Olfactory cues are sufficient to elicit social approach behaviors but not social transmission of food preference in C57BL/6J mice

Abstract

Mouse models for the study of autistic-like behaviors are increasingly needed to test hypotheses about the causes of autism, and to evaluate potential treatments. Both the automated three-chambered social approach test and social transmission of food preference have been proposed as mouse behavioral assays with face validity to diagnostic symptoms of autism, including aberrant reciprocal social interactions and impaired communication. Both assays measure aspects of normal social behavior in the mouse. However, little is known regarding the salient cues present in each assay that elicit normal social approach and communication. To deconstruct the critical components, we focused on delivering discrete social and non-social olfactory and visual cues within the context of each assay. Results indicate that social olfactory cues were sufficient to elicit normal sociability in the three-chambered social approach test. On social transmission of food preference, isolated social olfactory cues were sufficient to induce social investigation, but not sufficient to induce food preference. These findings indicate that olfactory cues are important in mouse social interaction, but that additional sensory cues are necessary in certain situations. The present evidence that both the three-chambered social approach assay and the social transmission of food preference assay require socially relevant cues to elicit normal behavior supports the use of these two assays to investigate autism-related behavioral phenotypes in mice.

Figure 1

Similar sociability when the stranger mouse is the same sex versus the opposite sex. Male C57BL/6J mice were given a choice between an unfamiliar male mouse (Male Stranger) and a non-social novel object, an empty inverted wire pencil cup. A second group of male C57BL/6J mice was given a choice between an unfamiliar female mouse (Female Stranger) and non-social novel object. (A) Amount of time spent in each of the three chambers during the ten minute test. (B) Amount of time spent sniffing the stranger mouse or the novel object. (C) Number of entries into each side chamber during the ten minute test. N = 10 in each group. In all Figure 1–Figure 8, data are expressed as mean + the standard error of the mean. * p < 0.05 vs. Novel Object.

Figure 2

Similar sociability in mice given a choice between an unfamiliar male mouse (Stranger Mouse) and a nestlet soiled with male mouse cage odors (Social Odor Nestlet). (A) Amount of time spent in each of the three chambers during the ten minute test. (B) Amount of time spent sniffing the wire cage containing either the unfamiliar male or social odor nestlet. (C) Number of entries into each outer chamber during the ten minute test. No significant differences were detected between the two side chambers for any of the three behavioral parameters. N = 22.

Figure 3

Higher approach to a nestlet soiled with male cage odors (Social Odor Nestlet) than to an almond-scented nestlet (Almond Nestlet). (A) Amount of time spent in each of the three chambers during the ten minute test. (B) Amount of time spent sniffing the wire cage containing either the soiled nestlet or the almond-scented nestlet. (C) Number of entries into each outer chamber during the ten minute test. * p < 0.05 cage vs. Almond Nestlet. N = 15.

Figure 4

Higher approach to a nestlet soiled with male mouse cage odors (Social Odor Nestlet) than to a clean nestlet (Clean Nestlet). (A) Amount of time spent in each of the three chambers during the ten minute test. (B) Amount of time spent sniffing the wire container with the soiled nestlet versus the wire container with the clean nestlet. (C) Number of entries into each side chamber during the ten minute test. * p < 0.05 vs. Clean Nestlet. N = 17.

Figure 5

Higher approach to a nestlet soiled with male mouse cage odors (Social Odor Nestlet) than to an unfamiliar male mouse enclosed in a solid Plexiglas cylinder that blocked olfactory cues (Mouse in Cylinder). (A) Amount of time spent in each of the three chambers during the ten minute test. (B) Amount of time spent sniffing the wire cage containing the soiled nestlet versus the cylinder containing the mouse. (C) Number of entries into each outer chamber during the ten minute test. * p < 0.05 vs. the Mouse in Cylinder. N = 13.

Figure 6

Similar approach to an unfamiliar male mouse enclosed in a solid Plexiglas cylinder that blocked any olfactory cues (Mouse in Cylinder) and an empty wire cage (Novel Object). (A) Amount of time spent in each of the three chambers during the ten minute test. (B) Amount of time spent sniffing the wire object versus the cylinder containing the mouse. (C) Number of entries into each outer chamber during the ten minute test. No significant differences were detected between the two side chambers for any of the three behavioral parameters. N = 11.

Figure 7

Significant social transmission of food preference in mice exposed to a cagemate that had previously eaten novel flavored powdered chow, but no significant social transmission of food preference in mice exposed to the novel flavored powdered chow associated with a social odor. One group of “observer” mice was exposed to a same-sex stranger “demonstrator” mouse that had previously eaten a flavored powdered chow and had the odor of cinnamon or cocoa flavored food on its breath (Demonstrator Mouse). A second group of mice was exposed to 1 gram of flavored powdered food mixed with 1 gram of bedding containing social cage odors (Powdered Food + Social Odor Bedding). After the 30 minute exposure period, the “observer” mice were immediately given 60 minutes of access to two food jars. One food jar contained powdered chow mixed with the novel flavor, e.g. cinnamon, previously consumed by the demonstrator or mixed with the bedding (Cued). The other jar contained powdered chow mixed with a completely novel flavor (Non-Cued), e.g. cocoa. Half the observers received cinnamon as the initial flavor cue and cocoa as the flavor for the uncued choice; the other half of the observers received cocoa as the cued initial flavor and cinnamon as the flavor for the uncued. (A) Amount of cued and non-cued food eaten. (B) Number of feeding bouts at each food jar during the preference test. (C) Number of investigative sniffs during the social interaction. * p < 0.05 vs. Non-Cued. N = 11 in both groups.

Figure 8

Social transmission of food preference in mice exposed to a demonstrator mouse, but not in mice exposed to only flavored food odor with no associated odor. One group of observer mice was exposed to 1 gram of powdered food mixed with either cocoa or cinnamon for 30 minutes and then immediately given a test to determine their preference for this cued food (Powdered Food + No Associated Odor). The other group of observer mice was exposed to a same-sex stranger mouse who had just consumed cinnamon or cocoa flavored food for 30 minutes (Demonstrator Mouse). Each group was then immediately given a 60 minute choice test to determine their preference for the cued food. (A) Amount of cued and non-cued food eaten. (B) Number of feeding bouts in each food jar during the preference test. (C) Number of investigative sniffs during the social interaction. * p < 0.05 vs. Non-Cued. N = 20 in the Powdered Food + No Associated Odor group and N = 15 in the Demonstrator Mouse group.