QTL analysis of macrophages from an AKR/JxDBA/2J intercross identified the Gpnmb gene as a modifier of lysosome function

Abstract

Our prior studies found differences in the AKR/J and DBA/2J strains in regard to atherosclerosis and macrophage phenotypes including cholesterol ester loading, cholesterol efflux, and autolysosome formation. The goal of this study was to determine if there were differences in macrophage lysosome function, and if so to use quantitative trait locus (QTL) analysis to identify the causal gene. Lysosome function was measured by incubation with an exogenous double-labeled ovalbumin indicator sensitive to proteolysis. DBA/2J vs. AKR/J bone marrow macrophages had significantly decreased lysosome function. Macrophages were cultured from 120 mice derived from an AKR/JxDBA/2J F4 intercross. We measured lysosome function and performed a high density genome scan. QTL analysis yielded two genome wide significant loci on chromosomes 6 and 17, called macrophage lysosome function modifier (Mlfm) loci Mlfm1 and Mlfm2. After adjusting for Mlfm1, two additional loci were identified. Based on proximity to the Mlfm1 peak, macrophage mRNA expression differences with AKR/J >> DBA/2J, and a protein coding nonsense variant in DBA/2J, the Gpnmb gene, encoding a lysosomal membrane protein, was our top candidate. To test this candidate, Gpnmb expression was knocked down with siRNA in AKR/J macrophages; and, to express the wildtype Gpnmb in DBA/2J macrophages, we obtained a DBA/2 substrain, DBA/2J-Gpnmb+/SjJ, which was isolated from the parental strain prior to its acquiring the nonsense mutation, and subsequently back crossed to the modern DBA/2J background. Knockdown of Gpnmb in AKR/J macrophages decreased lysosome function, while restoration of the wildtype Gpnmb allele in DBA/2J macrophages increased lysosome function. However, this modifier of lysosome function was not responsible for the strain differences in macrophage cholesterol ester loading or cholesterol efflux. In conclusion, we identified the Gpnmb gene as the major modifier of lysosome function and we showed that the QTL in a dish strategy is efficient in identifying modifier genes.\n\nAuthor SummaryInbred strains of mice differ in both their genetic backgrounds as well as in many traits; and, classical mouse genetics allows the mapping of genes responsible for these traits. We identified many traits that differ between the inbred strains AKR/J and DBA/2J, including atherosclerosis susceptibility, macrophage cholesterol metabolism, and in the current study, macrophage protein degradation via an organelle called the lysosome. Using mouse genetic mapping and bioinformatics we identified a candidate gene, called Gpnmb, responsible for modifying lysosome function; and, the DBA/2J strain carries a mutation in this gene. Here we demonstrate that the Gpnmb gene is a modifier of lysosome function by either correcting this Gpnmb mutation in DBA/2J macrophages, or by knocking down Gpnmb expression in AKR/J macrophages. This study is noteworthy as the human GPNMB gene has been implicated in many diseases including cancer, kidney injury, obesity, non-alcoholic steatohepatitis, Parkinson disease, osteoarthritis, and lysosome storage disorders.