Both anti-inflammatory and antiviral properties of novel drug candidate ABX464 are mediated by modulation of RNA splicing

Abstract

ABX464 is a first-in-class, clinical-stage, small molecule for oral administration that has shown strong anti-inflammatory effects in the DSS-model for inflammatory bowel disease (IBD) and also prevents replication of the HIV virus. ABX464 which binds to cap binding complex (CBC) has demonstrated safety and efficacy in a phase 2a proof-of-concept clinical trial in patients with Ulcerative colitis. Previously, with limited technologies, it was not possible to quantify the effect of ABX464 on viral and cellular RNA biogenesis. Here, using RNA CaptureSeq and deep sequencing, we report that ABX464 enhances the splicing of HIV RNA in infected PBMCs from six healthy individuals and also the expression and splicing of a single long noncoding RNA to generate the anti-inflammatory miR-124 both ex vivo and in HIV patients. While ABX464 has no effect on pre-mRNA splicing of cellular genes, depletion of CBC complex by RNAi leads to accumulation of intron retention transcripts. These results imply that ABX464 did not inhibit the function of CBC in splicing but rather strengthens it under pathological condition like inflammation and HIV infection. The specific dual ability of ABX464 to generate both anti-inflammatory miR-124 and spliced viral RNA may have applicability for the treatment of both inflammatory diseases and HIV infection.

Figure 1

Analysis of HIV splicing following ABX464 treatment. (A) Organization of the HIV-1 genome and different mRNA splicing products. The 5′ ss (D1–D4) and 3′ ss (A1–A7) are indicated. ORFs of the coding exons of each mRNA product are indicated with a different color code indicating the corresponding encoded proteins of the HIV genome. The non-coding exons are boxed in gray. (B) Inhibition of HIV-1 replication measured by p24 production in PBMCs from six donors. (C) Quantification of HIV splicing events using RNA CaptureSeq in PBMCs from six different donors that were infected and untreated (DMSO) or treated with ABX464 for 6 days (464). Counts of spliced and unspliced contigs are shown. The different splicing products are colored as in A. (D) Sequence of the new viral RNA generated by splicing.

Figure 2

ABX464 has no global effect on cellular splicing. (A) The effect of ABX464 on infected and uninfected CD4+ T cells was tested by a high-throughput RNAseq approach. Sixteen libraries were constructed using 4 conditions: uninfected (DMSO_NI), uninfected treated with ABX464 (ABX464_NI), infected (DMSO_I) and infected treated with ABX464 (464_I), corresponding to 4 donors. Approximately 38 million reads (more than 50% of the total raw reads) were aligned to exons of the human genome sequence in each of the samples. (B) Multidimensional scaling analysis (MDS) was used to interpret major trends in the data. (C) Alternative splicing (AS) events of cellular genes were classified into five major groups (left panel): Alternative 5′ splice site (A5SS, orange), Alternative 3′ splice site (A3SS, blue), Skipped exon (SE, gray), Mutually exclusive exons (MXE, gray) and Retained exon (RI, yellow). AS event counts comparing infected vs uninfected samples (DMSO_I vs DMSO_NI), uninfected vs uninfected treated by ABX464 (DMSO_NI vs 464_NI), infected vs infected treated by ABX464 (DMSO_I vs 464_I) and after 50% depletion CBC in IPS (IPS depletion of CBC by 50%) (right panel). (D) By comparing exon coverage reads of a common highly expressed gene (B2M) between the ABX464 and DMSO conditions in the 4 donors, we confirmed that ABX464 did not increase splicing events in B2M. (E) Volcano plot of DMSO_I vs DMSO_NI (upper panel), DMSO_NI vs 464_NI (middle panel), DMSO_I vs 464_I (lower panel). The gene expression variation generated by ABX464 treatment was very low in infected (6 downregulated genes) and uninfected (6 downregulated genes) samples.

Figure 3

ABX464 upregulates a single microRNA, the anti-inflammatory miR-124. (A) Microarray analysis of small RNAs from PBMCs from 6 donors. PBMC were infected with the YU-2 strain (I) or uninfected (NI) and treated with ABX464 or untreated (DMSO). Volcano plots show that infection leads to large variations in small noncoding RNAs (left panel), whereas ABX464 induced a reproducible upregulation of a single microRNA, miR-124, in infected and uninfected cells (right and middle panels, respectively). (B) Quantification of miR-124 expression using TaqMan Low Density Array technology in CD4+ T cells under the same conditions as in A. (C) Expression of miR-124 measured by qPCR in PBMCs, purified CD4+ and CD8 T cells, and macrophages treated with ABX464 in comparison to untreated cells (DMSO, fold change). (D) Expression of miR-124 in PBMCs treated with the antiretrovirals ABX464, ABX530, maraviroc, efavirenz, darunavir and AZT compared to untreated cells (DMSO, fold change). (E) Quantification of miR-124 in rectal biopsies of healthy participants (n = 10) and HIV patients undergoing ART at days 1 and 28 of treatment with ABX464 (n = 9). Individual graphs show the results for each patient in comparison with the results for healthy participants.

Figure 4

miR-124 upregulation by ABX464 originates from splicing of a long noncoding RNA at the miR-124-1 locus. (A) There are three genes encoding miR-124, miR-124-1, miR-124-2, and miR-124-3, located on chromosomes 8 and 20 in the human genome. (B) We employed a targeted RNA capture and sequencing strategy to determine which gene was induced by ABX464. In both infected and uninfected cells, treatment with ABX464 leads to upregulation of miR-124 from the miR.124.1 locus, whereas a control locus, miR-429, is not affected. (C) Locus miR-124-1 contains a long noncoding RNA (LncRNA 0599-205) whose splicing is stimulated by ABX464. (D) Counts of the reads at splice junctions (J1, J2, J3 and J4), exon-exon (J5 and J6) and the miR-124 75 bp region (miR-124) quantified by RNA CaptureSeq in PBMCs treated with ABX464.

Figure 5

Splicing of LncRNA 0599-205 is required for the production of miR-124. (A) Schematic representation of LncRNA 0599-205 precursor and primers used to amplify different derived RNAs. (B) Quantification of spliced and unspliced LncRNA 0599-205 in the presence or absence of ABX464. (C) Quantification of the expression of miR-124 following transfection of a wild type and splicing mutant of lncRNA0599-205 plasmids in HeLa cells in the presence or absence of ABX464. (D) Quantification of total wild type and splicing mutant lncRNA0599-205 in the presence or absence of ABX464.