UCSF

Mouse primordial germ cells (PGCs) are specified in the posterior embryo and migrate through hindgut endoderm and dorsal mesentery to the genital ridge. Although they move individually, PGCs do not migrate synchronously and are not all successful migrants. When migration ends at E11.5, we find approximately 6% of mouse PGCs outside the genital ridge; lacking survival cues, they are poised to undergo apoptosis. What heterogeneities distinguish a successful PGC, able to colonize the gonad, from a failed migrant? Are heterogeneities shared between clonally labeled mother and daughter cells? Our previous data show that mouse PGCs cultured ex vivo mount variable responses to migration cues, suggesting PGC behavioral variation may arise intrinsically; we also showed that proliferation is regulated by the somatic niche, increasing proliferation of PGCs as they progress along the migratory route. These data indicate that migratory speed and PGC-niche interactions may influence PGC gonadal colonization. To characterize intrinsic heterogeneities in migrating mouse PGCs, we conducted single cell RNA sequencing of purified PGCs and surrounding somatic niche for E9.5, E10.5, and E11.5 migratory timepoints. For E9.5 and E10.5, we generated libraries after splitting PGCs into anterior "leading" vs posterior "lagging" populations. Upon bioinformatic integration of all libraries, we analyzed differential gene expression probing positional differences as well as age related differences. We identified an E11.5 subcluster that may represent lagging PGCs transitioning medial/midline aorta-gonad-mesonephros tissues to the gonadal ridge. This cluster expresses imprinting genes. We have shown in vivo through HCR probes that cells expressing Igf2, a top imprinting gene marker of this cluster, are localized in the extragonadal space. We also probed receptor ligand interactions in migratory PGCs, and found several known receptor-ligand pairs, like Cxcl12-Cxcr4 survival signals. Interestingly, we identified potentially novel regulators of migrating germ cells, including ephrins. Additionally, re reanalyzed some human fetal PGCs and found that Slit-Robo signaling was implicated in both mouse and human PGC-somatic cell interactions.

Additionally, we investigated whether mitochondrial heteroplasmies might contribute to migratory heterogeneities we observed. Using our single cell libraries, we were able to sequence some regions deeply enough to call variants. In general, we found that at either timepoint, anterior PGCs did not have different levels of heteroplasmy compared to posterior PGCs. However, we did observe that heteroplasmies increased from E9.5 compared to E10.5. In early human PGCs, it is known that nonsynonymous mutations decrease over time; in mouse, however, we saw that nonsynonymous mutations increased at E10.5 compared to E9.5.

Finally, using fluorescent lineage labeling, we can test whether these heterogeneities are clonal. Interestingly, cells labeled while in the hindgut still localize near their clonal family in the genital ridge. Studying transcriptional profiles in migratory PGCs across different ages is revealing signaling mechanisms associated with specific niches as well as the extent of heterogeneity between cells.