Project Summary
This submission represents a collaborative application from Maureen Murphy PhD at the Wistar Institute in
Philadelphia and Kara Maxwell MD/PhD from the University of Pennsylvania Perelman School of Medicine.
The goal is to study the function of a novel p53 mutation that was first discovered in several highly cancer-
prone families seen in Dr. Maxwell’s clinic at the Hospital of the University of Pennsylvania. This variant,
G334R in p53 protein (G331R in mouse p53) was found in nine different families with multiple cancers in
multiple generations, including cousins with pediatric adrenal tumors; the latter are a hallmark of cancer-
predisposing mutations in TP53. Moreover, we find that several tumors from these patients show loss of
heterozygosity for p53. We show that the G334R variant is impaired for oligomerization, and for the induction
of a small subset of p53 target genes, several of which are themselves tumor suppressors (PCLO, PLXNB3,
and others).
Our functional analyses of the G334R variant suggest that this is not a traditional Li Fraumeni mutant; that is,
a mutation that completely inactivates p53 function. Rather, our functional data are most consistent with
G334R being a p53 “hypomorph”, or a genetic variant that shows impaired, but not completely inactive,
function. Specifically, 1) G334R possesses some ability to suppress colony formation in tumor cells, though
it shows less ability than wild type p53; 2) G334R is fully capable of transactivating the overwhelming majority
of p53-regulated genes, but is defective in the transactivation of approximately two dozen p53 target genes;
3) the peak incidence for breast cancer in G334R individuals is between the ages of 35-55, while most cases
of Li Fraumeni occur between the ages of 20-40, and sporadic breast cancer peaks between the ages of 55-
75.
The overarching goal of this proposal is to identify the mechanism(s) whereby G334R is impaired for function.
In a completely unexpected finding, we show that there are some activities of the G334R hypomorph that are
shared with two other cancer-predisposing p53 hypomorphs, P47S and Y107H. These activities include a
heightened propensity to misfold into a conformation that is specific for mutant p53, as well as increased
sensitivity to glutaminase inhibitors. We also show that using RNA sequencing data and machine learning
approaches, we have identified a 42-gene signature that is predictive of a p53 hypomorph, and can distinguish
a p53 hypomorph from WT p53 or a benign variant with 100% accuracy. The proposed research will help us
better understand tumor suppression by p53, and to identify other carriers of hypomorphic genetic variants of
p53.