Search

The olfactory and immune systems must perform optimally in the task of recognizing thousands of molecules to ensure survival. A particularly intriguing link between these systems is that animals can smell differences in the major histocompatibility complex (MHC), a cluster of highly polymorphic genes found on human chromosome 6 and mouse chromosome 17. Two different sets of compounds found in urine have been postulated to convey information on MHC haplotype: volatile compounds (odortypes) and MHC peptides. Here we argue for complementary roles for these chemosignals.

Odors that distinguish one individual from another member of the species and are determined by polymorphic genes are called odortypes. Odortypes and their considerable societal significance have been studied intimately only in mice and mainly with respect to the genes of the major histocompatibility complex. Further understanding and the matter of human relevance have been hampered by the apparent restriction of odortype expression to urine. The present finding that odorants comprising prerenal odortypes are already present in blood, albeit in masked form, affords the basis of a comprehensive view of odortypes. Accordingly, major histocompatibility complex and other polymorphic genes of antiquity are seen inter alia as agents of normal variation, which entails quantitative variation in output of odorant metabolites. Relatively few such normal variations should suffice for a vast range of compound odors whose specificity is determined by combinative assortment of the same set of individual volatile compounds.

Genes located within the major histocompatibility complex (MHC) of mice are responsible for individual differences in body odor (odor types). In this review we suggest that the MHC genes themselves are responsible for odor differences among MHC-congenic mice. Recent studies indicating that volatile carboxylic acids are at least in part responsible for the individual odors and what this finding implies about the pathway from gene to odorant are also reviewed. We suggest that odorants or their precursors are bound directly by MHC products and are released into serum and concentrated in urine. Finally, possible functions of MHC odor types in mice are enumerated and important future research questions are raised.

In the last decade, studies using approaches from molecular biology have substantially advanced our understanding of the early events in olfaction and taste. The many odorants that we can recognize may well interact with many distinct receptor proteins. Of the four taste qualities that we recognize, studies on salty and sour suggest that these tastes involve ion channels in the membrane of receptor cells while sweets and bitters bind to receptor proteins. Some volatiles (pheromones) play special roles in reproductive behavior via the vomeronasal organ (VNO) and the accessory olfactory system. Initial belief that humans lack a VNO has been questioned recently, thus raising the fascinating possibility of human pheromones. The roles that taste and smell play in the world of the newborn are very different. Acceptance of sweet and rejection of bitter appear to be hard-wired while the affect associated with odors depends much more on experience. Genetic variation may produce total losses (Kallman's syndrome produces anosmia and familial dysautonomia produces ageusia) or losses specific to certain stimuli. The best known of the specific anosmias is that for androstenone, which has no smell to some, a urinous smell to others, and a smell like sandalwood to still others. Analogous to the specific anosmias, some individuals are unable to taste PROP while others, supertasters, perceive PROP to be exceedingly bitter. Clinical studies reveal pathologies responsible for total or partial losses. The olfactory system, dependent on one cranial nerve, is more vulnerable than taste, and total anosmia is a relatively common clinical problem. Three cranial nerves carry taste and two of those nerves inhibit one another such that damage to one disinhibits the other and preserves over-all taste function. Total ageusia is very rare. Throughout these studies we see that taste and olfaction have different properties and often different functions (e.g. odor and reproduction). Yet taste and smell can also be integrated to determine what does or does not enter the body. In Adrian's words, "we are dealing with the sense organs which signal the quality of the air we breathe and that of the food and drink we propose to swallow."

Differential effects of one or more genes in the nonpseudoautosomal region of the DBA1 and C57BL10 Y chromosomes on offensive attack may be mediated, at least in part, by differential effects of this Y chromosomal gene(s) on the sending and receiving of discriminable urinary odortypes. This hypothesis is based, in part, on the report that urine from a pair of Y chromosomal congenic strains (C57BL6.AKR-Y and C57BL6) of mice can be discriminated in a Y maze test. The AKR and C57BL6 Y chromosomes come from two distinct species of mouse (Mus domesticus and Mus musculus, respectively). Thus, this Y chromosomal variant exists between mouse species. The DBA1 and C57BL10 Y chromosomes come from a single species, Mus musculus. Here we show that in the Y maze system previously used, urine from mice with the DBA1 and with the C57BL10 Y chromosomes can also be discriminated. Thus, there are discriminable urinary odortypes for this pair of Y chromosomes from the same species, Mus musculus.

In investigating the ability of the rat to discriminate among urine odors from mice of different genetic strains, we developed a technique for determining the extent of odor generalization. Trained rats performed a discrete trial, go/no-go procedure, touching a bar in the presence of an S+ odor for water reward. Rats easily learned this task and restricted their responding to trials with the S+ stimulus, even though there was no penalty for responding to S-. However, when presented with test stimuli that might be more similar to S+ or S-, some rats responded only to the S+, and not at all to S- or any generalization stimuli. We then created series of stimuli composed of different levels of adulteration of the S+ stimulus with generalization stimuli. With these stimuli, rats produced graded levels of responding, allowing comparisons of similarity among odorant stimuli. This technique can be applied to other types of odorant stimuli, particularly biological compounds of unknown concentration or purity, and should be of use in an operant laboratory studying odorant perception, or odorant characteristics.

Although the majority of human infants are breast-fed for the first few months of life, there is a paucity of information regarding the sensory qualities of human milk and how these qualities are affected by maternal diet. The present study investigated the effects of garlic ingestion by the mother on the odor of her breast milk and the suckling behavior of her infant. Evaluation of the milk samples by a sensory panel revealed garlic ingestion significantly and consistently increased the perceived intensity of the milk odor; this increase in odor intensity was not apparent 1 hour after ingestion, peaked in strength 2 hours after ingestion, and decreased thereafter. That the nursling detected these changes in mother's milk is suggested by the finding that infants were attached to the breast for longer periods of time and sucked more when the milk smelled like garlic. There was a tendency for infants to ingest more milk as well; the lack of a significant effect may be due to the inherent limitations on the total amount of milk available to the infant.

Genes in the extended major histocompatibility complex (MHC) of the mouse (H-2: Qa:T1a) impart to each mouse an odor that reflects its genetic constitution at this region of chromosome 17. Sensory recognition of these differential odors influences reproductive behavior and evokes neuroendocrine responses critical to the maintenance of pregnancy. To determine whether other parts of the mouse genome contribute to individual odor, and so may similarly exert a selective force on loci other than the MHC, mice differing genetically only in their X and/or Y chromosomes were tested for individuality of scent in the Y-maze system previously employed to investigate MHC-related scent distinctions. It was found that the X and Y chromosomes each confer individuality of scent related to genotype, but differences at the H-2 locus are considerably more salient. Nevertheless, chemosensory cues controlled by differences on the Y chromosome could play a role in individual recognition, mate selection, aggressive interactions, and perhaps other aspects of mouse chemosensory behavior.

Genetic polymorphism in the H-2:Qa:Tla region of chromosome 17 is associated with constitutive variation of bodily odor phenotypes which permit individual olfactory recognition among mice. To determine whether known genes in the H-2:Qa:Tla complex are concerned in the constitution of odor phenotypes, mice were tested for their ability to sense a difference between the B6/By (H-2b) and congeneic B6.C-H-2bm1 strains, which differ genetically by mutation of the H-2K gene. As in previous studies of the sensory discrimination of H-2:Qa:Tla phenotypes, mice were trained by reward in a Y maze to distinguish the odors of urine samples, and the successful distinctions of B6/By from B6.C-H-2bm1 were confirmed by transfer of training, without reward, to coded samples of urine from genetically equivalent urine donor mice which the trained mice had not previously encountered. Cosegregation of odor phenotype with H-2b and H-2bm1 was demonstrated by transfer of training to typed H-2b and H-2bm1 homozygous segregants of F2 generations of appropriate crosses. Although it is not excluded that the differences in odor phenotype which distinguish H-2b and H-2bm1 mice are directly related to the structure of the H-2b and H-2bm1 products, it is equally possible that H-2-related odor phenotypes arise from effects of H-2 genetic variation on metabolic pathways either directly, or indirectly through developmental polymorphism.

The vomeronasal organs of male guinea pigs were removed (VNX; n = 10) or males experienced sham surgery (Sham; n = 10). Subsequently a battery of chemosensory tests of investigatory responsiveness to conspecific urine was conducted. Additionally, male subjects were paired with female conspecifics for short and long periods and social and sexual behaviors were monitored. VNX males exhibited a depression in urine investigation and this depression became more profound following repeated testing and/or the passage of time. By 6.3 months following surgery, investigatory responsiveness to urine was practically eliminated. Maintenance of responsiveness to urine odors may require reinforcing input through the accessory olfactory system. In contrast to these effects on responsiveness to odors, VNX and Sham males were indistinguishable in their social and sexual behavior. These data indicate that male guinea pigs without a VNO: (1) Exhibit a depression of investigation of urine odors which is time dependent and which may involve an extinction-like process; (2) continue to discriminate classes of urine (e.g., urine from male vs urine from female conspecifics); and (3) exhibit normal sexual behavior. The vomeronasal organ in the male domestic guinea pig is apparently critical for the maintenance of normal responsiveness to sex odors but, in its absence, other sensory systems are capable of maintaining normal sexual behavior under conditions of laboratory testing.

Guinea pigs were allowed to investigate urine that contained rhodamine, a nonvolatile fluorescent dye. Guinea pigs given free access to dyed urine exhibited fluorescence in their vomeronasal and septal organs but not on their olfactory epithelium. Fluorescence was not seen when unadulterated urine was presented. Thus compounds of low volatility, which do not reach the olfactory epithelium, may stimulate the vomeronasal system and provide information that is normally not provided by gustation or olfaction.

Four experiments were conducted to determine the chemosensory modality that supports ultrasonic courtship vocalizations by male mice to females and to chemosignals from females. Both removal of the olfactory bulbs (Experiment 1) and removal of the vomeronasal organ (Experiments 2-4) produced similar deficits in the pattern of ultrasonic vocalizations elicited by conspecifics or their odors. Removal of the vomeronasal organ did not impair the ability to locate food buried under cage shavings. These results are consistent with the notion that the analysis of food related odors is subserved by olfaction and that vocalizations to sex chemosignals are elicited primarily by stimulation of the vomeronasal organ/accessory olfactory bulb. Removal of the vomeronasal organ did not induce seminal vesicle regression or lower plasma immunoreactive testosterone levels (Experiment 2) nor was an attempt to restore vocalizations with exogenous testosterone successful (Experiment 4). Thus the altered vocalization pattern following removal of the vomeronasal organ does not appear to arise as a motivational deficit mediated by androgens. Experiments 2 and 3 demonstrated that, in the absence of the vomeronasal organ, stimulation of other sensory systems can, to some extent, maintain the male's tendency to vocalize more to females or their odors than to males or their odors. However, this responsiveness to females may rely upon additional behavioral cues. Previous experience also plays a considerable role in the response to chemosensory gender cues by males who lack their vomeronasal organs. Removal of the vomeronasal organ prior to adult heterosexual encounters (Experiment 3) virtually eliminated the male's responsiveness to either anesthetized females or their chemosignals. Hence males require adult heterosexual experience with a functioning vomeronasal organ before other chemosensory systems acquire the ability to mediate gender recognition as measured by ultrasonic vocalizations.

Mice can recognize one another by individually characteristic body scents that reflect their genetic constitution at the extremely polymorphic major histocompatibility complex (MHC) of genes on chromosome 17. Reproductive behavioral manifestations of this sensory communication system include MHC-related mating preferences and neuroendocrine responses that affect preimplantation pregnancy and arise from the MHC-related scent of alien males. We have shown previously that mice can be trained in a Y maze to distinguish the scents of urine of congeneic mice that differ genetically only at the MHC. By means of an automated olfactometer, we now show that rats also can similarly distinguish the urinary scents of MHC congeneic mice. Thus, the mode of individual recognition that depends on scents determined by MHC genes can operate across species barriers.

Sensitivity to the odor of 5-androst-16-en-3-one (androstenone), a testosterone metabolite, shows wide variations among unrelated individuals. Analysis of correlations in sensitivity between monozygotic twin pairs, dizygotic twin pairs, and nontwin sib pairs now shows that at least a portion of this variation is genetically determined. However, although data from some mouse studies have suggested a relationship between olfaction and the murine histocompatibility system (H-2), we were unable to demonstrate any role of the human HLA system in explaining the wide individual variations in human sensitivity to androstenone. An additional analysis of HLA antigens among 61 human mating pairs also provided no evidence that HLA phenotypes play a role in human mating preference. These data fail to support a role for the human HLA system in the recognition of an odorant of potential biological significance.

It is known that mice can distinguish the H-2 haplotypes of other mice by scent and that urine is a potent source of the characteristic odors governed by H-2. It now is shown that genetic disparity confined to the Qa:Tla region of chromosome 17, adjacent to H-2, is accompanied by distinctive urinary scents that can be distinguished by trained mice in a Y maze. This was confirmed by the transfer of training procedure in which mice trained to distinguish the urine of B6 mice from the urine of recombinant congeneic B6-Tlaa mice successfully distinguished the urines of Tlaa and Tlab homozygous F2 segregants of the cross B6 X B6-Tlaa in the absence of reward, including blind trials with coded urine samples. It also is shown that genetic disparity confined to the K end of H-2 is accompanied by distinctive urinary scents recognized in the Y maze. Thus, mice trained to distinguish the urines of B10 from B10.A congeneic mice successfully distinguished various combinations of B10, B10.A, and B10.A (2R, 3R, 5R, and 18R) recombinant congeneic mice representing genetic differences limited to the K end of H-2. It is suggested that the individual scents of mice comprise several odorous components, at least some of which may be incidental to quantitative or qualitative metabolic variations arising from polymorphism of genes such as H-2.

A series of experiments revealed that humans can use olfaction to discriminate closely related strains of mice, differing genetically only at the major histocompatibility gene complex (H-2). In Experiment 1, subjects were asked to distinguish between the whole-body odors of live mice. In Experiments 2 and 3, the odor source was mouse fecal pellets, and in Experiments 4 and 5, the odor source was mouse urine.

Following vomeronasal organ removal or sham surgery at 4-7 days of age, male guinea pigs were tested for responsiveness to conspecific chemical uses as infants and again as adults. In the first experiment, vocalizations in response to soiled home cage bedding and male bedding were monitored twice prior to surgery and twice weekly for four weeks. Home cage cues elicited more vocalizations than did male bedding in both groups; however, there was no effect of vomeronasal organ removal. When tested as adults in a second experiment, animals without vomeronasal organs exhibited depressed investigative responsiveness and vocalizations to female genital smears. The data from the first experiment fail to indicate a role for the vomeronasal organ in infantile response to conspecific odor. However, the second experiment demonstrates that adult responses to similar odors are substantially depressed by an absence of the vomeronasal organ.

Some adult humans cannot detect the odor of androstenone (5 alpha-androst-16-en-3-one), a volatile steroid. To test for the presence of genetic variance associated with this trait, adult twins were tested for their ability to smell androstenone and another odorant, pyridine, that is readily perceived by most adults. Ascending concentration, two-sample (odor versus blank) forced choice tests were used to assess sensitivity to these odorants. Intraclass correlations for identical and fraternal twin detection thresholds to pyridine were small and not significantly different. However, intraclass correlations for thresholds to androstenone were significantly different, with the correlation for identical twins being greater than that for the fraternal twins. These data indicate a genetic component of variation in sensitivity to this odor. Investigations that use genetic variation could offer a new tool for studies of olfactory transduction mechanisms.

Nearly half the adult human population does not perceive an odor when sniffing androstenone (5 alpha-androst-16-en-3-one), a volatile steroid found in human perspiration, boar saliva, some pork products (e.g., bacon), truffles, and celery. This variation in ability to perceive androstenone has a significant heritable component, suggesting that androstenone insensitivity is in part determined genetically. We now report that the ability to perceive androstenone was induced in 10 of 20 initially insensitive subjects who were systematically exposed to androstenone. Since olfactory neurons of the olfactory epithelium undergo periodic replacement from differentiating basal cells, and assuming the induction of sensitivity to be peripheral, we propose that a portion of the apparently anosmic human population does in fact possess olfactory neurons with specific receptors for androstenone. Such neurons may undergo clonal expansion, or selection of lineages with more receptors or receptors of higher affinity, in response to androstenone stimulation, much in the manner of lymphocytes responding to antigenic stimulation, thus raising odor stimulation to the level of conscious perception. As a guide to further study of the genetics and mechanism of variation of androstenone perception, we provisionally envisage three categories of human subjects, the truly anosmic, the inducible, and those subjects who either are constitutionally sensitive or have already experienced incidental induction.

Multidimensional scaling was used to analyze odor similarity judgments obtained by monorhinic (single nostril) stimulation from normal subjects (N = 52), equally partitioned by sex and handedness. Neither sex nor handedness nor side of nose appeared to alter the position of stimuli on a two-dimensional map of odor similarity. However, women produced significantly more consistent maps than men. This result was not due to differential utilization of axes in the multidimensional perceptual space, nor to differences in verbal labeling. Left versus right nostril asymmetries were significantly greater in dextrals.

The possibility that male guinea pigs (Cavia porcellus) habituate to individual female conspecific urine odors following very brief, widely-spaced presentations was tested. On day 1 of test week 1, each male was exposed for 2 min to urine of one female, on day 3 to urine of a second female and on day 5 to urine of a third female. The number of seconds investigating the samples was recorded. This same procedure was repeated for 3 more weeks. On weeks 5 through 8, three novel females provided urine samples for testing which was conducted in an identical manner. On week 9, urine samples of the original 3 donors was tested. Results indicated that male guinea pigs habituated to urine odors presented for only 2 min 7 days apart and that dishabituation occurred following presentation of samples from novel donors. These data demonstrate an exquisite sensitivity to female urine odors, a long memory for individual differences in these odors, and a consistency over time in the composition of an individual's urine.

While it has been reported that most, if not all, very young children are able to detect the odor of 5 alpha-androst-16-en-3-one (androstenone), approximately 40-50% of human adults cannot detect its odor. The present study focused on changes in sensitivity to androstenone during adolescence, which may account for this discrepancy. Sensitivity to androstenone was determined in 247 subjects aged 6 to 50. There was a significant increase in the number of males anosmic to androstenone between 9-14 and 15-20 years of age, and a significant increase in threshold with age among males able to detect the odor. We infer that a smaller percentage of females than males becomes anosmic to the odor of androstenone during development, and those able to detect it apparently show a decrease in threshold with age. No age-related changes were observed in tests of pyridine or d,l-beta-phenylethylmethylethylcarbinol (PEMEC).

Previous studies of young children have failed to demonstrate adult-like odor preferences in children less than 5 years old. To test whether these results may have been due to inappropriate methods or to stimulus sets that were too limited in range to capture discriminations, a forced-choice procedure embedded in a simple game was used to contrast 3-year-olds' (n = 16) and adults' (n = 17) hedonic reactions to 9 odorants. Subjects indicated liking an odor by pointing to one puppet and disliking an odor by pointing to another puppet. Analyses revealed essentially the same pattern of preferences in both groups. In general, the odorant was a much better predictor of its hedonic quality than was the age of the subject. However, children and adults did differ in their ratings of some odors, and response patterns indicated that children may be more sensitive than adults to some odorants (e.g., the steroid androstenone). These data indicate that the predominant view that adult-like odor preferences and aversions do not exist until between 5 and 7 years of age must be reevaluated. [ABSTRACT FROM AUTHOR]