Restoration of hippocampal growth hormone reverses stress-induced hippocampal impairment

Abstract

Though growth hormone (GH) is synthesized by hippocampal neurons, where its expression is influenced by stress exposure, its function is poorly characterized. Here, we show that a regimen of chronic stress that impairs hippocampal function in rats also leads to a profound decrease in hippocampal GH levels. Restoration of hippocampal GH in the dorsal hippocampus via viral-mediated gene transfer completely reversed stress-related impairment of two hippocampus-dependent behavioral tasks, auditory trace fear conditioning, and contextual fear conditioning, without affecting hippocampal function in unstressed control rats. GH overexpression reversed stress-induced decrements in both fear acquisition and long-term fear memory. These results suggest that loss of hippocampal GH contributes to hippocampal dysfunction following prolonged stress and demonstrate that restoring hippocampal GH levels following stress can promote stress resilience.

Keywords: conditioning; fear; gene therapy; growth hormone; hippocampus; stress.

Figure 1

Chronic stress reduces hippocampal growth hormone (GH). Hippocampal GH levels were assayed seven days after a two week period of immobilization stress (STR; n = 4) or handling (NS; n = 4). GH levels were expressed as a percentage, relative to the NS control average. GH was significantly lower in the hippocampi of stressed rats relative to unstressed rats. ^*Indicates p < 0.05.

Figure 2

Construction of an HSV-1 viral vector to overexpress GH. (A) The full-length gene for presomatotropin was cloned into an HSV-1 amplicon under the control of the HSV α-4 promoter. eGFP was co-expressed via the HSV α-22 promoter. (B) GH protein expression was confirmed in vitro. Vero cells were infected with GH virus at increasing MOIs. As the MOI increased, progressively higher levels of both GH and eGFP were detected. (C) The viral vector was infused into the dorsal hippocampus of rats. A representative infection, showing high levels of expression in pyramidal cells of CA1, and sparse infection in the granule cell layer of the dentate gyrus, is shown. Scale bar = 100 microns. (D) GH protein expression was quantified in infected dorsal hippocampal slices four days following virus delivery. GH levels were expressed as a percentage, relative to the NS control average. Viral overexpression of GH led to an approximate doubling of GH protein. ^*Indicates p < 0.05.

Figure 3

Overexpression of hippocampal GH rescues stress-related impairments in auditory trace conditioning. Two weeks of daily immobilization stress (STR) or handling (NS) were administered to rats. Twenty-four hours after the last session, GH or GFP virus was infused bilaterally into the dorsal hippocampus. (NS-GFP, n = 9; NS-GH, n = 7; STR-GFP, n = 7; STR-GH, n = 6). (A) After three days of recovery, auditory trace conditioning was administered. Neither stress nor GH administration affected conditional freezing during training. (B) Long-term contextual fear memory was measured the next day by returning rats to the conditioning context for 5 m. Stress impaired contextual fear memory, and intra-hippocampal GH expression partially reversed this impairment. (C) Long-term auditory fear memory was measured the following day by placing the rats in a novel context and presenting four tones in the absence of footshock. Stress impaired auditory fear memory, and intra-hippocampal GH expression fully reversed this effect. ^*Indicates p < 0.05 in a post-hoc comparison.

Figure 4

Overexpression of hippocampal GH rescues stress-related impairments in contextual conditioning. Ten days of daily immobilization stress (STR) or handling (NS) were administered to rats. Twenty-four hours after the last session, GH or GFP virus was infused bilaterally into the dorsal hippocampus. (NS-GFP, n = 4; NS-GH, n = 3; STR-GFP, n = 4; STR-GH, n = 3). (A) After three days of recovery, contextual fear conditioning was administered. Stress slowed contextual fear acquisition (^*Indicates p < 0.05 in post-hoc comparisons between STR-GFP and other groups), and this was prevented by intra-hippocampal GH. In contrast, intra-hippocampal GH had no effect in unstressed control rats. (B) The next day, the rats were returned to the context for an 8 m context extinction session. Stress impaired long-term contextual fear memory, and this impairment was rescued by expression of GH in the dorsal hippocampus. In contrast, intra-hippocampal GH tended to produce a mild impairment of long-term contextual fear memory in unstressed control rats. ^*Indicates p < 0.1 in a post-hoc comparison.