Next-Generation Sequencing Techniques Reveal that Genomic Imprinting Is Absent in Day-Old Gallus gallus domesticus Brains

Abstract

Genomic imprinting is a phenomenon characterized by parent-of-origin-specific gene expression. While widely documented in viviparous mammals and plants, imprinting in oviparous birds remains controversial. Because genomic imprinting is temporal- and tissue-specific, we investigated this phenomenon only in the brain tissues of 1-day-old chickens (Gallus gallus). We used next-generation sequencing technology to compare four transcriptomes pooled from 11 chickens, generated from reciprocally crossed families, to the DNA sequences of their parents. Candidate imprinted genes were then selected from these sequence alignments and subjected to verification experiments that excluded all but one SNP. Subsequent experiments performed with two new sets of reciprocally crossed families resulted in the exclusion of that candidate SNP as well. Attempts to find evidence of genomic imprinting from long non-coding RNAs yielded negative results. We therefore conclude that genomic imprinting is absent in the brains of 1-day-old chickens. However, due to the temporal and tissue specificity of imprinting, our results cannot be extended to all growth stages and tissue types.

Fig 1. Reciprocal design.

We reciprocally crossed two inbred strains, Cornish (Cor) and White Leghorn (WL), to generate progenies. For each cross, the brains of 1-day-old chickens from each sex were collected and total RNA was extracted from the tissue samples. Same-sex RNA samples from the same family were pooled to establish an RNA-seq library. The male and female pools within the same family were regarded as biological replicates. DNA samples of four parents from each family were used for whole-genome re-sequencing.

Fig 2. Ratio of reads aligned to parental genomes of all SNPs.

Points in this figure reflect the ratio of reads aligned to the maternal versus the paternal genomes under the loose standard (SNP sequencing depth greater than 2×). The horizontal axis is log₂ (m:f) in cross I, while the vertical axis is log₂ (m:f) in cross II. Red points indicate the candidate maternal imprinted SNPs, while the blue points indicate candidate paternal imprinted SNPs.

Fig 3. Verification experiments excluded most of the candidate imprinted SNPs.

(A) Restriction endonuclease analysis. The cDNA pattern was similar to offspring DNA, which contained two bands of the same brightness (No. 4, gene XLOC_046149). (B) SNPs filtered using loose standards (sequencing depth > 2×) show heterozygosity in some parents (No. 5, ENSGALG00000000194). We designed different primers for our cDNA analysis versus our DNA analysis, due to the presence of an intron region near some SNPs in the DNA. (C) Direct Sanger sequencing of parental DNA confirming the heterozygosity of some SNPs. Sequencing results for gene XLOC_046722 (No. 8) are shown; the SNP is the third base. (D) Direct Sanger sequencing of offspring DNA and cDNA. Expression of some genes does not show evidence of imprinting. Sequencing results for gene ZSWIM7 (No. 13) are shown; the SNP is the third base. (E) Pyrosequencing results reveal that the gene expression patterns of most candidate genes do not show evidence of imprinting. Results for the cDNA of gene ACAA1 (No. 12) are shown.

Fig 4. Results of further verification experiments on the SNP most likely to exhibit imprinting.

The sequence near the SNP is CTCCCA/GAACGC. (A) This gene appeared to be expressed in a way indicative of genomic imprinting when examined with direct Sanger sequencing. However, the parent-of-origin characteristics were exhibited only in females. The relevant SNP is the third base. Differences between parent and offspring sequences in the first base are due to the introduction of mismatch bases in the offspring for dCAPS assays. (B) Pyrosequencing results are consistent with direct Sanger sequencing results. (C) Cross III and Cross IV, two different reciprocally crossed families, had parents homozygous at that SNP locus. This is in accordance with data from the previous four parents and was confirmed by Sanger sequencing. We detected the brain cDNA of offspring individually, using pyrosequencing. Neither males nor females in crosses III and IV exhibited parent-origin-specific gene expression patterns.