Our interest in miRNAs began with a desire to solve the fundamental question of how organs form at the right time during development [1, 4], and this core theme has since led us to a broader interest in the role of miRNAs in human disease and aging [5, 6]. We initially focused on using C. elegans to find conserved genes and molecules that control developmental timing and to understand the underlying mechanisms utilized, but more recently we have focused on translating this information to the analysis of more complex organisms. For example, since many of the C. elegans developmental timing genes are related to human cancer genes, we have moved to mammals to examine the role of their homologues in controlling timing of cell differentiation and disease. In one example, let-7, a founding member of the miRNA family, is known to control essential development processes in stem cells such as cellular differentiation, in large part through our work [2, 7, 8]. In humans, we have shown that miRNAs are key cancer genes (see below), and administration of tumor-suppressor miRNAs to mice with cancer causes dramatic tumor regression.
We demonstrated for the first time that a miRNA directly binds its target mRNA in vivo and described the minimal sequences necessary for miRNA control of target sequences [9, 10]. We used this information to design one of the first bioinformatics screens to identify novel miRNA targets in an effort to understand how these miRNAs control differentiation [3, 11]. Most of the dozen or so targets we identified encode transcription factors, leading to our assertion that these miRNAs are master temporal control genes. Another target of let-7 was the C. elegans homologue of the human proto-oncogene RAS (see below), which proved to be a conserved interaction in human cells [3]. This led to our intense analysis of roles for miRNAs in cancer pathways, in both human cells and in mouse models (see below).
Future work: We have taken an in-depth approach to investigating all aspects of the lin-4 and let-7 miRNAs as models for miRNA biology, leading to these miRNAs being the best understood of any miRNAs and we will continue to dig deeply into these models. Specifically, we are interested in what regulates these miRNAs, where and when are they expressed, what their roles are, what their targets are, what pathways they function in and the extent of their reach in C. elegans and mammals. We showed that both the lin-4 and let-7 RNAs are transcriptionally regulated [12] and begin to be expressed at critical times in development. We have dissected the promoter regions of lin-4 and let-7 and have identified temporal control elements that bind to proteins, which we are pursuing [12]. let-7 is regulated by LIN-28 protein in many animals and we have undertaken an unbiased screen for LIN-28 binding sites in RNAs. Our data shows that LIN-28 regulates many genes, including let-7. These are high priority for further study, especially since LIN-28 is a key stemness gene in C. elegans and mammals [13].
We conducted a genome-wide RNAi screen for suppressors of the lin-4 and let-7 miRNA mutants. We identified around 50 genes that can suppress the lethality of a let-7 mutant [11, 14]. These include important transcription factors; signal transduction molecules; RNA metabolism genes; and other genes with unknown functions but for which there are human disease homologues, like apl-1/APP. Given the lack of functional information about APP, a key gene in Alzheimer’s disease, we are investigating apl-1 in detail. We found that apl-1 acts in our developmental timing pathway, and its expression is temporally regulated late in development by heterochronic genes and miRNAs, like let-7 [15]. This work provides insights into the age-related increase in APP expression in Alzheimer’s disease. It also pointed to human APP as a target of miRNAs, and that these miRNAs may be potential Alzheimer’s therapeutics. This work also demonstrates the close relationship between developmental timing, aging and disease, an emerging field of which we are at the forefront.
As part of the modENCODE project we have used next-generation sequencing to identify hundreds of novel, temporally expressed miRNAs during development in C. elegans [16-18], most of which remain of unknown function. In addition to the lin-4 and let-7 families, recent studies have shown that another five out of thirty-one temporally regulated miRNAs in C. elegans possess mammalian miRNA homologues. Therefore, this group of potentially important, highly conserved miRNAs is being analyzed in my lab for informative expression patterns and phenotypes during development. The hypotheses we are testing are that conserved, temporally expressed miRNAs will be essential for proper growth and development in C. elegans. For example, we have found that mir-34, a homologue of a human tumor-suppressor miRNA activated by p53 [19], is temporally expressed in the vulva and plays a role in the radiation-sensitivity of these cells [20]. This observation was the first in vivo phenotype for this important miRNA. We also hypothesize that their homologs will play important roles during mouse development. Further, as part of this analysis we developed a protocol for in situ hybridization of mammalian miRNAs that is now standard in the field ([21]). Thus, I expect our “discovery” work in C. elegans will continue to inform studies in more complex organisms.