Evidence that a consensus element found in naturally intronless mRNAs promotes mRNA export

Abstract

We previously showed that mRNAs synthesized from three genes that naturally lack introns contain a portion of their coding sequence, known as a cytoplasmic accumulation region (CAR), which is essential for stable accumulation of the intronless mRNAs in the cytoplasm. The CAR in each mRNA is unexpectedly large, ranging in size from ∼160 to 285 nt. Here, we identified one or more copies of a 10-nt consensus sequence in each CAR. To determine whether this element (designated CAR-E) functions in cytoplasmic accumulation of intronless mRNA, we multimerized the most conserved CAR-E and inserted it upstream of β-globin cDNA, which is normally retained/degraded in the nucleus. Significantly, the tandem CAR-E, but not its antisense counterpart, rescued cytoplasmic accumulation of β-globin cDNA transcripts. Moreover, dinucleotide mutations in the CAR-E abolished this rescue. We show that the CAR-E, but not the mutant CAR-E, associates with components of the TREX mRNA export machinery, the Prp19 complex and U2AF2. Moreover, knockdown of these factors results in nuclear retention of the intronless mRNAs. Together, these data suggest that the CAR-E promotes export of intronless mRNA by sequence-dependent recruitment of the mRNA export machinery.

Figure 1.

CARs in naturally intronless mRNAs contain a conserved motif. (A) The 10-nt consensus element (CAR-E) identified in the CARs from HSPB3, c-Jun, IFNα1 or IFNβ1 is shown. (B) The locations of the CAR-Es in each CAR are indicated by the boxes. The height of box is proportional to −log(P-value), truncated at the height for a motif with a P-value of 1 × 10⁻¹⁰ (34). The numbering is relative to the first nucleotide of each CAR sequence. (C) The sequences of each of the CAR-Es (colored bases) surrounded by flanking sequence (black bases) are shown.

Figure 2.

Tandem CAR-E promotes cytoplasmic accumulation of β-globin cDNA and β-globin protein expression. (A) Schematic of CMV-β-globin cDNA construct indicating the position where the CAR-E and CAR-E_AS were inserted. The start codon (ATG), HA tag and BGH polyA signal (pA) are indicated. The sizes of each of the exons (in nucleotides) are shown. (B) FISH was used to determine the nucleocytoplasmic distribution of the indicated transcripts 24 h after transient transfection of the respective constructs into HeLa cells. DAPI staining was used to identify the nucleus. Scale bar: 10 µm. (C) Low magnification of the data shown in panel B. (D) Western analysis at 24 h (Day 1) or 45 h (Day 2) after transient transfection of the indicated constructs into HeLa cells. HA-β-globin, co-transfection control EGFP and tubulin (loading control) are indicated.

Figure 3.

Tandem CAR-E containing dinucleotide substitutions is non-functional in RNA export. (A) Schematic of CMV-β-globin cDNA construct showing sequences of the CAR-E and each of the mutants that were inserted into the cDNA. Mutated nucleotides are shown in lower case. (B) Western analysis at 24 h (Day 1) or 45 h (Day 2) after transient transfection of the indicated constructs into HeLa cells. HA-β-globin and tubulin (loading control) are indicated. (C) Nucleocytoplasmic distribution of CAR-E_m1-5 transcripts was determined by FISH. DAPI staining was used to identify the nucleus. Scale bar: 10 µm.

Figure 4.

RNAi of Prp19 complex components or U2AF2 blocks export of naturally intronless IFNβ1 mRNAs. (A) Schematic of CMV-IFNβ1 construct used for microinjection. The length of the 5′-UTR, coding region and 3′-UTR are indicated. Vector sequences are shown as gray lines. BGH pA: bovine growth hormone polyA site. (B, D, J) Western blots showing levels of XAB2 (B), AQR (D) or U2AF2 (J) after transfecting HeLa cells with targeting or non-targeting (Cntl) siRNAs as indicated. Tubulin was used as a loading control. (F and H) RT–PCR showing levels of ISY1 or CRNKL1 mRNAs after transfecting HeLa cells with targeting or non-targeting (Cntl) siRNAs as indicated. Endogenous DNAJB1 was amplified by PCR as a loading control. Reverse transcription was carried out in reactions that contained (+) or lacked (−) reverse transcriptase (RTase) followed by PCR. (C, E, G, I, K) CMV-IFNβ1 naturally intronless construct was microinjected into nuclei of the indicated knockdown or control cells. FISH was carried out to determine the nucleocytoplasmic distribution of the RNAs. Dextran 70 kDa was used an injection marker. Scale bar: 10 µm.