Systematic analysis of the phosphoproteome and kinase-substrate networks in the mouse testis

Abstract

Spermatogenesis is a complex process closely associated with the phosphorylation-orchestrated cell cycle. Elucidating the phosphorylation-based regulations should advance our understanding of the underlying molecular mechanisms. Here we present an integrative study of phosphorylation events in the testis. Large-scale phosphoproteome profiling in the adult mouse testis identified 17,829 phosphorylation sites in 3955 phosphoproteins. Although only approximately half of the phosphorylation sites enriched by IMAC were also captured by TiO2, both the phosphoprotein data sets identified by the two methods significantly enriched the functional annotation of spermatogenesis. Thus, the phosphoproteome profiled in this study is a highly useful snapshot of the phosphorylation events in spermatogenesis. To further understand phosphoregulation in the testis, the site-specific kinase-substrate relations were computationally predicted for reconstructing kinase-substrate phosphorylation networks. A core sub-kinase-substrate phosphorylation networks among the spermatogenesis-related proteins was retrieved and analyzed to explore the phosphoregulation during spermatogenesis. Moreover, network-based analyses demonstrated that a number of protein kinases such as MAPKs, CDK2, and CDC2 with statistically more site-specific kinase-substrate relations might have significantly higher activities and play an essential role in spermatogenesis, and the predictions were consistent with previous studies on the regulatory roles of these kinases. In particular, the analyses proposed that the activities of POLO-like kinases (PLKs) might be dramatically higher, while the prediction was experimentally validated by detecting and comparing the phosphorylation levels of pT210, an indicator of PLK1 activation, in testis and other tissues. Further experiments showed that the inhibition of POLO-like kinases decreases cell proliferation by inducing G2/M cell cycle arrest. Taken together, this systematic study provides a global landscape of phosphoregulation in the testis, and should prove to be of value in future studies of spermatogenesis.

Fig. 1.

Summarization of the phosphorylation sites identified in this study. A, The distribution of different residue types (S/T/Y) for the phosphorylation sites identified by IMAC or TiO₂ B, in this study. C, Summarization of the overlap of the phosphorylation sites identified by IMAC or TiO₂ for different residue types. D, Summarization of the overlap of the phosphorylation sites identified by IMAC and TiO₂ in this study and in Gygi's data set.

Fig. 2.

Statistical analysis of the functional annotations for the phosphorylation sites identified in this study. The top 15 enriched biological processes for the total mouse testis phosphoproteome identified in this study A, or phosphorylated proteins identified by IMAC B, or TiO₂ C, or phosphorylated testis proteins identified from the Gygi's data set D. E-ratio: the proportion in the specific data set divided by that in the mouse proteome.

Fig. 3.

Network analysis of phosphorylation regulation in the mouse testis phosphoproteome. A, The network constructed from the total mouse testis phosphoproteome identified in this study. B, The network constructed from the phosphorylated spermatogenesis-related proteins. Different colored proteins were from different data sets. “IMAC,” “TiO_2,” and “Kinase” in the colored box represent that the protein is a phosphoprotein identified by IMAC, TiO₂, and a kinase, respectively, whereas “/” means both.

Fig. 4.

The kinase activity analyses for the phosphoproteomes. The top15 significantly high-activity kinases for the total mouse testis phosphoproteome identified in this study A, or phosphorylated proteins identified by IMAC B, or TiO₂ C, or phosphorylated testis proteins identified from the Gygi's data set D, are presented (p value <0.05). E-ratio: the proportion in the specific data set divided by that in the mouse phosphoproteome atlas.

Fig. 5.

The phosphorylation network for the PLKs. A, The phosphorylation network among the PLKs and the substrates of the PLKs. B, The phosphorylation network among the PLKs and the PLKs phosphorylated spermatogenesis-related proteins. The phosphorylation network annotated with identified sites for PLK1 C, PLK2 D, PLK3 E, PLK4 F, and PLK5 G. Different colored proteins were from different data sets. “IMAC,” “TiO_2.” and “Kinase” in the colored box represent that the protein is a phosphoprotein identified by IMAC, TiO₂, and a kinase, respectively, whereas “/” means both.

Fig. 6.

The phosphorylation level of PLK1 and inhibition assay of the PLKs in GC2 cells. A, The expression and pT210 levels of PLK1 in testes and mixed tissues of different developmental stages were analyzed by Western blot, with β actin as a loading control. Mixed tissues is the pooling of equal amount of brain, brown fat, heart, liver, lung, kidney, pancreas, and spleen from 3-week or 8-week mice. The distribution patterns of GC2 cells in different phases of the cell cycle were compared after treatment with BI2536 and/or OA for 2 h B, 4 h C, and 6 h D.