Disrupting Cyclin Dependent Kinase 1 in Spermatocytes Causes Late Meiotic Arrest and Infertility in Mice

Abstract

While cyclin dependent kinase 1 (CDK1) has a critical role in controlling resumption of meiosis in oocytes, its role has not been investigated directly in spermatocytes. Unique aspects of male meiosis led us to hypothesize that its role is different in male meiosis than in female meiosis. We generated a conditional knockout (cKO) of the Cdk1 gene in mouse spermatocytes to test this hypothesis. We found that CDK1-null spermatocytes undergo synapsis, chiasmata formation, and desynapsis as is seen in oocytes. Additionally, CDK1-null spermatocytes relocalize SYCP3 to centromeric foci, express H3pSer10, and initiate chromosome condensation. However, CDK1-null spermatocytes fail to form condensed bivalent chromosomes in prophase of meiosis I and instead are arrested at prometaphase. Thus, CDK1 has an essential role in male meiosis that is consistent with what is known about the role of CDK1 in female meiosis, where it is required for formation of condensed bivalent metaphase chromosomes and progression to the first meiotic division. We found that cKO spermatocytes formed fully condensed bivalent chromosomes in the presence of okadaic acid, suggesting that cKO chromosomes are competent to condense, although they do not do so in vivo. Additionally, arrested cKO spermatocytes exhibited irregular cell shape, irregular large nuclei, and large distinctive nucleoli. These cells persist in the seminiferous epithelium through the next seminiferous epithelial cycle with a lack of stage XII checkpoint-associated cell death. This indicates that CDK1 is required upstream of a checkpoint-associated cell death as well as meiotic metaphase progression in mouse spermatocytes.

Keywords: CDK1; cell cycle; chromosome condensation; male meiosis; meiotic arrest; metaphase-promoting factor; okadaic acid; synaptonemal complex.

FIG. 1

Targeted deletion of Cdk1 in mouse spermatocytes. A) CDK1 levels in whole testis lysates of Cdk1 cKO mice are reduced compared to WT or HET testis lysates. B) The deletion specifically from meiotic, but not mitotic, male germ cells was confirmed by IHC and ISH. In WT seminiferous tubules, CDK1 localizes to mitotic and meiotic germ cells in the basal half of the seminiferous epithelium (Ba, brown labeling). In the cKO seminiferous epithelium, however, only mitotically dividing cells exhibit CDK1 localization (Bb, arrows). Bars = 50 μm. C) The fertility of HET mice is indicated by the average number of pups per litter compared to the cKO, which was infertile. D) Body and organ weights for WT, HET, and cKO males are represented in bar graph form. Error bars indicate SEM; asterisk (*) indicates statistically significant difference.

FIG. 2

Cdk1 cKO testes lacks postmeiotic germ cells. Periodic acid Schiff-hematoxylin (PAS-H)-stained tissue sections. A) WT testis sections with stages III and VIII tubules shown. B) The cKO testis section with three tubules shown, and approximate tubule stages indicated (II–V, XII–I, and VII–VIII). WT spermatocytes (C) and cKO spermatocytes (D) (arrow = PAS-positive Golgi). WT spermatids in stage 4 (E) and stage 6 (G) (arrowhead = steps 4 and 6 spermatids, respectively). F and H) Abnormal cKO cells from luminal portion of tubules. PAS staining is indicated in the cytoplasm (arrows) of cells in a tubule stage between II and V (F) and multinucleated giant cells (H). I) WT caput epididymal section containing sperm. J) The cKO caput epididymal section containing rounded cells and debris but lacking sperm. Bars = 40 μm in A, B, I, and J, and 10 μm in C–H. K) Sperm counts in WT, HET, and cKO mice. Error bars indicate SEM; asterisk (*) indicates statistically significant difference.

FIG. 3

Normal prophase I progression through diplonema in Cdk1 cKO testis with extrachromosomal SYCP3 aggregation at pachynema. SYCP3 (red) and SYCP1 (green) labeling of WT (A, E, G) and cKO (B, D, F, H) spermatocyte spreads. Fully synapsed chromosomes (colocalized SYCP3 and SYCP1 appear yellow) in WT (A) and cKO (B) indicative of pachynema. SYCP3-positive aggregates were also seen outside of the SC (D, arrows) in a significantly greater number (P = 0.024) of cKO pachytene-stage spreads compared to WT (C). Normal diplotene phase chromosome spreads with desynapsing chromosomes were observed in both WT (E) and cKO (F). SYCP1 was absent and SYCP3 was localized to confined foci, consistent with prometaphase, in both WT (G) and cKO (H) chromosome spread preparations. In addition to typical bright SYCP3 foci, numerous dim foci were observed in the cKO (H, crossed arrows). I) Bar graph of the number of bright, dim, and total SYCP3 foci in prometaphase spermatocyte spreads with average foci number included from at least 10 spermatocytes for three biological replicates in WT and cKO. J) Scatter plots of all the individual spermatocyte SYCP3 foci counts for 52 WT and 52 cKO spermatocyte spreads with mean values indicated with a bar. Asterisk (*) indicates statistically significant difference. Bars in D and H = 10 μm and are representative of all panels.

FIG. 4

Cdk1 cKO spermatocyte chromosomes initiate condensation and fully condense with OA treatment. HET (A and B) and cKO (C and D) IHC on testis tissue sections for H3pS10 (brown labeling). Localization to late prophase I/prometaphase spermatocytes is observed in both HET and cKO testis sections (D, diplotene phase; PM, prometaphase) whereas H3pS10 localization in metaphase and later meiotic cells was seen only in the HET (M, metaphase; A, anaphase). E and F) Air-dried chromosome preparations from WT males showing MI spreads with fully condensed bivalent chromosomes. G and H) Air-dried chromosome preparations from cKO males contained only partially condensed chromosomes. I and J) Air-dried chromosome preparations from germ cells pretreated with OA from both WT (I) and cKO (J) mice both contained MI bivalent chromosomes. Bars in A–D = 50 μm and in E–J = 5 μm.

FIG. 5

Increased TUNEL-positive cells in Cdk1 cKO testes. A and B) TUNEL staining of testis tubule sections reveals an increased number of apoptotic cells in the luminal half of the seminiferous epithelium in the cKO (B) compared to HET mice (A). Arrows indicate some TUNEL-positive cells; red hashed lines indicate tubule basal surface/perimeter. Bars = 20 μm.

FIG. 6

Germ cell populations in stages of cKO tubules indicate prolonged survival of meiotic arrested germ cells. A) Immunofluorescence localization of H2AFXγ (yellow) and SYCP3 (red), along with DAPI staining (blue) was used to group cKO seminiferous tubules into stage-groups and compared to WT. Arrows indicate H2AFXγ in pachytene-stage spermatocytes exhibiting X-Y body localization. Arrowheads indicate weak H2AFXγ chromatin localization in stage VII–VIII preleptotene-stage spermatocytes, strong labeling in stage IX–X leptotene-stage spermatocytes, and patchy labeling in stage XI–XII–I zygotene-stage and zygotene/pachytene-stage transitioning cells, characteristic of H2AFXγ relocalization to the forming XY body. Hashed lines in tubules of testis sections indicate the separation between spermatocytes in the basal portion of the seminiferous epithelium and the spermatids (WT) or abnormal arrested cells (cKO, crossed arrows) in the luminal portion. Cells contained in WT and cKO epididymis are also shown with abnormal cKO cells indicated (crossed arrows). Bars in the upper left and upper right panels are 10 μm and representative for all testis and epididymis panels. B) A bar graph summarizing the percent of seminiferous tubules containing TUNEL-positive cells in seminiferous tubules from stage groups II–VI, VII–VIII, IX–X, XI–I, and combined averages in testis sections from WT, HET, and cKO mice. C) A bar graph summarizing the number of TUNEL-positive cells per tubule containing such cells. Data for testis sections from each genotype and stage-group are indicated. Error bars represent SEM, asterisk (*) indicates statistically significant difference between genotypes at a given stage, # indicates statistically significant difference between stages for the given genotype. All indicated P-value calculations include Tukey multiple testing corrections.