Learning-mediated plasticity in cortical feedback projections to the olfactory bulb

Funding: NIH NIDCD K99 DC017754

Odor information in the olfactory bulb is bi-directional. In the feedforward pathway, incoming information is carried by the axons of olfactory sensory neurons, which activate mitral/tufted cells. In the feedback pathway, information is delivered by the axons of principal neurons in the piriform (olfactory) cortex to granule cells, which make reciprocal inhibitory synapses with mitral/tufted cells.

To reliably encode information about the environment, neurons must modify their activity profiles and even connectivity to accurately interpret complex stimuli. Cortical feedback projections to the olfactory bulb are uniquely positioned at the interface between detection-based processing that is driven by sensory input and analytical processing occurring in the piriform cortex. This arrangement makes these projections an ideal target to study how learning reshapes neuronal activity profiles. I have developed an approach that will allow for a comprehensive analysis of the axonal activity of principal neurons in the piriform cortex, while mice learn a task requiring them to identify a specific odor embedded in complex mixtures, thereby providing unique insight into olfactory scene analysis. My approach will also provide a detailed analysis of the connectivity between cortical axons and their postsynaptic targets in the olfactory bulb, which will reveal how the olfactory bulb integrates processed information from the piriform cortex. The outcomes of these studies will provide novel insight to how the brain to updates its stimulus-encoding scheme from a synthetic to analytical representation of a stimulus environment.


This work is ongoing and is the mentored phase of a National Institutes of Health Pathways to Independence Award. This project is co-mentored by Profs. Venkatesh Murthy and Naoshige Uchida at Harvard University.





Information coding in olfactory sensory neurons

Funding: NIH NIDCD F32 DC015938

Left, olfactory sensory neurons, found in the olfactory epithelium, are sensitive to molecular features of odor molecules. Each sensory neuron expresses a single receptor type and those of a common type project their axon to a common domain on the olfactory bulb called a glomerulus. Right, some odors can compete for the same receptor. Upon odor biding these receptors trigger a signaling cascade that results in action potential generation.

Animals, including humans, interact with their chemical environment through specialized receptor cells found within the nasal epithelium. Upon odorant binding, these neurons transmit information to the olfactory bulb, the initial site of sensory processing in the olfactory system. However, our understanding of how these cells ultimately communicate information about odor identity and concentration to the brain is limited.


In this project we first developed a theoretical framework describing how blends of odor stimuli are detected and encoded by olfactory sensory neurons, with a focus on antagonistic interactions. A key prediction of the model is that antagonism in sensory neurons could normalize input to the olfactory bulb. I then tested the hypotheses generated by our model using in vivo calcium imaging to measure the stimulus-evoked activity in olfactory sensory neurons in live animals. This work revealed that olfactory sensory neurons are far from simple relays and that their nonlinear interactions fundamentally affect olfactory processing. Another component of this project investigated the role of ion channels downstream of odor receptor binding to how olfactory sensory neurons detect odors and transmit odor information to the brain.


This work was completed as a postdoctoral research fellow in the laboratory of Prof. Venkatesh Murthy at Harvard University. These studies resulted in the following publications:

  1. Zak JD, Reddy G, Vergassola M, Murthy VN (2019) Antagonistic odor interactions in olfactory sensory neurons are widespread in freely breathing mice. bioRxiv. 847525
  2. Albeanu DF, Provost AC, Agarwal P, Soucy E, Zak JD, Murthy VN (2018) Olfactory marker protein (OMP) regulates formation and refinement of the olfactory glomerular map. Nat. Commun. 9:5073
  3. Zak JD, Grimaud J, Li R-C, Lin C-C, Murthy VN (2018) Calcium-activated chloride channels clamp odor-evoked spike activity in olfactory receptor neurons. Sci. Rep. 8:10600
  4. Reddy G**, Zak JD**, Vergassola M, Murthy VN (2018) Antagonism in olfactory receptor neurons and its implications for the perception of odor mixtures. eLife. 7:e35958 **equal contribution




Balancing excitation and inhibition within olfactory bulb glomeruli

Funding: NIH NIDCD F31 DC013480

Sensory input to olfactory bulb glomeruli is delivered by the axons of olfactory receptor neurons. These axons provide direct input to local excitatory neurons (eTCs) and some inhibitory neurons (PG cells) that surround glomeruli. eTCs also provide excitatory input to PG cells. PG cells supply feedback inhibition to eTCs and feedforward inhibition to mitral cells, an olfactory bulb output neuron. Although, sensory neurons synapse with mitral cells, much of their excitation is supplied by eTCs. This project explored the how relationship between eTC and PG cell activity relates to signal propagation through olfactory bulb glomeruli.

Input to the olfactory system arises from sensory neurons within the epithelium of the nasal cavity. Sensory neurons project axons to specialized networks in the surface of the olfactory bulb called glomeruli that each receive input from a single sensory neuron subtype. While each subtype of sensory neuron has a preferred odorant ligand, they may also bind other ligands across a range of affinities. Therefore, in the presence of any odor, a few glomeruli will receive strong input from sensory neurons, while many others will receive weaker input, thereby potentially obscuring the brain's interpretation the odor environment of an animal.


A potential mechanism to overcome this limitation is through filtering of weak signals from “non-preferred” ligands. In this project, I tested whether the local network of neurons that surround glomeruli could perform this function. These cells, which include glutamatergic external tufted cells (eTCs) and GABAergic periglomerular (PG) cells, act together as a signal detection mechanism that selectively filters weak sensory inputs in favor of strong signals. Two of the primary goals of this project were to (i) test whether PG cells or eTCs are preferentially activated by differential levels of sensory input. This allows for low-level olfactory input to selectively activate inhibitory PG cells, thereby filtering weak sensory input signals. (ii) To determine if PG cell activity can be modulated by local glutamate released from eTCs, which may increase the effectiveness of weak signals and produce temporal patterning in glomerular activity.


This work was completed as my dissertation work in the laboratory of Prof. Nathan Schoppa at the University of Colorado School of Medicine. These studies resulted in the following publications:

  1. Gire DH**, Zak JD** Bourne JN, Goodson N, Schoppa NE (2019) Balancing extrasynaptic excitation and synaptic inhibition within olfactory bulb glomeruli. eNeuro.6(4)0247-19.2019 **equal contribution
  2. Zak JD (2016) A computational framework for temporal sharpening of stimulus input in the olfactory system. J. Neurophysiol. 115(4):1749-51
  3. Zak JD, Whitesell JD, Schoppa NE (2015) Metabotropic glutamate receptors promote disinhibition of olfactory bulb glomeruli that scales with input strength. J. Neurophysiol. 113(6):1907-20.
  4. Gire DH, Franks KM, Zak JD, Tanaka KF, Whitesell JD, Mulligan AA, Hen R, Schoppa NE (2012) Mitral cells of the olfactory bulb are mainly excited through a multi-step signaling path. J. Neurosci. 32:2964-75