Overexpression or knockdown of rat tryptophan hyroxylase-2 has opposing effects on anxiety behavior in an estrogen-dependent manner

Abstract

Previous studies showed that chronic estrogen treatment increases tryptophan hydroxylase-2 (TpH2) mRNA in the caudal dorsal raphe nucleus (DRN), and this increase was associated with decreased anxiety. The present study explored the interaction of estrogen and targeted, bidirectional manipulation of TpH2 expression in the caudal DRN by knockdown or viral overexpression, to decrease or increase tryptophan hydroxylase expression respectively, on anxiety behavior. Rats were ovariectomized and replaced with empty or estradiol capsules (OVX, OVX/E, respectively). Animals received microinfusions of either antisense TpH2 or control morpholino oligonucleotides into caudal DRN and were later tested in the open field test. A separate group of animals were microinfused with TpH2-GFP or GFP-only herpes simplex viral vectors into caudal DRN and tested in the open field. The bidirectional impact of manipulations on TpH2 expression was confirmed using a combination of quantitative protein and mRNA measurements; TpH2 expression changes were limited to discrete subregions of DRN that were targeted by the manipulations. Estradiol decreased anxiety in all behavioral measures. In the OVX/E group, TpH2 knockdown significantly decreased time spent in the center of the open field, but not in the OVX group, suggesting that TpH2 knockdown reduced the anxiolytic effects of estrogen. Conversely, TpH2 overexpression in the OVX group mimicked the effects of estrogen, as measured by increased time spent in the center of the open field. These results suggest that estrogen and TpH2 in the caudal DRN have a critical interaction in regulating anxiety-like behavior.

Figure 1

PMOs were successfully taken up by the cells in the DRN. A representative fluorescent image of PMO injection in the DRN at 20X (A) and 40X (B) magnification. Scale bar, 500μm (A), 50μm (B).

Figure 2

TpH immunoreactivity is reduced by PMO infusions in the midrostral DRN. Injections of scrambled PMO in the midrostral DRN (B) did not change DAPI signals (A) or TpH immunoreactivity (C). On the other hand, injections of αTpH2 PMO (E) markedly reduced TpH immunoreactivity (F) without affecting DAPI signals (D). G and H show magnified view (40X) of the scrambled and αTpH2 PMO injection site, respectively. Dashed ovals encircle the region with lissamine-PMO injection. Scale bar, 500μm (A–F), 20μm (G, H).

Figure 3

TpH protein expression is reduced by PMO infusions in the midrostral DRN. Injections of αTpH2 PMO in the midrostral significantly reduced TpH immunoblot integrated density compared to the unoperated (Unop), saline or scrambled PMO injected groups (A). However, it had no effect on TpH integrated density in the neighboring caudal DRN (B), which was not targeted by these injections. TpH integrated density shown by mean ± SEM. * p=0.039.

Figure 4

TpH2 knockdown in the caudal DRN reversed the anxiolytic effects of estrogen in the open field test. Estrogen treatment in OVX rats significantly increased time spent in the center of the open field compared to all other groups (A). Percent time spent in the center (A), corner (B), and total distance traveled (C) are shown by mean ± SEM. * p=0.004.

Figure 5

TpH-GFP HSV injection targeting the caudal DRN selectively increased TpH mRNA expression in the caudal DRN. A and E show representative fluorescent image of TpH2-GFP viral vector targeting the caudal DRN at 40X magnification (Scale bar, 20μm). TpH2 viral vector infusions are restricted in the caudal DRN (A), as GFP-labeled cells were rare in the neighboring midrostral DRN (E). Representative phosphorimages of TpH2 in situ hybridization (ISHH) signal in the caudal DRN are compared between treatment groups with GFP-only injections (C) versus TpH-GFP injections (D). Analysis of the ISHH signals shows that TpH-GFP injection increased density of TpH mRNA in the caudal DRN (B), but not in the neighboring area, mid-DRN (F), compared to GFP-only HSV injection. TpH mRNA density expressed as mean ± SEM. *p=0.013.

Figure 6

TpH2 overexpression mimicked the anxiolytic effects of estrogen in the OVX rats. A) Infusion of TpH2-GFP viral vector in the caudal DRN of OVX rats increased the time spent in the center of the open field (#p=0.025) to a comparable level as the anxiolytic effects of estrogen treatment in the OVX rats injected with GFP-only viral vector (*p=0.003). B) Estrogen treatment similarly decreased time spent in the corner of the open field (*p=0.005) in GFP-only groups; however, TpH2 overexpression in the OVX/E rats was anxiogenic, as shown by increased time spent in the corner of the open field (#p=0.041). C) There were no significant differences between any of the treatment groups in overall locomotion. Percent time spent in the center (A), corner (B) and total distance traveled in cm (C) are shown by mean ± SEM.

Figure 7

Theoretical serotonin-anxiety interaction. Our data suggest a U-shaped relationship between serotonin and anxiety. Since TpH2 knockdown was anxiogenic in OVX/E rats (red arrow) while TpH2 overexpression was anxiolytic in OVX rats (green arrow), it suggests an inverse linear relationship between anxiety and 5-HT synthetic capacity as shown by the left half of the U-shaped curve. However, we also found that TpH2 overexpression was actually anxiogenic in OVX rats treated with chronic estrogen (dotted green arrow). These data suggest that increased 5-HT tone is anxiolytic at a lower range of serotonin activity, but is anxiogenic at higher level of serotonergic activity. Because chronic estrogen increases TPH2 mRNA in the caudal DRN, it is tempting to hypothesize that this supraphysiological overexpression TPH2 in OVX/E rats may mimic 5-HT level of male rats whose 5-HT levels are normally higher than females.