In initial studies, we discovered the odorant receptor (OR) family, which mediates odor detection in the nose. This is the largest protein family known, with ~1000 members in mice and 350 in humans. Each olfactory sensory neuron in the nose expresses a single OR gene and thousands with the same OR are dispersed. However, their axons all converge in a few OR-specific glomeruli in the olfactory bulb, producing a semi-stereotyped map of OR inputs. Our studies revealed that ORs are used in a combinatorial fashion, with different odorants detected, and thereby encoded, by different combinations of ORs. This combinatorial coding strategy explains how we can distinguish a multitude of odorants and how odorants with similar structures have different scents. Together, these findings indicated that the spatial code for an odorant in the nose is a dispersed ensemble of neurons expressing ORs comprising its receptor code whereas in the olfactory bulb it is a combination of glomeruli that receive input from those ORs and whose spatial arrangement is similar among individuals. How olfactory information is next encoded in the olfactory cortex and then higher levels of the nervous system is still largely a mystery.

We later discovered a very small family of trace amine-associated receptors (TAARs) in the nose that detect volatile amines, including one aversive to humans. We and others also identified two families of receptors in the vomeronasal organ, an accessory olfactory organ that can detect pheromones, as well as candidate receptors for bitter and sweet tastes on the tongue. These findings laid an additional foundation for exploring how environmental cues are initially detected and how they generate distinct physiological changes and behaviors.

To investigate neural circuits that underlie pheromone effects on reproduction, we developed a genetic method for tracing neural circuits with a plant lectin. When expressed in hypothalamic GnRH neurons, which control reproductive hormones, the lectin traveled to upstream and downstream neurons in numerous brain areas. These included the olfactory cortex, indicating that pheromones that influence reproduction can be detected not only in the vomeronasal organ, as previously thought, but also in the nose.   

Predator odors elicit instinctive fear responses in mice. These include fear behaviors and increases in blood stress hormones controlled by hypothalamic corticotropin releasing hormone neurons (CRHNs). To investigate how volatile predator odors detected in the nose stimulate CRHNs, we infected CRHNs with Cre recombinase-dependent Pseudorabies viruses (PRVs) we developed that travel retrogradely (backwards) through neural circuits. We observed neurons upstream of CRHNs in multiple brain areas, including the olfactory cortex. Predator odors activated neurons indirectly upstream of CRHNs in only one small area of the olfactory cortex, the AmPir.  Chemogenetic activation of AmPir induced stress hormone increases, mimicking a predator odor, and AmPir silencing inhibited the stress response to predator odors. Thus, AmPir, which comprises <5% of the olfactory cortex, plays a key role in the physiological fear response to predator odors.

These findings raise a number of intriguing questions we would like to answer. These include whether there are specific ORs in the nose and/or molecularly identifiable neuronal subsets in the olfactory bulb, AmPir, or a downstream relay that are involved in predator odor-induced effects on CRHNs. Another is whether different predator odor signals travel to CRHNs via parallel or the same circuits and the relationship between those circuits and those through which other stressors activate CRHNs.

Our viral tracing experiments revealed neurons directly upstream of CRHNs in 31 different brain areas. CRHNs control stress hormone responses to a variety of different stressors, including not only predator odors, but other external and internal stressors, such as psychological stressors and injury. We reasoned that the identification of molecular markers for subsets of upstream neurons would provide molecular/genetic tools for dissecting the roles of those subsets in responses to different stressors.

We recently devised two novel methods to define molecular markers for neurons upstream of CRHNs. In the first, called “Connect-seq”, we infected CRHNs with a Cre-dependent PRV expressing GFP. We then used flow cytometry and single cell RNA-sequencing (RNA-seq) to define the transcriptomes of single upstream neurons. Surprisingly, many upstream neurons contained combinations of signaling molecules, including different neuropeptides. In the second method, called “RAMUN” (receptor assisted mapping of upstream neurons), we employed single cell RNA-seq to define receptors expressed by CRHNs and determined those with known ligands. Following PRV infection of CRHNs, we examined upstream PRV-infected neurons for expression of molecules identified in upstream neurons using Connect-seq, or ligands of CRHN receptors. These studies revealed molecular markers for a number of subsets of neurons upstream of CRHNs in 17 different brain areas, laying a foundation for studies to dissect their functions in different stress responses. In these studies, we also pinpointed the locations of upstream neurons activated by different stressors and defined one subset involved in the stress hormone response to one psychological stressor but not another. We would now like to investigate the roles of other subsets in different stress responses, including those involved in reactions to different predator odors as well as blocking odors.

Our lab is also interested in how the olfactory system influences appetite. The arcuate nucleus of the hypothalamus contains AGRP (agouti-related peptide) neurons that enhance appetite as well as POMC (pro-opiomelanocortin) neurons that suppress appetite. Using PRVs to infect those two neuronal subsets, we found neurons indirectly upstream of both AGRP and POMC neurons in the olfactory cortex. This raises a number of questions regarding what kind of odor signals are being sent to AGRP and POMC neurons, the stimulatory or inhibitory effects of those signals, the circuits that carry olfactory cortical signals to AGRP and POMC neurons, as well as whether there are specific ORs or molecularly identifiable neurons involved in appetite control in the olfactory bulb or olfactory cortex.