In mice, the Ter mutation causes primordial germ cell (PGC) loss in all genetic backgrounds. Ter is also a potent modifier of spontaneous testicular germ cell tumour (TGCT) susceptibility in the 129 family of inbred strains, and markedly increases TGCT incidence in 129-Ter/Ter males. In 129-Ter/Ter mice, some of the remaining PGCs transform into undifferentiated pluripotent embryonal carcinoma cells, and after birth differentiate into various cells and tissues that compose TGCTs. Here, we report the positional cloning of Ter, revealing a point mutation that introduces a termination codon in the mouse orthologue (Dnd1) of the zebrafish dead end (dnd) gene. PGC deficiency is corrected both with bacterial artificial chromosomes that contain Dnd1 and with a Dnd1-encoding transgene. Dnd1 is expressed in fetal gonads during the critical period when TGCTs originate. DND1 has an RNA recognition motif and is most similar to the apobec complementation factor, a component of the cytidine to uridine RNA-editing complex. These results suggest that Ter may adversely affect essential aspects of RNA biology during PGC development. DND1 is the first protein known to have an RNA recognition motif directly implicated as a heritable cause of spontaneous tumorigenesis. TGCT development in the 129-Ter mouse strain models paediatric TGCT in humans. This work will have important implications for our understanding of the genetic control of TGCT pathogenesis and PGC biology.

The agouti-yellow (A(y)) deletion is the only genetic modifier known to suppress testicular germ cell tumor (TGCT) susceptibility in mice or humans. The A(y) mutation deletes Raly and Eif2s2, and induces the ectopic expression of agouti, all of which are potential TGCT-modifying mutations. Here we report that the reduced TGCT incidence of heterozygous A(y) males and the recessive embryonic lethality of A(y) are caused by the deletion of Eif2s2, the beta subunit of translation initiation factor eIF2. We found that the incidence of affected males was reduced 2-fold in mice that were partially deficient for Eif2s2 and that embryonic lethality occurred near the time of implantation in mice that were fully deficient for Eif2s2. In contrast, neither reduced expression of Raly in gene-trap mice nor ectopic expression of agouti in transgenic or viable-yellow (A(vy)) mutants affected TGCT incidence or embryonic viability. In addition, we provide evidence that partial deficiency of Eif2s2 attenuated germ cell proliferation and differentiation, both of which are important to TGCT formation. These results show that germ cell development and TGCT pathogenesis are sensitive to the availability of the eIF2 translation initiation complex and to changes in the rate of translation.

The genetic basis for susceptibility to testicular germ cell tumors (TGCTs) has been remarkably elusive. Although TGCTs are the most common cancer in young men and have an unusually strong familial risk, only one low-frequency susceptibility gene has been identified for this highly multigenic trait. In tests to determine whether pairs of genetic variants act epistatically to modulate susceptibility in the 129/Sv mouse model of spontaneous TGCTs, we discovered an unusual mode of inheritance that involved interactions between different genes in different generations. Any of six genetic variants, in either the female or male parent interacted with the Dnd1(Ter) mutation in male offspring to significantly increase both the frequency of affected Ter/+ males and the proportion of bilateral cases. Trans-generational epistasis is a novel mode of epigenetic inheritance that could account for the difficulty of finding TGCT susceptibility genes in humans and might represent a mechanism for transmitting information about genetic and environmental conditions from parents to offspring through the germline.

Testicular germ cell tumors (TGCTs) are the most common solid cancers affecting young men. Although the evidence for genetic predisposition to TGCTs in humans is compelling, the genetic control of susceptibility is poorly understood. The 129S1/SvImJ (129/Sv) inbred strain of mice is an excellent model for studying TGCT susceptibility. We previously reported a new mouse strain, the 129.MOLF-Chr19 chromosome substitution strain, which develops spontaneous TGCTs at a high frequency (70-80%) as compared with the much lower rate in the 129/Sv strain (5%). To characterize the genetic control of TGCT susceptibility, we created a panel of single- and double-congenic strains derived from 129.MOLF-Chr19. The frequency of TGCTs in these strains suggests that several genes with additive and epistatic effects located at distinct sites on chromosome 19 control susceptibility. However, an alternative interpretation involving epigenesis is based on a striking correlation between TGCT frequency and the length of the MOLF-derived congenic segment, regardless of their chromosomal location on Chr 19 in each congenic strain. We also show that bilateral TGCT cases result from the coincidental co-occurrence of unilateral TGCTs rather than from the action of distinct genes that control susceptibility to bilateral versus unilateral TGCT cases. Finally, we propose that these TGCTs result from disrupted testicular and spermatogenic developmental programs.

Susceptibility to spontaneous testicular germ cell tumors (TGCTs), a common cancer affecting young men, shows unusual genetic complexity. Despite remarkable progress in the genetics analysis of susceptibility to many cancers, TGCT susceptibility genes have not yet been identified. Various mutations that are inherited as Mendelian traits in laboratory mice affect susceptibility to spontaneous TGCTs on the 129/Sv inbred genetic background. We compared the frequency of spontaneous TGCTs in single-and doublemutant mice to identify combinations that show evidence of enhancer or suppressor effects. The lowerthan-expected TGCT frequencies in mice with partial deficiencies of TRP53 and MGF-SLJ and in 129.MOLFChr19 (M19) consomic mice that were heterozygous for the A y mutation suggest that either these genes complement each other to restore normal functionality in TGCT stem cells or together these genes activate mechanisms that suppress incipient TGCTs. By contrast, the higher-than-expected TGCT frequencies in Mgf Sl-J -M19 compound heterozygous mice suggest that these mutations exacerbate each other's effects. Together, these results provide clues to the genetic and molecular basis for susceptibility to TGCTs in mice and perhaps in humans.

Germ cell tumor development in humans has been proposed to be part of testicular dysgenesis syndrome (TDS), which manifests as undescended testes, sterility, hypospadias, and, in extreme cases, as germ cell tumors. Males of the Ter mouse strain show interesting parallels to TDS because they either lack germ cells and are sterile or develop testicular germ cell tumors. We found that these defects in Ter mice are due to mutational inactivation of the Dead-end (Dnd1) gene. Here we report that chromosome X modulates germ cell tumor development in Ter mice. We tested whether the X or the Y chromosome influences tumor incidence. We used chromosome substitution strains to generate two new mouse strains: 129-Ter/Ter that carry either a C57BL/6J (B6)-derived chromosome (Chr) X or Y. We found that Ter/Ter males with B6-Chr X, but not B6-Chr Y, showed a significant shift in propensity from testicular tumor development to sterile testes phenotype. Thus, our studies provide unambiguous evidence that genetic factors from Chr X modulate the incidence of germ cell tumors in mice with inactivated Dnd1.

Susceptibility to spontaneous testicular germ cell tumors (TGCTs), a common cancer affecting young men, shows unusual genetic complexity. Despite remarkable progress in the genetics analysis of susceptibility to many cancers, TGCT susceptibility genes have not yet been identified. Various mutations that are inherited as Mendelian traits in laboratory mice affect susceptibility to spontaneous TGCTs on the 129/Sv inbred genetic background. We compared the frequency of spontaneous TGCTs in single- and double-mutant mice to identify combinations that show evidence of enhancer or suppressor effects. The lower-than-expected TGCT frequencies in mice with partial deficiencies of TRP53 and MGF-SLJ and in 129.MOLF-Chr19 (M19) consomic mice that were heterozygous for the Ay mutation suggest that either these genes complement each other to restore normal functionality in TGCT stem cells or together these genes activate mechanisms that suppress incipient TGCTs. By contrast, the higher-than-expected TGCT frequencies in Mgf Sl-J-M19 compound heterozygous mice suggest that these mutations exacerbate each other’s effects. Together, these results provide clues to the genetic and molecular basis for susceptibility to TGCTs in mice and perhaps in humans.