Characterization and function of cks2-a cyclin dependent kinase regulatory subunit-during spermatogenesis
Cell division is controlled by cyclin dependent kinases (CDKs) that are activated to phosphorylate various downstream targets. My goal was to study the CDKs and interacting proteins that control meiosis. CDK2beta, the alternatively spliced isoform of cyclin dependent kinase 2 (CDK2), may be a key meiotic regulator due to its expression at prophase of meiosis I in spermatogenesis. Therefore, we used the yeast two-hybrid system to identify binding partners of CDK2beta. The mouse homolog of CKS2 (CDC28 Kinase Subunit 2) was identified as a CDK2beta interactor after screening an adult mouse testis cDNA library. CKS2 was confirmed to bind to CDK2beta. Cks2 was more abundant in day 17 after birth and adult mice, and was only weakly detected in mutants that lack germ cells. Further, Cks2 mRNA levels were significantly higher in pachytene spermatocytes than early and late spermatids, while Cks1 was nearly undetectable. CKS2 protein was detected with both CDK2beta and cyclin A1 in germ cells during the first meiotic division. The Cks2 gene has three exons. Cks2-/- mice were generated by another group by deletion of exon 1. Cks2-/- mice are sterile due to a block in spermatogenesis which arrests at metaphase I of meiosis. I show here that the major biochemical effect in the absence of Cks2 was an increase of CCNA1 and CDK2-associated kinase activity. Surprisingly, reverse transcription-PCR analysis showed that a mutant form of Cks2 was being transcribed in testes of these mice. We also found that Cks2-/- mice produced a protein that was immunoreactive with the CKS antibody, suggesting that a translatable mRNA was present in testes of Cks2-/- mice. Cloning of the mutant Cks2 cDNA indicated that it consisted of a portion of the Cks2 coding sequence (exons 2 and 3) fused to portions of introns 1 from the Cks2 gene and the plasmid used to generate the Cks2-/- ES cells. We postulate that this cDNA is derived from a mRNA transcribed from a cryptic promoter active in the testes of Cks2-/- mice. Interestingly, the aberrant protein did not possess all of the function attributed to the wild type protein, as Cks2-/- mice were sterile. However, these data indicate that the interpretation that this phenotype is due to complete loss of Cks2 must be interpreted with care. While not definitive, these data suggest that a truncated form of Cks2 was being expressed in testes of these mice, and suggest a more complicated explanation for the observed infertility phenotype. These data indicate that CKS2 appears to act as a CDK2beta/CCNA1 repressor and that Cks2-/- mice generate a Cks2 mutant that is being transcribed and possibly translated in the testis.