Adult C3H and C57Bl/6J mice were purchased from Jackson laboratories (Bar Harbor, ME, USA). Mice were bred and maintained according to the Yale University animal research requirements, and all procedures were approved by the Institutional Animal Care and Use Committee (protocol number # 2005-07222). Mice were fed at libitum and housed under a 12 h light cycle.

Mouse tail genomic DNA was extracted by standard methods (Laird et al., 1991). Briefly, tail tips were cut and digested overnight with proteinase K (100 µg/ml) in 0.5 ml lysis buffer (100 mM Tris HCl pH 8.5, 5 mM EDTA, 0.2% SDS, 200 mM NaCl) at 55°C with rocking. Following vortexing and centrifugation to pellet hair, 0.5 ml of 100% isopropanol was added to the supernatant and chromosomal DNA was aggregated into a visible clump by rocking manually at room temperature. The DNA was then spooled and dissolved in 0.5 ml TE buffer.

Genotyping for exon 9 single nucleotide polymorphism (SNP), rs55858A/G was performed by polymerase chain reaction (PCR) amplification of DNA using primers exon9F: 5′-TCTACAGCCCAGGCTGTGTAC-3′ and IVS9R: 5′-TAGCAATTTAGCCAAAAGAGTTTC-3′, followed by HaeIII restriction enzyme digestion (New England Biolabs, Ipswich, MA, USA). Amplifications were carried out using 35 cycles of PCR, in which the initial 5 min for denaturation at 94°C was followed by a touch-down program for 10 cycles of 92°C/30 s, 65°C/20 s (−1°C per cycle) and 72°C/30 s and then 25 cycles of 92°C/30 s, 55°C/20 s and 72°C/30 s, in a 50 µl reaction mix containing 1× PCR buffer (Ambion, Austin, TX, USA), 125 µM dNTP, 0.5 µM primers and 1 unit of SuperTaq polymerase (Ambion). Digested samples were run on a 2% agarose gel and visualized by ethidium bromide staining.

Total RNA was isolated after homogenization of adult mouse ovaries in Trizol (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions and kept at −80°C until use. RNA was treated for genomic DNA contamination using DNase I (Ambion). The quality and concentration of the RNA was determined by measuring the absorbance at 260 and 280 nm. Equal amounts of total RNA (2 µg) were reverse-transcribed to cDNA using oligo d(T) priming at 37°C for 1 h (Omniscript, Qiagen, Valencia, CA, USA). Then, PCRs were performed on cDNAs using Taq DNA polymerase (Qiagen) with specific primer pairs flanking the alternatively spliced exons 10 and 13, respectively, as described in Results. The sequences of primers were: exon8F: 5′-ACGAAACCGCTCTATGTGGCCC-3′, exon11R: 5′-TGGGGACTCCAGCAATAA CAGAAGC-3′ and exon14R: 5′-CCAGGAACCCACAGAGTTCAG-3′, PCR conditions were as above.

Spliced and unspliced forms were gel fractionated in 1.5% agarose and visualized by ethidium bromide staining. PCR fragments were extracted from agarose gel using a gel extraction kit (Roche Applied Science, Indianapolis, IN, USA) according to the manufacturer’s protocol, and sequenced at the W.M. Keck Foundation Biotechnology Resource Laboratory at Yale University. Sequencing results were compared with mouse Epab genomic sequence using Sequencher software version 4.7 (Gene Codes, Ann Arbor, MI, USA).

Gene expression studies by quantitative real-time PCR (qRT-PCR) were carried out on an iCycler (Bio-Rad, Hercules, CA, USA). Tissue cDNA, prepared as described above, was assayed in triplicate. Each experiment was repeated three times and five mutant animals were compared with five litter mate controls. Each 25 µl reaction contained 12.5 µl of 2× SybrGreen supermix (Biorad), 0.4 µM of each primer and 1 µl of template.

In order to quantitate the expression of different splicing variants of Epab, the following primers were used: exon9–11F: 5′-GGAACCTCAGCTGTACACAGTG-3′, exon 12R: 5′-CTGATGGCCAGGGACATACAC-3′, exon10F: 5′-AGCCAGGTGCCT CACACGCAG-3′, exon8F: 5′-ACGAAACCGCTCTATGTGGCCC-3′ and exon9R: 5′-CCTGGGGTCAGGTTGCATTGG-3′. Expression of the target gene was normalized to β-actin levels using primers 5′-GACCTCTATGCAACACAGT-3′ and 5′-TTGCTGATCCACATCTGCT-3′. The ΔΔCt (cycle threshold) method was used to calculate relative expression levels. Results were reported as a fold change in gene expression between different genotypes.

The ESE finder program (version 2.0, http://rulai.cshl.edu/cgi-bin/tools/ESE3/esefinder.cgi?process=home) (Cartegni et al., 2003) (Cold Spring Harbor Laboratory, NY, USA) was used to identify ESE consensus sequences.

Since the data from qRT-PCR were normally distributed (as determined by Kolmogorov–Smirnov test), the Student t-test was used. Statistical calculations were performed using SigmaStat for Windows, version 3.0 (Jandel Scientific Corp., San Rafael, CA, USA). Statistical significance was defined as P < 0.05.

A splicing variant of mouse Epab lacking exon 13 has previously been reported (Seli et al., 2005) (Fig. 1A). In the present study, RT–PCR in mouse ovary using primers on exons 8 and 11 revealed two Epab-specific bands (Fig. 1B). Sequencing of the lower band revealed an additional splicing variant of the Epab transcript lacking exon 10 (c.Ex10del).

Moreover, amplification from exon 8 to exon 14 flanking both alternatively spliced exons, produced four Epab-specific bands. Sequencing revealed that these bands corresponded to the full-length Epab, alternatively spliced forms lacking exon 10 or exon 13, alternatively spliced form lacking both exons 10 and 13 and a heteroduplex band (Fig. 1C). Domain analysis of the predicted protein revealed that the alternatively spliced form of mouse Epab mRNA without exon 13 lacks the PABP domain, whereas the spliced form of Epab without exon 10 encodes a full PABP domain (Fig. 2, Table I).

The mice used for the identification of the c.Ex10del variant were progeny of mated CH3 with C57Bl/6J mice and were heterozygous for an exon 9 SNP (rs55858A/G). SNP genotyping was performed as described in Materials and Methods, using genomic PCR with primers 9F and IVS9R (Fig. 3A and B), followed by HaeIII digestion (Fig. 3C and D). The presence of rs55858G created a new HaeIII site and resulted in a 240 bp fragment, whereas the rs55858A generated for a 307 bp fragment upon digestion (Fig. 3C and D). This SNP allowed us to determine that, consistent with bi-allelic gene expression observed for most genes, Epab mRNA containing exon 10 was expressed from both alleles (Fig. 4A). However, c.Ex10del Epab was expressed only from the allele containing the rs55858A at the exon 9 SNP (Fig. 4A). This observation could be explained by three possible mechanisms: imprinting, RNA editing or ESEs.

To determine if monoallellic expression of c.Ex10del Epab is mediated by imprinting, we examined heterozygous (rs55858A/G) mice that inherited the rs55858A allele from either the mother or the father (Fig. 4A and B). Regardless of the parental origin of the allele, the shorter Epab transcript lacking exon 10, contained the rs55858A SNP, ruling out imprinting as the regulatory mechanism (Fig. 4A and B).

Next, we tested the possibility of RNA editing mechanism, which is a post-transcriptional base substitution (C to U, A to I) catalyzed by enzymes called RNA-editing deaminases (Maydanovych and Beal, 2006). We bred homozygous (rs55858G/G) mice for the exon 9 SNP. In mice homozygous rs55858G/G SNP, we found that the c.Ex10del variant also contained (G). Therefore, while in mice heterozygous for the exon 9 SNP (rs55858A/G), exon 10 skipping preferentially occurred from the rs55858A allele (Fig. 4A and B), alternative splicing still occurred when both alleles contained G (rs55858G/G) at position rs55858 (Fig. 4C). Consequently, RNA editing mechanism was eliminated.

To identify the origin of the monoallelic exon skipping, we performed further sequence analysis and identified an additional SNP in exon 10 (rs56574G/C) that co-segregated with the exon 9 SNP (rs55858A/G). The two SNPs are in ‘linkage disequilibrium’ as rs55858A allele in exon 9 is always inherited together with rs56574G allele on exon 10, and rs55858G on exon 9 with rs56574C on exon 10.

Preferential splicing can be regulated by the presence of ESEs within the sequence. To identify ESEs that are recognized by individual splicing regulatory proteins (SF2/ASF, SC35, SRp40 and SRp55), we used the ESE finder program version 2.0 (http://rulai.cshl.edu/cgi-bin/tools/ESE3/esefinder.cgi?process=home). In this program, consensus motifs obtained with the four SR proteins are detected and the thresholds represent values above which a score for a given sequence is considered to be significant (high score motif). In silico analysis using ESE finder program revealed that the presence of rs56574G leads to the formation of a consensus sequence [CCTGAGG] characteristic of an ESE that binds splicing regulatory protein SRp40, resulting in efficient exclusion of exon 10, as the binding affinity of SRp40 protein for the CCTGAGG sequence is 3.19 whereas it is lower than thresholds value for CCTGAGC sequence.

The effect of the ESE on the efficiency of alternative splicing in different genetic backgrounds was evaluated by qRT-PCR. Mouse strain C57BL/6 is homozygous for rs55858A and rs56574G alleles, and C3H is homozygous for rs55858G and rs56574C alleles.

In order to quantitate the expression of the c.Ex10del variant, we used a forward primer (exon9–11F) that was complementary to exon 9 at its 5′ end and to exon 11 at its 3′ end, and would not amplify Epab mRNA that contains exon 10. Reverse primer (exon12R) was in exon 12. This primer set resulted in a robust and sensitive qRT-PCR reaction with a correlation coefficient of 0.942, and a slope of –3.356, corresponding to PCR efficiency of 98.6%.

Expression of Epab that contains exon 10 was quantified using primers on exons 10 and 12 (exon10F and exon12R). This PCR reaction had a standard curve with a correlation coefficient of 0.999, a slope of −3.395 and a PCR efficiency of 97%. Total Epab mRNA expression (including all forms) was quantified using forward and reverse primers at exons 8 and 9, respectively. This resulted in a standard curve with a correlation coefficient of 0.99, a slope of –3.310 and a PCR efficiency of 99.5%.

Quantitative RT-PCR revealed a 5-fold increase in the amount of the c.Ex10del variant in the animals carrying the enhancer (homozygous rs56574G/G) for the exon 10 SNP compared with the mice that did not (homozygous rs56574C/C) at the same locus (P < 0.001) (Fig. 5A). This was accompanied by a decrease in the amount of the unspliced Epab transcript in the animals that contained the ESE compared with mice that did not have the ESE (P < 0.01) (Fig. 5B). There was no significant difference in the total Epab mRNA content of these two different genotypes (Fig. 5C and D), suggesting that the rs56574G SNP resulted in an increase in alternative splicing without changing overall of Epab expression.