Poster Session B   |   7:00am Expo - Hall A & C   |   Poster ID #163

Human APOBEC3B promotes tumor development in vivo including signature mutations and metastases

Program:
Academic Research
Category:
Molecular and Cellular Biology, Genetics
FDA Status:
Not Applicable
CPRIT Grant:
Cancer Site(s):
All Cancers
Authors:
Cameron Durfee
The University of Texas Health Science Center at San Antonio
Nuri Temiz
University of Minnesota Twin Cities
Rena Levin-Klein
University of Minnesota Twin Cities
Prokopios P. Argyris
The University of Texas Health Science Center at San Antonio
Joshua Proehl
The University of Texas Health Science Center at San Antonio
Anna Holzhauer
University of Minnesota Twin Cities
Zachary Seeman
University of Minnesota Twin Cities
Xingyu Liu
The University of Texas Health Science Center at San Antonio
Yu-Hsiu Lin
The University of Texas Health Science Center at San Antonio
Rachel Vogel
University of Minnesota Twin Cities
Reuben Harris
The University of Texas Health Science Center at San Antonio

Introduction

The human APOBEC3 family of enzymes is comprised of seven distinct cytosine deaminases, which play important roles in virus restriction by generating C-to-U changes in viral DNA. Although most of these enzymes preferentially deaminate TC motifs in single-stranded (ss)DNA, a number of studies have converged on APOBEC3B (A3B) as one of the predominant sources of APOBEC signature mutations in cancer. Specifically, expression of A3B triggers an abundance of APOBEC signature mutations in human cells, and gene knockouts lower the capacity of cancer cell lines to accumulate the associated C-to-T and C-to-G mutation signatures. However, a major obstacle in assessing the overall impact of the enzyme in cancer is a lack of appropriate murine models, as mice lack direct equivalents of human A3B. Here, we report a novel genetically engineered mouse model which conditionally expresses human tumor-like levels of A3B. These high levels of A3B increase the rate of tumor formation and produce a variety of mutational outcomes.

Methods

A construct comprised of a strong transcriptional CAG promoter and lox-stop-lox cassette followed by a human A3B minigene was knocked into the Rosa26 locus in C57BL/6 embryonic stem cells. These were then implanted into adult C57BL/6 mice to produce fertile CAG-LSL-A3Bi mice. By crossing these mice with CMV-Cre animals, A3B expression was induced whole-body. Resulting A3B-expressing mice were monitored for tumor formation across their lifespan in parallel with wildtype mice. Tumors were collected for downstream processing including histology and genomic sequencing. Hematoxylin and eosin staining was performed to enable tumor grading by a trained pathologist, and immunohistochemistry was done for multiple proteins including A3B. RNA and DNA sequencing were also performed on resulting tumors from both A3B-expressing and wildtype mice. Bioinformatic software programs were used to characterize the genomic landscape in these tumors, including MutationalPatterns for mutational signature analysis and PyClone-VI for clonality analysis.

Results

Mice expressing high levels of A3B develop normally but have an accelerated rate of tumor formation along with an elevated frequency of metastasis. These malignancies also recapitulate phenotypes present in A3B-hot human lesions including A3B signature mutations and increased heterogeneity. A3B signature mutations are positively correlated with other mutagenic outcomes in these tumors such as small insertion and deletion mutations and structural variations. Similarly, RNA transcript expression of key base excision repair enzymes correlates with A3B signature mutations. Experiments done in parallel with a catalytic inactive mutant of A3B reveal a decreased tendency toward tumor development.

Conclusion

These experiments reveal that A3B is capable of driving tumor formation in vivo, and that this phenotype is dependent on its signature mutations. Here, the enzyme can generate the well-characterized point mutations from the enzyme along with larger-scale mutational outcome, both of which can occur in cancer-associated genes. This mouse model will be a useful tool for studying mechanisms by which A3B generates cancers, as well as for testing novel treatments targeting the deaminase.