Our lab's early work on catalytic RNAs (ribozymes) helped to establish that RNA is not restricted to being a passive carrier of genetic information, but can have an active role in cellular metabolism. We have now moved on from ribozymes to other large noncoding RNAs, where catalysis is carried out by RNPs (RNA-protein complexes).
Telomerase, an RNP enzyme critical for chromosome end-replication, provides the subject for much of our research. Telomerase is a fascinating biochemical system because DNA (the chromosome end) and RNA and proteins (telomerase) all work in concert. In addition, telomerase is biologically and medically important; telomerase mutations lead to diseases involving degeneration of highly proliferative tissues, and activation of expression of telomerase in adult human somatic cells is a step toward cancer. Finally, telomerase serves as a model for some large noncoding RNAs that have a very different function—the regulation of gene expression through modification of the local chromatin structure (see last section below).
Telomerase was discovered in 1985 by Elizabeth Blackburn and Carol Greider, who also cloned and sequenced its RNA subunit. The catalytic protein subunit, however, remained elusive. As a postdoctoral fellow in our lab, Joachim Lingner purified telomerase from the ciliated protozoanEuplotes, which has an astoundingly large number of telomeres, and found that it contained a reverse transcriptase subunit. TERT (telomerase reverse transcriptase) was quickly found in other organisms, including budding and fission yeast, Tetrahymena, and humans. Our current efforts are focused on the portions of TERT that differ from retroviral and retrotransposon reverse transcriptases and enable telomerase-specific functions, such as repeat-addition processivity (repetitive copying of a small template portion of the RNA).
The TEN (telomerase essential N-terminal) domain of TERT provides an anchor site to hold DNA preceding the chromosome end during telomerase extension, and it contributes to repeat-addition processivity. Steven Jacobs in the lab solved the crystal structure of the TEN domain ofTetrahymena TERT, and this structure has provided a framework for biochemical experiments that have identified its DNA-binding groove and additional regions that contribute to repeat-addition processivity.
The RNA subunit has been investigated in the budding yeast Saccharomyces cerevisiae, a system that facilitates testing hypotheses both in vivo and in vitro. David Zappulla in the lab derived a secondary structure model of the 1.2-kilobase RNA, while contemporaneous work in Raymund Wellinger's lab (Université de Sherbrooke, Canada) arrived at a very similar structure. In the structure, three long RNA arms, each binding a separate protein complex, protrude from a central core. The arms are not conserved in sequence even between different Saccharomyces species, and they can be shortened or even moved to different locations without destroying function. This led to the proposal that the RNA provides a flexible scaffold for bringing accessory proteins into the RNP complex. In the core of the structure, which contains the RNA template and the TERT-binding site, the RNA forms a triple helix that contributes to telomerase catalysis, an unexpected RNA-level function.
Cells must distinguish broken chromosome ends, which elicit a DNA-damage response, from the natural ends called telomeres. Telomeric DNA typically consists of multiple repeats of a short sequence, such as TTAGGG in mammals. These repeats are synthesized by telomerase. The DNA repeats recruit proteins that cap off the chromosome ends, preventing the DNA-damage response, and these proteins also regulate telomerase action. Some telomeric proteins bind double-stranded regions of the telomeric DNA, while others bind the single-stranded DNA "tail" at the very end of the chromosome.
Telomeric single-stranded DNA-binding proteins (TEBPs) were first identified in ciliated protozoa, facilitated by the extremely large number of minichromosomes present in these organisms. Our lab cloned and sequenced the genes for two Oxytricha nova telomere end-binding proteins in 1990 and 1991, but for a decade thereafter it was unclear whether humans even had corresponding proteins. Then in 2001, Peter Baumann in our lab identified candidate genes inSchizosaccharomyces pombe and in humans, which were named POT1 (protection of telomeres 1). The deletion of the gene for Pot1 in S. pombe deprotected the telomeres, leading to rapid loss of all telomeric DNA; rare survivors had circularized all three of their chromosomes, thereby circumventing the need for telomeres. X-ray crystal structures of Pot1 protein-telomeric DNA complexes from S. pombe and human by Ming Lei in our lab revealed the molecular basis of sequence-specific DNA recognition, while work in other labs identified a binding partner of Pot1, now called Tpp1. Thus, the ciliate TEBP α-β dimer has a human homolog, POT1-TPP1.
The expectation was that the binding of POT1-TPP1 to a telomeric DNA primer would inhibit telomerase access if the protein complex were near the DNA 3' end, or have no effect if the protein were binding farther from the DNA end. Instead, human POT1-TPP1 acts as a processivity factor, allowing telomerase to synthesize multiple DNA repeats before dissociation. Current goals of our research are to pinpoint the interactions between TPP1 and telomerase responsible for processivity and to test the importance of these interactions in vivo.
Our group will continue to employ structural biology, biochemistry, and yeast molecular genetics to understand the structure and function of telomeric DNA-protein complexes and of the telomerase enzyme. In addition, we are collaborating with medical researchers to better understand the roles of telomerase mutations in human disease.
Recently many large noncoding RNPs have been implicated in transcriptional regulation in mammals. Two major systems are the XIST RNA, required for X-chromosome inactivation in females, and HOTAIR, which regulates expression of the developmentally important HOX genes. We are applying some of the approaches developed for telomerase to understand the roles of such long noncoding RNPs, with special emphasis on RNA-protein interactions and molecular structure.