Comparative mitogenome assembly and structural evolution across Perilla frutescens and closely related taxa using whole-genome sequencing data

In the quiet corners of a cell, far from the nucleus where the main genetic instructions are kept, lie two tiny powerhouses: the mitochondria and the chloroplasts. These organelles have their own separate sets of DNA, remnants of ancient bacteria that once lived independently before merging with the first plant cells. While the DNA in the nucleus acts like a master blueprint for the whole organism, the DNA inside these powerhouses is more like a specialized toolkit, focused on energy production and photosynthesis. In plants, this mitochondrial DNA is particularly mysterious. Unlike the neat, linear chromosomes found in the nucleus, plant mitochondrial genomes are often messy, shuffling their pieces around, swapping parts with their neighbors, and changing size dramatically from one species to another. This structural chaos makes them difficult to study, yet they hold vital clues about how plants evolve, how they adapt to their environments, and how different species are related to one another. For scientists trying to improve crops or understand plant history, untangling this genetic knot is essential, but it requires looking at the data with fresh eyes and powerful new tools.

Perilla, a fragrant plant widely grown in East Asia for its culinary and medicinal uses, has long been a subject of interest for researchers. While scientists have already mapped out the plant's main nuclear genome and its chloroplast DNA, the story of its mitochondrial DNA remained incomplete. Previous studies had looked at Perilla in isolation, but no one had systematically compared its mitochondrial genome to those of its closest living relatives. This gap left researchers wondering: is the unique structure of Perilla's mitochondrial DNA a specific trait of the plant itself, or is it a shared feature among its family? To answer this, a team of researchers turned to a group of plants that includes Perilla and its close kin, such as the genus Mosla and the species Keiskea sinensis. By using advanced sequencing technology that reads long, continuous stretches of DNA, they were able to assemble the complete mitochondrial genomes of these five plants and compare them side by side, revealing a landscape of genetic stability mixed with surprising structural fluidity.

The researchers began by gathering fresh plant samples from the field and extracting their DNA. They then used a high-precision sequencing method known as PacBio HiFi, which generates long, accurate reads that can span the complex, repetitive regions that often confuse older sequencing techniques. This allowed them to build complete, gap-free maps of the mitochondrial DNA for Perilla frutescens, Keiskea sinensis, and three varieties of Mosla. Once these maps were assembled, the team examined the basic building blocks. They found that while the total size of these mitochondrial genomes varied significantly—ranging from about 300,000 to nearly 350,000 letters of genetic code—the actual number of genes they contained was remarkably consistent. Every plant carried the same core set of instructions for making energy, including genes for the machinery that powers the cell's respiration. The chemical composition of the DNA, specifically the balance of its building blocks, also stayed nearly the same across all five species. This suggested that the fundamental biological engine of these plants had not changed much, even as the surrounding genetic landscape shifted.

However, when the researchers looked beyond the genes to the spaces between them, the picture changed dramatically. The differences between the species were not in the genes themselves, but in the repetitive sequences and the overall arrangement of the DNA. The team discovered that these genomes were filled with scattered repeats, stretches of DNA that appeared in multiple places, acting like glue that could hold the genome together or cause it to rearrange. In some species, these repeats made up a small fraction of the genome, while in others, they occupied a much larger portion. When the researchers compared the order of the DNA blocks between Perilla and its relatives, they found that while the pieces were often the same, their arrangement was frequently different. Some sections were flipped upside down, and others were shuffled into new positions. This structural plasticity meant that two plants could have the same set of tools but arrange them in completely different ways, creating a unique genetic architecture for each species.

The study also uncovered evidence of genetic borrowing between the cell's different compartments. Plants are unique in that their mitochondria sometimes steal small pieces of DNA from their chloroplasts, the organelles responsible for photosynthesis. The researchers found that all five plants carried these stolen fragments, known as mitochondrial plastid DNA sequences, but the number and size of these fragments varied. In Perilla, for instance, there were twelve such fragments, ranging in size from a few hundred to over a thousand letters of code. Some of these stolen pieces landed right on top of existing genes, while others sat in the empty spaces. This suggests that the process of DNA transfer is an ongoing, dynamic event in these plants, contributing to the structural diversity seen across the group. Furthermore, the researchers looked at how the DNA is edited after it is copied, a process where specific letters in the genetic code are changed to ensure the proteins work correctly. They predicted hundreds of these editing sites in each plant, with the most common change being the swapping of one letter for another. While the total number of these sites varied, the pattern of where they occurred was largely similar, pointing to a shared mechanism for maintaining the integrity of the genetic code.

To understand how these genetic differences fit into the broader picture of evolution, the researchers built family trees based on both the mitochondrial and chloroplast DNA. Surprisingly, the trees drawn from the two different organelles told the same story for the close relatives they were studying: Perilla, Keiskea, and the Mosla species formed a tight-knit group. This agreement gave the researchers confidence that their genetic maps were accurate and that these plants are indeed closely related. However, the trees also hinted that at a broader level, the evolutionary history of these plants might be more complex, with different parts of the genome telling slightly different stories. This discordance is common in plants and often results from events like hybridization or the swapping of genetic material between species. The key takeaway from this study is that while the core functional genes of these plants have remained stable over time, the structural landscape surrounding them has been in a constant state of flux.

The implications of these findings extend beyond simple curiosity. For breeders and scientists working to improve crops like Perilla, understanding the structural variation in mitochondrial DNA provides new markers to distinguish between different varieties and to track the history of germplasm. The regions where the DNA is rearranged or where foreign DNA has been inserted offer a rich source of variation that could be linked to important traits, such as stress resistance or oil production. While the study did not prove that these structural differences directly cause specific traits, it identified the precise locations where such differences exist, setting the stage for future research. By mapping the complete mitochondrial genomes of these five plants, the researchers have provided a detailed reference that separates the shared, conserved features of the group from the unique, variable traits of each species. This work transforms our understanding of plant mitochondrial evolution, showing that even in a group of closely related plants, the genome is not a static map but a dynamic, shifting landscape where the arrangement of pieces is just as important as the pieces themselves.