Liver-specific transcriptional modules identified by genome-wide in silico analysis enable efficient gene therapy in mice and non-human primates

Abstract

The robustness and safety of liver-directed gene therapy can be substantially improved by enhancing expression of the therapeutic transgene in the liver. To achieve this, we developed a new approach of rational in silico vector design. This approach relies on a genome-wide bio-informatics strategy to identify cis-acting regulatory modules (CRMs) containing evolutionary conserved clusters of transcription factor binding site motifs that determine high tissue-specific gene expression. Incorporation of these CRMs into adeno-associated viral (AAV) and non-viral vectors enhanced gene expression in mice liver 10 to 100-fold, depending on the promoter used. Furthermore, these CRMs resulted in robust and sustained liver-specific expression of coagulation factor IX (FIX), validating their immediate therapeutic and translational relevance. Subsequent translational studies indicated that therapeutic FIX expression levels could be attained reaching 20-35% of normal levels after AAV-based liver-directed gene therapy in cynomolgus macaques. This study underscores the potential of rational vector design using computational approaches to improve their robustness and therefore allows for the use of lower and thus safer vector doses for gene therapy, while maximizing therapeutic efficacy.

Figure 1

Nucleotide sequence of the HS-CRM8 element located with the promoter of human Serpina1. This HS-CRM8 element is the most potent HS-CRM that was identified by the hydrodynamic transfection screen (see Figure 3). The evolutionary conservation is highlighted and the TFBS are shown. The TFBS include binding sites for LEF-1 (brown), LEF-1/TCF (dark green), CEBP (yellow), FOXA1 (light blue), HNF1 (light green), and MyoD (purple).

Figure 2

Schematic representation of vectors used. (a) AAV9-HS-CRM-TTR-FIX, (b) AAV9-HS-CRM-PALM-FIX, (c) AAV9-HS-CRM8-TTR-FIX, and (d) AAV9-HS-CRM8-TTR-FIXco. The expression cassettes in a–b were packaged in a single-stranded AAV9, flanked by the 5′ and 3′ AAV inverted terminal repeats (ITR). (a) The liver-specific minimal transthyretin (TTR) promoter drives the human FIX minigene. Alternatively, (b) a Palm (paralemmin) promoter was used that was only weakly expressed in liver to drive the FIX mini-gene. The hepatocyte-specific CRMs (i.e., HS-CRM1 to HS-CRM14) are located upstream of the (a) TTR promoter. Similarly, HS-CRM8 is located upstream of the (b) Palm promoter. The FIX first intron (intron A) and polyadenylation sites (pA) are also indicated. The control vectors AAV9-TTR-FIX and AAV9-Palm-FIX are identical to AAV9-HS-CRM8-TTR-FIX and AAV9-HS-CRM8-Palm-FIX, respectively, except for the absence of any HS-CRM elements. The vectors (c) AAV9-HS-CRM8-TTR-FIX and (d) AAV9-HS-CRM8-TTR-FIXco were packaged in a self-complimentary AAV9. Both vectors had a small MVM (minute virus of mice) intron cloned upstream of the hFIX gene.

Figure 3

In vivo validation of HS-CRMs. (a) Semi-high throughput HS-CRM screening in vivo after intravenous hydrodynamic liver-directed injection with pAAV-HS-CRM-TTR-FIX and pAAV-TTR-FIX plasmids at a dose of 2 µg DNA. Significant differences compared to the control without HS-CRM were indicated (t-test, *P ≤ 0.05, mean ± SD). (b) FIX expression levels after intravenous administration of AAV9-HS-CRM8-Palm-FIX and AAV9-Palm-FIX (1 × 10¹¹ vg/mouse) (n = 5 per cohort, C57Bl/6). The difference in FIX expression levels was significant (t-test, *P ≤ 0.0001). (c) FIX expression levels after intravenous administration of AAV9-HS-CRM8-TTR-FIX and AAV9-TTR-FIX control vectors (5 × 10⁹ vg/mouse) (n = 5 per cohort, C57Bl/6). The difference in FIX expression levels was significant (t-test, *P ≤ 0.00005). FIX levels were determined using a hFIX-specific ELISA. (d) Hepatocyte-specificity of AAV9 containing HS-CRM8. RT-PCR analysis on 20 ng total RNA from different organs of C57/Bl6 mice (n = 3) injected intravenously with AAV9-HS-CRM8-TTR-FIX vectors (3 × 10¹² vg/mouse); RNA liver samples were amplified with and without RT to exclude genomic DNA amplification. Amplification of hFIX mRNA was not detectable in these control samples without RT (data not shown). (e) Corresponding qRT-PCR analysis of hFIX mRNA levels in the different organs expressed relative to hFIX mRNA levels in the liver. Expression levels (mean ± SD) relative to liver are shown. H₂O, water control; MW, molecular weight marker; PBS liver, liver sample of PBS-injected control mice; RT, reverse transcription.

Figure 4

Optimization of the HS-CRM8-TTR-hFIX construct. The (a) pAAV-HS-CRM8-TTR-MVM-hFIX or (b) pAAV-HS-CRM8-TTR-FIXIA and (a,b) pAAV-HS-CRM8-TTR-MVM-hFIXco plasmids were hydrodynamically transfected into C57BL/6 mice at doses of (a) 2 µg/mouse or (b) 0.5 µg/mouse. FIX expression was measured using a hFIX-specific ELISA (n = 4) on plasma samples collected 1 or 2 days after transfection (**P < 0.01; ***P < 0.001).

Figure 5

Preclinical validation of efficacy and safety in non-human primates. The scAAV9-HS-CRM8-TTR-MVM-hFIXco vector was injected in two cynomolgus macaques at a dose of 9 × 10¹¹ vg/kg. (a) hFIX antigen expression, (b) anti-AAV9 capsid antibodies, and (c) anti-hFIX antibodies were measured by ELISA. (b) The drop in anti-AAV9 IgG during the immunosuppressive treatment corresponded to a drop in anti-AAV9 neutralization titer (from 1:1,000 to 1:316). (d) Inhibitor anti-hFIX antibodies were determined using Bethesda assays. The effect of the immunosuppressive regimen was monitored by determining the percentage of (e) CD20⁺ B cells and (f) CD3⁺ T cells. Open circles: animal OBEV7; filled squares: animal OUG6; gray area between dashed lines: cyclosporine A (CsA) administration; dotted line: rituximab (Rtx) administration.

Figure 6

Expression biodistribution and transduction efficiency analysis in non-human primates. (a) Expression was analyzed by qRT-PCR and (b) biodistribution and transduction efficiency in different organs was determined by qPCR. Background signals from non-injected animals were subtracted from the experimental values.