Number of animals and method of euthanasia
In accordance with Directive 2010/63/EU of the European Parliament and of the Council of 22 September 2010, all the animals were slaughtered by decapitation (as birds weighting less than 250g), and the number of animals was reduced to the maximum by setting the power of the test at 80% and the alpha risk at 5% using a bilateral test. From previous studies of gene expression, we expected a coefficient of variability around 50%, and wanted an inter group variation of 50%. These parameters lead us to calculate n=16 animals per group. Given the fertility and hatchability rates (estimated at 90% and 80% respectively), we chose to incubate 160 eggs (to assure 7 sample points of 16 individuals), and we were finally able to sample 20 individuals per group.
Animal and sample collections
A total of 160 mule duck eggs, from mothers aged 46 weeks (genotype H85, provided by Grimaud Frères Selection Company, Roussay, France), were kept at room temperature during 3 days, prior to incubation at 37.6°C, and 60% average relative humidity (RH) during the whole incubation period. All eggs were turned through 90° every 3 hours. Temperature and hygrometry were continuously measured by a sensor (KIMO). Unfertile eggs were excluded by candling at E10, with a sliding of remaining eggs to prevent local temperature disturbances caused by the appearance of holes. At E27, all eggs were placed in the same hatcher at 37.3°C and 80% RH. On day 2, the ducklings were transferred to a rearing room where the ambient temperature was adjusted to 26-28°C and the starting diet (PALMA07, Maïsadour, France) was available ad libitum. Livers from 20 randomly selected animals were sampled every 4 days from 12th embryonic day (E12) to the 4th day after birth (D4).
Samples were frozen in liquid nitrogen for RNA analysis. Total RNA was isolated from frozen tissue according to the Ribozol method (VWR Life Science). Total RNA concentration was measured by spectrophotometry (optical density at 260 nm) using a Biotek EPOCH 2 microplate reader with Take3 Plate, and all the samples were normalized at 500 ng/ μl. The integrity of total RNA was analyzed by electrophoresis. An amount of 3µg RNA was reverse-transcribed to cDNA with Iscript Reverse Transcription Supermix for RTqPCR (Bio-Rad, USA) with duplicates of samples. The absence of DNA contamination was prevented by DNase treatment. Reverse transcription reaction was done in CFX384 (Bio-Rad, USA) according to this program: 25 °C/5 min, 46 °C/20 min, 95 °C/1 min.
qPCR EvaGreen using BioMark
The mRNA levels of 50 genes coding for proteins involved in lipid metabolism, carbohydrate metabolism, stress and development were quantified. The primer sequences (listed in supplemental tables 1 to 4) used in the qPCR assays were created first by aligning the protein sequences of humans, mice and ducks (Anas platyrhynchos) on MultAlin (49) to identify the best-preserved exon. These exons were then treated on Primer3 (50,51) to build specific primers. Validation of their efficiency ranged between 1.90 and 2 was carried out by using cascade dilution of a cDNA pool. High throughput real-time quantitative PCR was performed using the Biomark microfluidic system from Fluidigm (GeT-PlaGe platform, Castanet-Tolosan, France) in which every sample-gene combination is quantified using a 96.96 Dynamic Array™ IFCs (BMK-M-96.96, Fluidigm,). Pre-amplification of the samples, chip loading and real time quantitative PCR were performed according to manufacturer’s protocol. Real time quantitative PCR results were analyzed using the Fluidigm real-time PCR analysis software v.4.1.3.
Firstly, 6.5 ng of each cDNA were initially preamplified (10 min 95°C activation and 14 PCR cycles (15 sec 95°C and 4 min 60°C) with PreAmp Master Mix (100-5581, Fluidigm) and a pool containing all the primers targeting all the genes (200nM), excluding the 16S rRNA primer sets. Preamplified sample were diluted at 1/5 after an exonuclease treatment (M02935, NEB). In order to prepare samples for loading into the integrated fluidic circuits (IFC), a mix was prepared consisting of 440 μL 2X TaqMan Master Mix (Applied Biosystem, 4369016), 44 μL 20× DNA Binding Dye Sample Loading Reagent (100-7609, Fluidigm), 44µl 20X Evagreen (31000, Biotium) plus 132 μL TE, and 6 μL of this mix was dispensed to each well of a 96-well assay plate. 2µL of preamplified and diluted cDNA sample was added to each well and the plate was briefly vortexed and centrifuged. For the assays, 5 μL of each Assay (5 μM each primer in primer-mix (2X assay loading reagent (100-7611, Fluidigm) and Tris EDTA) were dispensed to each Detector Inlet of the 96.96 IFC. Following priming of the IFC in the IFC Controller HX, 5 μL of the cDNA sample + reagent mix and 5 µl of Assay were dispensed to each Sample Inlet of the 96.96 IFC. After loading the assays and samples into the IFC in the IFC Controller HX, the IFC was transferred to the BioMark and PCR was performed using the following thermal protocol : Thermal Mix of 50 °C, 2 min; 70 °C, 30min; 25°C, 10min, Hot Start at 50 °C, 2 min; 95°C, 10 min, PCR Cycle of 35 cycles of (95 °C, 15 s; 60 °C, 60 s), and Melting analysis (60°C, 30s; 95°C,1°C/3s). Results were analyzed using the Fluidigm real-time PCR analysis software v.4.1.3.
Data pre-processing
The first part of the analysis is to clean up the data with the Fluidigm real-time PCR analysis software v.4.1.3. Data were pre-processed for expression analysis as follows: the cycle threshold (Ct) values registered from amplifications that generated melting curves with aberrant Tm (melting temperature) or with products giving rise to a double peak in melting curves (corresponding to a mixture of expected and aberrant PCR products) were removed.
Gene expression analysis
The selectHKgenes function with the "Vandesompele" method of the SLqPCR package was used with RStudio (Version 1.2.1335) to choose the five most stable housekeeping genes. The five housekeeping genes for the relative quantification of mRNA levels of target genes were SDHA, GLUT8, PDHA1, POL2 and Luciferase. Luciferase is an exogenous RNA (Promega), added to each sample during the reverse transcription (100 pg per inch) to allow normalization of the data, as previously described (52,53). The slope of a standard curve using serial dilutions of cDNA measured the efficiency (E) of PCR. In all cases, PCR efficiency values ranged between 1.90 and 2. The analyses were done with RStudio (54,55) with:
[Please see the supplementary files section to view the equations.]
Statistical analysis
Statistical analyses were done using the Graphpad Prism version 8 for Windows (GraphPad software, La Jolla California USA, www.graphaapd.com (serial number GP8-1598457-RJQD-5E2EC)). Data are presented with a box-and-whisker plot, boxes ranging from the 25th to the 75th percentiles, and whiskers ranging from the lowest to the highest value. When the data set presented a Normal distribution (assessed by Shapiro–Wilk test), parametric variance analysis (ANOVA) was performed followed by a Dunnett’s multiple comparison test as post hoc analysis. When normal distribution was not demonstrated, the Kruskal–Wallis non-parametric test was performed followed by a Dunn’s test as post hoc analysis. In every case, differences between the groups were considered statistically significant if the value of P < 0.05.
The heatmap.2 function from the gplots package was used to draw heatmaps with RStudio. The corrplot package was used to draw the correlation matrix; this package contains algorithms to reorder the matrix according to the degree of correlation between the variables.