Adar2 restricts line1 retrotransposition

Adenosine deaminases that act on RNA (ADARs) catalyse the hydrolytic deamination of Adenosine (A) to Inosine (I) within double-stranded RNA substrates. In mammals three ADAR enzymes are present: ADAR1, ADAR2 and ADAR3. ADAR1 and ADAR2 are expressed in many tissues and they are catalytically active; ADAR3 is expressed only in the brain and its catalytic activity has never been demonstrated. During my PhD thesis I focused my studies on the deaminase ADAR2. Two different isoforms of ADAR2 are expressed: ADAR2a of 701 amino acids in length, and ADAR2R of 750 amino acids in length, generated by alternative splicing. Both the isoforms possess two copies of dsRBDs and the C-terminus catalytically active domain, that contains a molecule of inositol hesakifosfate; moreover, ADAR2R has an N-terminal Rich-Arginine R domain. It was reported the role of ADAR2 as a positive regulator in the replication of HIV-1. In fact, in vitro experiments, under ADAR2 overexpression conditions, demonstrated that ADAR2 causes an increase in the release of HIV-1 virions; the deaminase edits some Adenosines localized in the 5′UTR of the HIV-1 transcripts. Moreover, by an editing-independent mechanism the overexpression of ADAR2 leads to an increase in the intracellular accumulation of the viral protein p24. The aim of my PhD research project was to identify protein factors (cellular and/or viral) that may affect the proviral activity of ADAR2. Starting from the lysate of HIV-1 expressing cells by using a Dual-Tag Affinity purification system followed by Mass Spectrometry analysis, ADAR2-binding proteins were identified. This approach allowed the identification of 10 putative non-ribosomal ADAR2-interacting proteins. Interestingly almost half of them were known to be involved in LINE-1 (Long Interspersed element-1) life cycle: DNA Topoisomerase I (TOP1), Nucleolin (NCL), ADAR1 and PSF. LINE-1 is the most abundant autonomous retrotransposon and, by using a copy-and-paste mechanism, is involved in shaping the human genome. The full-length LINE-1 transcript is about 6 kilobases (kb) and contains two nonoverlapping open reading frames, ORF1 and ORF2, that encode for the ORF1p and ORF2p proteins, essential in LINE-1 retrotransposition. ORF1p is a 40 KDa RNA binding protein with nucleic acid chaperone activity, while ORF2p is a 150 KDa protein that contains both endonuclease (EN) and reverse transcriptase (RT) activities. Although most of the LINE-1 copies are truncated and functionally inactive, about 80-100 copies are competent for the retrotransposition in the human genome. The results from Dual-Tag Affinity purification and Mass Spectrometry approach suggested a possible link between ADAR2 and LINE-1. To confirm this hypothesis, first we validated the Mass Spectrometry results by co-immunoprecipitation experiments (co-IP) followed by Western Blot (WB) analysis, in HIV-1 expressing cells. We validated the interaction between ADAR2 and some proteins, SFPQ/PSF, NCL, hnRNP-C1/C2, P54/NONO and ADAR1, identified by Mass Spectrometry. Furthermore, by a treatment with a mix of RNase (RNase A and RNase V1) prior the co-IP experiments, we demonstrated that the interactions between ADAR2 and the putative interacting factors are mostly RNA-dependent; moreover, these interactions were further validated in the absence of viral replication. Next step was to investigate whether ADAR2 could be involved in the regulation of LINE-1 activity. To this aim, in vitro cell culture LINE-1 retrotransposition assay (dual-luciferase assay) was employed to determine the effect of ADAR2-V5 overexpression on LINE-1 retrotransposition. We demonstrated that the overexpression of ADAR2-V5 causes a decrease in LINE-1 retrotransposition, suggesting a novel function of ADAR2 as suppressor of LINE-1 activity. Since the overexpression of ADAR2-V5 caused a reduction in LINE-1 retrotransposition, we expected that ADAR2 depletion could cause the opposite effect. For this reason, we employed the dual-luciferase assay in conditions of partial depletion of the endogenous ADAR2, obtained by taking advantage from the CRISPR-Cas9 technology. The partial depletion of the endogenous ADAR2, as expected, determines an increase of retrotransposition efficiency. To further investigate the mechanism by which ADAR2 controls LINE-1 retrotransposition we performed IP and RIP assays, to test whether ADAR2-V5 could bind the basal components of LINE-1 RNPs. By using this approach, we demonstrated that ADAR2-V5 interacts with LINE1 mRNA and with the protein ORF1p-T7. To further confirm the association between ADAR2 and the basal components of the LINE-1 RNP complex, we analysed their subcellular localization by an immunofluorescence (IF) assay. The IF revealed that ADAR2-V5 forms a ring-like structure around the nucleolus, thus probably leading to a sequestration of the ORF1pT7 in this subcellular compartment. Furthermore, to investigate whether the editing activity of ADAR2 is necessary for the inhibition of LINE-1 retrotransposition, we performed an additional dual-luciferase assay, using the ADAR2-V5 E/A mutant protein, which contains a single amino acid change in its catalytic domain (Glut396 to Ala396), that makes the enzyme catalytically inactive. The result of these experiments suggests that the catalytic activity of ADAR2 is critical for the suppression of LINE-1 activity. Since the ADAR2 catalytic domain seems to be required to suppress LINE1 retrotransposition and ADAR2 interacts with LINE-1 RNA, we hypothesized that LINE-1 full-length transcript might be a target of ADAR2 editing, that in turn affects LINE-1 activity; however, sequencing of the full length of LINE-1.3 RNA failed to detect any A-to-I editing events in the ~70% of the sequence analysed.