spermatogenesis
Two methods were used to assess spernatogenesis: enumeration of homogenization-resistant sperm/spermatids (39,40) and germ cell morphometric analysis (41). In the first method, phase-contrast microscopic analysis of sperm/spermatid nuclei in testicular homogenates yielded a general estimation of spermatogenesis. This method is usually accurate unless the gene mutation alters the physical properties of spermatid nuclei and renders them less resistant to homogenization. The spermatogenic efficiency can be expressed as sperm/ spermatids per milligram testis, thereby negating the effect of differences in testis weight. Interestingly, the experimental data provided by our homogenization method revealed that -200,000 sperm or spermatids/ mg testis was the value for normal spermatogenesis regardless of genotype.
If a perturbation in spermatogenesis was indicated by the homogenization data, sections of the perifusionfixed testes were morphometrically analyzed for the nuclear volume fraction (%) of five germ cell classes: spermatogonia, early spermatocytes (leptotene, zygotene), late spermatocytes (pachytene, diplotene), round spermatids, and elongated spermatids (42). More precise determinations of specific stages could be performed when the testis sections were stained with Sidman's acid fuchsin-toluidine blue (43).

Function of Male Genital Ducts and Accessory Sex Glands
The genital ducts comprise the epithelium-lined tubes that transport sperm from the testis to the exterior. These ducts also contribute to the maturation and protection of sperm during their transit. The male accessory sex glands, evaginations of the genital ducts, secrete the seminal fluid that mixes with sperm during ejaculation (44).
The function of these ducts and glands was investi-  gated by anatomical studies and by indirect assessments of function. The weights of seminal vesicles, preputial glands, epididymides, and penes were determined. Additionally, the morphology of penes was microscopically examined for the presence of key structures including the corpus spongiosum, corpus cavernosum, os penis, and epithelial papillae (45). The patency of efferent ductules and epididymal ducts was elucidated by determining the number of sperm/spermatids in epididymal

Sperm Function
Sperm function was assessed by in vivo and in vitro methods. The most critical test was the ability of the mice to sire litters. If the mice did not sire litters, we determined if the sperm could successfully enter the uterus by enumerating the sperm in uterine homogenates 1 hr after ejaculation. Additionally, we quantitatively measured the velocity of epididymal sperm by a videomicroscopic method similar to that reported by Katz and Overstreet (46). In our tests, we cut the cauda epididymides at their junction with the vas deferens and expressed the epididymal sperm into either modified Toyoda medium (47) or Alvarez medium (48). After incubation, the following parameters were measured: net velocity, average path velocity, and linear index (49). Also, the video recordings permitted analyses of sperm size and percent motility.
The fertilizability of epididymal sperm under uterine conditions was assessed by artificial insemination of hormonally-primed female mice and the subsequent quantification of live, near-term fetuses. Artificial insemination was performed similar to the vaginal route method reported by Wolfe (50)  homogenates with the aid of phase-contrast microscopy (38). Finally, the normal function of the entire system of ducts and accessory sex glands was indicated by the presence of a vaginal (copulatory) plug and sperm in the female reproductive tract following ejaculation.

Male Sexual Behavior
Male sexual behavior was analyzed quantitatively by observing male mice paired with females induced into behavioral estrus by hormonal injections (52). A series of behavioral parameters were observed and recorded during repetitive tests. These parameters included mount latency, intromission latency, ejaculation latency, number of head mounts, mounts without intromission (time per mount, total number), mounts with intromission (time per mount, total number, total number of intravaginal thrusts), and ejaculation duration (53,54). In the intervals between sex behavior tests, female mice paired with male mice were inspected for vaginal plugs. The presence of vaginal plugs was considered as evidence that the male mice had the capacity for male sexual behavior. The results from the sex behavior tests and natural breeding trials were compared to determine if the more stressful conditions of the sex behavior tests inhibited the male mice.

Utilization of Biological Markers
The biological markers ofmale reproduction were validated for use with mice. However, all of the biological markers were not assessed for each genotype. For example, if mouse testes were proven to be devoid ofgerm cells, we would not analyze sexual behavior or sperm function. A specific example from our stubby mouse studies (55) is cited to demonstrate the sequence in which the biological markers could be used to define the phenotypic effects of a gene mutation on male reproduction (Fig. 1). Animal Models of Male Infertility The biological markers of male reproduction were used initially to classify gene mutations according to their effect on testicular function. Based on these initial results, certain gene mutations were selected for additional study. A brief description of the reproductive biology of the mutant mice studied in our laboratory is presented in the following paragraphs. Please be aware of two caveats: male infertility may be one of several pleiotropic actions of the gene mutation and this report presents only a brief overview of selected data collected in our laboratory and other laboratories. The cited references provide more information.

Gene Mutations That Affect
Spermatogenesis Only (at, bs, olt, qk, Sid, Sxr, WV) Atrichosis (at/at). The atrichosis mutant mouse is an animal model of the Sertoli-cell-only syndrome (56). At present, the etiology of the absence of germ cells is unknown. However, our studies (57)  Blind-Sterile (bs/bs). Blind-sterile mouse testes contain seminiferous tubules with either Sertoli cells only or varying degrees of spermnatogenesis (58). Of interest is the deleterious effect of the gene mutation on the acrosomal development of spermatids (58). Testicular steroidogenesis is not affected (unpublished observations). Blind-sterile female mice are fertile (11).
Steel-Dickie (S1/Sld. Steel-Dickie mice have several deficiencies, including the total absence oftesticular and ovarian germ cells (7). Our studies (Fig. 4)  Sex Reversed (XIX Sxr). Although originally postulated to be a single-gene mutation, the sex-reversing factor (Sxr) is now known to be a distal portion of the Y chromosome that is translocated to the X chromosome during meiosis (14). Genotypic X/X female mice bearing Sxr develop a male phenotype (63). We provided evidence that the Sertoli-cell-only testes of X/X Sxr mice secrete the same amount of steroids as testes from X/ Y controls (64) (Figs. 5

and 6).
Viable dominant spotting (W/Wv). One ofthe pleiotropic effects of the WIW' genotype is the failure of primitive germ cells to proliferate and migrate to the embryonic gonads of both sexes (15). Few or no spermatogonia can be identified in the testes of WIW' mice. The small size of the testes prevented their in vitro perfusion; however, seminal vesicle weights (Fig. 7) support the finding of Amador and co-workers (65) that testicular steroidogenesis is not dramatically affected by the gene mutation.
Flipper-Arm (fl/fl). Flipper-arm mice have not been studied extensively because of their high neonatal mortality. Other than initial report of the mutation (7)   flipper-arm mice (38). One male survived to maturity during our breeding program. Steroidogenesis was decreased and mature spermatids were not produced in flipper-arm mouse testes. Hypogonadal (hpg/hpg). Male hypogonadal mice have immature reproductive tracts and small testes, and spermatogenesis rarely advances past the primary spermatocyte stage (68). This mouse has been used in several experiments (69,70)  of the molecular defects in the gene coding for luteinizing hornone-releasing hormone (71) and the reversal of the mutation effects by gene therapy (72). Our experiments confirmed the drastic effect of the hypogonadal gene mutation on male reproduction in the absence of a decrease in body size (Fig. 11). Hypogonadal female mice are infertile (68). Hybrid Sterility (Hst-1S/Hst-lW). One specific com-  Hst-1w8 and Hst-1wf for Mus musculus musculus) results in the production of sterile male progeny: male mice bearing the Hst-1i/Hst-lw genotype are sterile (73,74). Near-total spermatogenic arrest at the pachytene stage causes the sterility. Our studies (75) confirmed the spermatogenic arrest and provided evidence that the Hst-181Hst-1w8 genotype decreases testicular steroidogenesis. Female hybrids are unaffected. Testicular Feminization (Tfm/Y). Male mice bearing the testicular feminization mutation resemble females phenotypically but have an XY karyotype (76). The primary defect is androgen insensitivity due to the absence of effective androgen receptors (77). Testes of TfmlY mice are characterized by spermatogenesis arrested at the spermatocyte stage or earlier (76) and defective steroidogenesis (78). Our studies of TfmlY mice provided data that were in agreement with the earlier reports (Fig. 12). The testes could not be perfused in vitro but the significant decline in seminal vesicle and kidney weight reflects their inability to respond to androgen stimulation.
Gene Mutations That Do Not Affect Either Spermatogenesis or Steroidogenesis (Ht, Hx, hyt, lit, pg, stb) Hightail (Ht/+). Although the initial report of hightail mice (7) did not mention their fertility, investigators at Oak Ridge National Laboratory stated that hightail males are infertile while hightail females are unaffected (E. M. Kelly, personal communication). Our preliminary studies (38) indicated that hightail mice have normal testicular function. The cause of the reported male infertility is not known.
Hemimelic Extra Toes (Hlx +). The hemimelic extra toes gene mutation causes male-specific infertility (79). Our laboratory elucidated the effects of the mutation on testicular function (36) (Figs. 13 and 14). We concluded that testicular function is normal in the mutant mouse. One possible etiology of the male infertility is the incapacity to copulate effectively because of physical impairment (7).
Hypothyroid (hyt/hyt). Hypothyroid male and female mice have been reported to be infertile (12). The gene mutation causes primary hypothyroidism. Thyroxine-supplemented feed reverses the deleterious effects of the gene mutation (12). In previous studies (38), we demonstrated that testicular steroidogenesis and spermatogenesis of hypothyroid mice were normal and that the defective aspect of the male reproductive process had not been defined. Our most recent studies of the fertility and sexual behavior of hypothyroid male mice suggest that they are fertile, although puberty may be delayed (unpublished observations).
Little (lit/lit). Little male mice have been reported to exhibit a marked incidence of infertility, although little female mice are fertile (13). The primary defect of little mice is isolated growth hormone deficiency partly due to the altered function of somatotroph receptors for growth hormone-releasing factor (80). Our investigations revealed that testicular steroidogenesis and spermatogenesis are normal in little mice (38) (Fig. 15). In addition, videomicroscopic analyses of sperm motility and sperm length ( Table 2)   and natural breeding trials provided evidence that little mice were more accurately described as subfertile in contrast to infertile (unpublished observations). Together, the experimental data indicated that the primary etiological factor of the subfertility of little male mice is their diminutive size. Pygmy (pg/pg). The miniature size of pygmy mice results from tissue unresponsiveness to growth hormone (81). Although King (82) originally reported that both sexes of pygmy mice were infertile, it is now known that the pygmy gene mutation on genetic backgrounds favoring large body size may not induce infertility (7). The latter observation agrees with our experimental data demonstrating that spermatogenesis and testicular steroidogenesis were not significantly affected by the pygmy gene mutation (38). Consequently, the defect in male reproduction of pygmy mice may be due either to a physical barrier to copulation related to their small body size or to an insufficient quantity of spermn in the ejaculate due to their small testes.
Stubby (stb/stb). Stubby male mice were originally reported as infertile by Lane and Dickie (83). Female mice are unaffected by the gene mutation (83). We examined the testicular function of stubby mice and found that testicular steroidogenesis and spermatogenesis were normal (38) (Fig. 16) 43 53 aValues were derived by determining the mean of the included tests of each mouse and calculating the x ± SE of the means for all 12 mice of each genotype. b Significant difference (p <0.05, Wilcoxon Rank Sum Test). c Tests were not included in data analyses if more than one female was used or if the receptivity score for the female was < 3 since the effects of these situations on male sexual behavior could not be considered comparable to those of acceptable tests. onstrated that stubby mice were infertile during natural breeding trials, and furthermore, that they did not intromit or ejaculate during sex behavior tests (55) (Table  3). Next, we elucidated spern motility by quantitative videomicroscopy and tested sperm function by artificial insemination. Both of these tests supported the normal function of stubby mouse sperm (unpublished observations). Together, the experimental observations provided unequivocal evidence that stubby mice are infertile because they are impotent. The etiology of the impotence may involve the central nervous system.

Conclusions and Applications
Our studies have provided the first comprehensive assessment of the effects that the described gene mutations have on male reproduction in mice. The phenotypic effects were elucidated by the application of several biological markers of male reproduction. We propose that these same biological markers would be effective in defining the action of environmental toxins on male reproduction. Figures 17A and B contain our suggestions concerning the sequence in which the biological markers could be implemented.
Two hypothetical situations will be discussed. In the first situation, an environmental chemical has been determined to cause male infertility. The flow diagrams in Figures 17A and B could aid in pinpointing the deleterious action of the toxin. In the second scenario, the goal is to assess the reproductive toxicology of a chemical. Details about the dosage, time schedules, and protocols will not be discussed (84)(85)(86)(87). Initially, the experimental animals would be paired with behaviorallyprimed females. Since the end point is the presence of a copulatory plug at a defined time, one person could monitor a large number of animals for reproductive defects. Also, the same male mice could be tested repetitively. The male mice could be examined for sperm output by quantifying the number of ejaculated sperm; again, this involves simple analysis and repetitive observations of the same mice. After the initial noninvasive markers were used, the more time-consuming, invasive biological markers could be assessed to identify specific causes of male infertility.
Inbred mice offer important advantages in the animal bioassays of reproductive risks. The advantages include genetic uniformity, availability of different inbred strains, and the control of conditions affecting male reproduction (age, social experience, environment, disease, genotype). Additionally, mice have short generation times and are reproductively active under diverse conditions (88).
We have described biological markers that primarily reflect physiological functions. The markers could be used as screening tests for the selection ofmore complex analyses. These biochemical and molecular analyses could focus on in vitro fertilization (89), postimplantation development (90), testicular autoimmunity (91)  antigens (94), germ cell cytogenetics (85), or Leydig cell ultrastructure (95).

Future Directions
The mutant mice are novel animal models that may assist in the development of accurate diagnostic param-eters and effective therapies for male infertility. Furthermore, the mouse models represent a spectrum of reproductive defects and, as such, provide experimental tools for discovering new biological markers of male reproduction that could be evaluated noninvasively and be more sensitive than current markers. Future analysis of the gene mutations will be directed to deciphering the biochemical events that are nearer to the primary gene action.