Thermal conditioning of quail embryos has transgenerational and reversible long-term effects

Abstract

Background: In the current context of global warming, thermal manipulation of avian embryos has received increasing attention as a strategy to promote heat tolerance in avian species by simply increasing the egg incubation temperature. However, because of their likely epigenetic origin, thermal manipulation effects may last more than one generation with consequences for the poultry industry. In this work, a multigenerational and transgenerational analysis of thermal manipulation during embryogenesis was performed to uncover the long-term effects of such procedure.

Results: Thermal manipulation repeated during 4 generations had an effect on hatchability, body weight, and weight of eggs laid in Japanese quails, with some effects increasing in importance over generations. Moreover, the effects on body weight and egg weight could be transmitted transgenerationally, suggesting non-genetic inheritance mechanisms. This hypothesis is reinforced by the observed reversion of the effect on growth after five unexposed generations. Interestingly, a beneficial effect of thermal manipulation on heat tolerance was observed a few days after hatching, but this effect was not transgenerational.

Conclusions: Our multigenerational study showed that thermal conditioning of quail embryos has a beneficial effect on post-hatch heat tolerance hampered by transgenerational but reversible defects on growth. Assuming that no genetic variability underlies these changes, this study provides the first demonstration of epigenetic inheritance of traits induced by environmental temperature modification associated with long-term impacts in an avian species.

Keywords: Avian; Embryo; Programming; Temperature; Transgenerational inheritance.

Fig. 1

Schematic representation of the multigenerational experience plan. Eggs incubated in standard control conditions (37.8 °C) are shown in blue. Thermo-manipulated eggs (TM; + 1.7 °C 12 h/d from incubation d 0–13) are shown in red. The name of the treatments for each generation is indicated on the top right side of the egg as follows: Cn: n generations incubated in control conditions; TMn: n consecutive generations of TM; TM2Cn-2: 2 consecutive generations of TM followed by n-2 generations incubated in control conditions. Gn: Generation n

Fig. 2

TM affects the growth of hatched quails. A Summary of the statistical results of weight analyses at 4 weeks of age (left) and 5 weeks of age (right). For G1, the 4 weeks measurement has been performed at D31 instead of D28 for all other generations. Complete results including P-values (P) and the effects on the sex and the interaction between treatment and sex can be found in Additional file 1: Supplementary Tables 2–4 and 6. For clarity, results from multigenerational and transgenerational analyses (“multig.” and “transg.”, see methods) are shown on two different lines. ^*P < 0.05; ^**P < 0.01; ^***P < 0.001; ns: not significant. Color codes (light green to green and light red) reflect the significance of the results (*, **, *** and ns, respectively). The transgenerational effect of TM on weight is circled in purple and the reversion of the effect is circled in red. B–D Effects of thermal manipulation on quail weight at 1 d (B), 4 weeks (C) and 5 weeks (D) of age. Mean and confidence interval were plotted by group at each generation. The weight of quails at D1 is presented independently of sex, this factor having no significant impact (Additional file 1: Supplementary Table 2). For G1, the 4 week measurement has been performed at D31 instead of D28 for all other generations. Cn: n generations incubated in standard control conditions; TMn: n consecutive generations of TM; TM2Cn-2: 2 consecutive generations of TM followed by n-2 generations incubated in control conditions; 1: males; 2: females. Different letters indicate significant differences between groups at P < 0.05

Fig. 3

TM affects the weight of laid eggs. A Effects of TM on the weight of eggs laid by G1 to G4 quails. Mean and confidence interval were plotted by group at G1–G4 generations. Cn: n generations incubated in control conditions; TMn: n consecutive generations of TM; TM2Cn-2: 2 consecutive generations of TM followed by n-2 generations incubated in control conditions. The results of the general statistical model are available in Additional file 1: Supplementary Tables 2–6, the impact of treatment in the "multig." or "transg." models being always significant (P < 0.05) regardless of the generations of treatment considered. Different letters indicate significant differences between groups at P < 0.05. B Histogram representing the mean egg weight difference laid by TM1, TM2, TM3 or TM4 quails compared to C1, C2, C3 or C4 quails, respectively. C Effects of TM on the weight of eggs laid by G5 to G7 quails. Cn: n generations incubated in control conditions; TM2Cn-2: 2 consecutive generations of TM followed by n-2 generations incubated in control conditions. Different letters indicate significant differences between groups at P < 0.05

Fig. 4

TM had a non-transgenerational and beneficial effect on heat stroke survival during the first days of life. A Graph showing the theoretical set point temperature (in red) and the actual temperature (in blue) in the rearing cell as a function of time. The period of heat stress due to an uncontrolled heat wave during which mortality was measured is indicated by the double arrow. D: Day of rearing (the first day of post-hatch life being D1). B Stacked histograms showing the number of animals that died or survived between D3 and D6 of rearing during heat stroke. The width of the columns is proportional to the number of animals measured. The C3TM treatment, shown in Fig. 1, corresponds to third generation control animals whose eggs were TM for the first time. Cn: n generations incubated in control conditions; TM2C2: two consecutive generations of TM followed by 2 generations incubated in control conditions. The impact of the heat wave was assessed on quail mortality between D1 and D6 using a Chi2 test, giving a result of P = 0.01088. Different letters indicate significant differences between groups at P < 0.05, obtained by Fisher's tests