In an organism like the human body, every cell holds a replica of the individual’s genome. Although all cells share identical genetic information, they can take a variety of shapes and functions to build functional organs and tissue. This is the result of each cell generating proteins only from a specific subset of the ~25,000 genes encoded in the genome. The gene expression process (synthesizing mRNA and protein from a gene) constitutes the basis of differentiation, morphogenesis, and adaptability.
We are interested in understanding how gene expression works, focusing mainly on its first step, transcription (the synthesis of mRNA from a gene encoded in DNA). Evidence shows that transcription is highly regulated: for instance, some genes are transcribed at precise times in the developing embryo; other sets of genes turn on only when the environment changes. However, the order and predictability we observe at the tissue scale contrasts with the random behavior of individual molecules: synthesis of one mRNA copy requires that the transcription machinery components (dozens of proteins) find the precise sequence of DNA where transcription has to start. The randomness of the search process (it occurs through diffusion) coupled with the small number of molecules involved (one or two copies of the DNA) results in natural fluctuations in the RNA levels that can have consequences for the cell’s fate.
One crucial question is how ordered patterns of gene expression can emerge from random dynamics at the molecular scale. We are particularly interested in exploring the role of the vast portions of DNA that do not encode RNA or proteins. Elucidating the mechanisms that drive selective gene expression will better our ability to produce the different cell types needed to regenerate tissues; it will also provide clues to treat diseases in which gene expression is misregulated, such as cancer.
We believe combining approaches from distinct fields is key to linking molecular scale and downstream effects on cells and tissue behaviors. We develop advanced fluorescence imaging to measure the transcription of individual copies of mRNA in living cells. We combine it with genomics and computational biophysics tools to build predictive models.
Innovative imaging techniques to quantify transcription dynamics in living cells.
Building on our recent technological developments (new fluorescent dyes and photoactivatable dyes, imaging of selected loci,) and insights (the formation of clusters of PolII around transcribing genes), we are extending our exploration of the regulation of transcription at the molecular scale. This requires new creative ways to image and manipulate multiple factors at the gene locus in order to understand how the random, transient binding of factors at the gene can produce long-lasting, productive transcription events. Also of interest are the architecture and dynamics of the DNA itself, in order to understand how chromatin movements constrain or enhance transcription specificity.
Systems Analysis of Mechanosensing
Cells respond to a variety of signals from their environment by transcribing different sets of genes. In addition to chemical signals, mechanical signals such as forces or tissue stiffness can affect gene expression and cell fate. This is important for instance during wound healing when the increasing stiffness of the fibrin clot protecting the wound drives cells towards distinct roles; when misregulated, this process leads to scaring and fibrosis. To better understand this process, we use imaging to observe sequences of events in individual cells, from the initial deformation and signaling events to the activation of downstream genes. These datasets help us uncover the kinetics and robustness of the various steps, from which we generate mathematical models of the response process. Using genomics, we can explore the networks and molecular strategies that evolved to produce the observed responses. A mechanistic understanding of these processes has the potential to uncover novel therapeutic strategies for wound healing; it also holds the promise of harnessing mechanical means to more efficiently drive cellular differentiation into desired cell types.