Our Scientists

In most eukaryotes, the ends of chromosomes are capped by a series of short G-rich DNA repeats bound by a set of specific proteins. These so-called telomeres distinguish chromosome ends from DNA double-strand breaks and must continuously be replenished in dividing cells. Telomerase is the primary means of telomere elongation, and the enzyme is up-regulated in more than 80 percent of cancers. Conversely, mutations in telomerase components are linked to degenerative syndromes, including dyskeratosis congenita, aplastic anemia, and idiopathic pulmonary fibrosis. Much of our research is aimed at deciphering the molecular mechanisms that underlie telomere protection and telomerase biogenesis.

Chromosome End Protection
Cells that store genetic information on linear chromosomes face a fundamental challenge: chromosome ends must be distinguished from DNA double-strand breaks. We are interested in defining the molecular mechanisms that prevent DNA repair factors from acting at chromosome ends. Widely accepted models involve the presence of large complexes (telomeric heterochromatin) or telomeric loop structures (t-loops).

We have used a biochemical approach to define the minimal requirements for the protection of telomeric DNA ends from nonhomologous end joining (NHEJ), a major DNA double-strand break repair pathway in mammalian cells. Neither long, single-stranded overhangs nor t-loops were required to prevent illegitimate repair of telomeric ends. Instead, a tandem array of 12 telomeric repeats impeded NHEJ in a highly directional manner consistent with the orientation of naturally occurring telomeres. Subsequent biochemical and cell biological experiments revealed that telomere protection is mediated by a TRF2-hRAP1 complex, providing the first evidence for a direct role for human RAP1 in the protection of telomeric DNA from NHEJ. Our current and future efforts are aimed at defining the molecular mechanism by which hRAP1 blocks the NHEJ machinery.

Progressive telomere shortening in cells that lack active telomerase eventually causes chromosome ends to be recognized as DNA double-strand breaks. Fusions between the ends of different chromosomes produce unstable dicentric products, but a fusion between the two ends of one chromosome generates a relatively stable circular chromosome. Such ring chromosomes have been observed in cells from a variety of organisms, including humans.

The fact that the haploid fission yeast genome is distributed over only three chromosomes favors intrachromosomal fusions in this organism, compared to most others with higher chromosome number. When fission yeast cells fail to maintain telomeres, both inter- and intrachromosomal fusions can be observed. While cells harboring interchromosomal fusions succumb to mitotic catastrophes, cells with three circular chromosomes quickly emerge as stable survivors that are amenable to further analysis. Taking advantage of this phenomenon, we have identified the single-strand annealing pathway as the repair process responsible for telomere attrition-induced chromosome fusions. Furthermore, we found that the circumstances and causes of telomere dysfunction profoundly affect which DNA repair pathway is engaged at denuded chromosome ends. We are building on these results to understand how chromosome fusions triggered by various insults to telomeres contribute to genome instability.

Telomerase Biogenesis and Regulation
While studying telomere end protection, we noticed that certain mutations in the protein Pot1 (protection of telomeres) cause dramatic telomere lengthening, indicative of a role in regulating telomerase. This observation rekindled my interest in telomerase. Although the core components of telomerase have been known for some time, our understanding of telomerase assembly, function, and regulation is still limited. As part of the initial characterization of fission yeast telomerase, we established a precursor-product relationship between longer poly-adenylated RNAs and a shorter form of the RNA found in the active telomerase complex. We designed a series of experiments to examine how the precursors are trimmed to generate the mature length. These studies led us to the highly unexpected realization that the spliceosome has a function in 3'-end processing of this noncoding RNA. Instead of removing an intronic sequence in a two-step process, the first transesterification reaction alone generates the mature 3' end.

Termination after the first step of splicing would be highly detrimental if it occurred during intron removal due to the release of a truncated and nonfunctional RNA. For this reason the two steps in splicing are normally tightly coupled, and at least for some introns, mutations at the 3' splice site block the first step of splicing, which is consistent with the spliceosome not starting a job that it cannot finish. We are using biochemical assays and molecular genetics to gain insights into the mechanism by which the 5' exon is released after the first cleavage has occurred.

We are also examining other steps in telomerase biogenesis, such as transcriptional regulation, RNA editing, and assembly of the holoenzyme. Downstream from these events, the questions of how telomerase is recruited specifically to the shortest telomeres and how the activity of the enzyme is regulated via post-translational modifications remain to be answered. Our research goals in this area are guided by the belief that a better understanding of the dynamic interactions that occur at telomeres will ultimately enable us to identify compounds that modulate telomere length. Such reagents will have therapeutic use to limit the life span of tumor cells and to boost the proliferative potential of desired cell populations.

Parthenogenesis in Vertebrates
Although bisexual reproduction is by far the most common and by inference successful mode, unisexual species are found on many branches of the tree of life. In theory, unisexuals have a powerful advantage over their sexual counterparts, often referred to as the "twofold cost of sex." First, no energy is expended in searching for and choosing a mate. Second, each individual is capable of producing offspring, translating into faster population growth. In reality, however, bisexual species outnumber unisexual lineages by 1,000 to 1, a fact that illustrates that the former strategy provides substantial if not fully understood benefits. An obvious advantage for sexual reproduction is the genetic diversity that results from the reshuffling of alleles at every generation. This increased variation among individuals in sexual species aids in the long-term survival. In contrast, unisexual species are thought to be genetically isolated and clonal, decreasing their chances to compete successfully in a changing environment.

Nevertheless, some unisexual lineages have persisted for many millions of years, and unisexuality is not limited to more primitive life forms. On the contrary, more than 70 species of unisexual vertebrates have been described. We are specifically interested in whiptail lizards of the genus Aspidoscelis. It was demonstrated more than 30 years ago that parthenogenetic Aspidoscelis lineages originated by cross-hybridization of sexual species. In contrast to the sterility observed in most cross-species hybrids, some hybridization events gave rise to parthenogenetic lineages that have colonized large areas where they have been competing successfully with their bisexual relatives.

Surprisingly, the molecular events that permit parthenogenetic reproduction have barely been examined. We have now established captive colonies of several sexual and parthenogenetic species and are studying the mechanism underlying parthenogenetic reproduction. In particular, we are interested in how diploid, and in some species triploid, ova are generated, and how much—if any——genetic diversity exists among parthenogenetically produced offspring. In light of a growing number of reports of incidences of parthenogenesis in species that normally reproduce bisexually, we are also curious to examine the events that may trigger such a switch in reproductive strategy.

This research is supported in part by the Stowers Institute for Medical Research.

As of May 30, 2012