Pathological mechanism and antisense oligonucleotide-mediated rescue of a non-coding variant suppressing factor 9 RNA biogenesis leading to hemophilia B.

Loss-of-function mutations in the human coagulation factor 9 (F9) gene lead to hemophilia B. Here, we dissected the consequences and the pathomechanism of a non-coding mutation (c.2545A>G) in the F9 3' untranslated region. Using wild type and mutant factor IX (FIX) minigenes we revealed that the mutation leads to reduced F9 mRNA and FIX protein levels and to lower coagulation activity of cell culture supernatants. The phenotype could not be compensated by increased transcription. The pathomechanism comprises the de novo creation of a binding site for the spliceosomal component U1snRNP, which is able to suppress the nearby F9 poly(A) site. This second, splicing-independent function of U1snRNP was discovered previously and blockade of U1snRNP restored mutant F9 mRNA expression. In addition, we explored the vice versa approach and masked the mutation by antisense oligonucleotides resulting in significantly increased F9 mRNA expression and coagulation activity. This treatment may transform the moderate/severe hemophilia B into a mild or subclinical form in the patients. This antisense based strategy is applicable to other mutations in untranslated regions creating deleterious binding sites for cellular proteins.

Introduction

Variants in the non-coding part of the genome can result in severe diseases and phenotypes with a high expressivity [1] and the number of variants identifiable by GWAS can explain only a modest fraction of the phenotypic variation [2]. However, the pathomechanism underlying these variants are often difficult to dissect.

Factor IX (FIX) constitutes a serine protease in the enzyme cascade of the intrinsic coagulation pathway. The F9 gene is located on the X chromosome consisting of 8 exons [3]. Disease-causing variants in the F9 gene cause complete or partial deficiency of the corresponding coagulation factor, ultimately leading to hemophilia B (OMIM:306900) of varying severity with an incidence of 1 in 25.000 births [4]. The FIX mutation database (http://www.factorix.org;[5] lists 1083 different variants. Of these, 122 are non-coding and four occur in the 3’ untranslated region (3’UTR). These non-coding variants can have severe consequences and in addition, they are often part of regulatory sequences, which otherwise would escape our notice [1].

In 1993 a 3’UTR variant (Chr.X, NC_000023.11:g.139563228A>G) was described in four families with moderate to severe hemophilia B [6]. Affected individuals showed FIX activity between <1 and 4.5% of normal. At present, the FIX mutation database lists 21 patients from different geographic regions with this variant (http://www.factorix.org;[7, 8]; S1 Table). The variant may generate a 5’ splice site (5’SS) by an A to G transition at position +1157 downstream of the stop codon (c.2545A>G; NM_000133) in the 3’ UTR [6]. However, the underlying pathophysiological mechanism remained unsolved.

Eukaryotic pre-mRNAs undergo multiple processing steps. The removal of introns is executed by the spliceosome, which consists of five small nuclear U-rich ribonulceoproteins (UsnRNPs). The 5’SS is recognized by the RNA component of the U1snRNP [9]. The strength of a U1 binding site is primarily determined by its sequence complementarity to U1snRNA [10]. In addition to splicing, mRNAs are cleaved and polyadenylated at their 3’ end. Again, several conserved sequence features in close proximity to the hexameric poly(A) signal (PAS) direct the assembly of the cleavage and polyadenylation (CPA) machinery. The CPA is confronted with the task to identify the correct PAS among many close matches. Especially the long mammalian introns harbor many cryptic PAS. Recently, we and others have shown that U1snRNP is also part of a nuclear surveillance mechanism that suppresses cryptic polyadenylation sites and thereby ensures transcript integrity [11, 12].

For the LAMTOR2 (p14) gene, we demonstrated that a 3’UTR variant creating a U1 binding site leads to PAS suppression, thereby causing an immunodeficiency [12]. Thus, the generation of a de novo U1 binding site in the F9 3’UTR could have at least two distinct outcomes: aberrant splicing or failure of 3’ end processing and RNA degradation. Here, we show that our F9 variant generates a U1 binding site. Recruitment of U1snRNP does not initiate splicing, but rather leads to suppression of the downstream PAS and to a strong decrease in F9 mRNA levels. This phenotype is the primary cause for reduced FIX protein levels and lower coagulation activity. Using different antisense-oligonucleotide based strategies; we rescued the mutant phenotype, which again points to U1snRNP as the key player in the pathomechanism. Furthermore, we opened the venue to a gene therapeutic approach to target such UTR variants. Our study strengthens the importance of non-coding variants as disease-causing and demands their pathomechanistic analysis.

Results

The factor 9 variant in the 3’UTR reduces F9 mRNA expression

The previously described A to G transition at +1157 bp (c.2545A>G) downstream of the F9 stop codon may create a 5’SS by comparison to the consensus sequence [6]. Based on our results with a similar disease-causing mutation in the LAMTOR2 gene [12], we thought to revisit this particular variant. The 3’UTR variant is located upstream of the polyadenylation signal (PAS; Fig 1A). Bioinformatic analysis revealed a putative 3’SS downstream of the PAS in a similar distance followed by a second PAS (Fig 1A). Thus, we were left with five possible explanations: i) the variant creates a miRNA binding site; ii) splicing deletes PAS1 with insufficient PAS2 usage; iii) splicing in the 3’UTR more than 50 nts downstream of the F9 stop codon eliciting non-sense mediated RNA decay (NMD; [13]; iv) the variant creates or deletes a motif for an RNA-binding protein (RBP) and v) the site binds U1snRNP thereby preventing PAS recognition and 3’ end processing [14].

Fig 1

The variant leads to lower F9 RNA levels and creates a putative 5’ splice site.

A) Depiction of the SV40 driven Factor IX minigene. The F9 cDNA (Exon 1–8) is followed by its authentic 3’UTR. The position of the primer used for the 3’RACE is marked by a blue arrow. The variant is indicated by an asterisk at position 1156 downstream of the stop codon and 208 nts upstream of the polyadenylation site 1 (PAS1). Further downstream a putative 3’SS (p3’ss) as well as a potential second polyadenylation site (PAS2) was annotated. Another 421 nts of genomic sequence downstream of PAS2 were included. The sequence of the putative U1 binding site (5'SS) (FIXmut) is opposed to the corresponding regions of the wildtype and the 5’SS consensus sequence. The oblique outlines the exon/intron border within a common 5’SS. The numbers of possible base pairs to U1 snRNA are shown on the right, followed by the respective MaxEnt score. The position of each nucleotide within the consensus sequence is indicated by numbering. In addition, we cloned minigenes harboring either an optimized (FIXopt) or no U1 binding site (FIXdown). B) Northern blot using total RNA obtained from HEK 293T cells transfected with the indicated constructs. The membrane was hybridized with a P32-labelled probe corresponding to the F9 cDNA. The position of F9 mRNA is indicated on the right. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) serves as a loading control. Bands were densitometrically quantified using a phosphoimager (lower panel). The wildtype was set to 1. The standard deviation represents 3 independent experiments. An unpaired student’s T-test was performed (ns, non-significant; *** p<0.0001). To monitor transfection efficiency all experiments included transfection of a eGFP encoding plasmid. C) Agarose gel electrophoresis of 3’RACE products. Two gels are shown with markers on the left side. The presence of reverse transcriptase (RT) during cDNA synthesis is indicated (+/-). The band around 300bp represents the amplicon with the forward primer shown in (A). The signal marked with an asterisk in the untreated sample (mock) represents an amplicon of the MLX gene. In the lower panel Sanger sequencing of F9 specific amplicons showed polyadenylation at the correct F9 PAS1. D) Schematic of F9 ORF, 3’UTR and downstream sequences including ampicillin sequence, f1 ori and SV40 poly (A) signal. Suppressive effect of the variant (marked by an asterisk) on PAS1 is indicated in red. Green box symbolizes the qRT-PCR amplicon 1 detecting all F9 RNA species and the yellow box indicates amplicon 2 for measuring F9 RNA species derived from read-through. qRT-PCR results are depicted below the corresponding amplicons. Each qRT-PCR experiment was conducted in biological and technical triplicates. E) Tet-off F9 RNA half-life experiment. pTetbi FIXwt and pTetbi FIXmut plasmids were transfected into Hela TA cells. 36 hours post transfection, doxycycline was added to shut off transcription. Total RNA was collected 0, 2, 4, 6 and 8 hours post shut off and analyzed by Northern blot (n = 3). F) Quantification of F9 RNA obtained from (E). FIXwt was set to 1 and all FIX bands were GAPDH normalized. G) Depiction of relative F9 RNA decay rate, both FIXwt and FIXmut were set to 1 to allow direct comparison.

First, we generated a minigene consisting of the F9 cDNA followed by the 3’UTR and downstream genomic sequences (Fig 1A). The mutated sequence shows high complementarity to U1 snRNA with the A to G transition creating the invariant GU of spliceable U2 introns (Fig 1A;[9]. Next, we generated the variant and two other versions either increasing (FIXopt) or decreasing (FIXdown; Fig 1A) sequence identity to the U1snRNA consensus as expressed by the number of possible base pairs to U1snRNA and the MaxEnt score describing splice site strength[15]. The variant creates a high scoring U1 binding site (Fig 1A).

A comparison of F9 mRNA levels indicates a dramatic decrease of FIX expression in HEK 293T cells transfected with FIXmut (Fig 1B, lanes 3 and4). Thus, the variant exerts its phenotype by downregulating F9 mRNA amount. Suppression of the FIX PAS causes read-through into the plasmid backbone (Fig 1B and 1D). Notably, neither optimization (FIXopt; Fig 1B, lane 5) nor deoptimization (FIXdown; Fig 1B, lane 2) of the sequence did alter F9 mRNA levels in a significant manner compared to the mutant or the wild type, respectively. To analyze this in more detail and to definitely exclude miRNAs as causative factors, we cloned another version containing the variant combined with low complementarity to U1snRNA (mutdown, S1A Fig). Although this construct carries the variant, the inability to bind U1 prevents RNA decrease (S1B Fig). This demonstrates the relevance of the mutated position for U1snRNP recognition transforming a sequence motif into a strong U1 binding site causing FIX RNA decrease.

The F9 gene holds a potential 3’SS 198 nts downstream of its PAS (Fig 1A). This can give rise to a splicing event initiated from the newly created 5’SS and thereby removing the canonical PAS leading to nonsense-mediated RNA decay [13]] or to a general failure in 3’ end processing due to an insufficient PAS2. To examine these possibilities, a 3’Rapid Amplification of cDNA Ends (3’RACE) was conducted (Figs 1A and S1C). Sequencing indicates the absence of a splicing event, since the gene-specific forward primer is located upstream of the potential 5’SS (S1C Fig). In addition, the 3’RACE revealed appropriate polyadenylation at the annotated site in both FIXwt and FIXmut transfected cells (Fig 1C upper and lower panel). Importantly, these results show that the residual, non-read-through mutant F9 mRNA (Fig 1B, lane 4) still underwent 3’ end processing at the F9 PAS1. Polyadenylation at PAS2 occurred neither in the wildtype and nor in the mutant. To excluded splicing by a more direct approach we performed RT-PCR with primers located upstream of the potential 5’SS and downstream of the putative 3’SS. Again, no splicing could be detected (S1D Fig). Thus, the lack of splicing from the new, cryptic 5’SS excludes NMD and processing at PAS2 as pathomechanisms.

The PAS inhibition exerted by U1 is only transient and spatially limited to roughly 1 kb [16, 17] and allows processing at downstream sites as shown in Fig 1B. To quantify the extent of read-through we used qRT-PCR with one amplicon detecting all F9 mRNA and a second measuring only the read-through (Fig 1D). The variant increased read-through (amplicon 2, Fig 1D, right side) as already observed in the Northern blot analysis (Fig 1B). The downregulation of mutant F9 mRNA is not as pronounced, because all RNAs including the read-through are detected (amplicon 1, Fig 1D, left side).

The residual RNA expression observed in Fig 1B and 1D could be an artifact of the plasmid-based transient transfection. Thus, we introduced our FIX minigene into a pre-tagged, doxycycline-inducible locus to obtain a stable, single copy F9 HeLa cell line for the wild type and the mutant (S1E Fig). Even from this genomic localization a residual RNA expression for the mutant was detectable (S1E Fig).

The observed reduction in F9 mRNA could be due to co-transcriptional RNA decay as described for suppression of 3’ end processing [18]. Alternatively, the variant may alter the F9 mRNA stability post-transcriptionally. To analyze this, we assessed the steady-state level of F9 mRNA and its half-life. Using a doxycycline inducible version of the FIX minigenes, we shut off transcription 36 hours post transfection. The wild type F9 mRNA produced from this minigene decays with a half-life of approximately 3 hours (Fig 1E and 1F). The mutant F9 mRNA is expressed at lower levels already at time point zero (compare to Fig 1B). The half-life analysis suggests only a slightly accelerated decay (Fig 1G). Thus, the mutation does not alter the half-life of F9 mRNA.

A modified FIX minigene phenocopies poly(A) site suppression on the protein level

Our initial attempts to detect FIX protein and to measure active FIX in the supernatant of several cell lines transfected with our minigenes failed due to low expression levels (Fig 2B, lanes 2 and 3). The usage of the CMV promoter and the 5’UTR of pcDNA3.1 (Fig 2A) slightly enhanced the F9 mRNA expression, but still recapitulated the phenotype of the variant (Fig 2B, lanes 4 and 5). Now we observed FIX protein both in whole cell lysates (Fig 2C, lanes 2,3 and inset) and in the supernatant (Fig 2D, lanes 2 and3). Since FIX is synthesized in hepatocytes, we sought to verify our results in hepatic cell lines. To this end we transfected the human hepatoma cell line Huh7 with the CMV-driven FIX constructs (S2 Fig). Also, in this cell line the variant leads to 4-fold downregulation of F9 RNA (S2A Fig). The same is true for FIX protein in cell culture supernatants (S2B and S2C Fig).

Fig 2

Establishment of a FIX protein expression system.

A) Vector architecture of three different CMV driven FIX minigenes with or without 3’UTR variant. Two constructs (middle and lower) were equipped with a truncated FIX intron 1 located in between exon 1 and exon 2. The lower construct combines truncated intron 1 and the gain-of-function mutation denoted as “Padua”. B) Northern blot using total RNA of HEK 293T cells transfected with the indicated constructs. GAPDH served as a loading control. A higher exposure of lanes 2 and 3 is shown in the right panel. C) Western blot of HEK 293T cell lysate transfected with the respective constructs visualized by Licor Odyssey and two antibodies labelled with different fluorophores (FIX in green). GAPDH (red) serves as loading control. Cells were lysed 30 hours post transfection. A higher exposure of lanes 1 to 3 is shown in the right panel. A densitometric quantification is depicted in the lower panel. The CMV-driven FIXwt construct was set to 1. D) Immunoblot analysis of supernatant collected from transfected HEK 293T cells with the respective constructs at 30 hours post transfection. The densitometric analysis was performed as in C. Standard deviations represent three independent experiments.

We additionally inserted a truncated F9 intron 1 with its authentic splice sites at the correct position into the F9 cDNA (Fig 2A, middle panel;[19]. The intron is efficiently spliced (S3A Fig) and it strongly enhances F9 mRNA and protein levels (Fig 2B, lane 6 and 2C, lane 4). Elevated transcription rather than splicing seems a likely explanation for this increase, since incorporation of the terminal F9 intron did not lead to higher mRNA levels (S3B–S3D Fig) and work done by the Kurachi laboratory also implied the presence of a transcriptional enhancer [19]. Of note, the variant again led to lower RNA and protein levels even though the basal expression was higher (Fig 2B, lane 7 and 2C, lane 5). This argues for a high expressivity of the variant and a tight suppression of 3’ end processing. Finally, we took advantage of a previously published gain-of-function variant referred to as Padua [20] (Fig 2A, lower panel) yielding a hyperactive FIX. Since the Padua mutation exerts its effect on FIX protein activity, the mRNA levels remained unchanged (Fig 2B, lanes 8 and 9). However, protein levels in the supernatant were slightly reduced (Fig 2D, lanes 6 and7). Since FIX is highly modified on the post-translational level [21], we included recombinant FIX as a size-matched control showing a higher molecular weight (Fig 2D, lane 9) indicating insufficiently modified FIX protein in both the cell lysate and the supernatant (Fig 2C and 2D). Densitometric analysis revealed that the mutant phenotype gets blurred in the context of FIX intron and the Padua mutation (Fig 2D, lower panel). However, in the cell lysates a robust 4-fold decrease was observed on protein level (Fig 2C, lower panel) comparable to the reduction of F9 mRNA (Fig 1B, lower panel).

The improved FIX minigene phenocopies the mutant also in reduced coagulation activity

CHO cells were shown to produce active FIX in-vitro to levels almost comparable with primary hepatocytes and they are harnessed to purify FIX for injection into hemophilic patients [22]. And indeed, transfection of the minigenes into CHO cells could reproduce the phenotype at the level of RNA (Fig 3A) and intracellular FIX protein (Fig 3B, lanes 4,5 and 6,7). Transfection of CHO cells displayed a favourable ratio of modified vs. unmodified FIX protein (compare Figs 2C and 3B). In addition, we quantified FIX antigen (FIX:Ag) by ELISA and observed a 4-fold reduction for all mutant constructs (Fig 3C). To assess coagulation activity, we employed of a one-stage coagulation assay [23]. Again a 4-fold decrease in activity was observed, when comparing mutant with wildtype constructs (Fig 3D). This was also observed in hepatic Huh7 cells albeit on a lower level (S2D Fig). The Padua mutation displayed a clear gain-of-function phenotype as described previously (Fig 3D; [20]. Although protein and FIX:Ag levels were quite similar (Fig 3B and 3C), the activity was enhanced by 10 fold (Fig 3D). Again, the phenotype of our variant persisted in the hyperactive Padua context showing that the variant acts on a wide range of activity levels (Fig 3D).

Fig 3

The F9 3’UTR variant reduces FIX antigen and coagulation activity.

A) Northern blot of CHO total RNA collected 30 hours post transfection with CMV driven FIX minigenes. GAPDH serves as a loading control. B) Western blot of CHO cell lysate 30 hours post transfection with the respective minigenes. GAPDH was used as a loading control and recombinant FIX serves as a positive control (Lane 8). C) Supernatant of FIX transfected cells collected 96 hours post transfection was subjected to quantification using a FIX ELISA (n = 3). Significant differences determined via Bonferroni’s Multiple Comparison Test are indicated by an asterisk (p<0.05). (D) Coagulation activity of FIX transfected CHO supernatant was determined 96 hours post transfection (n = 3). Asterisks indicate statistical differences (p<0.05).

In summary, the non-coding variant creates a U1 binding site in close proximity to the F9 poly(A) site and this results in failure of 3’ end processing and ultimately F9 mRNA degradation. The decreased mRNA levels result in lower protein amounts, less secreted FIX and finally reduced FIX antigen and coagulation activity, thus partially reflecting the clinical phenotype observed in patients [24].

Blockade of U1snRNP rescues FIX expression

To demonstrate that the binding of U1snRNP to the mutated region is responsible for the decrease in F9 mRNA, we blocked U1snRNP using Antisense Morpholino Oligonucleotides (AMOs) targeting U1snRNA (Fig 4A, red sequence). U1 AMOs block 5’SS recognition by binding to the free U1 snRNA 5’ end and additional nucleotides thereby invading and disrupting the secondary structure of U1snRNA [16]. We transfected the respective FIX minigenes followed by introduction of the AMOs in HeLa cells since they proved to be more resilient to the treatment. The mutant phenotype remained unaltered in the presence of a control AMO (Fig 4B, lanes 2,3 and lowest panel). The transfection of U1-specific AMOs facilitated a strong increase in mutant F9 mRNA expression by 3-fold (Fig 4B, lane 5 and lowest panel). Thus, the blockade of U1snRNP rescues the F9 mutant mRNA expression. Interestingly, both FIXwt and FIXmut minigenes showed elevated expression (Fig 4B, lanes 4,5). However, the level of GAPDH mRNA was decreased due to splicing inhibition upon U1snRNP blockade (Fig 4B). The 3-fold decrease in GAPDH mRNA (Fig 4C) strongly correlated with its mRNA half-life in HeLa cells of 8 hours [25]. To rule out loading differences we took advantage of 18S rRNA as an unspliced, pol I derived transcript (Fig 4B). Interestingly, expression from a GFP-encoding plasmid is also increased by anti U1 AMOs (Fig 4B). Thus, the reason for enhanced FIX expression in the wildtype seems to be a general effect of the U1 AMOs on intron-less genes. Very recently U1 AMOs were shown to increase RNA pol II density at transcriptional start sites, which may explain our observation [26].

Fig 4

Blockade of U1snRNP rescues mutant F9 RNA expression.

A) U1snRNA and its 5’SS recognition sequence (blue). Binding properties of locked nucleic acids (LNAs), indicated in green and antisense morpholino oligonucleotides (AMOs), indicated in red. B) Depiction of SV40-driven F9 expression construct. Below, Northern blot using total RNA from FIXwt/mut transfected Hela cells performed as in Fig 1B. Control or anti U1 AMOs were transfected 14 h post FIX minigene transfection as indicated. The position of FIX RNA is marked on the right. The blot was re-hybridized with a GAPDH-specific, an eGFP-specific and an 18S rRNA-specific probe. Densitometric analysis of FIX mRNA expression normalized to 18S RNA is shown on the lowest panel. The FIXwt in presence of control AMOs was set to 1 (n = 3). C) Densitometric analysis of GAPDH mRNA expression. GAPDH mRNA of FIXmut transfected cells plus control AMOs was set to 1. D) Northern blot as performed in (B) except that control and anti U1 LNAs were used. The statistical analysis of three independent experiments is shown below. The asterisks indicate that the partial rescue due the blockade of U1 is significant as determined via Student’s T-test.

In a second approach, a blockade of U1snRNP has been performed using locked nucleic acids (LNAs) designed exclusively against the U1snRNA 5’ end (Fig 4A, green sequence), thereby retaining the U1snRNA secondary structure. Masking of the U1snRNA 5’ end led to a rescue of F9 mRNA expression in FIXmut transfected cells (Fig 4D, lanes 4 and 5), whereas cotransfection with a control LNA still showed reduced FIXmut mRNA expression. The rescue was incomplete due to the lower transfection efficiency of LNAs vs. AMOs. Again, the GAPDH expression was reduced due to the manipulation of splicing. Importantly, after LNA treatment the wildtype F9 mRNA expression remained constant (Fig 4D, lanes 2 and 4).

A U1snRNA suppressor mutant reduces wild type F9 RNA expression

In a third approach we sought to co-express a suppressor U1snRNA, which is designed to bind the wildtype sequence and to a lesser extent to the variant (Fig 5A; [12]. To this end we co-transfected our FIX minigenes along with the U1 suppressor mutant plasmid or an empty plasmid as a control. The retargeting of the suppressor U1snRNP to the wildtype minigene decreases F9 mRNA expression (Fig 5B). RNA expression from the mutant minigene remains constant (Fig 5B). A rescue of the mutant RNA by the suppressor U1 snRNA is not expected since a G•U base pair still allows binding of the suppressor to mutated sequence (Fig 5A and 5B). In addition, high levels of endogenous U1snRNP are still repressing 3’ end processing of mutant FIX minigene.

Fig 5

A U1 snRNA suppressor mutant reduces wild type F9 RNA expression.

(A) The U1 suppressor mutation is depicted. In addition, the wildtype sequence (FIXwt) is opposed to the corresponding regions of the variant (FIXmut) and the 5‘SS consensus sequence. The oblique outlines the exon/intron border. The number of possible base pairs to U1snRNA are shown on the right. The position of each nucleotide within the consensus sequence is indicated by numbering. (B) HEK 293T were transfected in triplicates with the indicated minigenes and U1 suppressor plasmids or control vectors. The Northern blot was hybridized with a F9 specific probe and re-hybridized with a GAPDH probe for normalization. A statistical analysis of three independent experiments is shown on the right side. ** = p<0.004.

This approach together with the data presented above clearly demonstrates a direct involvement of U1snRNP in the suppression of 3’ end processing in the context of the 3’UTR variant. In addition, we only found one miRNA binding site, which seed region overlaps with the variant but leading to a lower score (S4 Fig). Moreover, we employed the algorithm RBP map [27] to identify whether the variant changed binding motifs for RBPs. The variant, however, creates an ANKHD1 binding site, which is involved in signaling [28] and transcriptional regulation via histone binding [29] and is thus unlikely to be responsible for the observed phenotype.

Masking the mutated site rescues FIX expression

While the mechanistic approach clearly showed that the pathomechanism of the variant depends on U1snRNP recognition, blockade of U1 is not a therapeutic option. We followed a vice versa approach and aimed to mask the newly created U1 binding site by chemically modified RNA antisense oligonucleotides (2OMePS) displaying increased stability [30] and therapeutic potential in Duchenne muscular dystrophy (DMD;[31, 32]. We designed a 20mer 2OMePS antisense oligonucleotide hybridizing to both the wildtype and the mutated sequence (Fig 6A). To this end we transfected our original SV40-driven FIX minigenes along with different concentrations of control and targeting oligonucleotides into HeLa cells, which again are more resilient to the 2OMePS treatment (Fig 6B). This vector change was necessary, since our optimized constructs produced high amounts of F9 mRNA, which could not be rescued due to a lower efficiency of 2OMePS transfection and thus, a lower nuclear antisense oligonucleotide concentration in HeLa cells. While in transfections with a control oligo the decrease of mutated F9 mRNA persisted (Fig 6B, lane 2 and 3), masking the site to prevent association of U1snRNP leads to a rescue of mutant F9 mRNA expression (Fig 6B, lane 4 and 5). Higher concentrations of the oligo exhibited some toxicity as shown by the lower GAPDH signal (Fig 6B, lane 6). Interestingly, an additional F9 mRNA species is detectable in both wildtype and mutant context indicating a switch to a processing site beyond PAS2 (Fig 1A).

Fig 6

Masking of mutated region via two different antisense oligonucleotides.

A) Depiction of the mutationally generated U1 binding site being masked by a complementary 2’-O-Methyl oligonucleotide (2OMePS). B) Northern blot of total RNA after co-transfection of SV40 driven FIXwt and mut constructs with control or targeting 2OMePS oligonucleotides. Hela cells were harvested 16h post transfection. GAPDH serves as a loading control. C) Depiction of the modified U7snRNP targeting the mutated region on the FIX mRNA thereby masking the U1snRNP binding site. D) Northern blot of total RNA after cotransfection of SV40 driven FIX minigenes and U7-antisense expression vector in indicated molar ratio (1+2/1+3). Hela cells were harvested after 16 hours as in (B) and GAPDH serves as a loading control. E) Depiction of the AAV vector backbone, which was used in transient transfections (top). The inverted terminal repeats (ITR) delineate the vector-derived sequences. A cassette consisting of the U7 promoter (U7p) and a fusion of the FIX antisense sequence (AON) to U7 snRNA was inserted. The lower panel the RNA analysis by northern blot using total RNA from transfected CHO cells in presence of the AAV plasmid expressing the antisense oligonucleotide fused to U7snRNA or pUC19 as a control. GAPDH serves as a loading control. Below three independent experiments were quantified by phosphoimager analysis. The partial rescue in the presence of the antisense oligonucleotide is indicated by a double asterisk (p = 0.0025). F) Supernatant of FIX transfected CHO cells collected 96 hours post transfection was subjected to quantification using a FIX ELISA (n = 3). Significant differences determined via Bonferroni’s Multiple Comparison Test are indicated by an asterisk (p = 0.05). G) Coagulation activity in percent matched to a standard curve of recombinant factor IX. The tested supernatants were derived from CHO cells transfected with the indicated plasmids. Standard deviations represent three independent experiments. The asterisks indicate a statistically significant elevation of FIX activity (p<0.05).

The same strategy can also be applied using a viral vector derived system. Here, the snRNP system is ‘hijacked’ by replacing the 5’ part of natural snRNAs with an antisense sequence of choice. This strategy protects the antisense sequence of choice from degradation. U7snRNA is often used for this purpose, which normally ensures histone RNA processing (Fig 6C;[33, 34]. We constructed a plasmid based on an adeno-associated virus (AAV) vector backbone [35, 36] and inserted a longer antisense sequence of 40 nucleotides complementary to the F9 3’UTR fused to U7snRNA (Fig 6C and 6E; upper scheme). This sequence shows no hits in a Blast search (S4B Fig). Transfection along with the FIX minigenes showed a rescue of mutant F9 mRNA expression (Fig 6D, lanes 4–6). Again, a more downstream PAS is used when the U1 binding site is masked (Fig 6D, lane 6). Next, we introduced the FIX minigene and the U7 plasmid into CHO cells by transfection and monitored F9 mRNA by Northern blot (Fig 6E, middle panel). The co-transfection of the U7 plasmid led to a significant increase in mutant F9 mRNA expression (Fig 6E, lowest panel). This rescue is also detectable on the protein level in the supernatant measure by FIX ELISA (Fig 6F) and finally also the coagulation activity was increased by blocking U1 access to the mutant site (Fig 6G). The absolute activity was lower compared to Fig 3 due to altered molar ratios of the transfected FIX plasmid (Fig 6G). This confirms the potential of our strategy to be applied as a gene therapeutic approach to treat non-coding and disease-causing variants leading to hemophilia B in this case and other diseases.

Discussion

A reassessment of UTR variants in F9 revealed a point mutation possibly creating a 5’SS [6]. We promisingly recapitulated the phenotype demonstrating a downregulation of F9 mRNA in the presence of the variant. A key player of this pathomechanism is U1snRNP, which erroneously binds within the 3’UTR and leads to a PAS suppression resulting in poor mRNA expression. Thus, the variant misleads the U1-dependent surveillance mechanism, which normally suppresses intronic poly(A) sites [37] and a failure of 3’ end processing results in RNA degradation[18].

The F9 mRNA repression also translates into reduced FIX protein levels explaining the hemophilia B observed in the patients (http://www.factorix.org; [8]). While we clarified the pathomechanism, our minigene system is not able to perfectly mirror the reduction of FIX antigen and coagulation activity observed in the patients. This is mainly due to a higher residual mutant RNA expression. In addition, the read-through RNAs (Fig 1B and 1D) may also contribute to FIX protein, thus slightly compensating the effect of the variant. Also, a human hepatic cell line displayed the same residual RNA expression (S2 Fig) and so did a single integrated copy of F9 under control of a dox-inducible promoter (S1C Fig). We speculate that the genomic location on the X chromosome in conjunction with local regulatory sequences is the reason for this discrepancy. In addition, we always assayed for accumulation of FIX in the supernatant compared to secretion and circulation/consumption of FIX in the patients. Nevertheless, our F9 minigene proved to be a useful tool to uncover the pathomechanism of the non-coding variant.

Despite the different levels of transcription, the phenotype of the variant persisted, which again stresses its high expressivity also shown by the reduced coagulation activity of our mutant supernatants. At the molecular level this can be explained by the high excess of U1snRNP in the nucleus leading to a tight suppression of 3’ end processing. Due to issues on FIX modifications and activity during in-vitro experiments, we sought to use the Padua mutation as a “magnifier” to thoroughly test the expressivity of our non-coding variant. Now the coagulation assay reached the same activity as recombinant FIX and the 4-fold downregulation of the variant persisted also in this context.

We reasoned that shielding of the mutant site would prevent U1snRNP-binding and alleviate PAS suppression. Two types of antisense oligonucleotide (ASO) strategies previously demonstrated efficacy in other models: i) 2OMePS RNA oligos [32, 38] with some side effects reported [39] and ii) an plasmid based system expressing an antisense oligo fused to U7 snRNA [35]. In essence, both antisense strategies lead to a rescue of F9 mRNA expression and an increase in coagulation activity, which may be able to revert the phenotype into a mild or subclinical hemophilia B. Interestingly, the RNA duplex formation and association with U7snRNP did not hamper translation in the cytosol, which suggests detachment of the complex at the nuclear pore or during the pioneer round of translation. However, we will have to determine if the partial rescue was due to limitation of the cellular processing capacity of snRNP particles [40].

Most hemophilia B patients suffer from disease-causing variants in the coding region (http://www.factorix.org). Some of these patients have been treated by gene-addition therapy introducing a non-mutated F9 cDNA as it has been done using adeno-associated viral vectors (AAV) in clinical trials [41]. We constructed similar vectors expressing the antisense-U7snRNA cassette as mentioned above and produced recombinant AAV particles. However, cell culture experiments using AAV serotype 2 did not reach sufficient antisense RNA levels to reproduce the partial rescue of FIX activity observed by transfection of the AAV vector plasmid (Fig 5E). We speculate that the U7 hairpin structure may serve as a resolving signal for the AAV rep protein leading to the aberrant packaging of truncated genomes. Currently we are optimizing the vector design. In general, this current study might be a starting point for dissecting and treating other UTR variants, which exert their effects by misplacing cellular RNA-binding proteins on the mutant RNAs. These variants may cover a wide range of phenotypes from modulation of RNA half-life to mislocalization in the cytosol. For example, the Factor VIII mutation database list four 3’UTR variants. One advantage of the variant masking strategy compared to gene addition is the ongoing physiological expression of the mutated gene as only transcribed RNAs are targeted and rescued.

Materials and methods

Expression vectors

The human F9 cDNA (NM_000133.3) was cloned into pGL3 (Promega). The 3’UTR including the variant and 597 bp of downstream sequence was synthesized by MWG operon.

FIXwt was restored by PCR mutagenesis (see primer list in Table 1). FIXdown and FIXopt were generated by PCR with primers P9, P17 and P18 and cloned into pGL3 using EcoRI and XbaI. The 5’UTR of pcDNA3.1 and a CMV promoter were inserted using BglII and HindIII. A sequence comprising Factor IX Padua and a truncated F9 intron 1 was synthesized by MWG operon and inserted into the CMV driven factor IX minigene via HindIII and BamHI. The Padua mutation has been replaced by the authentic sequence using BstBI and MfeI in a partial digest. The vector coding for U7snRNA linked to a mutated factor IX complementary region descends from pSMD2 U7dtex23 [42]. F9 specific antisense nucleotides were inserted into the U7snRNA gene by three PCR reactions (P8+P10, P9+P11 and P10+P11) and implemented into pSMD2 GFP via XbaI/NheI. F9 vectors employed for the decay experiments and stable cell lines are derived from CMV FIXwt int Padua or CMV FIXmut int Padua. F9 cDNA sequence and 3’UTR were amplified using P16+P17 and transferred into a pTetbi promoter driven vector (kindly provided by V. Cordes, MPI Göttingen) via AgeI, BamHI and NcoI digest. FIX mutdown was generated via PCR mutagenesis using primers P1, P18, P19 and P20. Products of PCR1 (P1+P18) and PCR2 (P19 and P20) were employed for a fusion PCR using P1 and P20. The corresponding PCR product was cloned into FIXwt via EcoRI/XbaI, thereby replacing the wild type sequence and giving rise to FIX mutdown. Cloning of the U1 suppressor mutant was done by modifying the 5’SS recognition site of the U1snRNA to bind „GACATAAGT“. This modified U1snRNA including the authentic U1snRNA promoter was synthesized (IDT-DNA) and cloned into the U1 wt vector [12] via HindIII/BamHI.

Table 1

List of oligonucleotides.

Cloning of	P1 (EcoRI for)	5‘-GAGGAAGAATTCAACAGTGTGTCTTCAGC-3‘
F9 cDNA	P2 (FIX_opt_rev)	5‘-GGAGACATGATACTTACCTGCTCTGGTC-3‘
	P3 (FIX_down_rev)	5‘-GGAGACATGATACAGATGTCCTCTGGTC-3‘
Cloning of	P4 (FIX_ovl_DS_for_wt)	5‘-GTTGAAGTTGCCTAGACCAGAGGACATAAG-3‘
F9 downstream	P5 (FIX_ovl_US_rev_wt)	5‘-GTATGCTAGTTAAAGGAGACATGATACTTATGTCCTC-3‘
Sequences	P6 (3’UTR BamHI for)	5‘-GCGGATCCTGAAAGATGGATTTCCAAGGTTAATTCATTG-3‘
	P7 (FIX_ovl_flank_DS_rev)	5‘-GAGCATTGAGAAAGCGCCACGC-3‘
Cloning of	P8 (U7-FIX-Reverse)	5’-AGAGGACGTAAGTATCATGTCTTTGCGGAAGTGCGTCTGTA-3’
U7 construct	P9 (U7-FIX-Forward)	5’-TTACGTCCTCTGGTCTAGGAATTTTTGGAGCAGGTTTTCT -3’
	P10 (FU7-Xba)	5‘-GGGTCTAGATAACAACATAGGAGCTGTGA-3‘
	P11 (RU7-Nhe)	5‘-AAAGCTAGCCACAACGCGTTTCCTAGGA-3‘
Amplicon 1	P12 (F9_ORF_qPCR_for)	5’-ATTCCTATGAATGTTGGTGTCCCT-3′
	P13 (F9_ORF_qPCR_rev)	5’-GGGTGCTTTGAGTGATG TTATCCAA-3′
Amplicon 2	P14 (F9_readthrough_for)	5’-GTGCACCTATAATCCCAGCTACTGGGGAG-3‘
	P15 (F9_readthrough_rev)	5’- CAGTCATAAGTGCGGCGACG-3‘
Stable cell line	P16 (cl_F9NcoI_for)	5’- GCGCGCCATGGCAAGCTTACCACTTTCACAATCTGCTAGC-3’
	P17 (cl-F9AgeI_rev)	5’-CGCGACCGGTGCGAATTGGGATGCCTCTCCATG-3’
FIXmutdown	P18 (FIX mut down rev)	5’-GGAGACATGATTAAGACGACCTCTGGTC-3’
	P19 (FIX mut down for)	5’-GTCGTCTTAATCATGTCTCCTTTAAC-3’
	P20 (Xba rev)	5’- ACTCTCTAGAAGATTCAAGATAGAAG-3’
	P21 (cr ds primer)	5‘- CTGGGCCCAGCCAAGAAATTTAA-3‘
Antisense	AMO oligonucleotide—Control	5‘-CTTCTTACCTCAGTTACAATTTATA-3‘
Morpholinos	AMO oligonucleotide–anti U1	5’-GGTATCTCCCCTGCCAGGTAAGTAT-3’
LNAs	Anti U1 Locked Nucleic Acid	5’-G+CC+AG+GT+AA+GT+AT-3’
	Control Locked Nucleic Acid	5‘-G+CC+AA+CT+CA+CT+AT-3‘
2OMePs	2’O Methyl anti FIX oligo	5’-UGAUACUUACGUCCUCUGGU-3’
Oligos	2’-O-Methyl control oligo	5’-AAAAGAAAACAUUCACAAAAUGGG-3’

Cell culture and transfection

HEK 293T, HeLa and CHO cells were grown in DMEM supplemented with 10% fetal calf serum, 1mM sodium pyruvate, and 1% penicillin/streptomycin and 4μg/mL vitamin K at 37°C and 5% CO2. One day before transfection, 5×106 293T cells were seeded in a 10-cm dish to a confluency of 70–80%.

Transfections were performed using calcium phosphate and a total DNA amount of 15.5μg (5μg FIX plasmid, 0.5μg eGFP DNA and 10μg of pUC19 DNA). Cells were harvested 16h post transfection for RNA analysis. pTetbi FIXwt or pTetbi FIXmut were co-transfected with a Flp-recombinase vector into Hela-EM2-11ht cells [43]. Single cell clones were evaluated via Mammalian beta-Galactosidase Assay Kit (Thermofisher) according to the manufacturer’s protocol. Half-life experiments were conducted in HeLa TA cells harboring a stably integrated copy of the Tet-transactivator by the addition of 20 ng Doxycyclin. For U1 blockade, HeLa cells were transfected with FIXwt and FIXmut using Viafect (Promega). 18 hours post transfection, cells were PBS-washed and trypsinised. After centrifugation, 106 cells were resuspended in 50μL of Buffer “R” (Neon Transfection System, Life Technologies) and 2μL of 300nmol of antisense morpholino oligonucleotides (AMOs) were added. The cell suspension was successively exposed to two pulses of 35ms width and an input voltage of 1005 V. After electroporation, cells were transferred in 2mL of antibiotic-free medium and incubated overnight before harvesting. For U1 blockade via locked nucleic acids (LNAs) 5×105 HeLa cells were seeded in a six-well. A total DNA amount of 2μg was transfected using Viafect (Promega). 4 hours post transfection, 10nmol of anti U1 or control LNAs (Exiqon) were transfected using Lipofectamine 3000. Cells were harvested 20 hours post LNA transfection. For the U1 suppressor experiments, HEK 293T cells were seeded into 6-well plates. At a confluency of approximately 70%, cells were transfected with 0.7ug FIXwt or FIXmut and 1.8ug of either U1 suppressor vector or empty plasmid (pUC19) using Lipofectamine 3000. RNA was harvested 30 hours post transfection and subjected to Northern blot analysis.

For FIX activity analysis, 4×105 CHO cells were seeded in a six-well. At a confluency of about 50%, cells were transfected with various FIX minigenes using Viafect (Promega). 12 hours post transfection, medium was replaced with FCS-free DMEM, supplemented with 4μg/mL vitamin K, 2.5% BSA and 1% L-Glutamine and cells were incubated for 96 hours.

For the 2OMePS mediated rescue, 5×105 HeLa cells were seeded into a six-well one day before transfection. A total DNA amount of 2μg was transfected using Viafect (Promega). 4 hours post viafection, 400nmol of targeting (2OMePS) or fluorescent control oligo were transfected using 5μL of Lipofectamine 3000 (Invitrogen). RNA was harvested 16 hours after second transfection. U7 mediated rescue was performed via cotransfection of SV40 driven FIXwt or FIX mut and pSMD2-FIX-U7 into HeLa cells in a molar ratio of 1:3 using viafect. Medium was changed after 4 hours and 16h post transfection; cells were harvested for RNA isolation.

RNA preparation and analysis

RNA methods were performed as described previously [12]. For detection of F9 RNA, a specific probe corresponding to the F9 cDNA was generated from the FIX plasmid by HindIII/BamHI digestion. The GAPDH-specific probe was prepared as described [12]. cDNA synthesis for quantitative PCR (qPCR) was conducted using QuantiTect Reverse Transcription Kit (Qiagen) according to the manufacturer’s protocol. Prior to reverse transcription, RNA was treated with TURBO DNase (Invitrogen) and purified via RNeasy columns (Qiagen). qPCR was performed with QuantiTect SYBR Green PCR Kit (Qiagen) using primer pairs P12+P13 (amplicon 1) and P14+P15 (amplicon 2). For 3’RACE experiments, poly(A)+ -RNA of transfected cells was isolated and 50 ng were reverse transcribed. To this, the Invitrogen 3’-RACE protocol and the provided reagents were utilized. The PCR products were separated via agarose gel electrophoresis. After gel extraction, the DNA was transferred into pCR2.1 (Invitrogen) for sequencing.

Western blot analysis and factor IX activity measurement

For detection of FIX protein, the F9 monoclonal antibody (M01), clone 2C9 (Abnova) was employed in a dilution of 1:500. GAPDH protein expression was detected using GAPDH antibody (1:2000; Cell Signaling Technology). Detection was performed via licor Odyssey infrared imaging system. Recombinant FIX (BeneFix, Pfizer) was employed as positive control. Supernatant obtained from CHO cells was employed for ELISA-based quantification of FIX antigen (FIX:Ag) using ZYMUTEST Factor IX from HYPHEN BioMed according to the manufacturer’s protocol including a standard FIX calibrator concentration set to 100% which equals 100 U/dl. Thus, FIX:Ag levels are expressed as % FIX protein. FIX activity was determined using a one-stage coagulation assay. Samples were diluted 1:5 in imidazol buffer and mixed with an equal amount of FIX deficient plasma and Actin FS APTT reagent (all from Siemens Healthcare). After incubation at 37°C for 2 min, 0.025 M calcium-chloride was added, and the coagulation time recorded in an Amelung KC10 coagulometer. Calibration curves were generated with human standard plasma diluted in FIX deficient plasma.

Bioinformatic analysis and statistics

MicroRNA binding sites were analyzed using the online prediction tool miRDB [44]. RBPmap [27] is available here: http://rbpmap.technion.ac.il/. Splice site prediction was performed using NNSPLICE 0.9 version [45] and MaxEnt score [15]. Statistical analysis (Bonferroni’s Multiple Comparison test) was performed using the GraphPad Prism software.

Mutational analysis and stable FIX cell line.

(PDF)

Click here for additional data file.

Analysis of the phenotype in the hepatic cell line HuH-7.

(PDF)

Click here for additional data file.

Splicing of the first truncated and the terminal FIX intron.

(PDF)

Click here for additional data file.

miRNA analysis and BLAT screening for off-target effects.

(PDF)

Click here for additional data file.

Clinical and phenotypic data of hemophilia B patients carrying the same mutation.

(PDF)

Click here for additional data file.

8 Oct 2019

Dear Dr Bohne,

Thank you very much for submitting your Research Article entitled 'Pathological mechanism and antisense oligonucleotide-mediated rescue of a non-coding variant suppressing factor 9 RNA biogenesis leading to hemophilia B' to PLOS Genetics. Your manuscript was fully evaluated at the editorial level and by independent peer reviewers. The reviewers appreciated the attention to an important problem, but raised some substantial concerns about the current manuscript. Based on the reviews, we will not be able to accept this version of the manuscript, but we would be willing to review again a much-revised version. We cannot, of course, promise publication at that time.

Should you decide to revise the manuscript for further consideration here, your revisions should address the specific points made by each reviewer. We will also require a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript.

If you decide to revise the manuscript for further consideration at PLOS Genetics, please aim to resubmit within the next 60 days, unless it will take extra time to address the concerns of the reviewers, in which case we would appreciate an expected resubmission date by email to plosgenetics@plos.org.

If present, accompanying reviewer attachments are included with this email; please notify the journal office if any appear to be missing. They will also be available for download from the link below. You can use this link to log into the system when you are ready to submit a revised version, having first consulted our Submission Checklist.

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see our guidelines.

Please be aware that our data availability policy requires that all numerical data underlying graphs or summary statistics are included with the submission, and you will need to provide this upon resubmission if not already present. In addition, we do not permit the inclusion of phrases such as "data not shown" or "unpublished results" in manuscripts. All points should be backed up by data provided with the submission.

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool.  PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

PLOS has incorporated Similarity Check, powered by iThenticate, into its journal-wide submission system in order to screen submitted content for originality before publication. Each PLOS journal undertakes screening on a proportion of submitted articles. You will be contacted if needed following the screening process.

To resubmit, use the link below and 'Revise Submission' in the 'Submissions Needing Revision' folder.

[LINK]

We are sorry that we cannot be more positive about your manuscript at this stage. Please do not hesitate to contact us if you have any concerns or questions.

Yours sincerely,

Gregory M. Cooper, PhD

Associate Editor

PLOS Genetics

Gregory Barsh

Editor-in-Chief

PLOS Genetics

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: In this paper, the authors describe their examination of the mechanism of a specific variant in the 3’ UTR of the factor 9 (F9) gene that is known to cause hemophilia B. After presenting several potential mechanisms, they pursue the hypothesis that the variant creates a U1snRNP binding site that in turn leads to inappropriate binding and loss of 3’ end processing. Through the use of various synthetic F9 wild-type and mutant expression constructs, the authors demonstrate reduction of F9 mRNA levels corresponding with the presence of the variant. Additionally the authors demonstrate that U1snRNP is necessary for this mRNA reduction, partially rescuing the phenotype when U1snRNP is inhibited. Lastly, they present data from ASOs targeting the variant itself, again demonstrating partial rescue and thereby suggesting a potential therapeutic strategy.

Overall, the paper presents a convincing argument that the variant’s deleterious effect is through U1snRNP. They present the results of many well-controlled experiments that combine to create a consistent picture of U1snRNPs role. My main concern is the use of several different cell contexts. It seems that HEK 293T, CHO, Huh7, and HeLa cells are all used in different places. It is disappointing that better data was not obtained in the liver-derived Huh7 cells that might be conjectured to be the most relevant to factor 9 biology. It appears that some cell types were better suited for protein assays and others for mRNA assays. The authors must clarify the reasons for using different cell types in the various assays. Nevertheless, I believe with these changes and a few others this manuscript is suitable for publication in PLOS Genetics.

Some minor issues:

1. p.3 line 82: “At present, the FIX mutation database lists 21 patients from different geographic regions (http://www.factorix.org; (7, 8); supplementary table 1)”. I believe “with this variant” should be added to the end of the sentence.

2. p.3 line 91: “Again several conserved sequence features the hexameric poly(A) signal (PAS) direct the assembly of the cleavage and polyadenylation (CPA) machinery.” This is confusing and should be re-worked.

3. p. 7 line 191: “The intron is efficiently spliced (Fig.192 S3A) and it strongly enhances F9 RNA and protein levels (Fig. 2B, lane 6 and 2C, lane 4).” Is there evidence of an enhancer element in this intron?

4. P. 10 line 269 “This vector change was necessary, since our optimized constructs are highly overexpressed in HeLa cells making rescue experiments difficult (data not shown)” This is a bit concerning. It ties into my general concern about the switching of cell types throughout the paper.

Reviewer #2: The authors have investigated the mechanism by which a noncoding mutation in the 3’UTR of the Factor IX (FIX) serine protease of the coagulation pathway causes disease. The nucleotide substitution was described in 1993 as a cause of moderate to severe hemophilia B with FIX activity between <1 and 4.5 % of normal. The mutant site resembled a 5’ splice site suggesting a cryptic splicing event. The authors propose that while the resemblance to a 5’ splice site recruits the U1 snRNP, the effect is not splicing but suppression of polyadenylation that is another function of U1. They use antisense oligos that bind to the mutant site to block binding of U1 and rescue polyadenylation. The authors provided reasonable evidence that the mutation does not induce use of a cryptic 5’ splice site and that in a plasmid minigene the mutation causes reduced polyadenylation leading to readthrough transcription in the plasmid. This conclusion is also supported by qRT-PCR. Antisense oligos (ASOs) that bind U1 and block interactions with 5’ splice sites and the mutant site rescue expression of FIX. ASOs and a U7 snRNP directed block of the mutation in the FIX 3’UTR provided a partial rescue. One major concern is that all work was done using transfected, or rarely, integrated minigenes. Minigenes with different architectures gave similar results but still it would be particularly convincing if the results were shown in patient cells if they are available. The results are a proof of principle but given the partial rescues in Fig 5 and the reliance on only minigenes, the results are borderline preliminary.

The results in Figs 2 and 3 could be combined and there is redundancy with the use of different minigenes

Fig 2D – clarify “supernatant”. Presumbably this is protein from the media but not clear

Using and ASO to block U1 is an unusual strategy since there should be an effect on most splicing. However here the results support the contention that the mutation affects U1 binding rather than a microRNA.

While the rescue in Fig 5 was significant, there was not as strong as effect as one might expect. Why weren’t AMOs rather than 2OMe antisense oligos were used to block the mutant site on the minigene mRNA since the AMOs worked so well on U1 snRNA? We are number of ASOS tested? Perhaps there are secondary structures of other features that prevent sufficient interaction of ASOs with the mutant site.

Reviewer #3: In this interesting mns, the Authors explore the pathological consequence of a previously reported mutation associated to hemophilia B and located in the 3’UTR of the F9 gene. Through the identification of an U1 snRNP-mediated polyA suppression mechanism that reduce the amount of F9 transcripts the Authors developed an antisense oligonucleotide rescue strategy to correct this type of defect.

The pathological consequences of mutations in the 3’UTR are difficult to study and not always obvious. Here the Authors provide convincing evidence with reporter systems that a F9 mutation in the 3’UTR affect the amount of F9 transcripts through the creation of novel U1 consensus binding site that possibly supress the polyA. The Authors have previously reported a mutation with a similar disease causing-mechanism in another gene (REF 12) and here they provide novel evidence that this type of defects can be corrected by a novel ASO-based therapeutic strategy. Even if the effect of the mutation on transcript and F9 activity as well as the ASO therapeutic approach are clear and convincing, I feel that some additional experiments should be performed to better clarify the mechanistic connection between the U1 and polyA site suppression.

Major points.

1- The Authors do not completely exclude an effect on splicing. It is essential to prove that the mutation creates a functional 5' splice site (SS) and that its recognition by the spliceosomal component U1 snRNP causes F9 mRNA suppression in the absence of splicing. Specifically, (Page 6 lane from lane 149), the 3’RACE experiments do not exclude that the mutation induces a splicing-dependent removal of the first polyA site. This splicing event might occur through the usage of the upstream mutant-created 5’ss and the cryptic downstream 3’ss. To clarify this aspect, the Authors should perform a RT-PCR experiment with primers located before and after the two sites to see if this region is spliced. More in general this part is confused and difficult to follow and should be completely rewritten (for example fig S1C does not seem to evaluate “readthrough” but only the expression levels).

2-To have a more direct evidence that U1 is involved in polyA suppression (and not splicing), I feel that the Authors should necessarily perform experiment with suppressor U1s that bind to the WT region as previously done in ref 12 and show that the suppressor U1s reduce expression. This is important as the polyA site and the U1 binding site seems quite distant (208 nucleotides).

In addition, the formal prove that the U1 affect polyA is missing as they just show less transcript and “read through”. The experiments I suggest at point 1 and 2 will be useful to reinforce the main mechanistic conclusions that U1 directly suppresses polyA.

4- Some conclusions are overestimated. In the absence of specific experiments, the conclusions should be toned down. I provide here a list of sentences that, in my opinion, need attention.

- Page 11 lane 303. I do not feel that the experiments performed really prove that the variant affect the nuclear quality control and that a failure of 3’ end processing result in RNA degradation. I suggest to tone down/ explain better in light of the available experimental evidence provided.

-Page 12 lane 336. The evidence that the ASO binding to the mutant site induce a switch to the downstream polyA site is not evident from the experiment and should be better explained.

-Page 5 lane 149. As the mutation is in the non-coding 3’UTR, I do not understand why the splicing removal of the first polyA site should induce NMD.

Minor points

- Page 3 lane 88. I think that the snRNPs involved in splicing are five and not six. (U7 is in 3’ end processing of histones)

-The rationale of using the gain of function Padua mutant in the context of the 3’UTR variant is unclear, as it seems simply an additional control.

-The Authors claim that transcription is not involved? Why increased transcription does not compensate? This is strange.

-Page 4 lane 112. I think that the most appropriate verb is “reduces” and not “impedes” F9 expression

-The Authors identified with in silico approaches two polyA sites that were then tested in expression plasmids and the effect of the mutation is on the proximal one. Is this polyA site the most used or the only one used also in vivo in hepatocytes?

-Fig 1 B and S1B lack of appropriate statistical analysis. Fig 3b lacks quantification.

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the PLOS Genetics data availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

14 Jan 2020

Submitted filename: Kroos et al_Rebuttal_plos genetics_final.doc

Click here for additional data file.

3 Feb 2020

* Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out. *

Dear Dr Bohne,

Thank you very much for submitting your Research Article entitled 'Pathological mechanism and antisense oligonucleotide-mediated rescue of a non-coding variant suppressing factor 9 RNA biogenesis leading to hemophilia B' to PLOS Genetics. Your manuscript was fully evaluated at the editorial level and by independent peer reviewers. The reviewers appreciated the attention to an important topic but identified some aspects of the manuscript that should be improved.

We therefore ask you to modify the manuscript according to the review recommendations before we can consider your manuscript for acceptance.

You will notice that while two of the reviewers had no additional comments, one reviewer has some remaining concerns that primarily focus on a few interpretive and organizational issues.  We do not believe any additional experimentation or major alterations will be required to address these comments.  However, we are hopeful you can address these points with a few minor revisions, and we are likely to solicit one final review from this reviewer before making a final decision.

In addition we ask that you:

1) Provide a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript.

2) Upload a Striking Image with a corresponding caption to accompany your manuscript if one is available (either a new image or an existing one from within your manuscript). If this image is judged to be suitable, it may be featured on our website. Images should ideally be high resolution, eye-catching, single panel square images. For examples, please browse our archive. If your image is from someone other than yourself, please ensure that the artist has read and agreed to the terms and conditions of the Creative Commons Attribution License. Note: we cannot publish copyrighted images.

We hope to receive your revised manuscript within the next 30 days. If you anticipate any delay in its return, we would ask you to let us know the expected resubmission date by email to plosgenetics@plos.org.

If present, accompanying reviewer attachments should be included with this email; please notify the journal office if any appear to be missing. They will also be available for download from the link below. You can use this link to log into the system when you are ready to submit a revised version, having first consulted our Submission Checklist.

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

Please be aware that our data availability policy requires that all numerical data underlying graphs or summary statistics are included with the submission, and you will need to provide this upon resubmission if not already present. In addition, we do not permit the inclusion of phrases such as "data not shown" or "unpublished results" in manuscripts. All points should be backed up by data provided with the submission.

PLOS has incorporated Similarity Check, powered by iThenticate, into its journal-wide submission system in order to screen submitted content for originality before publication. Each PLOS journal undertakes screening on a proportion of submitted articles. You will be contacted if needed following the screening process.

To resubmit, you will need to go to the link below and 'Revise Submission' in the 'Submissions Needing Revision' folder.

[LINK]

Please let us know if you have any questions while making these revisions.

Yours sincerely,

Gregory M. Cooper, PhD

Associate Editor

PLOS Genetics

Gregory Barsh

Editor-in-Chief

PLOS Genetics

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: I am satisfied that the authors have addressed the concerns I brought up.

Reviewer #2: the authors have addressed my concerns

Reviewer #3: The mns is significantly improved and the Authors have appropriately addressed most of the criticism raised also performing novel experiments. However there are still some points that needs attention.

1 In my opinion the Authors have now sufficient evidence to prove that the creations of a consensus 5’ss in the 3’ UTR reduces the total amount of mRNA through an U1snRNP-dependent inhibition of polyadenylation as well as to exclude that the variant induces aberrant splicing. The other potential mechanisms the Authors exclude, miRNA, RBPs and NMD, are not experimentally validated (miRNA and RBPs only in silico and NMD just because splicing is not involved). In particular, for NMD the minigene systems in general (and in particular the one used here, which has no introns) are not NMD sensitive. My suggestion is to focus, in the result section, on the experimentally validated mechanism first and then say that the other mechanisms, based on indirect evidence, are very unlikely.

2 The Authors have correctly analyzed the effect of the suppressor U1 on the WT mutant clearly showing that it significantly reduces the amount of mRNA. I believe this is an important and critical experiment that prove the direct involvement of U1 and accordingly it could be inserted in the main figures. (If there is no space one of the two ASO approaches that block the U1 snRNP should go into supplementary). However, I feel that there is a mistake in the data interpretation at page 10 lanes 270-273 concerning the effect of suppressor U1s on the mutant minigene. I do not see an increase in RNA expression after transfection of the suppressor U1 and I do not understand why a “stronger F9 mutant mRNA increase” should be expected in this case. The Authors should note that the suppressor U1 makes a good complementarity also with the mutant sequence through a wobble G-U base pairing, thus I would expect a further reduction in the mRNA. In any case, the suppressor U1 experiment are not designed to block the U1 snRNP and thus this experiment should be presented in a separate result section.

3 Unfortunately, the half-life experiments do not prove a co-transcriptionally decay of the mutant F9 mRNA. The mutation clearly reduces but not completely disrupt the usage of the PAS, producing a low amount of normal mature F9 mRNA which has, as expected, a normal half-life. Thus, in the absence of more direct measurement (like Chip assay with PolII or nascent nuclear RNA) the conclusions should be better explained/modified accordingly.

4 The ASO and the U7 for masking the mutated site seem to induce additional bands (Fig 5 B and D) which are considered to originate from the use of downstream PASs. The Authors speculate in discussion that the ASO might prevent binding of a cellular factor. In the absence of more defined prove of their origin and considering that they are using hybrid minigenes, I suggest to be more cautious in their interpretation. I also see another unclear band in mutant untreated.

**********

Have all data underlying the figures and results presented in the manuscript been provided?

Large-scale datasets should be made available via a public repository as described in the PLOS Genetics data availability policy, and numerical data that underlies graphs or summary statistics should be provided in spreadsheet form as supporting information.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

13 Feb 2020

Submitted filename: Krooss et al_2nd_Rebuttal_plos genetics_final.doc

Click here for additional data file.

22 Feb 2020

Dear Dr Bohne,

We are pleased to inform you that your manuscript entitled "Pathological mechanism and antisense oligonucleotide-mediated rescue of a non-coding variant suppressing factor 9 RNA biogenesis leading to hemophilia B" has been editorially accepted for publication in PLOS Genetics. Congratulations!

Before your submission can be formally accepted and sent to production you will need to complete our formatting changes, which you will receive in a follow up email. Please be aware that it may take several days for you to receive this email; during this time no action is required by you. Please note: the accept date on your published article will reflect the date of this provisional accept, but your manuscript will not be scheduled for publication until the required changes have been made.

Once your paper is formally accepted, an uncorrected proof of your manuscript will be published online ahead of the final version, unless you’ve already opted out via the online submission form. If, for any reason, you do not want an earlier version of your manuscript published online or are unsure if you have already indicated as such, please let the journal staff know immediately at plosgenetics@plos.org.

In the meantime, please log into Editorial Manager at https://www.editorialmanager.com/pgenetics/, click the "Update My Information" link at the top of the page, and update your user information to ensure an efficient production and billing process. Note that PLOS requires an ORCID iD for all corresponding authors. Therefore, please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field.  This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager.

If you have a press-related query, or would like to know about one way to make your underlying data available (as you will be aware, this is required for publication), please see the end of this email. If your institution or institutions have a press office, please notify them about your upcoming article at this point, to enable them to help maximise its impact. Inform journal staff as soon as possible if you are preparing a press release for your article and need a publication date.

Thank you again for supporting open-access publishing; we are looking forward to publishing your work in PLOS Genetics!

Yours sincerely,

Gregory M. Cooper, PhD

Associate Editor

PLOS Genetics

Gregory Barsh

Editor-in-Chief

PLOS Genetics

www.plosgenetics.org

Twitter: @PLOSGenetics

----------------------------------------------------

Comments from the reviewers (if applicable):

----------------------------------------------------

Data Deposition

If you have submitted a Research Article or Front Matter that has associated data that are not suitable for deposition in a subject-specific public repository (such as GenBank or ArrayExpress), one way to make that data available is to deposit it in the Dryad Digital Repository. As you may recall, we ask all authors to agree to make data available; this is one way to achieve that. A full list of recommended repositories can be found on our website.

The following link will take you to the Dryad record for your article, so you won't have to re‐enter its bibliographic information, and can upload your files directly:

http://datadryad.org/submit?journalID=pgenetics&manu=PGENETICS-D-19-01514R2

More information about depositing data in Dryad is available at http://www.datadryad.org/depositing. If you experience any difficulties in submitting your data, please contact help@datadryad.org for support.

Additionally, please be aware that our data availability policy requires that all numerical data underlying display items are included with the submission, and you will need to provide this before we can formally accept your manuscript, if not already present.

----------------------------------------------------

Press Queries

If you or your institution will be preparing press materials for this manuscript, or if you need to know your paper's publication date for media purposes, please inform the journal staff as soon as possible so that your submission can be scheduled accordingly. Your manuscript will remain under a strict press embargo until the publication date and time. This means an early version of your manuscript will not be published ahead of your final version. PLOS Genetics may also choose to issue a press release for your article. If there's anything the journal should know or you'd like more information, please get in touch via plosgenetics@plos.org.

23 Mar 2020

PGENETICS-D-19-01514R2

Pathological mechanism and antisense oligonucleotide-mediated rescue of a non-coding variant suppressing factor 9 RNA biogenesis leading to hemophilia B

Dear Dr Bohne,

We are pleased to inform you that your manuscript entitled "Pathological mechanism and antisense oligonucleotide-mediated rescue of a non-coding variant suppressing factor 9 RNA biogenesis leading to hemophilia B" has been formally accepted for publication in PLOS Genetics! Your manuscript is now with our production department and you will be notified of the publication date in due course.

The corresponding author will soon be receiving a typeset proof for review, to ensure errors have not been introduced during production. Please review the PDF proof of your manuscript carefully, as this is the last chance to correct any errors. Please note that major changes, or those which affect the scientific understanding of the work, will likely cause delays to the publication date of your manuscript.

Soon after your final files are uploaded, unless you have opted out or your manuscript is a front-matter piece, the early version of your manuscript will be published online. The date of the early version will be your article's publication date. The final article will be published to the same URL, and all versions of the paper will be accessible to readers.

Thank you again for supporting PLOS Genetics and open-access publishing. We are looking forward to publishing your work!

With kind regards,

Jason Norris

PLOS Genetics

On behalf of:

The PLOS Genetics Team

Carlyle House, Carlyle Road, Cambridge CB4 3DN | United Kingdom

plosgenetics@plos.org | +44 (0) 1223-442823

plosgenetics.org | Twitter: @PLOSGenetics