Mol. Hum. Reprod. Advance Access originally published online on January 18, 2006
Molecular Human Reproduction 2005 11(12):871-880; doi:10.1093/molehr/gah251
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
A repository of ENU mutant mouse lines and their potential for male fertility research
1Centre for Molecular Reproduction and Development, Monash Institute of Medical Research, 2ARC Centre of Excellence in Biotechnology and Development, Monash University, Melbourne and 3The Australian Phenomics Facility, John Curtin School of Medical Research, Australian National University, Canberra, Australia
4 To whom correspondence should be addressed at: The ARC Centre of Excellence in Biotechnology and Development, The Centre for Molecular Reproduction and Development, Monash Institute of Medical Research, Monash University, 27-31 Wright Street, Clayton 3168, Australia. E-mail: moira.obryan{at}med.monash.edu.au
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Many of the proteins and their encoding genes involved in spermatogenesis are unknown, making the specific diagnosis and treatment of infertility in males difficult and highlighting the importance of identifying new genes that are involved in spermatogenesis. Through genome-wide chemical mutagenesis using N-ethyl-N-nitrosourea (ENU) and a three-generation breeding scheme to isolate recessive mutations, we have identified mouse lines with a range of abnormalities relevant to human male fertility. Abnormal phenotypes included hypospermatogenesis, Sertoli cell-only (SCO) seminiferous tubules, germ-cell arrest and abnormal spermiogenesis and were accompanied, in some, with abnormal serum levels of reproductive hormones. In total, from 65 mouse lines, 14 showed a reproductive phenotype consistent with a recessive mutation. This study shows that it is feasible to use ENU mutagenesis as an effective and rapid means of generating mouse models relevant to furthering our understanding of human male infertility. Spermatozoa and genomic DNA from all mouse lines, including those with abnormal reproductive tract parameters, have been cryopreserved for the regeneration of lines as required. This repository will form a valuable resource for the identification and analysis of key regulators of multiple aspects of male fertility.
Key words: FSH/infertility/inhibin/sperm/testis
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
In developed countries, such as Australia, approximately 1 in 25 men are infertile, and of these men in
30% the underlying cause of their spermatogenic failure remains unknown; however, it is expected that many will be genetic in origin (McLachlan et al., 1998
). Further, as many of the proteins, their encoding genes and regulatory pathways involved in spermatogenesis are also unknown, it is frequently difficult to make informed predictions of causal genes. As a consequence, there is a serious absence of specific diagnostic tests and a lack of treatment options for infertile men resulting in many of these men using ICSI, a technique which often provides an effective means of achieving pregnancies with small numbers of often compromised sperm. ICSI is not, however, a cure for infertility and has transmitted genetic defects that can cause infertility to subsequent generations (Mulhall et al., 1997; Cram et al., 2000). These points vividly highlight the need for the identification and characterization of novel genes/proteins that are involved in the production of functional spermatozoa.
Sperm production, the process of spermatogenesis, takes place in the seminiferous tubules and involves the migration of the primordial germ cells, a proliferative mitotic phase (spermatogonia), a meiotic phase (spermatocytes) and the differentiation of haploid cells (spermatids), the latter phase termed spermiogenesis (Russell et al., 1990; de Kretser and Kerr, 1994). The complex features of spermatozoa serve the specific purpose of delivering the male derived genetic material to the egg (Eddy, 1994). The hormonal requirements of spermatogenesis and the cytological changes during this process are well described (de Kretser and Kerr, 1994); however, many of the genes encoding specific components of this process and the underlying biochemical pathways remain unknown. Because there are no available in vitro techniques to successfully achieve fully functional spermatogenesis, investigation of the function of genes to achieve sperm production requires the development of expensive animal models.
N-ethyl-N-nitrosourea (ENU) is a potent mutagen and carcinogen in mammals (Justice et al., 2000). In recent years, laboratories have used ENU to identify novel genes that are involved in human disease. This approach provides a non-biased, phenotype-driven approach, as opposed to a genotype-driven approach using knockouts and targeted mutagenesis. ENU is a synthetic compound that causes random, single base pair mutations, acting directly on nucleic acids without any metabolic processing required for its activation (Singer and Dosanjh, 1990; Justice et al., 2000; Noveroske et al., 2000). Though it predominantly induces point mutations, a few small deletions or insertions have been reported. The most common mutations are AT to TA transversions and AT to GC transitions (>82% of sequenced mutations) (Popp et al., 1983; Noveroske et al., 2000). ENU is considered to be the most effective chemical mutagen on mice and most efficiently affects the spermatogonial stem cells in the testis (Russell et al., 1979). The very high mutation rate of 0.0015 per locus per gamete, using the regime described herein, means that examining fewer than 1000 pedigrees should on average identify a mutation at any given locus (Russell and Montgomery, 1982; Hitotsumachi et al., 1985).
Any individual ENU-treated mouse will have many different mutations that are simultaneously induced, i.e. silent, loss-of-function, gain-of-function, niche-filling (whereby the normal function of a protein is inactivated but does not prevent the assembly of a protein into higher order complexes) (Papathanasiou et al., 2003). Most functional gene defects that are induced by ENU will be loss-of-function variants. These behave as recessive traits and can only be identified after breeding to homozygosity or using mice carrying large deletions or chromosomal inversions (Justice et al., 1997; Kile et al., 2003). Using the strategy described herein, the G1 progeny (produced by mating a wild-type female to the ENU-treated male) are expected to carry 30 loss-of-function mutations and will pass on 15 mutant alleles to the heterozygous G2 generation (Goodnow, personal communication) (Figure 1). These mutations are bred to homozygosity by inter- or back-crossing. With the assumption that each G1 male carries 30 loss-of-function mutations, the resulting G3 offspring will be homozygous for 3–4 different loss-of-function mutations. One quarter of G3 mice derived from any particular G1 (pedigree) will be homozygous for the same mutation (Figure 1). Thus by examining many G3 mice from the same pedigree and determining the frequency of a particular phenotype, it is possible to distinguish mice with a monoallelic phenotype (in this case infertility) from those with multigenic phenotypes, assuming multiple mutations are not linked.
|
This study demonstrates the potential of ENU mutagenesis and a controlled three-generation breeding strategy to produce relevant models of human male infertility for the identification of key regulators of spermatogenesis. This repository will be continuously expanded and is open for use by the entire reproductive biology community.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The generation of ENU mutagenized pedigrees
All animal experiments were carried out in accordance with the National Health and Medical Research Council (NHMRC) Guidelines on Ethics in animal experimentation and approved by the John Curtin Medical School animal experimentation ethics committee and Monash Medical Centre animal ethics committee.
About 8- to 15-week-old male C57Bl/6J mice (G0) were treated three or four times, 1 week apart, with 85–100 mg/kg ENU (Sigma, St Louis, MO, USA) freshly dissolved in 10% ethanol in citrate buffer, pH 5.0 (Hitotsumachi et al., 1985). After 8 weeks of treatment, at a time when spermatogonia would have had a chance to sufficiently repopulate the testis (Nolan et al., 2000), G0 mice were breed with wild-type C57BL/6 females. A maximum of 20 G1 offspring from any ENU-treated male were utilized to minimize the chances of the same mutation appearing in any G1-derived pedigree (Justice, 2000). Subsequently, one of two breeding strategies was implemented to allow for the inclusion or exclusion of X-linked mutations.
Strategy 1 (biases against both X- and Y-linked mutations)
G1 (pedigree founders) males were mated with wild-type C57Bl/6 females. G2 females were mated with either their G1 father or their G2 brother (Figure 1). G3 males were culled, and samples analysed for male fertility-related parameters. This strategy would strongly bias against both X- and Y-linked mutations, critically affecting fertility at the G1 stage, as such males would be sterile or sub-fertile (reduced pup number). It is estimated that each G1 founder male carried 30 loss-of-function mutations, and as such it is estimated that the G3 offspring should be homozygous for a range of 3–4 different loss-of-function mutations.
Strategy 2 (includes X-linked but excludes Y-linked mutations)
G1 males were breed with unrelated G1 females (Figure 1). Subsequently, G2 females were back-crossed to their father or bred with siblings to produce G3 pups. Samples were harvested from G3 males for analysis of fertility-related parameters. G3 mice would be homozygous for an average of 3–4 different mutations and could include X-linked mutations derived from their G1 grandmother.
With the exception of X-linked mutations and mutations in male-specific genes produced in strategy 2, which will affect 50% of G3 males within a pedigree, all other fully penetrant single gene defects will affect 25% of G3 males.
Sample collection
Following CO2 asphyxiation, mouse body length and weight were recorded, and a 0.5–1 ml blood sample was collected by cardiac puncture and serum collected following clotting and centrifugation. A small blood sample was stored on Guthrie cards (Whatmann, UK) to allow for gDNA screening at a later date.
Analysis of testicular histology
Testes were collected and fixed overnight in Bouin’s fluid, weighed and processed through standard paraffin techniques (O’Bryan et al., 2000). Sections of 5 µM were stained using periodic acid Schiff (PAS). Testes were examined for histological abnormalities using the criteria defined in Russell et al. (1990). In particular, the diameter of the tubules and the presence or absence of peritubular fibrosis was noted, and the seminiferous epithelium was evaluated to identify whether all stages of germ-cell development were present. A qualitative assessment of the number of germ cells occupying the seminiferous epithelium was made, and if spermatogenesis ceased at a particular phase of germ-cell development, this was noted and defined according to the cytological detail. Several definitions were used (i) hypospermatogenesis was defined when all stages of spermatogenesis were present, but the numbers of germ cells were decreased in at least 50% of all tubules examined; (ii) germ-cell arrest was used when all tubules in the testis showed spermatogenic arrest at a particular stages, e.g. pachytene spermatocyte stage; (iii) Sertoli cell-only (SCO) phenotype was a term used when the only cell type in all the tubules was the SC, and no germ cells were present; and (iv) spermiation failure was a term used when elongated spermatids that should have been released into the lumen were not released and instead were drawn back towards the basement membrane and phagocytosed by the SCs.
Detection of apoptotic cells
For selected testis samples, cells undergoing apoptosis within the testis were immunostained, using the terminal deoxynucleotidyl transferase-mediated dUDP nick-end labelling (TUNEL) method, as recommended by the manufacturer (ApopTag kit, Intergen, NY, USA).
Sperm cryopreservation
Sperm were collected from every second mouse within a line for cryopreservation before which an assessment of sperm motility, morphology and concentration was made and recorded. It should be noted that any G3 mouse has a 75% chance of carrying a mutation causing any phenotype associated with that particular line (Figure 1), as such by cryopreserving sperm from every second mouse ensures that sperm carrying the mutation will be stored. The cauda epididymides were isolated by dissection, and a small slit was made in the epididymal duct and placed into a 1.5 ml microcentrifuge tube containing either 130 µl or 1.0 ml of a 0.3 mM raffinose and 3% skim milk solution for sperm cryopreserved as straws or pellets, respectively. The cauda and sperm were incubated for 10 min at 37°C to allow sperm to exit the tissue into the media. Following incubation, the cauda was removed, and the presence of sperm, their approximate count and total motility and progressive motility were determined by an experienced researcher. Initially, sperm were frozen following a modified Nakagata method (Nakagata, 2000; Sanchez-Partida, 2003); 10 µl samples from 130 µl sperm suspension were stored in cryostraws (IMV technologies, France), sealed with polyvinyl alcohol (Sigma) and cooled to –150°C, equilibrated for 10 min at –150 to –160°C, then transferred to liquid nitrogen. Alternatively, 100 µl aliquots from 1 ml of sperm suspension were placed in solid dry-ice wells (–79°C) for 60–90 s and transferred and stored in liquid nitrogen.
Serum hormone analyses
FSH radioimmunoassay
The concentration of FSH in mouse serum was determined using radioimmunoassay reagents kindly provided by Dr A. Parlow (NIDDK, Bethesda, MD, USA). The iodination preparations and antisera used were rFSH I-8 and anti-rFSH-S-11, respectively. Results are expressed by NIDDK mFSH-RP-1. The tracer was iodinated using iodogen reagent (Sigma). Goat anti-rabbit IgG (GAR#12; Monash Institute of Medical Research, Monash University, Melbourne, Australia) was used as the secondary antibody. The assay buffer used was 0.01 M phosphate-buffered saline (PBS) containing 0.5% bovine serum albumin (BSA) (Sigma).
A mouse serum pool diluted in a dose-dependent manner was included in parallel to the standard curve in each assay (data not shown). For the FSH radioimmunoassay, the lowest limit of detection was 1.04 ng/ml, the average within-assay percentage coefficient of variation (%CV) was 5.8%, and the inter-assay %CV was 7.7% (n = 9 assays) calculated using a pool of normal mouse serum used in each assay. The normal FSH range was determined using 28 normal wild-type C57Bl/6J male mice (ages 6–8 weeks).
Activin A enzyme-linked immunosorbent assay
Activin A was measured using a specific enzyme-linked immunosorbent assay (ELISA) (Knight et al., 1996), according to the manufacturer’s instructions (Oxford Bio-Innovations, Oxfordshire, UK) with some minor modifications. The standard used was human recombinant (hr) activin A, as described previously (Robertson et al., 1992), and is a valid standard as the amino acid sequence of human and mouse activin A is identical (Mason et al., 1985). Mouse serum diluted in a dose-dependent manner was included in parallel to the standard curve (data not shown). The average intra-plate %CV was 8.0%, the inter-plate %CV was 9.2% (n = 26 plates) based on a pool of normal mouse serum used in each plate, and the limit of detection for the assay was 0.01 ng/ml. The normal activin A range was determined for 28 wild-type C57Bl/6J male mice (ages 6–8 weeks).
Inhibin radioimmunoassay
Immunoreactive inhibin was measured by heterologous radioimmunoassay, as described previously (Robertson et al., 1988). Results are expressed by an in-house rat ovarian extract calibrated against hr-inhibin. Iodinated hr-inhibin was used as tracer. The antiserum used was rabbit antiserum (1989), which is directed towards the 
-subunit of inhibin and bound to both inhibin A and inhibin B, and cross-reacts 288% with pro-C, the pro-sequence of the inhibin -subunit (Robertson et al., 1989). GAR#12 was used as second antibody. The assay buffer used was 0.01 M PBS/0.5% BSA. A normal mouse serum pool diluted in a dose-dependent manner and was parallel to the standard curve in each assay (data not shown). The lowest limit of detection was 0.12 ng/ml. The average within-assay %CV was 9.4%. The inter-assay %CV was 6.4% (n = 9 assays) calculated using a pool of normal mouse serum used in each assay. The normal range for immunoreactive inhibin in serum was determined for 28 wild-type C57Bl/6J male mice (ages 6–8 weeks).
Testis weights
Testis weights were determined after fixation in Bouin’s solution, before being processed into paraffin. The normal range was determined from 20 wild-type C57Bl/6J male mice aged between 6 and 8 weeks.
Statistical analysis
The data from each mouse within a line was dot plotted. The mean for each parameter measured was determined for wild-type mice and for each line, recognizing that there were likely to be about 3–4 mutations within each line. The normal range for each parameter was defined as 2 SDs around the mean for wild-type mice. Examination of the serum FSH and inhibin data from wild-type mice suggested that there were significant elevations of both FSH and inhibin in all lines raising the possibility that this may be because of the age difference of the wild-type mice used to determine the normal range (6–8 weeks) compared with the G3 mice that were killed at 12–18 months, i.e. the FSH and inhibin rise may represent an age-related phenomenon, as seen in men (Baker and Hudson, 1983), rather than being related to the testicular phenotype. In view of this finding, we also calculated a ‘normal’ range (mean ± 2 SD) for all parameters from all G3 mice used in the study recognizing that the use of this wider range may blunt the sensitivity of the use of this parameter.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Mouse ENU mutagenesis is a highly effective means of developing mouse models of male infertility
At the outset of this project, ENU-mutagenized mice were assessed by the evaluation of testicular histopathology, but this was later expanded to include the measurement of key reproductive hormones and a simplified assessment of sperm parameters, including motility, morphology and number.
In total, 1583 male G3 mice from 143 different ENU-mutagenized pedigrees were generated using the strategy outlined in Figure 1 and described in the Materials and methods and examined for male fertility-related phenotypes. Our initial screening using testis histology evaluated 173 mice from 26 lines. Following the expansion of our screening criteria, a more comprehensive fertility-related analysis of 1410 mice from 117 ENU-mutagenized lines was completed including measurements of testis weight, sperm motility, crude sperm count and morphology, as well as the serum concentrations of the key fertility hormones, FSH, activin A and inhibin.
The data for the parameters assessed are given in Figures 2–4, Table I and supplementary Figures 1–10. In evaluating the data, it must be remembered that all of the mice in these groups do not represent a consistent phenotype associated with a single mutation but represent data from a pool of mice from each line in which it is reasonable to expect 3–4 homozygous point mutations per G3 mouse.
|
|
|
|
Abnormal fertility-related phenotypes were seen in varying frequencies. The G3 males that were screened result from few different randomly paired G2 animals, not all of which were carriers, thus several litters would consist of all wild type or heterozygous animals. Therefore, phenotypes that were observed several times, but not at one in four frequency, were considered to be potentially caused by a single mutation within one gene that was segregating in a Mendelian fashion; from this point forward, these will be referred to as recessive. Such a hypothesis would be confirmed upon further breeding should a line be chosen for study. Phenotypes that were seen at a very low rate (e.g. 1 of 19 mice) were considered to be caused by mutations within two or more genes (compound-interacting mutations) and were excluded. Phenotypes occurring at a rate higher than would be expected from a recessive mutation could be caused by variable penetrance of dominant mutations or multiple mutations affecting similar aspects of fertility.
With the parameters used in this screen, the phenotypes detected included seminiferous epithelium abnormalities, abnormal epididymal sperm counts and motility and altered hormonal levels. These data are considered separately, and then examples are provided showing the overall interrelationships that were sometimes seen. The data from this screen are presented in Table I and Figures 2–
4, as well as in the supplementary information.
Of the 117 lines screened with our more comprehensive screening procedure, there were 65 lines with 10 or more mice. Data for mouse lines not in Figure 4 can be found in the supplementary data accompanying this article (supplementary Figures 1–10). Abnormal reproductive parameters consistent with abnormal or compromised reproductive function were detected in a frequency consistent with a recessive mutation in 14 of 65 lines (Table I).
Seminiferous epithelium phenotypes as detected by microscopic evaluation
A wide range of seminiferous epithelium phenotypes were observed that would likely result in varying degrees of infertility ranging from sterility to significantly reduced pup numbers. These included hypospermatogenesis (4.5% of all mouse lines), germ-cell arrest (1.5% of all mouse lines), SCO (1.5% of all mouse lines) and spermiation failure (1.5% of all mouse lines) (see Material and methods section for definitions) and are shown in Figure 2. The most common phenotype observed was hypospermatogenesis typified by the presence of all germ cells within the seminiferous epithelium but with variably reduced numbers ranging from moderate to severely depleted populations in different tubules (Figure 2C). In this study, the classification of hypospermatogenesis was used when 50% or more tubule cross-sections were abnormal. This phenotype was seen in recessive frequencies in two independent lines, 4.60 and 4.69. Line 4.48 showed this phenotype in approximately half of the mice, and it is more likely that this phenotype was caused by a dominant mutation.
One line (4.1) was observed with an arrest of germ cells in meiosis with cessation at the primary spermatocyte stage (Figure 2G). The testes of these mice fulfilled the key criterion of germ-cell arrest in that no germ cells in the entire section examined progressed beyond the primary spermatocyte stage. As would be expected, mice showing an arrest in meiosis also displayed dramatically reduced testis weights (<1/4 of wild type) (Figure 4A). TUNEL staining revealed that germ-cell death was mediated by apoptosis (Figure 2H). This line showed a phenotypic frequency consistent with a recessive mutation.
Four lines with phenotypes affecting the later stages of spermatogenesis were seen with the partial or complete absence of sperm tails (4.5% of all mouse lines) and spermiation failure (1.5% of all mouse lines). Spermiation failure was classified when elongated spermatids are not released into the lumen, rather they are drawn back towards the basement membrane and phagocytosed by the SCs. Lines 4.29 and 4.91 (Figure 2D) showed a complete absence of sperm tails. Line 5.28 showed a reduction in both sperm tails and spermiation failure. These phenotypes were seen in recessive frequencies. Line 5.134 showed partial spermiation failure in nearly half of the mice screened, with some sperm being released into the lumen, but many failed to do so.
Testis weights
The testis weights for each line were compared with the testis weights for wild-type mice, where the mean and 2 SDs above and below the mean represented a normal range indicated on Figures 3 and 4, between 0.0689 and 0.1001 g. For completeness, the testis weight range for all G3 mice screened in this study (between 0.0458 and 0.111 g) has also been included in Figures 3 and 4. Examination of the dot plots of testis weights shown in Figure 3 demonstrates that several lines were distributed over a range that far exceeded the mean and 2 SDs of the wild-type mice, e.g. lines 4.1, 5.25 and 4.94. Such a pattern suggests that there may be several phenotypes within these lines that influence testis weights. For instance, line 4.1 shows three clusters, a group of six with severely reduced testis weights, a group of eight just below the lower end of the normal range and a group clustered around the mean of the normal range (Figure 4A). Such a proposal could be verified by comparing the histological features of the testes of these mice. In the case of line 4.1, the group of six severely reduced testis weights showed an arrest in meiosis. The group clustered around the mean of the normal range showed normal testis histology. The group of eight near the lower end of the normal range showed a range of histological profiles. Two of these mice exhibited a decrease in the number of elongating spermatids, and one mouse exhibited an increase in SCO tubules. It is possible that the 4.1 line contained two (or more) mutations affecting germ-cell development and testis weight. Alternatively, the intermediate testis weight cluster of mice may represent heterozygous mice in which spermatogenesis may be affected but not to the extent of an arrest in meiosis. This line provides us with an example of how one or more of the parameters measured can help to identify several groups within the one mouse line, highlighting which particular lines to analyse further.
A severe disruption in spermatogenesis can be identified in a simple screen of testis weights in which 25% of mice within a line display testis weights below the lower limit. An analysis of testis weights reliably detected an abnormal group of mice in line 4.1 which had testis weights that were 25% of wild-type mice as the result of an arrest in meiosis (Figures 3A and 4A). Further, the same approach for line 5.25 identified two of 13 mice that had testis weights less than the lower limit of wild-type mice because of a SCO phenotype (Figures 3A and 4B). In contrast, the simple measurement of testis weight could not detect all phenotypes with abnormal spermatogenesis. For example, testis weight alone was not sensitive enough to discern defects in spermiogenesis that were present in lines 4.29 and 4.91, i.e. loss of sperm tail structures (Table I and Figure 2D).
Sperm concentration and motility
The crude assessment of epididymal sperm counts can provide an indication of sperm output. It should, however, be remembered that spermatozoa retrieval from the epididymis is influenced by some factors, including testicular output, epididymal function and operator competence. Ultimately, sperm count and/or germ-cell type would be determined using a quantitative measurement of daily sperm count or stereological methods. Epididymal sperm counts are most valuable in assessing the severity of the phenotype classified as hypospermatogenesis. For instance, two lines (4.39 and 4.78) showed reduced epididymal sperm counts when compared with other mouse lines. Both lines showed qualitatively normal spermatogenesis by testicular histology. Line 4.48 had reduced sperm motility but relatively normal sperm numbers as judged by crude spermatozoa retrieval from the epididymis.
Hormone analysis (FSH, inhibin and activin A)
Serum hormone levels for each line were compared with the hormone levels for wild-type mice (Figure 3), where the mean and 2 SDs above and below the mean represented a normal range for each hormone. In normal wild-type mice, these ranges were 0.031–0.171 ng/ml for activin A, 7.98–10.68 ng/ml for FSH and 0.48–1.59 ng/ml for immunoreactive inhibin. Interestingly, the range for serum FSH levels derived from the wild-type mice was significantly lower than the range calculated using the levels of FSH from all the mice screened in this study (***P 
0.0001). This may result from the fact that there are many examples of mice with damage to the seminiferous epithelium amongst all mice in this screen, thereby raising the mean FSH and elevating the range. However, another possibility is the fact that the wild-type mice were of 6–8 weeks of age, whereas the mice from the ENU screen were killed at 18–24 months of age. This suggests that serum FSH levels increase with age in the mouse as has also been reported in men and horses (Baker and Hudson, 1983; Johnson and Thompson, 1983). Additionally, the normal range for immunoreactive inhibin in wild-type mice was also significantly lower than the range for all the mice screened in this study (**P = 0.0025). Again, as with FSH, this could reflect decreased SC function associated with spermatogenic failure (Rich et al., 1979) or may be related to an ageing phenomenon. No age significant difference was found for the activin A concentrations. For completeness, the hormone ranges for all mice in the study for FSH (between 8.08 and 16.46 ng/ml), inhibin B (between 0.497 and 1.829 ng/ml) and activin A (between 0 and 0.225) are provided in Figures 3 and 4, in addition to the normal ranges for wild-type mice. Additionally, the testis weight range for wild-type mice and that derived from the concentration of all mice in the study are also included in Figures 3 and 4.
As is shown in Figure 3, several lines showed clusters of mice outside the normal hormone ranges. Clinically, human male infertility is assessed by andrologists using a combination of histology and endocrine parameters. As such, several lines were chosen for an in-depth analysis to assess their relevance to human male infertility (Figure 4 and supplementary Figures 1–10).
The measurements of reproductive hormones aid in the phenotypic analysis in at least two ways, and this view is best demonstrated by reviewing the FSH levels in several lines. First, low concentrations of FSH result in a failure of testicular development, which in its most severe form would emerge as a phenotype, wherein the seminiferous tubules remained as cords with germ cells failing to develop beyond the gonocyte or spermatogonial stages. This occurrence would be further supported by SCs that failed to develop nucleoli, a feature of FSH action on the testis. None of the testes observed in this study fit this pattern. However, a less severe form of impaired spermatogenesis would emerge with low FSH levels accompanied by hypospermatogenesis as seen in line 4.71, wherein two of 13 mice showed borderline low FSH when compared with wild type or all mice within this study. These mice exhibited a mild hypospermatogenesis phenotype.
Inhibin is primarily defined by its ability to exert an inhibitory effect on FSH secretion (McLachlan et al., 1987). Thus, the measurement of both FSH and inhibin allows a comprehensive picture of the effect of one upon the other. A large increase in inhibin could lead to a decrease in FSH secretion, the effects of which are discussed above. Alternatively, a decrease in inhibin could result in a rise in FSH. This was not observed in our set of mice and was suggestive of a more complicated interaction between pituitary-derived FSH and testis-derived inhibin.
Serum activin A levels were measured because of the role of this protein in stimulating FSH (de Kretser et al., 2000) and its secretion by peritubular and SCs (Buzzard et al., 2003). Several lines, e.g. 4.24, 4.29 and 4.36, showed high mean activin A levels (4.5% of all lines) (Figure 3B). Although these could be linked to the fertility status of the mice, there is increasing evidence that this protein plays a significant role in inflammation (Jones et al., 2004). Because these mice were also screened for immunological phenotypes, we searched the database for information on these lines and noted that all three with high activin A levels also had immunological disorders, such as type 1 diabetes and systemic and autoimmune traits. Collectively therefore, although significant changes in serum activin A levels may be reflective of disturbed spermatogenesis, this is unlikely to be clear-cut because of the role of activin A in immune function and the complex interaction between the immune and reproductive systems (Hedger, 2002).
Examples of the utilization of multiple parameters in the male fertility assessment of the mouse lines
The value of using multiple parameters in screening these ENU treated mice is illustrated in a few examples.
- Line 4.68: Eight of 20 mice exhibited extremely raised FSH levels, 9 of 20 mice had low testis weights, 3 of 20 had low inhibin levels (though still in the normal range) and the majority of mice showed serum activin A levels below the mean for wild type (Figures 3 and 4C). The measurement of immunoreactive inhibin in these mice provided a crude index of SC function (Rich and de Kretser, 1977; Rich et al., 1979). There was a positive correlation between immunoreactive serum inhibin and testis weight. A negative correlation was shown between FSH levels and testis weight with high FSH levels correlating with low testis weights. Interestingly, this line showed a positive correlation between FSH and activin A. The strongest correlation was seen between testis weight and activin A, with the highest testis weights associated with the lowest activin A levels. This finding is quite unexpected as an increase in activin A would be expected to stimulate FSH, leading to an increase in FSH and most likely an increase in testis weight.
Line 4.68 also showed a cohort of animals with an infiltration of strongly PAS positive cells into the testicular interstitium (Figure 2F). Based on morphological criteria, these cells were likely to be multinucleated macrophages. Such testes had relatively normal seminiferous tubule histology, and all hormone parameters were mid-range. This is an example of a line, whereby the many abnormalities observed may be the result of more than one ENU-induced mutation.
- In some ENU mouse lines, such as 4.94 and 4.68, up to half of the mice had testis weights below the normal range (Figure 4D). For line 4.94, the low testis weights correlated with the presence of testicular damage of a heterogeneous pattern showing tubules with hypospermatogenesis and other tubules with SCO epithelium. Interestingly, both lines showed clusters of animals with serum activin A levels below the wild-type mean. The latter mice had normal testis histology.
- Line 4.29: Five of 36 mice from this line had elevated serum activin A levels (Figure 3B). All five had normal testis histology. As indicated earlier, this line had an immunological phenotype with a lupus-like disease and kidney pathology. However, a further five mice from this line had a testicular phenotype where the spermatids lacked tails (Figure 2D), with three of these exhibiting testis weights that were below the lower end of the normal range. All hormonal levels in these five mice were distributed around the middle of the normal range suggesting that the phenotype was because of an autonomous germ-cell defect rather than being related to hormone levels. It would seem likely that this line contained two independent mutations with one affecting male fertility parameters and one affecting the immunological status (and serum activin A levels).
- In line 5.137, over half the mice had serum inhibin levels above the normal range, with the mice exhibiting the high inhibin levels showing testis weights above the normal range (Figure 4F). Such a correlation suggests that the increased testicular weight may have resulted from an increased SC number, which in turn lead to a greater spermatogenic capacity and an increased inhibin output.
- Line 4.48 showed a cluster of mice with significantly decreased testis weights (Figure 4E) associated with the presence of SCO and germ-cell arrest tubules (hypospermatogenesis). These mice also frequently showed serum inhibin levels in the lower end of the normal range (Figures 3D and 4E). This was expected given the well-characterized relationship between serum inhibin levels and SC number and function (Ramaswamy et al., 1999; Sharpe et al., 1999). Further, as SCs are capable of supporting a finite number of germ cells, the observation of decreased inhibin production in the 4.48 line could be indicative of decreased SC number or alternatively, ‘sick’ SCs and a concomitant reduction in germ-cell number and overall testis weight. This line was observed to have three mice with decreased sperm motility and normal sperm concentration, all of which also displayed the hypospermatogenesis phenotype.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
This study shows that ENU mutagenesis can feasibly and successfully be used to generate mouse models of testicular dysfunction that will enable the identification of the responsible genes. Many of the phenotypes identified are similar to those seen in male infertility clinics, where, in about 40% of cases, the actual cause of the spermatogenic disorder is unknown, making therapeutic intervention difficult. This study also established the value of the parameters employed in evaluating the phenotypes and will assist those in choosing which lines should be studied further. In particular, testis weight and histology as well as serum inhibin and FSH concentrations and the simplified epididymal sperm analysis, allowed for the formulation of a comprehensive picture of testicular pathology within each pedigree, comparable with that obtained by andrologists clinically. Further, the deposition of these data and the cryopreservation of sperm carrying the mutations provide a valuable resource for the entire reproductive biology community. Assisted reproduction technologies (ART) such as IVF and ICSI allow for the regeneration of lines of interest and subsequent linkage analysis and causal gene identification at a later date. For lines in which the abnormal males exhibit no sperm or sperm that are incapable of achieving fertilization by IVF or ICSI, fertile male siblings (which carry a two in three chance of being a carrier of the causal mutation) can be used for line regeneration. Collectively therefore, the development of this repository promises to reveal a wealth of relevant data that will allow selection of lines for further study and to identify the relevant genetic defect thereby expanding our understanding of the control of spermatogenesis.
Traditionally, to map a causal mutation, researchers are required to outbreed the affected line with another strain of mouse. For non–fertility-related genes, this can be done from homozygous-affected animals; however, the limitations of a male fertility phenotype necessitate the ‘blind’ outbreeding of potentially heterozygous G3 litter mates (two of three fertile male litter mates will be heterozygotes) and the pairing of F1s siblings to produced phenotype bearing F2 mice. PCR-based linkage analysis is then performed on affected versus wild-type mice. Given the inability to genotype (and phenotype) heterozygous mice, the mapping of fertility-related phenotypes is significantly more time consuming than many of the previously published ENU-induced phenotypes. However, if sperm from homozygous mice are used in ART procedures, all F1 mice will be carriers. Subsequently, the random pairing of F1 mice will result in 25% of F2 being affected. Given the added complication of male infertility affecting only one in four male G3 pups, the preservation of affected sperm represents a substantial logistical and financial advantage to assist in the identification of the causal gene.
The screen described here has been highly effective in identifying a wide range of male fertility-related phenotypes that are clinically relevant and occurred in a frequency consistent with a recessive mutation. Interestingly, several lines were identified with two different phenotypes occurring at a frequency indicative of recessive mutations (e.g. line 4.29 with increased activin A levels and loss of tail structure). These phenotypes may be the consequence of two independent fertility affecting mutations occurring within the same pedigree and would ultimately be separated during the outbreeding and linkage process. Alternately, dominant mutations in male reproductive specific genes are theoretically possible and would be transmitted at a one in two frequency to male off-spring from heterozygous mothers, e.g., dominant mutations in male-specific genes. Phenotype frequencies significantly varying from one in four or one in two males are indicative of compound-interacting mutations or variable penetrance of dominant mutations (e.g. line 5.137 showing half of the mice with increased inhibin and testis weights).
During the process of analysing hormone levels, it was discovered that the mutagenized mice, aged 18–24 months at the time of screening, had statistically significantly higher FSH and inhibin serum levels than the much younger 6- to 8-week-old wild-type mice. The reasons for these differences may be complex. Because many of these mouse lines likely contain mice with testicular damage relative to genes expressed in germ cells, the elevated FSH levels may partly be explained by the well-recognized increase in FSH in response to seminiferous tubule dysfunction (de Kretser et al., 1974; Rich and de Kretser, 1977). For completeness therefore, all data have been presented with a comparison to wild-type 6- to 8-week-old mice and against all mice within the study.
Given the sensitivity of spermatogenesis in particular to the affects of systemic illness (Spratt et al., 1993), data on reproductive parameters in isolation may be misleading. For example, there is a recognized decline in spermatogenic output following systematic inflammation (Hedger, 2002). This is particularly critical when assessing the reproductive health of G3 males which on average are homozygous for 3–4 mutations. The outbreeding of such mice will, however, identify whether the infertility is related to a systemic illness or is caused by a separate mutation. Given the large personal and budgetary investment in animal models, it is highly desirable to avoid such secondary infertility where possible. An example of this phenomenon was demonstrated in this study with pedigrees showing high activin A levels linked to the presence of immunological disorders.
This screen is the most comprehensive of its type for male fertility (Ward et al., 2003; Clark et al., 2004); however, the inability to reliably and economically measure the key reproductive hormone testosterone in mice remains a serious limitation, i.e., current assays display high sample to sample variation, and consequently the achievement of statistical significance requires untenably high animal numbers for large-scale screening projects where only one in four male mice would be expected to display a phenotype. Initially, we attempted to use seminal vesicle weight as a physiological indicator of androgen levels. This, however, proved to be technically difficult on a large scale and biologically misleading because of large and unexpected variations in seminal vesical weights and morphology (data not shown). The reason for this variation is not known.
Using a wide range of measures to assess the function of spermatogenesis in a large number of mice from a recessive ENU mutagenesis breeding scheme has proved to be a valuable means of identifying a wide range of male fertility-related phenotypes caused by single gene mutations. Our thorough screening process and archiving of sperm will prove to be a valuable resource for the scientists interested in identifying new genetic mechanisms critical for successful spermatogenesis.
|
Acknowledgements |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The authors gratefully acknowledge the excellent technical assistance of Amy Herlihy, Amanda Eddy, Luke Swain, Miranda McEwan, Gabrielle Douglas, Rebecca Craythorn, Ann Davies, Adrienne McKenzie and Shirine Chaudhry. This work was supported by grants from the NHMRC (143786) and the ARC (CEO348239). MKOB is the recipient of a Monash University Fellowship. CK is the recipient of an APA scholarship from the Commonwealth of Australia.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Baker HW and Hudson B (1983) Changes in the pituitary-testicular axis with age. Monogr Endocrinol 25,71–83.[Medline]
Buzzard JJ, Farnworth PG, De Kretser DM, O’Connor AE, Wreford NG and Morrison JR (2003) Proliferative phase sertoli cells display a developmentally regulated response to activin in vitro. Endocrinology 144,474–483.
Clark AT, Firozi K and Justice MJ (2004) Mutations in a novel locus on mouse chromosome 11 resulting in male infertility associated with defects in microtubule assembly and sperm tail function. Biol Reprod 70,1317–1324.
Cram DS, Ma K, Bhasin S, Arias J, Pandjaitan M, Chu B, Audrins MS, Saunders D, Quinn F, de Kretser D et al. (2000) Y chromosome analysis of infertile men and their sons conceived through intracytoplasmic sperm injection: vertical transmission of deletions and rarity of de novo deletions. Fertil Steril 74,909–915.[CrossRef][Web of Science][Medline]
de Kretser DM and Kerr JB (1994) The cytology of the testis. In Knobil E and Neill D (eds), The Physiology of Reproduction, 2nd edn. Raven Press, New York, pp. 00–00.
de Kretser DM, Burger HG and Hudson B (1974) The relationship between germinal cells and serum FSH levels in males with infertility. J Clin Endocrinol Metab 38,787–793.
de Kretser DM, Meinhardt A, Meehan T, Phillips DJ, O’Bryan MK and Loveland KA (2000) The roles of inhibin and related peptides in gonadal function. Mol Cell Endocrinol 161,43–46.[CrossRef][Web of Science][Medline]
Eddy EM (1994) The Spermatozoon. In Knobil E and Neill D (eds), The Physiology of Reproduction. Raven Press, New York, pp. 00–00.
Hedger MP (2002) Macrophages and the immune responsiveness of the testis. J Reprod Immunol 57,19–34.[CrossRef][Web of Science][Medline]
Hitotsumachi S, Carpenter DA and Russell WL (1985) Dose-repetition increases the mutagenic effectiveness of N-ethyl-N-nitrosourea in mouse spermatogonia. Proc Natl Acad Sci USA 82,6619–6621.
Johnson L and Thompson DL Jr (1983) Age-related and seasonal variation in the Sertoli cell population, daily sperm production and serum concentrations of follicle-stimulating hormone, luteinizing hormone and testosterone in stallions. Biol Reprod 29,777–789.[Abstract]
Jones KL, de Kretser DM, Patella S and Phillips DJ (2004) Activin A and follistatin in systemic inflammation. Mol Cell Endocrinol 225,119–125.[CrossRef][Web of Science][Medline]
Justice MJ (2000) Capitalizing on large-scale mouse mutagenesis screens. Nat Rev Genet 1,109–115.[CrossRef][Web of Science][Medline]
Justice MJ, Zheng B, Woychik RP and Bradley A (1997) Using targeted large deletions and high-efficiency N-ethyl-N-nitrosourea mutagenesis for functional analyses of the mammalian genome. Methods 13,423–436.[CrossRef][Web of Science][Medline]
Justice MJ, Carpenter DA, Favor J, Neuhauser-Klaus A, Hrabe de Angelis M, Soewarto D, Moser A, Cordes S, Miller D, Chapman V et al. (2000) Effects of ENU dosage on mouse strains. Mamm Genome 11,484–488.[CrossRef][Web of Science][Medline]
Kile BT, Hentges KE, Clark AT, Nakamura H, Salinger AP, Liu B, Box N, Stockton DW, Johnson RL, Behringer RR et al. (2003) Functional genetic analysis of mouse chromosome 11. Nature 425,81–86.[CrossRef][Medline]
Knight PG, Muttukrishna S and Groome NP (1996) Development and application of a two-site enzyme immunoassay for the determination of ‘total’ activin-A concentrations in serum and follicular fluid. J Endocrinol 148,267–279.
Mason AJ, Hayflick JS, Ling N, Esch F, Ueno N, Ying SY, Guillemin R, Niall H and Seeburg PH (1985) Complementary DNA sequences of ovarian follicular fluid inhibin show precursor structure and homology with transforming growth factor-beta. Nature 318,659–663.[CrossRef][Medline]
McLachlan RI, Robertson DM, de Kretser D and Burger HG (1987) Inhibin – a non-steroidal regulator of pituitary follicle stimulating hormone. Baillieres Clin Endocrinol Metab 1,89–112.[CrossRef][Web of Science][Medline]
McLachlan RI, Mallidis C, Ma K, Bhasin S and de Kretser DM (1998) Genetic disorders and spermatogenesis. Reprod Fertil Dev 10,97–104.[CrossRef][Medline]
Mulhall JP, Reijo R, Alagappan R, Brown L, Page D, Carson R and Oates RD (1997) Azoospermic men with deletion of the DAZ gene cluster are capable of completing spermatogenesis: fertilization, normal embryonic development and pregnancy occur when retrieved testicular spermatozoa are used for intracytoplasmic sperm injection. Hum Reprod 12,503–508.
Nakagata N (2000) Cryopreservation of mouse spermatozoa. Mamm Genome 11,572–576.[CrossRef][Web of Science][Medline]
Nolan PM, Peters J, Vizor L, Strivens M, Washbourne R, Hough T, Wells C, Glenister P, Thornton C, Martin J et al. (2000) Implementation of a large-scale ENU mutagenesis program: towards increasing the mouse mutant resource. Mamm Genome 11,500–506.[CrossRef][Web of Science][Medline]
Noveroske JK, Weber JS and Justice MJ (2000) The mutagenic action of N-ethyl-N-nitrosourea in the mouse. Mamm Genome 11,478–483.[CrossRef][Web of Science][Medline]
O’Bryan MK, Sebire KL, Gerdprasert O, Hedger MP, Hearn MT and de Kretser DM (2000) Cloning and regulation of the rat activin betaE subunit. J Mol Endocrinol 24,409–418.[Abstract]
Papathanasiou P, Perkins AC, Cobb BS, Ferrini R, Sridharan R, Hoyne GF, Nelms KA, Smale ST and Goodnow CC (2003) Widespread failure of hematolymphoid differentiation caused by a recessive niche-filling allele of the Ikaros transcription factor. Immunity 19,131–144.[CrossRef][Web of Science][Medline]
Popp RA, Bailiff EG, Skow LC, Johnson FM and Lewis SE (1983) Analysis of a mouse alpha-globin gene mutation induced by ethylnitrosourea. Genetics 105,157–167.
Ramaswamy S, Marshall GR, McNeilly AS and Plant TM (1999) Evidence that in a physiological setting Sertoli cell number is the major determinant of circulating concentrations of inhibin B in the adult male rhesus monkey (Macaca mulatta) J Androl 20,430–434.
Rich KA and de Kretser DM (1977) Effect of differing degrees of destruction of the rat seminiferous epithelium on levels of serum follicle stimulating hormone and androgen binding protein. Endocrinology 101,959–968.
Rich KA, Kerr JB and de Kretser DM (1979) Evidence for Leydig cell dysfunction in rats with seminiferous tubule damage. Mol Cell Endocrinol 13,123–135.[CrossRef][Web of Science][Medline]
Robertson DM, Hayward S, Irby D, Jacobsen J, Clarke L, McLachlan RI and de Kretser DM (1988) Radioimmunoassay of rat serum inhibin: changes after PMSG stimulation and gonadectomy. Mol Cell Endocrinol 58,1–8.[CrossRef][Web of Science][Medline]
Robertson DM, Giacometti M, Foulds LM, Lahnstein J, Goss NH, Hearn MT and de Kretser DM (1989) Isolation of inhibin alpha-subunit precursor proteins from bovine follicular fluid. Endocrinology 125,2141–2149.
Robertson DM, Foulds LM, Prisk M and Hedger MP (1992) Inhibin/activin beta-subunit monomer: isolation and characterization. Endocrinology 130,1680–1687.
Russell LB and Montgomery CS (1982) Supermutagenicity of ethylnitrosourea in the mouse spot test: comparisons with methylnitrosourea and ethylnitrosourethane. Mutat Res 92,193–204.[CrossRef][Web of Science][Medline]
Russell WL, Kelly EM, Hunsicker PR, Bangham JW, Maddux SC and Phipps EL (1979) Specific-locus test shows ethylnitrosourea to be the most potent mutagen in the mouse. Proc Natl Acad Sci USA 76,5818–5819.
Russell LB, Ettlin R, Hikim A and Clegg E (1990) Histological and Histopathological Evaluation of the Testis. Cache River Press, New York.
Sanchez-Partida L-G (2003) Sperm cryopreservation. In Nagy A, Gertstein M, Vintersten K and Behringer R (eds), Manipulating the Mouse Embryo: A Laboratory Manual. CSHL Press, New York, pp. 00–00.
Sharpe RM, Turner KJ, McKinnell C, Groome NP, Atanassova N, Millar MR, Buchanan DL and Cooke PS (1999) Inhibin B levels in plasma of the male rat from birth to adulthood: effect of experimental manipulation of Sertoli cell number. J Androl 20,94–101.
Singer B and Dosanjh MK (1990) Site-directed mutagenesis for quantitation of base-base interactions at defined sites. Mutat Res 233,45–51.[Web of Science][Medline]
Spratt DI, Cox P, Orav J, Moloney J and Bigos T (1993) Reproductive axis suppression in acute illness is related to disease severity. J Clin Endocrinol Metab 76,1548–1554.[Abstract]
Ward JO, Reinholdt LG, Hartford SA, Wilson LA, Munroe RJ, Schimenti KJ, Libby BJ, O’Brien M, Pendola JK, Eppig J et al. (2003) Toward the genetics of mammalian reproduction: induction and mapping of gametogenesis mutants in mice. Biol Reprod 69,1615–1625.
Submitted on August 23, 2005; revised on November 22, 2005; accepted on November 28, 2005
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. L. Borg, K. M. Wolski, G. M. Gibbs, and M. K. O'Bryan Phenotyping male infertility in the mouse: how to get the most out of a 'non-performer' Hum. Reprod. Update, March 1, 2010; 16(2): 205 - 224. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P Sipila, J Jalkanen, I T Huhtaniemi, and M Poutanen Novel epididymal proteins as targets for the development of post-testicular male contraception Reproduction, March 1, 2009; 137(3): 379 - 389. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Harris, B. Marquez, S. Suarez, and J. Schimenti Sperm Motility Defects and Infertility in Male Mice with a Mutation in Nsun7, a Member of the Sun Domain-Containing Family of Putative RNA Methyltransferases Biol Reprod, August 1, 2007; 77(2): 376 - 382. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Furnes and J. Schimenti Fast forward to new genes in mammalian reproduction J. Physiol., January 1, 2007; 578(1): 25 - 32. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||