Nucleotide Excision Repair and Transcription-coupled DNA Repair Abrogate the Impact of DNA Damage on Transcription

Abstract

DNA adducts derived from carcinogenic polycyclic aromatic hydrocarbons like benzo[a]pyrene (B[a]P) and benzo[c]phenanthrene (B[c]Ph) impede replication and transcription, resulting in aberrant cell division and gene expression. Global nucleotide excision repair (NER) and transcription-coupled DNA repair (TCR) are among the DNA repair pathways that evolved to maintain genome integrity by removing DNA damage. The interplay between global NER and TCR in repairing the polycyclic aromatic hydrocarbon-derived DNA adducts (+)-trans-anti-B[a]P-N(6)-dA, which is subject to NER and blocks transcription in vitro, and (+)-trans-anti-B[c]Ph-N(6)-dA, which is a poor substrate for NER but also blocks transcription in vitro, was tested. The results show that both adducts inhibit transcription in human cells that lack both NER and TCR. The (+)-trans-anti-B[a]P-N(6)-dA lesion exhibited no detectable effect on transcription in cells proficient in NER but lacking TCR, indicating that NER can remove the lesion in the absence of TCR, which is consistent with in vitro data. In primary human cells lacking NER, (+)-trans-anti-B[a]P-N(6)-dA exhibited a deleterious effect on transcription that was less severe than in cells lacking both pathways, suggesting that TCR can repair the adduct but not as effectively as global NER. In contrast, (+)-trans-anti-B[c]Ph-N(6)-dA dramatically reduces transcript production in cells proficient in global NER but lacking TCR, indicating that TCR is necessary for the removal of this adduct, which is consistent with in vitro data showing that it is a poor substrate for NER. Hence, both global NER and TCR enhance the recovery of gene expression following DNA damage, and TCR plays an important role in removing DNA damage that is refractory to NER.

Keywords: DNA repair; RNA polymerase II; carcinogenesis; nucleotide excision repair; transcription.

FIGURE 1.

The structures of B[a]P-N⁶-dA and B[c]Ph-N⁶-dA and the site-specific modified DNA for in vitro transcription are illustrated. A, the B[a]P-N⁶-dA adduct disrupts base pairing at sites where it occurs in DNA, significantly distorting the double helix. In contrast, the B[c]Ph-N⁶-dA adduct does not disrupt base pairing at sites where it occurs in DNA. Both adducts are intercalated between base pairs, but the B[a]P-N⁶-dA adduct, which is planar, rigid, and more bulky, causes disruption of Watson-Crick base pairing, whereas the smaller, curved B[c]Ph ring system stacks between base pairs without rupturing the Watson-Crick pairs. B, templates for in vitro transcription contained the CMV immediate early promoter/enhancer, as shown schematically in black. The +1 start site for transcription is indicated, and the arrow designates the direction of transcription elongation. The region formed by annealing the 96-mer, the 90-mer, and the 11-mer is shown. The position of the B[a]P-N⁶-dA adduct within the 11-mer is indicated with a triangle on the transcribed strand. In the unmodified, control DNA template, the position indicated by the triangle contained dA. The size of the template was 1,141 bp following isolation from the paramagnetic beads. The recognition sites for I-PpoI and BsiWI are also shown.

FIGURE 2.

Schematic for the gapped duplex system to assemble the site-specifically modified vectors for transcription analysis in cells. The map of the parent vector for this work, pWLZG-I-Insert-R, is shown in detail and is divided into three functional regions. Region I, which is encoded in shades of green, has the following elements: the constitutive elongation factor 1-α (EF1-α) promoter, which drives transcription of the region following it to generate a polycistronic mRNA that encodes three proteins, each separated by an internal ribosomal entry sequence (IRES). The three encoded proteins include: tetR-derived reverse transactivator protein (rtTA), tetR-derived transrepressor protein (tTS), and GFP (ZsGreen1). IRES1 is derived from the polio virus genome, and IRES2 is derived from the encephalomyocarditis virus genome. Region II, which is encoded in shades of red, has the following elements: DsRed-express, which encodes RFP, and a tetracycline-responsive promoter element (TRE-tight) that drives Ds-Red-express transcription. Region III, which is encoded in shades of gray, contains components needed to propagate the plasmid in bacteria and prepare single-stranded DNA: an E. coli origin of replication (ColE1), an ampicillin resistance gene (Amp-r), and the F1 phage origin of replication (F1ori). Following transfection with the control or damaged vector, expression of the polycistronic mRNA in Region I produces transrepressor protein that binds to the tetracycline-responsive promoter element, repressing transcription of the RFP gene. The reverse transactivator protein and GFP are also expressed. After the addition of doxycycline to the transfected cells, the drug binds to transrepressor protein, which releases it from the tetracycline-responsive promoter element, thus removing the repressor; in concert, doxycycline binds to the reverse transactivator protein, creating a potent activator that binds to the tetracycline-responsive promoter element, permitting transcription of the RFP gene. Note that a chicken HS4 insulator element (Insulator), which is encoded in black, is positioned between Regions I and II to separate the two transcription units. The oligomer that ligated into the gapped duplex DNA is illustrated with a red line; the black triangle represents the presence of a site-specific DNA adduct.

FIGURE 3.

DNA templates for in vitro transcription were characterized following synthesis. A, I-PpoI digestion of DNA templates for in vitro transcription. Lane M, New England Biolabs 50-bp DNA ladder labeled with [³²P]phosphate; lane 1, I-PpoI digestion of unmodified control DNA template; lane 2, I-PpoI digestion of B[a]P-N⁶-dA-modified DNA template. B, BsiWI digestion of DNA templates for in vitro transcription. Lane M, IDT 20/100 DNA ladder labeled with [³²P]phosphate; lane 1, BsiWI digestion of unmodified control DNA template; lane 2, BsiWI digestion of B[a]P-N⁶-dA-modified DNA template.

FIGURE 4.

B[a]P-N⁶-dA blocks hRNAPII elongation. For each DNA template studied, four separate reactions were carried out as indicated above each lane in the gel: lane M, New England Biolabs 50-bp DNA ladder labeled with [³²P]phosphate; lanes 1–4, results for transcription using the unmodified DNA template illustrated in Fig. 1; lanes 5–8, transcription results using the DNA template modified with a B[a]P-N⁶-dA adduct, also illustrated in Fig. 1; lanes 9–12, results using the control DNA template containing a CMV immediate early promoter supplied with the HeLa nuclear extract. The components for each transcription reaction are indicated above the lanes.

FIGURE 5.

Site-specifically modified vectors for transcription in cells were characterized. The presence of either B[a]P-N⁶-dA (A) or B[c]Ph-N⁶-dA adducts (B) within a BsiWI restriction sight located in the vector prevents BsiWI digestion. Vectors were incubated with the restriction enzymes shown, and the products were resolved by agarose gel electrophoresis.

FIGURE 6.

B[c]Ph-N⁶-dA interferes with transcription elongation in all cells lacking NER or TCR, whereas B[a]P-N⁶-dA only exhibits an effect when NER is compromised. A, normal primary human fibroblasts; B, XPA^−/− primary human fibroblasts; C, XPC^−/− primary human fibroblasts; D, CSB^−/− primary human fibroblasts. Time courses for each cell type: unmodified control vector (▴); vector modified with B[a]P-N⁶-dA (■); vector modified with B[c]Ph-N⁶-dA (●). RFP mRNA levels were normalized as the percentage of the maximum average RFP mRNA level compared with that of the control vector in a given cell type.

FIGURE 7.

The effect of DNA repair background on recovery of RFP from vectors containing either B[a]P-N⁶-dA or B[c]Ph-N⁶-dA. Each cluster of three bars represents the mean and S.E. (error bars) of the mean of the normalized RFP signal for the indicated cell type, with a value of 100% for the RFP signal for the control. Shown are control vector (black), vector modified with B[a]P-N⁶-dA (gray), and vector modified B[c]Ph-N⁶-dA (white). *, statistically significant difference from the respective control (p < 0.05). Statistical tests were performed by one-way analysis of variance with contrasts.

FIGURE 8.

B[c]Ph-N⁶-dA, which is resistant to NER, is subject to TCR; B[a]P-N⁶-dA is subject to global NER, with the strong possibility that other repair pathways, including TCR, play a role in its removal. NER repairs B[a]P-N⁶-dA efficiently, permitting reactivation of gene expression. In the absence of global NER, B[a]P-N⁶-dA is repaired by other pathways, more than likely including TCR. In contrast, B[c]Ph-N⁶-dA is a very poor substrate for global NER, with TCR mediating its repair to reactivate gene expression.