Reproductive Function in Protein Kinase Inhibitor-Deficient Mice

Abstract

The protein kinase inhibitor (PKI) family includes three genes encoding small, heat-stable inhibitors of the cyclic AMP-dependent kinase PKA. Each PKI isoform contains a PKA inhibitory domain and a nuclear export domain, enabling PKI to both inhibit PKA and remove it from the nucleus. The PKIβ isoform, also known as testis PKI, is highly expressed in germ cells of the testis and is found at more modest levels in other tissues. In order to investigate its physiological role, we have generated PKIβ knockout mice by gene targeting. These mice exhibit a partial loss of PKI activity in testis but remain fertile with normal testis development and function. PKIβ knockout females also reproduce normally. The PKIβ mutants were crossed with our previously derived PKIα mutants to obtain double-knockout mice. Remarkably, these mice are also viable and fertile with no obvious physiological defects in either males or females.

Many hormones and neurotransmitters initiate their physiological actions by stimulating production of the intracellular second messenger, cyclic AMP (cAMP). The downstream target of cAMP is the cAMP-dependent kinase PKA. Many other proteins and enzymes modulate this pathway, including PKI, a highly specific inhibitor of PKA.

PKI is a small, heat-stable protein with high affinity for the catalytic subunit of PKA, and binding of PKI to the catalytic subunit inhibits its activity (15, 23). PKI also contains a leucine-rich region that has been shown to function as a nuclear export signal when PKI is bound to PKA (8, 24). PKI is capable of freely entering the nucleus and actively shuttling the catalytic subunit of PKA back to the cytoplasm, where PKA regulatory subunits are located. By facilitating nuclear export of PKA, PKI is thought to affect the kinetics and/or extent of PKA activity in the nucleus. PKI may, for example, terminate the transcriptional regulation by PKA of specific genes and rapidly reset the PKA system for subsequent gene induction responses.

There are three distinct PKI genes encoding homologous isoforms, referred to as PKIα, PKIβ, and PKIγ (7, 12, 14, 19, 20). Each of these isoforms has a unique tissue expression pattern (2, 7, 16, 19). The PKIα isoform is highly expressed in skeletal muscle, heart, cerebral cortex, and cerebellum, whereas the PKIβ isoform (originally called testis PKI) is most highly expressed in testis, with a small amount of expression in brain and little to none elsewhere. PKIγ mRNA is widely expressed and found most highly expressed in heart and testis. Some tissues possess multiple isoforms of PKI, in which case the expression pattern is cell specific. In the testis, for example, PKIα is localized to the Sertoli cells and PKIβ is localized to the germ cells (18).

We previously reported on the gene targeting of the PKIα gene in mice (9). Despite the presence of PKIα in Sertoli cells of the wild-type testis, PKIα knockout mice showed no defect in testis development, spermatogenesis, or fertility, demonstrating the expendability of PKIα for Sertoli cell function. In contrast, defects in skeletal muscle, where PKIα is most highly expressed in the wild type, were observed. Knockout skeletal muscle showed a complete absence of PKI activity, suggesting a lack of any compensation by other PKI isoforms. Surprisingly, the mice exhibited a counterintuitive decrease in basal PKA activity and a reduction in both basal and isoproterenol-induced gene expression, apparently as a consequence of diminished phosphorylation of the transcription factor CREB. These results challenged the prevailing view that PKI is required simply to maintain low basal PKA activity and terminate the nuclear actions of PKA.

The important role of the cAMP-PKA pathway in testis development and function has a long history of investigation. Follicle-stimulating hormone, a gonadotropin that signals through cAMP, is essential for normal development of the testis and production of normal numbers of sperm. The motility of mature sperm is stimulated by cAMP and phosphodiesterase inhibitors (10, 17), and this stimulation is likely to be mediated by PKA (21). Similarly, capacitation and the acrosome reaction involve PKA (11, 22). Based on these and other studies, it was believed that PKA played a pivotal role in testis function. It was unexpected, therefore, when gene knockouts of the testicular PKA subunit isoforms RIIα (6) and RIIβ (4) produced mice with normal fertility, sperm development, and function. Of the five viable knockout mouse lines with mutations in individual PKA subunit isoforms (4–6, 13), only the Cα catalytic subunit knockout displayed testicular dysfunction (B. S. Skålhegg, Y. Huang, T. Su, R. L. Idzerda, G. S. McKnight, and K. A. Burton, submitted for publication). The Cα knockout mice had sperm that were nearly devoid of PKA activity and lacked forward motility. Analysis of the other four various PKA knockout mouse lines suggested that compensation by other PKA isoforms allowed for normal cAMP signaling, thus preventing manifestations of testicular defects.

The high level of PKIβ in germ cells of the testis suggests that it may play an important role in cAMP signaling in these cells. To elucidate the physiological function of PKIβ, we have created PKIβ-deficient mice by homologous recombination in embryonic stem (ES) cells. By interbreeding with our previously derived PKIα knockout mice (9), we have also produced PKIα/β double-knockout mice. We report here on the phenotypes of these mouse mutants.

MATERIALS AND METHODS

Construction of the PKIβ targeting vector and generation of mutant mice.

A PKIβ genomic clone was isolated from a 129SV/J mouse genomic library and given to us by M. Uhler (University of Michigan). The 4.7-kb EcoRI-AflII fragment, spanning exon 3, was flanked with thymidine kinase cassettes to facilitate negative selection. The 0.5-kb BamHI-SpeI fragment encompassing exon 3 was replaced with a 1.9-kb neomycin phosphotransferase cassette to facilitate positive selection. This strategy deleted the inhibitory and nuclear export domains of PKIβ.

Gene targeting in ES cells was performed essentially as described previously (3), using E14.1 cells (provided by W. Muller and K. Rajewsky) derived from 129/Ola mice. G418- and ganciclovir-resistant colonies were picked, isolated, and screened by genomic Southern blot analysis. Four clones were identified as having undergone specific homologous recombination and were microinjected into C57BL/6 blastocysts, yielding nine chimeric male mice. The chimeras were bred to C57BL/6 females, and several PKIβ heterozygous offspring were obtained. Heterozygotes were interbred to produce homozygous null mice. Mice were genotyped by genomic Southern blot analysis of DNA isolated from tail biopsies. All experiments comparing wild-type and mutant mice used age- and sex-matched animals on the C57BL/6 × 129Ola hybrid background. Animal care and experimentation complied with all guidelines set forth by the National Institutes of Health, the University of Washington, and the Veterans Affairs Puget Sound Health Care System.

Northern blot analysis.

Total RNA was isolated from testis, and Northern blots were run with 10 μg of RNA per lane as described previously (4). Blots were stained with methylene blue to visualize the 28S and 18S RNAs and also probed with the housekeeping gene encoding cyclophilin to confirm that all lanes had equivalent amounts of intact RNA. Riboprobes were made from cDNA clones of PKIγ and PKIβ (gifts from M. Uhler), protamine 1 (gift from R. Braun), and testis ACE, RT7, and TP1 (Incyte Genomics) using standard techniques.

Sperm counts and motility.

The sperm counts and motility of cauda epididymal sperm from male mice, 12 to 15 weeks of age, were assessed in a medium containing 135 mM NaCl, 5 mM KCl, 1 mM MgSO₄, 2 mM CaCl₂, 10 mM lactic acid, 1 mM sodium pyruvate, 30 mM HEPES, pH 7.4, and 20 mg of bovine serum albumin (fraction V; Sigma)/ml. A Neubauer chamber was used to determine sperm number and motility.

PKI activity assays.

Whole testis homogenates were prepared, and PKI activity was assayed essentially as described previously (9). The homogenates were heated for 10 min at 95°C to inactivate endogenous kinases and centrifuged for 10 min at 12,000 × g at 4°C. Equal amounts of heat-inactivated supernatant proteins from three individual mice were pooled, and then 47 μg was added to a kinase assay mixture containing 1 nM purified bovine heart C subunit (Sigma) in 50 μl. Each sample was assayed in triplicate, and the results are reported as percent inhibition of C subunit activity.

RESULTS

Generation of mice lacking a functional PKIβ gene.

A gene targeting approach was used to inactivate the PKIβ gene in embryonic stem cells. Figure 1A depicts the targeting vector, along with the resulting mutant PKIβ locus. Exon 3 of PKIβ, which encodes both the inhibitory and nuclear export domains, was deleted and replaced by the neomycin resistance cassette. The mutant ES cells were used to generate heterozygous mutant mice, which were then bred to produce knockout mice. Genotyping of the litters from heterozygous crosses (shown in Fig. 1B) demonstrated the presence of wild-type, knockout, and heterozygous offspring at the expected Mendelian ratio, indicating no embryonic lethality associated with the mutation. Northern blot analysis confirmed the loss of a functional PKIβ gene; Fig. 1C demonstrates the absence of the PKIβ transcript in homozygous testis, the tissue with the highest amount of PKIβ mRNA in wild-type mice. Knockout mice appeared morphologically normal, with no obvious detrimental effects on health or life span. A 12-week body weight study revealed normal weight gain for the PKIβ knockout mice (data not shown).

FIG. 1.

Targeted disruption of PKIβ in mice. (A) Targeting strategy at the PKIβ locus. (Top) Restriction map of the genomic region encompassing exon 3 (ex3) of wild-type PKIβ. (Middle) Targeting vector, in which exon 3 is replaced by the neomycin resistance cassette (NeoR). (Bottom) Mutant locus resulting from homologous recombination. (B) Genomic Southern blot of offspring from a heterozygote cross. A restriction digest with NcoI yields a 5.2-kb wild-type and 3.5-kb recombinant band when hybridized with a 0.5-kb AflII-EcoRI probe from the 3′ wild-type genomic region. Wild-type (+/+), homozygous null (−/−), and heterozygous (+/−) genotypes are indicated. (C) Northern blot of testis total RNA from two knockout and two wild-type mice. Note the absence of the PKIβ transcript in the knockouts.

PKI activity in PKIβ knockout and PKIα/β double-knockout mice.

The PKIβ knockout mice were interbred with PKIα knockout mice that we reported on previously (9) to generate double knockouts. These mice exhibited no noticeable ill effects of having inactivating mutations in both genes and exhibited normal weight gain (data not shown). PKI activity was assayed in both lines of mice. Homogenates were made from testis, where PKIβ is expressed at high levels in germ cells and PKIα is found in Sertoli cells (18). Homogenates were heated to inactivate endogenous kinases and then tested for PKA inhibitory activity. As shown in Fig. 2, wild-type mouse homogenates inhibited nearly 70% of the PKA in the assay, while the PKIβ and PKIα/β double-knockout mouse homogenates each inhibited about 47% of the PKA. Thus, inactivation of the PKIβ gene led to a significant loss of PKI activity in testis, confirming the functional loss of the PKIβ gene. However, the additional loss of PKIα did not lead to a detectable further decrement in PKI activity, probably due to the small amount of PKIα present in the wild-type testis. This result is consistent with our previous studies of PKIα knockout mice, which demonstrated no change in PKI activity in testis (data not shown). The substantial amount of residual inhibitory activity in the PKIβ and α/β double-mutant testes is presumably contributed by the only other known isoform of PKI, PKIγ. Northern blot analysis revealed no compensatory change in PKIγ mRNA levels in the knockout mice (data not shown), suggesting that either the normal endogenous level of PKIγ is sufficient for functional compensation or that compensatory increases in PKIγ occur at the protein level.

FIG. 2.

Loss of PKI activity in PKIβ knockout and PKIα/β double-knockout testes. Heat-inactivated extracts of testis from mice of the specified genotype were added to an assay measuring phosphorylation of the PKA substrate, Kemptide, by exogenous PKA (C subunit). PKI activity is expressed as % kinase inhibition. Extracts from PKIβ knockouts (αWT/βKO) and the double knockouts (αKO/βKO) show similar, significant reductions in PKI activity compared with wild-type (WT/WT) mice. Error bars represent standard errors of the means.

Reproductive function.

The presence of PKI isoforms in distinct cell types within the testis suggests an important role in sperm development and/or function. Several parameters of testicular function were examined in PKIβ knockout and PKIα/β double-knockout mice. As displayed in Table 1, testis weights were indistinguishable among wild-type mice and the two mutant mouse lines. To assess litter size, males and females of the same genotype were interbred, and the resulting litters showed no significant differences in size from wild-type, PKIβ knockout, or PKIα/β double-knockout breeders. In addition, histological analysis of testes from PKIβ single and PKIα/β double knockouts uncovered no abnormalities (data not shown). Thus, apparently normal reproductive function can occur in both male and female mice in the absence of both PKIα and PKIβ. We previously reported similar findings for PKIα single-knockout mice (9). Further sperm analysis was performed on the PKIβ knockout mice, revealing no change in sperm number or percent motility compared with wild-type mice (Table 1).

TABLE 1.

Reproductive function of PKI knockout mice

Genotype	Testes weight, mg (n)a	Litter size (n)b	Sperm count (n)c	Motile sperm, %d(n)
PKIα+/+β+/+	228 ± 6 (5)	8.0 ± 0.8 (15)	6.1 ± 0.5 (6)	38 ± 2.5 (6)
PKIα+/+β−/−	227 ± 5 (7)	7.9 ± 0.8 (15)	6.8 ± 0.3 (5)	34 ± 1.5 (5)
PKIα−/−β−/−	228 ± 10 (4)	6.9 ± 0.9 (11)	NDe	ND

^aMean ± standard error of the mean (SEM) of testis pairs from n number of mice. 

^bMean ± SEM from seven breeding pairs of the specified mutant or three breeding pairs of wild-type mice. 

^cMean sperm count per cauda ± SEM from n number of mice, multiplied by 10⁻⁶. 

^dMean of the percentage of sperm that are motile ± SEM. 

^eND, not done. 

Testicular gene expression.

Because PKI has a nuclear export signal that serves to chaperone the active catalytic subunit of PKA out of the nucleus, it is believed that PKI might function to terminate PKA phosphorylation of nuclear transcription factors, thereby shutting down transcriptional regulation of specific genes by PKA. Our previous studies of PKIα knockout mice revealed an unexpected decrease in expression of genes normally induced by PKA in skeletal muscle (9), the tissue with the highest PKIα level in wild-type mice. These results made it difficult to predict the outcome in PKIβ mutant mice. Figure 3 depicts gene expression analysis in PKIβ knockout testis. Four germ cell genes that are known to be regulated by PKA (or potentially regulated by virtue of cAMP response elements) were examined by Northern blot analysis in four PKIβ knockout and four wild-type mice. No consistent differences between genotypes were evident for any of the genes, protamine 1, encoding the testis isoform of angiotensin-converting enzyme, RT7, or transition protein 1. Together, these data demonstrate that PKIβ is not essential for spermatogenesis and fertility, and its loss in germ cells of the testis has no detectable effect on gene expression in those cells.

FIG. 3.

Gene expression in testis. Northern blots were performed with total RNA from four PKIβ knockout (KO) and four wild-type (WT) mice. Blots were probed for protamine 1 (Prm1), testis angiotensin-converting enzyme (ACE), RT7, and transition protein 1 (TP1).

DISCUSSION

Decades of research support the idea that the cAMP-PKA pathway is essential for normal testis development and function. The lack of reproductive phenotypes in most of the PKA isoform knockout mice has been attributed to compensation by other PKA isoforms (1, 4–6, 13). It appears that a similar phenomenon occurs within the PKI system. Despite substantial amounts of PKIβ in testis, our results demonstrate that it is clearly dispensable for reproductive function. The PKIβ knockout mice exhibit normal fertility and normal sperm production and motility (Table 1). Furthermore, expression of putative PKA-regulated germ cell genes is normal (Fig. 3), suggesting that there is no perturbation of PKA activity and its regulation.

The PKIβ knockouts and PKIα/β double knockouts both show a significant loss of PKI activity in testis compared with wild-type mice (Fig. 2). The remaining PKI activity is almost certainly contributed by PKIγ, although it is formally possible that another undiscovered PKI is expressed in testis. In wild-type mice, PKIγ and PKIβ mRNA levels in testis are approximately equal (7). If PKI activity levels roughly correspond to mRNA levels, then PKIβ and PKIγ each contribute about half of the PKI activity normally found in testis. Therefore, in the PKIβ knockout, one might have predicted a 50% loss of PKI activity, but only a 30% reduction was observed. It is possible that testis PKIγ protein levels had increased to compensate for the loss of PKIβ. We ascertained that there was no change in testis PKIγ mRNA, so any compensation would be at the protein level. Unfortunately, antibodies are not available to directly assess PKIγ protein levels by Western blot. Whether or not PKIγ levels increased in response to PKIβ loss, PKIγ clearly is able to functionally compensate, consistent with a built-in redundancy in the PKI system. PKIβ is known to be expressed within the germ cell compartment of the testis, while the cellular localization of PKIγ has not been reported. However, the functional compensation of PKIγ for PKIβ suggests germ cell expression.

The small amount of PKIα present in Sertoli cells does not contribute significantly to the overall PKI activity in testis, since PKIα knockout mice show no decrement in testis PKI activity (data not shown). Furthermore, we found no difference in testis PKI activity in PKIβ knockout mice compared with PKIα/β double-knockout mice (Fig. 2). That the PKIα knockout mice display no reproductive deficiency indicates that PKIα is not needed for normal follicle-stimulating hormone signaling in Sertoli cells. Whether PKIγ is expressed in Sertoli cells is unknown. It is possible that PKIγ is expressed throughout the testis and is able to compensate for the missing PKIα as well as PKIβ. A mouse knockout of PKIγ is needed in order to clarify the role of PKI in reproduction and in other physiological processes as well.