Elucidating the Mechanism of Precision in Vertebral Segmentation

Abstract Despite unavoidable fluctuations in gene expression, embryonic development is robust and reproducible, which necessitates several mechanisms buffering stochastic gene expression. An intriguing example of robust spatiotemporal patterning is the rhythmic segmentation of somites, which are precursors of the vertebral column. Periodic segmentation of somites is controlled by the oscillatory expression of the Hes/Her gene family; known as the vertebrate segmentation clock. To measure the amplitude of oscillations and their cell-to- cell variability (noise), we counted RNA molecules transcribed by two master segmentation clock genes (her1 and her7) using single molecule fluorescent in situ hybridization (smFISH). We found low amplitudes, high noise and transcriptional bursts of her1 and her7 transcription in wild-type embryos. In Notch-signaling mutants, amplitudes of oscillations decreased due to reduced transcriptional bursts, and variability increased due to increased gene extrinsic noise. Furthermore, transcriptional noise increased from the posterior progenitor zone towards the anterior segmentation zone, in wild-type embryos. Loss of several factors involved in the basic machinery of transcription resulted in segmentation defects and reduced transcription of clock genes. These proteins, including Rtf1, Ctr9 and Spt6, release proximal-promoter paused Pol-II. The underlying mechanism remains elusive but our preliminary, yeast-two-hybrid data show that Spt6 interacts with Her7, one of the master segmentation clock regulators. In this proposal, we will test the following hypotheses built on our extensive preliminary data and literature: 1) polymerase pausing at the proximal promoters of clock genes causes bursts of transcription; the frequency of Pol II pausing is controlled by Her1/7 repressors and Notch activators, 2) the posterior-to-anterior gradients of Fgf, Wnt, and RA signaling activity control the observed spatial profile of transcriptional noise, 3) gene expression noise is buffered by redundancy in the clock machinery, as well as short- and long-distance cell-to-cell signaling: Aim 1. Determine the sources of stochastic fluctuations in the expression of segmentation clock genes. Aim 2. Investigate how signaling gradients buffer expression noise in the segmentation clock. Aim 3. Understand how noise propagation is suppressed downstream of the segmentation clock. Oscillations of Hes/Her proteins control the switch from proliferation to differentiation in various tissues. Their expression has been detected in certain cancers, while their inhibition restores differentiation. Elucidating the molecular mechanisms that guide their expression in somitogenesis is significant for understanding and potentially preventing vertebral malformations, but also for enhancing stem cell proliferation and developing therapies against cancer. Therefore, this application has strong relevance to the mission of the National Institute of Health.