Maize’s extra chromosomes are active in pollen—and may echo across the genome

A chromosome is usually thought of as part of an organism’s genome, copied and passed on through reproduction. Maize B chromosomes complicate that picture. They are additional chromosomes, separate from the standard A chromosomes, and can gain an inheritance advantage through chromosome drive. During pollen development, their sister copies can fail to separate at a cell division; later, sperm carrying B chromosomes are preferentially involved in fertilization. The precise molecular details of this process remain unresolved.

The study asks a related question: are B chromosomes active in mature pollen, and are molecular differences associated with their presence also detectable on standard chromosomes? The researchers compared maize inbred line B73 with no B chromosomes (0B) and with two (2B). For ATAC-seq, they isolated pollen nuclei using flow cytometry, which sorts particles by measured properties. ATAC-seq identifies DNA regions that are relatively accessible to molecular machinery. Separately, they extracted RNA from pollen for RNA-seq, which measures gene transcripts and helps indicate which genes are expressed.

Each condition had three biological replicates. For ATAC-seq, the team sorted 50,000 pollen nuclei per sample. The RNA-seq measurements, however, came from pollen—not from those sorted nuclei. The study therefore combines chromatin-accessibility data from sorted nuclei with gene-expression data from pollen samples. Neither assay separately examined B-carrying sperm during fertilization.

The researchers detected transcripts from 103 of the 689 predicted B-chromosome genes in mature pollen. Their differential-expression analysis identified 88 upregulated genes on the B chromosome in the 2B comparison. This result is inherently one-sided: 0B plants lack B chromosomes, so B-linked genes cannot be meaningfully classified as downregulated relative to their presence in 2B plants. On the standard A chromosomes, the analysis found 42 upregulated genes and three downregulated genes in 2B samples. These are differences between conditions; they do not establish that B chromosomes directly switch particular A-chromosome genes on or off.

The accessibility data add another layer. Most accessible regions were shared between conditions, consistent with the samples’ common pollen identity. The analysis identified 2,905 regions more accessible in 2B samples and 12 less accessible. Most regions classified as more accessible were on the B chromosome. That needs careful interpretation: 0B plants do not contain B chromosomes, so B-linked regions can appear as newly present and accessible in 2B samples. They cannot be compared as equivalent DNA regions in both groups. The authors note that this large B-chromosome signal is not evidence of a genome-wide reshuffling of accessibility.

There were also signals on the standard chromosomes. The team reported 203 regions more accessible and 12 less accessible on A chromosomes. When they compared accessibility and expression results, 87 genes showed both increased accessibility and higher transcript abundance; 83 were B-linked and four were on A chromosomes. This overlap is consistent with accessible DNA and gene activity occurring together at some loci. It does not show that accessibility caused the expression differences, or explain how B chromosomes might be connected to the four A-chromosome genes.

The clearest finding is that B chromosomes are not simply silent passengers in mature maize pollen: they contain accessible regions and expressed genes. Evidence for effects beyond the B chromosome is more limited, and the authors describe the influence on the A genome as comparatively modest. Accessibility alone also did not account for every expression difference, leaving other regulatory processes to investigate.

The results do not show that B-chromosome genes cause chromosome drive, improve pollen performance, or determine which sperm fertilizes the egg. The study compares one maize genetic background with zero versus two B chromosomes, using three replicates per condition, and measures molecular patterns rather than fertilization outcomes. Further work is needed to test specific genes and mechanisms and to establish how broadly the results apply across maize backgrounds or reproductive stages.

This report is a bioRxiv preprint posted in November 2026 and is identified as not certified by peer review. Its conclusions should be considered provisional. The authors provide sequencing data and supplementary analyses, but independent evaluation and follow-up experiments will be important in assessing the findings.