New insights into the molecular control of plant endoreduplication

Cells that copy their genomes over and over without ever dividing may sound like a recipe for genomic disaster, but in plants this process, known as endoreduplication, is one of the most widespread and productive developmental programs in nature. It is the reason a single Arabidopsis leaf hair can balloon into a giant branched cell, why tomato fruits swell to edible sizes, and how tissues can grow when cell division is no longer an option. A new review published in Plant Molecular Biology by Takashi Ishida of Kumamoto University synthesizes decades of work into a coherent conceptual framework, arguing that plant endoreduplication is best understood as a regulated “endocycle oscillator,” a periodic control logic that drives repeated rounds of DNA replication while reliably bypassing mitosis.

The core idea is elegant in its simplicity. In the canonical mitotic cell cycle, a cell passes through G1, replicates its DNA in S phase, prepares in G2, and then enters mitosis, the M phase, before dividing. Endoreduplication severs the link between the first half of this cycle and the second. G1/S progression, the machinery that licenses and executes DNA replication, continues to run, but M-phase entry is suppressed. The result is a cell that doubles its nuclear DNA content with each pass, cycling from 2C to 4C to 8C and beyond, a state called endopolyploidy. Ishida emphasizes that this is distinct from endomitosis, where a cell enters but fails to complete mitosis, and from chemically induced polyploidization; endoreduplication arises from an intrinsic reconfiguration of the cell cycle control network itself.

Much of the conceptual scaffolding for this field was built in animal systems, particularly in the fruit fly Drosophila, whose larval tissues are famously polytene. Studies there established that a functional endocycle requires an oscillator: a set of feedback loops that alternately activate S phase and suppress mitosis, period after period. The critical question the review poses is how far these principles map onto plants, whose cell cycle machinery, while conserved in outline, is regulated by a distinctly plant-specific cast of characters. The available evidence suggests the logic is transferable, but the implementation in angiosperms has its own architecture, organized around two pillars: repeated S-phase entry and robust mitotic suppression.

The first pillar depends on DNA replication licensing. Before any genome can be copied, the origins of replication must be loaded with the pre-replicative complex, a process involving the origin recognition complex (ORC), CDC6, and CDT1. In Arabidopsis, the expression and stability of these licensing factors are tightly tied to endoreduplication competence. CDC6, for example, is an E2F transcription factor target whose accumulation correlates with endocycling cells, and the ORC1 genes are expressed differently in proliferating versus endoreduplicating tissues. F-box protein FBL17 drives the degradation of the licensing inhibitor CDT1a in early S phase, preventing genome instability, while Cullin 4-based ubiquitin ligases shape the timing of licensing cycles in trichomes. On the enzymatic side, DNA topoisomerase VI, along with accessory components such as BIN4, RHL1, and MIDGET, is essential; mutants in this complex cannot sustain endoreduplication, presumably because untangling and decatenating ever-larger replicated genomes becomes impossible without it. Together these findings portray repeated genome duplication as a carefully staged licensing problem that the endocycle must solve anew with every round.

The second pillar is the suppression of mitosis, and here the anaphase-promoting complex/cyclosome (APC/C) takes center stage. This E3 ubiquitin ligase, when activated by the CCS52A proteins, targets mitotic cyclins for destruction and locks cells out of M phase. Loss-of-function mutations in CCS52A1 restore mitotic cycling in cells that should have become endocycling, underscoring its role as a gatekeeper. In parallel, the B-type cyclin-dependent kinases, CDKB1 complexes paired with A-type cyclins such as CYCA2;3, actively suppress endocycle onset; their inactivation permits the transition to endoreduplication. The plant-specific CDK inhibitor SIAMESE (SIM), named for the multi-headed trichomes produced by its mutants, targets both CDKA and CDKB complexes to establish the endocycle in trichomes, while its relatives, the SIAMESE-RELATED (SMR) proteins, extend this control into stress responses. Another layer of mitotic braking comes from DEL1, a DP-E2F-like transcriptional repressor that keeps the APC/C activator CCS52A1 in check, and from APC/C inhibitors UVI4 and GIGAS CELL1, which preserve mitotic competence where division is still needed. Deubiquitinating enzymes add further tuning: the UBP14–CDKB1;1–CDKG2 cascade illustrates how a proteolytic regulator, a mitotic kinase, and a CDK-like protein can be wired into a single pathway governing ploidy and growth.

What makes the endocycle oscillator concept useful is that it frames these components as elements of a periodic system rather than a one-way switch. For a cell to cycle through repeated S phases, the machinery must reset: licensing factors must be reloaded, inhibitors cleared, and mitotic suppressors re-established in the right sequence. Feedback loops generating this periodicity are well defined in Drosophila, but in plants, Ishida argues, they remain incompletely mapped. Identifying the loops that generate oscillation within the plant endocycle is one of the field’s central open problems, and resolving it will likely require moving beyond static measurements of ploidy toward dynamic, system-level analyses that track the cell cycle machinery in living tissues over time.

Remarkably, this core oscillator does not operate in isolation. A diverse array of upstream signals converges on it, allowing plants to tune ploidy in response to their development and environment. Light is a classic example: phytochrome photoreceptors control the number of endoreduplication cycles in the Arabidopsis hypocotyl, linking photomorphogenesis directly to ploidy. Phytohormones provide another route. Cytokinin promotes endocycle onset in roots by inducing expression of the APC/C activator CCS52A1, while auxin modulates the transition from the mitotic cycle to the endocycle, and single-cell RNA sequencing has revealed auxin fluctuation during the endocycle itself. D-type cyclins of the CYCD3 class, which respond to cytokinin, act as rate-limiting factors that keep cells in the mitotic cycle, so their regulation effectively gates entry into endoreduplication.

Stress and DNA damage add yet another dimension. When Arabidopsis experiences DNA double-strand breaks, it can respond by deliberately inducing endoreduplication, a programmed shift mediated by the DNA damage response. The ATM kinase phosphorylates the transcription factor SOG1, which can trigger early onset of endoreduplication by inducing CCS52A1 expression in roots, while the SMR5 and SMR7 CDK inhibitors regulate the DNA damage checkpoint in response to reactive oxygen species. Mechanical signals participate as well; recent work shows that endoreplication mediates cell size control through mechanochemical signaling from the cell wall, and DELLA proteins, the growth-repressing components of gibberellin signaling, mediate stress-induced cell differentiation by modulating APC/C activity. The transcriptional layer is equally rich: TCP family transcription factors, including TCP14 and TCP15, modulate cell-cycle gene expression and are themselves regulated by ubiquitin receptors DA1, DAR1, and DAR2, while the trihelix factor GTL1 and the fluctuating epidermal regulator ATML1 shape ploidy-dependent patterns of giant cells in leaves and sepals.

Why does all this matter beyond the mechanics? Endoreduplication is increasingly recognized as a flexible mechanism supporting developmental plasticity. The relationship between ploidy and cell size is one of the oldest observations in plant biology, with the karyoplasmic ratio theory invoked to explain fruit growth in tomato, and it underlies the striking mosaics of giant and small pavement cells that pattern the leaf surface. Because endoreduplication allows cells to grow large without division, it provides a rapid, energetically efficient way to build organs, and its responsiveness to light, hormones, and stress suggests it functions as an integration point where environmental information is converted into cellular architecture. Polyploid cells may also buffer genome stress, offering a survival strategy under adverse conditions.

The review closes with a forward-looking agenda. Two priorities stand out: first, experimentally identifying the feedback loops that generate periodicity in the plant endocycle oscillator, work that will benefit from live-cell reporters of cell cycle phase and from quantitative imaging approaches capable of inferring spatial ploidy patterns across tissues; and second, advancing from static ploidy snapshots to dynamic, context-resolved analyses spanning developmental, environmental, and stress-related conditions. Such tools, including fluorescent cell cycle sensors and time-series single-cell transcriptomics, are already available in Arabidopsis and are poised to reveal how the oscillator behaves in real time. As Ishida’s synthesis makes clear, plants have turned an apparent cell cycle accident into a precision instrument, and decoding its internal clock promises to explain how one of biology’s most unusual cell cycles is truly run.