A simple and reliable test of olfactory learning and memory in mice.

The present paper describes a quick and efficient method for assessing olfactory discrimination learning in mice. In training mice received trials in which one odor (CS+) was paired with sugar and another odor (CS-) was paired with no sugar. When the mice were subsequently placed in a chamber with CS+ odor at one end and CS- odor at the other, they spent more time digging in CS+ than in CS- odor. In Experiment 2 mice trained with this procedure and tested after 60 days also spent more time digging in CS+ than CS- in the test phase, indicating that this olfactory discrimination task is effective for assessing long-term memory. In addition to the outbred strain of CD1 mice used in Experiments 1 and 2, C57Bl/6NCr/BR and DBA/2NCr/BR mice used in Experiment 3 also acquired this learned odor discrimination. Moreover, Experiment 4 showed that DBA animals were capable of acquiring this odor discrimination after receiving only two training trials (one exposure each to CS+ and CS-) per day for 4 days.

Effect of food restriction on acquisition and expression of a conditioned odor discrimination in mice.

Level of food restriction was manipulated in mice to assess its importance for the acquisition and expression of a conditioned odor discrimination. In training, animals were exposed to odors (either rose or lemon) presented on a piece of filter paper in a pot covered in bedding. For half of the conditioning trials, group paired received one odor (CS+) with sucrose, the unconditioned stimulus (us), under the bedding. For the remaining trials, they received the other odor (CS-) alone. Group CS-alone was also exposed to both odors, but neither odor was paired with sugar on any of the conditioning trials. During training, Group Paired mice that were food-restricted tended to dig more readily and longer in the odors, especially in the CS+ odor, than animals that were not restricted. Both restricted and nonrestricted PAIRED GROUPS dug more in the CS+ than in the CS- by the end of training, but the CS-alone mice dug very little in either. Following training, mice were exposed to both odors simultaneously in a discrimination test. Half the mice in each training food restriction condition were tested under food restriction, and half were not. Only PAIRED animals that were food-restricted in the test expressed an odor discrimination, digging only in the CS+. This occurred regardless of their previous restriction state in training. These data suggest that both food-restricted and nonrestricted mice can acquire an odor discrimination; however, expression of this odor discrimination depends on food restriction.

Selective depletion of bacteria alters but does not eliminate odors of individuality in Rattus norvegicus.

To determine if odors of individuality are influenced by the removal of Gram-negative or Gram-positive gut bacteria, Long-Evans rats were trained in an operant olfactometer to discriminate between the odors of two individual conspecifics and their operant responses to three different odors in randomly presented probe trials were analyzed. Significantly more responses were made to the probe odors from two known individuals than to the probe odors from known individuals with their Gram-negative bacteria eliminated (Experiment 1) or their Gram-positive bacteria eliminated (Experiment 2). Responses to the probe odors from known rats with bacterial selectively depleted did not differ significantly from responses to probe odors from unknown rats. These results support the hypothesis that the urinary odor of an individual rat is altered by the removal of specific gut bacteria. In Experiments 3 and 4, subjects made fewer errors in learning to discriminate between the odors of the familiar rats whose bacteria had been selectively depleted than between the odors of unknown rats. This "savings effect" indicates that some components of the individual urinary odors were retained after the removal of specific gut bacteria. Thus, the eliminated bacteria were not totally responsible for the odors of individuality. The outcome of Experiments 3 and 4 also indicates that conclusions regarding the recognition of odors by rats should not be made on the outcome of probe trial experiments alone.

Patterns of expression of the immediate-early gene egr-1 in the accessory olfactory bulb of female mice exposed to pheromonal constituents of male urine.

Male mice excrete large quantities of major urinary proteins that have been proposed to have an important pheromonal role either alone or by way of their bound ligands. We have found that these major urinary proteins are not only likely to mediate the pregnancy blocking effects of male urine, but that they also convey the strain recognition signal of the male pheromone. Recent molecular biological investigations have characterized two classes of pheromonal receptor in the vomeronasal organ that appear to project separately to anterior and posterior regions of the accessory olfactory bulb. However, it is not known whether these separate pathways handle fundamentally different types of pheromonal information. We have attempted to investigate this question using the expression of the immediate-early gene egr-1 as a marker for activity of neurons in the accessory olfactory bulb of female mice in response to putative pheromonal constituents. Exposure to 2,3 dihydro-exo-brevicomin and 2-sec-butyl-4,5-dihydro-thiazole, the main ligands bound to the major urinary proteins, elicited expression of egr-1 in clusters of presumed mitral neurons at the medial and lateral margins of the posterior accessory olfactory bulb. Whole male urine and a preparation of major urinary proteins that had been stripped of their ligands induced egr-1 expression in mitral cells of the anterior half of the accessory olfactory bulb in addition to the posterior clusters. This would suggest that the anterior and posterior halves of the accessory olfactory bulb are processing different aspects of the male pheromone signal with the anterior region, which responds preferentially to major urinary proteins, being principally concerned with the strain recognition component.

Changes in neurotransmitter release in the main olfactory bulb following an olfactory conditioning procedure in mice.

Olfactory learning is associated with substantial neural changes at the level of the accessory and main olfactory bulb, during both pheromonal learning in mated mice and lamb odour recognition in post partum sheep. These forms of learning occur during "sensitive periods" and an important question is whether similar neural changes occur in the olfactory bulb at other times. We used a classical conditioning procedure to establish an olfactory discrimination in adult mice and then measured changes in neurotransmitter levels in the main olfactory bulb in response to the presentation of the conditioned odours. Presentation of the conditioned, but not the non-conditioned, odour resulted in significant increases in the levels of certain transmitters, including glutamate from the mitral/tufted cells, GABA from the granule and periglomerular cells and noradrenaline from the centrifugal projection from the locus coeruleus. Overall, there was a decrease in the ratio of excitatory to inhibitory neurotransmitters in the olfactory bulb in response to the conditioned, but not the non-conditioned odour. Moreover, the magnitude of the decrease in this ratio was correlated with the level of behavioural response to the conditioned odour. These findings support the hypothesis that changes in the gain of the reciprocal synapses between mitral/tufted neurons and their inhibitory interneurons are a general feature of olfactory learning.

Behavioural studies of MHC-congenic mice.

Behavioural studies of MHC-congenic mice and rats have focused primarily on mate choice and the ability to discriminate between strains by their urine odours, but these strains may differ in other behaviours, such as activity and ultrasonic vocalizations. Ivanyi (1978, Proc. Roy. Soc. Lord. 202, 117-158) has reviewed the physiological differences associated with the MHC, many of which could influence behaviour. We have started a systematic study of behavioural development and adult behaviour in MHC-congenic mice. A developmental test battery (growth, rate, locomotion, grooming, eye opening, ultrasonic vocalizations, etc.) was used to examine differences between C57BL/6J vs. B6-H-2bml and C57BL/10SnJ vs. B10.BR/sgSnJ mice. A test battery of spontaneous behaviours (activity, exploration, ultrasonic vocalizations, etc.) was used to examine behavioural differences between adult C57BL/6J vs. B6-H-2bml; and C57BL/10SnJ vs. B10.BR/sgSnJ mice. Differences in development and in adult behaviours between these MHC-congenic strains is discussed in relation to possible neural, endocrine and immune system differences. Future studies will compare MHC-congenic mice on levels of anxiety, sociosexual behaviour and on learning paradigms.

The influence of dietary and genetic cues on the ability of rats to discriminate between the urinary odors of MHC-congenic mice.

This study directly compared the role of genetic and dietary factors in the production of individual urinary odors in mice by investigating the ability of male Long-Evans hooded rats to discriminate between urine samples from two strains of mice (C57BL/6-H-2Kb/J and C57BL/6-H-2Kbml/ByJ), which differ at only one locus of the major histocompatibility complex (MHC) and which were maintained on two different diets. Groups of rats were trained in an olfactometer on a go/no-go operant task with water as reinforcement to discriminate between one of the following four pairs of mouse urinary odors: individual mice differing at the MHC and maintained on the same diet (task 1), individual mice differing at the MHC and maintained on different diets (task 2), individual mice of the same MHC type maintained on the same diet (task 3), or individual mice of the same MHC type maintained on different diets (task 4). The urinary odors of mice on different diets (task 2 and 4) were more readily discriminable than those of mice maintained on the same diet (tasks 1 and 3), irrespective of genetic differences at the MHC. It was more difficult to discriminate between urinary odors of mice on the same diet whether they were genetically identical (task 4) or differed at the MHC (task 1). A second experiment revealed that it was no less difficult for rats to discriminate between the urinary odors of mice that differed at three MHC loci than it was for rats to discriminate between two mice that differed at one MHC locus. Thus, the results from Experiment 1 were not caused by the degree of genetic difference at the MHC. Overall, these results provide further evidence that dietary as well as genetic cues contribute to individual odors in rodents, and demonstrate the ability of rats to make very subtle discriminations between odors when trained in an olfactometer.

Odors of individuality of germfree mice are not discriminated by rats in a habituation-dishabituation procedure.

This study investigates whether differences in urinary odors of two major histocompatibility complex (MHC) congenic strains of mice reared in germfree conditions could be detected by male Sprague-Dawley rats in a habituation-dishabituation procedure. Rats were tested for their ability to discriminate between urine odors from one of the following pairs: (a) MHC congenic mice (B6 and B6-H-2k) housed in conventional conditions; (b) MHC congenic mice (B6 and B6-H-2k) raised in a germfree environment; (c) genetically identical mice maintained in conventional housing; or (d) genetically identical mice raised in a germfree environment. The urine odors of mice from the two MHC congenic strains raised in conventional housing were discriminable, but the urine odors of MHC congenic mice raised in a germfree environment were not discriminable. The odors of genetically identical mice, whether housed conventionally or under germfree conditions, were not discriminable. These results indicate that germfree rearing influences the urinary odors of individuality in MHC congenic mice. We discuss the possibility that the disparate findings from previous studies are a result of the methodologies employed.

Odor-induced sexual maturation and expression of c-fos in the olfactory system of juvenile female mice.

Exposure to the urine or soiled bedding odors of novel adult males is known to accelerate puberty in juvenile female mice. To determine what part of the olfactory system is activated by these odors, the expression of c-fos in the main olfactory bulb (MOB) and the accessory olfactory bulb (AOB) of juvenile female mice was examined after their exposure to male-soiled bedding, peppermint odor or their own bedding. Fos-like immunoreactivity was found throughout the AOB of the juvenile female mice exposed to the bedding odors of adult males for 3 h. In contrast, dense staining was found in the granular cell layer of the MOB of mice exposed to peppermint odors for 3 h, whereas mice exposed to their own bedding failed to show immunostaining in the AOB and only slight or no staining in the MOB. These results indicate that social odors stimulate the expression of c-fos in the AOB while non-social odors activate the MOB. This method allows the identification of individual cells activated by the different odors and will be useful in locating other areas of the brain involved in the neuroendocrine changes underlying odor-induced precocious puberty.

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.

A go/no-go operant task was used to assess the ability of male Long-Evans rats to discriminate between the urine odors from pairs of intact MHC congenic mice (C57BL/6-H-2Kb/J and C57BL/6-H-2Kbm1/ByJ), intact Y congenic mice (DBA1 and DBA1.C57BL10-Y), and castrated Y congenic mice of these two strains. The MHC congenic strains differ in alleles of the H-2 K locus, while the Y congenic strains differ in the nonrecombining part of the Y chromosome. Analysis of the number of correct responses to a criterion of 85% correct on each block of 20 trials revealed that the ability of the subjects to discriminate between urine odors did not differ whether samples were from pairs of intact MHC congenic mice, intact Y congenic mice, or castrated Y congenic mice. These findings are consistent with the hypothesis that individually unique urine odors may be influenced both by genes in the nonrecombining part of the Y chromosome and by genes in the major histocompatibility complex of chromosome 17. These odors are not androgen dependent. Such urinary chemical signals may be involved in pregnancy block (the Bruce effect), aggression, and other mouse social behaviors.

Rats can discriminate between the urine odors of genetically identical mice maintained on different diets.

Male Long-Evans hooded rats were tested in a habituation-dishabituation procedure for their ability to discriminate between the urine odors of male C57BL/6J mice maintained on two different diets. There were discriminable differences between the urinary odors of two individual mice maintained on different diets. The rats did not dishabituate when presented with urine odors from two individuals on the same diet or two odor samples from the same individual. These results indicate that individual urinary odors of genetically identical mice are influenced by diet. We discuss the hypothesis that diet may act together with genetic differences, commensal bacteria, and hormonal changes to convey olfactory information used for individual recognition.