|dc.description.abstract||Apolipoprotein B mRNA-editing enzyme-catalytic, polypeptide-like 3 (APOBEC3) enzymes are a family of single-stranded (ss)DNA cytosine deaminases that serve as a host restriction factors for retrotransposons and viruses that contain a ssDNA intermediate. While the APOBEC3 family was originally expanded to restrict endogenous retroelements, the majority of these targets are now inactivated and the family has been found to act on different targets, such as viral ssDNA as a mechanism of viral defense. Some of the family members also catalyze “off-target” deaminations in human ssDNA and have been implicated in somatic mutagenesis that can lead to cancer.
My PhD thesis research investigated the biochemical mechanisms underlying both the benefits and the risks of the A3 family of enzymes. The central hypothesis of this work is that A3 enzymes have evolved distinct biochemical mechanisms to deaminate ssDNA that differentiate A3 restriction of retroelements and retroviruses from A3-induced somatic mutagenesis. Therefore, we investigated the biochemical mechanisms the A3 enzymes utilize during restriction of Human Immunodeficiency Virus (HIV) in both deamination-dependent and deamination-independent modes, as well as the unique mechanisms employed during mutation of genomic DNA. There are seven APOBEC3 members (A-H, excluding E) and of these, four members (A3D, A3F, A3G, A3H) have been identified to restrict HIV replication in CD4 T+ cells, and currently three members (A3A, A3B and A3H haplotype I) have been implicated in somatic mutagenesis.
The APOBEC3 enzymes that restrict HIV replication function optimally in the absence of the HIV viral infectivity factor (Vif). Vif targets APOBEC3 for degradation via the proteasome in HIV infected cells. For APOBEC3s that are able to fortuitously escape Vif mediated degradation and become encapsidated, the enzymes can deaminate cytosines to form uracils in viral (-)DNA. Replication of (-)DNA to (+)DNA causes HIV-1 reverse transcriptase to incorporate adenines opposite uracils which creates C/G→T/A transition mutations. While restriction of HIV most commonly occurs through this deamination-dependent mechanism, APOBEC3s have also been identified to interfere with HIV reverse transcriptase processes in a deamination-independent manner, although this mechanism is secondary to the deamination-dependent mode. In order to restrict retroelements and viruses such as HIV, the A3 enzymes require an efficient processive ssDNA scanning mechanism that allows them to search for, and locate cytosines on ssDNA in the finite amount of time the substrate is available during formation of the double-stranded DNA provirus. To understand if this requirement was lost in A3 enzymes unable to restrict HIV, we studied A3C. Human A3C is not able to restrict HIV, in comparison to the more active gorilla and chimpanzee A3C orthologs that can restrict HIV, however an explanation for this difference was lacking. A polymorphism, S188I, in human A3C, which is found in less than 3% of the global population leads to increases enzymatic activity and HIV restriction. However, chimp and gorilla also possess the S188 allele suggesting a secondary residue responsible for increasing the activity of the A3C enzyme, in addition to residue 188. We determined that both I188 as well as N115 increases the activity of A3C by promoting dimerization of A3C. The dimerization of A3C increases its processivity on ssDNA and correlates with higher levels of mutagenesis during HIV reverse transcription, as is known for other A3 members studied. Since these amino acid changes also lie within the known interface for A3C interaction with Vif, we examined whether the different oligomerization states changed the ability of Vif to degrade A3C and found that Vif was able to induce A3C degradation regardless of its oligomerization state. Nevertheless, we determined that dimerization of A3C was able to predict the activity of the enzyme.
Similarly, it is known that A3 enzymes exhibiting reduced mutagenesis due to inefficient cytosine deamination may compensate by having an increased deamination-independent antiviral activity. A3 mediated inhibition of reverse transcriptase movement along the template may have an effect on processes such as HIV reverse transcriptase (RT) template switching and insertion fidelity, all of which would inhibit HIV replication. By examining these HIV RT processes in the presence of the A3 enzymes we were able to determine that A3F, which binds RNA and DNA with high affinity, promoted HIV RT template switching by blocking the progression of RT and forcing the RT to switch templates. Other A3s such as A3G and A3C, which bind the template with lower affinity, did not affect template switching. Interestingly, A3G decreased the fidelity of RT, causing misincorporation of nucleotides. Both of these outcomes, an increase in template switching and a decrease in polymerase fidelity, may promote virus evolution and emphasizes the importance of viral inactivation through the deamination-dependent mode of restriction.
If APOBEC3s are expressed in the wrong cell at the wrong time, they can facilitate mutagenesis of human genomic DNA and contribute to the C/G→T/A mutations evident in multiple human cancers. A3 enzymes induce mutations during processes that generate ssDNA, such as transcription and replication, and we sought to determine the biochemical features of A3 enzymes that cause somatic mutations. Interestingly, instead of enzyme processivity, which is required for efficient restriction of HIV, the determining characteristic for mutagenesis of genomic DNA was the ability of A3 enzymes to cycle rapidly between ssDNA substrates. For example, A3A, a non-processive enzyme, readily cycled between substrates and induced higher levels of mutations than A3G, a processive enzyme that is unable to cycle. In addition to substrate cycling, enzyme oligomerization also influences its ability to induce mutations in cells due to a size limitation imposed by the dynamic transcription bubble. Deamination during transcription by A3B tetramers was poor, but A3A monomers, and A3H dimers were efficient. Therefore, the biochemical properties of the enzymes, in addition to availability of ssDNA, determine whether A3s will be able to induce mutagenesis in cells.
Taken together, these data have allowed us to better understand the biochemical mechanisms behind A3 enzyme evolution that has influenced their ability to restrict HIV in hominids, their ability to manipulate retroviral polymerases, and their capacity to induce somatic mutagenesis in human genomes.||