1. IntroductionCystic fibrosis (CF) is a life-shortening multi-organ genetic disease affecting ~ 90,000 individuals worldwide, caused by mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene that encodes a chloride (Cl−) channel located at the surface of epithelial cells resulting in impaired ion transport across tissues of the exocrine system (reviewed in [[1]Rowe S.M. Miller S. Sorscher E.J. ]). The impaired CFTR activity alters electrolytes and hydration balance across epithelia, causing the accumulation of thick mucus in the lung bronchial tree leading to a chronic progressive lung disease, which is the major cause of morbidity and mortality (reviewed in [[1]Rowe S.M. Miller S. Sorscher E.J. ]).The last decade has witnessed developments of genotype-specific, targeted drugs that improve CFTR protein folding, stability and gating defects, leading to increased amounts of mutant CFTR reaching the cell surface and to restoration of the ion transport. These CFTR modulators offer therapeutic opportunities mainly for patients that carry gating mutations or the F508del mutation (reviewed in [[2]Bell S.C. Mall M.A. Gutierrez H. Macek M. Madge S. Davies J.C. et al.The future of cystic fibrosis care: a global perspective.]). Still, there are CF patients carrying CFTR mutations for which the current CFTR modulators are unlikely to provide a clinical benefit. Among these are patients carrying mutations abrogating the production of CFTR proteins, such as deletions, nonsense and splicing mutations.Among the ~2000 reported CFTR sequence variations, a significant fraction (10–15%) affect splicing of the precursor messenger RNA (pre-mRNA), by either creating or abolishing canonical splice sites, commonly leading to skipping over the exon. There is another group of mutations altering exonic and intronic regulatory splicing motifs throughout the gene [[3]Garcia-Blanco M.A. Baraniak A.P. Lasda E.L. Alternative splicing in disease and therapy.], leading to variable levels of both aberrantly and correctly spliced transcripts from these mutated alleles. This group includes the splicing mutations 3849+10 kb C-to-T (c.3717+12191C-to-T), 1811+1.6 kb A-to-G (c.1679+1634A>G), 3272–26A-to-G (c.3140–26A>G), IVS8–5T and 2789+5G-to-A (c.2657+5G>A) (reviewed in [[4]Pharmacological induction of CFTR function in patients with cystic fibrosis: mutation-specific therapy.]). Most patients carrying these later splicing mutations are pancreatic sufficient, however, clinical studies show that their lung function is variable and is similar to that observed among patients with CF carrying minimal function mutations (https://cftr2.org/) [5Augarten A. Yahav Y. Szeinberg A. Noiman S. Gazit E. Kerem B.-.S. et al.Mild cystic fibrosis and normal or borderline sweat test in patients with the 3849 + 10 kb C → T mutation., 6Kerem E. Rave-Harel N. Augarten A. Madgar I. Nissim-Rafinia M. Yahav Y. et al.A cystic fibrosis transmembrane conductance regulator splice variant with partial penetrance associated with variable cystic fibrosis presentations., 7Chillón M. Casals T. Mercier B. Bassas L. Lissens W. Silber S. et al.Mutations in the cystic fibrosis gene in patients with congenital absence of the vas deferens.].The 3849+10 kb C-to-T splicing mutation, generates an aberrant 5′ splice site, deep in intron 22 of the CFTR pre-mRNA. This activates a cryptic 3′ splice site 84 nucleotides upstream, resulting in the inclusion of 84 intronic nucleotides that constitute a cryptic exon in the CFTR mRNA [[8]Highsmith W.E. Burch L.H. Zhou Z. Olsen J.C. Boat T.E. Spock A. et al.A novel mutation in the cystic fibrosis gene in patients with pulmonary disease but normal sweat chloride concentrations.]. This 84 bp cryptic exon contains an in-frame stop codon, leading to degradation of a significant fraction of the mRNA by the nonsense-mediated mRNA decay (NMD) mechanism, as well as to the production of truncated non-functional CFTR proteins [[9]Linde L. Boelz S. Nissim-Rafinia M. Oren Y.S. Wilschanski M. Yaacov Y. et al.Nonsense-mediated mRNA decay affects nonsense transcript levels and governs response of cystic fibrosis patients to gentamicin.]. Importantly, the 3849+10 kb C-to-T mutation does not alter the wild-type (WT) splice sites and can enable the generation of both aberrantly and correctly spliced transcripts. Since the normal CFTR splice site sequences are intact, the involved pre-mRNA retains the potential for normal splicing, if usage of the aberrant splice sites could be inhibited. This mutation is the 7th most common CFTR mutation in the US and 8th in Europe, carried by >1400 CF patients worldwide [,]. In several populations, the mutation is highly prevalent, such as in Ashkenazi Jews and CF patients in Slovenia, Poland and Italy. Since this mutation is associated with reduced amount of normal CFTR, clinical trials investigated the effect of the modulator ivacaftor alone or together with tezacaftor and showed a modest clinical benefit [[12]Rowe S.M. Daines C. Ringshausen F.C. Kerem E. Wilson J. Tullis E. et al.Tezacaftor–ivacaftor in residual-function heterozygotes with cystic fibrosis.,[13]Kerem E. Cohen-Cymberknoh M. Tsabari R. Wilschanski M. Reiter J. Shoseyov D. et al.Ivacaftor in people with cystic fibrosis and a 3849+10 kb C → T or D1152H residual function mutation.]. Therefore, another approach is required in order to restore the CFTR function and significantly improve the disease in patients carrying alternative splicing mutations.Importantly, a correlation between lung disease severity as measured by lung function and the level of correctly spliced CFTR transcripts was found for patients carrying various splicing mutations, including the 3849+10 kb C-to-T mutation [[8]Highsmith W.E. Burch L.H. Zhou Z. Olsen J.C. Boat T.E. Spock A. et al.A novel mutation in the cystic fibrosis gene in patients with pulmonary disease but normal sweat chloride concentrations.,14Ramalho A.S. Beck S. Meyer M. Penque D. Cutting G.R. Amaral M.D. Five Percent of Normal Cystic Fibrosis Transmembrane Conductance Regulator mRNA Ameliorates the Severity of Pulmonary Disease in Cystic Fibrosis., 15Highsmith W.E. Burch L.H. Zhou Z. Olsen J.C. Strong T.V. Smith T. et al.Identification of a splice site mutation (2789+5 G>A) associated with small amounts of normal CFTRmRNA and mild cystic fibrosis., 16Chiba-Falek O. Kerem E. Shoshani T. Aviram M. Augarten A. Bentur L. et al.The molecular basis of disease variability among cystic fibrosis patients carrying the 3849+10 kb C→T mutation., 17Rave-Harel N. Kerem E. Nissim-Rafinia M. Madjar I. Goshen R. Augarten A. et al.The molecular basis of partial penetrance of splicing mutations in cystic fibrosis.] as reviewed in [[18]The molecular basis for disease variability in cystic fibrosis.]. This correlation is found also among patients with the same genotype [[16]Chiba-Falek O. Kerem E. Shoshani T. Aviram M. Augarten A. Bentur L. et al.The molecular basis of disease variability among cystic fibrosis patients carrying the 3849+10 kb C→T mutation.,[17]Rave-Harel N. Kerem E. Nissim-Rafinia M. Madjar I. Goshen R. Augarten A. et al.The molecular basis of partial penetrance of splicing mutations in cystic fibrosis.]. The ability of the splicing machinary to act as a disease modifier was demonstrated in several models of genetic diseases caused by splicing mutations (reviewed in [[19]Nissim-Rafinia M. Kerem B. Splicing regulation as a potential genetic modifier.,[20]Nissim-Rafinia M. Kerem B. The splicing machinery is a genetic modifier of disease severity.]). For example, overexpression of splicing factors was able to increase the level of correctly spliced CFTR RNA transcribed from the 3849+10 kb C-to-T allele and to promote the restoration of CFTR channel function [[21]Nissim-Rafinia M. Aviram M. Randell S.H. Shushi L. Ozeri E. Chiba-Falek O. et al.Restoration of the cystic fibrosis transmembrane conductance regulator function by splicing modulation.]. These observations, which highlight the potential of splicing modulation as a therapeutic approach, and the therapeutic need which still exists for CF patients carrying splicing mutations [[12]Rowe S.M. Daines C. Ringshausen F.C. Kerem E. Wilson J. Tullis E. et al.Tezacaftor–ivacaftor in residual-function heterozygotes with cystic fibrosis.,[13]Kerem E. Cohen-Cymberknoh M. Tsabari R. Wilschanski M. Reiter J. Shoseyov D. et al.Ivacaftor in people with cystic fibrosis and a 3849+10 kb C → T or D1152H residual function mutation.], encouraged us to develop drug candidates with a specific splice-switching potential.A specific therapeutic approach for splicing modulation is based on the administration of single-stranded short Antisense Oligonucleotides (ASOs) designed to hybridize to specific elements within target RNAs (reviewed in [[22]Kole R. Krainer A.R. Altman S. RNA therapeutics: beyond RNA interference and antisense oligonucleotides.,[23]RNA Targeting Therapeutics: molecular Mechanisms of Antisense Oligonucleotides as a Therapeutic Platform.]). Splice switching ASO-based therapies are designed to inhibit or activate specific splicing events by a steric blockade of the recognition of specific splicing elements and preventing the recruitment of effectors to these sites. The potential of ASOs to modulate the splicing pattern generated due to CFTR splicing mutations was shown in cellular systems overexpressing full-length mutated CFTR cDNA constructs. ASO transfection of epithelial cell cultures, expressing a CFTR cDNA vector harbouring a mini-intron 22 with the 3849+10 kb C-to-T locus, enhanced normal CFTR splicing and increased the production of normally processed CFTR proteins [[24]Friedman K.J. Kole J. Cohn J.A. Knowles M.R. Silverman L.M. Kole R. Correction of aberrant splicing of the Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) gene by antisense oligonucleotides.]. Similarly, ASO transfection of epithelial cells expressing a cDNA harbouring the c.2657+5G>A (2789+ 5G>A) splicing mutation, which causes the generation of CFTR transcripts lacking exon 16, increased the amount of correctly spliced CFTR proteins localized at the plasma membrane and restored CFTR function [[25]Igreja S. Clarke L.A. Botelho H.M. Marques L. Amaral M.D. Correction of a cystic fibrosis splicing mutation by antisense oligonucleotides.]. The ability of ASOs to modulate the splicing of the endogenous 3849+10 kb C-to-T allele was recently demonstrated by Michaels et al., showing that transfecting primary Human Bronchial Epithelial cells (HBEs) with a phosphorodiamidate morpholino oligomer (PMO) targeted to mask the cryptic splice site was able to block aberrant splicing and to improve CFTR function [[26]Michaels W.E. Bridges R.J. Hastings M.L. Antisense oligonucleotide-mediated correction of CFTR splicing improves chloride secretion in cystic fibrosis patient-derived bronchial epithelial cells.].ASO-based drugs modulating splicing are already approved for Spinal muscular atrophy (SMA) and Duchenne muscular dystrophy (DMD) (reviewed in [[27]Groen E.J.N. Talbot K. Gillingwater T.H. Advances in therapy for spinal muscular atrophy: promises and challenges.,[28]Optimization of antisense-mediated exon skipping for Duchenne muscular dystrophy.]). The exciting clinical data suggests that ASO-mediated splicing modulation is able to improve protein function and slow disease progression. In light of this data, modulating the level of correctly spliced CFTR transcripts using an ASO-based approach has a great therapeutic potential for CF patients.

Here we focused on the development of drug candidates for patients carrying the 3849+10Kb C-to-T splicing mutation, using chemically modified ASOs targeted to prevent the recognition of splicing elements involved in the cryptic exon inclusion. We have identified a lead ASO able to significantly increase the level of correctly spliced mRNA and restore the production of normal and functional CFTR channels by a free ASO uptake in well differentiated polarized Human Nasal Epithelial (HNE) and HBE cells, from patients carrying the 3849+10Kb C-to-T allele. Our promising results are aimed to serve as a basis for clinical evaluation of the lead ASO.

4. DiscussionIn the present study, we have described a developmental path which led to the identification of a lead ASO drug candidate able to correct the aberrant splicing and restore normal full-length functional CFTR protein in primary, well-differentiated HNEs and HBEs from patients carrying the 3849+10 kb C-to-T mutation. Our results clearly show that the lead ASO drug candidate is highly effective and potent, displaying a significant and consistent splicing modulation effect which leads to restoration of CFTR function, reaching levels expected to confer a significant clinical benefit and improved quality of life for patients carrying this splicing mutation (reviewed in [[43]Pharmacologic therapy for stop mutations: how much CFTR activity is enough?.]).The 84 bp cryptic exon region harbors various ESE motifs (Supplementary Figure 2), that may contribute to the recognition of this sequence as a cryptic exon, in alleles carrying the C-to-T mutation which generates a new alternative donor site. Indeed, previous studies by us and others [[21]Nissim-Rafinia M. Aviram M. Randell S.H. Shushi L. Ozeri E. Chiba-Falek O. et al.Restoration of the cystic fibrosis transmembrane conductance regulator function by splicing modulation.,[24]Friedman K.J. Kole J. Cohn J.A. Knowles M.R. Silverman L.M. Kole R. Correction of aberrant splicing of the Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) gene by antisense oligonucleotides.,[26]Michaels W.E. Bridges R.J. Hastings M.L. Antisense oligonucleotide-mediated correction of CFTR splicing improves chloride secretion in cystic fibrosis patient-derived bronchial epithelial cells.] demonstrated the ability of splicing modulation to rescue CFTR protein processing and function in cells carrying the 3849+10 kb C-to-T mutation. We have previously shown that over-expression of splicing factors in nasal epithelial cells carrying the 3849+10 kb C-to-T mutation increased the level of correctly spliced transcripts and restored CFTR channel function [[21]Nissim-Rafinia M. Aviram M. Randell S.H. Shushi L. Ozeri E. Chiba-Falek O. et al.Restoration of the cystic fibrosis transmembrane conductance regulator function by splicing modulation.]. A recent study further supported these findings, showing that transfection of a 25-mer PMO (targeting the donor 5′ splice site generated by the 3849+10 kb C-to-T mutation) was able to reduce aberrant splicing and to improve CFTR channel activity in HBEs derived from three CF patients carrying at least one allele of the 3849+10 kb C-to-T mutation [[26]Michaels W.E. Bridges R.J. Hastings M.L. Antisense oligonucleotide-mediated correction of CFTR splicing improves chloride secretion in cystic fibrosis patient-derived bronchial epithelial cells.]. In line with this important recent data, our present work, performed both in cells over-expressing the 3849+10 kb C-to-T mutated CFTR allele and in respiratory primary cells derived from seven CF patients homozygous for the mutation or heterozygote with various second alleles, establishes that splicing modulation is a powerful tool for restoration of CFTR in cells carrying a splicing abnormality.In the current work, we describe development of a highly potent ASO-based drug candidate, a 2′-MOE/PS modified 20 nt ASO that was efficiently delivered by free uptake into well differentiated HNEs and was able to completely restore correct splicing and function at low nanomolar concentrations. It is worth noting that achieving efficient in vivo oligonucleotide delivery remains a major challenge in the field of oligonucleotide therapeutics, and the majority of oligonucleotide therapeutics are designed for local delivery or delivery to highly vascularized tissues. Moreover, most of the approved therapies are using ‘naked ASOs’ (lacking a delivery vehicle) and are dependent on chemical modification to enhance their tissue delivery (reviewed in [[44]Roberts T.C. Langer R. Wood M.J.A. Advances in oligonucleotide drug delivery.]). In the present study, we have treated primary HNEs and HBEs with naked 2′-MOE/PS modified ASOs, resembling the conditions following in vivo administration. This enables the translation of the ASO effect from the primary patients’ cells into the clinic, whereas previous studies either have not studied primary patients’ respiratory cells [[24]Friedman K.J. Kole J. Cohn J.A. Knowles M.R. Silverman L.M. Kole R. Correction of aberrant splicing of the Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) gene by antisense oligonucleotides.] or have used transfection for ASO delivery [[26]Michaels W.E. Bridges R.J. Hastings M.L. Antisense oligonucleotide-mediated correction of CFTR splicing improves chloride secretion in cystic fibrosis patient-derived bronchial epithelial cells.] and thus the translation of their results is limited.Masking ESE motifs by ASOs has the potential to interfere with their recognition and facilitate skipping over the cryptic exon. Still, the presence of an ESE motif in a sequence does not necessarily identifies that sequence as a functional ESE [[38]Cartegni L. Wang J. Zhu Z. Zhang M.Q. Krainer A.R. ESEfinder: a web resource to identify exonic splicing enhancers.]. Future experiments are required for revealing the contribution of each ESE motif and the combination of them in the 84 bp cryptic region to the splicing modulation by ASOs in respiratory tissues.The lead ASO was selected by screening consecutive ASOs covering the entire 84 bp cryptic exon and flanking intronic sequences, targeting all sequence elements with a potential role in recognition of the cryptic exon. This “ASO walk” is a central step in drug development path as it enables to identify and select the optimal ASOs. Previous studies however, have studied the effect of only a very limited number of ASOs, and lacked the important selection of the optimal ASO required for development of a potent drug [[24]Friedman K.J. Kole J. Cohn J.A. Knowles M.R. Silverman L.M. Kole R. Correction of aberrant splicing of the Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) gene by antisense oligonucleotides.,[26]Michaels W.E. Bridges R.J. Hastings M.L. Antisense oligonucleotide-mediated correction of CFTR splicing improves chloride secretion in cystic fibrosis patient-derived bronchial epithelial cells.]. As part of the selection process, the ASOs were also tested for their off-target specificity, immunogenicity potential using peripheral blood mononucleated cells, cytokines release activation model and their potential in vivo toxicity. Lastly their stability and penetration ability through mucus, both in vitro and in lungs of ENaC mice were carefully assessed (data not shown).

The lead ASO candidate rescued CFTR expression and function to WT levels in HNEs from a homozygote patient and reached an average of 43% of WT levels in HNEs and HBEs from six heterozygote patients, 5 of which achieving levels ≥37% of WT. The ASO ability to modulate the splicing pattern in several different patients carrying various second CFTR mutated allele (3849+10 kb C-to-T/F508del; 3849+10 kb C-to-T/W1282X; 3849+10 kb C-to-T/405+1G-to-A; 3849+10 kb C-to-T/3849+10 kb C-to-T) highlights its general therapeutic potential for a repertoire of patients and genotypes.

Correlations between CFTR functional modulation in patient-derived HNE cells with clinical efficacy showed that reaching CFTR-dependent Cl− secretion levels >10% of WT function corresponds to a significant improvement in the respiratory activity measured as change in ppFEV1 [[31]Pranke I.M. Hatton A. Simonin J. Jais J.P. Le Pimpec-Barthes F. Carsin A. et al.Correction of CFTR function in nasal epithelial cells from cystic fibrosis patients predicts improvement of respiratory function by CFTR modulators.]. Indeed, exceeding 10% of WT CFTR activity is associated with pancreatic sufficiency and less elevated sweat chloride levels (reviewed in [[45]Assessing the disease-liability of mutations in CFTR.]). Our results therefore provide a solid basis for advancement of our lead ASO, expected to confer significant clinical benefit.It is important to note that ASOs with 2′-MOE chemistry are already in clinical use. The FDA-approved SPINRAZA®, a splice-switching ASO indicated for SMA, composed of the 2′-MOE modification, demonstrates high efficacy in patients with early and later-onset disease [[46]Big win possible for Ionis/Biogen antisense drug in muscular atrophy.]. The favorable benefit-to-risk profile of this and other 2′-MOE-based drugs [[35]The chemical evolution of oligonucleotide therapies of clinical utility.], prompted us to evaluate the efficacy of our lead ASO with 2′-MOE chemical modification. In previous studies, aiming to correct the splicing pattern generated due to the 3849+10 kb C-to-T mutation, the 2′-OMe [[24]Friedman K.J. Kole J. Cohn J.A. Knowles M.R. Silverman L.M. Kole R. Correction of aberrant splicing of the Cystic Fibrosis Transmembrane Conductance Regulator (CFTR) gene by antisense oligonucleotides.] or the PMO [[26]Michaels W.E. Bridges R.J. Hastings M.L. Antisense oligonucleotide-mediated correction of CFTR splicing improves chloride secretion in cystic fibrosis patient-derived bronchial epithelial cells.] chemical modifications were used. As presented in the current study, enhanced ASO efficiency was achieved by the 2′-MOE chemical modification, relative to the 2′-OMe modification, which allowed the use of significantly reduced concentrations for complete CFTR restoration (Fig. 6).Based on our results, we aim to develop an inhaled ASO drug, for increased lung exposure with minimal systemic exposure. Previous studies showed that inhaled ASOs are widely distributed and active in mice and non-human primates lungs, reaching even the most distal (alveolar) regions, with minimal systemic exposure and good tolerability as well as high stability with an estimated half-life of 13–14 days following a single ASO exposure [47Karras J.G. Crosby J.R. Guha M. Tung D. Miller D.A. Gaarde W.A. et al.Anti-inflammatory activity of inhaled IL-4 receptor-α antisense oligonucleotide in mice., 48Fey R.A. Templin M.V. McDonald J.D. Yu R.Z. Hutt J.A. Gigliotti A.P. et al.Local and systemic tolerability of a 2′O-methoxyethyl antisense oligonucleotide targeting interleukin-4 receptor-α delivery by inhalation in mouse and monkey., 49Guimond A. Viau E. Aubé P. Renzi P.M. Paquet L. Ferrari N. Advantageous toxicity profile of inhaled antisense oligonucleotides following chronic dosing in non-human primates., 50Oligonucleotide therapies for the lung: ready to return to the clinic?.]. Importantly, in a recent study, Crosby et al. demonstrated that ASOs dissolved in saline traverse CF-like mucus and distribute throughout mouse lung [[51]Crosby J.R. Zhao C. Jiang C. Bai D. Katz M. Greenlee S. et al.Inhaled ENaC antisense oligonucleotide ameliorates cystic fibrosis-like lung disease in mice.]. Moreover, the ASOs were found to be effective and lead to a reduction of the target mRNA. These studies support the notion that inhaled ASO-based drugs can be efficiently delivered to the lungs of CF patients.The 3849+10 kb C-to-T mutation is one of the ten most common CFTR mutations, carried by >1400 individuals worldwide. The CFTR modulators Kalydeco® (Ivacaftor/VX-770) and Symdeco® (Tezacaftor/Ivacaftor) were approved recently in some countries for CF patients carrying the 3849+10 kb C-to-T splicing mutation. While both ivacaftor and tezacaftor/ivacaftor are clinically approved for the treatment of people with CF caused by the 3849+10 kb C-to-T mutation, with or without the F508del mutation, clinical trials demonstrate only modest response compared to placebo. Changes in spirometry (i.e. FEV1) ranged from ~2.6–5.8% [