Tissue-specific modifier alleles determine Mertk loss-of-function traits

Reviewer #1 (Recommendations for the authors):

As mentioned in the public review, TYRO3 protein levels appear to be elevated in Mertk-/- v2 or Mertk-/- v3 compared to Mertk-/- v1 (readily notable in the WB presented Figure 5a) although no significant changes are reported at mRNA. This point should be re-evaluated by the authors and discussed in their conclusions as a compensatory increase in TYRO3 may, at least in part, contribute to masking some of the Mertk-/- v2 or Mertk-/- v3 phenotypes, especially given previous work, demonstrating that even hypomorphic expression of TYRO3 in Tyro3 B6/129 can suppress the phenotypes of Mertk-/- v1 loss-of-function. WB quantification of Figure 5a may help solve this apparent discrepancy between the data presented in the figure and the main text in which it is stated that " TYRO3 levels in these newly generated mice are comparable to BL6".

We sincerely apologize for the confusion. Quantitation of 5 independent immunoblots for TYRO3 from Mertk ^-/-V2 and Mertk ^-/-V3 RPEs failed to show statistically significant differences. These results are consistent with the mRNA expression data, which also failed to reveal statistically significant differences in Tyro3 expression between WT, Mertk ^-/-V2 and Mertk ^-/-V3 RPEs. Only Mertk ^-/-V1 RPEs showed significantly less Tyro3 mRNA and protein compared to WT.

Can the authors provide additional data as to the cellular mediators of the protective phenotype observed in Mertk-/-v1? What is the status of these functionally relevant cell types in the newly generated Mertk deficiency models, Mertk-/- v2, Mertk-/- v3 at steady state and upon tumor challenge?

This is an excellent question. Understanding the cellular and molecular basis of anti-tumor protection in Mertk ^-/-V1 would be highly significant. We have been indeed pursuing this remarkably important but extremely complex question. So far, we have been able to determine that the anti-tumor resistance is entirely dependent upon T cells. Depletion of CD8⁺ or CD4⁺ T cells completely reverts this anti-tumor phenotype in Mertk ^-/-V1. Furthermore, we have performed extensive bone-marrow chimera experiments followed by single cell RNA sequencing to individually compare and contrast the hematopoietic and stromal compartments of Mertk ^-/-V1 with WT B6 in tumor implanted mice, as well as performed linkage mapping to investigate the Mendelian inheritance of the element that determines the anti-tumor immune phenotype. We are currently in the process of preparing a separate manuscript to describe these results. We hope that the Reviewer will agree that these findings are beyond the scope of this manuscript.