Review
Lesions of structures showing FOS expression to cat presentation: Effects on responsivity to a Cat, Cat odor, and nonpredator threat

https://doi.org/10.1016/j.neubiorev.2005.04.019Get rights and content

Abstract

Exposure of rats to a cat elicits Fos activity in a number of brain areas or structures. Based on hodological relationships of these, Canteras has proposed a medial hypothalamic defense system, with input from several forebrain sites. Both electrolytic and neurotoxic lesions of the dorsal premammillary nucleus, which shows the strongest Fos response to cat exposure, produce striking decrements in a number of defensive behaviors to a cat or to cat odor stimuli, but do not have a major effect on either postshock freezing, or responsivity to the odor of a female in estrus. Neurotoxic lesions of the medial amygdala produce decrements in defensiveness to predator stimuli, particularly odor stimuli, that are consistent with a view of this structure as involved with allomonal cues. While dorsal hippocampal lesions had little effect on responsivity to predator stimuli, neurotoxic lesions of the ventral hippocampus reduced freezing and enhanced a variety of nondefensive behaviors to both cat odor and footshock, with similar reductions in defensiveness during context conditioning tests for cat odor, cat exposure and footshock.

These results support the view that the dorsal premammillary nucleus is strongly and selectively involved in control of responsivity to predator stimuli. Structures with important input into the medial hypothalamic defense system appear also to be functionally involved with antipredator defensive behaviors, and these lesion studies may suggest specific hypotheses as to the particular defense functions of different areas.

Introduction

In a study of Fos immunoreactivity in rats displaying defensive responses during cat exposure, Fos expression was enhanced in the anterior hypothalamic nucleus (AHN), dorsomedial part of the ventromedial nucleus (VMHdm), and dorsal premammillary nucleus (PMd) (Canteras et al., 1997). Previous anatomical studies had shown that these cell groups in the medial hypothalamus are interconnected into a distinct subsystem. Putting together anatomical and functional findings, it was possible to bring into focus a distinct medial hypothalamic circuit, conceptualized as the medial hypothalamic zone (MHZ) defensive system. Within this system, the PMd emerged as the most responsive nucleus to the predatory threat, and ibotenic acid lesions in the PMd drastically reduced the escape and freezing seen in rats during exposure to a cat (Canteras et al., 1997).

The MHZ defensive system receives inputs from widely distributed areas in the forebrain, and to a lesser extent, from the brainstem as well. The major telencephalic sources of inputs arise in the lateral septum and amygdala. Projections from the lateral septum to the MHZ defensive system preferentially target the AHN and arise from specific regions of the lateral septal nucleus (Canteras et al., 2001). The septal districts particularly mobilized during cat exposure receive inputs from intermediate regions of field CA1 and subiculum (Canteras et al., 2001), thus forming a specific septohippocampal domain potentially involved in modulating anti-predatory defensive behaviors. In line with this view, previous behavioral studies have shown that hippocampal lesions alter defensive responses to cat exposure, reducing freezing (Blanchard and Blanchard, 1972). In addition, recent findings indicate that subjects exposed to a context where the animals had previously encountered a predator presented a clear increase in Fos expression in the AHN and the PMd, but not in the VMHdm (Ribeiro-Barbosa and Canteras, personal observations). Taken together, the evidence suggests that the septohippocampal–AHN path is likely to be involved in modulating both unconditioned and contextual conditioned responses linked to the predator.

The VMHdm, in turn, is particularly mobilized in the response to the predator, and not to predator contextual clues (Canteras et al., 2001; Staples et al., this issue; Ribeiro-Barbosa and Canteras, personal observations). This view is consonant with the findings that the VMHdm mostly integrates information from amygdalar sites responding to the presence of the predator, including the posteroventral part of the medial nucleus (MEApv) and the posterior basomedial nucleus (BLAp). The MEApv presents a very strong and selective activation following exposure to cat odor, suggesting that this particular region of the ‘vomeronasal amygdala’ is involved in pheromone-like processing of predator odor (Dielenberg et al., 2001). The BLAp, on the other hand, receives massive inputs from the caudal regions of lateral amygdalar nucleus (LA), and this amygdalar path is mobilized during cat exposure, but not to cat odor alone (Canteras, 2002, Dielenberg et al., 2001). The caudal regions of the LA and, to a lesser extent, the BMAp, integrate inputs from visual and auditory association areas (McDonald, 1998), and are, therefore, in a position to respond to a wealth of predator-related sensory information rather than to odor alone.

In the present report, we provide a series of experiments examining how a wide spectrum of defenses to the cat, as well as defensive behaviors to other types of threat stimuli, are affected by selective lesions of discrete elements of these septohippocampal- and amygdalar-hypothalamic paths. The predator stimuli used in these studies were the odor of a cat, obtained by rubbing a domestic cat and using the rubbing cloth as the stimulus; and actual exposure to the same cat. These stimuli were always presented in the same order; odor first, followed by cat exposure on another day. While this may be regarded as susceptible to order effects, the cat is a much more potent threat stimulus than is cat odor alone, and to balance presentation of these stimuli would run a clear risk of a sensitized response to odor in the group receiving the cat first and odor later, potentially contributing to variance and obscuring lesion effects on these responses. Moreover, as will be seen, the data were such that an interpretation of possible sensitization effects did not easily apply.

In addition to the same ordering of cat odor and cat exposure tests, the protocols for these tests were generally similar, with some exceptions that will be noted under individual lesion studies. Briefly (additional details are given in Blanchard et al., 2003, Markham et al., 2004) cat odor tests involved individual placement of animals in a previously habituated alley with sections marked off on the floor. The odor block was placed at one end of the alley. Avoidance of the odor stimulus, freezing, and risk assessment (stretch attend—risk assessment from a distance—and flatback or stretch approach to the stimulus) were the major defensive behaviors measured. Rearing, grooming and locomotion provided additional indices of activity, less clearly related to defense. The cat exposure test, while run in a different apparatus, involved similar measures, plus contact with a screen separating the cat from the subject.

An important focus of the program was to determine the cross-stimulus generality of any deficit in defensiveness to predator stimuli that might be obtained with a particular lesion. Thus lesions in each site were also evaluated in terms of effects on nonpredator threat or danger stimuli, in a footshock test that was run after the predator exposure tests. This involved 3, 1.0 mA, 1-s footshocks, given at 1-min intervals, following a 3-min habituation to the test situation. Freezing was evaluated during a 5-min test period. One of two sets of studies done with the dorsal premammillary nucleus also added a test of olfactory sensitivity, evaluating the responses of lesioned animals to the scent of a female rat. In this test, soiled bedding from a female rat in estrus was placed in one of two Petri dishes while a second dish had an equal quantity of unsoiled bedding. These were presented simultaneously, and the number and duration of contacts with each Petri dish were evaluated to determine if lesioned rats favored the soiled bedding.

For simplicity and ease of comparison freezing will be the focal measure presented in this report. As each test reported was 10 min or 600 s in length, maximum possible duration of freezing was 600-s in each case. A variety of other measures were also taken and significant effects of lesions on these will be briefly presented.

Section snippets

Dorsal premammillary nucleus (PMd)-I

In an initial PMd study (Blanchard et al., 2003) 14 Long-Evans rats received bilateral electrolytic lesions of this structure, while seven sham-lesioned controls had electrodes lowered to 1 mm above the PMd, without passage of current. Histological analysis indicated that seven lesioned animals had acceptable bilateral PMd lesions and all analyses compared these to the sham lesioned group. All rats also had partial damage to the arcuate nucleus and lateral hypothalamus, either unilateral or in

The dorsal premammillary nucleus (PMd)-II

A second study of PMd effects (Markham et al., 2004) involved ibotenic acid lesions in 11 animals and nine sham-operated controls. Animals that showed damage to the ventral premammillary nucleus (PMv) or the mammillary bodies were not included in the analysis, leaving six animals with complete bilateral damage to the PMd. This study repeated the cat odor and cat exposure tests used in the initial PMd study, and added a test of responsivity to the odor of a female rat. The female odor test was

Medial amygdala (MeA)

The MeA receives heavy projections from the vomeronasal organ (VNO), a structure known (Miller and Gutzke, 1999) to play a critical role in the detection of pheromonal or allomonal odors, including those associated with a predator. Further, one of the principle targets of the MeA is the dorsomedial portion of the ventromedial hypothalamic nucleus, which, as previously mentioned, together with the anterior hypothalamus and the dorsal premammillary nucleus, forms the MHZ defensive system (

Dorsal hippocampus (DH)

A study of the effects of ibotenic acid lesions of the dorsal hippocampus (Pentkowski, 2004) utilized 10 lesioned and 15 sham-lesioned rats. All subjects included in the experimental group sustained extensive, bilateral lesions of the DH formation, with minimal damage to surrounding structures. This minimal damage was largely restricted to cannula tract marks and was therefore equivalent to the damage found in the sham-lesioned controls. Nearly all of the remaining cells were found in the most

Ventral hippocampus (VH)

The VH study used 10 rats with verified ibotenic acid lesions of the ventral hippocampus, compared with seven sham-lesioned controls (Pentkowski, 2004). All subjects included in the experimental group suffered complete, bilateral lesions of the VH formation, with minimal damage to surrounding structures. Nearly all the remaining cells were found in the most posterior ventral and dorsal portions of the VH (CA-2 region). Thus, most of the VH was destroyed including virtually all of the granular

PMd studies I and II

The first and second PMd studies, involving electrolytic and neurotoxic lesions, respectively, showed substantial agreement in terms of effects on defensive behaviors to cat odor or to a cat. This was particularly true with reference to the cat exposure test. In the cat odor test, PMd-lesioned rats in both studies showed greater activity and fewer crouch-immobile (freezing) responses. Electrolytic PMd lesions (only) also produced decreased avoidance and a reduction in stretch attend, while

Relationship of these findings to specific predictions of the Canteras defense system

These studies provide findings strongly supporting the Canteras (2002) suggestion that functional mapping studies, in conjunction with analysis of hodological relationships among brain structures, can be used to outline a schema for the neural systems modulating anti-predator defense. Moreover, the specific findings of these studies are generally in agreement with particular predictions from this schema.

The two PMd studies (Blanchard et al., 2003, Markham et al., 2004) confirm a strategic

Cited by (154)

  • Coyote urine, but not 2-phenylethylamine, induces a complete profile of unconditioned anti-predator defensive behaviors.

    2021, Physiology and Behavior
    Citation Excerpt :

    Similarly, the present results are congruent with a previous study that observed increased avoidance, but not risk assessment following PEA exposure in mice [24]. In line with the framework outlined above, at least one lab has observed aversive conditioning following coyote urine exposure in mice [28], which is consistent with behavior observed in rats and mice following cat fur/skin odor exposure [3,29,35,39,41]. Notably, to the best of our knowledge, PEA and the other predator urines (e.g., wolf, lion, bobcat etc.) have not been tested for their ability to support aversive conditioning.

View all citing articles on Scopus
View full text