Targeted Mutagenesis of the Hypophysiotropic Gnrh3 in Zebrafish (Danio rerio) Reveals No Effects on Reproductive Performance

Abstract

Gnrh is the major neuropeptide regulator of vertebrate reproduction, triggering a cascade of events in the pituitary-gonadal axis that result in reproductive competence. Previous research in mice and humans has demonstrated that Gnrh/GNRH null mutations result in hypogonadotropic hypogonadism and infertility. The goal of this study was to eliminate gnrh3 (the hypophysiotropic Gnrh form) function in zebrafish (Danio rerio) to determine how ontogeny and reproductive performance are affected, as well as factors downstream of Gnrh3 along the reproductive axis. Using the TALEN technology, we developed a gnrh3-/- zebrafish line that harbors a 62 bp deletion in the gnrh3 gene. Our gnrh3-/- zebrafish line represents the first targeted and heritable mutation of a Gnrh isoform in any organism. Using immunohistochemistry, we verified that gnrh3-/- fish do not possess Gnrh3 peptide in any regions of the brain. However, other than changes in mRNA levels of pituitary gonadotropin genes (fshb, lhb, and cga) during early development, which are corrected by adulthood, there were no changes in ontogeny and reproduction in gnrh3-/- fish. The gnrh3-/- zebrafish are fertile, displaying normal gametogenesis and reproductive performance in males and females. Together with our previous results that Gnrh3 cell ablation causes infertility, these results indicate that a compensatory mechanism is being activated, which is probably primed early on upon Gnrh3 neuron differentiation and possibly confined to Gnrh3 neurons. Potential compensation factors and sensitive windows of time for compensation during development and puberty should be explored.

Fig 1. Verification of targeted, heritable mutation in gnrh3 by gene and transcript.

(Aa) The gnrh3^-/- deletion is localized to exon 2 (highlighted in gray) of the gnrh3 gene and includes the start codon (highlighted in green) and a portion of the nucleotide sequence encoding the Gnrh3 decapeptide (highlighted in pink). The underlined regions represent the TALEN recognition sites. WT = gnrh3^+/+. M = gnrh3^-/-. (Ab-c) Sequencing chromatograms of gnrh3^+/+ and gnrh3^-/- gDNA, demonstrating the homogenous sequence of the gnrh3^-/- mutation. The purple regions represent the gDNA sequences of gnrh3^+/+ that were deleted in the gnrh3^-/- gene. Black arrows represent where single base pair substitutions occurred. (B) The gnrh3^-/- cDNA reveals no change in the gnrh3^-/- mutation after transcription. (Ba) The pink region represents the nucleotide sequence of the Gnrh3 decapeptide, while the underlined regions represent the TALEN recognition sites. WT = gnrh3^+/+. M = gnrh3^-/-. (Bb-c) Chromatograms of gnrh3^+/+ and gnrh3^-/- cDNA indicate the homogenous sequence of the gnrh3^-/- mutation.

Fig 2. Anti-zebrafish Gnrh3 Gap polyclonal antibody is specific to zebrafish Gnrh3.

(A) The immunostaining of Gnrh3-ir soma and fibers (white arrow; Aa) in the brain with anti-zebrafish Gnrh3 Gap was eliminated when the antibody was substituted with pre-immune serum (Ab). (B) Immunostaining with anti-Gnrh3 Gap (green; Ba) on brain sections of gnrh3:tdTomato (red; Bb) adults shows co-localization between Gnrh3-ir soma and gnrh3:tdTomato-labeled soma (Bc), indicating the specificity of the antibody to Gnrh3. White arrows indicate Gnrh3-tdTomato-expressing soma that were positively stained by anti-Gnrh3 Gap. (C) Immunostaining with (Ca-Cc) anti-zebrafish Gnrh3 Gap or (Cd) pre-immune serum in COS7 cells transfected with (Ca) control pcDNA3.1 plasmid, (Cb) gnrh2-pcDNA3.1 plasmid, or (Cc,d) gnrh3-pcDNA3.1 plasmid. The cells that express zebrafish Gnrh3 and are immunostained with anti-zebrafish Gnrh3 Gap (red) are indicated by white arrows (Cc). Scale bars = 50 μm.

Fig 3. gnrh3 mRNA but not Gnrh3 protein detectable in gnrh3^-/- fish.

(A) Immunohistochemistry on adult coronal brain sections (Aa) using anti-zebrafish Gnrh3 Gap (green) demonstrates the presence of Gnrh3 signal in the form of somas (white arrows) and fibers in the gnrh3^+/+ pre-optic area of the brain (Ab). However, no Gnrh3 signal was found in the gnrh3^-/- pre-optic area (Ac) or in any other region of the brain. (B) In situ hybridization on adult sagittal brain sections (Bd) using gnrh3 DIG-labeled riboprobes. The anti-sense gnrh3 riboprobe demonstrated mRNA (red) in the ventral telencephalon and pre-optic area of both gnrh3^+/+ (Ba) and gnrh3^-/- (Bb) fish. The sense gnrh3 riboprobe demonstrated no gnrh3 mRNA signal in any brain regions in the gnrh3^+/+ fish (Bc). Scale bars = 50 μm.

Fig 4. Developmental mRNA levels of pituitary gonadotropin genes tend to be higher in gnrh3^-/- fish than in gnrh3^+/+ fish.

mRNA levels at different time points of (A) fshb, (B) lhb, and (C) cga were measured in pooled samples of whole embryos/larvae/juveniles. Absolute mRNA levels were normalized to ef1a levels and are presented as mean ± SEM. Differences between genotypes at a specific time point were determined by a one-tailed, homoscedastic Student t-test and are considered statistically significant when *P ≤ 0.05. gnrh3^+/+, grey bars. gnrh3^-/-, white bars.

Fig 5. Male and female adult mRNA levels of pituitary gonadotropin genes show no differences between gnrh3^+/+ and gnrh3^-/- fish.

Pituitary mRNA levels of (A) fshb, (B) lhb, and (C) cga were determined using QPCR. Absolute mRNA levels were normalized to ef1a levels and are presented as mean ± SEM. Differences between genotypes for each sex were determined by a one-tailed, homoscedastic Student t-test and are considered statistically significant when *P ≤ 0.05. gnrh3^+/+, grey bars. gnrh3^-/-, white bars.

Fig 6. Adult male and female Lh levels in the pituitaries of gnrh3^-/- fish do not differ from gnrh3^+/+ fish.

Lh levels in the pituitaries of male and female adults (n = 4) were measured by ELISA. Values are presented as mean ± SEM. Differences between genotypes for each sex were determined by a one-tailed, homoscedastic Student t-test and are considered statistically significant when *P ≤ 0.05. gnrh3^+/+, grey bars. gnrh3^-/-, white bars.

Fig 7. gnrh3^-/- juveniles exhibit normal Gnrh3 neuronal migration.

In both gnrh3^+/+ gnrh3:tdTomato (B) and gnrh3^-/- gnrh3:tdTomato (C) fish, Gnrh3-tdTomato-expressing soma located in the olfactory region and ventral telencephalon project fibers that extend posteriorly toward the hypothalamus and pituitary stalk (pituitaries not shown; A). Z-stack images were taken at 20x magnification on a Leica Microsystems DMi8 confocal microscope with a resolution of 1024 x 1024 and a z-step size of 0.10. All images were analyzed and assembled with Image J and Adobe Photoshop. Scale bars = 100 μm. OB = olfactory bulb. T = telencephalon. OT = optic tectum. ON = optic nerves. POA = pre-optic area. P = pituitary.

Fig 8. Adult gonadal morphology and gametogenesis do not differ between gnrh3^+/+ and gnrh3^-/- fish for both males and females.

(A-D) Gross gonadal morphology of adult male (A,C) and female (B,D) fish of both genotypes: gnrh3^+/+ (A,B) and gnrh3^-/- (C,D). (E-H) Gonadal histology with hematoxylin and eosin staining, demonstrating no differences between gnrh3^+/+ and gnrh3^-/- gametogenesis for both males and females. Male gnrh3^-/- testes contain spermatozoa in the lumens of testicular cysts (G), similar to gnrh3^+/+ testes (E). Female gnrh3^-/- ovaries contain all stages of vitellogenesis (H), including late-vitellogenic oocytes, which is similar to gnrh3^+/+ ovaries (F). White asterisks = mature spermatozoa in lumens of spermatocysts. Black stars = mature, late vitellogenic oocytes. Scale bars = 125 μm (testes) and 250 μm (ovaries).

Fig 9. Reproductive capacity of gnrh3^-/- adult fish is not different from that of gnrh3^+/+ fish.

(A) Fecundity, (B) fertilization rate, and (C) offspring survival (to 2 dpf) were determined in the spawning of four different crossing combinations: gnrh3^+/+ male x gnrh3^+/+ female, gnrh3^+/+ male x gnrh3^-/- female, gnrh3^-/- male x gnrh3^+/+ female, and gnrh3^-/- male x gnrh3^-/- female. No differences between any of the parameters assessed were observed between any of the spawning combinations. All values are presented as mean ± SEM with 3–6 replicates per group, were subjected to a one-way ANOVA, and were considered statistically significant when *P ≤ 0.05. WT = gnrh3^+/+. KO = gnrh3^-/-. M = male. F = female.