Chemical Transduction in Taste and Smell

Chemical transduction refers to the conversion of a chemical stimulus — a tastant dissolved in saliva or an odorant molecule in inhaled air — into a graded receptor potential and ultimately a train of action potentials. In taste, the receptor cells are specialized epithelial cells clustered into taste buds on the tongue, palate, and epiglottis; each bud contains 50–100 cells representing multiple taste modalities. In olfaction, the receptor cells are true bipolar neurons, the olfactory sensory neurons (OSNs), embedded in the olfactory epithelium high in the nasal cavity, whose axons project directly into the central nervous system — one of the few places a neuron is one synapse away from the outside world. Five basic taste qualities are recognized: sweet, umami (savory/glutamate), bitter, sour, and salty. The first three use seven-transmembrane GPCRs from the T1R and T2R gene families; sour and salty use dedicated ion channels. This is a fundamentally important distinction: taste, superficially unified in the "five tastes" concept, is actually built from two mechanistically distinct transduction logics operating in the same organ. Olfaction differs by scale and coding strategy rather than by mechanism — nearly all of it runs through GPCR signaling. Mammals possess large multigene families of odorant receptors (ORs) — about 1000 genes in mouse, roughly 400 intact (functional) genes in human, the rest pseudogenes. Any single odorant molecule can bind several receptor types with different affinities, and any single receptor can bind several odorant molecules; the resulting combinatorial code — which subset of receptors fired, and how strongly — is what specifies odor identity, not any single "grandmother receptor" for rose or coffee. Critically, each mature OSN expresses functionally only one allele of one OR gene ("one neuron, one receptor"), and this receptor identity determines both its odorant tuning and its axonal target in the brain, linking peripheral chemistry to central topography — a preview of how Sensory Maps and Cortical Topography are built from receptor-level rules.

Taste transduction: sweet, umami, bitter (GPCR pathway). 1. Tastants dissolve in saliva and reach the apical microvilli of taste receptor cells (Type II cells) through pores in the taste bud. 2. Sweet compounds (sugars, some proteins, artificial sweeteners) bind a T1R2+T1R3 heterodimer. Umami compounds (glutamate, aspartate, often potentiated by IMP/GMP ribonucleotides) bind a T1R1+T1R3 heterodimer. Bitter compounds bind one or more of ~25 different T2R receptors in humans, each broadly tuned to structurally diverse toxins (alkaloids, thioureas, etc.). 3. Ligand binding activates the receptor's associated heterotrimeric G-protein, gustducin (α-subunit closely related to transducin, the phototransduction G-protein), together with Gβ3γ13. 4. The Gβγ complex activates phospholipase C-β2 (PLCβ2), hydrolyzing PIP2 into IP3 and DAG. 5. IP3 binds IP3 receptors (type 3) on the endoplasmic reticulum, releasing Ca²⁺ into the cytosol. 6. Rising Ca²⁺ opens TRPM5, a Ca²⁺-activated monovalent cation channel, depolarizing the cell. 7. Depolarization opens voltage-gated Na⁺ channels and eventually CALHM1 channels, which release ATP directly through the plasma membrane (Type II taste cells lack conventional synaptic vesicles for this step) — an unusual, non-vesicular transmitter release mechanism. 8. ATP excites purinergic P2X2/P2X3 receptors on the afferent gustatory nerve fibers (chorda tympani, glossopharyngeal), generating the sensory signal to the brainstem (nucleus of the solitary tract).

Taste transduction: sour and salty (ionotropic pathway).

• Salty taste (at least the low-salt, appetitive component) is transduced largely through the epithelial sodium channel ENaC on Type I taste cells: Na⁺ enters directly, depolarizing the cell. This is blocked by the diuretic amiloride in rodents, a classic pharmacological signature, though human ENaC is less amiloride-sensitive.
• Sour taste detects acidity (H⁺ concentration). The key sour receptor, identified in 2018 (Zhang, Ryba, and colleagues), is OTOP1, a proton-selective ion channel. Intracellular acidification through direct H⁺ entry via OTOP1 depolarizes Type III taste cells, which then release serotonin and norepinephrine via conventional synaptic vesicles onto afferent fibers.

Olfactory transduction (GPCR pathway, amplifying cascade). 1. Odorant molecules partition into the mucus layer overlying the olfactory epithelium, often assisted by odorant-binding proteins that solubilize hydrophobic molecules and help deliver them to receptors. 2. The odorant binds an odorant receptor (OR), a class-A GPCR embedded in the cilia membrane of the OSN dendritic knob. Each odorant typically activates a combination of several receptor types with graded affinity, and the same receptor is typically activated by several structurally related odorants. 3. Ligand-bound OR activates Golf, an olfactory-specific stimulatory G-protein closely related to Gs. 4. Golf-α stimulates adenylyl cyclase type III (ACIII), converting ATP to cAMP. 5. cAMP directly binds and opens cyclic nucleotide-gated (CNG) channels, permeable to Na⁺ and Ca²⁺, producing an initial depolarizing receptor current. 6. Ca²⁺ entering through CNG channels activates a Ca²⁺-activated chloride channel (ANO2/TMEM16B). Because OSNs maintain unusually high intracellular Cl⁻ (via the NKCC1 transporter), the chloride reversal potential is depolarizing rather than hyperpolarizing, so Cl⁻ efflux amplifies the depolarization roughly two- to five-fold — an amplification step that is crucial because odorant concentrations at threshold are extremely low. 7. Ca²⁺ also drives adaptation: it activates calmodulin, which reduces CNG channel cAMP sensitivity, and it feeds back on ACIII, limiting the response and allowing rapid resetting after a sniff. 8. Sufficient depolarization triggers voltage-gated Na⁺ channels at the axon initial segment, producing action potentials that propagate, unmyelinated, along the OSN axon through the cribriform plate directly into the olfactory bulb.

The glomerular map: from receptor identity to spatial code. Each OSN expresses one functional OR gene, chosen through a stochastic, mutually exclusive process (involving negative feedback from the expressed receptor protein itself on further OR gene transcription, plus epigenetic silencing of the other alleles). All OSNs scattered through the epithelium that express the same receptor gene—despite being physically dispersed across a large sheet of tissue—send axons that converge onto the same one or two glomeruli, discrete balls of neuropil on the surface of the olfactory bulb, out of roughly 1800 glomeruli per bulb in mouse. Convergence ratios are extreme: thousands of OSNs per receptor type converge onto a single glomerulus, improving signal-to-noise by averaging over many noisy inputs. This produces a genetically hard-wired, receptor-identity map: bulb position becomes a code for receptor type, which downstream cortex can read as a stable spatial pattern — the conceptual seed for how the brain builds ordered sensory maps generally.

Discovery of the odorant receptor gene family (Buck & Axel, 1991). Before 1991, it was not known whether olfactory receptors were a small handful of proteins or a vast family, nor how a nervous system could discriminate among thousands of odorants using so few sensory neuron types. Linda Buck and Richard Axel, working in rat olfactory epithelium, reasoned that odorant receptors should be GPCRs (by analogy to other chemosensory and light receptors known at the time) expressed specifically in olfactory tissue, with a diverse but related sequence (to bind diverse ligands) coupled to a G-protein already known to be enriched in olfactory cilia (Golf, identified by Randall Reed's laboratory shortly before). Using degenerate PCR primers targeting conserved transmembrane GPCR motifs, restricted to cDNA from olfactory epithelium and not other tissues, they isolated a strikingly large and diverse multigene family — they estimated 500–1000 distinct genes in rat, an order of magnitude larger than any previously known receptor family. This work, published in Cell in 1991, earned Buck and Axel the 2004 Nobel Prize in Physiology or Medicine. It directly established that the size of the odorant receptor repertoire, not specialized central circuitry alone, is a major solution to the discrimination problem, and it gave the field the actual molecular identity of the receptors whose transduction cascade (Golf–ACIII–cAMP–CNG) had been partly worked out biochemically by Reed and colleagues in the same years.

The "one neuron, one receptor" rule and the glomerular map (Ressler, Sullivan & Buck; Vassar, Chao, Sitcheran, Nunez, Vosshall & Axel; Mombaerts et al., 1994–1996). A separate and equally important question was whether each OSN expresses many receptor genes (making its response promiscuous) or essentially one. Using in situ hybridization with probes for individual OR genes on sections of rat and mouse olfactory epithelium, Ressler et al. and Vassar et al. showed that any given OR mRNA is expressed only in a small, scattered subset (roughly 0.1–1%) of OSNs within a defined epithelial zone, and that individual neurons rarely if ever showed two different OR signals simultaneously — establishing the singular expression rule. The decisive anatomical proof of convergence came from Peter Mombaerts and colleagues (1996), who used gene targeting to insert a tau-lacZ reporter into the coding sequence of a specific mouse OR gene (P2), so that every neuron expressing that receptor also stained blue for β-galactosidase. Even though P2-expressing neurons were scattered across a large expanse of epithelium, their axons converged with striking precision onto just two glomeruli, one on the medial and one on the lateral surface of each olfactory bulb, in every animal examined. This proved directly that receptor identity determines axonal target, converting a spatially disordered epithelial signal into an ordered, genetically reproducible glomerular map — the anatomical basis for treating glomerular position as a stable code for chemical identity.

Taste receptor gene discovery (Zuker, Ryba and colleagues, late 1990s–2000s). The T1R and T2R gene families were identified through a combination of expression cloning, genetic mapping to loci already known from mouse behavioral genetics to affect sweet and bitter sensitivity, and heterologous expression assays. Critically, Charles Zuker's and Nicholas Ryba's groups performed genetic swap experiments: expressing human T2R bitter receptors in mice made the mice averse to compounds that are normally tasteless to them but bitter to humans, and, conversely, engineering mice to express a single ectopic receptor (e.g., a capsaicin or opioid receptor) specifically in bitter-receptor-expressing taste cells, then testing whether the animal now treated the receptor's normal ligand as aversive like a bitter substance. These "ectopic receptor" experiments showed that the identity of the cell, not the identity of the receptor protein, determines the taste quality perceived — a labeled-line logic downstream of an initially combinatorial receptor step — directly supporting a coding model in which activating T2R-expressing cells produces "bitter" regardless of which chemical or engineered receptor triggered them.

Example 1 — Psychophysical dose-response and receptor occupancy for a bitter compound. Suppose a bitter alkaloid X activates a single human T2R receptor with a dissociation constant Kd = 4 µM, and detection threshold in a panel of tasters is defined as the concentration producing 10% of maximal T2R activation (a simple occupancy model, ignoring cooperativity, for teaching purposes). Occupancy (fraction of receptor bound), assuming simple bimolecular binding: ``` Occupancy = [X] / ([X] + Kd) ``` Set occupancy = 0.10: ``` 0.10 = [X] / ([X] + 4 µM) 0.10 ([X] + 4) = [X] 0.4 = [X] − 0.10[X] = 0.90 [X] [X] = 0.444 µM ``` So a threshold concentration near 0.44 µM should suffice for detection under this model — consistent with the general empirical finding that bitter thresholds (denatonium, quinine, etc.) sit in the sub-micromolar to low-micromolar range, orders of magnitude below sweet or salty thresholds (which are typically millimolar). This asymmetry has an evolutionary rationale often discussed in the field: bitter compounds mark potentially toxic plant alkaloids, so extreme sensitivity even at trace concentration confers a large survival advantage, whereas sugars and salts are needed in bulk as nutrients, so a coarser, higher-threshold detector suffices for their behavioral role. (This functional reasoning explains threshold evolution, not any developmental "purpose" of the receptor itself.) Now suppose a second compound, Y, is a partial agonist occupying the same T2R but producing at saturation only 40% of the maximal Gα-gustducin activation that X produces. Even at occupancy near 100%, compound Y will taste distinctly less intensely bitter than saturating X — illustrating that psychophysical intensity depends on both occupancy and the receptor's efficacy for a given ligand, a distinction directly analogous to pharmacological partial agonism at any GPCR.

Example 2 — Glomerular convergence and estimating discrimination capacity. A mouse has roughly 1000 functional OR genes and about 1800 glomeruli per olfactory bulb (so each receptor typically maps to one or two glomeruli, consistent with the ~2:1 ratio found for the P2 receptor). Suppose the olfactory epithelium contains roughly 10 million OSNs, evenly divided among the 1000 receptor types. 1. Average number of OSNs expressing a given receptor: ``` 10,000,000 / 1000 = 10,000 OSNs per receptor type ``` 2. If all of these converge onto 2 glomeruli, the average convergence ratio per glomerulus: ``` 10,000 / 2 = 5,000 OSNs → 1 glomerulus ``` This huge convergence (thousands-to-one) is thought to improve the reliability of the bulb's estimate of a receptor type's activation level: individual OSN spike trains are noisy (a Poisson-like process with a low mean rate at threshold odorant concentrations), but averaging synaptic input from 5,000 independent noisy afferents onto the same mitral/tufted cells reduces the relative error of the pooled signal by roughly the square root of the convergence number: ``` Relative noise reduction ≈ 1/√5000 ≈ 0.014 (a ~70-fold improvement in signal-to-noise over a single OSN) ``` This calculation (a standard population-averaging argument, not a report of a specific measured SNR value) illustrates why convergence architecture is functionally useful, independent of the map's role in preserving spatial identity: it turns a small, unreliable single-cell signal into a robust, averaged population signal that a small number of downstream mitral cells can read out reliably, one mitral/tufted cell cluster effectively representing "how strongly was receptor type 247 activated on this sniff."

Linda B. Buck trained as an immunologist before moving into Richard Axel's laboratory at Columbia University in the mid-1980s, where she took on the then-unfashionable problem of odorant receptor identity — many considered it nearly intractable given how little sequence information existed for any chemosensory receptor. Her decision to combine degenerate-primer PCR (targeting conserved GPCR transmembrane domains) with tissue-restricted expression screening was a methodological gamble that paid off spectacularly, revealing a multigene family far larger than any receptor family known at the time. Buck went on to run her own laboratory, first at Harvard Medical School and later at the Fred Hutchinson Cancer Research Center in Seattle, where she extended the work to the logic of odor coding, pheromone receptors, and the neural circuits linking olfaction to physiology and lifespan in animal models.

Loss of smell (anosmia) and distorted smell (parosmia) became prominent clinical signs during COVID-19, where SARS-CoV-2 infection of sustentacular support cells (which express the ACE2 receptor, rather than the OSNs themselves) disrupts the local environment needed for OR gene expression and OSN function, producing often-reversible but sometimes prolonged anosmia — a direct clinical application of understanding OSN biology and epithelial support cell dependence. Anosmia is also an early, sometimes pre-motor sign in Parkinson's disease and can predict cognitive decline in Alzheimer's disease, making olfactory testing a low-cost, non-invasive early screening tool under active clinical investigation. In agriculture and food science, knowledge of the T2R bitter receptor family underlies deliberate breeding-out of bitterness in crops such as certain cucurbits and brassicas, since some natural cucurbitacins and glucosinolates are potent bitter T2R agonists that reduce palatability and consumer acceptance; conversely, understanding T1R sweet receptor pharmacology drives the design of high-potency non-caloric sweeteners for the food industry, screened using heterologous cell lines expressing cloned human T1R2/T1R3. In pharmacology, bitter receptor agonism and antagonism are used to mask unpleasant drug tastes in pediatric formulations, an important practical problem in medication adherence, using T2R-expressing cell-based assays to screen bitter-blocking excipients before a drug ever reaches a human tongue. In biotechnology and neuroscience research, the "one-neuron-one-receptor" and glomerular convergence principles have become a general model system for studying stochastic gene choice and axon guidance, informing broader questions about how genetically identical neurons in a population can adopt distinct, reproducible identities — a theme echoed later when considering how any Sensory Map in cortex arises from orderly convergence of peripheral inputs, and how olfactory transduction's reliance on GPCR cascades connects directly back to the general principles of Principles of Sensory Transduction as a unifying logic across modalities, including vision's phototransduction cascade, to which the gustducin/Golf systems are evolutionarily and mechanistically related.

Ask the tutor: How does the stochastic choice of a single odorant receptor gene get stabilized in an olfactory sensory neuron, and what happens if that choice mechanism fails?