Defensive behaviors and brain regional activation changes in rats confronting a snake

https://doi.org/10.1016/j.bbr.2020.112469Get rights and content

Abstract

In the present study, we examined behavioral and brain regional activation changes of rats). To a nonmammalian predator, a wild rattler snake (Crotalus durissus terrificus). Accordingly, during snake threat, rat subjects showed a striking and highly significant behavioral response of freezing, stretch attend, and, especially, spatial avoidance of this threat. The brain regional activation patterns for these rats were in broad outline similar to those of rats encountering other predator threats, showing Fos activation of sites in the amygdala, hypothalamus, and periaqueductal gray matter. In the amygdala, only the lateral nucleus showed significant activation, although the medial nucleus, highly responsive to olfaction, also showed higher activation. Importantly, the hypothalamus, in particular, was somewhat different, with significant Fos increases in the anterior and central parts of the ventromedial hypothalamic nucleus (VMH), in contrast to patterns of enhanced Fos expression in the dorsomedial VMH to cat predators, and in the ventrolateral VMH to an attacking conspecific. In addition, the juxtodorsalmedial region of the lateral hypothalamus showed enhanced Fos activation, where inputs from the septo-hippocampal system may suggest the potential involvement of hippocampal boundary cells in the very strong spatial avoidance of the snake and the area it occupied. Notably, these two hypothalamic paths appear to merge into the dorsomedial part of the dorsal premammillary nucleus and dorsomedial and lateral parts of the periaqueductal gray, all of which present significant increases in Fos expression and are likely to be critical for the expression of defensive behaviors in responses to the snake threat.

Introduction

Recent attention to defense systems in rodents suggests that these involve a common set of structures, including the amygdala, hypothalamus, and midbrain periaqueductal gray, but with specific differences related to the type of threat stimulus and situation [[1], [2], [3], [4], [5]]. In particular, exposure to a carnivore predator (i.e., a cat), or to an aggressive conspecific, elicits activation in parallel circuits in the medial amygdala (MeA), the ventromedial nucleus of the hypothalamus (VMH), the dorsal premammillary nucleus (PMD) and the periaqueductal nucleus of the midbrain (PAG). In each nucleus, however, the specific areas activated appear to be different: For a cat predator, the posteroventral portion of the MeA (MeApv); the dorsomedial portion of the VMH (dmVMH); the ventrolateral portion of the PMD (PMDvl), and the dorsolateral portion of the PAG (PAGdl) [1] are involved, while the presence of an attacking conspecific involves activation in each of these same nuclei; but in different portions, posterodorsal MeA, ventrolateral VMH, dorsomedial PMD, and dorsomedial and lateral PAG (see Ref. [1]).

Silva et al. [4] utilized pharmacogenetic neural inhibition tools to inhibit neurons in the VMH, for mice confronted by these two types of threat, a rat predator or an aggressive conspecific. Inhibition of cells in the dmVMH sharply reduced defensive behaviors in the presence of a predator, without altering defensiveness to the aggressive conspecific, whereas inhibition of cells in the VMHvl selectively reduced defensiveness to the conspecific. Foot shock, in the same situation in which the predator and aggressive conspecific were presented, elicited little activity in the VMH, and selective activation of these portions of the VMH did not alter defensiveness to the shock [4].

On the other hand, conditioned stimuli previously paired with pain are associated with activation in the central nucleus of the amygdala (CeA) and in the vlPAG [1,4]. Viellard et al. [5] further examined the systems associated with shock-based contextual fear conditioning, utilizing two very different situations; confinement inside the shock chamber itself, or, placement in an apparatus outside the shock chamber, but with access to it. This difference produced a profound alteration in the defensive behaviors seen, with freezing in the shock chamber, and very little freezing but high levels of risk assessment, exploration, and piling of substrate against the shock chamber, in the animals placed outside it. This difference is compatible with a view of risk assessment as a major defensive response to less intense threatening stimuli, particularly when an element of ambiguity is involved [6]. Moreover, when the animals were placed in the apparatus outside the shock chamber, but with access to it, there was a Fos upregulation in the dorsomedial PMD and dorsomedial and lateral PAG, in a way similar to what has been described for animals exposed to aggressive conspecifics.

These findings indicate that different classes of threat stimuli, although involving some or many of the same structures, nonetheless show differences that suggest they involve separate circuits. How may these specific systems be related to potential differences in the behaviors associated with defense? Silva et al. [4] found that a predator (rat), an attacking conspecific, and foot shock produced increased stretch postures and immobility, as well as reduced locomotion, in mice. However, Viellard et al. [5] reported that behaviors to threat were very different in the two conditioned fear contexts (inside or outside with access to, the shock chamber). These findings suggest that an adequate analysis of the behavioral and brain system correlates of defense may require different threat stimuli, and a wider range of test situations and behaviors measured.

The current study involves a snake predator, a larger test chamber with a number of objects in which the rat subjects can shelter, or, on which they can climb. The use of a living snake to elicit defensive behaviors has been validated in a number of laboratory rodents, including rats, mice, and hamsters [[7], [8], [9], [10], [11]]. Although the specific behaviors measured and reported are virtually identical to those used previously, additional spatial data were intended to capture a flight-hiding dimension that has not been possible in many previous studies of neural systems involved in fear/defense related behaviors. As with the other types of threat that have been used, this study includes analysis of activation patterns of septal, amygdalar, hypothalamic and brainstem sites to search a possible neural substrate for the snake-induced defensive behavior.

Section snippets

Animals

The threat stimulus was a wild rattler snake (Crotalis durissus terrificus), a species native to Brazil, weighing approximately 1.6 kg. The snake was maintained in captivity in a snake pit in the animal house of the Ribeirão Preto Medical School of the University of São Paulo (FMRP-USP). It was fed every 24 h with rodents previously killed by CO2 inhalation.

Subjects were 10 adult male Long-Evans laboratory rats (R. norvegicus) weighing 250−350 g. from the animal facility of the Department of

Behavior of the snake stimulus

The Brazilian rattler was very active and made strikes at each of the 5 rats that encountered it. For 3 of these rats (Rs2, Rs3 and Rs5), the strike did make contact, but the contact was extremely brief and no injury to any of these rats was apparent.

Behavior of the rats

Individual behaviors for control and rattler-exposed rats are presented in Table 1.

Controls: The rats confronting the toy (rubber) snake appeared to treat it as an object of some interest but showed little defensiveness toward it. Each of them

Discussion

Several snake species, including Crotalis durissus terrificus, elicit a range of defensive reactions in laboratory rodents (e.g. [14,15,7,11]). These include a stretch-attend or “flat-back-approach” behavior involving orientation to the threat or potential threat along with a stretched-out body; freezing or defensive immobility; and escape and/or avoidance of the threat. With animate threats such as predators or attacking conspecifics, sniffing at the threat when it is nearby is also common.

Acknowledgements and Funding

This research was supported by Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP) Research Grants #2014/05432-9 (NSC), # 2017/11855-8 (NCC) and #2017/12881-1 (DCB) - Conselho Nacional de Desenvolvimento Científico e Tecnológico (awarded to M.V.C.B., N.C.C. and N.S.C) - Pro-Rectory of the University of São Paulo (USP) Research grant (awarded to N.C.C). J. Mendes-Gomes was supported by FAPESP Post-doctoral fellowship #2013/13398-2); R.P. Bindi was supported by FAPESP Doctoral fellowship

References (35)

  • A. Pérez-Gómez et al.

    Innate predator odor aversion driven by parallel olfactory subsystems that converge in the ventromedial hypothalamus

    Curr. Biol.

    (2015)
  • R. Bandler et al.

    Columnar organization in the midbrain periaqueductal gray and the integration of emotional expression

    Prog. Brain Res.

    (1996)
  • A.S. Bittencourt et al.

    Organization of single components of defensive behaviors within distinct columns of periaqueductal gray matter of the rat: role of N-methyl-D-aspartic acid glutamate receptors

    Neuroscience.

    (2004)
  • P. Carrive

    The periaqueductal gray and defensive behavior: functional representation and neuronal organization

    Behav. Brain Res.

    (1993)
  • C.T. Gross et al.

    The many paths to fear

    Nat. Rev. Neurosci.

    (2012)
  • R. de Andrade Rufino et al.

    The rostrodorsal periaqueductal gray influences both innate fear responses and acquisition of fear memory in animals exposed to a live predator

    Brain Struct. Funct.

    (2019)
  • B.A. Silva et al.

    Independent hypothalamic circuits for social and predator fear

    Nat. Neurosci.

    (2013)
  • Cited by (0)

    View full text