The immediate early gene early growth response gene 3 mediates adaptation to stress and novelty

Abstract

Stress and exploration of novel environments induce neural expression of immediate early gene transcription factors (IEG-TFs). However, as yet no IEG-TF has been shown to be required for the normal biological or behavioral responses to these stimuli. Here we show that mice deficient for the IEG-TF early growth response gene (Egr) 3, display accentuated behavioral responses to the mild stress of handling paralleled by increased release of the stress hormone corticosterone. Egr3-/- mice also display abnormal responses to novelty, including heightened reactivity to novel environments and failure to habituate to social cues or startling acoustic stimuli. In a Y-maze spontaneous alternation task, they perform fewer sequential arm entries than controls, suggesting defects in immediate memory. Because stress and novelty stimulate hippocampal long-term depression (LTD), and because abnormalities in habituation to novelty and Y-maze performance have been associated with LTD deficits, we examined this form of synaptic plasticity in Egr3-/- mice. We found that Egr3-/- mice fail to establish hippocampal LTD in response to low frequency stimulation and exhibit dysfunction of an ifenprodil-sensitive (NR1/NR2B) N-methyl-d-aspartate receptor subclass. Long term potentiation induction was not altered. The NR2B-dependent dysfunction does not result from transcriptional regulation of this subunit by Egr3, because NR2B mRNA levels did not differ in the hippocampi of Egr3-/- and control mice. These findings are the first demonstration of the requirement for an IEG-TF in mediating the response to stress and novelty, and in the establishment of LTD.

Figures

Fig. 1. Egr3−/− mice are hypersensitive to handling stress

a, Egr3−/− mice were significantly (*) more reactive to handling on each of 3 consecutive test days. b, Baseline corticosterone levels in Egr3−/− mice did not differ from controls, but handling induced significantly greater corticosterone release in Egr3−/− mice (*p = 0.002), indicating heightened physiologic response to mild stress.

Fig. 2. Egr3−/− mice display heightened reactivity to novelty stress

Egr3−/− mice were more active in a 1-hour open-field activity test compared to WT control mice (p = 0.001). This effect was greatest early in the test and decreased to control levels over time, indicating that the heightened activity was a response to the novel environment (p values for blocks 1–6 respectively: p

Fig. 3. Egr3−/− mice show increased social investigation of familiar mice suggesting failure to habituate to a social stimulus

a, b, Egr3−/− mice showed significantly more social investigation of familiar cagemates compared to that of WT mice (n = 8 WT, n = 7 Egr3−/−). a, This effect was greatest for investigation of the anogenital region (p < 0.001) but b, was also significantly different in the head region (p = 0.015). c, the hidden cookie test of olfactory function revealed that Egr3−/− mice took significantly longer to find and eat hidden food when tested in a novel cage (*, p = 0.028; n = 5 WT, n = 5 Egr3−/−), but showed no difference when tested in a familiar cage (n = 12 WT, n = 7 Egr3−/−; see methods for explanation of variance in n), indicating that Egr3−/− mice are able to smell but their behavior is highly affected by novelty stress.

Fig. 4. Egr3−/− mice fail to habituate to a startling sound and show deficits in spontaneous alternation in a Y-maze

a, Initial startle response, displayed as means of blocks of 5 startle trials, did not differ between Egr3−/− and WT mice. However, while the response of WT mice decreased significantly over time (†), Egr3−/− mice failed to habituate to the stimulus, resulting in significantly (*) increased startle response compared to WT in blocks 3 and 4 (p < 0.001, and p = 0.004 respectively). b, Egr3−/− mice made significantly fewer successful alternations in the Y-maze (* p < 0.05, students t-test; † p < 0.05, Mann-Whitney U test).

Fig. 5. Impaired induction of hippocampal LTD, but not LTP, in Egr3−/− mice

a, Time course of change in EPSPs in the CA1 region following LFS in slices from WT (filled circles; n = 5) and Egr3−/− mice (open circles; n = 5) demonstrate a deficit in LTD in Egr3−/− mice. Traces to the right depict EPSPs obtained before (solid line) and 60 min after (dotted line) LFS in slices from an Erg3−/− mouse (upper) and a WT mouse (lower). b, Change in CA1 region EPSPs following 100 Hz × 1 s HFS of the Shaffer Collaterals in slices from WT (filled circles; n = 5) and Egr3−/− mice (open circles; n = 5) demonstrate no difference in LTP based on Egr3 genotype. Traces to the right depict EPSPs obtained before (solid line) and 60 min after (dotted line) HFS in slices from an Erg3−/− mouse (upper) and a WT mouse (lower).

Fig. 6. Dysfunction of NMDARs in Egr3−/− mice

a, Loss of ifenprodil sensitivity of synaptic NMDARs in the CA1 region of slices from Egr3−/− mice. The time course of change in isolated NMDAR-mediated EPSPs in slices from WT mice (filled circles) and Egr3 −/− mice (open circle) following treatment with 10 µm ifenprodil. b, WT mice treated with 3mg/kg ifenprodil made significantly fewer successful alternations in the Y-maze than vehicle-treated WT mice (* p < 0.05, students t-test; † p < 0.05, Mann-Whitney U test).