Gene co-regulation in the developing olfactory system

The olfactory system is a primal sensory system for most animals. The sense of smell is required in order for animals to detect food sources, avoid predators, and find mates. Our laboratory is interested in the development of the mammalian olfactory system in consideration of two specific questions: (1) How do olfactory neurons develop the capability to detect and distinguish among the thousands of olfactory cues in the environment? (2) How are olfactory neurons and supporting olfactory tissues maintained given constant insult from toxins, viruses, and bacteria inhaled into the nose?

Odorant receptor transcriptional gene regulation underlies the ability to distinguish odorants

Encoded in the animal’s genome are about 1,000 odorant receptor (OR) proteins that are expressed in the olfactory sensory neurons of the nose. Each OR protein interacts with a particular chemical structure present in an odor, and upon binding that structure, the neuron fires an action potential to communicate the presence of that chemical structure to olfactory processing centers of the brain. A critical organizing principle during development of olfactory sensory neurons is that each neuron expresses one, and only one, OR protein. Thus, each sensory neuron is tuned to detect a specific chemical structure; each “smell” is recognized as a composition of several specific chemical structures. A major component of our research is focused on the gene regulatory problem that underlies sensory neuronal specialization: how does each neuron express only one OR gene and silence all the other ~999 OR genes?


LSD1 protein as the “gatekeeper” for OR transcription.

A. LSD1 protein (red) is broadly distributed in the nucleus of an early developing olfactory sensory neuron (a single nucleus is shown). B. LSD1 protein (red) becomes consolidated into a single spot in the nucleus at later stages, and this LSD1 compartment associates with only one OR gene (all OR gene loci are shown in green). The inset panel zooms in on the single red LSD1 compartment with only one OR gene locus in association (yellow spot).

C. Only one OR gene is transcribed per cell (single green spot detecting OR RNA at the one transcribed locus in the center of this cell). D. When LSD1 protein is knocked down using RNAi techniques, multiple OR loci are transcribed (multiple green spots detecting OR RNA at numerous transcribed loci at once in this cell).


An epigenetic model for OR co-regulation

Work in our lab and elsewhere has clarified that the selection of one and only one OR gene in each developing neuron is established by licensing only one OR locus with permissive chromatin, while maintaining repressive chromatin states at other non-selected loci. How does one OR locus become licensed in this way? Recent work has focused on a key chromatin regulatory protein, the lysine-specific demethylase-1 (LSD1), whose biochemical function is to remove methyl groups from histone-3 side chains within chromatin. Interestingly, LSD1 is able to function as either a repressor, through removing the activating histone methylation mark on the H3K4 residue, or an activator, through removing the repressive histone methylation mark on the H3K9 residue. As shown in the figure (Panels A&B), we have identified a shift in LSD1 protein distribution, where we hypothesize that the former stage is associated with globally repressing OR expression (via H3K4 demethylation) and the latter stage associated with selectively activating only one OR gene (via exclusive H3K9 demethylation) To further investigate, we are currently using transgenic/CRISPR-Cas9 techniques to perturb LSD1 function (see Panels C&D) in order to test the hypothesis that LSD1 functions as the key master regulator in OR selection (e.g., does LSD1 genetic perturbation result in premature OR activation at earlier stages and/or failure to select at later stages?). We are also using proteomic approaches to understand how (e.g., through LSD1 post-translational protein modifications and/or modulating binding partners?) and why (e.g., caused by some developmental or cell cycle cue that instructs the neuron to take this developmental step?) the LSD1 protein switches its functionality during this critical moment of choice during neuronal maturation.

G9a protein as a “gatekeeper” for cell-type stability? A. Cell differentiation depicted in a landscape metaphor, where normal differentiation of cell types flows irreversibly downhill along one valley. Does G9a perturbation cause up-hill “de-differentiation” into a more pluripotent stem cell type? B. G9a perturbation results in changes in cell morphology into a more spherical cell type (arrows) from a typical pre-neuronal cell shape (flattened background cells not marked by arrows). C. and D. Differentiated olfactory neurons have nuclei with heterochromatin distributed in various chromocenters (red blobs, panel C), whereas cells transformed by G9a perturbation have heterochromatin distributed at the nuclear periphery (panel D). E. and F. A core marker of stem cells, Sox2, is up-regulated in transformed cells (panel F) as compared to untreated neurons (panel E). G. and H. Another core marker of stem cells, Oct4, is up-regulated in transformed cells (panel H) as compared to untreated neurons (panel G).

Regenerative capability of the olfactory system

The nose is a messy environment, full of toxins, viruses, and bacteria. As observed with the recent Covid pandemic, temporary loss of the sense of smell is a common symptom of infection, as the virus destroys olfactory cell types as it enters the body through the inhalation of viral particles. Why is this loss of smell only temporary? The olfactory system is one of the few neurological systems capable of regeneration, accomplished primarily through a population of stem cells that reside at the base of the olfactory epithelium. Moreover, it appears that olfactory cell types are atypically “plastic” in their ability to trans-differentiate from one cell type to another (typically, differentiated cell types are robust and stable), presumably in order to meet pressing needs to replenish a certain cell type. Our lab is interested in this developmental “plasticity”. We have focused on a second chromatin regulating protein call G9a, which adds repressive H3K9 methylation marks to histone proteins. We have discovered that CRISPR-Cas9 depletion of the G9a protein can cause transformation of an olfactory neuronal cell type to a cell that has stem-cell like properties, including stem-cell like growth properties, changes in shape and nucleus size, gross modification of chromatin organization within the nucleus, and up-regulation of core stem-cell gene networks (e.g., Sox, Nanog, etc.). We are currently using RNA-seq to investigate the activation/deactivation of key genetic networks during this transformation process. In addition, we are testing the hypothesis that these transformed olfactory cell types have de-differentiated and have adopted stem cell characteristics (i.e., multipotency) by attempting to re-differentiate these transformed cells using common stem cell biological cell culturing techniques.

Overview of research goals for students

Our lab uses a wide range of molecular biological techniques ranging from genetics/transgenics, cell biology (cell culturing and microscopy), genomics and bioinformatics, proteomics (protein studies), and chromatin analysis in order to study the development of olfactory sensory neurons. The specific models we are testing should contribute to our understanding of epigenomics – the relationship between chromatin state, gene expression, and cell differentiation – an emerging field important for development of clinical strategies in gene therapy and stem cell research, as well as for our understanding of numerous chromatin-influenced diseases (e.g., cancer).

Current lab members (2022): Ghazia Abbas (Ph.D.), Tyler Burdick, Eleanor Walsh, Jenny Margolis.

Recent lab members (past 3 years): Rutesh Vyas (Ph.D.), Joyce Noble (Ph.D.), Spencer Tang (BA/MA), James Farber (BA/MA), Kate Louderback, Erica Horowitz.