Date: December 14th, 2006

Speaker's Name and Affiliation:

Alan Spector, University of Florida
John Glendenning, Barnard College
Anthony Sclafani, Brooklyn College
Danielle Reed, Monell Chemical Senses

Title: A Symposium on Taste: The Genetic and Molecular Bases of Taste

Presiding Chair: Harry R. Kissileff, Ph.D.

Rapporteur: Kathleen L. Keller, Ph.D.

Attendees and their Affiliation:

Kathleen Keller	Columbia/Obesity Research Center
Harry Kissileff	Columbia/Obesity Research Center
Jennifer Nasser	New York Obesity Research Center
Chris Ochner	New York Obesity Research Center
Laurence Nolan	Wagner College
Emmanuel Pothos	Tufts University
Joel Grinker	University of Michigan
Allen Geliebter	New York Obesity Research Center
Janell Mensinger	SUNY Downstate Medical Center
Sarah Weinberger-Litman	Brooklyn College
Khalid Touzani	Brooklyn College
*Alan Spector	University of Florida
*Tony Sclafani	Brooklyn College
*Danielle Reed	Monell Chemical Senses
*John Glendinning	Barnard
Karen Acroff	Brooklyn College
Steven Zukerman	Brooklyn College
Carol Maggio	New York Obesity Research Center
Katherine Halmi	Cornell
Elizabeth Watson	Columbia
Nao Wabar	Columbia
Joe Vasselli	Columbia
Tim Kowalski	Sharing Plough
Gerry Smith	Weill Medical College
John G. Kral	SUNY Downstate
Rhoda Gruen	Columbia
Beverly Tepper	Rutgers
Xavier Pi-Sunyer	St. Luke's
George Collier	Rutgers

• Alan Spector, "Rewiring the Gustatory System: A Behavioral Analysis of Functional Consequences"

  Summary:

  The oral cavity is innervated by a number of different nerves, including the greater superficial petrosal (GPS), the chorda tympani (CT), the glossopharyngeal (GL), and the superior laryngeal (SL). In this presentation, Dr. Spector reviewed a wealth of data to suggest that regeneration of taste nerves can occur following transection of the GPS, CT, and GL in a rat model. Transection of the CT significantly shifts NaCl detection thresholds. Moreover, CT neurotomy impairs a rat's ability to discriminate sodium from potassium, a finding which is consistent with previous research noting that the branches of the 7th nerve are critical in qualitative discrimination. Interestingly, these functions return to normal after regeneration of the CT (Kopka and Spector, 2002), despite the fact that a range of other anatomical consequences remain, as reported in the literature. These include a 25% reduction in number of taste buds, a 40% reduction in taste bud volume, a 67% reduction in the number of myelinated fibers in the CT, and a reduction in the density of CT terminal projections in the nucleus of the solitary tract.

  The fact that some processes, such as gaping, survive cross-regeneration, while the ability to discriminate potassium from sodium does not, sheds some light on the functional organization of the gustatory pathways. In cases where a taste function "survives" cross-regeneration, there must be some "matching" mechanism for taste receptor cells and regenerational axonal fibers that allows for signals to be appropriately channeled through central signals.

  Discussion:
  

  Q. Does the water come from the same tube?
  A. In this version, we had a separate water spout that was rotated into a centralized access slot. We've since changed the procedure. Now, we have separate water tubes, one for each type of response.

  Q. Does the peripheral end of the nerve have any control over the receptor that it binds to?
  A. It remains unclear whether regenerating peripheral nerve fibers have the capacity to identify appropriate receptor cells. The fact that unconditioned gapes to quinine are normal after the central CT cross-reinnervates the posterior tongue suggests some matching mechanism must exist.

  Q. When you cut the nerve, are there stem cells in taste buds that you sever?
  A. I don't really know how you'd be able to show that. It is thought that, for the most part, the presence of morphologically intact taste buds requires an intact nerve supply.

  Q. If bitter taste detection is supposed to be protective, why would bitter tastes only stimulate the back of the tongue?
  A. I don't know. It's an interesting question, but difficult to answer. Bitter tastes stimulate all of the receptor fields, but they stimulate the back of the tongue the best in rats One possibility is that stimulation of the back of the tongue represents the last opportunity for the animal to expel the oral contents. In animals that are unable to vomit such as the rat, it is important to have that safeguard.

  Q. What do you think about the notion that there are taste receptors in the gut?
  A. Dr. (Gerald) Smith actually proposed that notion when I was a graduate student. I think you'll hear more about that in the talks that follow mine (Dr. Sclafani's and Glendenning's). There is definitely evidence of taste receptors and some intermediary enzymes important in taste transduction expressed in the gut. It remains unclear, however, what function they are serving.

  Q. If a function can be determined by the receptor field alone, what function does the brain serve?
  A. Well, the function is not completely determined by the receptor field, only partially. The brain is clearly important. The results of these experiments reflect the organization of the system as a whole. The brain needs the proper input; the input must be appropriately channeled to allow for normal behavioral responses to occur. In the case, of the cross-regeneration results from the experiment testing NaCl vs. KCl discrimination, it is likely that the brain is either not receiving the input from the proper receptors (i.e., chorda tympani cross-reinnervating posterior tongue) or the input from the proper receptors is not appropriately channeled centrally (i.e., glossopharyngeal cross-reinnervating the anterior tongue).

• John Glendinning, "Sugar Appetite in Mice: Oral, Post-Oral, and Genetic Factors"

  Summary:

  Humans have an innate drive to accept sugar that is present from birth. However, there are also genetic variations in sweetener intakes across humans and animals. Dr. Glendinning is using a mouse model to study the genetic and physiological bases for individual differences in sweetener consumption. In the 70s and 80s, Fuller and Lush used Mendelian approaches to determine that strain differences in saccharin preferences assort with allelic differences at a single gene, the Sac locus on the distal end of chromosome 4. In 2001, several laboratories discovered that the Tas1r3 gene is the Sac locus, and it encodes a G-protein coupled sweet taste receptor (T1r3). T1r3 dimerizes with T1r2 to form a functional sweet taste receptor, a homolog of which has been found in humans.

  In order to understand variations in intake of sweeteners, polymorphisms of Tas1r3 have been studied. Max et al. noted that two amino acid substitutions denote a "non-taster" strain, with respect to sweet taste. Two different strains, the 129 allele of Tas1r3 and the B6 allele of this gene explain a significant proportion of the daily intake of sucrose. The remainder of the talk discussed two central questions: 1) can strain differences in sweetener intake be explained by Tas1r3 polymorphisms alone, and 2) how does the Tas1r3 genotype contribute to strain differences in long-term intake of sweeteners.

  Data to address these central questions suggests that B6 and 129 differ with respect to their intake, preference, and motivation to consume sweeteners. B6 mice consume more non-caloric and caloric sweeteners over 24 hours, they have higher sensitivities to perithreshold concentrations of sweeteners, and they have high CT responses to sweeteners. In contrast, B6 animals lick less vigorously for high sweetener concentrations and they show less motivation to work for sucrose after protracted feeding experience with it.

  To further elucidate the relationship between genetic differences and sweetener intake, congenic strains of mice have been used to determine the role of the B6 allele. Specifically, does expression of the B6 allele in 129 mice "rescue" the B6 phenotype? Apparently, this congenic strain shows an increased CT responsiveness to some, but not all, sweeteners, and an increased lick responsiveness to low, but not high concentrations of sweeteners. Thus, the expression of the B6 allele partially, but not completely, restores the B6 phenotype, and additional genes likely contribute to strain differences in sweetener intake. In addition, post-oral positive feedback mechanisms and experience also contribute to differences in sweetener intake.

  Discussion:
  

  Q. If you increase the concentration of sucrose, could you recreate this intake pattern?
  A. Yes, up until the post-ingestive satiety kicks in.

  Q. Are there differences in the number of papillae in tasters and non-tasters?
  A. We didn't find any?
  

  Q. Do you have "motivation" evidence in your animals?
  A. I'll show you some data later, but I probably misspoke when I said "motivation.

  Q. The allelic differences between strains, what are the functional consequences at the receptor?
  A. One thought is that binding affinity for sweetners at the receptor site impairs either dimerization or ligand binding.

  Q. Do these strain differences (B6 and 129) affect other taste responses, or just response to sweet?
  A. Possibly, it will affect unami taste, but not sour, salty, or bitter.

  Q. When you look at 24 hour intake, do you take into account eating and drinking?
  A. We could monitor both eating and drinking, but chose not to do so in the present experiments.

  Q. When you refer to post-oral stimuli as positive feedback, are these truly "positive feedback?" Dr. Smith has referred to these as negative feedback.
  A. Now, data suggests there can be both positive and negative feedback.

• Tony Sclafani, "The Taste for Fat and Carbohydrate: Studies of Knockout Mice"

  Summary:

  Dr. Sclafani presented data from knockout mice to suggest that fat, starch, and polycose might be included among the basic tastes, at least in mice. There is accumulating evidence, first from Tim Gilbertson's lab in the late 90s, that fatty acids are detected by taste cells in the oral cavity. Additional studies suggest that the protein CD36, found in taste cells (Fukuwatari et al., 1997) might play a role in detection of fatty acids (Laugerette et al., 2005). CD36 knockout (KO) mice fail to display a preference for linoleic acid, as do wild-type (WT) mice. Sclafani observed that CD36 KO also did not prefer soybean oil at low concentrations but developed a strong oil preference at higher concentrations which appears to be mediated by the known post-oral reinforcing effect of fat. Additional experiments were conducted with Trpm5, gustducin (Gus) and P2X knockout mice. Gustducin and Trpm5 are part of the signaling pathway involved in sweet, umami and bitter taste transduction whereas P2X2 and P2X3 are the purinergic receptors on gustatory nerves activated by taste cells (Finger et al., 2005). Like CD36 KO mice, Trpm5 and P2X KO mice did not prefer soybean oil at low concentrations but did so at high concentrations. Gus KO mice, in contrast, were similar to WT in their preference for soybean oil at low to high concentrations.

  Gus KO mice showed an attenuated preference for saccharin, compared to WT mice, whereas Trpm5 KO and P2X KO mice were indifferent to saccharin. Gus KO, Trpm5 KO, and P2X KO mice showed no preference for polycose (a polysaccharide) at low concentrations preferred by WT mice. Trpm5 KO and P2X KO mice were also indifferent to corn starch at low concentrations whereas Gus KO mice were similar to WT mice in their starch preference. All KO mice developed a strong preference for polycose at high concentrations which is attributed to the post-oral reinforcing effect of the carbohydrate.

  In summary, the results obtained with CD36 KO, Trpm5 KO and P2X KO mice provide additional evidence for the "fatty" taste hypothesis. Gustducin, although co-localized with CD36 in taste cells, apparently is not involved in the gustatory response to fat. The failure of Trpm5 KO, Gus KO, and P2X KO mice to prefer dilute Polycose solutions supports the idea that rodents can taste polysaccharides as well as sugars. The finding that Trpm5 KO and P2X KO mice, but not Gus KO mice, also failed to prefer starch suggests that different receptors mediate the preference for polycose and starch. All mice developed strong preferences for concentrated soybean oil emulsions and Polycose solutions which indicates that CD36, Trpm5, gustducin and P2X signaling proteins in the gut are not essential for post-oral conditioned nutrient preferences.

  Discussion:
  

  Q. What happened to the ability to taste sour in your CD36 knockouts?
  A. We didn't look at that.

  Q. Why do you think acceptance and preference data are different in your animals?
  A. CD36 deficient animals have a deficit in fatty acid uptake in the upper intestine.

  Q. Did you also test what would happen if you increased concentrations of olestra?
  A. We did not increase concentrations of olestra. It can have negative ingestive consequences, as you are aware.

  Q. If the animals are detecting fat based on texture, might they confuse olestra for soybean oil, and then consume more of the olestra?
  A. Yes, that has happened in our experiments. There is some carry over.

  Q. What distinguishes polycose from amylose?
  A. Polycose contains shorter polysaccharide molecules and is soluble in water. Amylose contains much longer molecules and is insoluble in water.

• Danielle Reed, "Individuality of Human Taste Perception"

  The peripheral receptors for sweet, savory (umami), and bitter have recently been discovered, and what is known about their function was reviewed, drawing on the effects of naturally occurring and man-made alleles of these receptors, their shape and function based on receptor modeling techniques, and how they differ across animal species that vary in their ability to taste certain qualities. How taste genes may differ among people and give rise to individuated taste experience is not well-described as yet, but there is a high correlation between alleles of one bitter receptor, TAS2R38 and the perception of bitterness from the compound PTC. During the seminar, attendees tasted a low concentration of PTC and rated its bitterness on a four-point scale. Based on the perceptual ratings, allele frequency of the non-taster haplotype was inferred. In the attendees, the data suggested that the non-taster haplotype (composed of three amino acid substitutions at position 49, 262 and 296) was 54%, similar to other US samples. A highly concentrated sweet solution was also rated, with 50% of the attendees displaying a sweet tooth by this measure, finding the sweetness to be 'just right'. The molecular basis of this response and its degree of heritability is not known in humans, but individual differences are present.

  Discussion:
  

  Q. Does that mean if you go back in our ancestry, you will see the same type of variation with respect to sweet taste?
  A. Yes. Ancestors to dogs are able to taste sweet, but ancestors to cats cannot.

  Q. Are these data all collected from people in and around Philadelphia?
  A. Yes, but people from Philadelphia are very mixed in ethnicity, and they come from all over the world.

  Q. When you talk about frequency of AVI, are you talking about homozygotes and heterozygotes?
  A. Yes

  Q. What is the test-retest reliability on the PTC taste tests?
  A. It is very good. The ability to taste PTC is very reliable. The ratings of sweet taste are not as reliable.

  Q. I'm not clear on the molecular relationship between sweet and bitter.
  A. TAS1R receptors respond to sweet and umami. TAS2RS only respond to bitter compounds.