U1 snRNP protects pre-mRNAs from premature cleavage and polyadenylation

Abstract

In eukaryotes, U1 small nuclear ribonucleoprotein (snRNP) forms spliceosomes in equal stoichiometry with U2, U4, U5 and U6 snRNPs; however, its abundance in human far exceeds that of the other snRNPs. Here we used antisense morpholino oligonucleotide to U1 snRNA to achieve functional U1 snRNP knockdown in HeLa cells, and identified accumulated unspliced pre-mRNAs by genomic tiling microarrays. In addition to inhibiting splicing, U1 snRNP knockdown caused premature cleavage and polyadenylation in numerous pre-mRNAs at cryptic polyadenylation signals, frequently in introns near (<5 kilobases) the start of the transcript. This did not occur when splicing was inhibited with U2 snRNA antisense morpholino oligonucleotide or the U2-snRNP-inactivating drug spliceostatin A unless U1 antisense morpholino oligonucleotide was also included. We further show that U1 snRNA-pre-mRNA base pairing was required to suppress premature cleavage and polyadenylation from nearby cryptic polyadenylation signals located in introns. These findings reveal a critical splicing-independent function for U1 snRNP in protecting the transcriptome, which we propose explains its overabundance.

Figure 1. U1 AMO binds to the 5′ sequence of U1 snRNA and inhibits its splicing activity

(a) HeLa cells were transfected with the indicated concentrations of control and U1 AMO for 8 hrs. RNase H protection assay was performed using total cell extracts and U1 snRNA was detected by Northern blotting. (b) In situ hybridization was performed on HeLa cells transfected with varying concentrations of U1 AMO as indicated for 8 hrs using a biotin-labeled LNA probe to the U1 snRNA (left panels) followed by fluorescent Alexa Fluor 594 streptavidin conjugate detection. Nuclei were visualized with DAPI (middle panels) and merged images are shown (right panels). (c) [α–³²P] UTP-labeled Ad2ΔIVS pre-mRNA was spliced in vitro in the presence of control or U1 AMOs at the indicated concentrations. Splicing product identities are depicted to the right of the gel.

Figure 2. Genomic tiling arrays identify unspliced pre-mRNAs following U1 AMO and spliceostatin A (SSA) treatment

RNA samples prepared from control or U1 AMOs-transfected cells (7.5 μM, 8 hrs) or SSA (100 ng/ml, 8 hrs)-treated cells were analyzed using genomic tiling array. Fold-changes (log2) of signal intensities of U1 AMO-transfected and SSA-treated cells compared to control cells are shown above the corresponding structure of each gene. With a scale shown below, gene structures are depicted in red, with horizontal lines indicating introns and boxes indicating exons. The middle part of CBFB gene (~35 kb) was removed. White arrows indicate points showing abrupt drop of the signal (inflection points).

Figure 3. Premature termination in introns of pre-mRNAs in U1 AMO transfected cells

Representative examples of genes differentially affected by U1 AMO and SSA. The sudden drop in signals in U1 AMO-transfected cells is indicated by green arrow heads. With a scale shown below, gene structures are depicted in red with horizontal lines indicating introns and boxes indicating exons.

Figure 4. The prematurely terminated pre-mRNAs are polyadenylated from cryptic PASs in introns

(a) 3′ RACE using nested PCR was performed to detect polyadenylated mRNAs using total RNA from U1 AMO (7.5 μM, 8 hrs)-transfected cells. Sequencing results of the 3′ RACE product for the NR3C1 and STK17A genes are shown with the corresponding genomic sequence in black. The poly(A) tails are shaded and the putative PASs are indicated in red line boxes. (b) HeLa cells were transfected with the wild-type and PAS mutant (mutated from AAUAAA to GAAUUC) NR3C1 mini-gene construct followed by either control or U1 AMOs. 3′ RACE was performed as described in (a). The mini-gene structure is depicted to the above of the gel, with a blue arrow indicating the forward primer for 3′ RACE. Gray arrowhead: PAS mutation. PCPA: premature cleavage and polyadenylation. Band sizes (bp) are indicated to the left of the gel.

Figure 5. U1 snRNP suppression of premature cleavage and polyadenylation from a nearby cryptic PAS is splicing independent and requires base pairing

(a) 3′ RACE was carried out as described in Figure 4a on the endogenous NR3C1 and STK17A genes using RNA samples from HeLa cells transfected with control, U1 (7.5 μM, 8 hrs) or U2 AMO (25 μM, 8 hrs), or treated with SSA (100 ng/ml, 8hrs). (b) 3′ RACE was carried out as described in Figure 4a using RNA samples from HeLa cells transfected with control or U1 AMOs (7.5 μM) with or without SSA (100 ng/ml) for 8 hrs. (c) HeLa cells were transfected with wild-type and 5′ splice site mutant NR3C1 (mutated from AAGGTAAGA to GTCCATTCA) mini-gene. 3′ RACE was performed as described in Figure 4a. The mini-gene structure is depicted. Blue arrow: forward primer to detect PCPA; Black arrow: forward primer to detect polyadenylation at 3′ end; Gray arrowhead: 5′ splice site mutation; Green and yellow arrowheads: polyadenylation signals in intron and at 3′ end, respectively. An unspliced and normally cleaved product was too large to detect. (d) Quantitation of PCPA and normal polyadenylation at 3′ end in control and U1 AMO treated cells was performed using real-time PCR. Error bars indicate s.d. (n=3). PCPA in panel a, b, c and d: premature cleavage and polyadenylation.