Chemogenetic Depletion of Hypophysiotropic GnRH Neurons Does Not Affect Fertility in Mature Female Zebrafish

Abstract

The hypophysiotropic gonadotropin-releasing hormone (GnRH) and its neurons are crucial for vertebrate reproduction, primarily in regulating luteinizing hormone (LH) secretion and ovulation. However, in zebrafish, which lack GnRH1, and instead possess GnRH3 as the hypophysiotropic form, GnRH3 gene knockout did not affect reproduction. However, early-stage ablation of all GnRH3 neurons causes infertility in females, implicating GnRH3 neurons, rather than GnRH3 peptides in female reproduction. To determine the role of GnRH3 neurons in the reproduction of adult females, a Tg(gnrh3:Gal4ff; UAS:nfsb-mCherry) line was generated to facilitate a chemogenetic conditional ablation of GnRH3 neurons. Following ablation, there was a reduction of preoptic area GnRH3 neurons by an average of 85.3%, which was associated with reduced pituitary projections and gnrh3 mRNA levels. However, plasma LH levels were unaffected, and the ablated females displayed normal reproductive capacity. There was no correlation between the number of remaining GnRH3 neurons and reproductive performance. Though it is possible that the few remaining GnRH3 neurons can still induce an LH surge, our findings are consistent with the idea that GnRH and its neurons are likely dispensable for LH surge in zebrafish. Altogether, our results resurrected questions regarding the functional homology of the hypophysiotropic GnRH1 and GnRH3 in controlling ovulation.

Keywords: GnRH; LH; fertility; ovulation; zebrafish.

Figure 1

GnRH3 neuronal ablation is validated in the brain at 28 days of treatment. (A) Treatment and analysis schedule. Fish were exposed to 2 mM Mtz daily during the dark phase for 28 days, and were sampled at 14 and 28 days into the treatment. (B) Top panel: confocal images of GnRH3 neurons expressing NTR-mCherry fusion protein (magenta) in whole brains from the Tg(gnrh3:Gal4ff; UAS:nfsb-mCherry) fish before the treatment, and at 14 and 28 days of treatment. Bottom panel: POA-zoomed images showing the boxed area from the top images. Scale bars, 200 µm. (C) Confocal images of co-immunostaining of GnRH3-associated peptide (green) and mCherry (magenta) in the whole brain of Tg-Mtz (top panels) and Tg-Cont (bottom panels) at 28 days of treatment. GnRH3 neuron cell bodies co-localized with mCherry in Tg-Mtz are labeled by arrowheads. Scale bars, 100 µm. Insets, zoomed images of GnRH3-ir and mCherry-ir neuron cell bodies co-localization showing the boxed areas of the merged image in Tg-Mtz. Scale bars, 10 µm. (D) Numbers of GnRH3-positive cells (left) and mCherry-positive cells (right) in the POA (n = 6–7). (E) Relative gnrh3 mRNA levels in the whole brain (n = 4 each). OB, olfactory bulb; TL, telencephalon; POA, preoptic area; ir, immunoreactive; WT-Mtz, WT siblings treated with 2 mM metronidazole (Mtz); Tg-Cont, Tg(gnrh3:Gal4ff; UAS:nfsb-mCherry) treated with vehicle; Tg-Mtz, Tg(gnrh3:Gal4ff; UAS:nfsb-mCherry) treated with 2 mM Mtz. *, p < 0.05; **, p < 0.01; ***, p < 0.001.

Figure 2

GnRH3 neuronal projections are reduced at 28 days of treatment in the pituitary of GnRH3 neuron-ablated females. (A) Confocal images of immunostaining with GnRH3-associated peptide antibody in the whole pituitary (posterior to top left), showing GnRH3 neuronal projections (green). Scale bars, 1 mm. The bottom panel of Tg-Mtz, an image with enhanced signals with levels and tone curve adjustments to demonstrate that GnRH3 is still detectable in the pituitary. (B) LH levels in plasma (n = 9–14). WT-Mtz, WT siblings treated with 2 mM metronidazole (Mtz); Tg-Cont, Tg(gnrh3:Gal4ff; UAS:nfsb-mCherry) treated with vehicle; Tg-Mtz, Tg(gnrh3:Gal4ff; UAS:nfsb-mCherry) treated with 2 mM Mtz. p > 0.05.

Figure 3

GnRH3 neuron-ablated females exhibit normal fertility. (A) Number of fertilized eggs per spawn (fecundity), percentage of fertilized eggs (fertility), and hatching rate from the experimental females paired with WT males before treatment, and at 28 days of treatment. (B) Gonadosomatic index (GSI) (n = 7–11). (C) Representative ovarian sections stained with hematoxylin and eosin. Scale bars, 500 µm. Bottom panels, images of the intact ovaries with the corresponding GSI (mean ± SEM) adjacent to each image. Scale bars, 1 mm. WT-Mtz, WT siblings treated with 2 mM metronidazole (Mtz); Tg-Cont, Tg(gnrh3:Gal4ff; UAS:nfsb-mCherry) treated with vehicle; Tg-Mtz, Tg(gnrh3:Gal4ff; UAS:nfsb-mCherry) treated with 2 mM Mtz. p > 0.05.

Figure 4

No correlation is observed between the number of detectable POA GnRH3 neurons and each reproductive parameter at 28 days of treatment. (A) Table depicting the number of remaining detectable GnRH3 immunoreactive (ir) neurons in the preoptic area, and the results of reproductive performance test for individual females from 0 days (before treatment) to 28 days of treatment. (B) Pearson’s correlation coefficient analysis between the number of POA GnRH3-ir cells and each reproductive parameter at 28 days of treatment. R, correlation coefficient. P, p-value. Data from A. WT-Mtz, WT siblings treated with 2 mM metronidazole (Mtz); Tg-Cont, Tg(gnrh3:Gal4ff; UAS:nfsb-mCherry) treated with vehicle; Tg-Mtz, Tg(gnrh3:Gal4ff; UAS:nfsb-mCherry) treated with 2 mM Mtz.

Figure 4

No correlation is observed between the number of detectable POA GnRH3 neurons and each reproductive parameter at 28 days of treatment. (A) Table depicting the number of remaining detectable GnRH3 immunoreactive (ir) neurons in the preoptic area, and the results of reproductive performance test for individual females from 0 days (before treatment) to 28 days of treatment. (B) Pearson’s correlation coefficient analysis between the number of POA GnRH3-ir cells and each reproductive parameter at 28 days of treatment. R, correlation coefficient. P, p-value. Data from A. WT-Mtz, WT siblings treated with 2 mM metronidazole (Mtz); Tg-Cont, Tg(gnrh3:Gal4ff; UAS:nfsb-mCherry) treated with vehicle; Tg-Mtz, Tg(gnrh3:Gal4ff; UAS:nfsb-mCherry) treated with 2 mM Mtz.

Figure 5

Schematic diagram of the different possible pathways in inducing the pre-ovulatory LH surge along the HPG axis in female zebrafish. In general, the accepted dogma is that the hypophysiotropic GnRH and its neurons are critical for LH surge, and ultimately oocyte maturation, ovulation, and spawning (A). Our current results suggest that GnRH3 neuron-ablated zebrafish females display normal fertility, and previous studies suggest four alternative pathways (B–E): I. Hypophysiotropic GnRH3 has no role, but other brain factors (e.g., vasoactive intestinal peptide, kisspeptin, neurokinin B, and secretoneurin) trigger LH surge (B); II. Functional redundancy—GnRH3 has a role, but is not the sole and only player, thus sharing this function with other factors (C); III. Ovarian estradiol-17b (E₂) directly induces the LH surge and controls ovulation (D); IV. In GnRH3 neuron-ablated females, a few remaining hypophysiotropic GnRH3 neurons are possibly sufficient to elicit LH surge (E). The question marks indicate that these pathways remain to be established in zebrafish.