An Exon-Specific U1snRNA Induces a Robust Factor IX Activity in Mice Expressing Multiple Human FIX Splicing Mutants

Abstract

In cellular models we have demonstrated that a unique U1snRNA targeting an intronic region downstream of a defective exon (Exon-specific U1snRNA, ExSpeU1) can rescue multiple exon-skipping mutations, a relevant cause of genetic disease. Here, we explored in mice the ExSpeU1 U1fix9 toward two model Hemophilia B-causing mutations at the 5' (c.519A > G) or 3' (c.392-8T > G) splice sites of F9 exon 5. Hydrodynamic injection of wt-BALB/C mice with plasmids expressing the wt and mutant (hFIX-2G^5'ss and hFIX-8G^3'ss) splicing-competent human factor IX (hFIX) cassettes resulted in the expression of hFIX transcripts lacking exon 5 in liver, and in low plasma levels of inactive hFIX. Coinjection of U1fix9, but not of U1wt, restored exon inclusion of variants and in the intrinsically weak FIXwt context. This resulted in appreciable circulating hFIX levels (mean ± SD; hFIX-2G5'ss, 1.0 ± 0.5 µg/ml; hFIX-8G3'ss, 1.2 ± 0.3 µg/ml; and hFIXwt, 1.9 ± 0.6 µg/ml), leading to a striking shortening (from ~100 seconds of untreated mice to ~80 seconds) of FIX-dependent coagulation times, indicating a hFIX with normal specific activity. This is the first proof-of-concept in vivo that a unique ExSpeU1 can efficiently rescue gene expression impaired by distinct exon-skipping variants, which extends the applicability of ExSpeU1s to panels of mutations and thus cohort of patients.

Figure 1

Features of the human FIX exon 5 contexts and FIX mRNA splicing patterns. (a) Schematic representation of the full-length splicing-competent cassette and of the modified U1snRNA (U1fix and U1fix9) with the 5′ tail sequence placed above the corresponding hFIX target. U1fix9 base pairs with the intronic sequence located 9 nt downstream the exon 5, 5′ss. The U1wt, used as control is also shown for the sake of completeness. Exonic and intronic hFIX sequences are represented by boxes and lines, respectively. Upper and lower dotted lines indicate the normal and aberrant splicing patterns of exon 5. The sequences of the 5′ss and 3′ss splicing junctions, with exonic and intronic regions in upper and lower cases, are also reported together with mutations under investigation (arrows). Scores of the 5′ss and 3′ss, calculated using the Splice Site Prediction by Neural Network tool (http://www.fruitfly.org/seq_tools/splice.html), are reported in normal and mutated conditions. “n.d”, not determined by the computational program as 5′ss. pSV40, SV40 promoter; SV40pA, SV40 poly A signal. (b) Evaluation of hFIX alternative splicing patterns in healthy human liver. The schematic representation of the normal and aberrant transcripts, and of primers used for the RT-PCR (arrows) is reported in the right panel. The histogram reports the relative percentage of correct transcripts from three independent liver samples, and results expressed as mean ± SD. Amplified products were separated on 2% agarose gel. M, 100 bp molecular weight marker. RT-PCR, reverse transcription polymerase chain reaction; hFIX, human factor IX; SD, standard deviation.

Figure 2

Evaluation of the U1fix9-mediated rescue of hFIX mRNA splicing in mice. (a) Representative example of hFIX alternative splicing patterns in liver from BALB/c mice injected with the pFIX variants alone (−) or in combination (+) with a molar excess (1.5X) of the pU1wt, pU1fix or pU1fix9. The schematic representation of the normal and aberrant transcripts, and of primers used for the RT-PCR (arrows), is reported in the right panel. Amplified products were separated on 2% agarose gel. M, 100 bp molecular weight marker. The histograms report the relative percentage of correct transcripts in liver from mice injected at the different conditions (n = 4 for group 1–2, 4–5, 7–10; n = 5 and n = 7 for groups 3 and 6, respectively), and results are expressed as mean ± SD. (b) Representative example of hFIX splicing patterns in various organs of mice coinjected with pFIX-2G^5′ss and pU1fix9. (c) Evaluation (middle and lower panels) and comparison (upper panel) of the expression of the endogenous (U1end, white histograms) and ExSpeU1 (U1fix9, gray histograms) by qPCR in liver from mice treated as in panel A. For qPCR, the results are expressed as mean ± SD from three mice for each condition. Values are normalized for the mouse GAPDH and are indicated as fold increase compared to the reference indicated in the Y-axis, set to 1. No transcripts were detected in samples without reverse transcriptase. ExSpeU1, Exon-specific U1snRNA; hFIX, human factor IX; qPCR, quantitative PCR; RT-PCR, reverse transcription polymerase chain reaction; SD, standard deviation.

Figure 3

Evaluation of U1fix9-mediated rescue of hFIX protein and coagulant properties in mice. (a) Plasma hFIX antigen levels (µg/ml) in mice injected with the pFIX variants alone (−) or in combination (+) with a molar excess (1.5X) of pU1wt, pU1fix or pU1fix9. Negative controls are represented by mice injected with saline solution alone. To evaluate hFIX antigen in mouse plasma, samples were diluted 1:10 or 1:20 and evaluated in duplicate. A standard curve was created by adding known amounts of hFIX to mouse plasma, and the sensitivity threshold was 0.40 ng/ml of hFIX. Histograms reports the results obtained in the different mouse groups (n = 4, for group 1–2, 4–5, and 7–10; n = 5 and n = 7, for groups 3 and 6, respectively) and are expressed as mean ± SD. (b) Western Blotting analysis performed with the anti-hFIX antibody of the hFIX isoforms in plasma from mice treated as in panel a. mFIX, mouse FIX; hFIXwt, human hFIX. (c) FIX-dependent activated partial thromboplastin time (aPTT) values (s) in plasma from mice treated as in panel A or in plasma from mice (two independent) expressing the mutated hFIX variants supplemented with known amounts of purified rhFIX (0.25, 0.5 or 1 µg/ml). All samples have been tested in duplicate and the mean reported as single symbols. hFIX, human factor IX; rhFIX, recombinant human FIX.

Figure 4

Evaluation of the U1fix9-mediated mechanism. Schematic representation of the human F9 exon/intron 5 sequence (in bold and lowercase, respectively) and of antisense oligoribonucleotides (AONfix20 and AONfix25) and of the U7fix9 masking the intronic region recognized by the U1fix9. The gels report the evaluation on alternative splicing patterns in cells transfected with pFIX variants alone (−) or in combination with 1 mmol/l (+) and 100 mmol/l (++) of the AONfix20 and AONfix25, or with a molar excess (1.5X) of the pU1fix9 or pU7fix9. The schematic representation of the normal and aberrant transcripts, and of primers used for the RT-PCR (arrows), is reported in the right panel. Amplified products were separated on 2% agarose gel. RT-PCR, reverse transcription polymerase chain reaction.