The kinase VRK1 is required for normal meiotic ... - Semantic Scholar

MECHANISMS OF DEVELOPMENT

1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

available at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/modo

The kinase VRK1 is required for normal meiotic progression in mammalian oogenesis Carolyn S. Schober a, Fulya Aydiner a b c

b,1

, Carmen J. Booth c, Emre Seli b, Valerie Reinke

a,*

Department of Genetics, Yale University School of Medicine, New Haven, CT, USA Department of Obstetrics, Gynecology and Reproductive Sciences, Yale University School of Medicine, New Haven, CT, USA Section of Comparative Medicine, Yale University School of Medicine, New Haven, CT, USA

A R T I C L E I N F O

A B S T R A C T

Article history:

The kinase VRK1 has been implicated in mitotic and meiotic progression in invertebrate

Received 2 December 2010

species, but whether it mediates these events during mammalian gametogenesis is not

Received in revised form

completely understood. Previous work has demonstrated a role for mammalian VRK1 in

19 January 2011

proliferation of male spermatogonia, yet whether VRK1 plays a role in meiotic progres-

Accepted 20 January 2011

sion, as seen in Drosophila, has not been determined. Here, we have established a mouse

Available online 26 January 2011

strain bearing a gene trap insertion in the VRK1 locus that disrupts Vrk1 expression. In addition to the male proliferation defects, we find that reduction of VRK1 activity causes

Keywords: VRK1 Oogenesis Meiosis

a delay in meiotic progression during oogenesis, results in the presence of lagging chromosomes during formation of the metaphase plate, and ultimately leads to the failure of oocytes to be fertilized. The activity of at least one phosphorylation substrate of VRK1, p53, is not required for these defects. These results are consistent with previously defined functions of VRK1 in meiotic progression in Drosophila oogenesis, and indicate a con-

p53

served role for VRK1 in coordinating proper chromosomal configuration in female

Mouse

meiosis. 2011 Elsevier Ireland Ltd. All rights reserved.

1.

Introduction

Gametogenesis is a fundamental biological process and a necessary precursor to sexual reproduction. In mammals, the production of oocytes in females or sperm in males involves a sexually dimorphic progression of germ cells through meiosis. However, errors that arise during this intricate process can result in infertility. Errors during female meiosis most often lead to aneuploidy, which is estimated to cause approximately one-third of human spontaneous abortions, or result in physical and/or mental disabilities if the conception results in a full term pregnancy (Hassold and Hunt, 2001). Defects that occur during the process of spermatogenesis in males can also

lead to infertility, although most diagnoses are often idiopathic in nature (Matzuk and Lamb, 2002, 2008). Given the sensitive nature of infertility, along with the ethical implications of working with human material, the study of mammalian gametogenesis has relied heavily upon rodent models. However, many of the genes that govern this process, particularly those controlling meiotic events, remain elusive (Handel and Schimenti, 2010). Consequently, gametogenesis requires continued investigation in order to further understand the properly coordinated production of oocytes and sperm and ultimately, infertility. Vaccinia related kinase-1 (VRK1) is a serine/threonine kinase that has recently been implicated in gametogenesis

* Corresponding author. E-mail address: [email protected] (V. Reinke). 1 Present address: Department of Obstetrics, Gynecology and Reproductive Sciences, Dartmouth Hitchcock Medical Center, Lebanon, NH, USA. 0925-4773/$ - see front matter 2011 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.mod.2011.01.004


in multiple species. Evidence from the study of the Drosophila homolog of VRK1, nucleosomal histone kinase-1 (NHK-1), has demonstrated that mutations in NHK-1/VRK1 lead to abnormal chromosome structure and improper spindle formation during female meiosis (Cullen et al., 2005; Ivanovska et al., 2005). Furthermore, a recent report has proposed that NHK1/VRK1 acts as a downstream effector of the meiotic checkpoint in Drosophila oocytes (Lancaster et al., 2010). Additional evidence implicating a role for VRK1 in germ cell development has also been demonstrated in the germline stem cells of Caenorhabditis elegans where a reduction in proliferation, but not an increase in apoptosis, causes the sterility defect observed in vrk-1 mutants (Gorjanacz et al., 2007; Waters et al., 2010). In mammals, VRK1 has been shown to affect germ cell proliferation in males, leading to sterility (Choi et al., 2010; Wiebe et al., 2010). Female mice lacking VRK1 function are reported to be sterile as well, but the basis for this phenotype is unknown. Multiple candidate phosphorylation substrates have been identified for VRK1 (reviewed in (Klerkx et al., 2009)). Mitotic chromatin modifications attributed to VRK1 include phosphorylation of histones H3 at Ser10 (Kang et al., 2007) and H2A at Thr119 (Aihara et al., 2004; Brittle et al., 2007). Phosphorylation of H2A at Thr119 is dependent upon NHK-1/ VRK1 during female meiosis in Drosophila (Ivanovska et al., 2005). The nuclear pore protein barrier-to-autointegration factor (BAF), which also functions during mitosis to regulate the association of chromatin with the nuclear envelope, is another phosphorylation substrate of VRK1 (Gorjanacz et al., 2007; Lancaster et al., 2007; Nichols et al., 2006). Finally, VRK1 has been shown to affect transcription factors involved in cell cycle progression by targeting CREB (Kang et al., 2008), ATF2 (Sevilla et al., 2004b), c-JUN (Sevilla et al., 2004a), and the tumor suppressor p53 (Lopez-Borges and Lazo, 2000). Most recently, the functional relationship between p53 and VRK1 was investigated in vivo within the germ line of C. elegans, where VRK1 acts to inhibit the cell cycle arrest activity of p53 (Waters et al., 2010). However, whether this inhibitory relationship between VRK1 and p53 is conserved within the mammalian germ line remains unknown. In this study, we investigated the role of VRK1 during mammalian gametogenesis by establishing mice carrying a gene trap insertion within the Vrk1 locus that disrupts Vrk1 expression. Our results show that a reduction in VRK1 causes sterility in both male and female mice. Male mice homozygous for the Vrk1 gene trap allele, while fertile at first, display a progressive loss of spermatogenesis that results in early onset sterility, similar to previous reports (Choi et al., 2010; Wiebe et al., 2010). Female homozygous mice are completely sterile and display multiple defects during oocyte development, including delays in meiotic progression and lagging chromosomes during metaphase II. We also evaluated the relationship between VRK1 and p53, and found that in contrast to the inhibitory relationship seen in C. elegans germ cells, VRK1 and p53 have additive effects in mammalian germ cells. In sum, our results demonstrate that VRK1 has both conserved and species-specific functions in gametogenesis.

1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

2.

Results

2.1.

Generation of Vrk1 mutant mice

179

In an effort to expand our studies of Vrk1 into a mammalian system, we acquired the ES cell line RRR178 from the International Gene Trap Consortium. This cell line is heterozygous for an insertion between exons three and four of the Vrk1 locus of the pGT0Lxf vector, which includes a strong splice acceptor at the 5 0 end, a bgeo fusion transcript, and a poly A site at the 3 0 end (Fig. 1A). Splicing of the trap into Vrk1 results in an mRNA transcript that contains only the first three exons of Vrk1 followed by the bgeo sequence, thereby excluding the kinase active domain, which is found in exon seven of the Vrk1 locus. Based on the work described below, this truncated transcript is sufficient to cause a reduction of Vrk1 wild-type expression. RRR178 ES cells were injected into mouse blastocysts and chimeric male offspring with germ cell transmission were crossed to B6 females in order to establish a colony of mice containing the Vrk1Gt(RRR178)Byg allele (hereafter referred to as Vrk1Gt). We determined the exact site of the gene trap within Vrk1 by conducting PCR with primer pairs spanning the length of the intron using DNA from a sample of the ES cells and then sequencing the product. The site of the gene trap insertion was at +6176 of intron 3. We then established a PCR genotyping strategy (Fig. 1B). Offspring from crosses between heterozygous Vrk1Gt mice were born at the expected Mendelian ratio of 1:2:1 of wild type (Vrk1+/+), heterozygous (Vrk1Gt/+) and homozygous mutant (Vrk1Gt/Gt) genotypes. All genotypes were viable and healthy and both sexes were equally represented.

2.2. VRK1 is broadly expressed with high expression in the testes and ovaries We first determined the expression profile of VRK1. Semiquantitative RT-PCR conducted on Vrk1+/+ somatic tissues and

Fig. 1 – Generation of Vrk1 gene trapped mice. (A) The Vrk1 locus is disrupted by insertion of the gene trap within the third intron. Exons are depicted as numbered boxes. Primers used for genotyping are shown as lettered arrows. (B) An example of the products produced from a genotyping PCR using the primer set shown in A. Primers ‘a’ and ‘b’ span the gene trap insertion site and create a product in the absence of the gene trap from either one or both of the Vrk1 alleles present in tail biopsies. Primers ‘a’ and ‘c’ create a product in the presence of the gene trap.

180


LacZ staining of whole mount Vrk1Gt/Gt embryonic day 13.5 embryos revealed broad expression of VRK1 throughout the mouse (data not shown, (Choi et al., 2010; Wiebe et al., 2010)). Northern blot analysis utilizing a probe directed to the 3 0 end of Vrk1 demonstrated that the transcript is expressed in all tissues tested, with highest expression in both the testes and ovaries as well as in the thymus (Fig. 2A). When compared to Vrk1+/+ testes, Vrk1Gt/Gt testes have approximately 70% reduction of Vrk1 expression, and Vrk1Gt/Gt ovaries display almost 80% reduction. This expression pattern and reduction of Vrk1 is consistent with a previous report of an independently generated mouse line carrying the same gene trap (Wiebe et al., 2010). Previous analysis of Vrk1 expression within the testes indicated that Vrk1 is present in both the somatic Sertoli cells as well as in the spermatogonia (Choi et al., 2010; Wiebe et al., 2010). We analyzed Vrk1 expression in the ovary of Vrk1Gt/Gt females by staining for LacZ, and found that LacZ staining was restricted to the developing oocyte, with no staining within the surrounding somatic granulosa cells (Fig. 2B). Notably, this staining only reflects the cells in which the Vrk1Gt transcript is expressed, and does not reflect endogenous protein localization, as the truncated Vrk1Gt transcript lacks the nuclear localization signal found in exon 11 of Vrk1. Nevertheless, this LacZ staining indicates that Vrk1 is expressed specifically within the oocyte.

2.3.

Vrk1Gt/Gt males display early onset sterility

No initial phenotypes were observed in Vrk1Gt/Gt mice at birth. However, after reaching sexual maturity, Vrk1Gt/Gt males were only briefly fertile, and became sterile by ten weeks of

1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

age. In order to determine the severity of this phenotype, eight Vrk1Gt/Gt males and eight Vrk1Gt/+ males were each mated with either Vrk1+/+ or Vrk1Gt/+ females over an eight month period beginning at 35 d.p.p. While the Vrk1Gt/+ males were able to sire litters with 100% success through 200 d.p.p., and 75% success through 275 d.p.p., the Vrk1Gt/Gt males never sired a litter after 69 d.p.p. (Fig. 3A), despite the continued presence of copulatory plugs. Notably, while the male Vrk1Gt/Gt mice used in this study were all fertile at the onset of sexual maturity, an independently derived mouse line that utilized the same gene trap resulted in a slightly more severe phenotype, in which only a single litter was obtained from one homozygous mutant male (Wiebe et al., 2010). The slight discrepancy between studies is most likely owing to the nature of the gene trap and the amount of wild-type VRK1 that is expressed due to leaky splicing around the gene trap insertion in the two separate mouse lines. Vrk1Gt/Gt testes weigh significantly less than Vrk1+/+ testes (Fig. 3B), although there is no significant difference in body mass between Vrk1+/+ and Vrk1Gt/Gt males at any age (p > 0.1, data not shown). The Drosophila homolog of VRK1, NHK-1, is required during meiosis, and has been implicated in the disassembly of the synaptonemal complex (SC) (Ivanovska et al., 2005; Lancaster et al., 2010). The SC mediates proper pairing, synapsis and recombination of homologous chromosomes prior to exit from meiotic prophase and entry into metaphase I (G2/MI transition), and failure to make this transition during mammalian spermatogenesis results in a meiotic arrest at prophase I and a decrease in testes weight (Sun et al., 2010). To address whether Vrk1Gt/Gt males might have meiotic defects, we assessed the formation of the SC both prior to sexual maturity at 17 d.p.p. as well as at

Fig. 2 – Expression analysis of Vrk1 gene trapped mice. (A) Northern blot depicting knockdown of the Vrk1 transcript (top) using bactin as a loading control. Normalized expression is shown below. Tissue abbreviations are as follows: Te, testis; Ov, ovary; Sm In, small intestine; Li, liver; Ki, kidney; He, heart; Th, thymus; Lu, lung; and Br, brain. (B) LacZ expression of the Vrk1 gene trap is observed within the oocytes of frozen ovarian sections from a Vrk1Gt/Gt mouse (right), but not in control sections from a Vrk1+/+ mouse (left). Original magnification is 25·.


1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

181

Fig. 3 – Vrk1Gt/Gt males become sterile due to spermatogenesis defects. (A) Sterility of Vrk1Gt/Gt males was assessed during the age ranges shown and successful mating attempts were recorded. n = 8 per genotype. (B) Testes from mice were obtained at the ages shown and weighed. Error bars indicate standard deviation. n P 4 per each genotype per each age. Student’s t test p values *p < 0.0035, **p < 0.00034. (C) Spermatocytes from Vrk1+/+ and Vrk1Gt/Gt mice were stained with an anti-SCP1 (green) and DAPI. Scale bars are 10 lm. (D) Representative sections of testes from Vrk1+/+ and Vrk1Gt/Gt mice reveal a progressive decline in spermatogenesis and loss/degeneration of the seminiferous tubular epithelium (arrowheads). Representative spermatids are circled. Scale bars are 50 lm.

77 d.p.p., after the onset of sterility. Using an antibody against the lateral unit of the SC, we found that the SC appears to form normally on chromosomes in Vrk1Gt/Gt spermatocytes (Fig. 3C). Meiotic defects that prevent a proper G2/MI transition can lead to an arrest of spermatogenesis at prophase I, which would result in the presence of only spermatogonia and spermatocytes within the seminiferous tubules. However, we initially observe all stages of spermatogenesis, including mature spermatids, within the seminiferous tubules of Vrk1Gt/Gt males during the first several weeks after sexual maturity (Fig. 3D, 6 and 10 weeks). Instead, histologic evaluation of testes sections revealed a progressive reduction of all stages of spermatogenesis within Vrk1Gt/Gt testes beginning at 10 weeks, that resulted in a significant loss of germ cells at

all stages of meiosis and differentiation, with marked loss/ degeneration of the seminiferous tubular epithelium, by 14 weeks (Fig. 3D, 10 and 14 weeks). In sum, we observe no detectable meiotic arrest, but rather a progressive loss of germ cells at all stages of differentiation. Two previous reports demonstrated that the spermatogenesis defect in mice with decreased VRK1 function in the germ line can be attributed to a loss of both differentiated and undifferentiated spermatogonia, and a decrease in overall proliferation within the testes (Choi et al., 2010; Wiebe et al., 2010). This data is consistent with our findings, although the onset of sterility is slightly delayed in our Vrk1Gt/Gt mice. Overall the phenotypes closely mirror each other, and provide independent evidence that a deficiency in VRK1 activity impedes the process of spermatogenesis over time.

182

2.4.


Vrk1Gt/Gt females are infertile

Unlike the male Vrk1Gt/Gt mice, the female Vrk1Gt/Gt mice never produced any litters. To monitor this phenotype more precisely, we set up a mating study in which Vrk1Gt/Gt females were mated with Vrk1+/+ males of proven fertility over the course of six months. To assess female fertility at the earliest point upon sexual maturity, six Vrk1Gt/Gt females were first mated at five weeks of age. Additionally, five Vrk1Gt/Gt females were first mated at eight weeks of age, for a total of eleven Vrk1Gt/Gt females tested. Deposition of seminal fluid was observed each time a female was placed with a male, indicating that a copulation event took place, yet there were no signs of pregnancy and no litters ever resulted. To further explore this phenotype, we examined ovarian histology. Vrk1Gt/Gt ovaries contain all stages of folliculogenesis including corpora lutea, which indicate successful ovulation events. Histologic comparison of Vrk1+/+ and Vrk1Gt/Gt ovaries did not uncover any obvious differences (Fig. 4). This observation led us to hypothesize that the sterility of Vrk1Gt/ Gt females might occur because of post-ovulation defects.

2.5.

1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

stage, but not beyond. Moreover, unmated Vrk1+/+ control females produced similar numbers of two-cell embryos (p > 0.2), indicating that such embryos were probably the result of parthenogenesis. To investigate whether oocytes from Vrk1Gt/Gt females were fertilized, we examined formation of pronuclei. While zygotes with visible maternal and paternal pronuclei were retrieved from both Vrk1+/+ and Vrk1Gt/+ mice, we never observed both maternal and paternal pronuclei in oocytes obtained from Vrk1Gt/Gt mice (Fig. 5C). Moreover, we noted that many oocytes from Vrk1Gt/Gt mice failed to reach MII arrest, as there were also earlier stages of meiosis represented (Fig 5B and note that the oocyte in Fig. 5C is actually a GV oocyte). Therefore, oocytes from Vrk1Gt/Gt mice do not appear to be fertilized despite the fact that sperm are observed within the oviduct of Vrk1Gt/Gt females, and in contact with oocytes (black arrowheads in Fig. 5A), when oocytes are obtained from the ampulla. Together, these findings led us to conclude that defects in oogenesis impair fertilization but not ovulation. This conclusion differs from Drosophila, where disruption of NHK-1/VRK1 does not impede fertilization (Ivanovska et al., 2005).

Oocytes from Vrk1Gt/Gt females are not fertilized

In order to assess embryonic development, we mated superovulated females with Vrk1+/+ males and then obtained potentially fertilized oocytes from the ampullae of plugged females the following morning. These oocytes were cultured over a four day period to monitor embryonic development. Embryos from both the Vrk1+/+ and Vrk1Gt/+ females developed past the two-cell and morula stages and formed blastocysts (Fig. 5A and B). However, no embryonic development was detected in most oocytes from Vrk1Gt/Gt mice. Some oocytes obtained from Vrk1Gt/Gt mice did progress to the two-cell

Fig. 4 – Ovarian histopathology from Vrk1+/+ and Vrk1Gt/Gt mice. Shown here are representative sections of ovaries stained with hematoxylin and eosin from 4 week-old Vrk1+/ + (left) and Vrk1Gt/Gt mice (right). No significant morphologic differences in preantral follicles (arrowheads), mature follicles (*), or corpora lutea (CL) were detected in serial sections of ovaries from multiple mice from each genotype. Scale bars are 500 lm.

2.6. Oocytes from Vrk1Gt/Gt mice have developmental deficiencies Superovulation of Vrk1+/+ and Vrk1Gt/Gt mice produces comparable numbers of oocytes (22.1 ± 9.5 versus 21.5 ± 13.6, respectively, n = 13 stimulations). However, the prevalence of oocytes displaying proper polar body extrusion was significantly less in oocytes from Vrk1Gt/Gt mice compared to Vrk1+/+ oocytes (25% versus 87.7%, respectively, p < 0.0005). To assess oocyte development more precisely, we hormonally stimulated Vrk1+/+ and Vrk1Gt/Gt mice, obtained GV stage oocytes from antral follicles, and followed oocyte development in culture. After incubation for 20 h in vitro, we found a significant difference in developmental progression between oocytes from Vrk1+/+ and Vrk1Gt/Gt mice. While 59.9% of the oocytes from Vrk1+/+ females displayed polar body extrusion, only 27.9% of oocytes from Vrk1Gt/Gt females progressed to this stage (p < 0.0005). Although germinal vesicle breakdown (GVBD) appeared to take place normally in oocytes from Vrk1Gt/Gt mice (p > 0.02), 67.7% progressed through GVBD but failed to extrude a polar body versus only 39.2% from Vrk1+/ + females (p < 0.0005), indicating meiotic delays. To distinguish more precisely the stage at which these delays arise during oogenesis in Vrk1Gt/Gt mice, we followed meiotic oocyte development in vitro from the GV stage through metaphase II (Fig. 6A). We initially detected similar numbers of rimmed and condensed oocytes, as described in (Can et al., 2003) from both Vrk1+/+ and Vrk1Gt/Gt mice. Subsequently, significantly fewer oocytes from Vrk1Gt/Gt females reached the ‘‘parachute’’ stage, a point at the completion of GVDB during which bivalents clump together prior to entrance into prometaphase I as described in (Debey et al., 1993), within 2.5 h, compared to Vrk1+/+ females (15.6% versus 49.1% respectively, p < 0.0005). Yet, we again found comparable numbers of oocytes from Vrk1+/+ and Vrk1Gt/Gt females at later time points corresponding to prometaphase and metaphase I, indicating that the mutant oocytes had recovered


1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

183

Fig. 5 – Vrk1Gt/Gt oocytes fail to be fertilized. (A) Potentially fertilized oocytes were cultured in vitro and development was assessed at time points indicated. Embryonic development was observed for potentially fertilized oocytes from Vrk1+/+ and Vrk1Gt/+females, but not from Vrk1Gt/Gt females. n = 65 for Vrk1+/+, 52 for Vrk1Gt/+, and 84 for Vrk1Gt/Gt oocytes. White arrowheads indicate pronuclei. Black arrowheads indicate sperm. Scale bars are 10 lm. (B) Graphical representation of the developmental stages observed at each time point during culture for Vrk1+/+ and Vrk1Gt/Gt females. Vrk1Gt/+ numbers are very similar to Vrk1+/+ and are not included. (C) Potentially fertilized oocytes were stained with DAPI. Pronuclei were observed in Vrk1+/+ and Vrk1Gt/+oocytes, but not in Vrk1Gt/Gt oocytes. Scale bars are 10 lm.

from the earlier delay at the parachute stage. However, significantly fewer oocytes from Vrk1Gt/Gt mice ultimately achieved metaphase II, (22.5% versus 54.5% in Vrk1+/+ mice, p < 0.0005). Furthermore, of those oocytes from Vrk1Gt/Gt mice that had reached metaphase II, a significant number displayed lagging chromosomes (42.9% versus 2.7% from Vrk1+/+ mice, p < 0.0005) (Fig. 6B). Consequently, less than 13% of oocytes from Vrk1Gt/Gt females are able to proficiently reach metaphase II. Despite the presence of lagging chromosomes, formation of the spindle appears normal and spindle length is equivalent between Vrk1Gt/Gt and Vrk1+/+ spindles (33.04 ± 6.79 lm versus 29.82 ± 2.56 lm, respectively, p > 0.3). Furthermore, spindle length is similar between Vrk1Gt/Gt oocytes with lagging chromosomes and those without (p > 0.6). We next assessed chromosomal spreads for the formation and number of bivalents at metaphase I, and univalents at metaphase II, to determine whether these lagging chromosomes lead to mis-segregation and aneuploidy in Vrk1Gt/Gt oocytes. At metaphase I, we observed normal pairing of bivalents, and at metaphase II, Vrk1Gt/Gt oocytes displayed 20 univalent chromosomes (Fig. 7A). These findings lead us to conclude that even though lagging chromosomes can be detected in approximately 40% of the Vrk1Gt/Gt oocytes that reach metaphase II, the lagging chromosome(s) eventually reach the

spindle, and ultimately, the overall chromosome number is unaffected.

2.7. Known phosphorylation targets of VRK1 are not affected in Vrk1Gt/Gt oocytes Given that the majority of Vrk1Gt/Gt oocytes fail to reach metaphase II properly, we wondered whether known phosphorylation targets of VRK1 were properly modified in the developing Vrk1Gt/Gt oocytes. Histone H3 Ser10 has been established as a mitotic target of VRK1 in mammalian cell culture (Kang et al., 2007), and phosphorylated H3 Ser10 is present during mouse oocyte meiotic progression (Swain et al., 2007). When we tracked the phosphorylation status of H3Ser10 during oocyte development from GV through metaphase II by immunofluorescence, we saw that H3 phosphoSer10 staining was comparable between Vrk1+/+ and Vrk1Gt/ Gt oocytes (Fig. 7B). Histone H2A Thr119 (equivalent to mouse H2A Thr120) is a target of Drosophila NHK-1/VRK1, both during mitosis (Aihara et al., 2004; Brittle et al., 2007; Cullen et al., 2005) as well as during female meiosis (Ivanovska et al., 2005). Using an antibody directed against phosphorylated histone H2A Thr120, we observed phosphorylation in the developing oocyte beginning at the GV stage and persisting through meiosis to metaphase

184


1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

Fig. 6 – Vrk1Gt/Gt oocytes have defects at the parachute and metaphase II stages of development. (A) Vrk1+/+ and Vrk1Gt/Gt GV oocytes were obtained from antral follicles and cultured in vitro for the time indicated at the left, then fixed and stained with DAPI. The expected chromosomal configurations at each time point are also indicated at the left. A representative image for each genotype at each time point is shown, with the percent of oocytes in that stage for each genotype listed to the right. n P 45 for each genotype at each stage. (B) Vrk1+/+ and Vrk1Gt/Gt metaphase II oocytes were stained with DAPI and anti-atubulin (green). White arrowheads indicate lagging chromosomes. Scale bars are 10 lm.

II. Levels of phosphorylated histone H2A Thr120 staining, like histone H3 Ser10, were similar between Vrk1+/+ and Vrk1Gt/Gt oocytes (Fig. 8). Therefore, the two histone modifications previously implicated as regulated by VRK1 are in fact globally unaffected in Vrk1Gt/Gt oocytes.

2.8.

Vrk1Gt/Gt phenotypes are not rescued by loss of p53

In C. elegans, the proliferation defect found in the vrk-1 mutant germ line can be partially rescued by loss of the C. elegans homolog of the tumor suppressor p53 (Waters et al., 2010). This finding suggests that VRK1 acts to inhibit the cell cycle arrest activity of p53. Additionally, in vitro studies indicate that murine VRK1 is able to phosphorylate p53 at Thr18 within the region that mediates the regulatory interaction of p53 with MDM2 (Lopez-Borges and Lazo, 2000). We wondered if this inhibitory relationship could be detected in our mice and perhaps rescue the sterility observed in Vrk1Gt/Gt animals. We therefore created mice that carried both the Vrk1 gene trap mutation as well as the Trp53tm1Tyj mutation. Mice homozygous mutant for both Vrk1 and Trp53tm1Tyj (hereafter referred to as Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj) are viable and healthy at birth with no visible phenotypes. As seen for p53 single mutant mice, the Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj mice begin to succumb to tumors around 3 months of age (Jacks et al., 1994).

Because of the rapid onset of tumors, the long term fertility studies that we carried out for Vrk1Gt/Gt animals were not feasible. However, given that the sterility in Vrk1Gt/Gt males occurred by 70 d.p.p., we were still able to assess whether Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj males were rescued for sterility. To that end, four Vrk1Gt/+, Trp53tm1Tyj/tm1Tyj males and three Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj males were each mated with Vrk1Gt/+, Trp53tm1Tyj/+ females beginning at 35 d.p.p. The Vrk1Gt/+, Trp53tm1Tyj/ tm1Tyj males were able to sire litters with 100% success through 80 d.p.p., and 75% success through 136 d.p.p. All three of the Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj males tested were able to sire a litter between the ages of 35 and 50 d.p.p, and two sired litters after 50 d.p.p.; one at 68 d.p.p. and one at 69 d.p.p. However, none sired litters after this time, despite the continued presence of copulatory plugs. Thus, the male sterility phenotype persists even upon the loss of p53 activity. Evaluation of testes weight (Fig. 9A) and histologic sections (Fig. 9B) demonstrated that the proliferation defect that results in loss of spermatogenesis is not rescued by the p53 mutation in Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj males. In fact, the severity of the phenotype is slightly increased; loss of spermatogenesis occurs earlier in the Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj males. We next evaluated sterility of the Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj females by mating four Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj females to


1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

185

Fig. 7 – Vrk1Gt/Gt oocytes have grossly normal histone H3 Ser10 phosphorylation. (A) Vrk1+/+ and Vrk1Gt/Gt GV oocytes were obtained from antral follicles and cultured in vitro to metaphase I (top) or metaphase II (bottom) and then stained with DAPI (green) and phosphorylated histone H3 Ser10 (red). Bivalent and univalent formation and number were also noted at metaphase I and metaphase II, respectively. (B) Vrk1+/+ and Vrk1Gt/Gt GV oocytes were obtained from antral follicles and cultured in vitro to each stage noted, then fixed and stained with DAPI (green) and phosphorylated histone H3 Ser10 (red). Scale bars are 10 lm. Vrk1+/+, Trp53+/+ males. Although we obtained litters from Vrk1Gt/+, Trp53tm1Tyj/tm1Tyj female controls, no Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj females produced any litters or showed any signs of pregnancy, despite the presence of copulatory plugs. To date, we have not been able to obtain a sufficient number of Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj females to assess meiotic progression. Together, these data argue that the inhibitory relationship observed between VRK1 and p53 in C. elegans is not crucial for mouse germ cell development.

2.9. The Vrk1Gt/Gt phenotype is consistent in multiple backgrounds To test whether the phenotypes described herein are dependent upon a particular strain background, we backcrossed our Vrk1Gt(RRR178)Byg mouse line eight times onto both the FVB and C57BL/6 backgrounds. Vrk1Gt/Gt mice on both backgrounds are viable and have no apparent mutant phenotypes at birth. We have conducted comparable female fertility studies as mentioned above using three six-week-old Vrk1Gt/Gt C57BL/6 females and two six-week-old Vrk1Gt/Gt FVB females. We have not obtained any litters as of yet despite the pres-

ence of copulatory plugs; moreover, sibling Vrk1Gt/+ FVB and C57BL/6 control females are fertile. Additionally, we have also evaluated the FVB and C57BL/6 Vrk1Gt/Gt males. We have obtained one litter each from two C57BL/6 Vrk1Gt/Gt males mated at 42 d.p.p. but they have not produced a second litter, despite reaching 98 d.p.p. Therefore, it appears that the sterile Vrk1Gt/ Gt phenotypes of both sexes still occur on different strain backgrounds.

3.

Discussion

3.1. VRK1 has different yet essential roles in male and female germ cell development Here, we present evidence that VRK1 functions in a conserved role during mammalian female gametogenesis. We additionally provide an independent demonstration that a reduction of VRK1 leads to a progressive loss of spermatogenesis. Our data are consistent with previously defined functions of VRK1 in the germ cells of other organisms. The proliferation defect observed during spermatogenesis in Vrk1Gt/Gt mice is reminiscent of the germline stem cell defect

186


1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

Fig. 8 – Vrk1Gt/Gt oocytes have grossly normal histone H2A Thr120 phosphorylation. Vrk1+/+ and Vrk1Gt/Gt GV oocytes were obtained from antral follicles and cultured in vitro to each stage noted. Then, they were fixed and stained with DAPI (green) and phosphorylated histone H2A Thr120 (red). Scale bars are 10 lm.

observed in C. elegans (Waters et al., 2010), while the defect observed in oogenesis in Vrk1Gt/Gt mice is similar to the meiotic defects described in Drosophila (Cullen et al., 2005; Ivanovska et al., 2005). The sexual dimorphism between these two phenotypes is likely due to the fact that mitotic proliferation occurs in spermatogonia throughout male fecundity, but not in the oogonia of sexually mature females. Furthermore, given that we were able to rule out any obvious meiotic defects in male Vrk1Gt/Gt mice, our findings point to a female-specific role for VRK1 during meiosis.

3.2.

VRK1 in female meiosis

Along with previous work in Drosophila oocytes, our study implicates a conserved function for VRK1 during mammalian female meiosis. A prolonged prophase I and the metaphase II arrest seen in vertebrates are both highly conserved, critical characteristics of female meiosis, while a metaphase I arrest is specific to insects. The developmental defects that we observe during Vrk1Gt/Gt oocyte meiotic progression, specifically during the clustering of chromosomes at the end of prophase I and during the formation of a metaphase II oocyte, are superficially similar to the phenotype seen in Drosophila nhk-1/Vrk1 mutant oocytes, which display developmental errors in both prophase I and the metaphase I arrest (Ivanovska et al., 2005). The clumping of chromosomes into a parachute formation during mammalian meiosis is reminiscent of the karyosome that is formed in the nucleus of Drosophila oocytes undergoing meiosis. Both configurations bring partially condensed chromosomes into close proximity prior to full condensation and further meiotic progression. The fact that we see significantly fewer Vrk1Gt/Gt oocytes reach this stage, along with the defect in karyosome formation and maintenance of Drosophila nhk-1/Vrk1 mutants (Cullen et al., 2005; Ivanovska et al.,

2005), further points to a role for VRK1 during prophase I of female meiosis. Additionally, it has been suggested that the meiotic checkpoint suppresses NHK-1/VRK1 function in Drosophila oocytes, thereby preventing its activity in chromosomal reorganization until the proper stage of prophase I is reached (Lancaster et al., 2010). Finally, as previously mentioned, during female meiosis, insects have an arrest at metaphase I prior to fertilization, while vertebrates arrest at metaphase II. In Drosophila nhk-1/Vrk1 mutants, 50% of oocytes have chromosomes that fail to congress properly at metaphase I (Ivanovska et al., 2005) and we found that only 13% of Vrk1Gt/Gt oocytes display a proper metaphase II chromosome orientation. Since we observe lagging chromosomes in Vrk1Gt/Gt oocytes, it is possible that VRK1 acts along with other proteins to trigger the spindle checkpoint. Together, this evidence suggests that VRK1 may mediate proper chromosomal orientation during both prophase I and the metaphase arrest of female meiosis, whether it occurs in meiosis I or meiosis II. Given that we observed significantly fewer Vrk1Gt/Gt oocytes that progressed correctly to metaphase II due to the involvement of VRK1 in orchestrating proper chromosomal configurations in meiosis, it is understandable that there may be a decrease in fertilized Vrk1Gt/Gt oocytes. However, our data do not completely explain the complete lack of any discernable fertilization in Vrk1Gt/Gt oocytes, as a low percentage of oocytes appear to have progressed through meiosis normally. Possibly, other defects have occurred that are not apparent from our assays. The residual Vrk1 expression present in Vrk1Gt/Gt oocytes might mask a more severe defect. Additionally, we cannot exclude the possibility that VRK1 might function in non-ovarian reproductive tissues, or in sperm–egg recognition, to also influence fertilization. Generation of tissue-specific null alleles of Vrk1 in the future would help to address these issues.


1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

187

Fig. 9 – Vrk1Gt/Gt, Trp53tm1Tyj/ tm1Tyj males become sterile due to spermatogenesis defects. (A) Testes from mice were obtained at the ages shown and weighed. Error bars indicate standard deviation. n = 4 per each genotype per each age. Student’s t test p values **p < 0.00034. (B) Representative sections of testes from Vrk1Gt/+, Trp53tm1Tyj/tm1Tyj and Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj mice reveal a progressive decline in spermatogenesis and loss/degeneration of the seminiferous tubular epithelium (arrowheads). Representative spermatids are circled. Scale bars are 50 lm.

3.3. Non-conserved aspects of VRK1 function in female meiosis A measurable decrease in histone H3 Ser10 phosphorylation in mammalian cells with reduced VRK1 activity has been previously reported (Kang et al., 2007). Furthermore, the phosphorylation of histone H2A Thr119 requires NHK-1/VRK1 in Drosophila oocytes (Ivanovska et al., 2005). However, we found the phosphorylation of these histone targets to be apparently unaffected in oocytes with reduced VRK1 function. Possibly, the loss of VRK1 is supplemented by functional redundancy with other kinases. For example, Aurora kinases have been shown to target histone H3 Ser10 within the oocyte (Swain et al., 2008; Wang et al., 2006) and it has also been demonstrated that H2A Thr119 phosphorylation by NHK-1/VRK1 in Drosophila requires Aurora B kinase (Brittle et al., 2007). Additionally, Bub1 kinase has been shown to target histone H2A Thr120 in mouse spermatocytes (Kawashima et al., 2010) and although this action has not been documented in female meiosis, Bub1 activity has been implicated during the spindle checkpoint in mouse oocytes (Yin et al., 2006). Alternatively, the persistence of phosphorylation in Vrk1Gt/Gt oocytes could be due to proper VRK1 function stemming from the approximately 20% wild-type expression that is still present in the Vrk1Gt/Gt mice. In an effort to evaluate the activity of VRK1 within mammalian gametogenesis we looked to determine whether a previously described inhibitory relationship observed between

VRK1 and p53 in the germ line of C. elegans was conserved in mammals. Ultimately, the genetic evidence demonstrated in C. elegans only allowed for the defect in germline proliferation to be restored and did not affect the sterility phenotype of the vrk-1 mutant (Waters et al., 2010). Our results were analogous to the C. elegans data, as we found that Vrk1Gt/Gt, Trp53tm1Tyj/tm1Tyj mice remain sterile. However, the histologic evaluation in males revealed that the proliferation defect observed in Vrk1Gt/Gt testes persisted in Vrk1Gt/Gt, Trp53tm1Tyj/ tm1Tyj testes, unlike C. elegans, where loss of p53 rescued the proliferation defects of vrk-1 mutants. Therefore, our findings suggest that the inhibitory relationship between VRK1 and p53 is not likely to be conserved, or at least essential in mammalian germ cells. The relationship between VRK1 and p53 is undoubtedly very complex, as it has also been suggested that VRK1 acts to stabilize p53 in the absence of DNA damage to allow for a basal level of p53 activity (Vega et al., 2004). Future investigation of the interaction between VRK1 and p53 within the mammalian germ line will help to elucidate whether these proteins work together to control the regulation of gametogenesis. By generating a mouse model with a gene trap insertion in the Vrk1 locus, we have provided the first evidence for a conserved role of VRK1 within mammalian female gametogenesis. Further exploration of the role of VRK1 will allow for an increased understanding of the choreography of proper chromosomal configurations during mammalian meiosis and perhaps even in the mitotic cell cycle.

188


4.

Experimental procedures

4.1.

Animals

Protocols for maintenance of mouse colonies as well as all experimental procedures performed were approved by Yale University’s Institutional Animal Care and Use Committee (IACUC). Mice heterozygous for the Trp53tm1Tyj mutation were obtained from Dr. Junjie Chen, Yale University School of Medicine, New Haven, CT. C57BL/6 mice and FVB/NJ mice were obtained from the Jackson Laboratory (Bar Harbor, ME).

4.2.

Generation of Vrk1 gene trapped mice

The Vrk1 gene trapped ES cell line RRR178 (MGI Allele: Vrk1Gt(RRR178)Byg) was obtained from BayGenomics (now part of the International Gene Trap Consortium, Davis, CA). Chimeric mice were generated by Animal Genomics Services at Yale University. Briefly, the ES cells containing the gene trapped locus were injected into C57BL/6 blastocysts which were then transferred into pseudopregnant CD-1 females. Resulting chimeric males were assessed by coat color and subsequently mated to C57BL/6 females. Agouti offspring from these crosses were further evaluated in order to determine germline transmission of the gene trapped locus.

4.3.

Genotyping

DNA was isolated from tail biopsies taken from animals ranging in age from 10 to 21 days post partum (d.p.p.). Tissue was incubated in digest solution (10 mM Tris pH 8.0, 25 mM EDTA pH 8.0, 100 mM NaCl, 1% SDS, and 0.2 mg/ml Proteinase K) and DNA was subsequently extracted using phenol:chloroform (1:1) and isopropanol. Primer sequences for the PCR genotyping strategy described in Fig. 1 are as follows: (a) 5 0 -GG AGAAACTTTGTACAGCTTCG-3 0 , (b) 5 0 -AGGTGACGGAGTTCATT CTTGG-3 0 , (c) 5 0 -CCAAAGGGCAAACCCAAAAGGG-3 0 .

1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

embedded in paraffin, sectioned at 5 lm, and stained with hematoxylin and eosin. Testes were processed similarly, but sectioned at 7 lm. Samples were evaluated by investigators blinded to the experimental group. Digital light microscopic images (CJB) were acquired using a Zeiss Axio Imager A1 microscope, an AxioCam MRc 5 camera, and AxioVision 4.7.1 imaging software (Carl Zeiss Micro Imaging, Inc., Thornwood, NY). The resulting images were optimized using Adobe Photoshop 8.0 (San Jose, CA).

4.6.

Staining for LacZ activity

Whole ovaries were incubated in 4% paraformaldehyde with 2 mM MgCl2 and 5 mM EGTA for 1.5 h. Ovaries were then rinsed in detergent solution (2 mM MgCl2, 0.01% sodium deoxycholate, 0.02% Nonidet P-40 in 1xPBS) and stained overnight with X-gal (detergent solution plus 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 1 mg/ml X-gal, and 20 mM Tris) (Nagy, 2003) at 37 C in a humidity chamber. Afterward, ovaries were rinsed in 1xPBS, and incubated overnight in 1· PBS with 30% sucrose and 2 mM MgCl2. Next, ovaries were incubated in 30% sucrose/OCT 1:1 at 4 C, followed by incubation in OCT at 4 C and finally, tissues were then immersed in fresh OCT and frozen. Ovaries were cryosectioned at 10 lm, rinsed with H2O and then stained with Nuclear Fast Red (Sigma, St. Louis, MO). Images were acquired (CSS) using a Zeiss Axiovert 200 M microscope, a Photometrics CoolSNAP HQ2 camera, and IP Lab 4.0.8 imaging software (BD BioSciences, Rockville, Maryland). The resulting images were optimized (CJB) using Adobe Photoshop 8.0 (San Jose, CA).

4.7.

Analysis of pre-implantation development

Tissues were isolated from 3 month old mice and total RNA was extracted with TRIzol (Invitrogen, Carlsbad, CA). Forty micrograms of RNA was separated on a 1% agarose gel containing 3% formaldehyde buffered with 0.2 mM MOPS/ NaOH (pH 7.0). Following electrophoresis, RNA was transferred to a Hybond-N membrane (Amersham, Pascataway, NJ) and crosslinked by baking at 80 C. The probe used to detect Vrk1 an RNA levels was directed to the 3 0 end of the transcript and was made with the following primers: 5 0 -GGTGAGATCGCTAAGTACATGG-3 0 and 5 0 -CCAGACTTTATTG TCACAACG-3 0 . The b-actin control probe was made with the following primers: 5 0 -GCGAGCACAGCTTCTTTGCAGC-3 0 and 5 0 -CGCAGCTCAGTAACAGTCC-3 0 . Probe generation, hybridization and subsequent washes were carried out as previously described (Reinke and Lozano, 1997).

For early embryo studies, female mice were superovulated by intraperitoneal injection of 5 IU of PMSG (Sigma, St. Louis, MO) and 24 h later, 5 IU of hCG (Sigma, St. Louis, MO), followed directly by natural mating with a wild-type male. Cumulus oocyte complexes were acquired from the oviducts of plugged females 16 h later and oocytes were isolated by incubation in M2 culture media containing hyaluronidase (Millipore Corporation, Phillipsburg, NJ). To assess fertilization, oocytes were fixed in 4% paraformaldehyde, permeabilized in 0.5% Nonidet P-40, stained with DAPI and washed in 1· PBS with 0.1% Tween and 0.01% TritonX-100. Slides were mounted with DABCO (Sigma, St. Louis, MO) and images were obtained using a Zeiss Axioplan 2 microscope, an AxioCam MRm camera and AxioVision 4.6 imaging software (Carl Zeiss Micro Imaging Inc., Thornwood, NY). To evaluate developmental potential, oocytes were cultured at 37 C with 5% CO2 first in M16 media (Millipore Corporation, Phillipsburg, NJ) and then transferred to KSOM media (Millipore Corporation, Phillipsburg, NJ). Images were acquired with a Zeiss Axiovert 200 M microscope, a Photometrics CoolSNAP HQ2 camera, and IP Lab 4.0.8 imaging software (BD BioSciences, Rockville, Maryland).

4.5.

4.8.

4.4.

Northern blot analysis

Histology

Ovaries were fixed in Bouin’s fixative (Ricca Chemical Corporation, Arlington, TX), processed by routine methods,

Oocyte development

To collect germinal vesicle (GV) oocytes, mice were superovulated by intraperitoneal injection of 5 IU of PMSG (Sigma,


St. Louis, MO). Twenty-two hours later, GV staged oocytes were retrieved by puncturing antral follicles under a dissecting microscope. Oocytes were cultured to the desired time point at 37 C with 5% CO2 in culture media (a-MEM-glutaMax, 5% FBS, 1· Insulin–Transferrin–Selenium (ITS), 100· Penicillin– Streptomycin (GIBCO, Invitrogen, Carlsbad, CA), and 108 IU/ ml recombinant FSH (a kind gift from M. Lalioti)). Independent cohorts of mice were analyzed for each timepoint, thus, the analysis does not represent an actual ‘‘timecourse’’. Therefore, the assessment of the fraction of oocytes at each stage shown in Fig. 6A does not total 100% for a given genotype because only the oocytes at a single timepoint were analyzed.

4.9.

Immunofluorescence

Oocytes at the desired stage were fixed in 4% paraformaldehyde in 1· PBS for 30 min, permeabilized in 0.5% Nonidet P-40 in 1· PBS for 30 min, blocked in 1% BSA with 0.1% Tween in 1· PBS for 1 h at room temperature, incubated with primary antibody overnight at 4 C, rinsed three times in 0.1% Tween with 0.01% Triton-X-100 in 1· PBS, incubated with a fluorescent secondary antibody (1:500, Molecular Probes, Carlsbad, CA) for 1 h at room temperature, and rinsed as above but with 10 lg/ml DAPI added to the first rinse. Slides were mounted with DABCO (Sigma, St. Louis, MO) and images were obtained using a Zeiss Axioplan 2 microscope, an AxioCam MRm camera and AxioVision 4.6 imaging software (Carl Zeiss Micro Imaging Inc., Thornwood, NY). The following primary antibodies and dilutions were used: H3 phospho-Ser10 (1:200, Upstate, Billerica, MA), H2A phospho-Thr120 (1:500, Active Motif, Carlsbad, CA), a-tubulin (1:200, Sigma, St. Louis, MO), and SCP-1 (1:250, Abcam, Cambridge, MA).

4.10.

Chromosomal spreading

Oocytes were cultured to metaphase I or metaphase II as described in Section 4.8 and then processed for chromosomal spreading as previously described (Hodges and Hunt, 2002; Susiarjo et al., 2009). Following spreading, slides were processed for immunofluorescence and imaged as mentioned in Section 4.9.

Acknowledgements The authors thank the Brueckner Lab, the Bogue Lab, the Bale Lab, the Cooley Lab, the Lin Lab, Maria Lalioti, Scott Weatherbee, and Yale Animal Genomics Services for useful discussion and sharing their reagents and equipment. We would also like to thank the members of the Reinke Lab for comments on this manuscript and support. This work was funded by the March of Dimes and the Connecticut Stem Cell Research Grants Program.

R E F E R E N C E S

Aihara, H., Nakagawa, T., Yasui, K., Ohta, T., Hirose, S., Dhomae, N., Takio, K., Kaneko, M., Takeshima, Y., Muramatsu, M., Ito, T.,

1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

189

2004. Nucleosomal histone kinase-1 phosphorylates H2A Thr 119 during mitosis in the early Drosophila embryo. Genes Dev. 18, 877–888. Brittle, A.L., Nanba, Y., Ito, T., Ohkura, H., 2007. Concerted action of Aurora B, Polo and NHK-1 kinases in centromere-specific histone 2A phosphorylation. Exp. Cell Res. 313, 2780–2785. Can, A., Semiz, O., Cinar, O., 2003. Centrosome and microtubule dynamics during early stages of meiosis in mouse oocytes. Mol. Hum. Reprod. 9, 749–756. Choi, Y.H., Park, C.H., Kim, W., Ling, H., Kang, A., Chang, M.W., Im, S.K., Jeong, H.W., Kong, Y.Y., Kim, K.T., 2010. Vaccinia-related kinase 1 is required for the maintenance of undifferentiated spermatogonia in mouse male germ cells. PLoS ONE 5, e15254. Cullen, C.F., Brittle, A.L., Ito, T., Ohkura, H., 2005. The conserved kinase NHK-1 is essential for mitotic progression and unifying acentrosomal meiotic spindles in Drosophila melanogaster. J. Cell Biol. 171, 593–602. Debey, P., Szollosi, M.S., Szollosi, D., Vautier, D., Girousse, A., Besombes, D., 1993. Competent mouse oocytes isolated from antral follicles exhibit different chromatin organization and follow different maturation dynamics. Mol. Reprod. Dev. 36, 59–74. Gorjanacz, M., Klerkx, E.P., Galy, V., Santarella, R., Lopez-Iglesias, C., Askjaer, P., Mattaj, I.W., 2007. Caenorhabditis elegans BAF-1 and its kinase VRK-1 participate directly in post-mitotic nuclear envelope assembly. EMBO J. 26, 132–143. Handel, M.A., Schimenti, J.C., 2010. Genetics of mammalian meiosis: regulation, dynamics and impact on fertility. Nat. Rev. Genet. 11, 124–136. Hassold, T., Hunt, P., 2001. To err (meiotically) is human: the genesis of human aneuploidy. Nat. Rev. Genet. 2, 280–291. Hodges, C.A., Hunt, P.A., 2002. Simultaneous analysis of chromosomes and chromosome-associated proteins in mammalian oocytes and embryos. Chromosoma 111, 165– 169. Ivanovska, I., Khandan, T., Ito, T., Orr-Weaver, T.L., 2005. A histone code in meiosis: the histone kinase, NHK-1, is required for proper chromosomal architecture in Drosophila oocytes. Genes Dev. 19, 2571–2582. Jacks, T., Remington, L., Williams, B.O., Schmitt, E.M., Halachmi, S., Bronson, R.T., Weinberg, R.A., 1994. Tumor spectrum analysis in p53-mutant mice. Curr. Biol. 4, 1–7. Kang, T.H., Park, D.Y., Choi, Y.H., Kim, K.J., Yoon, H.S., Kim, K.T., 2007. Mitotic histone H3 phosphorylation by vaccinia-related kinase 1 in mammalian cells. Mol. Cell. Biol. 27, 8533–8546. Kang, T.H., Park, D.Y., Kim, W., Kim, K.T., 2008. VRK1 phosphorylates CREB and mediates CCND1 expression. J. Cell Sci. 121, 3035–3041. Kawashima, S.A., Yamagishi, Y., Honda, T., Ishiguro, K., Watanabe, Y., 2010. Phosphorylation of H2A by Bub1 prevents chromosomal instability through localizing shugoshin. Science 327, 172–177. Klerkx, E.P., Lazo, P.A., Askjaer, P., 2009. Emerging biological functions of the vaccinia-related kinase (VRK) family. Histol. Histopathol. 24, 749–759. Lancaster, O.M., Breuer, M., Cullen, C.F., Ito, T., Ohkura, H., 2010. The meiotic recombination checkpoint suppresses nhk-1 kinase to prevent reorganisation of the oocyte nucleus in Drosophila. PLoS Genet. 6, e1001179. Lancaster, O.M., Cullen, C.F., Ohkura, H., 2007. NHK-1 phosphorylates BAF to allow karyosome formation in the Drosophila oocyte nucleus. J. Cell Biol. 179, 817–824. Lopez-Borges, S., Lazo, P.A., 2000. The human vaccinia-related kinase 1 (VRK1) phosphorylates threonine-18 within the mdm-2 binding site of the p53 tumour suppressor protein. Oncogene 19, 3656–3664. Matzuk, M.M., Lamb, D.J., 2002. Genetic dissection of mammalian fertility pathways. Nat. Cell Biol. 4, s41–49.

190


Matzuk, M.M., Lamb, D.J., 2008. The biology of infertility: research advances and clinical challenges. Nat. Med. 14, 1197–1213. Nagy, A., 2003. Manipulating the mouse embryo: a laboratory manual, third ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY. Nichols, R.J., Wiebe, M.S., Traktman, P., 2006. The vaccinia-related kinases phosphorylate the N 0 terminus of BAF, regulating its interaction with DNA and its retention in the nucleus. Mol. Biol. Cell 17, 2451–2464. Reinke, V., Lozano, G., 1997. The p53 targets mdm2 and Fas are not required as mediators of apoptosis in vivo. Oncogene 15, 1527– 1534. Sevilla, A., Santos, C.R., Barcia, R., Vega, F.M., Lazo, P.A., 2004a. CJun phosphorylation by the human vaccinia-related kinase 1 (VRK1) and its cooperation with the N-terminal kinase of c-Jun (JNK). Oncogene 23, 8950–8958. Sevilla, A., Santos, C.R., Vega, F.M., Lazo, P.A., 2004b. Human vaccinia-related kinase 1 (VRK1) activates the ATF2 transcriptional activity by novel phosphorylation on Thr-73 and Ser-62 and cooperates with JNK. J. Biol. Chem. 279, 27458– 27465. Sun, F., Palmer, K., Handel, M.A., 2010. Mutation of Eif4g3, encoding a eukaryotic translation initiation factor, causes male infertility and meiotic arrest of mouse spermatocytes. Development 137, 1699–1707. Susiarjo, M., Rubio, C., Hunt, P., 2009. Analyzing mammalian female meiosis. Methods Mol. Biol. 558, 339–354.

1 2 8 ( 2 0 1 1 ) 1 7 8 –1 9 0

Swain, J.E., Ding, J., Brautigan, D.L., Villa-Moruzzi, E., Smith, G.D., 2007. Proper chromatin condensation and maintenance of histone H3 phosphorylation during mouse oocyte meiosis requires protein phosphatase activity. Biol. Reprod. 76, 628– 638. Swain, J.E., Ding, J., Wu, J., Smith, G.D., 2008. Regulation of spindle and chromatin dynamics during early and late stages of oocyte maturation by aurora kinases. Mol. Hum. Reprod. 14, 291–299. Vega, F.M., Sevilla, A., Lazo, P.A., 2004. p53 Stabilization and accumulation induced by human vaccinia-related kinase 1. Mol. Cell. Biol. 24, 10366–10380. Wang, Q., Wang, C.M., Ai, J.S., Xiong, B., Yin, S., Hou, Y., Chen, D.Y., Schatten, H., Sun, Q.Y., 2006. Histone phosphorylation and pericentromeric histone modifications in oocyte meiosis. Cell Cycle 5, 1974–1982. Waters, K., Yang, A.Z., Reinke, V., 2010. Genome-wide analysis of germ cell proliferation in C. elegans identifies VRK-1 as a key regulator of CEP-1/p53. Dev. Biol. 344, 1011–1025. Wiebe, M.S., Nichols, R.J., Molitor, T.P., Lindgren, J.K., Traktman, P., 2010. Mice deficient in the serine/threonine protein kinase VRK1 are infertile due to a progressive loss of spermatogonia. Biol. Reprod. 82, 182–193. Yin, S., Wang, Q., Liu, J.H., Ai, J.S., Liang, C.G., Hou, Y., Chen, D.Y., Schatten, H., Sun, Q.Y., 2006. Bub1 prevents chromosome misalignment and precocious anaphase during mouse oocyte meiosis. Cell Cycle 5, 2130–2137.