# Effects of 17α‑methyltestosterone on sexual morphology, sex ratios and behaviour of Gambusia holbrooki

Tran, NK ORCID: 0000-0001-5831-7738 2021 , 'Effects of 17α‑methyltestosterone on sexual morphology, sex ratios and behaviour of Gambusia holbrooki', PhD thesis, University of Tasmania.

 Preview
PDF (Whole thesis)

| Preview

## Abstract

The eastern mosquitofish Gambusia holbrooki was introduced to Australia in 1928 from USA as a bio-control method for mosquitoes and has spread out over freshwater coastal and inland waters of Australia. It is an ineffective mosquito control agent but rather is responsible for depletion of several native and endangered fish and amphibians. Because of its severe negative impacts on native species, G. holbrooki is classed as an invasive species requiring control and eradication efforts. The Trojan Sex Chromosome (TSC) strategy has been suggested in recent years as a genetic control method for this noxious invasive species. The TSC theory relies on the regular release of sex-reversed fish into target wild populations to produce single sex offspring after mating with wild fish; hence biasing the sex ratios leading to eventual extinction of the target populations. As a step towards applying the TSC strategy in G. holbrooki, this study investigated the sex reversal of genetic females into phenotypic males at three different life stages: adults, neonates and 30-day old juveniles. The study observed the timing of sex differentiation in this species to inform the best time for applying hormonal treatments before systematically optimizing hormone doses and determining appropriate life stages for hormonal masculinization as well as observing sexual morphology, behavioural interactions, and reproductive fitness of neo-males.
Firstly, the study identified the window when sex differentiation occurred in G. holbrooki to determine the best time for hormonal induction in this species. The embryonic stages (segmentation, pharyngula and hatching) and juveniles from 0 to 150 days after parturition (DAP) were sampled for gonadal histology. Morphometrically, the results showed that there was no significant difference in cell diameters of PGC (t-test, t = 1.71, df = 23, p = 0.1) and gonocytes (t-test, t = 0.46, df = 14, p = 0.66) between males (PGC, 13.9 ± 1.2 μm; gonocytes, 11.7 ± 1.1) and females (PGC, 14.7 ± 1.3 μm; gonocytes, 11.9 ± 1.2). However, significant differences in cell diameters between PGC, gonocytes, oogonia and spermatogonia were detected (one-way ANOVA, F (3, 85) = 113.3, p < 0.0005, ƞ$$^2$$ = 0.8). There were significant decreases in size of germ cells during their development (Tukey post hoc analysis). Cell diameter decreased from PGCs (14.25 ± 1.37 μm), to gonocytes (11.81 ± 1.14 μm), to oogonia (9.16 ± 0.94 μm), to spermatogonia (8.27 ± 1.5 μm), in that order. Several migrating PCG, were clearly distinguished from the somatic cells in the somatic layer at the mid-segmentation stage. However, they have not yet coalesced into a nest. At late somitogenesis, PGC nested and formed gonadal primordia before becoming undifferentiated gonads with two distinct clusters of gonial germ cells attached to the epithelial swim bladder. In female embryos, the first signs of ovarian differentiation, e.g., the presence of mitotic oogonia and chromatin-nucleolus oocytes were found in early pharyngula stage. In males, the gonads of newly born juveniles were sexually indifferent and the morphological differentiation became apparent only about 10 DAP with the appearance of spermatogonia in clusters of more than 4 cells enclosed by Sertoli precursor cells and stromal cell aggregations. All the information related to timing of sex differentiation and gonad development was used to design the following sex reversal experiments.
Sex reversal experiments administering 17α-Methyltestosterone (MT) via feed targeted three life stages separately: (i) adults, (ii) 30-day old juveniles and (iii) neonates. The concentrations of MT tested at adult stages (re-maturing and maiden females) ranged between 50 to 200 mg MT/kg feed. The treated females were fed control or respective MT incorporated feed to satiation for 50 days. Effects of the hormone on external sexual morphology (gonopodial development, internal gonad morphology), expression of the Anti-Müllerian Hormone (amh) gene and sexual behaviours of the treated re-maturing and maiden females were observed. Specifically, the treated females were anesthetized, and their anal fins photographed every seven days (from 0 to 50 days of the treatment period) for observing the effects of MT on the development of gonopodia. At the end of the 50-day treatment period, the maiden females in the 50 mg MT/kg diet group showing the most advanced gonopodial development were used to observe MT effects on sexual behaviour and the remaining females were euthanized to determine MT effects on gonad morphology and expression of the amh. The results indicated that MT at the dose of 50 mg/kg feed stimulated sexual character development, upregulated expression of amh, formation of testicular tissue and a change in the behaviour of the treated maiden females.
Like adult females, MT enriched feed at concentrations ranging from 5 to 40 mg/kg diet was administered to two different life stages: 30-day old juveniles and neonates. The experiment of 30-day old juveniles included one control and four MT treatment doses (10, 20, 30, 40 mg/kg diet). At 150 DAP, there was no significant difference in survival rates of juveniles fed on MT treatment and control diets (Kruskal-Wallis H test, F = 3.05, df = 4, 20, p = 0.55). The gonadal sex ratio in the control group was not significantly different from the expected 50: 50 sex ratio (Chi-square goodness of fit, χ$$^2$$ = 3.26, df = 1, p = 0.07), although the gonad sex ratios of the treatment groups of 10, 20, 30 and 40 mg/kg diet were statistically different from that of the control. The percentages of neo-males were significantly different between MT treatment groups (Kruskal-Wallis H test, χ$$^2$$ = 10.44, df = 4, p = 0.034). Post hoc tests revealed significant differences in the median percentages of neo-males between control group (0 %) and 10 mg MT/kg diet group (40 %, p = 0.04), but not between any other group combinations. Therefore, the lowest MT dose (10 mg/kg diet) was the best dose rate for MT treatment on 30-day old juveniles. In the experiment of MT treatment on neonates, there were one control and three MT treatment doses (5, 10, 20 mg/kg diet). Clutches of neonates were fed respective MT enriched or control spirulina. The hormonal treatments started at 1 DAP and lasted for 30 days during which the neonates were fed twice daily to satiation. At 150 DAP, there was no significant difference in survival rates of neonates fed on MT treatment and control diets (Kruskal-Wallis H test, F = 5.7, df = 3, p = 0.13). The body morphology sex ratio in the control groups was not significantly different from the expected 50: 50 sex ratio (Chi-square goodness of fit, χ$$^2$$ = 0.23, df = 1, p = 0.63). There was no significant difference in the sex ratios based on body morphology between the treatment and control groups. Intersex females were found in the 20 mg MT/kg diet group, whereas no testicular tissue was seen on the gonads of fish in the other treatments (5 and 10 mg MT/kg diet) or control groups. The percentages of neo-males out of phenotypic males were significantly different between various MT treatment groups (One-way ANOVA, F = 5.1, df = 3, 14, p = 0.014, ƞ$$^2$$ = 0.52). Tukey post hoc analysis revealed that the percentages of neo-males in 5 mg MT/kg group (45.9 ± 22.2 %) and 10 mg MT/kg group (51.1 ± 26.6 %) were statistically significant to that of the control (0 %) (p = 0.28 and p = 0.14, respectively), but no other group differences were statistically significant. Therefore, the neo-males from 5, 10 and 20 mg MT/kg diet group were used to observe their gonopodial and gonad morphology.
Sexual morphology of the neo-males was observed and compared thoroughly. Thirty wild males, five control males and 15 neo-males were anesthetized and their gonopodia were photographed for morphometric measurements. Nine normal males and ten neo-males were euthanized for histological observation and 3D gonopodial and testis reconstructions. The results showed that neo-males in the 20 mg MT/kg diet group possessed an under-developed gonopodium with no serrae and hooks, while neo-males in the two lowest doses (5 and 10 mg MT/kg diet) were stimulated to develop complete gonopodia. There were significant differences in the width ratio of fin rays 3 & 4 between neo-male groups, control males and wild males (Welch ANOVA, F ($${_3,}$$ $$_{9.4}$$) = 14.4, p = 0.001) in which the width ratios in control males (2.54 ± 0.13) and wild males (2.76 ± 0.14) were significantly larger than those of neo-males of 5 and 10 mg MT/kg diet (2.2 ± 0.11 and 2.15 ± 0.09, respectively). The 3D simulation of the basal gonopodium was novel and shed light on the mechanism of transferring sperm packages from males to females in poeciliids. The results of histological observation and 3D reconstruction of testes determined that neo-males in the lowest dose group (5 mg MT/kg diet) developed funnel-shaped testes similar to testes found in control males and the structures of testicular lobes were comparable to those of control males. However, neo-males in the 10 and 20 mg MT/kg groups formed oval-shaped testes and abnormal development of testicular tissue. These observations demonstrated that MT at the dose of 5 mg/kg diet effectively sex reversed genetic females to phenotypic males with the complete development of gonopodia and testes as normal males. The neo-males in the 5 and 10 mg MT/kg diet groups showed well-developed gonopodia and were subsequently used to determine their behavioural interactions and reproductive fitness with normal virgin females.
Behavioural interactions, reproductive fitness assessment and sperm packet expression were conducted on the two groups of neo-males (fed on 5 and 10 mg MT/kg diet) and one control group. Fish from control or neo-male groups (test fish) were individually paired with a virgin female (stimulus fish) and their behaviours were recorded for 60 minutes using an in-front camera. The footage was analysed to observe behavioural interactions of the test fish and stimulus fish. After recording behaviour interactions, each pair of a control male/neo-male and a virgin female was kept together in an aquarium until the signs of female pregnancy were evident or at least 6 months before the females were separated to observe reproductive fitness. The results indicated that there were significant differences in median numbers of mating attempts between control and neo-male groups (Kruskal-Wallis H test, χ$$^2$$$$_{(2)}$$ = 6.6, p = 0.036). Post hoc tests revealed that there was a significant difference in median numbers of mating attempts between neo-males of 5 mg MT/kg diet (42) and neo-males of 10 mg MT/kg diet (17, p = 0.043), but no significant difference was found between the neo-male groups and control group (p = 0.17 and p = 1.0 respectively for the neo-males of 5 and 10 mg MT/kg diet groups). Within the 60 minutes of behavioural observations, most control males and neo-males did not illustrate any aggressive behaviours to virgin females. The number of aggressive behaviours by the virgin females increased over the time. In terms of reproductive fitness, 60 % of virgin females paired with control males gave birth to young, while only 20% of virgin females paired with the neo-males of 5 mg MT/kg diet group parturated to immature neonates all of which died one day later. No virgin females paired with the neo-males of 10 mg MT/kg diet group gave birth or showed signs of pregnancy. Like control males, spermatozeugmata could be expressed from all neo-males in the 5 mg MT/kg treatment, but these sperm packets could be squeezed from only 55.6% of neo-males of 10 mg MT/kg diet group. The observations indicated that MT at the dose of 5 mg/kg diet effectively sex reversed genetic females to phenotypic males with behaviour interactions and production of sperms packets as normal males, but the results for their reproductive fitness was uncertain because of time limitation of the PhD project. Therefore, further investigations of the reproductive fitness of neo-males in this group are warranted.
In conclusion, masculinization of G. holbrooki was successfully achieved by using MT, although a 100 % male population could not be attained. The study established a protocol to masculinize this species at optimum dose of MT through oral administration to neonates (5 mg/kg diet). The protocol and information produced in this recent study provides key knowledge towards producing functional neo-males for developing the TSC strategy to control and eradicate this pest species. The results presented in this thesis will also benefit other studies on fish sex differentiation especially in live-bearing species as well as research on biology, ecotoxicology and pest fish management.

Item Type: Thesis - PhD Tran, NK Sex-reversal, masculinization, mosquitofish, sex differentiation, gonad morphology, 3D simulation, behaviour https://doi.org/10.25959/100.00045943 Copyright 2021 the author View statistics for this item