Arabidopsis DEAD-box RNA helicase UAP56 interacts with both RNA and DNA as well as with mRNA export factors

Abstract

The DEAD-box protein UAP56 (U2AF65-associcated protein) is an RNA helicase that in yeast and metazoa is critically involved in mRNA splicing and export. In Arabidopsis, two adjacent genes code for an identical UAP56 protein, and both genes are expressed. In case one of the genes is inactivated by a T-DNA insertion, wild type transcript level is maintained by the other intact gene. In contrast to other organisms that are severely affected by elevated UAP56 levels, Arabidopsis plants that overexpress UAP56 have wild type appearance. UAP56 localises predominantly to euchromatic regions of Arabidopsis nuclei, and associates with genes transcribed by RNA polymerase II independently from the presence of introns, while it is not detected at non-transcribed loci. Biochemical characterisation revealed that in addition to ssRNA and dsRNA, UAP56 interacts with dsDNA, but not with ssDNA. Moreover, the enzyme displays ATPase activity that is stimulated by RNA and dsDNA and it has ATP-dependent RNA helicase activity unwinding dsRNA, whereas it does not unwind dsDNA. Protein interaction studies showed that UAP56 directly interacts with the mRNA export factors ALY2 and MOS11, suggesting that it is involved in mRNA export from plant cell nuclei.

Figure 1. Expression of the UAP56 genes.

A rtPCR analysis of RNA isolated from Col-0, uap56a-3 and uap56b-1 plants using primers P7 and P8 (cf. Figure S3), annealing at regions that are common to both UAP56 genes. As a reference the transcript of the house keeping gene UBQ5 was amplified and as negative control a reaction without addition of reverse transcriptase is shown. PCR amplification was allowed to proceed for a different number of cycles (27x, 30x, 33x). B The UAP56 sequences of the cDNA fragments amplified from the three genotypes (cf. Part A of this figure) were determined by DNA sequencing. Within the shown region, the sequences of UAP56a and UAP56b are identical except for the three positions indicated by arrows. C Production of recombinant UAP56 and an UAP56 antiserum. Analysis of purified recombinant UAP56 (and the reference protein HMGB2) by SDS-PAGE in a 12% polyacrylamide gel, which was stained with Coomassie (left panel). Immunoblot analysis of recombinant UAP56 (and HMGB2 as a reference) using the UAP56-antiserum (right panel). Full-length recombinant UAP56 is indicated by an arrow. D Comparable amounts of protein extracts of Col-0 and the two T-DNA insertion mutant lines were analysed by SDS-PAGE and Coomassie staining (left panel) and by immunoblot analysis using the UAP56-antiserum (right panel). Arabidopsis UAP56 detected by the antibody is indicated by an arrow.

Figure 2. Arabidopsis UAP56 localises to the nucleus in protoplasts and root cells.

Tobacco BY2 cell protoplasts were transformed with constructs driving the expression of the indicated GFP fusion proteins and GFP fluorescence was visualised by CLSM (A,B; size bar: 10 µm). GFP fluorescence of a nuclear HMGB protein and overlay with the corresponding bright field image (A). GFP fluorescence of a UAP56-GFP fusion (B). Analysis of Arabidopsis Col-0 root tip cells by immunofluorescence microscopy using different antibodies and DAPI staining (E–H, size bar: 5 µm). Immunostaining of fibrillarin and UAP56, as well as visualisation of the DNA by DAPI staining of the same nucleus (C–E). A merge of the three images is shown in (F) and examples of brightly DAPI-stained heterochromatic chromocenters are indicated by arrows. Immunostaining of RNAPII and UAP56 of the same nucleus (G,H).

Figure 3. Arabidopsis UAP56 binds ssRNA, dsRNA and dsDNA, but not ssDNA.

A Interaction with nucleic acids measured by MST. Fuorescently labelled 29-nt oligonucleotides of dsRNA, ssRNA, dsDNA and ssDNA were incubated with increasing concentrations (0.01–80 µM) of UAP56 (or as a reference with 0.2–411 µM Df31). Protein-nucleic acid interactions were quantified by MST and binding data are plotted using the Hill equation. Data represent the mean +/− SD of at least three technical replicates. B Interaction with nucleic acids analysed by EMSA. ³²P-labelled 25-nt oligonucleotides were incubated without protein addition or with increasing concentrations (0.25–5 µM, except in case of ssDNA 10–30 µM) of UAP56 (or as a reference with 35 µM Df31). Protein binding was detected after electrophoresis by phosphorimaging and appearing protein-nucleic acid complexes are indicated by arrow heads.

Figure 4. UAP56 associates with loci transcribed by RNA polymerase II independent from the presence of introns, but not with non-transcribed regions.

A Chromatin was immunoprecipitated with the UAP56 antiserum and the corresponding preimmune serum (PI). DNA purified from the precipitated samples or input chromatin was examined by PCR using gene-specific primer combinations (cf. Table S1). B Chromatin was immunoprecipitated without addition of antibody (mock) or using antibodies against histone H3, the SPT16 subunit of FACT and UAP56. DNA purified from the precipitated samples was examined by PCR and the presence/absence of introns at the respective locus is indicated. For comparison two non-transcribed retrotransposons were analysed. Representative PCR analyses based on three independent ChIP experiments are shown.

Figure 5. Arabidopsis UAP56 has ATPase acitivity that is stimulated by RNA and dsDNA.

A ATP hydrolysis by UAP56 (or the control protein HMGB2, each 8 µM) was tested in the presence (absence) of ssRNA (R13, 50 µM). ATP and ADP were separated by TLC and analysed by phosphorimaging. B Stimulation of the ATPase activity of UAP56 (8 µM) by different amounts of ssRNA was quantified from three independently repeated experiments, and error bars indicate standard diviations. C ATPase activity of different concentrations of recombinant UAP56 in the presence of ssRNA (R13, 50 µM). D Stimulation of the ATPase activity by different RNAs (cf. Table S2). The effect of ssRNA vs. dsRNA and of short (13-nt) vs. longer (25-nt) RNA (50 µM each) on the ATPase activity of UAP56 (4 µM). As evaluated using Students t-test, the different RNAs stimulate the ATPase activity to different extents (P<0.05).E ATP hydrolysis of UAP56 in the absence of nucleic acids and in the presence of 13-nt ssRNA or 25-bp dsDNA.

Figure 6. Arabidopsis UAP56 has ATP-dependent RNA helicase activity.

A dsRNA unwinding activity of recombinant UAP56 (12 µM) in the presence of R13ds RNA in the absence or presence of different NTPs (3 mM, as indicated). dsRNA and ssRNA were separated by polyacrylamide electrophoresis and analysed by phosphorimaging. B The dsRNA unwinding activity of different amounts of recombinant UAP56 (0–12 µM) was analysed in the presence of ATP and a 13-nt dsRNA substrate (R13ds). Helicase activity was quantified from three independently repeated experiments, and error bars indicate the calculated standard diviations. C The dsRNA unwinding activity of 12 µM UAP56 was monitored for different times in the presence of ATP and a 13-nt dsRNA substrate (R13ds). D dsRNA unwinding activity of UAP56 (0 or 12 µM) in the presence of ATP and different dsRNAs (cf.Table S2). Helicase activity was quantified from three independently repeated experiments, and error bars indicate the calculated standard diviations. E The dsDNA unwinding activity of different amounts of recombinant UAP56 (0–20 µM) was analysed in the presence of ATP and a 13-nt dsDNA substrate (D13ds).

Figure 7. Arabidopsis UAP56 interacts with mRNA export factors.

A Yeast two-hybrid assays. Yeast cells cotransformed with the indicated plasmids were grown on SD/-Leu/-Trp/-His medium (left panels). Cells grown on the same medium were used to detect β-galactosidase reporter gene activity by colony-lift filter assays and incubation of the filters with the X-gal substrate (right panels). B Purified GST and GST-UAP56 used for pull-down assays were analysed by SDS-PAGE and Coomassie staining. C Pull-down assays with GST and GST-UAP56. Recombinant MOS11, HMGB2 or in vitro translated ³⁵S-Met-labelled ALY2 were incubated in the presence of ATP with GST and GST-UAP56, which were bound to glutathione cellulose beads. After washing the beads, eluted proteins were analysed by SDS-PAGE. 6xHis-MOS11 and 6xHis-HMGB2 were detected by immunoblotting using an antibody directed against the 6xHis-tag and ALY2 by phosphorimaging. Aliquots of the protein input of MOS11, HMGB2 (10%) and ALY2 (25%) are also shown.