PiggyBac Transposon-Mediated Mutagenesis in Rats Reveals a Crucial Role of Bbx in Growth and Male Fertility.

Bobby sox homolog (Bbx) is an evolutionally conserved gene, but its biological function remains elusive. Here, we characterized defects of Bbx mutant rats that were created by PiggyBac-mediated insertional mutagenesis. Smaller body size and male infertility were the two major phenotypes of homozygous Bbx mutants. Bbx expression profile analysis showed that Bbx was more highly expressed in the testis and pituitary gland than in other organs. Histology and hormonal gene expression analysis of control and Bbx-null pituitary glands showed that loss of Bbx appeared to be dispensable for pituitary histogenesis and the expression of major hormones. BBX was localized in the nuclei of postmeiotic spermatids and Sertoli cells in wild-type testes, but absent in mutant testes. An increased presence of aberrant multinuclear giant cells and apoptotic cells was observed in mutant seminiferous tubules. TUNEL-positive cells costained with CREM (round spermatid marker), but not PLZF (spermatogonia marker), gammaH2Ax (meiotic spermatocyte marker), or GATA4 (Sertoli cell marker). Finally, there were drastically reduced numbers and motility of epididymal sperm from Bbx-null rats. These results suggest that loss of BBX induces apoptosis of postmeiotic spermatids and results in spermiogenesis defects and infertility.

INTRODUCTION

In contrast to the many available lines of genetically engineered mice, a large collection of genetically modified rats as alternative models for the functional study of human disease is currently lacking. To create additional genetically modified rats to model human diseases, transposon-tagged insertion in the rat genome has been developed to achieve mutations via nonbiased mutagenesis [1-3]. Previously, we created and demonstrated hybrid Sleeping Beauty and PiggyBac (PB) transposon-mediated insertional mutagenesis in conjunction with a visible coat color marker (tyrosinase) that can be mobilized in the rat genome in the presence of corresponding transposases using a simple breeding scheme [4]. Rats carrying homozygous mutations are ultimately generated for phenotype-driven studies. Lethality and growth restriction can easily be identified during the phenotypic screen if the transposon-disrupted genes are critical for development.

The impairment of fertility by the mutation of specific genes can be directly measured by the capability of progeny production of a test cross of homozygous mutant rats. Male infertility can be caused by a deficit in male germ cell production. Development of male germ cells, the spermatozoa, takes place in the seminiferous tubular epithelium of the testes. This process of spermatogenesis can be subdivided into three phases: the proliferation phase for self-renewal of spermatogonial stem cells and generation of progenitors; the meiotic phase for reduction of chromosome number in spermatocytes; and the spermiogenesis phase for differentiation of haploid spermatids [5, 6]. During these phases, Sertoli cells line the seminiferous tubules, making contact with male germ cells and supporting their development. In sequential cross-sections of seminiferous tubules, distinct morphologies of developing germ cells occur in a nonrandom manner and are defined as stages that vary across species [7]. For example, there are 14 stages in rats [8], 12 stages in mice [9], and 6 stages in humans [10, 11]. The sequential progression of spermatogenic stages over time is called a cycle, and approximately four cycles are required for the completion of spermatogenesis. Because the duration of the spermatogenic process is about 13 days in rats, 8.6 days in mice, and 16 days in humans [5, 7], the entire spermatogenic process from a given spermatogonial stem cell to the generation of spermatocytes, spermatids, and finally spermatozoa requires about 52 days in rats, 33 days in mice, and 64 days in humans [7]. Spermatozoa are then released from contact with Sertoli cells into the seminiferous tubal lumen, travel through the caput and corpus epididymis, and finally reach and are stored in the caudal epididymis.

Bobby sox homolog (Bbx) belongs to the high-mobility group (HMG) superfamily of transcription factors and is an evolutionally conserved gene found in worms, flies, fishes, birds, and mammals [12]. BBX is also known as HMG box-containing protein 2, initially identified by functional complementation of defective replication in yeast cells, suggesting that BBX acts as a positive regulator of G1/S transition [13]. Bbx is also expressed in ventricular zone progenitor cells in the mouse developing neocortex and hippocampus [12, 14]. Bbx is repressed by the transcription factor, nuclear factor one X, which plays a key role during neocortex and hippocampus development. These findings imply that Bbx regulates the self-renewal and differentiation of neuronal progenitor cells [14]. A recent study also showed that BBX is expressed and involved in odontoblast differentiation of dental pulp stem cells and progenitors [15]. Although these previous studies addressed the role of BBX in the differentiation of some types of stem cells and progenitors, the biological function of Bbx remains largely unknown.

Here, we describe a new PB insertion into intron 2 of the Bbx locus, referred to as Bbx. We found that the most prominent phenotypes of Bbx rats were growth restriction and male sterility. Bbx was highly expressed in the testes and pituitary gland. Although the loss of Bbx appeared to be dispensable for the expression of many pituitary hormonal genes, Bbx male rats exhibited impairments in testicular function compared with wild-type rats.

MATERIALS AND METHODS

Rats

Wistar Kyoto (WKY/CrlBltw) rats were purchased from BioLASCO Taiwan. The progeny from WKY/CrlBltw rats bred with seed rats, which were generated by crossing PB transposon (Bhr7)-carrying rats and UBC-PB transposase transgenic rats, were screened by splinkerette PCR (see below) to identify new insertions of Bhr7. We identified a male founder carrying the Bhr7 insertion in intron 2 of the Bbx locus on chromosome 11. Subsequently, the Bbx founder was bred with WKY/CrlBltw females for generating Bbx males and females. Bbx males and females were intercrossed and produced Bbx rats and their littermate controls (wild type and Bbx). All animal experiments were performed with approval of the Institutional Animal Care and Use Committee at National Yang-Ming University.

Splinkerette PCR

We performed splinkerette PCR to identify the location of transposon insertion in the rat genome. The screening procedure was modified from that of Potter and Luo [16]. The primer sequences used in splinkerette PCR are listed in Table 1. First, EcoRI-, BamHI-, or XbaI-digested genomic DNA was individually ligated to double-stranded splinkerette adaptors, which were prepared by annealing of the primer, SpaA, with SpbB_EcoRI, SpbB_BamHI, or SpbB_XbaI. The primers, Sp0F and BacE, designed from adaptor and transposon sequences, respectively, were used for the first round of PCR. The products from the first round of PCR were amplified by nested PCR using the primers, Sp1F and BacF.

TABLE 1

Primers used for splinkerette PCR and genotyping.

Genotyping

Genotyping was performed by PCR of genomic DNA from 8-day-old rat toes. The Bbx wild-type allele was detected by the primers, Bbx-F and Bbx-R, yielding a 513-bp product. The transposon-inserted allele was detected by the primers, Bbx-F and PB-3′TR (AS), yielding a 300-bp product. The primer sequences used for genotyping are listed in Table 1.

Tissue Processing and Histology

Testes were fixed in modified Davidson fixative solution (30% of 37% solution of formaldehyde, 5% glacial acetic acid, and 15% ethanol, and 50% double-distilled H2O) at 4°C overnight [17]. Pituitary glands were fixed in 10% buffered formalin at 4°C overnight. Next, tissues were washed with 1× PBS and stored in 70% ethanol. Tissues were processed and embedded as previously described [18, 19]. Embedded tissues were sectioned into 5-μm slices. Slides containing the sections were deparaffinized, rehydrated, and stained with Harris hematoxylin and eosin Y (H&E; Sigma-Aldrich) as previously described [18, 19].

Immunohistochemistry

Detailed immunohistochemical and immunofluorescence procedures were previously described [20-22]. Sections were incubated overnight with primary antibodies against BBX (rabbit IgG, 1:200; Proteintech) and growth hormone (GH; goat IgG, 1:100; Santa Cruz Biotechnology) at 4°C overnight. For immunofluorescence, sections were stained with BBX (rabbit IgG, 1:200; Proteintech), GH (goat IgG, 1:100; Santa Cruz Biotechnology), GATA4 (mouse IgG, 1:100; Santa Cruz Biotechnology), γH2AX (mouse IgG, 1:600; BD Pharmingen), PLZF (rabbit IgG, 1:100; Santa Cruz Biotechnology), or CREM1 (rabbit IgG, 1:50; Santa Cruz Biotechnology) at 4°C overnight. After washing, sections were incubated with secondary antibodies conjugated with Alexa Fluor 488 or Alexa Fluor 568 (1:200; Invitrogen) at room temperature for 1 h. Finally, sections were counterstained with 0.5 ng/ml DAPI (4′,6-diamidino-2-phenylindole) for 3 min followed by mounting with fluorescent mounting medium (DakoCytomation).

TUNEL Assay

TUNEL staining was performed using the FragEL DNA Detection Kit (Calbiochem) followed by staining with primary antibodies against GATA4 (mouse IgG, 1:100; Santa Cruz Biotechnology), PLZF (rabbit IgG, 1:100; Santa Cruz Biotechnology), γH2AX (mouse IgG, 1:600; BD Pharmingen), or CREM1 (rabbit IgG, 1:50; Santa Cruz Biotechnology). Nuclei were counterstained with DAPI.

Real-Time RT-PCR

Total RNA from the testes and pituitary glands of 10-wk-old wild-type, Tn/+, and Tn/Tn rats were extracted by TRIzol reagent (Invitrogen) and reverse transcribed to cDNA using the SuperScript III First-Strand Synthesis System (Invitrogen). Real-time PCR reactions were carried out using a Kapa SYBR Fast qPCR kit (Kapa Biosystems). The primers used for real-time PCR are listed in Table 2.

TABLE 2

Primers used for real-time RT-PCR.

F, forward; R, reverse.

Western Blotting

Protein lysates were extracted from tissue ground with lysis buffer (50 mM Tris-HCl [pH 7.9], 150 mM NaCl, 0.1% SDS, 1% NP40, 0.5% sodium deoxycholate, protease inhibitor cocktail [Roche Applied Science], and phosphatase inhibitor cocktail [Sigma-Aldrich]). Approximately 30 μg of protein lysate per lane was used for SDS-PAGE and immunoblotting. Membranes were blocked with 5% bovine serum albumin or 5% milk in TBS-T and incubated with primary antibodies against BBX (rabbit IgG, 1:1500; Proteintech) or βactin (mouse IgG, 1:5000; Sigma-Aldrich).

Sperm Analysis

The epididymis was dissected into small pieces in warm 1× PBS and incubated in a 37°C water bath for 30 min to release epididymal content, which was collected for sperm counting, sperm smear, and sperm motility analysis. The content was placed into culture dishes, and sperm movement videos were recorded under a microscope. Total sperm numbers were counted with a hemocytometer. To examine sperm morphology, 20 μl of epididymal content was dropped onto slides for sperm smears. The slides were immersed in fixative solution (95% alcohol and 5% acetic acid) for 3 min, washed twice with 70% ethanol, and stained with H&E.

Statistical Analysis

Statistical analysis was performed using paired Student t-tests. Data are shown as mean ± SD. Statistical significance was set at P < 0.05.

RESULTS

Insertional Mutation by Bhr7 Transposon Creates Bbx-Null Rats

The Bhr7 transposon was previously described [4]. Briefly, it contains a splice acceptor IRESrtTApA cassette followed by the Tyrosinase gene. Insertion into an intron should truncate the endogenous transcript and result in pigmentation using an albino genetic background. Using the splinkerette PCR method [16], we identified one dark-pigmented male carrying the Bhr7 insertion at the 51 711 655 nucleotide position in intron 2 (51 708 413–51 712 818 nucleotides) of the Bbx gene (Fig. 1A) on chromosome 11 (http://www.ensembl.org/) [23]; this founder was referred to as Bbx F1 progeny were obtained by breeding the Bbx founder with Wky female rats. Intercrosses of Bbx F1 males and females generated F2 rats that underwent PCR-based gene-specific genotyping (Fig. 1B). From the genotype analysis of 202 F2 rats (33 litters) at Postnatal Days 7–10, the numbers of wild-type, Bbx, and Bbx rats were 49 (24.3%), 127 (62.9%), and 26 (12.9%). The actual numbers of Bbx rats were lower than expected, suggesting that some mutant rats died before genotyping. Bbx rats had significantly smaller body sizes and weights compared with their littermates at 10 and 12 wk of age (Fig. 1, C and D). Next, we examined whether Bbx rats were capable of producing progeny. Five Bbx females mated with Bbx males gave birth to a total of 35 pups. However, no Bbx males (n = 5) mated with wild-type females yielded progeny, indicating that Bbx males were infertile. To investigate the function of Bbx in rats, we examined Bbx expression patterns in various organs using real-time quantitative RT-PCR (RT-qPCR). We found that Bbx was highly expressed in the testes and moderately expressed in both male and female pituitary glands compared with other internal organs (Fig. 1E) and brain subregions (cortex, hippocampus, thalamus, and cerebellum; Supplemental Fig. S1; Supplemental Data are available online at www.biolreprod.org). In the testes and pituitary glands, Bbx expression levels were reduced by approximately 50% in heterozygous mutants and were undetectable in homozygous mutants (Fig. 1, F and G). These results suggest that Bhr7 insertion into the Bbx locus creates a null allele.

FIG. 1

Transposon insertion and expression of the Bbx locus in rats. A) Graphic illustration showing the Bhr7 transposon insertion into intron 2 of rat Bbx. B) Representative Bbx-specific genotyping by PCR. C) Representative image showing body size of wild-type, Bbx, and Bbx rats at 12 wk of age. D) Body weight of wild-type (21 male and 6 female), Bbx (22 male and 10 female), and Bbx (15 male and 7 female) rats at 10 wk of age. E) RT-qPCR showed relative expression levels of Bbx in different organs of Wky male and female rats (n = 3) at 10 wk of age. F) RT-qPCR showed Bbx expression in the testes of wild-type (+/+), Bbx, and Bbx rats (n = 3 per genotype) at 10 wk of age. G) RT-qPCR showed Bbx expression in the pituitary glands of wild-type (+/+), Bbx, and Bbx male and female rats (n = 3 per genotype) at 10 wk of age. *P < 0.05; **P < 0.01; ***P < 0.001.

Bbx Appears Dispensable for the Expression of Many Pituitary Hormones

To detect Bbx expression in the pituitary gland, we performed immunohistochemistry using an antibody against BBX. We found that BBX localized in the nuclei of control pituitary glands, but was undetectable in Bbx-null mutants (Fig. 2A). Loss of BBX in the Bbx pituitary was also confirmed by Western blotting (Fig. 2B). Using coimmunofluorescent staining of BBX and GH, we found that BBX was expressed in both GH-expressing and GH-nonexpressing cells (Fig. 2C). However, Bbx loss appeared to be dispensable for expression of Prop1 and Pou1f1 (Fig. 2D), both of which control pituitary organogenesis and direct hormonal gene expression [24]. Furthermore, we examined the expression of several hormonal genes, including Gh1, Lhb, LSHb, Pomc, Prl, and Tshb, using RT-qPCR in Bbx, Bbx, and Bbx males (n = 3 per genotype) and females (n = 3 per genotype). We found no significant difference in the expression of these hormonal genes except for a reduction of Lhb expression in Bbx females (Fig. 2E). Thus, most hormonal genes were unaffected in Bbx-null mutants.

FIG. 2

Expression of Bbx and hormonal genes in pituitary glands of Bbx wild-type and mutant rats. A) Immunohistochemistry analysis showed the presence of nuclear BBX in the wild-type pituitary glands, but the lack of BBX in the Bbx pituitary gland, at 10 wk of age. B) Western blot analysis showed the presence of BBX in wild-type and Bbx pituitary glands, but the lack of BBX in the Bbx pituitary gland. βActin served as an internal control. C) Coimmunofluorescent staining of BBX (green) and GH (red) showed the coexpression of BBX and GH in the wild-type pituitary gland, but the lack of BBX in the Bbx pituitary gland. Nuclei were counterstained with DAPI. D) RT-qPCR showed relative levels of Prop1 and Pou1f1, which encode transcription factors for pituitary organogenesis and hormonal gene expression, in the pituitary glands of wild-type (+/+), Bbx, and Bbx males (n = 3 per genotype) and females (n = 3 per genotype) at 10 wk of age. E) RT-qPCR showed relative expression levels of GH1, Lhb (luteinizing hormone subunit β), Fshb (follicle-stimulating hormone subunit β), Pomc (pro-opiomelanocortin), Prl (prolactin), and Tshb (thyroid-stimulating hormone subunit β). *P < 0.05.

Bbx Loss Impairs Spermiogenesis

Because BBX is highly expressed in the testes, we focused on potential testicular defects in Bbx-null rats. We dissected the male reproductive system and found that testes in Bbx rats were smaller than those in their littermates (Bbx and Bbx) at 12 wk of age (Fig. 3A). We measured testes weight normalized by body weight in Bbx, Bbx, and Bbx rats at 3, 8, 10, 12, 15, and 20 wk of age. We observed a drastic decrease in the testes weight of Bbx rats after 8 wk of age (Fig. 3B). Using immunohistochemistry, we found that BBX was highly expressed in developing spermatids in Bbx+/+ testes, more specifically in round spermatids at stages V–VIII, elongated spermatids at stages IX–XI, and Sertoli cells (Supplemental Fig. S2). By contrast, BBX was undetectable in Bbx-null seminiferous tubules (Fig. 3C). Using Western blotting, we confirmed that BBX was expressed in Bbx and Bbx testes and absent in Bbx mutant testes (Fig. 3D). We next examined the histology of Bbx-null and wild-type testes at 8, 10, and 20 wk of age. In wild-type testes, multiple cellular layers in seminiferous tubule cross-sections were revealed by H&E staining (Fig. 3E), and high-magnification views showed orderly arrangement of germ cells within the tubules (Fig. 3E). Similarly, normal appearance of seminiferous tubules was observed in 8-wk Bbx-null testes. However, the seminiferous tubules of Bbx testes appeared to be abnormal, with clear atrophy after 10 wk of age (Fig. 3E). In high-magnification views, abnormal cells lacking the appearance of mature spermatozoa emerged at the lumens of the seminiferous tubules at 8 wk of age (Fig. 3E). From 10 wk of age onward, many tubules of the Bbx testes showed germ cell loss and an increase in multinucleated giant cells (Fig. 3E). To further confirm that Bbx is expressed in round spermatids and Sertoli cells in the seminiferous epithelium, we performed coimmunofluorescence staining of antibodies against BBX and GATA4, which is expressed in postnatal mouse Sertoli cells [25]. We detected BBX in multiple layers of round spermatids and its coexpression with GATA4 in the nuclei of Sertoli cells in Bbx seminiferous tubules (Fig. 3F). In Bbx seminiferous tubules, Bbx was undetectable, whereas GATA4 was expressed in Sertoli cells (Fig. 3F). We further costained testes sections of 10-wk-old rats with antibodies against BBX and γH2AX, which marks DNA breaks during meiosis in spermatocytes [26, 27]. We found exclusive BBX staining in round spermatids, which were distinct from γH2AX-positive spermatocytes (Fig. 3F). In Bbx testes, γH2AX-stained cells were disorganized and detached in the lumens of atrophic tubules (Fig. 3F). Taken together, these results suggest that Bbx is expressed in Sertoli cells and postmeiotic spermatids in the seminiferous tubules, indicative of its role in spermiogenesis.

FIG. 3

Loss of Bbx causes testicular atrophy. A) Gross appearances of testes of wild-type (+/+), Bbx, and Bbx rats at 12 wk of age. B) Quantification of testes weight normalized by body weight for wild-type (+/+), Bbx, and Bbx rats at 3 wk (+/+, n = 3; Tn/+, n = 8; Tn/Tn, n = 6), 8 wk (+/+, n = 3; Tn/+, n = 9; Tn/Tn, n = 4), 10 wk (+/+, n = 9; Tn/+, n = 11; Tn/Tn, n = 5), 12 wk (+/+, n = 5; Tn/+, n = 8; Tn/Tn = 4), 15 wk (+/+, n = 3; Tn/+, n = 4; Tn/Tn, n = 3), and 20 wk (+/+, n = 4; Tn/+, n = 3; Tn/Tn, n = 3) of age. **P < 0.01; ***P < 0.001. C) Immunohistochemical analysis of testes sections showed the presence of nuclear BBX in wild-type seminiferous tubules, but the lack of BBX in Bbx seminiferous tubules at 10 wk of age. D) Western blot analysis showed the presence of BBX in wild-type (n = 2) and Bbx (n = 3) testes, but not in Bbx (n = 3) testes; βactin served as an internal control. E) H&E-stained sections of testes from wild-type and Bbx rats at 8, 10, and 20 wk of age. Red asterisks indicate atrophic seminiferous tubules; black arrows indicate aberrant multinuclear giant cells. F) Coimmunofluorescent staining of BBX (green) and GATA4 (red; Sertoli cell marker) or γH2Ax (red; meiotic spermatocyte marker) shows BBX specifically in Sertoli cells and postmeiotic spermatids in the wild-type testis, but the lack of BBX in the Bbx testis. Nuclei were counterstained by DAPI.

Increased TUNEL-Positive Cells in Bbx-Deficient Seminiferous Tubules

We further examined whether apoptosis might cause germ cell loss in Bbx-null mutant rats. Bbx seminiferous tubules exhibited more TUNEL-positive cells at 8 wk of age and even more cells at 10 and 20 wk of age compared with wild-type seminiferous tubules, which showed undetectable TUNEL staining or only a few TUNEL-positive cells (Fig. 4A). When we quantified TUNEL-positive cells in the seminiferous tubules of Bbx, Bbx, and Bbx rats at 8, 10, 12, and 20 wk of age, we found a significantly greater number of TUNEL-positive cells in Bbx seminiferous tubules than in wild-type or Bbx tubules at all time points (Fig. 4B). We further examined whether Bbx Sertoli cells also underwent apoptosis. Immunofluorescent staining revealed that GATA4-positive cells were TUNEL-negative cells (Fig. 4C), suggesting that Bbx Sertoli cells are apoptosis-resistant cells. Moreover, we performed TUNEL assay and immunofluorescent staining of PLZF [28] for marking spermatogonia, γH2AX for marking spermatocytes [26, 27], and CREM for marking postmeiotic spermatids [29, 30]. We found that TUNEL-positive signals mainly overlapped with CREM-positive spermatids, but not with GATA4-, PLZF-, or γH2AX-positive cells (Fig. 4C). These findings suggest that loss of Bbx causes death of postmeiotic spermatids and impairs spermiogenesis.

FIG. 4

Increased TUNEL-positive spermatids in Bbx testes. A) Bbx seminiferous tubules exhibited more TUNEL-positive cells (green) than wild-type seminiferous tubules at 8, 10, and 20 wk of age. B) Quantification of TUNEL-positive cells in wild-type, Bbx, and Bbx seminiferous tubules at 8 wk (n = 3 per genotype), 10 wk (+/+, n = 4; Tn/+, n = 3; Tn/Tn, n = 4), and 20 wk (n = 3 per genotype) of age. *P < 0.05; ***P < 0.001. C) Low-magnification (upper panels) and high-magnification (lower panels; boxed areas of corresponding upper images) views of coimmunofluorescent staining of TUNEL (green) and GATA4 (red; Sertoli cell marker), PLZF (red; spermatogonial cell marker), γH2Ax (red; meiotic spermatocyte marker), or CREM (red; postmeiotic spermatid marker) showed greater coexpression of TUNEL and CREM1-positive signals in the seminiferous tubules of Bbx rats than those of wild-type rats at 10 wk of age. Nuclei were counterstained by DAPI.

Impaired Sperm Production and Motility in Bbx-Null Rats

By counting sperm number in the epididymis of wild-type, Bbx, and Bbx rats, we found a drastic reduction of sperm number in Bbx rats compared with wild-type or Bbx rats at 10 and 12 wk of age (Fig. 5A). After 15 wk of age, Bbx rats exhibited azoospermia, whereas Bbx and wild-type rats had substantial amounts of spermatozoa (Fig. 5A). At 10 wk of age, spermatozoa were detected in the epididymis of Bbx rats, but were largely abnormal, with misshapen sperm heads in contrast to the hook-shaped heads of wild-type sperm (Fig. 5B). The amount of abnormal spermatozoa produced in Bbx rats was significantly higher than that produced in wild-type rats (Fig. 5C). In addition, we observed that the spermatozoa produced in Bbx rats were immotile (Supplemental Movies S1 and S2). Taken together, our results demonstrate that the infertility of Bbx rats was mainly caused by impaired spermiogenesis.

FIG. 5

Impaired sperm production in Bbx rats. A) Number of sperm in the epididymis of wild-type (+/+), Bbx, and Bbx rats at 10 wk (+/+, n = 8; Tn/+, n = 8; Tn/Tn, n = 5), 12 wk (+/+, n = 4; Tn/+, n = 6; Tn/Tn, n = 10), 15 wk (+/+, n = 3; Tn/+, n = 4; Tn/Tn, n = 3), and 20 wk (n = 3 per genotype) of age. B) Images of abnormal sperm morphology in Bbx rats compared with wild-type rats at 12 wk of age. C) Quantification of abnormal sperm in the epididymis of wild-type (+/+) and Bbx rats at 10 and 12 wk of age (n = 3 per genotype and time point). *P < 0.05; ***P < 0.001.

DISCUSSION

In this study, we used PB-based mutagenesis screening [4] to identify Bbx-null rats, which were smaller than wild-type rats and exhibited male infertility. Two major organs, the pituitary gland and testes, were specifically analyzed due to their higher Bbx expression compared with other organs in wild-type Wky rats. Loss of Bbx in the pituitary glands appeared to be dispensable for pituitary organogenesis and hormonal gene expression. Although previous studies [12, 14] suggested that Bbx has a role in cortex and hippocampus, we did not observe obvious behavioral phenotypes in Bbx rats compared to controls (data not shown). In the testes, Bbx was expressed in the Sertoli cells and postmeiotic round spermatids. By 8 wk of age, histology of the testes appeared to be normal in Bbx rats, suggesting that Bbx is dispensable for the early spermatogenesis. After 10 wk of age, loss of Bbx resulted in the presence of aberrant multinucleated giant cells and apoptotic spermatids in the seminiferous tubules. These defects might be caused by Bbx deficiency in either the Sertoli cells or round spermatids, or both cell types, at the adult stage. Subsequently, loss of Bbx caused azoospermia and infertility. Aberrant multinucleated giant cells were previously observed in rats with vasectomy-induced spermatogenic impairments [31]. In addition, genetically modified mice, such as Crem- [29, 30], Trf2- [32], and Fdnc3a-deficient mice [33], display spermatogenic defects with aberrant multinucleated giant cells in the seminiferous tubules.

A recent public phenotyping screen (the Wellcome Trust Sanger Institute Mouse Portal; http://www.sanger.ac.uk/mouseportal/) [34], which identified Bbx knockout mice carrying a splicing acceptor reporter-tagged insertion in intron 5 of the Bbx locus, revealed decreased lean mass in homozygous males and females; decreased body length, heart weight, and bone mineral density/content in homozygous females; and abnormal tooth morphology in homozygous mutant males. However, the reproductive ability of Bbx mutant mice appeared to be unaffected. As previously described, the complete spermatogenic process varies among species [5, 7], which could lead to different outcomes of the loss of Bbx between mice and rats. It is also possible that the insertion into intron 5 in the mouse does not create a null allele. Thus, genetically altered rats could serve as alternative animal models for better understanding of comparative gene function across different species.

In summary, our findings demonstrate the fundamental role of BBX in spermiogenesis in rats using transposon-mediated insertional mutagenesis. Our study provides an entry for future studies of the BBX transcription factor and its potential target genes for regulating spermiogenesis. In addition, future clinical studies could examine whether BBX is associated with azoospermia in humans.