Our lab studies RNA processing in human cells. We are interested in basic mechanisms that contribute to variation in gene expression in normal physiologic conditions and in response to stress. We use molecular and computational methods to study cellular processes in normal and diseased cells.
Genetics of Human Gene Expression
It is well known that individuals differ at the DNA sequence level; however, the effect of DNA sequence variants on phenotypes remains largely unknown. Since the expression level of genes has important effects on cellular phenotypes, we examined the extent of individual variation in gene expression. We found extensive variability and a heritable component to this variation. This allows us to treat expression levels of genes as quantitative traits and to screen the genome for variants that influence these gene expression phenotypes.
Using genome-wide linkage and association analyses in large families as well as molecular studies, we have identified polymorphic regulators that influence expression levels of a few thousand human genes. These include cis- and trans-acting regulators, as well as regulators that influence the expression of many genes. More than 60 percent of the regulators were not previously known to play a role in gene expression regulation. Gene knockdown, metabolic perturbation assays, and studies of RNA-protein and protein-protein interactions have allowed us to uncover how these regulators influence gene expression. We are continuing these genetic studies and are extending our approach to measure gene expression more precisely and to characterize the regulatory landscape of gene expression in human cells at baseline and following cellular stress.
RNA-DNA Sequence Differences
It is generally believed that RNA sequences are identical to their corresponding DNA sequences. However, we uncovered RNA-DNA sequence differences (RDDs) in human cells, beyond the known adenosine deaminase acting on RNA (ADAR) and apolipoprotein B mRNA editing enzyme, catalytic polypeptide-like (APOBEC) mediated RNA editing. We found all 12 types of RDDs, including transversions. Our results show individual differences in the frequencies and levels of RDDs, thus, demonstrating a layer of genetic variation beyond those in the DNA sequences. We found RDDs lead to peptides that differ from their corresponding DNA. Using mass spectrometry, we identified peptides that correspond to both the RNA and DNA sequences at RDD sites. In addition, preliminary results show that, like splicing, the frequency of some RDDs is affected by cellular perturbations.
In collaboration with John Lis, Cornell University, we studied the sequences of nascent RNAs and found that RDD formation occurs soon after transcription—about 55 nucleotides from the RNA polymerase II active site. RDDs are formed after RNA synthesis and they are not likely a direct consequence of modified base incorporation. However, they are formed before ADAR-mediated RNA editing.
We are now examining other co-transcriptional events, such as R-loop formation, to determine if they co-occur with RDD formation. Results show that enzymes that resolve R-loops, such as topoisomerase I and ribonuclease H, affect RDDs thus suggesting that R-loops and RDDs are coupled. In parallel, in collaboration with Kenneth Fischbeck, NIH, we are studying a rare form of amyotrophic lateral sclerosis (ALS) that results from mutations in senataxin, which encodes a DNA/RNA helicase. Senataxin has multiple functions, among them, it resolves R-loops. RDD patterns in patient cells with aberrant senataxin further suggest the RDD formation is coupled to R-loops.
Genetics of Cellular Response to Stress
Appropriate response to stress is critical for maintaining cellular homeostasis. Our work focuses on two types of stress: endoplasmic reticulum (ER) stress as a result of excessive unfolded protein, and radiation-induced DNA and other cellular damage. Secretory cells such as human B cells must be able to handle large protein loads. Failure to respond to ER stress can lead to diseases such as immunodeficiency and neurodegenerative diseases. Similarly, human cells are increasingly exposed to radiation through the environment and in medical settings. Radiation-based tools have significant medical benefits; however, cellular damage can result from exposure to radiation. We and others have identified genes that change in expression in response to ER stress and radiation exposure. We have shown that these gene expression responses differ across individuals. Treating these changes in gene expression as quantitative traits, we identified polymorphic regulators that influence gene expression response and determined gene interactions by network analyses. To advance these findings into a mechanistic understanding of how stress response affects cell fate, we are deciphering the transcriptional steps that regulate these responses. We are studying how RNA polymerase activity and splicing are regulated to fine-tune cellular responses to stress. The results should advance our basic understanding of transcription and identify risk factors of diseases characterized by inadequate stress responses.
This research was supported by grants from the National Institutes of Health and funds from the Frederick G. L. Huetwell endowment.
As of March 4, 2016