RNA chaperone activity of protein components of human Ro RNPs

Abstract

Ro ribonucleoprotein (RNP) complexes are composed of one molecule of a small noncoding cytoplasmic RNA, termed Y RNA, and the two proteins Ro60 and La. Additional proteins such as hnRNP I, hnRNP K, or nucleolin have recently been shown to be associated with subpopulations of Y RNAs. Ro RNPs appear to be localized in the cytoplasm of all higher eukaryotic cells but their functions have remained elusive. To shed light on possible functions of Ro RNPs, we tested protein components of these complexes for RNA chaperone properties employing two in vitro chaperone assays and additionally an in vivo chaperone assay. In these assays the splicing activity of a group I intron is measured. La showed pronounced RNA chaperone activity in the cis-splicing assay in vitro and also in vivo, whereas no activity was seen in the trans-splicing assay in vitro. Both hnRNP I and hnRNP K exhibited strong chaperone activity in the two in vitro assays, however, proved to be cytotoxic in the in vivo assay. No chaperone activity was observed for Ro60 in vitro and a moderate activity was detected in vivo. In vitro chaperone activities of La and hnRNP I were completely inhibited upon binding of Y RNA. Taken together, these data suggest that the Ro RNP components La, hnRNP K, and hnRNP I possess RNA chaperone activity, while Ro60-Y RNA complexes might function as transporters, bringing other Y RNA binding proteins to their specific targets.

FIGURE 1.

La, hnRNP K, and hnRNP I increase splicing in a cis-splicing assay. In this assay a modified construct of the pre-mRNA of the td gene of the T4 phage containing a group I intron was used and the decrease of the amount of pre-mRNA with time was measured. In the absence of protein, ~30% of the RNA molecules spliced efficiently (black line). The presence of the proteins La (blue line), hnRNP I (green line), and hnRNP K (red line) significantly increased the fraction of fast-reacting RNA molecules, whereas Ro60 (purple line) had no effect on the splicing reaction.

FIGURE 2.

hnRNP K and hnRNP I increase splicing in a trans-splicing assay. (A) Scheme for the trans-splicing assay. The two separately transcribed constructs H1 (encoding exon1 + intron1) and H2 (encoding intron2 and exon2) have to fold into a splicing competent structure. The splicing reaction is started upon addition of ³²P-labeled GTP. (B) Representative autoradiograph of a 5% polyacrylamide gel. The 5′ part of the intron with the added ³²P-guanosine (G-I1) and the ligated exons (E1E2) are indicated, as well as the substrates H1 and H2. At 55°C splicing occurred efficiently (first set of experiments), whereas at 37°C splicing was largely reduced (second set). Trans-splicing was measured at 37°C in the presence of 2 μM La, hnRNP K, hnRNP I, and Ro60. (Prot.b.) Addition of an equivalent volume of protein storage buffer to the splicing reaction (negative control). A slightly different storage buffer was used for Ro60 (“prot.b. (Ro60)”). Note the appearance of the splice product G-I1 in the presence of hnRNP K and hnRNP I and its complete absence in the presence of La and Ro60. (C) Calculated relative splicing rates of the positive control (55°C), the negative control (37°C), and the trans-splicing reaction at 37°C in the presence of hnRNP I or hnRNP K. Relative splicing rates were calculated from (nx − n37)/(n55 − n37), “n55” and “n37” being splicing rates at 55°C or 37°C and “nx” the splicing rate of each evaluated time course.

FIGURE 3.

Increased trans-splicing rates upon addition of hnRNP I are observed at low temperatures but not at 55°C. The influence of hnRNP I on trans-splicing was investigated at 55°C and at 25°C. The labeling of the bands is identical to Figure 2 and increase of the splice-product G-I1 was used as read-out for measuring the splicing reaction. At 55°C the presence of hnRNP I did not further stimulate trans-splicing (compare the first set of reactions with the second set). The third set of trans-splicing reactions at 55°C shows splicing in the presence of protein storage buffer (prot. b.), which slightly inhibited the splicing reaction (i.e., formation of G-I1) and almost completely abolished exon ligation. At 25°C splicing was hardly observed but strongly enhanced in the presence of hnRNP I (compare fourth and fifth sets of splicing reactions).

FIGURE 4.

Effect of Y RNA binding to individual components of Ro RNPs on their chaperone activity. (A) Comparison of the cis-splicing activity of free La and Ro60 with La-hY1 RNA and Ro60-hY1 RNA complexes. Two micromolar protein or pre-bound Y RNA–protein complex were tested. As shown previously free La protein increased cis-splicing (filled blue squares); however, when bound to hY1 RNA the RNA chaperone activity was lost (open blue squares). When Y RNA and La were added to the cis-splicing assay without pre-incubation, La could still stimulate the reaction, however to a lesser extent (filled blue diamonds). Ro60 was inactive and also the association of hY1 RNA did not stimulate the reaction (open red triangle); as expected, hY1 RNA by itself did not show any effect on cis-splicing (open circle). (B) Comparison of trans-splicing activity of free hnRNP I and La protein to pre-bound hnRNP I-hY1 RNA and La-hY1 RNA complexes. The bar graph shows relative splicing rates calculated as described in Figure 2 by the formula (nx − n37)/(n55 − n37), where “nx” is the splicing rate of the respective protein or protein–Y RNA complex, “n55” the splicing rate at 55°C in the absence of protein (left bar), and “n37” the splicing rate at 37°C in the absence of protein (second bar from the left). Only free hnRNP I stimulated trans-splicing, whereas the hY1 RNA–hnRNP I complex was completely inactive. A moderate decrease in splicing was seen when hnRNP I and hY1 RNA were added separately (second bar from the right). No activity was observed for La or La-hY1 RNA complexes.

FIGURE 5.

Overexpression of La rescues splicing of a splicing deficient mutant in vivo. (A) Reverse transcription reaction analyzing splicing of tdwt and tdSH1 RNA. The DNA primer hybridizes in exon2 and by the addition of dATP, dGTP, dCTP, and ddTTP in the presence of reverse transcriptase either a primer + 5-nt transcript is produced (for pre-mRNA), a primer + 16-nt (for mRNA) or a primer + 8-nt (for cryptic-mRNA). (SJ) Splice junction, (3′ ss) 3′ splice site. (B) Representative autoradiograph of a polyacrylamide gel. Reverse transcription products of mRNA, pre-mRNA, and a cryptic-mRNA are indicated. Cryptic splicing occurs when an alternative fold of the 5′ splice site is formed. The double mutant tdSH1 is significantly reduced in splicing (lane 2). In the presence of La or E. coli protein StpA (lanes 3,5) splicing was significantly increased. In the presence of Ro60 splicing was moderately increased (lane 4). The last lane shows splicing of the wild-type td mRNA. (C) Percentage (mean value ±standard deviation) of splicing [mRNA + cryptic-mRNA] ×100/[mRNA + cryptic-mRNA + pre-mRNA] of three individual experiments.