The primary focus of my Section is on the role of DNA damage in human diseases, including both neurological diseases and cancer. Specifically, the major focus is on two different DNA lesions. The first DNA lesion is 8, 5-cyclo-deoxyadenosine (cyclo-dA). We have presented several lines of evidence indicating that this lesion is responsible, at least in part, for the neurodegeneration in patients with the hereditary DNA repair diseases xeroderma pigmentosum (see Brooks, 2008). Our efforts with both of these lesions has shifted towards a more mechanistic and structural biology approach. This shift was prompted by our discovery in 2007 that the oxidative DNA lesion 8, 5-cyclo-dA could produce different types of mutant RNA transcripts in vivo in human cells (Marietta et al, 2007). This observation led us to begin focusing on the mechanistic basis of these observations, as well as more generally how DNA lesions impact the transcription process. Towards this goal, we have taken a reductionist approach to the question by using a minimal transcription system composed of an RNA primer, a DNA template, and purified polymerase. Using the minimal system, we found that all multisubunit polymerases incorporated UMP opposite the lesion, consistent with our results in vivo. In the process of these studies, however, we found that the sequence content of the template DNA scaffold has a dramatic effect on transcrption past the lesion, an unexpected result that has important implication for the basic mechanisms of transcription itself. We have also used this comparative minimal system to examine the effects of N2-ethyl-dG, an acetaldehyde-derived DNA lesion, on transcription. In our paper published in JBC , we showed that this lesion blocks transcription by both multisubunit and single subunit polymerases, but at different steps. As with cyclo-dA, eukaryotic RNA Pol II inserted the correct nucleotide, CTP, opposite the N2-ethyl-dG, prior to stalling. However, kinetic studies showed that the addition ethyl group showed incorporation by a factor of approx 1500 fold. Using molecular modeling we were able to identify two mutually exclusive configurations for the ethyl group within the active site of the polymerase. In contrast to the effect of N2-ethyl-dG on transcription by multisubunit RNA polymerases, we found that T7 RNA polymerase was unable to incorporating any rNTP opposite the lesion (Cheng et al, 2008). Thus, the lesion blocks the two different types of polymerase at different mechanistic steps. Subsequent molecular modeling studies have indicated that the blocking effect of the lesion on T7 RNAP is due in large part to a steric clash between the ethyl group of the adduct and a specific His residue located within the active site of the enzyme. Using site-directed mutagenesis, TF Cheng in my lab has has obtained evidence in support of this prediction. We proposed that the cause of the unique forms of neuropathology in this disease may not be due to DNA repair defects, as generally believed, but to defects in the transcription of certain cell-type specific genes (Brooks et al, 2008). We are now pursuing this idea by analyzing postmortem brain tissue samples from CS patients. We had earlier proposed a key role for the Fanconi anemia BRCA1 (FA-BRCA) pathway in protecting against the genotoxic effects of acetaldehyde, the first metabolite of ethanol (Brooks and Theruvathu, 2005). In work completed this year, we have reported that acetaldehdye does in fact activate the FA-BRCA netowrk (Marietta et al, 2009), as predicted by our hypothesis.
