Abnormal development of the olfactory bulb and reproductive system in mice lacking prokineticin receptor PKR2

Abstract

Prokineticins, multifunctional secreted proteins, activate two endogenous G protein-coupled receptors PKR1 and PKR2. From in situ analysis of the mouse brain, we discovered that PKR2 is predominantly expressed in the olfactory bulb (OB). To examine the role of PKR2 in the OB, we created PKR1- and PKR2-gene-disrupted mice (Pkr1^−/− and Pkr2^−/−, respectively). Phenotypic analysis indicated that not Pkr1^−/−but Pkr2^−/−mice exhibited hypoplasia of the OB. This abnormality was observed in the early developmental stages of fetal OB in the Pkr2^−/− mice. In addition, the Pkr2^−/− mice showed severe atrophy of the reproductive system, including the testis, ovary, uterus, vagina, and mammary gland. In the Pkr2^−/− mice, the plasma levels of testosterone and follicle-stimulating hormone were decreased, and the mRNA transcription levels of gonadotropin-releasing hormone in the hypothalamus and luteinizing hormone and follicle-stimulating hormone in the pituitary were also significantly reduced. Immunohistochemical analysis revealed that gonadotropin-releasing hormone neurons were absent in the hypothalamus in the Pkr2^−/− mice. The phenotype of the Pkr2^−/− mice showed similarity to the clinical features of Kallmann syndrome, a human disease characterized by association of hypogonadotropic hypogonadism and anosmia. Our current findings demonstrated that physiological activation of PKR2 is essential for normal development of the OB and sexual maturation.

Prokineticins (PKs), comprising PK1 (also called EG-VEGF) and PK2 (also called Bv8), are secreted bioactive proteins possessing 10 conserved cysteines that form five disulfide bonds (1). The mature forms of PK1 and PK2 consist of 86 and 81 amino acids, respectively, and share ≈40% amino acid identity. The N-terminal six-amino acid sequence (AVITGA) of mature PKs is completely conserved during molecular evolution and is essential for the bioactivities of PKs (2). Two endogenous PK receptors (PKRs), termed PKR1 and PKR2, both of which are G protein-coupled receptors, mediate signal transduction of PKs (3–5). PKR1 and PKR2 show strong similarity (87% homology) in their primary structure. In humans, both PKRs are predominantly expressed in the testis. In addition, PKR1 shows preferential distribution in the peripheral tissues, whereas PKR2 shows relatively localized distribution in the CNS. Analysis using receptor-transfected mammalian cell lines showed that PK2 binds PKRs with higher affinity than does PK1, suggesting that PK2 is the stronger agonist for the PK/PKR system under physiological conditions (3–5). As a result of binding with PKs, PKRs, which can couple to Gq protein, promote intracellular Ca²⁺ mobilization (3–5). Moreover, PKRs also couple to Gqi and Gs proteins, indicating that PKRs activate multiple intracellular signal-transduction pathways (6, 7). Activation of PKRs influences several physiological events in the CNS and peripheral tissues (8), including intestinal contraction (9), hyperalgesia (10, 11), spermatogenesis (12), neuronal survival (13), circadian rhythm (14, 15), angiogenesis (16–18), ingestive behavior (19), and hematopoiesis (20). Recently, Ng et al (21) reported the generation of knockout mice lacking Pk2 (Pk2^−/− mice). Their Pk2^−/− mice showed marked reduction in the olfactory bulb (OB) size, loss of normal OB architecture, and accumulation of neuronal progenitors in the rostral migratory stream. Based on the abnormal phenotype observed in Pk2^−/− mice, they demonstrated that PK2 plays critical roles in OB morphogenesis and OB neurogenesis. To date, however, which PKR activation is essential for normal OB development remains unknown.

In this study, we have generated two murine lines that are gene-disrupted for Pkr1 and Pkr2, respectively. Phenotypic analysis of these two lines demonstrated that the Pkr2^−/− mice showed abnormal development of the OB and reproductive system, whereas Pkr1^−/− mice did not.

Results

In Situ Hybridization Analysis for PKR2 Using Radiolabeled Probes.

Based on our previous finding that PKR2 has strong expression in the human fetal brain, we analyzed its mRNA expression in the fetal mouse head region, upper cephalic portion of the embryo, or whole brain from embryonic day (E)9.5 to postnatal day (P)10 (see Fig. 5A, which is published as supporting information on the PNAS web site). Quantitative PCR study indicated that PKR2 was expressed strongly in both tissues during the embryonic stage from E9.5 to E18.5. Next, to study the detailed expression pattern of PKR2 in the mouse brain, we performed in situ hybridization analysis for PKR2 using radiolabeled probes of mouse Pkr2 cDNA. In accordance with the previous findings of Cheng et al. (14), PKR2 was strongly expressed in the suprachiasmatic nucleus and moderately in the paraventricular thalamic nucleus (Fig. 5B). In addition, PKR2 was predominantly expressed in the OB, especially in the ependyma and subependymal layer of the olfactory ventricle (Fig. 5 C and D).

Generation of Pkr1^−/− and Pkr2^−/− Mice.

To study the roles of PKRs in the OB, we generated gene-disrupted mice for Pkr1 (see Fig. 6 A–C, which is published as supporting information on the PNAS web site) and Pkr2 (Fig. 6 D–F), respectively. For designing the targeting vectors for Pkr1 and Pkr2, an exon containing the initiating ATG codon corresponding to the first Met of each ORF was deleted, and a reverse-oriented Neo-cassette was inserted, respectively (Fig. 6 A–D). Generation of heterozygous and homozygous mice for Pkr1 (Fig. 6B) and Pkr2 (Fig. 6E) was confirmed by Southern blot analysis. Pkr1^−/− mice were obtained at the expected Mendelian rate, whereas >50% of Pkr2^−/−mice died at an early neonatal stage from respective heterozygous mouse mating. By performing quantitative PCR analysis, we also confirmed that the mRNA expression levels of PKR1 (Fig. 6C) and PKR2 (Fig. 6F) were decreased by ≈50% and 100% in the heterozygous and homozygous mice, respectively.

Malformation of the OB in Pkr2^−/− Mice.

Anatomical analysis of the brain demonstrated that all the Pkr2^−/− mice (n = 13) displayed a decreased OB size compared with the control wild-type littermates (n = 10) (Fig. 1A Right). No significant abnormality of the OB size was observed in Pkr2^+/− heterozygous mice (data not shown). There was a wide-open space between the right and left OB in Pkr2^−/− mice. In contrast, no remarkable morphological differences in the OB were observed between Pkr1^−/− and wild-type mice (n = 4 each) (Fig. 1A Left). Histopathologically, the typical layered structure of the OB was absent and malformed in Pkr2^−/− mice (Fig. 1B). The glomerular layer was indiscernible in the OB of Pkr2^−/− mice. We next examined the morphogenesis of the developmental OB. During the embryonic stage at E14.5 (Fig. 1C), which is the developmental stage after establishment of the first olfactory sensory connections from the olfactory epithelium to the OB (22), no remarkable macroscopic differences could be observed between the Pkr2^−/− mice and their wild-type littermates (n = 19). However, at E16.5 and E18.5 as well as at P0, projection of the OB from the telencephalon was evident in all of the wild-type mice, whereas this projection was very slight in all of the Pkr2^−/− mice (n = 5, 7, and 5 at E16.5, E18.5, and P0, respectively).

Fig. 1.

Macro- and microscopic analyses of OB from Pkr2^−/− mice. (A) Macroscopic view of the male brain at 8 weeks of age from the control wild-type littermates and Pkr1^−/− mice (Left), and control wild-type littermates and Pkr2^−/− mice (Right). The OB size is small in the Pkr2^−/− mice. (Scale bars, 5 mm.) (B) Nissl-stained sagittal sections of the male OB at 8 weeks of age from the wild-type littermates (Upper) and Pkr2^−/− mice (Lower). Higher magnifications of the OB (boxes, Left) from each genotypic mouse are shown (Right), respectively. The arrow (Upper Right) indicates the glomerular layer (GL) of the OB from wild-type littermates. Note that no glomerular layer is discernible in the OB from the Pkr2^−/− mice (Lower Right). [Scale bars, 0.5 mm (Left) and 0.1 mm (Right).] (C) OB morphogenesis during the embryonic stage at E14.5 (Upper Left), E16.5 (Upper Right), E18.5 (Lower Left), and P0 (Lower Right). (Scale bars, 1 mm.)

Abnormal Development of the Reproductive System in Pkr2^−/− Mice.

Anatomical analysis of the non-CNS tissues of Pkr1^−/− and Pkr2^−/− mice indicated that both the male and female reproductive organ weights were reduced in Pkr2^−/− mice, but not in Pkr1^−/− mice (mean weight of testis, 114.9 ± 11.8 mg in wild-type mice and 2.8 ± 0.5 mg in mutant mice, P < 0.001; mean weight of ovary, 3.4 ± 0.7 mg in wild-type mice and <0.1 mg in mutant mice, P < 0.001) (Fig. 2A; and see Table 1, which is published as supporting information on the PNAS web site). However, no significant differences were observed in the reproductive systems of the male and female Pkr2^+/− heterozygous mice compared with the wild-type mice (data not shown). Histopathological analysis revealed overt atrophy of the testis of all Pkr2^−/− mice examined (n = 5) (Fig. 2B). The seminiferous tubules were reduced in diameter. No sperm were present in the lumen of the tubules in the mutant testis. Spermatogenic cells and spermatocytes were present, but spermatids were not observed. The interstitium was sparse, and the few Leydig cells observed were small. These findings suggest that spermatogenesis was arrested in the pachytene spermatocyte stage. Apparent atrophy was also observed in the ovary, uterus, vagina, and mammary gland of all female Pkr2^−/− mice (n = 5). The ovary contained mainly undeveloped follicles, and growth seemed to be arrested in the preantral phase of follicle development (Fig. 2C). Only a few follicles showed antral formation, but it was incomplete. No corpora lutea were seen, and the interstitium was severely atrophic. These findings suggest abnormalities in the maintenance and growth of the follicles. A high level of atrophy was also observed in the endometrium, the muscle layer of the uterus, and in the epithelium of the vagina.

Fig. 2.

Hypoplasia of the reproductive organs in Pkr2^−/− mice. (A) Macroscopic view of the testes at 8 weeks of age from the control wild-type littermates and Pkr1^−/− mice (Left), control wild-type littermates and Pkr2^−/− mice (Center), and the ovary and uterus (Right) at 12 weeks of age from the wild-type, Pkr1^−/−, and Pkr2^−/− mice. The male and female reproductive organs are small in the Pkr2^−/− mice. (Scale bars, 5 mm.) (B) HE-stained sections of wild-type and Pkr2^−/− testis at 20 weeks of age. Higher magnification (rectangle) shows normal Leydig cells in wild-type testis (Lower Left), whereas the mutant testis shows small Leydig cells and no spermatid (Lower Right). SPG, spermatogenic cell; SPC, spermatocytes; SPT, spermatid; LC, Leydig cells. [Scale bars, 100 μm (Upper) and 10 μm (Lower).] (C) HE-stained sections of wild-type and Pkr2^−/− ovary at 20 weeks of age. A magnified region (rectangle) shows normal antral follicle (arrow) in wild-type ovary (Lower Left). A magnified region (rectangle) shows undeveloped follicles (arrows) in Pkr2^−/− ovary (Lower Right). [Scale bars, 200 μm (Upper) and 50 μm (Lower).]

Analysis of plasma hormone concentrations revealed that testosterone (see Fig. 7A, which is published as supporting information on the PNAS web site) and follicle-stimulating hormone (FSH) (Fig. 7B) were lower in male Pkr2^−/− mice than in the age-matched wild-type mice (P < 0.001 for testosterone compared in male mice, and P < 0.001 for FSH compared in male mice and P = 0.078 in female mice, respectively). On the other hand, no statistically significant difference was found for the plasma luteinizing hormone (LH) concentration (Fig. 7C). In addition to the plasma levels of those sex hormones, the mRNA transcriptional levels for pituitary FSH (P = 0.004 and P = 0.01 for male and female mice, respectively) (Fig. 7D) and LH (P < 0.001 and P = 0.009 for male and female mice, respectively) (Fig. 7E) were lower in the Pkr2^−/− mice.

Disappearance of Gonadotropin-Releasing Hormone (GnRH) Neurons in Pkr2^−/− Mice.

To detect GnRH neurons in the brain, we performed GnRH immunohistochemistry on both the Pkr2^−/− mice and their control wild-type littermates. In contrast to the finding that GnRH-immunoreactive neurons were present in the preoptic region and median eminence of the hypothalamus of wild-type mice [n = 4 (male, 2; female, 2)] (Fig. 3A Left and B), no such reactivity was found in the corresponding regions of the Pkr2^−/− mice [n = 4 (male, 2, female, 2)] (Fig. 3A Right). GnRH-immunoreactive neurons were also observed in the hypothalamus of the Pkr2^+/− heterozygous mice (data not shown). In parallel with this observation, the mRNA transcriptional level for hypothalamic GnRH was significantly decreased in the Pkr2^−/− mice (P < 0.001 and P = 0.001 for males and females, respectively) (Fig. 3C).

Fig. 3.

Analysis of GnRH neurons in Pkr2^−/− mice. (A) Immunohistochemical study of GnRH neurons in the wild-type (Left) and Pkr2^−/− mice (Right). Coronal sections of the preoptic region (Top and Middle) and median eminence (Bottom). (Scale bars, 0.1 mm.) (B) Higher magnification of the preoptic region [box (A Middle Left)] from a wild-type mouse. The arrow indicates GnRH-positive neuronal cell bodies. (Scale bar, 0.2 mm.) (C) Mean ± SD of transcriptional mRNA level of GnRH in the hypothalamus of Pkr2^−/− mice. Quantitative PCR was performed to compare the mRNA level between the wild-type littermates (male, n = 6; female, n = 6) and Pkr2^−/− mice (male, n = 5; female, n = 6) at 20 weeks of age. Data shown are representative of two experiments.

To study the mechanism of abnormal GnRH neuron and OB development observed in the Pkr2^−/− mice, histopathological analysis was performed in the upper nasal region in the mutant and wild-type fetal mice at E12.5 and E13.5. In the Pkr2^−/− mice at E12.5 (Fig. 4B, F, and J) and E13.5 (Fig. 4 D, H, and L), the olfactory/vomeronasal axons in mutant mice failed to reach the forebrain, and the terminals of the axons formed a large tangled sphere-shaped structure. In contrast, the wild-type littermates at E12.5 (Fig. 4 A, E, and I), E13.5 (Fig. 4 C, G, and K) showed no such tangled fibers, and the olfactory/vomeronasal axons reached the forebrain.

Fig. 4.

Abnormal morphology in the upper nasal region in Pkr2^−/− mice at E12.5 and E13.5. HE-stained parasagittal sections of the upper nasal region in wild-type littermates at E12.5 (A, E, and I), Pkr2^−/− mice at E12.5 (B, F, and J), wild-type littermates at E13.5 (C, G, and K) and Pkr2^−/− mice at E13.5 (D, H, and L). Note that a sphere-shaped structure was observed at the region between the olfactory pit and forebrain in Pkr2^−/− mice at both E12.5 and E13.5 (arrowhead in B and D). This structure is particularly evident when viewed at higher magnification (arrowhead in F, H, J, and L). OP, olfactory pit, F, forebrain. [Scale bars, 200 μm (A–D), 100 μm (E–H), and 50 μm (I–L).]

Discussion

Here, we report that Pkr2^−/− mice exhibited hypoplasia in both the OB and the reproductive system, whereas Pkr1^−/− mice did not.

In situ hybridization analysis showed that PKR2 was predominantly expressed in the ependyma and subependymal cell layer of the adult OB. It is well known that development of the OB requires generation and differentiation of several lines of cells in the early developmental stage (22). Together with our current finding that PKR2 was strongly expressed in the mouse fetal brain as well as in the human fetal brain (5), we deduce that the PK/PKR2 system plays important roles in OB morphogenesis. The discovery of OB malformation, which originated in the embryonic stage, only in the Pkr2^−/− mice suggests that not PKR1 but PKR2 activation is essential to drive the normal development of the OB during embryogenesis.

In addition to the abnormal development of the OB, the Pkr2^−/− mice exhibited hypoplasia of the reproductive system. Our assays of the plasma hormones revealed that the testosterone and FSH levels were considerably decreased in the Pkr2^−/− mice. We also found that the mRNA transcription levels of FSH and LH were significantly lower in the Pkr2^−/− mice in the pituitary, which is the source of circulating FSH and LH. In addition to these circulating hormones, it is well known that sexual maturation is tightly regulated by hypothalamic GnRH (23). Moreover, the pathological findings for the testis and ovary of the Pkr2^−/− mice were highly consistent with those seen in GnRH-deficient mice (24). Accordingly, the defective sexual development in the Pkr2^−/− mice is considered to be the result of possible defects in GnRH secretion. In the present comparison of the plasma FSH level, a statistically significant difference was not found between the female Pkr2^−/− mice and their control wild-type littermates. We surmise that, because, in rodents, the plasma FSH level is generally lower in females than in males (25, 26), individual variability in the plasma FSH level masked the true difference between the mutant and wild-type mice. A similar explanation could be applied to our result that the plasma LH level showed no statistically significant difference (25). In view of the fact that PKR2 is also expressed in the gonads (4, 5, 18), we cannot exclude the possibility that the hypoplasia of the reproductive systems arose from the loss of a direct role of PKR2 in the development of these tissues. Further investigation is needed to confirm such an additional activity for PKR2 on gonadal development.

In contrast to the report of Ng et al. (21), indicating that ≈50% of Pk2^−/− mice displayed asymmetric OB formation, all of our Pkr2^−/− mice exhibited symmetrical OB malformation. Moreover, in their article, there was no description of any abnormalities of the reproductive system in their Pk2^−/− mice. It may be argued that PK1, another PKR2 ligand, compensated for the absence of PK2 bioactivity and played roles in OB morphogenesis and pubertal maturation in the Pk2^−/− mice. It is not clear, at this moment, what is the main cause of the frequent neonatal lethality observed in Pkr2^−/− mice. One possibility is, as described previously in regard to gene-disrupted mice for FGF receptor 1 (27) and lysophosphatidic acid receptor (28), that dysfunction of the olfactory system of Pkr2^−/− mice might lead to inability to suckle milk because of smelling disability. Additional studies would be required to elucidate this issue directly.

The pathology observed in Pkr2^−/− mice bears a striking resemblance to the clinical manifestations of Kallmann syndrome (KS), which is a human developmental disease with combined features of hypogonadotrophic hypogonadism and anosmia (29–32). KS is categorized into three types, KAL1, KAL2, and KAL3, which are an X-linked form (Online Mendelian Inheritance in Man (OMIM) entry no. 308700), an autosomal dominant form (OMIM entry no. 147950), and an autosomal recessive form (OMIM entry no. 244200), respectively (29). The gene responsible for KAL1 encodes an extracellular matrix protein, anosmin-1 (33, 34), whereas KAL2 is caused by loss-of-function mutations in the gene encoding FGF receptor 1 (35). On the other hand, the gene responsible for KAL3 remains unknown. In KS patients, OB development is completely or partially absent, and failed puberty is often the first manifestation of the disease in both sexes (31). Based on histopathological examinations, the pathogenesis of KS has been proposed to be that malformation of the OB causes abnormal localization of GnRH neurons, leading to the hypogonadotropic hypogonadism through loss of GnRH activity in the hypothalamus (29–32, 36, 37). The remarkable phenotypic similarities to KS, i.e., simultaneous hypoplasia in the OB and the reproductive system, observed in all of our Pkr2^−/− mice, inspired us to hypothesize that the hypothalamic GnRH neurons have a developmental abnormality in these mice. Our immunohistochemical study confirmed this hypothesis that GnRH neurons are, indeed, absent in the hypothalamus of Pkr2^−/− mice. The absence of GnRH neurons in the hypothalamus could be attributed to impaired migration of the GnRH neurons from the olfactory pit to the brain. We observed that the olfactory/vomeronasal axon runs into a sphere-shaped structure in the upper nasal region in Pkr2^−/− fetal mice at E12 and E13 and failed to reach the OB. The structure seemed like a fibrocellular mass (FCM) that has been observed in mutant extratoes mice (38) and Arx homeobox-gene-deficient mice (39). In these mice, most of the migrating olfactory axons failed to reach the OB and terminated in an axon-tangled FCM to cause failure of OB morphogenesis. It is well documented that the prior establishment of a migrating pathway by the olfactory/vomeronasal axons is essential for developing GnRH neurons to move properly from the olfactory pit to the rostral forebrain (31, 40). Therefore, in Pkr2^−/− mice, it is possible that the defect in the olfactory/vomeronasal axonal route in the nasal region blocked GnRH neurons from migrating into the forebrain. In line with this consideration, it is tempting to speculate that olfactory neurons express PKR2 and that their axons would be guided by the PK1/PK2 to establish the axonal route during embryogenesis.

The current results suggested that, in Pkr2^−/− mice, the failure of development of GnRH neurons took place in conjunction with developmental agenesis of the OB during the embryonic period and eventually led to the defects in sexual maturation. In the genetically engineered mice, however, the OB malformation does not necessarily cause abnormal GnRH localization. No pathological features of a hypoplastic reproductive system were observed in all hitherto-reported mouse lines possessing a pathological OB (31). Our Pkr2^−/− mouse is a genetically engineered murine line that shows hypoplasia in both the OB and the reproductive system.

Recent reports demonstrated that knockout mice lacking Gpr54 (Gpr54^−/− mice) showed a striking phenotype of isolated hypothalamic hypogonadism (25, 41). Similar to PKRs, GPR54 is also a member of the G protein-coupled receptor family. GPR54 activation leads to direct release of GnRH from hypothalamic GnRH neurons and is essential for mammalian puberty, including in man (26, 42–44). In Gpr54^−/− mice, which possess intact hypothalamic GnRH neurons (45), decreased pituitary LH/FSH secretion causes the characteristic hypogonadism due to the loss of GPR54 activation (25, 43). In view of the fact that neither agenesis of the OB nor abnormality of the hypothalamic GnRH neurons was observed in the Gpr54^−/− mice, we surmise that the physiological roles of GPR54 are different from those of PKR2.

It remains unclear whether PKR2 activation regulates OB development directly or indirectly. Based on the results of Ng et al. (21), demonstrating that PK2 promoted postnatal and adult OB neurogenesis, the current result infers that PKR2 activation is also necessary for embryonic OB morphogenesis, including the development of OB projection neurons. Moreover, the fact that PK2 has binding affinity for heparan sulfate proteoglycans (HSPG) (18) has something in common with the character of anosmin-1 (KAL1 product), and FGFs (endogenous ligands for the KAL2 product FGF receptor 1 (FGFR1) also interact with HSPG (30). It might be possible that PK–PKR complexes might influence the physiological roles of these KS-responsible products, including that there might be some interactions between FGFR1 and PKR2 signaling pathways.

In conclusion, we report, in this study, that activation of PKR2 and not PKR1 is critical for OB morphogenesis and localization of GnRH neurons, which is vital for maturation of the reproductive organs. It is also noteworthy that this is an animal model for KS that manifests simultaneous hypoplasia for both the OB and reproductive organs. The Pkr2^−/− mouse can be a useful model for studying this human disorder.

Materials and Methods

In Situ Hybridization Using Radiolabeled Probes.

Radiolabeled probes for 670 bp of mouse Pkr2 ORF-cDNA were made by using [³⁵S]UTP (PerkinElmer) with a standard protocol for cRNA synthesis. Mice were deeply anesthetized with ether and intracardially perfused with 10 ml of saline and 20 ml of a fixative containing 4% paraformaldehyde in 0.1 M phosphate buffer (PB), pH 7.4. Brains were postfixed in the same fixative for 24 h at 4°C, soaked in 0.1 M PB, pH 7.4, containing 20% sucrose for 48 h, and, finally, stored frozen at −70°C. Stored tissues were cut to a thickness of 40 μm on the coronal plane by using a cryostat. The in situ hybridization of mRNA was performed as described in detail in refs. 46 and 47.

Gene Targeting of Pkr1 or Pkr2.

To construct a Pkr1 targeting vector, the short arm of the 2.1-kb region, which includes exon 1, containing initiating methionine and an intron, and the long arm of the 7.8-kb region, which includes an intron, exon 2, and 3′ UTR, of the mouse Pkr1 gene, and, for the Pkr2 targeting vector, the long arm of the 6.6-kb region, which is 5′ UTR, and the short arm of the 2.0-kb region, which includes an intron of the Pkr2 gene, were ligated into the targeting vector pPNT (48). The targeting vector was transfected into TT2 cells (49) by electroporation, and clones resistant to G418 and ganciclovir were selected and further screened by Southern hybridization. Chimeric male mice were mated with C57BL/6 J females (CLEA Japan, Tokyo) to obtain F1 heterozygotes. Genotyping using the tails for detecting the genes of Pkr1, Pkr2, and neomycin was performed by PCR with the primer sets of (5′-ATGGAGACCACTGTCGGGGCTCTGGGTG-3′ and 5′-CCTGTCAATGGCAATGGCCAGTAGGGCG-3′); (5′-CATGGGACCCCAGAACAGAAACACTAGC-3′ and 5′-CCTGTCAATAGCGATGGCCAGCAGAGCG-3′); and (5′-TATGGGATCGGCCATTGAAC-3′ and 5′-CCTCAGAAGAACTCGTCAAG-3′), respectively. F₁ heterozygous mice were crossed with C57BL/6 mice to produce a large number of F₂ heterozygous mice, which were then intercrossed to produce homozygous mice for analysis. In this study, we used mice of the F₂ generations, and, in all experiments, comparisons were made with littermate wild-type mice. All experiments were performed in compliance with the regulations of the Animal Ethics Committee of Astellas Pharma, Inc.

Quantitative PCR.

DNase-treated total RNA was isolated from the head of mice from E9.5 to E13.5, the whole brain of mice from E14.5 to P10 (male, n = 2; female, n = 2), and the hypothalamus and pituitary of mice at 20 weeks of age (wild-type littermates (male, n = 6; female, n = 6) and Pkr2^−/−mice (male, n = 5; female, n = 6). Tissue expression of PKR2 in the head or brain, GnRH in the hypothalamus, and FSH and LH in the pituitary were quantitatively analyzed by using a Prism 7700 Sequence Detector (Applied Biosystems) as described in ref. 5. To confirm the disappearance of the mRNA corresponding to the target genes, total RNA was isolated from the hypothalamus of male mice at 5 weeks of age (wild-type littermates, heterozygous and homozygous mice (n = 2 for both Pkr1- and Pkr2-mutant mice). The oligonucleotide primer sets used for PCR were as follows: 5′-AGCACTGGTCCTATGGGTTGC-3′ and 5′-AGTGTTCAGTGTTTCTCTTTCCCC-3′ for GnRH; 5′-GTAGCCACTGAATGTCACTGTGG-3′ and 5′-GCAGTCAGTGCTGTCGCTGT-3′ for FSH; 5′-CGGCTCAGTAGCTCTGACTGTG-3′ and 5′-ACAGGCCATTGGTTGAGTCC-3′ for LH; 5′-GCCCCTGGATGAAGAGGAAG-3′ and 5′-GCAGCAAAGAAAGTCCGAGAA-3′ for Pkr1; and 5′-ACCAACCTCCTCATTGCTAACC-3′ and 5′-GATCGCCACCAGGAAGTCAG-3′ for Pkr2. The relative abundance of transcripts was normalized to the constitutive expression of G3PDH mRNA.

Histopathological Examination.

The brain, testis, ovary, uterus, vagina, and mammary gland were dissected, fixed, and preserved in 10% neutral buffered formalin (the testis was fixed in Bouin’s solution). Sections of these tissues were stained with hematoxylin and eosin (HE) and observed microscopically. For the fetal brain preparation, the time of performance of in vitro fertilization and embryo transfer was considered as E0, and E14.5–E18.5 embryos were collected from pregnant mice by Cesarean section. For the analysis of the upper nasal region in fetal mice, embryos from the 12.5–13.5 days of gestation were collected from pregnant mice by Cesarean section. E12.5–13.5 embryos were immersion-fixed in 4% paraformaldehyde in 0.1M PB overnight at 4° C. Specimens were dehydrated through graded alcohols and xylene and embedded in paraffin. Serial sections (8-μm thick) were cut in the sagittal planes and were mounted on silane-coated slides. These sections were stained with HE and observed microscopically.

Hormone Assays.

Plasma testosterone, FSH, and LH were measured with commercially available kits: DPC Total Testosterone kit (Diagnostic Products), rat FSH [¹²⁵I] RIA System (Amersham Pharmacia Biosciences), and rat LH EIA System (Amersham Pharmacia Biosciences), respectively.

Immunohistochemistry.

Immunohistochemical analysis was performed as described in ref. 50. Briefly, adult mice were anesthetized with diethyl ether and intracardially perfused with PB containing 4% paraformaldehyde. Brains were removed, frozen in dry ice, and coronally sectioned with a cryostat at a thickness of 20 μm. The sections were incubated for 2 days at 4°C in primary anti-GnRH (Chemicon) diluted 1:60,000 in PBS containing 0.3% Triton X-100. The sections were sequentially incubated with biotinylated anti-rabbit goat IgG (Vector Laboratories) diluted 1:1,000 and avidin-conjugated horseradish peroxidase (Vector Laboratories) diluted 1:1,000. They were then treated with 0.035% diaminobenzidine/0.05 M Tris·HCl buffer (pH 7.4), dehydrated with a graded series of ethanol rinses, immersed in xylene, and embedded in Entellan (Merck).

Statistical Analysis.

Results are expressed as the mean ± SD. Differences between groups were examined for statistical significance by using Student’s two-tailed unpaired t test.

Abbreviations

En

embryonic day n

FSH

follicle-stimulating hormone

GnRH

gonadotropin-releasing hormone

HE

hematoxylin and eosin

KS

Kallmann syndrome

LH

luteinizing hormone

OB

olfactory bulb

Pn

postnatal day n

PK

prokineticin

PKR

PK receptor.