Characterising the Modulatory Role of Ikebana in Flower-Dependent Cell Competition

Tissues encompass a quality control mechanism that promotes their optimal state. This mechanism, designated cell competition, is characterised by the elimin ation of suboptimal yet viable cells when they are near healthier cells within the same tissue compartment. This study explores Flower-dependent cell competition and introduces Ikebana as a novel player. The differential expression of the flower isoforms labels cells as winners or losers , influencing their fate in diverse contexts, including eye development, traumatic brain injury, and Alzheimer's disease. Ikebana, ubiquitously produced in wing imaginal discs and adult brains, modulates loser cell elimination. Reduction of ikebana expression correlates with an increased number of loser cells, while its overexpression in the Alzheimer’s disease model reduces the number of Flower LoseB-positive cells. We suggest that Ikebana protects loser cell elimination, particularly when excessive elimination of loser cells c an compromise tissue function. Thus, Ikebana might be a potential therapeutic target for modulating Flower LoseB expression.

Ikebana, Flower, Cell Competition, Drosophila melanogaster Abbreviations Ab42 – Amyloid beta 42; ACI – After Clone Induction; bp – base pair; ctr – control; Diap1 – Death- associated inhibitor of apoptosis 1; KI – knockin; KO – knockout; LAPTM – Lysosomal Protein Transmembrane

The Darwinian principle of "survival of the fittest" has transcended the animal kingdom to encompass the cellular landscape within their bodies (Moreno and Rhiner 2014) . C urrent knowledge recognises that cells within the same tissue compartment engage in a phenomenon known as Cell Competition, a process elucidated as early as 1975 with the observation of slow- proliferating minute ribosomal mutants being outcompeted by normally proliferating cells (Morata and Ripoll 1975). This phenomenon extends beyond mere cell proliferation differences (Böhni et al. 1999) , leading to a broader definition of cell competition — the elimination of viable yet suboptimal cells when juxtaposed with fitter counterparts within the same tissue compartment. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 2 Cell competition has been documented across various tissues in organisms ranging from Drosophila and mice to zebrafish and humans, with implications in neurodegenerative diseases and cancer biology (Akieda et al. 2019; Coelho et al. 2018; Eisenhoffer et al. 2012; Madan et al. 2019; Oliver et al. 2004; Villa del Campo et al. 2014; Walderich et al. 2016) . The prevailing consensus in the field identifies three major mechanisms of cell competition. First is competition for survival factors, exemplified by decapentaplegic capture during wing imaginal disc development, where cells closer to essential factors , or more able to capture them, gain a competitive advantage (Moreno, Basler, and Morata 2002). Second, mechanical cell competition, where cells compete for space, and the resistance to mechanical forces determines their fate, as seen in the elimination of cells converging to the midline in the Drosophila thorax (Brás-Pereira and Moreno 2018; Levayer, Dupont, and Moreno 2016; Moreno et al. 2019) . Lastly, cells may compete based on their fitness state, influenced by factors such as ageing (Merino et al. 2015) or exposure to toxic environments (Coelho et al. 2018). This leads to the elimination of suboptimal cells (losers) and their potential replacement by fitter cells when in the vicinity of healthier (winner) cells. Fitness fingerprints are proteins that recognise and define fitness states and are thought to play crucial roles in the execution of cell competition. The fitness fingerprints that have been identified to date include Flower (Rhiner et al. 2010), Slit-Robo-Ena (Vaughen and Igaki 2016), Spätzle-Toll (Alpar, Bergantiños, and Johnston 2018; Germani et al. 2018; Tan et al. 2008), Sas- PTP10D (Yamamoto et al. 2017), and Flamingo (Bosch, Cho, and Axelrod n.d.). This study delves into Flower -dependent cell competition, which relies on the transmembrane protein Flower (Rhiner et al. 2010). Although Flower has been initially described as a calcium channel (Yao et al. 2009, 2017), such function appears irrelevant to cell competition (Coelho and Moreno 2020; Madan et al. 2019) . In Drosophila, Flower exists in three isoforms – Flower LoseA and Flower LoseB , marking loser cells, and Flower Ubi, marking winner cells (Rhiner et al. 2010). The differential expression of these isoforms determines the fate of a cell as a winner or loser. The expression of flower lose isoforms in Drosophila causes the elimination of suboptimal cells in various contexts, including during eye development (Merino et al. 2013) , in response to traumatic brain injury (Moreno et al. 2015) , and in an Alzheimer’s disease model (Coelho et al. 2018; Coelho and Moreno 2019, 2020). The significance of Flower extends beyond Drosophila, with orthologues described in mice and humans (Madan et al. 2019; Petrova et al. 2012). In Drosophila, loser cells can experience various fates: if they express SPARC, they are protected from elimination; on the other hand, if they express the fitness checkpoint azot, they are then marked to die (Merino et al. 2015; Portela et al. 2010) . Azot, a predicted calcium- calmodulin, activates hid and the subsequent machinery, leading to apoptosis (Merino et al. 2015). However, the mechanisms underlying neighbour recognition as winners or losers, the criteria for selecting loser cells for elimination, and the interplay between Flower, Azot, and apoptosis remain areas of ongoing exploration. This study introduces Ikebana as a new participant in the landscape of cell competition, which acts as a key regulator in determining cell fate. In scenarios where competition is .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 3 intensified, as seen in Alzheimer's disease, increasing ikebana expression reduces the number of cells marked as losers. Our findings position Ikebana as a modulator of loser cell elimination, which might be particularly relevant in contexts where excessive labelling of cells as losers could compromise tissue function. With predicted human orthologs, Ikebana presents a potential avenue for developing novel therapies to modulate Flower expression and cell competition.

Ikebana is predicted to interact with Flower and is basally produced in the wing-imaginal discs and adult brain. Our understanding of the Flower-dependent cell competition pathway remains limited, so our objective was to characterise novel participants in this process . Based on a co -affinity precipitation assay coupled with mass spectrometry, we focused on proteins predicted to interact with Flower, which revealed 34 proteins as potential interactors (Guruharsha et al. 2011). From this list, only Mec2 and CG15098 were expected to be transmembrane proteins, which is crucial as communication and fitness status labelling likely occur at the plasma membrane. Out of these two proteins, we previously found that only CG15098 is upregulated in loser cells in a microarray analysis designed to identify genes differentially expressed during cell competition (Rhiner et al. 2010). We have named this gene ikebana, drawing inspiration from the Japanese art of flower arrangements. Ikebana has four transmembrane domains, with the -N and -C terminus predicted to be intracellular (Figure 1A), and features two predicted domains—Mtp and DUF4728—placing it within the family of tetraspanin-pasiflora proteins (Deligiannaki et al. 2015). To evaluate the expression of ikebana, we employed CRISPR-Cas9 technology, coupled with homologous recombination to generate a genetic ikebana knockout (KO) line and introduce a genetic marker to allow its identification (Baena-Lopez et al. 2013; Huang et al. 2009) . In this ikebana KO line, we inserted a LexA::p65 sequence under the control of the ikebana endogenous promoter, which, in conjunction with LexAop-GFP, facilitated the specific labelling of cells trying to express ikebana (Figure 1B). We found that ikebana exhibits endogenous widespread expression across third-instar larvae wing imaginal discs and within the adult brain (Figure 1C). Ikebana partially protects loser cells from elimination To elucidate the role of Ikebana in cell competition, we conducted experiments to investigate the consequences of modulating the expression of ikebana in loser clones. For this purpose, we used the supercompetition assay , which relies on the tub>dmyc>Gal4 cassette (Moreno and Basler 2004) . Upon activating a heat shock Flippase, this system facilitates the recombination of FRT sites, allowing the expression of both GFP and a construct of interest. By manipulating the heat shock duration, we generated GFP-marked clones scattered within cells carrying an additional copy of dmyc. This supplementary myc copy designates the background cells as winners, while the clones lacking this additional copy are losers whose elimination can be prevented by downregulating the flower lose isoforms or overexpressing diap1 (Death- associated inhibitor of apoptosis 1). .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 4 Contrary to downregulating flower loseA/B, downregulating ikebana in the loser clones did not block their elimination at 72 hours after clone induction (ACI), as shown in Figure 2A-D. However, the overexpression of ikebana in the loser clones significantly reduced their elimination compared to the negative control at 72 hours ACI (Figure 2E-H). Notably, this protective effect was not as pronounced as when overexpressing diap1, suggesting that Ikebana confers partial protection to loser cells from elimination. To ascertain that this effect is specific to cell competition, we conducted experiments to exclude the possibility that Ikebana is a general apoptosis regulator. First, we expressed eiger – cell death initiator via JNK signalling in the eye, which resulted in a reduced eye phenotype (Igaki et al. 2002) . We then examined whether Ikebana could rescue the normal phenotype (Figure S1A-L). Additionally, considering that the rotation of the fly terminalia during development is apoptosis-dependent (Benitez et al. 2010) , we investigated whether overexpressing or downregulating ikebana in this region would impact normal terminalia rotation, manifesting as incomplete terminalia rotation in the adult fly (Figure S 1M-S). Results show that modulating ikebana expression does not interfere with the eye phenotype or terminalia rotation (Figure S1), indicating that Ikebana is not a general regulator of apoptosis, and that the effects observed in clones are indeed attributable to cell competition. Ikebana regulates Flower LoseB expression Given the basal expression of ikebana in the wing imaginal discs of the third instar larvae, we explored whether downregulating its expression with ikebana-RNAi can induce cell competition. U sing the act>y+>Gal4 cassette, we generated GFP -labeled wild -type clones surrounded by cells expressing an additional copy of yellow, which are also wild-type in the context of cell competition. As a negative control, we expressed white-RNAi in the clones, which is not predicted to interfere with cell competition, and, as a positive control, we overexpressed flower loseB, which will confer a loser state to the clones, resulting in their elimination over time. At 72 hours ACI, ikebana-RNAi expression resulted in a reduced clonal area compared to the white-RNAi control. Such reduction of the clonal area was even more pronounced than the one observed with the overexpression of flower loseB (Figure 3A-D). This suggests that decreasing ikebana expression is sufficient to induce a loser state, although more experiments are required to prove that this is due to cell competition. We then asked whether Ikebana regulates the loser state of a cell by influencing Flower expression. Using a FlowerLoseB::mCherry reporter line, we assessed its production in an ikebana KO scenario (Figure 3E-K). For the ikebana KO condition, we crossed the ikebana KOT with the ikebana KOL, a transheterozygous allelic version, because we observed the presence of off-target effects in the ikebana KOL line (data not shown). In the optic lobe of ikebana KO flies, we found a 2.5-fold increase in Flower LoseB-positive cells compared to the control wild type for ikebana (Figure 3J). This rise in the number of loser cells was accompanied by increased Dcp1- positive cells (Figure 3K), meaning these cells were actively being eliminated. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 5 We then tested an Alzheimer’s disease model scenario where we expect ed more loser cells (Coelho et al. 2018). Intriguingly, in this scenario, removing ikebana did not further increase the number of loser cells in the optic lobe (Figure 3J). However, it did increase the number of dying cells (Figure 3K), as measured via Dcp-1 positivity. We also examined whether the same happened for Azot expression, and indeed, the absence of ikebana is sufficient to increase the number of Azot-positive cells in basal cell competition but not in an Alzheimer’s disease model (Figure S2). These data suggest that the absence of ikebana alone is often sufficient to increase the number of loser cells. However, in an Alzheimer’s disease model, Ikebana prevents cell elimination without affecting Flower LoseB expression, as seen by the increase in dying cells in the ikebana KO scenario without an increase in the number of loser cells. Lastly, we investigated whether overexpressing ikebana in the optic lobe, using the GMR driver, would decrease the number of Flower LoseB-positive cells (Figure 3L-P). Under basal levels of cell competition, overexpressing ikebana did not affect the number of Flower LoseB - positive cells. However, in an Alzheimer’s disease model, overexpressing ikebana in the GMR region le d to a 0.7 -fold decrease in the number of cells producing Flower LoseB::mCherry compared to the GFP control (Figure 3P). This suggests that high levels of Ikebana can reverse the low fitness status of cells, indicating that Ikebana is sufficient to partially block the loser fate specification in the Alzheimer’s disease model.

Ikebana, anticipated as a transmembrane protein which physically interacts with Flower, emerges as a pivotal contributor to cell competition dynamics. Using a clonal assay, we observed that localised reduction of ikebana expression within clones results in a diminished clon al area, similar to local overexpression of flower loseB, which forces the cells into a loser state. Given the seemingly ubiquitous expression of ikebana in wing imaginal discs, we propose that, for a cell to adopt a loser state, it must first downregulate ikebana. We hypothesise that this decrease in Ikebana production will promote the production of Flower LoseB, leading to cell elimination. In adult brains, the absence of ikebana proves sufficient to increase the number of cells positive for Flower LoseB or Azot, further supporting this hypothesis. In scenarios anticipating a high number of loser cells , as in the Alzheimer’s disease model , overexpressing ikebana correlates with a decrease in the number of Flower LoseB-positive cells, implying that Ikebana can revert the low fitness of the cells - a refined mechanism which likely operates to selectively eliminate the necessary number of losers, preserving tissue function. In such cases, Ikebana can be re - expressed in loser cells, reverting their loser state by decreasing Flower LoseB expression and sparing selected loser cells from elimination (Figure 4). To strengthen the model, additional experiments are needed to exclude interference with cell proliferation. Overexpressing or downregulating ikebana in specific regions and comparing sizes against a control without ikebana manipulation, coupled with immunostaining for cell death in clones, would bolster the robustness of our conclusions. Although these experiments were not conducted, observational evidence using the actin-Gal4 driver indicates normal development time .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 6 and adult fly size when manipulating ikebana (data not shown). This supports our conviction that the observed effects are due to cell competition rather than mere differences in proliferation rates. Ikebana joins a growing list of proteins, such as SPARC, implicated in protecting loser cells during Flower-dependent cell competition (Portela et al. 2010). However, SPARC does not belong to the Flower-dependent cell competition canonical pathway because manipulating flower in loser or wild -type clones does not change the levels of SPARC (Portela et al. 2010). On the contrary, Ikebana is proposed to interact directly with Flower. Additionally, SPARC is exclusively expressed in loser cells to prevent their elimination, whereas ikebana appears ubiquitously expressed. Interestingly, Ikebana was previously identified as a member of the tetraspanins-pasiflora family of proteins, characterised by four transmembrane domains and conserved regions (Deligiannaki et al. 2015) . In Drosophila, only two other proteins, Pasiflora and Fire Exit, are characterised, albeit in embryos , and th eir role in cell competition remains unexplored (Deligiannaki et al. 2015; With et al. 2003). This protein family extends to humans, including the lysosomal-associated proteins LAPTM4A and LAPTM4B. Ikebana shares more structural similarities with LAPTM4A (Deligiannaki et al. 2015) . This human protein, associated with the clearance of proteins from the plasma membrane via endocytosis and implicated in inducing multidrug resistance in Saccharomyces cerevisiae, presents a promising avenue for therapeutic exploration (Grabner et al. 2011; Hogue, Kerby, and Ling 1999). Future work is needed to confirm LAPTM4A as the human orthologue of Ikebana and test if it is involved in the clearance of Flower lose isoforms from the cell membrane. If so, manipulation of LAPTM4A could open therapeutic avenues to prevent the unnecessary elimination of loser cells. Experimental Procedures Drosophila Genetics and experimental setups Stocks and crosses were kept at 25ºC in Vienna standard media with extra yeast. All stocks were obtained from Bloomington Stock Center unless specified. The following RNAi lines from VDRC were used: UAS-ikebana-RNAi (ID 111608, Chr2, viable), UAS-eiger-RNAi (ID 12658, Chr3, viable), UAS-azot-RNAi (ID 7219, Ch3, viable). For the overexpression of ikebana, a UAS- ikebana stock was generated according to standard procedures. For the supercompetition assay, the following stocks were used: tub>dmyc>Gal4, UAS- GFP, UAS-white-RNAi, UAS-flowerloseA/B-RNAi, UAS-lacZ, and UAS-diap1. The larvae were given a 17-minute heat shock at 37ºC, and the vials were placed at 29ºC until dissected. For the wild -type clones in the wild -type background, the following stocks were used: act>y+>Gal4, UAS-GFP, and UAS-flowerloseB. The larvae were given an 8-minute heat shock at 37ºC, and the vials were placed at 29ºC until dissected. For the experiment to understand the effects of ikebana KO in the number of Flower LoseB or Azot-positive cells, the following stocks were used: white; ikebana{KO}T/ikebana{KO}L; GMR-Gal4, UAS-Ab42; flower{KO;KI-flowerloseB::mCherry}; azot::mCherry. After hatching, female flies were kept at 29ºC until 2 weeks old, when they were dissected. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 7 For the experiment to understand the effects of overexpressing ikebana in the number of Flower LoseB-positive cells, the following stocks were used: GMR-Gal4 and UAS-GFP. After hatching, female flies were kept at 29ºC until 2 weeks old, when they were dissected. For the experiments to understand if Ikebana is a general regulator of apoptosis, the following stocks were used: GMR-Gal4, UAS-eiger and Engrailed-Gal4. The eyes or genitalia were observed in the first week of adulthood. Ikebana knockout-L Generation CRISPR-mediated mutagenesis followed by homologous recombination was performed by the Fly Platform at the Champalimaud Foundation, using the methods highlighted in Baena- Lopez et al. 2013 . In brief, the upstream gRNA sequence ACCAACTGCTTGAACCA[GTC] and the downstream gRNA sequence [ GCT]ACACCAAGATTTAAGCT were cloned into a pCFD5 vector. The cassette 3xPax3::mCherry contains one attP site, a floxed 3xPax3::mCherry, and two homology arms were cloned into pTV3 as a donor template for repair. CG15098-targeting gRNAs and a donor plasmid were microinjected into embryos nos - Cas9. F1 flies carrying the selection marker 3xPax3::mCherry were further validated by genomic PCR and sequenced. CRISPR generates an 1139 -bp deletion allele of CG15098, deleting the partial 5’ UTR/3’ UTR and entire CDS of the CG15098 gene and replacing it with cassette 3xPax3::mCherry. This cassette was then removed by Cre -lox recombination, leaving only the attP site and the loxP. Ikebana knockout-T Generation CRISPR-mediated mutagenesis was performed by WellGenetics Inc. using modified

of Kondo and Ueda 2013 . Briefly, the upstream gRNA sequence GTGAATCCAGAATGCTGTCC[AGG] and the downstream gRNA sequence GGCCAAACGGGAAGCTACAC[TGG] were cloned into U6 promoter plasmid(s) separately. Cassette RMCE-3xPax3::RFP contains two attP sites, a floxed 3xPax3::RFP, and two homology arms were cloned into pUC57-Kan as donor templates for repair. CG15098-targeting gRNAs and hs -Cas9 were supplied in DNA plasmids and a donor plasmid for microinjection into embryos of control strain w[1118]. F1 flies carrying the selection marker 3xP ax3::RFP were further validated by genomic PCR and sequencing. CRISPR generates a 1029-bp deletion allele of CG15098, deleting partial 5’ UTR/3’ UTR and entire CDS of CG15098 gene and is replaced by cassette RMCE -3xPax3::RFP. This cassette was then removed by Cre-lox recombination, leaving only the attP sites and the loxP. Ikebana knockin Generation For the generation of the ikebana{KO; KI -LexA::p65}, the cDNA of LexA::p65 was generated and inserted into a RIV Cherry vector (Baena-Lopez et al. 2013) , and the knockin was generated as described in Huang et al. 2009. Primer sequences are available upon request. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 8 Immunohistochemistry and image acquisition Wing imaginal discs of third -instar larvae were dissected in chilled PBS, fixed for 20 minutes in formaldehyde (4% v/v in PBS), and permeabilised with PBT 0.4% Triton. Adult brains were dissected in chilled PBS; the samples were fixated for 30 minutes in formaldehyde (4% v/v in PBS) and permeabilised with PBT 1% Triton. The wing imaginal discs or brains were then incubated with rabbit a-Dcp1 (1:50) from Cell Signaling (#9578). Samples were mounted in Vectashield with DAPI (Vectorlabs). Confocal images were acquired with Zeiss LSM 880 using the Plan-Apochromat 20x/0.8 M27 dry objective for the case of the wing imaginal discs and the Plan-Apochromat 40x/1.4 Oil DIC M27 objective for the case of the adult brains. Maximum intensity projections of the wing imaginal discs or 41-µm-wide images of the adult brains were obtained with Zeiss Zen Black. To capture images of the adult eyes and genitalia, the Leica S9 I Stereomicroscope was used. Flies were either anaesthetised or imaged alive in a CO2 station. Quantification and statistical analysis Fiji-ImageJ macros developed in this work, available upon request, quantified the clone areas as the number of Flower LoseB, Azot, or Dcp1-positive cells. The areas of the adult eyes were measured in Fiji-ImageJ. For each condition, a minimum of 10 wing imaginal discs, 23 optic lobes, and 12 adult eyes were analysed. Statistical significance between groups was calculated using the nonparametric Kruskal- Wallis test, and Dunn’s test was applied for multiple comparisons between genotypes. Specifically in Figure 3P, statistical significance between groups was calculated using the Brown -Forsythe and Welch ANOVA tests , and Dunnett ’s T3 was applied for multiple comparisons between genotypes. Author Contributions M.M.R., A.G.G., and E.M. designed the experiments. M.M.R. performed and analysed the experiments. A.G.G. helped with image acquisition and statistical analysis. C.B.P. helped with data analysis and design of the transgenic constructs. C.F.C., I.A. and M.S.E obtained preliminary data. B.H. developed the UAS-ikebana transgenic flies. M.M.R. wrote the manuscript.

We thank Bloomington Stock Center for flies; WellGenetics Inc. for the generation of the ikebana KOT; the technicians at the Champalimaud Fly Platform for support with stock maintenance; Catarina Craveiro and the MTT platform for support in the generation of transgenic flies, the ABBE platform for microscopy support and Andrea Spinazzola for his suggestions and comments on the manuscript. M.R was supported by an FCT - Fundação para a Ciência e a Tecnologia - PhD studentship (SFRH/BD/138537/2018). This study was supported by Portuguese national funds, through FCT in the context of the project UIDB/04443/2020 and the European Research Council (Consolidator Grant to E.M.: ‘‘Active Mechanisms of Cell Selection: From Cell Competition to Cell .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 9 Fitness’’). Fly and MTT platforms were funded by the research infrastructure CONGENTO, co - financed by Lisboa Regional Operational Programme (Lisboa2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF) and Fundação para a Ciência e Tecnologia (Portugal) under the project LISBOA-01-0145-FEDER- 022170. The Portuguese Platform of Bioimaging funded ABBE platform - LISBOA-01-0145- FEDER-022122.

Akieda, Yuki, Shohei Ogamino, Hironobu Furuie, Shizuka Ishitani, Ryutaro Akiyoshi, Jumpei Nogami, Takamasa Masuda, Nobuyuki Shimizu, Yasuyuki Ohkawa, and Tohru Ishitani. 2019. “Cell Competition Corrects Noisy Wnt Morphogen Gradients to Achieve Robust Patterning in the Zebrafish Embryo.” Nature Communications 10(1):1–17. doi: 10.1038/s41467-019-12609-4. Alpar, Lale, Cora Bergantiños, and Laura A. Johnston. 2018. “Spatially Restricted Regulation of Spätzle/Toll Signaling during Cell Competition.” Developmental Cell 46(6):706-719.e5. doi: 10.1016/j.devcel.2018.08.001. Baena-Lopez, Luis Alberto, Cyrille Alexandre, Alice Mitchell, Laurynas Pasakarnis, and Jean Paul Vincent. 2013. “Accelerated Homologous Recombination and Subsequent Genome Modification in Drosophila.” Development (Cambridge) 140(23):4818–25. doi: 10.1242/dev.100933. Benitez, Sergio, Claudia Sosa, Nicolás Tomasini, and Ana Macías. 2010. “Both JNK and Apoptosis Pathways Regulate Growth and Terminalia Rotation during Drosophila Genital Disc Development.” International Journal of Developmental Biology 54(4):643–53. doi: 10.1387/ijdb.082724sb. Böhni, Ruth, Juan Riesgo-Escovar, Sean Oldham, Walter Brogiolo, Hugo Stocker, Bernard F. Andruss, Kathy Beckingham, and Ernst Hafen. 1999. “Autonomous Control of Cell and Organ Size by CHICO, a Drosophila Homolog of Vertebrate IRS1-4.” Cell 97(7):865–75. doi: 10.1016/S0092-8674(00)80799-0. Bosch, Pablo Sanchez, Bomsoo Cho, and Jeffrey D. Axelrod. n.d. “Flamingo Participates in Multiple Models of Cell Competition.” doi: 10.1101/2023.09.24.559197. Brás-Pereira, Catarina, and Eduardo Moreno. 2018. “Mechanical Cell Competition.” Current Opinion in Cell Biology 51:15–21. doi: 10.1016/j.ceb.2017.10.003. Coelho, Dina S., and Eduardo Moreno. 2019. “Emerging Links between Cell Competition and Alzheimer’s Disease.” Journal of Cell Science 132(13). doi: 10.1242/JCS.231258. Coelho, Dina S., and Eduardo Moreno. 2020. “Neuronal Selection Based on Relative Fitness Comparison Detects and Eliminates Amyloid-β-Induced Hyperactive Neurons in Drosophila.” IScience 23(9):101468. doi: 10.1016/j.isci.2020.101468. Coelho, Dina S., Silvia Schwartz, Marisa M. Merino, Barbara Hauert, Barbara Topfel, Colin Tieche, Christa Rhiner, and Eduardo Moreno. 2018. “Culling Less Fit Neurons Protects .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 10 against Amyloid-β-Induced Brain Damage and Cognitive and Motor Decline.” Cell Reports 25(13):3661-3673.e3. doi: 10.1016/j.celrep.2018.11.098. Deligiannaki, M., A. L. Casper, C. Jung, and U. Gaul. 2015. “Pasiflora Proteins Are Novel Core Components of the Septate Junction.” Development 142(17):3046–57. doi: 10.1242/dev.119412. Eisenhoffer, George T., Patrick D. Loftus, Masaaki Yoshigi, Hideo Otsuna, Chi Bin Chien, Paul A. Morcos, and Jody Rosenblatt. 2012. “Crowding Induces Live Cell Extrusion to Maintain Homeostatic Cell Numbers in Epithelia.” Nature 484(7395):546–49. doi: 10.1038/nature10999. Germani, Federico, Daniel Hain, Denise Sternlicht, Eduardo Moreno, and Konrad Basler. 2018. “The Toll Pathway Inhibits Tissue Growth and Regulates Cell Fitness in an Infection- Dependent Manner.” ELife 7:1–10. doi: 10.7554/eLife.39939. Grabner, A., S. Brast, S. Sucic, S. Bierer, B. Hirsch, H. Pavenstädt, H. H. Sitte, E. Schlatter, and G. Ciarimboli. 2011. “LAPTM4A Interacts with HOCT2 and Regulates Its Endocytotic Recruitment.” Cellular and Molecular Life Sciences 68(24):4079–90. doi: 10.1007/s00018- 011-0694-6. Guruharsha, K. G., Jean-François Rual, Bo Zhai, Julian Mintseris, Pujita Vaidya, Namita Vaidya, Chapman Beekman, Christina Wong, David Y. Rhee, Odise Cenaj, Emily McKillip, Saumini Shah, Mark Stapleton, Kenneth H. Wan, Charles Yu, Bayan Parsa, Joseph W. Carlson, Xiao Chen, Bhaveen Kapadia, K. VijayRaghavan, Steven P. Gygi, Susan E. Celniker, Robert A. Obar, and Spyros Artavanis-Tsakonas. 2011. “A Protein Complex Network of Drosophila Melanogaster.” Cell 147(3):690–703. doi: 10.1016/j.cell.2011.08.047. Hogue, Douglas L., Lilli Kerby, and Victor Ling. 1999. “A Mammalian Lysosomal Membrane Protein Confers Multidrug Resistance upon Expression in Saccharomyces Cerevisiae *.” 274(18):12877–82. Huang, Juan, Wenke Zhou, Wei Dong, Annie M. Watson, and Yang Hong. 2009. “Directed, Efficient, and Versatile Modifications of the Drosophila Genome by Genomic Engineering.” Proceedings of the National Academy of Sciences of the United States of America 106(20):8284–89. doi: 10.1073/pnas.0900641106. Igaki, Tatsushi, Hiroshi Kanda, Yuki Yamamoto-Goto, Hirotaka Kanuka, Erina Kuranaga, Toshiro Aigaki, and Masayuki Miura. 2002. “Eiger, a TNF Superfamily Ligand That Triggers the Drosophila JNK Pathway.” The EMBO Journal 21(12):3009–18. doi: 10.1093/emboj/cdf306. Kondo, Shu, and Ryu Ueda. 2013. “Highly Improved Gene Targeting by Germline-Specific Cas9 Expression in Drosophila.” Genetics 195(3):715–21. doi: 10.1534/genetics.113.156737. Levayer, Romain, Carole Dupont, and Eduardo Moreno. 2016. “Tissue Crowding Induces Caspase-Dependent Competition for Space.” Current Biology 26(5):670–77. doi: 10.1016/j.cub.2015.12.072. Madan, Esha, Christopher J. Pelham, Masaki Nagane, Taylor M. Parker, Rita Canas-Marques, Kimberly Fazio, Kranti Shaik, Youzhong Yuan, Vanessa Henriques, Antonio Galzerano, .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 11 Tadashi Yamashita, Miguel Alexandre Ferreira Pinto, Antonio M. Palma, Denise Camacho, Ana Vieira, David Soldini, Harikrishna Nakshatri, Steven R. Post, Christa Rhiner, Hiroko Yamashita, Davide Accardi, Laura A. Hansen, Carlos Carvalho, Antonio L. Beltran, Periannan Kuppusamy, Rajan Gogna, and Eduardo Moreno. 2019. “Flower Isoforms Promote Competitive Growth in Cancer.” Nature 572(7768):260–64. doi: 10.1038/s41586- 019-1429-3. Merino, Marisa M., Christa Rhiner, Jesus M. Lopez-Gay, David Buechel, Barbara Hauert, and Eduardo Moreno. 2015. “Elimination of Unfit Cells Maintains Tissue Health and Prolongs Lifespan.” Cell 160(3):461–76. doi: 10.1016/j.cell.2014.12.017. Merino, Marisa M., Christa Rhiner, Marta Portela, and Eduardo Moreno. 2013. “‘Fitness Fingerprints’ Mediate Physiological Culling of Unwanted Neurons in Drosophila.” Current Biology 23(14):1300–1309. doi: 10.1016/j.cub.2013.05.053. Morata, G., and P. Ripoll. 1975. “Minutes: Mutants of Drosophila Autonomously Affecting Cell Division Rate.” Developmental Biology 42(2):211–21. doi: 10.1016/0012-1606(75)90330-9. Moreno, Eduardo, and Konrad Basler. 2004. “DMyc Transforms Cells into Super-Competitors.” Cell 117(1):117–29. doi: 10.1016/s0092-8674(04)00262-4. Moreno, Eduardo, Konrad Basler, and Ginés Morata. 2002. “Cells Compete for Decapentaplegic Survival Factor to Prevent Apoptosis in Drosophila Wing Development.” Nature 416(6882):755–59. doi: 10.1038/416755a. Moreno, Eduardo, Yuniel Fernandez-Marrero, Patricia Meyer, and Christa Rhiner. 2015. “Brain Regeneration in Drosophila Involves Comparison of Neuronal Fitness.” Current Biology 25(7):955–63. doi: 10.1016/j.cub.2015.02.014. Moreno, Eduardo, and Christa Rhiner. 2014. “Darwin’s Multicellularity: From Neurotrophic Theories and Cell Competition to Fitness Fingerprints.” Current Opinion in Cell Biology 31(1):16–22. doi: 10.1016/j.ceb.2014.06.011. Moreno, Eduardo, Léo Valon, Florence Levillayer, and Romain Levayer. 2019. “Competition for Space Induces Cell Elimination through Compaction-Driven ERK Downregulation.” Current Biology 29:1–12. doi: 10.1016/j.cub.2018.11.007. Oliver, Edward R., Thomas L. Saunders, Susan A. Tarlé, and Tom Glaser. 2004. “Ribosomal Protein L24 Defect in Belly Spot and Tail (Bst), a Mouse Minute.” Development 131(16):3907–20. doi: 10.1242/dev.01268. Petrova, Evgeniya, Jesús M. López-Gay, Christa Rhiner, and Eduardo Moreno. 2012. “Flower- Deficient Mice Have Reduced Susceptibility to Skin Papilloma Formation.” DMM Disease Models and Mechanisms 5(4):553–61. doi: 10.1242/dmm.008623. Portela, Marta, Sergio Casas-Tinto, Christa Rhiner, Jesús M. López-Gay, Orlando Domínguez, Davide Soldini, and Eduardo Moreno. 2010. “Drosophila SPARC Is a Self-Protective Signal Expressed by Loser Cells during Cell Competition.” Developmental Cell 19(4):562– 73. doi: 10.1016/j.devcel.2010.09.004. Rhiner, Christa, Jesús M. López-Gay, Davide Soldini, Sergio Casas-Tinto, Francisco A. Martín, Luis Lombardía, and Eduardo Moreno. 2010. “Flower Forms an Extracellular Code That .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 12 Reveals the Fitness of a Cell to Its Neighbors in Drosophila.” Developmental Cell 18(6):985–98. doi: 10.1016/j.devcel.2010.05.010. Tan, Lihua, Paul Schedl, Ho Juhn Song, Dan Garza, and Mary Konsolaki. 2008. “The Toll→NFκB Signaling Pathway Mediates the Neuropathological Effects of the Human Alzheimer’s Aβ42 Polypeptide in Drosophila.” PLoS ONE 3(12). doi: 10.1371/journal.pone.0003966. Vaughen, John, and Tatsushi Igaki. 2016. “Slit-Robo Repulsive Signaling Extrudes Tumorigenic Cells from Epithelia.” Developmental Cell 39(6):683–95. doi: 10.1016/j.devcel.2016.11.015. Villa del Campo, Cristina, Cristina Clavería, Rocío Sierra, and Miguel Torres. 2014. “Cell Competition Promotes Phenotypically Silent Cardiomyocyte Replacement in the Mammalian Heart.” Cell Reports 8(6):1741–51. doi: 10.1016/j.celrep.2014.08.005. Walderich, Brigitte, Ajeet Pratap Singh, Prateek Mahalwar, and Christiane Nüsslein-Volhard. 2016. “Homotypic Cell Competition Regulates Proliferation and Tiling of Zebrafish Pigment Cells during Colour Pattern Formation.” Nature Communications 7:1–15. doi: 10.1038/ncomms11462. With, Sheila, Tiffany Rice, Claudia Salinas, and Vanessa Auld. 2003. “Fire Exit Is a Potential Four Transmembrane Protein Expressed in Developing Drosophila Glia.” Genesis 35(3):143–52. doi: 10.1002/gene.10177. Yamamoto, Masatoshi, Shizue Ohsawa, Kei Kunimasa, Tatsushi Igaki, and Jun N-terminal. 2017. “The Ligand Sas and Its Receptor PTP10D Drive Tumour-Suppressive Cell Competition.” Nature 542:246–50. doi: 10.1038/nature21033. Yao, Chi Kuang, Yong Qi Lin, Cindy V. Ly, Tomoko Ohyama, Claire M. Haueter, Vera Y. Moiseenkova-Bell, Theodore G. Wensel, and Hugo J. Bellen. 2009. “A Synaptic Vesicle- Associated Ca2+ Channel Promotes Endocytosis and Couples Exocytosis to Endocytosis.” Cell 138(5):947–60. doi: 10.1016/j.cell.2009.06.033. Yao, Chi Kuang, Yu Tzu Liu, I. Chi Lee, You Tung Wang, and Ping Yen Wu. 2017. “A Ca2+channel Differentially Regulates Clathrin-Mediated and Activity-Dependent Bulk Endocytosis.” PLoS Biology 15(4). doi: 10.1371/journal.pbio.2000931. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 13 Figures Figure 1 – Ikebana is basally expressed in several cells of the wing imaginal disc and adult brain (A) Predicted configuration of the Ikebana protein at the cell membrane. The green letters a-d indicate the four transmembrane domains represented by the green letters a-d. Model done with Protter. (B) Strategy for the generation of the ikebana{KO}L and, consequently, the ikebana{KO; KI -LexA::p65} transgenic lines. The entire ikebana coding sequence in the genome is removed by CRISPR with two sgRNA (Baena-Lopez et al. 2013) . Then the genome is repaired, by homologous recombination, allowing the

of the 3xPax3-mCherry selection marker (Founder line 1) (Huang et al. 2009) . For the ikebana{KO}L transgenic line, the 3xPax3-mCherry selection marker was removed by Cre-lox recombination. For the generation of the ikebana{KO; KI -LexA::p65} transgenic line, a knockin construct containing the LexA::p65 sequence under the control of the endogenous ikebana promoter, was integrated into the ikebana knockout locus. The pTV3 vector backbone was removed in the final knockin line. (C-D) Expression of Ikebana in the wing imaginal discs of third-instar larvae (C) and the adult brain (D). GFP, in green, marks the cells trying to express ikebana, and DAPI, in blue, marks the cell nuclei. Scale bars, 50µm. Genotype: ywF; ikebana{KO; KI-LexA::p65}/+; 26xLexAop-CD8::GFP/+. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 14 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 15 Figure 2 – Ikebana partially protects loser cells from elimination (A-C) Supercompetition assay in the wing imaginal discs of third instar larvae in which loser wild-type clones (GFP) are outcompeted by dmyc overexpressing cells (Rhiner et al. 2010). UAS-white-RNAi (A), UAS-flower lose A/B-RNAi (B) and UAS-ikebana-RNAi (C) are expressed in the loser clones 72h after clone induction (ACI). DAPI is represented in blue. Scale bars, 100 µm. (D) Quantification of the clone area over the total area of the wing pouch when UAS-white-RNAi, UAS-flower lose A/B-RNAi, and UAS-ikebana-RNAi are expressed in the loser clones 72h ACI. We normalised the ratio at 12 hours ACI (data not shown), and subsequently, the 72h time point is normalised relative to their respective conditions at 12 hours. (E-G) Supercompetition assay in the wing imaginal discs of third instar larvae in which loser wild-type clones (GFP) are outcompeted by dmyc overexpressing cells (Rhiner et al. 2010). UAS-lacZ (A), UAS-diap1 (B) and UAS-ikebana (C) are expressed in the loser clones 72h ACI. DAPI is represented in blue. Scale bars, 100 µm. (H) Quantification of the clone area over the total area of the wing pouch when UAS-lacZ, UAS-diap1, and UAS-ikebana are expressed in the loser clones 72h ACI. We normalised the ratio at 24 hours ACI (data not shown), and subsequently, the 72h time point is normalised relative to their respective conditions at 24 hours. The numbers after the genotypes indicate the number of discs analysed. Error bars indicate SD; *P<0.05; ***P<0.001; ****P<0.0001. Statistical significance between groups was calculated using the nonparametric Kruskal-Wallis test and a Dunn’s test was applied for multiple comparisons between genotypes. Genotypes: ywF/w; tub>dmyc>Gal4, UAS-GFP/+; UAS-white-RNAi/MKRS (A); ywF/w; tub>dmyc>Gal4, UAS- GFP/+; UAS-flowerloseA/B-RNAi/MKRS (B); ywF/w; tub>dmyc>Gal4, UAS-GFP/UAS-ikebana-RNAi; MKRS/+ (C); ywF/w; tub>dmyc>Gal4, UAS-GFP/UAS-lacZ; MKRS/+ (E); ywF/w; tub>dmyc>Gal4, UAS-GFP/+; UAS- diap1/MKRS (F); ywF/w; tub>dmyc>Gal4, UAS-GFP/+; UAS-ikebana/MKRS (G); .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 16 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 17 Figure 3 – Ikebana modulates Flower LoseB expression (A) Competition assay in the wing imaginal discs of third instar larvae in which wild-type clones (GFP) are in a

of yellow expressing cells (Rhiner et al. 2010) . UAS-white-RNAi (A), UAS-flowerloseB (B) and UAS-ikebana-RNAi (C) are expressed in the wild -type clones 72h ACI. DAPI is represented in blue. Scale bars, 100 µm. Genotypes: ywF; act>y>Gal4, UAS-GFP/+; UAS-white-RNAi/MKRS (A); ywF; act>y>Gal4, UAS-GFP/UAS- flowerloseB; MKRS/+ (B); ywF; act>y>Gal4, UAS-GFP/UAS-ikebana-RNAi; MKRS/+ (C) (D) Quantification of the clone area over the total area of the wing pouch UAS-white-RNAi, UAS-flowerloseB and UAS-ikebana-RNAi are expressed in the loser clones 72h ACI. We normalised the ratio at 24 hours ACI (data not shown), and subsequently, the 72h time point is normalised relative to their respective conditions at 24 hours. (E) Strategy for the generation of the ikebana{KO}T transgenic line. The entire ikebana coding sequence in the genome is removed by CRISPR with two sgRNA (Baena-Lopez et al. 2013). Then the genome is repaired, by homologous recombination, allowing the introduction of the 3xPax3-RFP selection marker (Founder line 2) (Huang et al. 2009). The 3xPax3-RFP marker was removed by Cre-lox recombination. (F-I) Expression of Flower LoseB and Dcp1 in ikebana KO adult optic lobes. Optic lobe of control flies (F); ikebana{KO} (G); GMR>Aβ42 (H) and GMR>Aβ42, ikebana{KO} (I). Flower LoseB is marked in red; a-Dcp1 is marked in magenta; DAPI, in blue, marks the cell nuclei. 41-μm image projections. Scale bars, 50 μm. Genotypes: w; ; flower{KO; KI-flowerloseB::mCherry}/+ (F); w; ikebana{KO}T/ikebana{KO}L; flower{KO; KI- flowerloseB::mCherry}/+ (G); w; GMR-Gal4, UAS -Ab42/+; flower{KO; KI -flowerloseB::mCherry}/+ (H); w; GMR-Gal4, UAS-Ab42, ikebana{KO}T/ikebana{KO}L; flower{KO; KI-flowerloseB::mCherry}/+ (I). (J) Quantification of the number of Flower LoseB positive cells in the optic lobes normalised against the control (ctr) +. (K) Quantification of the number of Dcp1 positive cells in the optic lobes normalised against the ctr +. (L-O) Expression of Flower LoseB in the adult optic lobes when ikebana is overexpressed. Optic lobe of control flies overexpressing GFP (L) or ikebana (M), or flies expressing the A β42 peptide in the GMR region, also overexpressing GFP (N) or ikebana (O). Flower LoseB is marked in red; DAPI, in blue, marks the cell nuclei. 41-μm image projections. Scale bars, 50 μm. Genotypes: w; GMR-Gal4/UAS-CD8::GFP; flower{KO; KI -flowerloseB::mCherry}/+ (L); w; GMR-Gal4/+; flower{KO; KI-flowerloseB::mCherry}/UAS-ikebana (M); w; GMR-Gal4, UAS-Ab42/UAS-CD8::GFP; flower{KO; KI- flowerloseB::mCherry}/+ (N); w; GMR-Gal4, UAS-Ab42/+; flower{KO; KI -flowerloseB::mCherry}/UAS- ikebana (O). .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 18 (P) Quantification of the number of Flower LoseB positive cells normalised against the control GFP in the ctr scenario. The numbers after the genotypes indicate the number of discs or optic lobes analysed. Error bars indicate SD; ns indicates non -significant; *P<0.05; **P<0.01; ***P<0.001; ****P<0.0001. Statistical significance between groups was calculated using the nonparametric Kruskal-Wallis test and a Dunn’s test was applied for multiple comparisons between genotypes (D, J and K). Statistical significance between groups was calculated using the Brown-Forsythe test and a Welch ANOVA test and Dunnette’s T3 was applied for multiple comparisons between genotypes (P). .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 19 Figure 4 - Model of the action of Ikebana in cell competition When an insult occurs, ikebana is downregulated, causing the suboptimal cells to express Flower LoseB and be labelled as loser (A). The loser cells that are not supposed to be eliminated express ikebana again and revert their loser status (B). The loser cells that should be eliminated express Azot and activate the apoptotic machinery (C). The loser cells are eliminated and, in some cases, replaced by fitter cells (D). Model done with Biorender.com. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 20 .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 21 Figure S1 – Ikebana is not a general regulator of apoptosis (A-F) Representative eyes of Drosophila when UAS-white-RNAi (A), UAS-diap1 (B), UAS-eiger-RNAi (C), UAS-azot-RNAi (D), UAS-flowerloseA/B-RNAi (E) and UAS-ikebana-RNAi (F) are co-expressed with Eiger in the GMR domain. Genotypes: yw/w; GMR-Gal4, UAS-eiger/+; UAS -white-RNAi/+ (A); yw/w; GMR -Gal4, UAS -eiger/+; UAS- diap1/+ (B); yw/w; GMR-Gal4, UAS-eiger/+; UAS-eiger-RNAi/+ (C) yw/w; GMR-Gal4, UAS-eiger/+; UAS-azot- RNAi/+ (D); yw/w; GMR-Gal4, UAS -eiger/+; UAS -flowerloseA/B-RNAi/+; (E) yw/w; GMR-Gal4, UAS - eiger/UAS-ikebana-RNAi; +/+ (F). (G) Quantification of the eye area of the flies (A-F) normalised for white-RNAi. (H-K) Representative eyes of Drosophila when UAS-white-RNAi (H), UAS-lacZ (I), UAS-diap1 (J) and UAS- ikebana (K) are co-expressed with Eiger in the GMR domain. Genotypes: yw/w; GMR-Gal4, UAS-eiger/+; UAS-white-RNAi/+ (H); yw/w; GMR-Gal4, UAS-eiger/UAS-lacZ; +/+ (I); yw/w; GMR-Gal4, UAS-eiger/+; UAS-diap1/+ (J); yw/w; GMR-Gal4, UAS-eiger/+; UAS-ikebana/+ (K); (K) Quantification of the eye area of the flies (H-K) normalised for white-RNAi. Horizontal line when eye area = 1 arbitrary unit (a.u) to show the differences of the other genotypes against the control. The numbers after the genotypes indicate the number of eyes analysed. Error bars indicate SD; ***P<0.001; ****P<0.0001. Statistical significance between groups was calculated using the nonparametric Kruskal-Wallis test and a Dunn’s test was applied for multiple comparisons between genotypes. (M-S) Representative images of the genitalia of the fly when UAS-white-RNAi (M), UAS-lacZ (N), UAS-diap1 (O), UAS-azot-RNAi (P), UAS-flowerloseA/B-RNAi (Q), UAS-ikebana-RNAi (R) and UAS -ikebana (S) are expressed under the control of the engrailed driver Genotypes: ywF/w; engrailed-Gal4, UAS-GFP/+; UAS-white-RNAi/+ (M); ywF/w; engrailed-Gal4, UAS- GFP/UAS-lacZ; +/+ (N); ywF/w; engrailed-Gal4, UAS-GFP/+; UAS-diap1/+ (O); ywF/w; engrailed-Gal4, UAS- GFP/+; UAS-azot-RNAi/+ (P); ywF/w; engrailed-Gal4, UAS-GFP/+; UAS-flowerloseA/B-RNAi/+ (Q); ywF/w; engrailed-Gal4, UAS-GFP/UAS-ikebana-RNAi; +/+ (R); ywF/w; engrailed-Gal4, UAS-GFP/+; UAS-ikebana/+ (S); .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint 22 Figure S2 – Ikebana modulates Azot expression (A-D) Expression of Azot in ikebana KO adult optic lobes. Optic lobe of control flies (A); ikebana{KO} (B); GMR>Aβ42 (C) and GMR>Aβ42, ikebana{KO} (D). Azot is marked in red; DAPI, in blue, marks the cell nuclei. 41-μm image projections. Scale bars, 50 μm. Genotypes: w; ; azot::mCherry/+ (A); w; ikebana{KO}T/ikebana{KO}L; azot::mCherry/+ (B); w; GMR-Gal4, UAS-Ab42/+; azot::mCherry/+ (C); w; GMR-Gal4, UAS-Ab42, ikebana{KO}T/ikebana{KO}L; azot::mCherry/+ (D). (E) Quantification of the number of Azot positive cells in the optic lobes normalised against the ctr +. The numbers after the genotypes indicate the number of discs or optic lobes analysed. Error bars indicate SD; ns indicates non-significant; ****P<0.0001. Statistical significance between groups was calculated using the nonparametric Kruskal-Wallis test and a Dunn’s test was applied for multiple comparisons between genotypes. .CC-BY-NC-ND 4.0 International licensemade available under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is The copyright holder for this preprintthis version posted April 16, 2024. ; https://doi.org/10.1101/2024.04.13.589362doi: bioRxiv preprint