Characterization of transcription from TATA-less promoters: identification of a new core promoter element XCPE2 and analysis of factor requirements

Abstract

Background: More than 80% of mammalian protein-coding genes are driven by TATA-less promoters which often show multiple transcriptional start sites (TSSs). However, little is known about the core promoter DNA sequences or mechanisms of transcriptional initiation for this class of promoters.

Methodology/principal findings: Here we identify a new core promoter element XCPE2 (X core promoter element 2) (consensus sequence: A/C/G-C-C/T-C-G/A-T-T-G/A-C-C/A(+1)-C/T) that can direct specific transcription from the second TSS of hepatitis B virus X gene mRNA. XCPE2 sequences can also be found in human promoter regions and typically appear to drive one of the start sites within multiple TSS-containing TATA-less promoters. To gain insight into mechanisms of transcriptional initiation from this class of promoters, we examined requirements of several general transcription factors by in vitro transcription experiments using immunodepleted nuclear extracts and purified factors. Our results show that XCPE2-driven transcription uses at least TFIIB, either TFIID or free TBP, RNA polymerase II (RNA pol II) and the MED26-containing mediator complex but not Gcn5. Therefore, XCPE2-driven transcription can be carried out by a mechanism which differs from previously described TAF-dependent mechanisms for initiator (Inr)- or downstream promoter element (DPE)-containing promoters, the TBP- and SAGA (Spt-Ada-Gcn5-acetyltransferase)-dependent mechanism for yeast TATA-containing promoters, or the TFTC (TBP-free-TAF-containing complex)-dependent mechanism for certain Inr-containing TATA-less promoters. EMSA assays using XCPE2 promoter and purified factors further suggest that XCPE2 promoter recognition requires a set of factors different from those for TATA box, Inr, or DPE promoter recognition.

Conclusions/significance: We identified a new core promoter element XCPE2 that are found in multiple TSS-containing TATA-less promoters. Mechanisms of promoter recognition and transcriptional initiation for XCPE2-driven promoters appear different from previously shown mechanisms for classical promoters that show single "focused" TSSs. Our studies provide insight into novel mechanisms of RNA Pol II transcription from multiple TSS-containing TATA-less promoters.

Figure 1. Determination of the core promoter element driving transcription from HBV X mRNA Start site 2.

(A) Schematic of the HBV enhancer 1-X promoter region and the deletion mutants used for mapping of the minimal promoter for transcription from Start site 2. Different HBV enhancer 1-X promoter fragments (nucleotide numbers of the boundaries shown) were cloned into a CAT reporter plasmid and used as the template for in vitro transcription. Summary of the in vitro transcription analyses are schematically shown. The bent arrows show that accurate transcription from Start site 2 , was detected with the indicated constructs. Being the same thicknesses of the bent arrows indicate about the levels of transcription detected by in vitro transcription assays. Absence of the bent arrows on the promoter region indicates no detectable transcription from the particular DNA fragments in the right direction. On the right ends, relative transcription activity of each construct in vivo (in transfected HepG2 cells) measured by CAT assay is shown as “SS2 txn activity in vivo”. The template pXStNc·ΔSS1CAT has mutation at the Start site 1 core promoter so that only transcription from Start site 2 can be measured. The DNA sequence around the Start site 2 minimal promoter region is shown at the bottom. The 13-bp minimal promoter region is underlined and the position of Start site 2 is shown by a bent arrow. (B) In vitro transcription assays of the wild-type and point-mutated minimal promoters for the Start site 2. The nucleotides within the 13-bp minimal promoter region were individually mutated into three other nucleotides, and the transcription activity of the mutants was assayed in vitro. Primer extension products corresponding to Start site 2 are shown. The levels of transcription from mutant templates were quantified by phosphor-imager analysis. (C) Summary of site-directed mutagenesis. The in vitro transcription assays were repeated at least three times with several independent preparations of template DNAs, and average activity of each mutant relative to that of the wild-type minimal promoter was calculated. Promoter activities of mutants are categorized into three groups based on their relative activity to the wild-type promoter: pink, ≥50%; yellow, 50–25%; gray, <25%. The consensus sequence deduced from our analysis is shown at the bottom of the figure.

Figure 2. XCPE2 drives transcription from human promoters.

(A) In vitro transcription analyses of wild-type and XCPE2-mutated promoter templates. Sequence ladders were made using the same sets of templates and primers as those for the primer extension analyses. Arrows and red asterisks (*) show the TSSs at the position expected to be driven by XCPE2. Green asterisks show the nucleotide positions of other start sites detected in our in vitro transcription assays. Black asterisks show the nucleotide positions of start sites recorded in the DBTSS database. (B) Primer extension analysis of Ankyrin repeat and SOCS box-containing protein mRNA produced in transfectecd HepG2 cells, showing that the same TSSs driven by XCPE2 as in vitro transcription assays were detected. Arrows and red asterisks show the TSSs driven by XCPE2.

Figure 3. TFIID present in nuclear extracts contributes in X mRNA transcription, but TAF11 is dispensable.

(A) In vitro transcription assays of immmunodepleted nuclear extracts (NEs). To examine importance of TFIID for X gene transcription, TFIID was depleted from NEs using antibodies raised against different TFIID subunits and the depleted NEs were tested for X gene transcription activity using the HBV enhancer-X promoter construct. SS1 and SS2: primer extension products showing transcription from Start site 1 and Start site 2, respectively. The asterisk shows a primer extension product that was not consistently observed. (B) Western blot analyses of the immunodepleted NEs. Depleted NEs were examined for the levels of TAF1, TAF4, TAF6, TAF11, and TBP.

Figure 4. Transcription from X mRNA Start site 2 can use either a free TBP or the TFIID complex.

(A) Examination of TBP concentration in HeLa NE. Indicated amounts of HeLa NE and purified recombinant TBP were loaded on a SDS-PAGE gel, and TBP was detected by anti-TBP western blotting. (B) In vitro transcription assays of the X gene (from Start site 2) and the Sp1-TATA templates (in a single two-template reaction). Control (lane 1) or TBP-depleted (lanes 2–8) NEs were tested for X gene or Sp1-TATA transcription in the absence (lane 2) or presence of 1 µl (lane 3) or 3 µl (lane 4) of purified TFIID or the presence of 1 ng (lane 5), 3 ng (lane 6), 10 ng (lane 7), or 30 ng (lane 8) of purified recombinant TBP (1 µl of the TFIID used in this experiment contained about 1 ng of TBP [6]). The enhancer-X promoter template was used to measure transcription from Start site 2. (C) Quantification results of the transcription assay. Transcription assays were performed twice for the reaction with 30 ng TBP (lane 8 of panel A) and four times for all of the other reactions. Brackets show standard errors of means.

Figure 5. A free form of TBP can drive X gene transcription in the absence of TAF4.

HeLa NE was depleted with nothing (lane 1) or with anti-TAF4 (lanes 2–5) and was mixed with the indicated amounts of purified recombinant TBP (lanes 3–5), then tested for X gene transcription activity.

Figure 6. TAF1 is marginally important for X gene transcription in ts13 cells.

ts13 cells were transfected with a firefly luciferase reporter plasmid driven by the X gene core promoter 2 or the cyclin A promoter. After 16 hr of incubation at the permissive (33.5°C) or nonpermissive (39.5°C) temperature, the luciferase activity in transfected cells were measured and normalized for transfection efficiency. Brackets show standard error of means.

Figure 7. Transcription from X mRNA Start site 2 requires MED26-containing mediator and TFIIB but not Gcn5.

(A) HeLa cell nuclear extract (NE) was immunodepleted with control or anti-Gcn5 antibody. The depleted NEs were then tested for the level of depletion by western blotting and for X gene transcription activity from the two start sites using the enhancer-X promoter construct. (B) HeLa NE was immunodepleted with control or anti-MED26 antibody. The depleted NEs were then tested for the level of depletion and for X gene Start site 2 transcription activity using either the enhancer-X promoter construct or the minimal promoter construct in the absence (lane 2) or presence (lanes 3 and 4) of mediator complexes purified from P.5 (lane 3) or P1.0 (lane 4) phosphocellulose fractions. The response of the transcription from Start site 1 has been reported . (C) HeLa NE was immunodepleted with control or anti-TFIIB antibody. The depleted NEs were then tested for depletion and for X gene transcription activity from the two start sites using the enhancer-X promoter construct in the absence (lane 2) or presence (lanes 3–5) of the purified TFIIB (10, 30, or 100 ng).

Figure 8. X gene transcription with free TBP requires mediator, TFIIB, and RNA pol II but not TAFs or Gcn5.

HeLa NE was immunodepleted of the indicated factors and tested for X gene transcription activity with or without addition of free TBP. Abolishment of X gene transcription by TAF4-, TAF6-, TBP-, MED26-, TFIIB-, or RNA pol II-depletion (lane 1 vs. lanes 3, 5, 7, 9, 11, and 13) confirms that these factors contribute to the X gene transcription but Gcn5 doesn't (lane 1 vs. lane 15). The strong activation of X gene transcription by addition of a higher-than-endogenous level of free TBP to the mock-, TAF4-, TAF6-, TBP-, and Gcn5-depleted NEs (lanes 2, 4, 6, 8, and 16) indicates that these depleted factors were not necessary for X gene transcription by the mechanism using free TBP. In contrast, the absence of such activation by addition of free TBP to MED26-, TFIIB-, or RNA pol II-depleted NE (lanes 10, 12, and 14) indicates that these three factors are required for the free TBP-driven transcription to occur. For more details, see text.

Figure 9. XCPE1 and XCPE2-containing cellular promoters show the same GTF requirements as HBV X gene promoter.

(A) Western blot analyses of depleted NEs. HeLa NE was immunodepleted as indicated and the levels of target factors as well as other GTFs were examined by western blotting. (B) In vitro transcription assays of the immunodepleted NEs. Transcription activities for the XCPE2-containing promoters (ENST268533, NM_16114, and NM_32267), an XCPE1-containing promoter (NM_024811), and the X gene Start site 2 minimal promoter were examined. Arrows show the transcripts starting at the positions expected to be driven by XCPE2 or XCPE1.

Figure 10. Sequence-nonspecific stable complex formation by TBP, TFIIB, RNA pol II, TFIIF, and mediator on X gene Start site 2 promoter.

(A) SDS-PAGE analyses of purified factors. Purified recombinant TFIIB, TFIIF, and TBP were analyzed by Comassie staining. RNA pol II and mediator were analyzed by silver staining. Bands are labeled by comparing the previously published patterns. The mediator complexes shown are bound to M2 beads, from which they were eluted with FLAG peptides and used for EMSA. (B) EMSA with the X gene Start site 2 minimal promoter probe. The ³²P-labeled probe was mixed with indicated factors without carrier DNA. B, TFIIB; D, TBP; P, RNA pol II; F, TFIIF; and M, mediator. The mixtures were incubated for 30 min before electrophoresis as described in Materials and Methods. The experiments shown were performed using the mediator complex from the phosphocellulose fraction P1.0, but the complexes from the P.5 fraction showed the same results. (C) XCPE2 DNA and TFIIB/TBP/Pol II/TFIIF form an unstable complex. The same EMSA as is shown in Fig. 10B was performed except that the binding mixtures were incubated for 1 hr 20 min before electrophoresis. (D) The XCPE2 DNA/TFIIB/TBP/Pol II/TFIIF/mediator complex was not observed in the presence of poly(dG-dG)· poly(dG-dC). (E) EMSA showing TFIIB/TBP/Pol II/TFIIF complex formation on the adenovirus major late (AdML) promoter. The binding mixtures were incubated for 30 min before electrophoresis. The same results were obtained when the mixtures were incubated for a longer period (2 hrs, data not shown).