Lesions of the ventral hippocampus, but not the dorsal hippocampus, impair conditioned fear expression and inhibitory avoidance on the elevated T-maze

Abstract

The hippocampus has been suggested to be involved in spatial (or configural) memory and also in the inhibition of certain response or goal alternatives. An increasing number of anatomical, physiological, and behavioral studies indicate that the hippocampus is functionally heterogeneous along the dorsal–ventral axis. Identification of distinct behavioral roles for the dorsal (DH) and ventral (VH) hippocampus may resolve differences between the various theoretical accounts of hippocampal function. The present study examined the effects of electrolytic lesions restricted to the DH or VH on fear-conditioned freezing, passive avoidance on the elevated T-maze (ETM) test of anxiety, and general activity in male Sprague–Dawley (Charles-River derived) albino rats. We found that rats with lesions of the VH, but not DH showed reduced freezing to both context and tone conditioned stimuli (CS). Rats with VH lesions also showed a reduced latency to emerge from the enclosed arm on trials 2 and 3 of the ETM (indicating reduced anxious behavior), while having no effect on the latency to escape from the open arms on trial 4. There were no differences in activity between the groups. These results indicate that the VH and DH are differentially involved in passive avoidance on the ETM and conditioned freezing to context and tone CS. We suggest that the VH may be specifically involved in modulating goal-oriented, defensive behavior expression through hypothalamic and amygdaloid connections.

Introduction

Theories of hippocampal function have sought a unitary conceptualization for the role of the hippocampus in behavior by incorporating the large body of findings from lesion, stimulation, and recording studies. The hippocampus has been proposed to be involved in the acquisition and consolidation of a representation of the environment based on multimodal stimuli (i.e., spatial, sensory, configural, and temporal stimuli) (O’Keefe & Nadel, 1978; Sutherland & Rudy, 1989; see also Anagnostaras, Gale, & Fanselow, 2001; Hirsh, 1980; Honey & Good, 1993) and also in the inhibition of certain response or goal alternatives (Chan, Morell, Jarrard, & Davidson, 2001; Gray, 1982; Gray & McNaughton, 2000; Kimble, 1968). Recently, there has been renewed interest in the suggestion (e.g., Isaacson, 1974; Nadel, 1968; Stevens & Cowey, 1973) of differential function along the longitudinal axis of the hippocampus (dorsal–ventral in the rat and posterior to anterior in humans).

Moser and Moser (1998a) reviewed the increasing anatomical and behavioral evidence for functional differentiation between the dorsal (DH) and ventral (VH) portions of the hippocampus. Anatomically, the hippocampus includes CA1, CA2, CA3, and dentate gyrus subfields with the pattern of efferent and afferent connectivity changing between DH and VH (Andersen, Bliss, & Skrede, 1971). Petrovich, Canteras, and Swanson (2001) have divided the hippocampus in terms of afferent connectivity into 5 parallel zones with Zone 1 encompassing the dorsal half of CA1 and Zones 2–5 the ventral CA1/subiculum. Hippocampal connections to subcortical structures such as amygdala, hypothalamus, and nucleus accumbens are predominantly restricted to the ventral two-thirds of the hippocampus (Canteras & Swanson, 1992; Ottersen, 1982; Petrovich et al., 2001). These connections are more extensive than those reported by other authors (e.g., Amaral & Witter, 1989; Pitkanen, Pikkarainen, Nurminen, & Ylinen, 2000). In contrast, sensory input from association, entorhinal, and perirhinal cortices terminates primarily in the DH (Amaral & Witter, 1989; Ruth, Collier, & Routtenberg, 1982; Witter & Groenewegen, 1984). Furthermore, the density of monoaminergic and cholinergic terminals is higher in the VH compared to the DH (Gage & Thompson, 1980; Hortnagl, Berger, Sperk, & Pifl, 1991; Verney et al., 1985), and DH and VH also exhibit different physiological responses (e.g., Jung, Wiener, & McNaughton, 1994; Maruki, Izaki, Nomura, & Yamauchi, 2001; Papatheodoropoulos & Kostopoulos, 2000). Identification of distinct functional roles for the DH and VH may help to resolve differences between the various theoretical accounts of hippocampal involvement in behavior.

The clear role of the hippocampus in spatial navigation (e.g., Morris, Garrud, Rawlins, & O’ Keefe, 1982) has been shown to be dependent on the DH but not the VH. Moser, Moser, and Andersen (1993) reported that ablation of the DH, removing between 20 and 37% of the total hippocampal volume resulted in profound impairment in spatial location of a submerged platform in the Morris water maze. In contrast, similar size ablations restricted to the VH produced no deficits in learning compared to control animals. Several subsequent studies have confirmed this dissociation between the DH and VH on the Morris water maze using axon-sparing neurotoxic lesions (Bannerman et al., 1999; Moser & Moser, 1998b; Moser, Moser, Forrest, Andersen, & Morris, 1995; but see also Richmond et al., 1999; Stubley-Weatherly, Harding, & Wright, 1996) and extended this dissociation to rewarded, spatial alternation in a T-maze (e.g., Bannerman et al., 1999; Hock & Bunsey, 1998).

Recent studies using fear conditioning paradigms have begun to characterize a greater role for the VH in conditioned freezing compared to the DH. Lesions of the DH impair acquisition of conditioned freezing to contextual cues (Kim, Rison, & Fanselow, 1993; Maren & Fanselow, 1997; Phillips and LeDoux, 1992, Phillips and LeDoux, 1994; but see Maren, Aharonov, & Fanselow, 1997; Richmond et al., 1999), but not conditioned freezing to a discrete CS (Richmond et al., 1999; but see Maren et al., 1997). Lesions of the VH reduce conditioned freezing to both context and discrete CS under certain conditions (e.g., low foot shock intensity) in which DH lesions do not (Richmond et al., 1999); the authors attributed the VH lesion effect to “enhanced locomotor activity under conditions of mild stress (p. 1202)” rather than to reduced fear. However, in a separate study, these authors found no effect of VH lesions on locomotor activity in novel cages before or after amphetamine treatment (Bannerman et al., 1999).

The role of the DH in expression of conditioned freezing to context appears to be time-limited. That is, permanent DH lesions made shortly after fear conditioning impair expression of conditioned freezing to context cues (Anagnostaras, Maren, & Fanselow, 1999; Kim & Fanselow, 1992; but see Maren et al., 1997), whereas lesions made 50 days or more after fear conditioning have no effect on expression of conditioned freezing to context (Anagnostaras et al., 1999; Maren et al., 1997). To date, no published study has examined the effects of post-training VH lesions on expression of conditioned freezing to contextual or discrete CS.

Recently, it has been suggested that these differential effects of DH and VH lesions indicate a more prevalent role for the VH in aversively motivated behaviors (e.g., avoidance, conditioned freezing, and anxiety) because of its dense anatomical connections with subcortical structures (hypothalamus, amygdala, and nucleus accumbens) involved in defense and emotion (see Bast, Zhang, & Feldon, 2001; Moser & Moser, 1998a; Petrovich et al., 2001). Therefore, studies comparing the effects of DH and VH lesions in rodents on measures of fear or anxious behavior that are sensitive to complete hippocampal lesions will help to further characterize the functional dissociation between the DH and the VH, and help to identify a more specific functional role for the VH in fear and anxiety. One such measure of anxious behavior in rodents that has been found to be sensitive to complete damage of the hippocampus is the elevated T-maze test of anxiety (ETM). The ETM capitalizes on the natural fear of rodents to elevated, open places (Montgomery, 1955) that is expressed as acquisition (in one or two trials) of passive avoidance for the elevated, open arms (see Graeff, Netto, & Zangrossi, 1998; Viana, Tomaz, & Graeff, 1994). Ablation of the entire hippocampus reduces passive avoidance of the open arms in the ETM (Coover, Trivedi, & Heldt, 2002; Trivedi, Heldt, & Coover, 2001) indicating a reduction in anxious behavior.

The central purpose of the present series of experiments was to examine the effects of selective lesions to the DH and VH on (1) the expression of freezing to both context and discrete CS conditioned 1 day (recent) and 42 days (remote) before lesions; (2) anxious behavior in the elevated T-maze test of anxiety; and (3) the possibility that our lesions increased activity levels.