bioRxiv Science⌕ Search

bioRxiv · 10.64898/2026.10.20.760594

Cytoplasmic genomes change allelic expression bias in plant seeds independently of parental imprinting: evidence for cytonuclear regulatory epistasis in Arabidopsis thaliana

Abstract

Cytoplasmic genomes play critical roles in plant cells, yet their potential contribution to regulation of the nuclear genome remains largely unexplored. Here, we examine how cytoplasmic genotype affects allelic dominance at nuclear loci by identifying allele-specific expression in hybrid plant lines in which cytoplasmic genomes have been swapped between accessions. After developing a bias-free ASE analysis pipeline based on dual pseudo-genome alignment, we determined ASE across 48 RNA-seq datasets of the model plant, Arabidopsis thaliana and distinguished between allelic bias under the control of the genotype and or the cytoplasm. Of 7,997 nuclear genes for which informative SNPs could be mapped, 2,702 displayed allelic imbalances, the great majority controlled by the nuclear background. However cytoplasmic interactions with nuclear genotype were found to regulate ASE at 24 genes. We also tested interactions with genomic imprinting, a specific form of ASE in which parental inheritance determines expression state, within seed transcriptomes: while we detected both novel and previously-reported imprinted genes, only three genes (VFB1, ANNAT3, ENODL2) exhibited both parent-of-origin bias and cytoplasm-responsiveness. This indicates that these two parental inheritance mechanisms are largely independent, with implications for maternal-cytoplasmic coadaptation theories. Interestingly, genes subject to cytoplasmic ASE were enriched in the ascorbate and aldarate metabolism pathway, with the highest ASE biases found in GAE4 and UGD3. We conclude that cytoplasmic variation can selectively alter expression of nuclear-encoded alleles and that this may represent a novel source of control of cellular energy balance and metabolic signalling pathways.

Explore related subjects

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

McKeown, P. C., Motienoparvar, P., Lai, L., Gondalia, N., Dupouy, G., Fort, A., Spillane, C.. 2026-10-26. Cytoplasmic genomes change allelic expression bias in plant seeds independently of parental imprinting: evidence for cytonuclear regulatory epistasis in Arabidopsis thaliana. https://doi.org/10.64898/2026.10.20.760594

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related preprints

Mapping, stacking and characterization of a novel resistance haplotype against GpaVvrn-virulent Globodera pallida derived from wild potato relative Solanum brevicaule

Potato production is threatened by pests and pathogens. Among the most harmful are the potato cyst nematodes (PCN), including Globodera pallida. The major control measures are potato cultivars harbouring resistance genes. However, virulent nematode populations are breaking through the most broadly applied resistance genes, and therefore novel resistance genes are required to keep control of G. pallida. In this study we map a resistance locus in wild potato relative Solanum brevicaule to the bottom of chromosome 11 (42.83 - 42.85 Mb) using a k-mer based bulked segregant analysis and name the resistant haplotype GpaXIbrc. The presence of the GpaXIbrc marker in the potato genome significantly increases resistance against a GpaVvrn-virulent G. pallida population (AMPOP01). Furthermore, the GpaXIbrc haplotype was stacked with three other resistance haplotypes: GpaXItar, GpaVvrn, and GpaIXvrn. Stacking of the novel GpaXIbrc haplotype to these other resistance haplotypes resulted in a significant increase in resistance. The largest effect was seen in combinations with 'major' resistance haplotypes GpaVvrn + GpaIXvrn and GpaXItar + GpaXIbrc. To determine whether we are dealing with a novel allele, the resistance mechanism underlying the GpaXIbrc haplotype was described after performing an in vitro bioassay. The resistance mechanism was different from GpaXIspl as described in literature, which is also localized at the bottom of chromosome 11. Therefore, we conclude that we found a novel resistance allele originating from S. brevicaule.

genetics↗

Multiple domains of Mata1 transcription factor contribute to binding site specificity

Among homologous transcription factors with the same function, DNA-binding specificity can diverge. Mat1 is a high mobility group (HMG) box transcription factor in budding yeast whose Saccharomyces cerevisiae and Candida albicans homologs regulate -specific genes but bind distinct cis-regulatory sequences. To identify the protein features underlying this divergence, we generated structure-guided chimeras and measured their activity using species-specific reporters. Replacing any of the four Mat1 regions reduced activation of the S. cerevisiae reporter, while all chimeras retained activity on the C. albicans reporter. Thus, all four regions of Mat - not just the HMG box domain - are important for sequence specificity.

genetics↗

A multi-tissue RNA-seq atlas of adult medaka (Oryzias latipes) under salinity and winter-summer sampling contrasts

Medaka (Oryzias latipes) is a euryhaline and eurythermal teleost widely used in genetics and environmental biology. We generated an RNA-sequencing atlas of 11 adult tissues (brain, eye, gill, heart, intestine, kidney, liver, muscle, spleen, ovary, testis) of the strain d-rR/TOKYO, sampled under two independent environmental contrasts: salinity (freshwater versus seawater) and season (winter versus summer). The freshwater and summer groups, reared indoors and outdoors under long days, provide a third contrast between laboratory and outdoor rearing. Typically, one male and one female were sampled per condition, with ten tissues taken from each individual. Ninety libraries were quantified against the medaka RefSeq transcript set (GCF_053564925.1). The resource comprises gene-level TPM and count matrices, differential expression tables for each tissue and contrast, tissue-specificity indices, human ortholog annotations, sample metadata, and a self-contained interactive browser. Because replication is limited and each condition was sampled on its own date, the differential expression tables are intended for hypothesis generation. The dataset supports studies of gene-expression plasticity and comparisons between medaka and humans.

genetics↗