Int. J Immunopharmac., Vol. 17, No. 8, pp. 655-661, 1995 Elsevier Science Ltd

Pergamon Copyright © 1995 International Society for Immunopharmacology Printed in Great Britain, All rights reserved

0192-056t /95 $9.50 + .00

0192-0561(95)00052-6

WHAT IS THE ROLE OF THE IMMUNE SYSTEM IN DETERMINING INDIVIDUALLY DISTINCT BODY ODOURS?

RICHARD E. BROWN

Psychology Department, Dalhousie University, Halifax, Nova Scotia, Canada B3H 4J 1

(Received for publication 12 May 1995)

A b s t r a c t - - Genetically inbred mice and rats which are identical except for the genes of the major histocompatibility complex (MHC) produce unique urinary odours which can be discriminated by other animals. Congenic strains differing in both the Class I and Class lI regions of the MHC produce distinct urinary odours. These urine odours can be used for mate selection and parental recognition, and it has been suggested that they provide a unique genetic mechanism for kin recognition. However, the non-MHC genes and the X and Y chromosomes also modulate the urinary odours of rodents, and rearing rats in a bacteria-free environment inhibits the production of unique MHC-related odours. We have found that dietary differences produce a greater effect on individual odours than differences at one MHC locus. These results suggest that the MHC, commensal bacteria, and dietary products interact to produce urinary odours which can be used for individual recognition in rodents. The problem is: what is the role of the immune system in determining individually distinct body odours? A model that suggests possible answers to this question is proposed.

When two animals, such as dogs or rats, first meet they usually go through an initial olfactory investigation before engaging in social interactions (Macdonald & Brown, 1985). During this olfactory investigation, each animal gains information about the other through its body odours, emitted from the urine, feces, scent glands, saliva, etc. These odours contain information about the animal on the dimensions summarized in Table 1 (Brown, 1979). This paper focuses on the role of the immune system and other factors which determine the production of individually unique odours in rodents and the importance of these odours in social behaviour.

I N D I V I D U A L O D O U R S A N D T H E I M M U N E S Y S T E M

Many mammals can be identified by their individual odours or "olfactory fingerprints" (Brown, 1979) and these individual odours have been linked to genetic differences at the major histocompatibility complex (MHC). The original study on the role of the immune system in the production of unique urine odours was done by Yamazaki and his colleagues, who found that males of two strains of mice, which were genetically identical except for the genes of the MHC, preferred to

mate with females of the opposite strain (Yamazaki et al., 1976). The ability to discriminate between the two strains was due to differences in their urine odours and mice could be trained to discriminate between the urine odours of the two MHC congenic strains in a Y maze (Yamaguchi et al., 1981). A summary of these mouse studies is given in Yamazaki et al. (1991).

In 1985, Bruce Roser and Prim Singh found that the classical class I MHC antigens of rats were excreted in the urine, and they hypothesized that these class I MHC antigens could provide the mechanism for the unique urine odours of MHC congenic rats. Using habitua- tion-dishabituation tests, we demonstrated that MHC congenic rats could be discriminated by their urine odours (Singh et al., 1987, 1988). However, the hypo- thesis that the class I antigens themselves are the source of the odour of individuality was disproven as removal of the class I antigens from the urine of rats did not remove the individuality signal (Brown et al., 1987).

The MHC of the mouse and rat have a similar organization, each having three different regions (Brown et al., 1990; Yamazaki et al., 1990a). In a series of experiments, we have shown that different regions of the MHC complex all contribute to the unique individual odours of MHC congenic rats (Brown et al., 1989, 1990) and that the infusion of different MHC antigens alters the urine odour of the rat (Pearse-Pratt et al., 1992).

655

656 R. E. BROWN

Table 1. Information provided about mammals by their body odour

Species Sex

Age class: infant, juvenile, adult Individual identity Colony or family membership Reproductive status: rut or heat Social status Stress: fear or alarm state Maternal status: pregnancy/lactation

From Brown (1979).

MHC-related body odours have also been demonstrated in humans (Ferstl et al. , 1992). The MHC, however, is not the only genetic region which determines individual odours. The sex chromosomes and the background genes (i.e. non-MHC regions) also contribute to the individual odours of mice (Yamazaki e t al. , 1986; Beauchamp et al. , 1990; Schellinck e t al. , 1993).

THE ROLE OF C O M M E N S A L BACTERIA IN THE P R O D U C T I O N OF THE ODOURS OF I N D I V I D U A L I T Y

IN MHC C O N G E N I C RATS

Bacteria play an important role in the production of the odours of the anal glands of foxes, the perineal secretions of guinea-pigs, the inguinal glands of rabbits, the vaginal secretions of hamsters and rhesus monkeys and the axillary odours of humans as well as the odours of a number of other mammalian secretions (Albone, 1984; Brown, 1979). Gastrointestinal (GI) bacteria are essential for the production of the maternal odour in the feces of lactating female rats, mice and gerbils, and antibiotics, such as tetracycline or neomycin, which inhibit GI bacteria, prevent the synthesis of the maternal odours (Leon, 1974, 1975; Moltz & Lee, 1981; Skeen & Thiessen, 1977).

The urine of germ-free rats differs from that of con- ventionally housed rats in a number of volatile compo- nents (Holland et al. , 1983) and some of these volatiles may provide the basis of the individually unique odours. GI bacteria have the ability to degrade amino acids into their metabolites which are then excreted in the urine (Borud et al. , 1973). For example, caffeic acid, a constit- uent of vegetable matter in food, is metabolized to form dihydrocaffeic acid and hydroxypropionic acid by the bacteria of the GI tract, and these metabolites are excre- ted in the urine of conventionally housed, but not germ- free, rats (Peppercorn & Goldman, 1972). Thus, individual differences in the urine odours of rats and mice may be due to differences in the urinary volatiles

produced by the GI microflora acting on dietary amino acids.

Since the classical class I MHC antigens did not determine the urine odours of MHC congenic rats, we tested the hypothesis that commensal bacterial flora may provide the urinary volatiles which are unique to each individual (Howard, 1977). To test the bacterial hypo- thesis, we reared MHC congenic rats in germ-free condi- tions and presented their urine to subjects in a habituation-dishabituation test. The results of this experiment indicated that germ-free rearing suppresses the production of the odour of individuality in rats (Singh et al. , 1990; Roser et al . , 1991). We repeated this experiment using an olfactory discrimination learning paradigm and found that when odours were presented in an olfactometer, it was significantly more difficult for subjects to learn to discriminate between the urine odours of germ-free MHC congenic rats than between the urine odours of conventionally housed MHC congenic rats (Schellinck et al. , 1991). Thus, it would appear that the MHC type of the rats interacts with the commensal bacteria to produce individually distinct urine odours.

The bacterial hypothesis, however, does not appear to account for the MHC-related individual odours in mice (Yamazaki e t al . , 1990b; Schellinck & Brown, 1992). Why the individual urine odours of rats, but not of mice, are eliminated by germ-free rearing has not been determined. Nevertheless, our results with germ- free rats led us to investigate the possible interactions between MHC type, bacterial flora and diet in the production of individual odours.

I N T E R A C T I O N A M O N G MHC TYPE, C O M M E N S A L BACTERIA AND DIET IN THE P R O D U C T I O N OF

INDIVIDUAL ODOURS

Dietary factors influence the GI flora of an animal and a number of the animal's physiological functions (Sacquet, 1981). The maternal odour of rats, which is dependent on GI bacteria, is altered by changes in diet as different foods provide different amino acids for these bacteria to metabolize. Moltz & Lee (1981) suggested that the maternal odour arises as a result of intestinal microorganisms breaking down cholic acid into a number of metabolites which provide the basis of the maternal odour. To account for the finding that conven- tionally housed, but not germ-free, MHC congenic rats are discriminated by their urine odours, we hypothesized that the GI bacteria metabolize dietary products to produce a pool of volatile molecules and that the class I antigens select a cocktail of these volatile metabolites and deliver them to the urine (Roser et al. , 1991; Singh

Body Odours 657

Dietary Exogenous Bacteria Amino Acids

Immune System ~ M H C Anigen Filter,

Tolerant Intolerant

Microflora MicrofFora

J J L Antibody Production

@ e 1 Breakdown of Synthesis dietary amino of new Bacteria

acids into volatiles attached to volatile from dietary Antibodies

metabolites amino acids.

I I l

® Volatile

sex pheromones

of bacteria

I MHC class I and II antigen filter ?

1 1 1 Urinary Volatiles

Fig. 1. Possible interactions between the immune system, diet and commensal bacteria in determining individual urinary odours. The commensal bacteria may produce volatile metabolites by breaking down dietary amino acids or by synthesizing new volatiles from the dietary amino acids. Volatile bacterial sex pheromones may also be produced. These volatiles may bind to MHC antigens or fragments of these antigens and be excreted in the urine. Exogenous bacteria must pass through the immune system and the resident microflora in order to invade the body, where they may activate antibody production. Antibiotics eliminate both foreign and commensal bacterial

populations (see Roser et al., 1991).

et al., 1990). This selection may result in a unique urinary odour because each individual has a unique bacterial population or because all individuals have the same bacterial population and dietary differences provide the bacteria with different proteins to metabo- lize (see Fig. 1).

There is considerable evidence that the GI flora of humans show significant individual differences, are stable over long periods of time, and are altered by major dietary changes (Holdeman et al., 1976). Furthermore, genetic differences in the hosts may influence the types of bacterial flora in the GI tract. For example, the bacterial flora of monozygotic twins are more similar than those of dizygotic twins. Unrelated adults, even if they live in close proximity, have significantly different GI flora (Holdeman et al., 1976; Van de Merwe et al., 1983). Rodents from different litters have different GI bacterial populations which persist for long periods of time (van der Waaij, 1986).

We have shown that altering the diet of genetically

identical mice will alter their urine odour. We collected urine from conventionally housed adult male C57BL/ 6J mice fed on two different diets. When adult male Long-Evans rats were presented with urine from these mice in a habituation-dishabituation test, they could readily discriminate between the urine odours of two mice on different diets, but not between the odours of two mice on the same diet (Schellinck et al., 1992). Thus, the bacterial flora and diet may interact to produce unique urine odours.

If the urine odours of rats depend on their commensal gastrointestinal flora, we should be able to manipulate the flora and alter the urine odours. This could be done in two ways: by injecting exogenous bacteria or by de- pleting the bacterial flora through the use of antibiotics (Fig. 1). It is difficult to introduce new bacteria into rats because the commensal bacteria, as well as the immune system, defend the animal from foreign bacteria (Freter, 1981).

We are now conducting an experiment to investigate whether the depletion of particular classes of GI bacteria using specific antibiotics eliminates the individual odours of rats. Following the partial depletion of commensal gut flora by antibiotics, however, it is possible for the GI tract to be repopulated by different bacteria, leading to a long-lasting change in the popu- lation of GI microorganisms (Rosebury, 1969). In order to alter the GI bacteria, therefore, it may be necessary to eliminate all GI bacteria and then inoculate MHC- congenic rats or mice with different microorganisms. It should also be possible to produce different urine odours in genetically identical germ-free rats by inoculating them with different bacterial cocktails (see Singh et al., 1990).

The intestinal bacteria can influence urine odours by breaking down dietary amino acids into volatile metabolites, synthesizing new volatiles from dietary amino acids, or by the production of volatile sex phero- mones (Fig. 1). E. coli and E. f aeca l e s , for example, produce peptide sex pheromones which can bind to receptors on rat neurophils (Sannomiya et al., 1990) and may be transported to the urine. Thus, the interaction between the diet, GI bacteria and the immune system on the production of individually unique odours merits further investigation (see Brown & Schellinck, 1992).

DEVELOPMENT OF INDIVIDUALLY UNIQUE ODOURS

It is also possible to look at the interaction between the MHC, commensal bacteria and diet in the develop- ment of individual odours (Fig. 2). When the rat is born,

658 R.E. BROWN

Antibodies in Proteins Ingested mother's milk in diet b a c t e r i a

" 0oL0 Genotype

Intact Thymus Gland

Development of T and B cells

1 Tolerance to

Resident Microflora

Recognition of "Foreign Microflora

Antibody production to "Foreign" Microflora

Fig. 2. The interaction of genotype, maternal antibodies, early diet, ingested bacteria, and the developing immune system in determining the establishment of a population of commensal

microflora (based on van der Waaij, 1984, 1986).

it has a particular MHC genotype, but has a bacteria- free gut. Within the first 24 h of life, the gut is invaded by microflora, but the population of gut microflora is unstable until after weaning (Brunel & Gouet, 1981; Foo et al., 1974; Van der Waaij, 1986). The nature of the neonatal gut microflora depends on the infant's genotype, its mother's antibody production and her commensal bacteria, the type of food in the diet, and the proper development of the thymus gland and gut associated lymphatic tissue (Peyer's patches) (van der Waaij, 1986, 1989). The MHC type of an individual may influence the initial microflora which colonize the GI tract. The microflora of the gut are established before weaning and these resident microflora then become tolerated by the immune system (i.e. treated as self). The immune system does not produce antibodies to the body's own commensal microflora, as it does to viral, bacterial and protozoon infections.

The development of the commensai microflora in the gut also depends on the diet (Smith, 1965). Mal- nourished infants may fail to develop commensal micro- flora and thus be at risk from infection by pathogens, since they lack the normal microflora which defend them against these pathogens (van der Waaij, 1984). The presence of these early microflora plays an import- ant role in the development of the immune system (Berg & Savage, 1973; Van der Waaij, 1991). Thus, siblings may have similar gut microflora because they have genetically related immune systems, they acquire their

initial microflora from the same mother, and they eat the same food. An important question to investigate is how the immune system, commensal bacteria and early diet interact in the development of the unique individual body odour.

INDIVIDUAL ODOURS AS SIGNALS OF HEALTH STATUS

Several diseases in humans are associated with a characteristic body odour, as shown in Table 2 (Liddell, 1976), and there is anecdotal evidence that doctors with well-trained noses can detect diseases by smelling their patients. One such anecdote states that "Ben Franklin said he could diagnose certain disorders by the smell of the patient's urine" (Bedichek, 1960, p. 60), but the particular disorders so diagnosed are not mentioned.

Hamilton & Zuk (1982) proposed the hypothesis that secondary sexual characteristics and elaborate courtship displays are performed to advertise the health status of the displaying animal to potential mates (see Zuk, 1992). Support for this hypothesis has been provided through a number of studies on the effects of parasites on the effectiveness of courtship displays in birds and fish. For example, red jungle fowl which are injected with gut parasites have less colourful red feathers, shorter comb length and lower testes weights than healthy birds, and females are less attracted to these infected males (Zuk et al., 1990). Similarly, male sage grouse which are infected with avian malaria or lice are less active in their leks and have significantly lower reproductive success than healthy males. Males given antibiotics to reduce parasite infections are chosen more often by females than untreated males (Boyce, 1990). Likewise, male stickleback fish which carry parasites which reduce their red colouration are less attractive to females than healthy males, thus they have lower reproductive success (Milinski & Bakker, 1990),

Since mammals use urine and other scent gland secretions as sex attractants, it is possible that the odours of these secretions provide information on the health of the animal. It has been proposed that "the urine acts primarily as a carrier of information concerning, not rank as such, but the male's physical condition as indicated by his own metabolic by-products excreted via the urine" (see Coblentz, 1976, p. 551). Hamilton & Zuk (1982) suggest that animals attempting to choose disease-free mates should, like physicians, examine the urine and faecal samples of potential mates.

The MHC-related odour of the urine may, therefore, be a reflection of the health status of the individual,

Body Odours

Table 2. The odours of disease

659

Disease Characteristic body odour

Diabetic ketosis Gout Typhoid Diphtheria Smallpox Scurvy Scrofula Yellow fever Schizophrenia Phenylketonuria Defective metabolism of

amino acids (valine, leucine and isoleucine)

Inability to metabolize methionine

Hyperaminoaciduria Inability to metabolize

Breath and sweat has the fruity aroma of decomposing apples Characteristic sweat odour Body smell of freshly baked brown bread Sweetish odour "The stench of smallpox is well known'" Sweat has a putrid odour A smell like that of stale beer An odour which resembles that of a butcher shop Pungent body odour due to increased amounts of trans-3-methyl-2-hexanoic acid in the sweat A musty odour, resembling stale, sweaty locker room towels Smell of maple syrup or caramelized sugar in sweat and urine (Maple syrup urine disease)

Odour of boiled cabbage in breath, sweat and urine

Smell of dried malt or hops (Oast house syndrome) The odour of the sweaty feet syndrome

butyric and hexanoic acids

From Liddell (1976).

and if Hamilton & Zuk's (1982) hypothesis is correct, the urine of diseased animals should be less attractive to potential mates than the odours of healthy individuals. Some recent evidence supports this hypothesis. For example, female mice are able to distinguish between the odours of healthy males and males which have been infected with a protozoan parasite, and find the odours of the parasitized males aversive (Kavaliers & Colwell, 1992). If diseases influence urine odours and thus reduce an animal's ability to attract mates, the role of the MHC may be to inhibit infections rapidly and maintain a "healthy" urine odour. Thus, the genetic diversity of the MHC may be the result of evolution in response both to disease and to reproductive mechanisms such as mate selection (see Potts & Wakeland, 1990). Since urine odours act as sex attractants in many rodents (Brown, 1985), and the odours of sick animals are aversive, the sick animals may be less likely to mate than healthy animals and sexual selection may operate to produce offspring with the most viable MHC types (Potts et al., 1991). Thus, in addition to the information shown in Table 1, body odours may provide information on the health status of potential mates.

SUMMARY

The body odours of mammals provide information about their individual identity as well as their species, sex, dominance status and sexual readiness. Individually unique body odours are determined by differences at the MHC and other genes, by the commensal bacteria and by the diet of an animal. Hypothetical mechanisms are suggested for the interaction of the immune system, commensal bacteria and early diet in the development of the individual odour of rodents, and the changes in body odour which occur in sick and infected animals are discussed. Diseases bourne by exogenous bacteria may also alter the body odour of mammals and reduce their attractiveness to potential mates. Thus the role of the immune system may be to ensure a "healthy" body odour by eliminating parasites, bacteria and viruses which would alter the animal 's olfactory signature. Body odour may thus be used for sexual selection, with animals choosing the most healthy mates.

Acknowledgement - - This research was supported by NSERC of Canada grant A7441.

REFERENCES

Albone, E. S. (1984). Mammalian Semiochemistry. Wiley, New York. Beauchamp, G. K., Yamazaki, K., Duncan, H., Bard, J. & Boyse, E. A. (1990). Genetic determination of individual mouse odour.

In Chemical Signals in Vertebrates 5 (eds MacDonald, D. W. et al.). Oxford University Press, Oxford. Bedichek, R. (1960). The Sense o f Smell. Michael Joseph, London.

660 R.E. BROWN

Berg. R. D. & Savage, D. C. (1972). Immunological responses and microorganisms indigenous to the gastrointestinal tract. Am. J. clin. Nutr., 25, 1364.

Borud, O., Midtvedt, T. & Gjessing, L. R. (1973). Phenolic metabolites in urine and faeces from rats given radioactive ~4C-L- DOPA. Acta Pharm. Toxic., 33,308.

Boyce, M. S. (1990). The red queen visits sage grouse leks. Am. Zool., 30, 263. Brown, R. E. (1979). Mammalian social odours, Adv. Study Behav., 10, 103. Brown, R. E. (1985). The rodents lI: suborder myomorpha. In Social Odours in Mammals (eds Brown, R. E. and Macdonald, D.

W.). Clarendon Press, Oxford. Brown, R. E., Roser, B. & Singh, P. B. (1989). Class I and class I1 regions of the major histocompatibility complex both contribute

to individual odors in congenic inbred strains of rats. Behav. Genet., 19, 659. Brown, R. E., Roser, B. & Singh, P. B. (1990). The M HC and individual odours in rats. In Chemical Signals in Vertebrates 5 (eds

Macdonald, D. W. et al.). Oxford University Press, Oxford. Brown, R. E. & Schellinck, H. M. (1992). Interactions among the MHC, diet, and bacteria in the production of social odors in

rodents. In Chemical Signals in Vertebrates 6 (eds Doty, R. L. and Mi~ller-Schwarze, D.). Plenum Press, New York. Brown, R. E., Singh, P. B. & Roser, B. (1987). The major histocompatibility complex and the chemosensory recognition of

individuality in rats. Physiol. Behav., 40, 65. Brunel, A. & Gouet, Ph. (1981). Kinetic bacterial implantation in the intestine of newborn rats. In Recent Advances in Germfree

Research (eds Sasaki, S., Ozawa, A. and Hashimoto, K.). Tokai University Press, Kanegawa. Coblentz, B. E. (1976). Functions of scent-urination in ungulates with special reference to feral goats (Capra hircus, L.). Am. Nat.,

110, 549. Ferstl, R., Eggert, F., Westphal, E., Zavazava, N. & Muller-Ruchholtz, W. (1992). MHC-related odors in humans. In Chemical

Signals in Vertebrates 6 (eds Doty, R. L. and Miiller-Schwarze, D.). Plenum Press, New York. Freter, R. (1981). Mechanisms by which the intestinal flora controls invading microorganisms. In Recent Advances in Germfree

Research (eds Sasaki, S., Ozawa, A. and Hashimoto, K.). Tokai University Press, Kanegawa. Foo, M. C., Lee, A. & Cooper, G. N. (1974). Natural antibodies and the intestinal flora of rodents. Austr. ,1. exp. Biol. Med. Sci., 52,

321. Hamilton, W. D. & Zuk, M. (1982). Heritable true fitness and bright birds: a role for parasites? Science, 218, 384. Holdeman, L. W., Good, I. J. & Moore, W. E. C. (1976). Human fecal flora: variation in bacterial composition within individuals

and a possible effect of emotional stress. AppL environ. Microbiol., 31,359. Holland, M., Rhodes, G., DalleAve, M., Wiesler, D. & Novotny, M. (1983). Urinary profiles of volatile and acid metabolites in

germfree and conventional rats. Life Sci., 32, 787. Howard, J. C. (1977). H-2 and mating preferences. Nature, 266, 406. Kavaliers, M. & Colwell, D. D. (1992). Exposure to the scent of male mice infected with the protozoan parasite, Eimeria vermiformis,

induces opioid- and nonopioid-mediated analgesia in female mice. Physiol. Behav., 52, 373. Leon, M. (1974). Maternal pheromone. Physiol. Behav., 13,441. Leon, M. (1975). Dietary control of maternal pheromone in the lactating rat, Physiol. Behav., 14, 311. Liddell, K. (1976). Smell as a diagnostic marker. Postgrad. meal J., 52, 136. Macdonald, D. W. & Brown, R. E. (1985). Introduction: The pheromone concept in mammalian chemical communication. In

Social Odours in Mammals (eds Brown, R. E. and Macdonald, D. W.). Clarendon Press, Oxford. Milinski, M. & Bakker, T. C. M. (1990). Female sticklebacks use male coloration in mate choice and hence avoid parasitized

males. Nature, 344, 330. Moltz, H. & Lee, T. M. (1981). The maternal pheromone of the rat: identity and functional significance. Physiol. Behav., 26, 301. Pearse-Pratt, R., Schellinck, H., Brown, R. & Roser, B. (1992). Evolutionary and immunological implications of the role of the

MHC in olfactory signalling. In ChemicalSignals in Vertebrates 6 (eds Doty, R. L. and Mfiller-Schwarze, D.). Plenum Press, New York.

Peppercorn, M. A. & Goldman, P. (1972). Caffeic acid metabolism by gnotobiotic rats and their intestinal bacteria. Proc. natn. Acad. Sci. U.S.A., 69, 1413.

Ports, W. K., Manning, C. J. & Wakeland, E. K. ( 1991). Mating patterns in semi-natural populations of mice influenced by MHC genotype. Nature, 352, 619.

Potts, W. K. & Wakeland, E. K. (1990). Evolution of diversity at the major histocompatibility complex. Trends Ecol. Evol., 5, 18 I. Rosebury, T. (1969). Life on Man. Martin Secker and Warburg, London. Roser, B., Brown, R. E. & Singh, P. B. ( 1991). Excretion of transplantation antigens as signals of genetic individuality. In Chemical

Senses (eds Wysocki, C. J. and Kare, M. R.), Vol. 3. Marcel Dekker, New York. Sacquet, E. ( 1981 ). Role of the diet in gnotobiotic research. In Recent Advances in Germfree Research (eds Sasaki, S., Ozawa, A.

and Hashimoto, K.). Tokai University Press, Kanegawa. Sannomiya, P., Craig, R. A., Clewell, D. B., Suzuki, A., Fujino, M., Till, G. O. & Marasco, W. A. (1990). Characterization of a

class of nonformylated Enterococcus faecalis-derived neutrophil chemotactic peptides: the sex pheromones. Proc. natn. Acad. Sci. U.S.A., 87, 66.

Schellinck, H. M. & Brown, R. E. (1992). Why does germfree rearing eliminate the odors of individuality in rats but not in mice?

Body Odours 661

In Chemical Signals in Vertebrates 6 (eds Doty, R. L. and Mfiller-Schwarze, D.). Plenum Press, New York. Schellinck, H. M., Brown, R. E. & Slotnick, B. M. ( 1991 ). Training rats to discriminate between the odors of individual conspecifics.

Anita. Learn. Behav., 19,223. Schellinck, H. M., Monahan, E., Brown, R. E. & Maxson, S. C. (1993). A comparison of the contribution of the major

histocompatibility complex (MHC) and Y chromosomes to the discriminability of individual urine odors of mice by Long- Evans rats. Behav. Genet., 23,257.

Schellinck, H. M., West, A. M. & Brown, R. E. (1992). The ability of rats to discriminate between the urine odors of genetically identical mice maintained on different diets. Physiol. Behav., 51, 1079.

Singh, P. B., Brown, R. E. & Roser, B. (1987). MHC antigens in urine as olfactory recognition cues. Nature, 327, 161. Singh, P. B., Brown, R. E. & Roser, B. (1988). Class I transplantation antigens in solution in body fluids and in the urine: individuality

signals to the environment. J. exp. Med., 168, 195. Singh, P. B., Herbert, J., Roser, B., Amott, L., Tucker, D. K. & Brown, R. E. (1990). Rearing rats in a germ-free environment

eliminates their odors of individuality. J. chem. Ecol., 16, 1667. Skeen, J. T. & Thiessen, D. D. (1977). Scent of gerbil cuisine. Physiol. Behav., 19, 11. Smith, H. W. (1965). The development of the flora of the alimentary tract in young animals. J. Path. Bact., 90, 495. Van de Merwe, J. P., Stegeman, J. H. & Hazenberg, M. P. (1983). The resident faecal flora is determined by genetic characteristics

of the host. Implications for Crohn's disease? Antonie van Leeuwenhock, 49, 119. Van der Waaij, D. (1984). The immunoregulation of the intestinal flora: experimental investigations on the development and the

composition of the microflora in normal and thymusless mice. Microecol. Ther., 14, 63. Van der Waaij, D. (1986). The apparent role of the mucous membrane and the gut-associated lymphoid tissue in the selection of

the normal resident flora of the digestive tract. Clin. lmmun. Newslett., 7, 4-7. Van der Waaij, D. (1989). Bior6gulation de la microflore du tractus digestif. Rev. sci. tech. Off. int. Epiz., 8, 347. Van der Waaij, D. (1991). The microflora of the gut: recent findings and implications. Dig. Dis., 9, 36. Yamaguchi, M., Yamazaki, K., Beauchamp, G. K., Bard, J., Thomas, L. & Boyse, E. A. ( 1981 ). Distinctive urinary odors governed

by the major histocompatibility locus of the mouse. Proc. hath. Acad. Sci. U.S.A., 78, 5817. Yamazaki, K., Beauchamp, G. K., Bard, J. & Boyse, E. A. (1990a). Single MHC gene mutations alter urine odour constitution in

mice. In Chemical Signals in Vertebrates 5 (eds MacDonald, D. W., Miiller-Schwarze, D. and Natynczuk, S. E.). Oxford University Press, Oxford.

Yamazaki, K., Beauchamp, G. K., Bard, J,, Boyse, E. A. & Thomas, L. (1991). Chemosensory identity and immune function in mice. In Chemical Senses (eds Wysocki, C. J. and Kate, M. R.), Vol. 3. Marcel Dekker, New York.

Yamazaki, K., Beauchamp, G. K., lmai, Y., Bard, J., Phelan, S. P., Thomas, L. & Boyse, E. A. (1990b). Odortypes determined by the major histocompatibility complex in germfree mice. Proc. hath. Acad. Sci. U.S.A., 87, 8413.

Yamazaki, K., Beauchamp, G. K., Matsuzaki, O., Bard, J., Thomas, L. & Boyse, E. A. (1986). Participation of the murine X and Y chromosomes in genetically determined chemosensory identity. Proc. natn. Acad. Sci. U.S.A., 83, 4438.

Yamazaki, K., Boyse, E. A., Mike, V., Thaler, H. T., Mathieson, B. J., Abbott, J., Boyse, E., Zoyas, Z. A. & Thomas, L. (1976). Control of mating preferences in mice by genes in the major histocompatibility complex. J. exp. Med., 144, 1324.

Zuk, M. (1992). The role of parasites in sexual selection: current evidence and future directions. Adv. Study Behav., 21, 39. Zuk, M., Johnson, K., Thornhill, R. & Ligon, J. D. (1990). Parasites and male ornaments in free-ranging and captive red jungle

fowl. Behaviour, 114, 332.