Skip to main content
NCBI home page
Proceedings of the National Academy of Sciences of the United States of America logoLink to Proceedings of the National Academy of Sciences of the United States of America
. 2026 May 26;123(22):e2522067123. doi: 10.1073/pnas.2522067123

Epithelial cell fusion is required for tissue repair following UV-A irradiation

Minqi Shen a, Lillie G Mitchell a, Lydia W Boer a, Lydia M Bischoff a, Vicki P Losick a,1
PMCID: PMC13229259  PMID: 42189976

Significance

Polyploid cells are life’s stress responders as the more than doubling of a cell’s genome has been shown to enable resistance to genotoxic stress. Cells are exposed to various sources of genotoxic stress, including ultraviolet light. Here, we find that ultraviolet light induces DNA damage, causing apoptosis and the subsequent generation of giant, multinucleated polyploid cells in the fruit fly epithelium. Unlike in other models studied to date, cell fusion is the predominant response to genotoxic stress and appears to be essential for tissue repair. This study also determines that polyploidization post UV stress is dependent on caspase activation, suggesting a conserved mechanism to initiate cellular multinucleation both in development as well as in response to life’s stressors.

Keywords: polyploidy, genotoxic stress, cell fusion, tissue repair, Drosophila

Abstract

Cell cycle–dependent and independent mechanisms lead to the generation of mononucleated and multinucleated polyploid cells. The more than doubling of a cell’s nuclear genome by endoreplication has been found to be an adaptation to genotoxic stress, enabling cell survival despite DNA damage. However, it remains unknown whether cells that increase ploidy via multinucleation also arise in response to genotoxic stress. Here, we use ultraviolet light A (UV-A) to induce permanent DNA damage in cells within the adult fruit fly epithelium. UV-A irradiation causes an injury-like response where giant multinucleated, polyploid cells arise following cell death. The epithelial cells undergo endoreplication, which is required to compensate for cell loss, but is surprisingly dispensable for tissue repair. UV-A irradiation also induces cell fusion, which generates multinucleated cells that encompass almost the entire epithelial area post injury. Cell fusion can be inhibited by expression of a dominant negative Rac or Cdc42 GTPase, which then blocks epithelial tissue repair postirradiation. Apoptotic nuclei were detected at the site of cell junction breakdown, suggesting that apoptosis itself or an apoptotic signal is required for polyploidization in this model. Expression of the effector caspase inhibitor, p35, led to inhibition of apoptosis, the endocycle, and cell fusion post UV-A. Therefore, we have found that caspase activation is necessary for polyploidization post injury and enhancing cell ploidy via multinucleation is another strategy to enable cell survival and tissue repair following genotoxic stress.

All organisms undergo various forms of tissue damage throughout their lifetimes from injury, irradiation, and/or toxins. Tissues must compensate for damage-induced cell loss to restore tissue integrity and physiology. In cases where cell division is limited, polyploid cell growth has been shown to be a driver of tissue repair (1, 2). Polyploidy is the more than doubling of a cell’s diploid genome, which can arise by an incomplete cell cycle, known as endoreplication, or by cell fusion. The increase in cell size and DNA content is sufficient to restore tissue mass in several animal tissues studied to date, including the mouse kidney and liver, where cells become polyploid through endoreplication to restore organ mass postinjury (3, 4). Likewise, several tissues in the fruit fly, Drosophila melanogaster, rely on polyploidization to compensate for cell loss and restore organ function (5–8). The genetic tractability of the fruit fly has made it a useful model to elucidate the role and regulation of polyploidy in tissue repair.

Polyploid cells arise by endoreplication in most fly tissues studied to date, except the cuticular epithelium, in which polyploid cells are generated by both cell fusion and the endocycle (6, 9–11). Thus, the Drosophila epithelium is a beneficial model to find how stress induces different mechanisms of polyploidization and, in turn, how different ploidy states (i.e., multinucleated versus mononucleated polyploid cells) enable adaptation to a particular stress. The epithelium in the adult fruit fly is composed of a monolayer of postmitotic diploid cells that, when stressed by a puncture injury, will generate multinucleated polyploid cells. These polyploid cells are advantageous for tissue repair as they enable healing in the presence of genotoxic stress (12).

Previous studies proved that endocycling cells in the Drosophila ovary are resistant to DNA damage caused by gamma irradiation (13). The p53-dependent DNA damage response pathway was epigenetically silenced, hence, the polyploid cells could accumulate double-stranded DNA breaks and not undergo apoptosis postirradiation. Polyploid neurons in the adult fly brain have also been shown to survive UV-induced DNA damage with age (14). Additionally, there is evidence from cancer studies that polyploidy may enable resistance against genotoxic stress (15, 16). However, it is unknown to what extent the different mechanisms of polyploid cell generation (i.e., cell fusion, endocycle, or endomitosis) contribute to genotoxic stress resistance.

Here, we establish an ultraviolet light (UV) injury model in the adult fruit fly where epithelial cells are known to become polyploid in response to stress (6). Irradiating the fly’s ventral abdomen with UV-A (a physiological source of DNA damage) at a low dose induces apoptosis of the postmitotic diploid cells. While roughly half the cells die, the other cells are able to survive and heal the epithelium within 7 d. Similar to a local injury (i.e., needle puncture wound), the epithelial cells both endoreplicate and fuse, although cell fusion, but not endoreplication, is required for tissue repair following UV-A irradiation. In addition, we find that irradiation-induced apoptosis is required to initiate both cell fusion and the endocycle as expression of effector caspase inhibitor, p35, is sufficient to block cell death and suppress polyploidization. In conclusion, we have found that UV-A irradiation can induce cell fusion, generating giant multinucleated, polyploid cells that are essential for cell survival and healing in response to this genotoxic stress.

Results

Giant Multinucleated, Polyploid Cells Are Generated in the Fly Epithelium in Response to UV-A Irradiation.

To examine how polyploid cells respond and contribute to tissue repair following genotoxic stress in the postmitotic cells, we used a fruit fly model of UV-A induced injury. To do so, adult fruit flies were irradiated with a low dose (25 mJ) of UV-A and the abdominal tissue was dissected, fixed, and immunostained with epithelial-specific antibodies to detect cell junctions (FasIII, a septate junction protein) and epithelial nuclei (GFP, epithelial-specific expression (epi-Gal4>nlsGFP)) revealing the epithelial tissue organization in uninjured (0 mJ) and days post (dp) UV-A irradiation (Fig. 1A). We first validated that UV-A irradiation was sufficient to induce DNA damage by immunostaining for gamma phospho-Histone 2Av (γH2Av), which binds the genome at sites of double-stranded DNA breaks (SI Appendix, Fig. S1) (17). As expected, the uninjured tissue had low, but detectable γH2Av-labeled foci in both epithelial and muscle cell nuclei (SI Appendix, Fig. S1B) (12, 18). By 2 dp UV-A, we observed a twofold increase in the nuclear γH2Av intensity compared to unirradiated cells (SI Appendix, Fig. S1 B and C). The γH2Av expression persisted in the UV-A irradiated epithelial nuclei, suggesting that DNA damage is not repaired even by 10 dp UV-A. A high dose (50 mJ) of UV-A irradiation was also tested in the adult flies and led to a dosage-dependent effect. The tissue showed a higher level of γH2Av within epithelial nuclei at 2 dp when compared to the 25 mJ treated samples (SI Appendix, Fig. S1D). However, the 50 mJ irradiated flies exhibited more severe tissue damage, which disrupted the tissue during dissection and therefore could not be used for further analysis.

Fig. 1.

Figure shows immunofluorescent images of epithelial cells pre- and post UV-A. Graphs show nuclei count, multinucleation, and nuclear ploidy.

Open in a new tab

UV-A-induced polyploidy in the fly epithelium. (A) Immunofluorescent images of the fly epithelium taken in uninjured (0 mJ) and dp UV-A irradiation (25 mJ). Cells are stained for septate junctions (FasIII, magenta) and epithelial nuclei (nGFP, green). Marked are examples of discontinuous (dashed arrows) and continuous (solid arrows) cell junctions and enlarged nuclei (arrowheads). Quantification of epithelial (B) nuclear number, (C) multinucleation at 10 dp, and (D) epithelial nuclear ploidy. Data represent the SEM with significance determined by One-way ANOVA or Student t test (C) with P-value *<0.05, **<0.01, ***<0.001, and ****<0.0001.

Next, we examined how UV-A irradiation affected the epithelium in terms of the number of nuclei, multinucleation, and ploidy, as genotoxic stress can lead to apoptosis (Fig. 1) (19). The UV-A irradiated epithelium started to show nuclear loss at 2 dp and continued to a ~40% reduction in the number of epithelial nuclei by 5 dp. The number of epithelial nuclei then remained steady with no further reduction at 7 and 10 dp, even though nuclear DNA damage (γH2Av nuclear foci) persisted (Fig. 1 A and B and SI Appendix, Fig. S1B). The significant reduction in epithelial nuclear number led to a noticeable change in epithelial organization as the septate junctions were disrupted (sparse and discontinuous) by 5 dp UV-A (Fig. 1A, dashed arrowheads). By 7 dp, the epithelial septate junctions were restored and more continuously outlined very large multinucleated cells. The multinucleated cells often extended beyond the image area and so, we quantified the total area occupied by multinucleated cells (Methods and Materials). Only ~17% of epithelium is multinucleated in wild-type, uninjured flies, but by 10 dp nearly ~95% of the epithelium is repaired via generation of multinucleated cells (Fig. 1 A and C).

In addition to the generation of giant, multinucleated cells, we also observed enlarged nuclei by 5 to 10 dp UV-A irradiation. Nuclear size is known to correlate with the DNA content, therefore we measured nuclear ploidy to determine if the epithelial nuclei underwent whole genome duplication (Fig. 1A) (20). Epithelial nuclear ploidy was measured using our previously reported method, where nuclear ploidy is determined based on DAPI intensity (21, 22). In doing so, we observed that epithelial nuclei gradually became more polyploid as time progressed post UV-A irradiation (Fig. 1D). There was a significant increase in epithelial nuclear ploidy (>3C) by 5 dp when ~40% of the epithelial nuclear number was reduced (Fig. 1 B and D). The epithelial nuclear ploidy further increased in terms of both the percentage of nuclei and total DNA content of each nucleus. At 7 to 10 dp, epithelial nuclear ploidy reached more than 12C with up to 40% of epithelial nuclei exhibiting a >3C DNA content per nucleus. Thus, UV-A irradiation appears to lead to an injury-like response generating both mono- and multinucleated, polyploid cells (6).

Endoreplication Is Dispensable for Epithelial Repair Post UV-A Irradiation.

The increased nuclear ploidy post UV-A indicated that the epithelial cells reentered an incomplete cell cycle to endoreplicate. To determine when cells entered S-phase, we used the reporter strain, PCNA-EGFP, and conducted a time course post UV-A (SI Appendix, Fig. S2A) (23). As expected, no PCNA positive epithelial nuclei were observed in the 0 mJ control. In UV-A irradiated flies, PCNA+ nuclei gradually peaked at 4 dp (~30, PCNA+ nuclei) and then reduced in number by 6 dp (SI Appendix, Fig. S2 B and C). Epithelial-specific expression was confirmed based on colocalization with the epithelial specific transcription factor, Grh (21, 22). Endoreplication can occur by the endocycle, which bypasses M-phase completely, or by endomitosis in which cells start, but do not finish M phase via failed karyokinesis or cytokinesis. To distinguish between the two incomplete cell cycles, the tissues were immunostained with the M-phase marker Cyclin B (CycB). Compared to a positive control, in which CycB was overexpressed in the epithelium, there was no detectable CycB expression pre- or post UV-A irradiation (SI Appendix, Fig. S2D). Therefore, the fly epithelial cells appear to endocycle and not endomitose post UV-A irradiation, leading to whole genome duplication.

Endoreplication contributes to tissue repair as it restores tissue mass and permits healing in the presence of genotoxic stress (12). Thus, we investigated the contribution of endoreplication to epithelial repair post UV-A irradiation, by knocking down E2F1 in the epithelium using the Gal4/UAS system (epi>E2F1RNAi) (SI Appendix, Fig. S3). First, we confirmed that E2F1 knockdown inhibited endoreplication post UV-A irradiation as epithelial nuclear ploidy was significantly reduced and remained in the diploid range (1-3C) for ~97% of the nuclei (SI Appendix, Fig. S3 A–C). Next, we quantified the number of epithelial nuclei pre- and post UV-A irradiation and observed a similar reduction in the nuclear number for both wild-type (epi>+) and E2F1RNAi strains (SI Appendix, Fig. S3D). Similarly, we observed that epithelial cell junctions remained continuous and intact at 7 dp UV-A even when endoreplication was inhibited (SI Appendix, Fig. S3 A and B). Thus, it appears that endoreplication is dispensable for tissue repair post UV-A irradiation. However, endoreplication may still be necessary to compensate for cell loss. Therefore, we measured total tissue ploidy by summing the epithelial nuclear ploidy values in wild-type and E2F1 knockdown strains pre- and post UV-A irradiation. As observed in other injury models, endoreplication is also necessary to restore lost synthetic capacity post UV-A irradiation (SI Appendix, Fig. S3E) (5, 6, 21).

Epithelial Cells Fuse in Response to UV-A Irradiation.

Next, we examined the role and regulation of the giant, multinucleated epithelial cells that encompassed the epithelium in its entirety post UV-A. Multinucleated cells can arise via endomitosis or cell fusion (1). Since the epithelium is prone to fuse in response to stress and we did not observe expression of the M-phase protein, CycB, we hypothesized that these multinucleated cells were generated by cell fusion post UV-A stress (6, 24). Cytoplasmic sharing is a reliable means to measure fusion events in vivo, so we used the dBrainbow lineage system to mosaically induce expression of the cytoplasmic fluorescent proteins EGFP-HSV or EBFP2-HA following Cre-mediated recombination (24, 25). As a result, any bicolored area in the fly epithelium would be indicative of a cell fusion event (Fig. 2A). The uninjured (0 mJ) epithelium displayed distinct color regions expressing either EGFP or EBFP while the fluorescent proteins in the 7 dp UV-A irradiated strains frequently overlapped (Fig. 2B). The epithelial area expressing EGFP was ~45% in the control (0 mJ) and significantly increased to ~85% at 7 dp UV-A irradiation (Fig. 2C). Comparing the epithelial cell junctions to bicolored areas revealed enlarged cells sharing different levels of EGFP and EBFP expression due to the merging of various number of cells which generated the multinucleated cells (Fig. 2 B and 2 B′). To confirm that the bicolored cells arose by cell fusion and not endoreplication, we generated fly strains where endoreplication could be inhibited (epi>E2F1RNAiand epi>CycERNAi). Blocking endoreplication did not significantly affect EGFP area when compared with wild-type strain post UV-A (SI Appendix, Fig. S4). Thus, cytoplasmic sharing of fluorescent proteins following UV-A irradiation appears to be primarily due to cell-to-cell fusion.

Fig. 2.

Figure shows immunofluorescent images of mosaically label cells, which upon fusion express more than one fluorescent protein.

Open in a new tab

Epithelial cells fuse in response to UV-A irradiation. (A) Illustration of mosaically labeled tissue where cells expressing more than one color (pink) are indicative of cell fusion. (B) Immunofluorescent images of the fly epithelium at 7 dp ± UV-A treatment with fluorescent proteins, EGFP (yellow) and EBFP2 (magenta). Cell junctions (FasIII, cyan). (B′) Zoom example of a multinucleated, bicolored cell (dashed white line). (C) Quantification of epithelial area expressing EGFP. Student t test with P-value ****<0.0001.

Epithelial Cell Fusion Appears to be Essential for Tissue Repair Post UV-A.

Cell fusion appears to be the prominent response to UV-A irradiation; thus, we tested whether cell fusion is required for tissue repair. Previous studies have shown that epithelial-specific expression of dominant negative Rac GTPase (RacDN) is sufficient to inhibit the generation of multinucleated cells in response to injury or aging (6, 24). Epithelial-specific expression of RacDN does not affect epithelial development, but after UV-A irradiation, only mononucleated epithelial cells were observed on the epithelial periphery (Fig. 3 A and B). The central region of the epithelium was devoid of nuclei in the RacDN strain, while big multinucleated cells were observed throughout the epithelium from the wild-type strain. To determine if the fly epithelial integrity was compromised, an epithelial-specific membrane RFP (mCD8.RFP) was expressed in both wild-type and RacDN strains (SI Appendix, Fig. S5). Prior to UV-A irradiation, a continuous epithelial sheet is visible in both fly strains (SI Appendix, Fig. S5 A and C). At 10 dp UV-A irradiation, the epithelial membrane appears healed in the wild-type strain, albeit with some puncta aggregates (SI Appendix, Fig. S5B). However, the membrane RFP was abolished in areas devoid of epithelial nuclei in the fly epithelium expressing RacDN and only the elongated muscle nuclei persisted (SI Appendix, Fig. S5 D and D′). By measuring the membrane RFP intensity, we confirmed that the fly epithelial membrane was repaired post UV-A irradiation. However, inhibiting cell fusion via RacDN appears to compromise epithelial integrity due to the loss of epithelial nuclei (Fig. 3C and SI Appendix, Fig. S5).

Fig. 3.

Figure shows immunofluorescent images of fly epithelium when cell fusion is inhibited with graphs quantifying number of viable and apoptotic nuclei.

Open in a new tab

Expression of dominant negative Rac GTPase sensitizes epithelial cells to apoptosis post UV-A irradiation. (A) Model of Rac GTPase–dependent cell fusion in the fly epithelial cells. (B) Immunofluorescent images of the fly epithelium at 7 dp UV-A irradiation. Epithelial cell junctions (FasIII, magenta) and nuclei (Grh, green) are shown. Denoted are nuclei within the syncytia (arrows) and mononucleated cells (arrowheads) as well as an area devoid of cells/nuclei (*). (C) Quantification of epithelial nuclei number, which is severely reduced post UV-A when dominant negative Rac (RacDN) is expressed. (D and E) Immunofluorescent images of fly epithelium ± UV-A at indicated dp. Boxed regions show examples of pyknotic nuclei (arrowheads) at 2 and 3 dp UV-A. All nuclei (DAPI, gray). (F and G) Quantification of epithelial nuclei and apoptosis, respectively. Data represent the SEM analyzed by Student t test (C) or Two-way ANOVA with P-value *<0.05, ***<0.001, and ****<0.0001.

Next, we performed a time course at 0, 1, 2, and 3 dp UV-A irradiation to determine the cause of the dramatic reduction in the number of epithelial nuclei. To start, we observed no significant difference in epithelial organization at 0 or 1 dp UV-A irradiation, as the total number of epithelial nuclei was comparable in both strains and time points (Fig. 3 D–F). However, by 2 dp UV-A, noticeable changes in the epithelial organization were evident. There were areas devoid of cell junctions and nuclei, suggesting that apoptosis occurred. Zoom-ins (Fig. 3 D′, D″, E′ and E″, boxed regions) revealed fragments or condensed puncta known as pyknotic nuclei at the center of areas devoid of FasIII septate junction staining. The condensed or fragment nuclei are characteristic of programmed cell death/apoptosis (26). In most cases, the epithelial nuclei lost their Grh staining but were scored based on their position and nuclear morphology using the DNA intercalating dye, DAPI. The frequency of pyknotic nuclei was further enhanced by RacDN expression with several more epithelial areas devoid of cell junctions (Fig. 3E). Quantifying the number of pyknotic nuclei in UV-A irradiated wild-type and RacDN strains revealed that the epithelial cells expressing RacDN had substantially more apoptotic nuclei than the wild-type cells (Fig. 3G). This was further enhanced by 3 dp UV-A irradiation as the RacDN expressing cells had a greater than twofold increase in pyknotic nuclei. As a result, a significant decrease in the number of viable epithelial nuclei was observed by 3 dp UV-A in RacDN expressing cells (Fig. 3F).

The induction of apoptosis and cell fusion appear to occur simultaneously, making it difficult to determine the cause versus the consequence of the epithelial tissue repair defect post UV-A irradiation. The changes in epithelial cell junction morphology suggest that a defect in cell fusion may precede apoptosis as the septate junction protein, FasIII, becomes enriched at the intersection of multiple cell borders in the wild-type strain (Fig. 3 D′ and D″). Meanwhile, in the cells expressing RacDN, there is no indication of FasIII remodeling as the septate junctions appear broken and continue to disintegrate over the time course (Fig. 3 E′ and E″). To further confirm these observations, we thought to identify other regulators of epithelial cell fusion.

Rac GTPase is a member of the Rho family of GTPases, which regulates actin polymerization in cellular development and wound healing (27–29). The Rho family of GTPase have been shown to regulate cell fusion in muscle and fungi, thus we tested if epithelial expression of a dominant negative Cdc42 or Rho would also sensitize epithelial cells to UV-A irradiation (SI Appendix, Fig. S6) (30–32). The epithelial-specific expression of Cdc42DN, similar to RacDN, resulted in enhanced nuclear loss by 7 dp UV-A irradiation (SI Appendix, Fig. S6B). We observed tissue areas devoid of nuclei with mononucleated cells at the borders, while RhoDN expressing cells were evenly dispersed across the tissue with the overlaying lateral muscle fibers as visualized by the DAPI stained nuclei (SI Appendix, Fig. S6A). Expression of RhoDN did not significantly impact the epithelium as a comparable number of nuclei were lost relative to the wild-type (SI Appendix, Fig. S6B). In conclusion, we have found that Cdc42, like Rac, appears to be required for epithelial cell fusion and tissue repair following UV-A irradiation.

Caspase-Dependent Cell Death Is Required for Polyploidization Post UV-A Irradiation.

To elucidate the link between cell apoptosis and polyploidy, we blocked cell apoptosis by overexpressing the baculoviral antiapoptotic protein, p35 (epi>p35OE), which counteracts the effector caspases in the apoptotic pathway (33). As compared to the wild-type (epi>+) strain, p35OE did not appear to affect the development of the fly epithelium as the cellular organization was normal despite a slight increase in the number of nuclei per unit area (Fig. 4 A–C). Following UV-A irradiation, however, p35OE significantly reduced nuclear loss (Fig. 4 B and C). Fewer enlarged nuclei were also observed in the p35OE strain, and nuclear ploidy measurements confirmed that the endocycle was inhibited (Fig. 4D).

Fig. 4.

Figure shows effect of caspase inhibition on cell death, ploidy, and cell fusion using dBrainbow mosaic labeling.

Open in a new tab

Caspase-dependent apoptosis is necessary for polyploidization post UV-A irradiation. (A and B) Immunofluorescent images of the fly epithelium at 7 dp ± UV-A in the strains indicated. Epithelial nuclei (Grh, green) and cell junctions (FasIII, magenta). Quantification of (C) epithelial nuclei number and (D) ploidy in both strains at 7 dp. (E and F) Immunofluorescent images of the fly epithelium at 7 dp ± UV-A in the dBrainbow strains with fluorescent proteins EGFP (yellow) and EBFP (magenta). Cell junctions (FasIII, cyan). (G) Quantification of epithelial area expressing EGFP in dBrainbow strains. Data represent the SEM analyzed by Two-way ANOVA with P-value *<0.05, ***<0.001, and ****<0.0001.

The epithelial septate junctions as stained with FasIII were still expressed but were disordered post UV-A irradiation making it difficult to quantify the extent of multinucleation. Instead, we used dBrainbow mosaic labeling technique to determine whether cell fusion was also inhibited. As shown previously, the wild-type strain had extensive overlap between EGFP-HSV and EBFP2-HA fluorescent proteins post UV-A (Figs. 2B and 4E). The epithelial area expressing EGFP significantly increased to ~89% at 7 dp UV-A compared to 41% in unirradiated (0 mJ) wild-type (epi>+; dBrainbow) strain (Fig. 4G). In contrast, both the uninjured (0 mJ) and UV-A irradiated p35OE strains displayed predominantly one fluorescent protein with regions expressing either EGFP or EBFP (Fig. 4 F and G). Therefore, inhibition of apoptosis through expression of this effector caspase inhibitor is sufficient to strongly reduce cell fusion post UV-A irradiation.

Next, we questioned if it was cell death itself or a caspase-dependent signal that initiated polyploidization. To do so, apoptosis was directly activated by conditionally expressing the proapoptotic Drosophila gene, hid (33–35). We used a Gal80 construct that has the auxin inducible degron to conditionally regulate Gal4-dependent hid expression in the fly epithelium for 1 or 3 d after the flies eclosed (SI Appendix, Fig. S7A) (36). This led to a dosage effect, where flies expressing hid for 1 d lost ~48% of their epithelial nuclei, whereas flies expressing hid for 3 d lost ~87% of their epithelial nuclei (SI Appendix, Fig. S7 B and C). Interestingly, giant multinucleated cells were observed in both apoptotic conditions, but the nuclear ploidy was only significantly increased when there was severe nuclear loss in the epithelium (SI Appendix, Fig. S7 C and D). As a result, we can conclude that an apoptotic signal is necessary for epithelial cells to fuse, but endoreplication requires a threshold of nuclear loss within the epithelium to be triggered.

Discussion

Here, we created a fruit fly model of ultraviolet light-induced injury. Exposing the adult fruit flies to a low dose of UV-A was sufficient to cause significant cell loss. The epithelium, in turn, responded by polyploidization instead of dividing the surviving cells. The enlarged epithelial cells predominantly arose by cell fusion and appeared to span the entire tissue area. Unlike with a puncture wound, where cell fusion and endocycling synergize to close the wound, the endocycle appears to be dispensable following UV-A irradiation (6). This is likely because cell fusion provides a more rapid response to widespread damage, accelerating wound closure (1, 6).

Activation of the Endocycle and Compensating for Cell Loss.

Even though endoreplication appears to be dispensable for epithelial wound healing in this UV-A injury model, it is still required to compensate for cell loss. Measuring the DNA content (C-value) of the remaining nuclei indicated that the total tissue ploidy was restored with fewer nuclei post UV-A irradiation as the surviving nuclei contained higher ploidy (≥3C). In addition, the endocycle has been shown to enable resistance to genotoxic stress and DNA damage, yet here we find that the endocycle is dispensable (13, 14, 16). Apoptosis, as measured by the number of epithelial nuclei that survive UV-A irradiation, was not changed when the endocycle was inhibited compared to the wild-type strain. This may be due to the dual mechanisms of polyploidization in the adult epithelium, as cell fusion, which is the predominant response to UV-A irradiation, appears to be essential to protect the epithelial cells against apoptosis.

DNA damage has been shown to be the initiator of endoreplication for a variety of cell types in response to differentiation, physiological, and pathological cues. This includes mouse mammary gland cells in which endoreplication occurs following replication stress, as well as giant macrophage cells that arise in response to DNA damage from reactive oxygen species (37–39). In these cases, genotoxic stress leads to endoreplication by truncating or bypassing mitosis; however, it remains unknown how DNA damage would result in cell fusion. Here, we suspect it may be through activation of canonical caspase-dependent apoptotic pathway (40).

The Crosstalk between Apoptosis and Cell Fusion.

Studies on syncytial cell types, including the skeletal muscle, bone-resorbing osteoclasts, and syncytiotrophoblasts in mammals, have shown that activation of effector caspases, including caspase 3, is necessary for cell-to-cell fusion (41–43). Similarly, we find that overexpression of the effector caspase inhibitor, p35, is sufficient to inhibit epithelial cell fusion post UV-A irradiation (44, 45). In osteoclasts, the effector caspase 3 was shown to cleave the La protein, which facilitates cell fusion through phosphatidylserine exposure on the plasma membrane (42, 46). However, it remains to be determined whether this mechanism is conserved in other animal cell types to initiate cell fusion in development or in response to stress.

This fits with our findings, as we were not able to untangle Rac GTPase’s role in apoptosis and fusion post UV-A irradiation. Apoptosis, as observed by pyknotic nuclei, occurred simultaneously with epithelial cell junctional remodeling. Epithelial border breakdown occurred at the site of pyknotic nuclei in both the wild-type and RacDN-expressing epithelial cells. However, the frequency of pyknotic nuclei was significantly enhanced by RacDN expression, suggesting that Rac inhibits apoptotic signaling in the process of regulating cell fusion in this model.

Interestingly, overexpression of the effector caspase inhibitor, p35, was also sufficient to inhibit endoreplication post UV-A irradiation. The adult Drosophila epithelial cells are postmitotic and do not express M-phase genes (6, 12). Therefore, the epithelial cells are unlikely to experience replication stress that would truncate the mitotic cell cycle (16). Instead, previous studies have shown that the degree of nuclear ploidy is dependent on the tissue’s cellular density, which is regulated by the conserved Hippo-Yki signal transduction pathway (21). As a result, we suspect that it is apoptosis itself that is necessary for endoreplication, as severe nuclear loss was needed to increase nuclear ploidy when cell death was genetically activated through hid expression. And in the case of cell fusion, it is more likely to be dependent on a nonapoptotic role of caspases as observed in cells, such as osteoclasts, that undergo cell fusion during development.

Cell Fusion as a Stress Response.

While cell fusion is known to be required for the development of a number of multinucleated cells, its role and regulation as an adaptation to stress remains more poorly understood. Multinucleated cells, also known as syncytia, have been reported to arise most commonly upon viral infection due to an encoded fusogen (47). There are also reports that cell fusion contributes to the generation of giant polyploid cancer cells (48, 49). However, most mammalian tissues are complex, and a common challenge is to distinguish whether the multinucleated cells identified arose by failed cytokinesis or cell fusion.

The fruit fly, Drosophila, has emerged as a model to study how stress triggers polyploidization, particularly in epithelial cell fusion. The epithelial cells underlying the cuticle exoskeleton have been found to undergo cell fusion at all postembryonic stages of life (i.e., larval, pupa, and adult) postinjury and with age (6, 9, 10, 24, 50). Here, we have identified another stress that induces cell fusion, UV-A irradiation. Utilizing Drosophila’s extensive genetic toolkit, the extent of cell fusion was determined through expression of the dBrainbow mosaic labeling system. A week post UV-A irradiation led to near complete overlap of the dBrainbow fluorescent proteins, whereas before irradiation, the fluorescent proteins rarely overlapped. Inhibiting effector caspase activity with p35 kept fluorescent proteins separate, similar to the unirradiated flies. The dBrainbow technique was essential to validate whether cell fusion was inhibited by p35 expression, as the localization of FasIII septate junction marker became disorganized post UV-A irradiation, making it difficult to discern the cell boundaries, which complicated the analysis.

Finally, we determined that cell fusion post UV-A irradiation is dependent on the Rho family GTPases, Rac and Cdc42, which play a conserved role in cell fusion. Both Rho GTPases regulate actin polymerization, an essential process in cell fusion from fungal species to the development of syncytial muscle cells (30–32). Epithelial cells on the periphery of the UV-A induced injury remained mononucleated when either Rac or Cdc42 dominant negative mutants were expressed. Thus, their GTPase activity appears to be necessary for cell fusion post UV-A irradiation, although we cannot rule out additional roles for Rac or Cdc42 in epithelial cell survival or wound healing in this model. While the molecular mechanism of epithelial cell fusion in response to stress remains to be identified, here we find a link between apoptosis and cell fusion, demonstrating that an apoptosis-dependent mechanism is necessary for polyploidization post UV-A irradiation.

Methods and Materials

Fly Husbandry and Strains.

All D. melanogaster strains and crosses used in these experiments were kept in vials containing a standard corn syrup and soy food (Archon Scientific) at 25 °C, 60% humidity, and a 12-h light/dark cycle. See SI Appendix, Table S1 for a complete list of strains used in this study. All crosses were created using standard mating schemes in Drosophila, beginning with 10 virgin females and 5 males. The Gal4/UAS system was used to express the dominant negative genes and gene-specific RNAi using an epithelial-specific GMR51F10-Gal4 driver (6, 21).

UV-A Irradiation.

Flies were collected, aged to 3 d, and anesthetized using Fly-Nap (Caroline Biologicals, cat#173010) for 3 min to ensure they remained asleep during UV-A irradiation. Once asleep, the Drosophila strains were immobilized with their ventral abdomen facing up using double-sided tape and irradiated at 4 cm from lamps in the Fisherbrand Stratalinker with UV-A bulbs using a dose of 25 mJ. The flies were returned to food vials and left to recover for the indicated times in dp irradiation.

Drosophila Dissection and Immunostaining.

Drosophila abdominal dissection and immunostaining were performed on female flies as previously reported (22). The fly abdomens were immobilized on a Sylgard plate, then fixed in 4% paraformaldehyde for 30 min, washed, and permeabilized with 1× PBS, 0.3% Triton X-100, and 0.3% BSA. The samples were then stained overnight in primary followed by secondary antibody solutions. See SI Appendix, Table S2 for a complete list of all antibodies used in this study. Finally, all samples were stained with DAPI (1:5,000) to visualize all nuclei.

Fly Imaging and analysis.

The fixed, stained, and mounted abdomen samples were imaged using a Zeiss AxioImagerM2 microscope with an Apotome and a 40× dry objective with a Hamamatsu Orca-Flash 4.0 camera. Samples for Figs. 1, 2, and 4 were imaged using the 40× oil (Zeiss Immersol 514F) objective on a Zeiss LSM 880 Airyscan. Z-stack images were taken with a slice size of 0.5 µm. Using ImageJ/FIJI, images were flattened into MAX intensity stacks for all except the DAPI stain, which was flattened as a SUM intensity stack, and then merged to generate the composite images that are shown.

Nuclei Counting.

Images were rotated and cropped to 100 × 100 μm2, 200 × 200 μm2, or 212 × 212 µm2, as indicated on the y-axis of the figure graphs. The cell counter tool in ImageJ/FIJI was used to quantify the viable nuclei, which colocalized with epithelial-specific Grh or nuclear GFP and DAPI stains. Nuclei that overlapped with the image border were excluded from the analysis. Apoptosis leads to condensation and fragmentation of nuclei, generating pyknotic nuclei. Pyknotic nuclei were identified as bright DAPI fragments or spots in the epithelial row that did not costain with Grh. The number of pyknotic nuclei may be higher than viable nuclei loss from tissue damage, as nuclei often fragment into multiple pieces during cell death.

Epithelial Membrane Integrity.

Flies expressing a membrane-RFP (UAS-mCD8.RFP) under epi-Gal4 control were dissected at the indicated time points pre- or post UV-A irradiation, fixed, and immunostained with antibodies to RFP and FasIII. All samples were imaged under the same conditions and settings. Using FIJI, the SUM projection of the RFP channel was compiled for each strain, and the RFP raw intensity was measured in 4, 50 × 50 µm2 areas. For the RacDN fly strain post UV-A, only areas devoid of epithelial nuclei that still retain intact muscle (elongated nuclei) were measured. This quantification served as a means to determine if the epithelial membrane was compromised in the areas devoid of epithelial nuclei.

Ploidy Analysis.

Epithelial nuclear ploidy was measured as previously reported (22). Using FIJI, regions of interest (ROIs) were outlined based on epithelial-specific nuclear GFP or Grh and mapped to the SUM projection of the Z-stack DAPI image to measure the DAPI intensity. Overlapping nuclei were excluded from the analysis. The raw DAPI intensity for each nucleus was measured by subtracting the background signals. The uninjured epithelial nuclei served as the diploid (2C) control to calculate the nuclear ploidy number based on the average from at least three samples per condition. Nuclei were grouped based on ploidy class of 1-3C, 3-6C, 6-12C, and >12C.

Statistical Analysis.

Each experiment was repeated in at least duplicate with the indicated replicate number (n) for each Drosophila strain in the Dataset. Statistical analysis was performed using Microsoft Excel and GraphPad Prism software. The tests conducted included Tukey’s One-way or Two-way ANOVA and Student unpaired t test with P-values as follows: *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.00001.

Supplementary Material

Appendix 01 (PDF)

pnas.2522067123.sapp.pdf (4.7MB, pdf)

Dataset S01 (XLSX)

pnas.2522067123.sd01.xlsx (70.5KB, xlsx)

Acknowledgments

We would like to thank the members of the Losick lab, including Stefania Bonnani, Loiselle Gonzalez-Baez, Sophie Jalkut, and Jack Weidenbach for critical review of this manuscript. We also thank the lab alumni, Benjamin Lamarre, for his help on the preliminary analysis as well as Dr. Ula Jurkunas (Harvard Medical School) for her help and advice on this project. The fly strains used in this study were purchased from (with support by) the Bloomington Drosophila Stock Center (NIH P40OD018537) and Vienna Drosophila Resource Center. The FasIII antibody used in this study was purchased from the Developmental Studies Hybridoma Bank (created by the NICHD and maintained at The University of Iowa, Department of Biology, Iowa City, IA 52242). Images were acquired using microscope supported by the NSF under Grant No. 1626072 with additional thanks to Bret Judson whom directs the Boston College Imaging Core for infrastructure and support. Research reported in this publication was supported by Boston College and the National Institute of General Medical Sciences of the NIH under Award No. R35GM124691 to V.P.L. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH.

Author contributions

M.S. and V.P.L. designed research; M.S., L.G.M., L.W.B., L.M.B., and V.P.L. performed research; M.S., L.G.M., L.W.B., L.M.B., and V.P.L. analyzed data; M.S., L.G.M., L.W.B., L.M.B., and V.P.L. reviewed and edited manuscript; and M.S., L.G.M., and V.P.L. wrote the paper.

Competing interests

The authors declare no competing interest.

Footnotes

This article is a PNAS Direct Submission. D.F. is a guest editor invited by the Editorial Board.

Data, Materials, and Software Availability

The authors affirm that all data necessary for verifying the conclusions are present within the article, figures, and SI Appendix. All raw data points reported in figures can be found in Dataset S1. Additionally, the Drosophila strains and antibodies used in this study are listed SI Appendix, Tables S1 and S2 and are available upon request or as referenced. All other data are included in the manuscript and/or supporting information.

Supporting Information

References

  • 1.Bailey E. C., Kobielski S., Park J., Losick V. P., Polyploidy in tissue repair and regeneration. Cold Spring Harb. Perspect. Biol. 13, a040881 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Lazzeri E., Angelotti M. L., Conte C., Anders H.-J., Romagnani P., Surviving acute organ failure: Cell polyploidization and progenitor proliferation. Trends Mol. Med. 25, 366–381 (2019). [DOI] [PubMed] [Google Scholar]
  • 3.Lazzeri E., et al. , Endocycle-related tubular cell hypertrophy and progenitor proliferation recover renal function after acute kidney injury. Nat. Commun. 9, 1344 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Zhang S., et al. , The polyploid state plays a tumor-suppressive role in the liver. Dev. Cell 44, 447–459.e5 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Cohen E., Allen S. R., Sawyer J. K., Fox D. T., Fizzy-related dictates a cell cycle switch during organ repair and tissue growth responses in the Drosophila hindgut. eLife 7, e38327 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Losick V. P., Fox D. T., Spradling A. C., Polyploidization and cell fusion contribute to wound healing in the adult Drosophila epithelium. Curr. Biol. 23, 2224–2232 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Tamori Y., Deng W.-M., Tissue repair through cell competition and compensatory cellular hypertrophy in postmitotic epithelia. Dev. Cell 25, 350–363 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Losick V. P., Duhaime L. G., The endocycle restores tissue tension in the Drosophila abdomen post wound repair. Cell Rep. 37, 109827 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Galko M. J., Krasnow M. A., Cellular and genetic analysis of wound healing in Drosophila larvae. PLoS Biol. 2, e239 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.White J. S., et al. , Wound-induced syncytia outpace mononucleate neighbors during Drosophila wound repair. bioRxiv [Preprint] (2023). https://www.biorxiv.org/content/10.1101/2023.06.25.546442v2 (Accessed 11 September 2025). [DOI] [PMC free article] [PubMed]
  • 11.White J. S., Hutson M. S., Page-McCaw A., Wounding increases nuclear ploidy in wound-proximal epidermal cells of the Drosophila pupal notum. MicroPubl. Biol. 2024, 10.17912/micropub.biology.001067 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Grendler J., Lowgren S., Mills M., Losick V. P., Wound-induced polyploidization is driven by Myc and supports tissue repair in the presence of DNA damage. Dev. Camb. 146, dev173005 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Zhang B., Mehrotra S., Ng W. L., Calvi B. R., Low levels of p53 protein and chromatin silencing of p53 target genes repress apoptosis in Drosophila endocycling cells. PLoS Genet. 10, e1004581 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Nandakumar S., Grushko O., Buttitta L. A., Polyploidy in the adult Drosophila brain. eLife 9, e54385 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Conway P. J., Dao J., Kovalskyy D., Mahadevan D., Dray E., Polyploidy in cancer: Causal mechanisms, cancer-specific consequences, and emerging treatments. Mol. Cancer Ther. 23, 638–647 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Herriage H. C., Huang Y.-T., Calvi B. R., The antagonistic relationship between apoptosis and polyploidy in development and cancer. Semin. Cell Dev. Biol. 156, 35–43 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Madigan J. P., Chotkowski H. L., Glaser R. L., DNA double-strand break-induced phosphorylation of Drosophila histone variant H2Av helps prevent radiation-induced apoptosis. Nucleic Acids Res. 30, 3698–3705 (2002). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Bou Saada Y., et al. , Control of DNA integrity in skeletal muscle under physiological and pathological conditions. Cell. Mol. Life Sci. 74, 3439–3449 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Borges H. L., Linden R., Wang J. Y., DNA damage-induced cell death. Cell Res. 18, 17–26 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Heijo H., et al. , DNA content contributes to nuclear size control in Xenopus laevis. Mol. Biol. Cell 31, 2703–2717 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Losick V. P., Jun A. S., Spradling A. C., Wound-induced polyploidization: Regulation by Hippo and JNK signaling and conservation in mammals. PLoS One 11, e0151251 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Bailey E. C., Dehn A. S., Gjelsvik K. J., Besen-McNally R., Losick V. P., A Drosophila model to study wound-induced polyploidization. J. Vis. Exp. 160, e61252 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Blythe S. A., Wieschaus E. F., Establishment and maintenance of heritable chromatin structure during early Drosophila embryogenesis. eLife 5, e20148 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Dehn A. S., et al. , Epithelial mechanics are maintained by inhibiting cell fusion with age in Drosophila. J. Cell Sci. 136, jcs260974 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Hampel S., et al. , Drosophila brainbow: A recombinase-based fluorescence labeling technique to subdivide neural expression patterns. Nat. Methods 8, 253–259 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Elmore S., Apoptosis: A review of programmed cell death. Toxicol. Pathol. 35, 495–516 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Hui J., Stjepić V., Nakamura M., Parkhurst S. M., Wrangling actin assemblies: Actin ring dynamics during cell wound repair. Cells 11, 2777 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Ma N., Xu E., Luo Q., Song G., Rac1: A regulator of cell migration and a potential target for cancer therapy. Mol. Basel Switz. 28, 2976 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Nakamura M., Hui J., Stjepić V., Parkhurst S. M., Scar/WAVE has Rac GTPase-independent functions during cell wound repair. Sci. Rep. 13, 4763 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Kim J. H., Chen E. H., The fusogenic synapse at a glance. J. Cell Sci. 132, jcs213124 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Noviello C., et al. , RhoA is a crucial regulator of myoblast fusion. Cells 12, 2673 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Sieber B., Coronas-Serna J. M., Martin S. G., A focus on yeast mating: From pheromone signaling to cell-cell fusion. Semin. Cell Dev. Biol. 133, 83–95 (2023). [DOI] [PubMed] [Google Scholar]
  • 33.Vucic D., Kaiser W. J., Miller L. K., Inhibitor of apoptosis proteins physically interact with and block apoptosis induced by Drosophila proteins HID and GRIM. Mol. Cell. Biol. 18, 3300–3309 (1998). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Bilak A., Su T. T., Regulation of Drosophila melanogaster pro-apoptotic gene hid. Apoptosis 14, 943–949 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Goyal L., McCall K., Agapite J., Hartwieg E., Steller H., Induction of apoptosis by Drosophila reaper, hid and grim through inhibition of IAP function. EMBO J. 19, 589–597 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.McClure C. D., et al. , An auxin-inducible, GAL4-compatible, gene expression system for Drosophila. eLife 11, e67598 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Molinuevo R., et al. , Physiological DNA damage promotes functional endoreplication of mammary gland alveolar cells during lactation. Nat. Commun. 15, 3288 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Herrtwich L., et al. , DNA damage signaling instructs polyploid macrophage fate in granulomas. Cell 167, 1264–1280.e18 (2016). [DOI] [PubMed] [Google Scholar]
  • 39.Horn V., Triantafyllopoulou A., DNA damage signaling and polyploid macrophages in chronic inflammation. Curr. Opin. Immunol. 50, 55–63 (2018). [DOI] [PubMed] [Google Scholar]
  • 40.Nakajima Y.-I., Kuranaga E., Caspase-dependent non-apoptotic processes in development. Cell Death Differ. 24, 1422–1430 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Gauster M., Huppertz B., The paradox of caspase 8 in human villous trophoblast fusion. Placenta 31, 82–88 (2010). [DOI] [PubMed] [Google Scholar]
  • 42.Whitlock J. M., Chernomordik L. V., Cell–cell fusion: To lose one life and begin another. BioEssays 47, 2400206 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Dehkordi M. H., Tashakor A., O’Connell E., Fearnhead H. O., Apoptosome-dependent myotube formation involves activation of caspase-3 in differentiating myoblasts. Cell Death Dis. 11, 1–12 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Hay B. A., Wolff T., Rubin G. M., Expression of baculovirus P35 prevents cell death in Drosophila. Development 120, 2121–2129 (1994). [DOI] [PubMed] [Google Scholar]
  • 45.Callus B. A., Vaux D. L., Caspase inhibitors: Viral, cellular and chemical. Cell Death Differ. 14, 73–78 (2007). [DOI] [PubMed] [Google Scholar]
  • 46.Leikina E., et al. , Formation of multinucleated osteoclasts depends on an oxidized species of cell surface-associated La protein. eLife 13, RP98665 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Leroy H., et al. , Virus-mediated cell–cell fusion. Int. J. Mol. Sci. 21, 9644 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.Song Y., Zhao Y., Deng Z., Zhao R., Huang Q., Stress-induced polyploid giant cancer cells: Unique way of formation and non-negligible characteristics. Front. Oncol. 11, 724781 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 49.Was H., et al. , Polyploidy formation in cancer cells: How a Trojan horse is born. Semin. Cancer Biol. 81, 24–36 (2022). [DOI] [PubMed] [Google Scholar]
  • 50.Scherfer C., Han V. C., Wang Y., Anderson A. E., Galko M. J., Autophagy drives epidermal deterioration in a Drosophila model of tissue aging. Aging 5, 276–287 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Appendix 01 (PDF)

pnas.2522067123.sapp.pdf (4.7MB, pdf)

Dataset S01 (XLSX)

pnas.2522067123.sd01.xlsx (70.5KB, xlsx)

Data Availability Statement

The authors affirm that all data necessary for verifying the conclusions are present within the article, figures, and SI Appendix. All raw data points reported in figures can be found in Dataset S1. Additionally, the Drosophila strains and antibodies used in this study are listed SI Appendix, Tables S1 and S2 and are available upon request or as referenced. All other data are included in the manuscript and/or supporting information.

Articles from Proceedings of the National Academy of Sciences of the United States of America are provided here courtesy of National Academy of Sciences

RESOURCES