Roles of human parainfluenza virus type 3 bases 13 to 78 in replication and transcription: identification of an additional replication promoter element and evidence for internal transcription initiation

Abstract

The genomic promoter of human parainfluenza virus type 3 (HPIV3) contains multiple cis-elements controlling transcription and replication. Previous work showed that regions 1 to 12 and 79 to 96 were critical in promoting replication of an HPIV3 minireplicon, while the intergenic sequence and N gene start signal (IS/Ngs, bases 49 to 61) were important for transcription. Because these data were collected primarily using point mutations, not every base from position 1 to 96 was analyzed, and some important control elements may have been missed. To clarify the role of bases 13 to 78 in transcription and replication, a series of mutations were made which collectively scanned this entire region. Mutation of bases 13 to 28 resulted in markedly decreased HPIV3 minireplicon replication, indicating these bases constitute an additional cis-element involved in the synthesis of the HPIV3 antigenomic RNA. The position dependence of the IS/Ngs was also examined. Analysis of mutants in which the IS/Ngs was shifted 5' or 3' showed that this segment could be moved without significantly disrupting transcription initiation. Additional mutants which contained two successive IS/Ngs segments were created to test whether the polymerase accessed the gene start signal by proceeding along the template 3' to 5' or by binding internally at the gene start signal. Based on analysis of the double gene start mutants, we propose a model of internal transcription initiation in which the polymerase enters the template at approximately the location of the natural N gene start but then scans the template bidirectionally to find a gene start signal and initiate transcription.

FIG. 1.

Scanning mutagenesis of the 13-78 region of the HPIV3 genomic promoter. (A) Sequences of the 13-78 scanning mutations and replication and transcription relative to WT. Bases changed for each of the mutants are underlined, and the N gene start signal is boxed. (B) Representative primer extension experiments. For analysis of RNA synthesis, total RNA (for transcription products) or S7-treated RNA (for replication products) was isolated from cells transfected with the indicated minigenome and N-, P-, and L-encoding support plasmids. The RNA was analyzed by primer extension using a negative-sense primer to detect the transcribed luciferase mRNA or replicated antigenomic RNA. Averages were based on three or more experiments with each mutant. The lane marked -L indicates the L plasmid was omitted in a transfection with the WT minigenome.

FIG. 2.

Defining important bases within the 13-28 region of the genomic promoter. S7-treated RNA extracts from cells transfected with the indicated minigenomes were analyzed by primer extension using a negative-sense primer to detect antigenomic RNA. Bases changed for each of the mutants are underlined. Averages were based on three or more experiments with each mutant. Lanes marked -L indicate the L plasmid was omitted in transfections with the WT minigenome. (A) Sequence and primer extension analysis of minigenome mutants scanning the 13-28 region. (B) Sequence and primer extension analysis of additional 13-28 minigenome mutants.

FIG. 3.

Positional and functional analysis of the 13-28 element. (A) Position dependence of the 13-28 element. The 13-28 element (underlined) was shifted 3 to 12 bases away from the 3′ end of the genomic RNA in a series of minigenome mutants. S7-treated RNA extracts from cells transfected with the mutants were analyzed by primer extension analysis using a negative-sense primer to detect antigenomic RNA. Averages were based on two experiments with each mutant. (B) Primer extension analysis of the 13-28 mutant using total RNA from transfected cells. Total RNA extracts from cells transfected with the WT minigenome or the 13-28 mutant were analyzed by primer extension analysis using a negative-sense primer to detect RNA with a 5′ terminus identical to that of the antigenomic RNA. The average was based on three experiments. Lanes marked -L indicate that the L plasmid was omitted in the transfection with the WT minigenome.

FIG. 4.

Effects of changing the position of the intergenic sequence/N gene start. A series of minigenome mutants was made in which the intergenic sequence/N gene start was moved toward (-Ngs mutants) or away (+Ngs mutants) from the 3′ end of the genome. For analysis of RNA synthesis, total RNA (for transcription products) or S7-treated RNA (for replication products) was isolated from cells transfected with the indicated minigenome and N-, P-, and L-encoding support plasmids. The total RNA and S7-treated RNA extracts were analyzed by primer extension using a negative-sense primer to detect the transcribed luciferase mRNA or replicated antigenomic RNA, respectively. S7-treated RNA was also analyzed by primer extension using a positive-sense primer to detect replicated genomic RNA. (A) Summary of replication and transcription of mutants. The antigenome synthesis is shown as a percentage relative to WT. The percent WT transcription efficiency is the amount of transcript relative to the amount of template (genomic RNA) and was compared to WT. Average RNA amounts were based on at least three determinations of each RNA species. (B) Representative gels showing primer extension experiments detecting the antigenomic RNA (AG), genomic RNA (G), and luciferase mRNA (mRNA). The image showing detection of the antigenome is overexposed to show the low levels of antigenomic RNA synthesis with the −13Ngs mutant. The background seen in the +6Ngs transcription lane was seen occasionally with different mutants, but the nature of this background is unknown. Data from such lanes were not used for quantitation purposes. The lane marked -L indicates the L plasmid was omitted in the transfection with the WT minigenome.

FIG. 5.

Transcription start site usage with the double gene start constructs. (A) Sequence of the double gene start mutants and predicted and actual gene start 1/gene start 2 (GS1/GS2) utilization ratios. The gene start sites are underlined. The calculation of the expected GS1/GS2 ratio with polymerase scanning from the 3′ end was done as follows: TEGS1/[(1 − TEGS1)(TEGS2)], where TE refers to the percent WT transcription efficiency of the corresponding + and − Ngs mutants shown in Fig. 4. The denominator in this equation reflects the assumption that transcription initiating at GS1 would not initiate at GS2. The expected GS1/GS2 ratio with internal transcription initiation was TEGS1/TEGS2. The resulting ratios were then normalized so that GS1 was equal to 1. The actual GS1/GS2 ratio was determined by phosphorimager analysis of three or more experiments. (B) Representative gels of primer extension experiments detecting transcription products. The lane marked -L indicates the L plasmid was omitted in the transfection with the WT minigenome.