The scaRNA2 is produced by an independent transcription unit and its processing is directed by the encoding region

Abstract

The C/D box scaRNA2 is predicted to guide specific 2'-O-methylation of U2 snRNA. In contrast to other SCARNA genes, SCARNA2 appears to be independently transcribed. By transient expression of SCARNA2-reporter gene constructs, we have demonstrated that this gene is transcribed by RNA polymerase II and that the promoter elements responsible for its transcription are contained within a 161 bp region upstream of the transcription start site. In mammals, we have identified four cross species conserved promoter elements, a TATA motif, an hStaf/ZNF143 binding site and two novel elements that are required for full promoter activity. Binding of the human hStaf/ZNF143 transcription factor to its target sequence is required for promoter activity, suggesting that hStaf/ZNF143 is a fundamental regulator of the SCARNA2 gene. We also showed that RNA polymerase II continues transcription past the 3'-end of the mature RNA, irrespective of the identity of the Pol II promoter. The 3'-end processing and accumulation are governed by the sole information contained in the scaRNA2 encoding region, the maturation occurring via a particular pathway incompatible with that of mRNA or snRNA production.

Figure 1.

Identification and functional analysis of the elements in the human SCARNA2 promoter. (A) Detection of promoter activity in the 5′-flanking sequence of the human SCARNA2 gene. Schematic representation of the SCARNA2 constructs fused to a luciferase (Luc) reporter and their activities in COS-7 cells. For each construct, the length of the sequence upstream and downstream of the +1 is shown to the left and the luciferase activity to the right. Cells were transiently transfected with the constructs and assayed for luciferase activity. The firefly luciferase activity was normalized to that of the renilla luciferase provided by co-transfection of the control vector. Data for each construct were plotted as fold-activation which represents the ratio between the normalized luciferase activity of each reporter construct to that of the empty luciferase vector (pFLASH1). Data are presented as the mean ± SD of three separate experiments. (B) Sequences of the wild-type and mutant X, Y, SBS and TATA elements of the SCARNA2 promoter. Mutations are highlighted in gray. (C) Contribution of the X, Y, SBS and TATA elements to SCARNA2 promoter activity. Left panel: schematic representation of the SCARNA2 promoter-luciferase (Luc) constructs substituted individually or simultaneously in the X, Y, SBS and TATA elements. Open and solid boxes represent wild-type and mutant elements, respectively. Right panel: COS-7 cells were transfected with the reporter constructs and the luciferase activity was normalized to that of the renilla lucifease obtained from the co-transfected control vector. The normalized luciferase activity of the wild-type construct −250/+145 was arbitrarily set to 100%. Data are presented as the mean ± SD of three separate experiments.

Figure 2.

Transcription assays of wild-type and mutated human SCARNA2 gene promoters and polymerase transcribing status of the SCARNA2 gene. (A) Structures of the templates used in the in vivo analyses. The +1 indicates the start of transcription. Open boxes, wild-type elements; solid boxes, mutant elements. The solid box in the transcribed region represents the 6 bp insertion creating the maxigene. Probes 1 and 2 and their hybridization regions in the nontranscribed strand are indicated (wt gene numbering). (B) In vivo expression of wild-type and mutated gene promoters. HeLa cells were transfected with the constructs indicated above the lanes together with plamid pα1 as the internal control. RNA recovered 48 h after transfection were analyzed by RNase protection assay with probe 1 and probe anti pα1 complementary to the α-globin standard. scaRNA2 and α are the protected RNAs derived from the scaRNA2 and pα1 globin internal standard. Results from lanes 1 to 6 and 7 to 9 arose from separate experiments. Normalized transcription activity values: wt, 100%; Xsub, 32 ± 7%; Ysub, 90 ± 9%; SBSsub, 48 ± 10%; TATAsub, 95 ± 7%. (C) Identification of altered TSS of the wt and mutants SCARNA2 promoters. RPA with probe 2 was used to determine whether transcription of the TATAsub mutant gene initiated at the same position. HeLa cells were transfected with the constructs indicated above the lanes; a, b, c and d indicate protected fragments. (D) Pol II transcription of the SCARNA2 gene. HeLa cells were transiently transfected with the constructs indicated above the lanes and treated simultaneously with low α-amanitin concentration (+α-amanitin, lanes 3, 4 and 7) or untreated (−α-amanitin, lanes 2 and 6). Lanes 1–4, RPA performed as in (B) with probe 1 and anti pα1. The lengths of protected fragments are indicated (nt). Lanes 5–7, primer extension assay performed with RNA isolated from transfected or untransfected cells and a ³²P-labeled oligonucleotide complementary to a region located downstream of the 16 bp insertion creating the maxi U6 gene. Data are representative views of experiments performed in triplicate. Results from lanes 1 to 4 and 5 to 7 arose from separate experiments.

Figure 3.

Identification of an occupied hStaf/ZNF143 transcription factor binding site in the human SCARNA2 promoter. (A) Gel retardation assays were performed with ³²P-labeled promoter fragments encompassing positions −88/+145 of the SCARNA2 promoter and containing the wild-type SBS sequence (wt probe, lanes 1–6), mutant SBS (SBSsub probe, lanes 7 and 8). In the SBSsub probe, the −70 CTCCCA−65 sequence was replaced by GATATC. Probes were incubated in the absence (lanes 1 and 7) or presence of hStaf/ZNF143 (2.5 µl of programmed reticulocyte lysate in lanes 2, 4, 5 and 8; 5 µl in lane 3). Reactions in lanes 4 and 5 were performed in the presence of a 1000-fold molar excess of unlabeled specific competitor (wt SBS) and unspecific competitor (unspe). Reaction in lane 6 was performed in the presence of 5 µl of unprogrammed reticulocyte lysate. Lanes 1–6 and 7–8 are from separate experiments. (B–D) Immunoprecipitation with hStaf/ZNF143 antibody of the SCARNA2 promoter from formaldehyde cross-linked chromatin prepared from HeLa cells. DNA fragments recovered from input chromatin, or ChIP assays with anti-hStaf/ZNF143 or control experiment without antibody, were analyzed by semi-quantitative PCR with specific primer pairs in the presence of (α−³²P) dCTP. Lanes 1, 2 and 3, 4: serial dilutions of DNA immunoprecipitated with anti-hStaf/ZNF143 or no antibody, respectively. Lanes 5–7: serial dilutions of input material to demonstrate that the assays were within the linear range of PCR amplification. Lane 8 (C, PCR control): PCR lacking DNA template. (B) PCR was performed with specific primer pairs for the human SCARNA2 promoter or (C) for the tRNA^Sec promoter acting as the positive control. (D) Negative control with the PP1 primer pair targeting a region located upstream of the tRNA^Sec gene and lacking SBS.

Figure 4.

The 3′ flanking genomic region of the SCARNA2 gene is not required for the processing step and the mature scaRNA2 can be produced from various Pol II promoters. (A) Structures of the various templates used in the analyses. Probes 1, 3–6 and their hybridization regions in the nontranscribed strand are indicated (wt gene numbering). (B) Various Pol II promoters produce correctly 3′-end processed scaRNA2. HeLa cells were transfected with the constructs indicated above the lanes. After quantitation, the signal/mock ratios in lanes 3–9 relative to the mature RNA are: 3.25 ± 0.25, 5 ± 0.3, 1 ± 0.1, 6.7 ± 0.35, 13 ± 0.5, 9.25 ± 0.4, 11 ± 0.25. Results from lane 2 arose from a longer exposure. (C) Deletion of the 3′ flanking genomic region of the SCARNA2 gene does not impede production of the mature scaRNA2. Signal/mock ratio relative to the mature RNA in lane 2: 2.5 ± 0.3 (D) The mature scaRNA2 can be obtained with a construct harboring the 3′-flanking region in an inverted orientation. Signal/mock ratio relative to the mature RNA in lane 2: 3.5 ± 0.5. (E) The U3 snoRNA and CMV promoters are compatible with scaRNA2 production. RNA extracted from cells transfected with the constructs indicated above the lanes were analyzed by RPA with the indicated probe. pα1: vector used as the internal transfection control; anti pα1: probe monitoring the RNA produced from the transcribed pα1 gene. scaRNA2 and α: protected RNAs derived from the scaRNA2 and pα1 globin internal standard. Normalized transcriptional activity values: wt, 100%; pCMV-maxiSCARNA2, 300 ± 25%; pU3-maxiSCARNA2, 145 ± 15%. The lengths of protected fragments are indicated (nt). Data presented in (B–D) are representative views of experiments performed in triplicate.

Figure 5.

Introduction of 3′-end processing elements inhibits production of mature scaRNA2. (A) Schematic representation of the structures of the various constructs used in the assays. (B and C) In vivo expression of the various constructs depicted in (A). HeLa cells were transfected with the constructs indicated above the lanes. The recovered RNAs were analyzed by RNase protection assay with the indicated probes. pα1: vector used as the internal transfection control; anti pα1: probe monitoring the RNA produced from the transcribed pα1 gene. scaRNA2 and α: protected RNAs as in Figure 2. The lengths of protected fragments are indicated (nt). Results from lanes 1–3 and lane 4 in (C) arose from separated experiments. Data presented are representative views of experiments performed in triplicate. Normalized transcription activity values for lanes 1–9 in (B): lane 1, 100%; lane 2, 61 ± 25%; lane 3, 34 ± 15%; lane 4, 100 %; lane 5, 75 ± 17%; lane 6, 66 ± 9%; lane 7, 100%; lane 8, 90 ± 9%; lane 9, 85 ± 10%.