Sex ratio theory suggests that females should bias offspring sex ratios based on maternal condition and the availability of critical food resources. Work in birds indicates that females do, indeed, bias sex ratios according to maternal condition and food quality and quantity. Yet it is unknown whether these sex ratio skews occur due to fluctuations in particular micro- or macro-nutrients, caloric content overall, or even the perception of food availability. We hypothesized that dietary fats may drive biases in offspring sex ratios, because measures of maternal condition often reflect fat reserves, and fats are critical for the process of egg-laying in birds. To test this, we provided breeding Japanese quail, a species that biases sex ratios in response to maternal condition, with either a control breeding diet or a diet supplemented with two oils (safflower oil and flaxseed oil). These oils were chosen for their high omega-3 and omega-6 fatty acid content as well as their importance in mammalian sex allocation. We then measured influences of these diets on the sex ratio of offspring, the change in maternal weight, and the laying rates of female quail. The dietary oil supplements increased weight gains in quail but decreased the number of eggs laid during the experiment. There was no influence of the oil supplements on offspring sex ratios. This indicates that fat may not be a macro-nutrient involved in the process of sex ratio adjustment in quail.

Housing and Bird Care

Sexually mature Japanese quail (Coturnix coturnix japonica) (n=65 males and 65 females) were pair-housed in wire cages (6”x12”x10”) with one male and one female per cage. Quail had ad libitum access to water and feed throughout the entire experiment through nipple drinkers and trough feeders. The quail cages were housed in a single, climate-controlled room with a light clock schedule of 14:10 hours of light to dark. These quail were part of a breeding colony maintained by the University of Georgia and were available for inclusion in this experiment when they were in peak lay, at approximately 24 weeks old. Japanese quail lay one egg per day, and clutch sizes range from 10–14 eggs per clutch (Lukanov and Pavlova 2020) with an average lag of 21.6h between successive clutches (Aggrey et al. 1993). In our breeding colony, we have noted similar clutch sizes, but rarely, we have observed females lay more than one egg per day, a phenomenon that has never been officially reported on in quail but has been observed previously in chickens (Navara and Wrobel 2019). The average fertility rates in previous studies of domestic quail were around 87% (reviewed in Lukanov and Pavlova 2020), though it is not uncommon in our breeding colony to see lower fertility rates, between 60 and 70%, in unmanipulated birds. It is still unknown whether wild Japanese quail are monogamous or polygynous; instead, it is likely that they can exhibit either mating strategy, but we maintain our birds in monogamous pairs, and female quail conduct all parental care in both wild and domestic situations.

Design and Dietary Treatments

The control diet was the standard quail layer diet used at the University of Georgia Poultry Research Facility (Table 1). Our high-fat diet was formulated by a poultry nutritionist at the University of Georgia (Dr. Woo Kim); it included 5% safflower seed oil (Hollywood® Safflower Oil) and 5% flaxseed oil (Puritan’s Pride® Natural Organic Flaxseed Oil) by weight and a reduction of carbohydrate content to account for the increased caloric load of the two added oils. We chose these amounts because it was the largest change in dietary fat that we could achieve without reducing a majority of the other critical nutrients in the diet other than carbohydrates. Overall, the formulation effectively elevated the fat content of the diet while simultaneously decreasing the carbohydrate content. We chose to decrease the carbohydrate content rather than increase the total caloric content of the diet because it would have been impossible to determine whether any effects caused were because of caloric content or a particular macronutrient. Unfortunately, it was impossible to adjust fat content by itself, so results will be interpreted with the understanding that carbohydrate content was reduced in this study as well, with the assumption that any effects we saw would need to be further examined to ensure that the decrease in carbohydrate content was not playing a role. We allocated 30 quail pairs to the fat-supplemented group and 35 pairs were allocated to stay on the control diet. The two groups remained on these diets for the remainder of the experiment. After two weeks on the dietary treatments, eggs were then collected for 14 days (Figure 1). We waited this two-week period because quail eggs can take anywhere from 4–7 days to complete rapid yolk deposition (Bacon and Koontz 1971); we wanted to be sure that all birds had acclimated to the treatment for at least two weeks and all eggs we measured were influenced by the dietary supplementation. We collected a total of 496 eggs from control females and 312 eggs from females on the experimental diet. Female body weights were measured both before and at the end of the experiment using a digital scale (accuracy 0.01g).

Sexing of Offspring

After collecting eggs, we stored them in a cooler at 4℃ for a maximum of seven days before transferring them to an incubator at 37.5℃ at 58% relative humidity for four days. The incubated eggs were then removed and frozen at -50℃. While some suggest that sexing unincubated eggs is a better method of detecting primary sex ratios (Klein et al., 2003), there have been questions about whether contamination with maternal granulosa cells may influence the results (Arnold et al. 2003a). We opted to incubate for four days, as we have in previous studies, because this provided ample embryonic tissue for DNA extraction (Gam et al. 2011, Pinson et al. 2015). A total of 332 eggs from control females and 187 eggs from females on the experimental diet were fertile, and embryos were collected from these eggs. The remaining 164 eggs from control females and 125 eggs from females on the experimental diet were infertile and did not yield embryonic material for sexing.

To extract DNA from embryos, we used a standard salt extraction according to procedures described in Lambert et al. (2000). While eggs were still frozen, we removed their eggshells and weighed out 10–20mg of embryonic tissue. DNA amplification was focused around the CHD-1 alleles to visualize male and female sex chromosomes (Fridolfsson and Ellegren 1999). PCR primers and reaction concentrations were the same as specified in Pinson et al. (2015). Reaction times and temperatures were as described in Fridolfsson and Ellegren (1999). Primers used were 2550F (5'-GTTACTGATTCGTCTACGAGA-3') and 2718R (5'-ATTGAAATGATCCAGTGCTTG-3'). PCR products were visualized utilizing ethidium bromide staining of a 3% agarose gel. Male products presented as a single band while female products presented as two bands. Eggs for which there was no evidence of embryonic development were deemed to be infertile.

Statistical Analyses

To test whether the treatment influenced the sex ratio of embryos produced by females, embryos were coded as “1” for female and “0” for male. We then conducted a generalized linear mixed effects model, including dietary treatment, the change in weight, the log-transformed value of initial weight, and the interactions of these variables as fixed factors and female ID as a random effect. We conducted a similar analysis to test the effects of treatment on whether the eggs laid were fertile (fertile eggs were coded as “1” while infertile eggs were coded as “0”), and whether an egg was laid on a given day, since quail generally lay one egg per day (egg laid was coded as “1” while egg not laid was coded as “0”). Because quail in our population occasionally lay more than one egg per day, we also tested whether treatment influenced the incidence of double eggs using a logistic regression analysis; females that laid 2 eggs in one day were coded as “1” and females that did not were coded as “0”. Next, we tested whether the initial weight and/or change in weight of the females was related to the number of eggs they laid using general linear models.

We tested whether female body weights were different between the two treatment groups at the beginning and the end of the experiment using a repeated measures ANOVA. The residuals of the initial and final weights were both non-normally distributed based on Shapiro-Wilks tests and needed to be log-transformed for analysis. We also tested whether treatment influenced the change in weight over the duration of the experiment using a general linear model with dietary treatment as the predictor variable and change in weight as the dependent variable.

Differences were considered significant at p < 0.05 and results are reported below with means ± standard deviations. Statistical analyses were carried out using RStudio (version 4.2.1), using the lmer package for sex ratio analyses.

Microsoft Excel