Dr. Bogunovic and his team then asked, why is this happening and what is the underlying mechanism? They performed digital droplet PCR using a reference allele (PLCG2) expressed in a monoallelic pattern. By introducing JMJD1 short non-coding RNA, which perturbs H3K27 trimethylation, they produced a biallelic clone. Likewise, using a biallelic PLCG2 clone, they were able to promote monoallelic expression by manipulating DNA methylation. Dr. Bogunovic explained that these results are somewhat misleading in that they have not fully elucidated the mechanism of MAE. For unexplained reasons, even in the same individual, some clones did not respond to the above perturbations, although others did, suggesting multiple mechanisms are involved in regulation of MAE.
MAE as a Driver of Clinical Phenotypes
Dr. Bogunovic returned to the patient from New York with the mutation in JAK1 that inspired this work. His laboratory colleague Conor Gruber, MD, PhD, was able to demonstrate that JAK1 was a gain-of-function mutation, with downstream signaling even in the absence of cytokines. Given that this particular patient had a de novo mutation, they could use this case to test for incomplete penetrance; however, they have since discovered hundreds of JAK1 gain-of-function patients, including some with incomplete penetrance.
To this point, he highlighted the pedigree of another family with gain of function mutations in JAK1. The grandfather was an asymptomatic carrier of the mutation, and the mother—the index case—was a 34-year-old woman who had been healthy until a toxoplasmosis infection at age 28. She then developed common variable immune deficiency (CVID), hyper IgM syndrome, splenomegaly and pancytopenia. This patient has a daughter who was a carrier of the variant, but also completely healthy. His team took PBMCs from the daughter and the mother and used digital droplet PCR to look for allelic bias.
They found the healthy carrier—the daughter—only had 22% of the mutant allele expressed, and the affected carrier—the mother—expressed 36% of the mutant allele. They sorted PBMCs into cell types and noted a propensity to suppress the gain of function allele, except in maternal T cells. “This was really the first example, telling us that monoallelic expression or expressional bias in heterozygous disease correlates well with disease penetrance,” Dr. Bogunovic said.
Dr. Bogunovic and his team wondered how widespread this mechanism was. Is it common? They then looked to PLCG2 mutations in a large family pedigree. They examined two cousins—patient A and patient B—with normal and low IgG production, respectively. Dr. Bogunovic wondered if monoallelic expression might be at play.
