S -nitrosylation of the CLIP-protease, Persephone, Regulates Drosophila innate immunity

63 Nitric oxide (NO) modulates innate immunity, but its molecular targets in Drosophila 64 melanogaster are largely undefined. A key mechanism of NO signalling is S-65 nitrosylation, the modification of cysteine thiols. The homeostasis of S-nitrosylation is 66 maintained by S-nitrosoglutathione reductase (Gsnor), encoded by the fdh gene in D. 67 melanogaster. As reduced Gsnor activity enhances pathogen sensitivity in plants, we 68 investigated its role in D. melanogaster. Here, we show that flies lacking fdh exhibit 69 increased susceptibility to the fungus Beauveria bassiana and the bacterium 70 Staphylococcus aureus, pathogens combatted by the Toll pathway. This immune 71 deficiency correlates with impaired Toll-dependent antimicrobial peptide expression. 72 We demonstrate that the Toll pathway protease Persephone (Psh) is S-nitrosylated in 73 vivo and that loss of Gsnor prevents its proteolytic activation following infection. We 74 propose a model where Gsnor-mediated regulation of NO is essential for Toll 75 activation, preventing excessive S-nitrosylation of Psh and ensuring a robust immune 76 response. 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint 95

96 The innate immune system is the evolutionarily ancient, first line of defence against 97 invading pathogens, and its core mechanisms are conserved across all metazoan life1. 98 The fruit fly, Drosophila melanogaster, has long been a genetic model for dissecting 99 the molecular architecture of innate immunity2. Its relative simplicity, characterized 100 by the absence of a confounding adaptive immune system, and the profound 101 evolutionary conservation of its core signalling pathways have yielded foundational 102 insights into mammalian immunity, most notably the discovery of Toll-like receptors 103 (TLRs) and their role in pathogen recognition3. 104 In D. melanogaster, host defence against systemic infection is orchestrated primarily 105 by two major humoral signalling cascades: the Toll pathway and the Immune 106 Deficiency (IMD) pathway4. These pathways exhibit a remarkable degree of 107 specificity, with the Toll pathway mounting a defence predominantly against fungi 108 and Gram-positive bacteria, while the IMD pathway is activated mainly in response to 109 Gram-negative bacteria5. Activation of these cascades culminates in the nuclear 110 translocation of distinct NF-κB transcription factors, Dif and Dorsal for the Toll 111 pathway, and Relish (rel) for the IMD pathway, which in turn drive the expression of a 112 battery of effector genes, including those encoding potent antimicrobial peptides 113 (AMPs) such as Drosomycin (Drs) and Metchnikowin (Mtk)6. 114 The activation of the Toll pathway is a sophisticated process initiated by two 115 mechanistically distinct upstream branches. The canonical branch involves the direct 116 recognition of pathogen-associated molecular patterns (PAMPs), such as Lys-type 117 peptidoglycan from Gram-positive bacteria or β-glucan from fungi, by circulating 118 pattern recognition receptors (PRRs) like PGRP-SA and GNBP3. This recognition event 119 triggers a self-amplifying extracellular serine protease cascade that ultimately leads 120 to the cleavage and activation of the cytokine-like ligand, Spätzle (Spz), which then 121 binds to the Toll receptor7. In parallel, a second, non-canonical branch functions as a 122 "danger-sensing" system. This pathway is not triggered by static molecular patterns 123 but rather by the functional activity of virulence factors, specifically the proteases 124 secreted by invading microbes. At the heart of this danger-sensing module is 125 Persephone (Psh), a CLIP-domain serine protease that circulates as an inactive 126 zymogen. The pro-Persephone contains a unique "bait" region that is susceptible to 127 cleavage by a broad range of microbial proteases8. This initial cleavage licenses pro-128 Psh for subsequent maturation by endogenous host proteases, unleashing its active 129 form and initiating the downstream cascade that converges on Spz processing8, 9, 10. 130 This dual-input architecture allows the Toll pathway to respond not only to the 131 presence of microbes but also to their pathogenic actions. 132 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint Superimposed upon these well-defined signalling networks are layers of regulation 133 mediated by pleiotropic signalling molecules. One such molecule is nitric oxide (NO), 134 a highly reactive and diffusible gas radical that functions as a key signalling 135 intermediate in a vast array of physiological processes, from neurotransmission to 136 vasodilation and immunity11, 12. A primary mechanism through which NO exerts its 137 biological effects is S-nitrosylation, the covalent addition of a NO moiety to the thiol 138 group of a reactive cysteine residue, forming an S-nitrosothiol (SNO). This reversible, 139 redox-based post-translational modification acts as a molecular switch, altering 140 protein activity, localization, and stability13. Cellular homeostasis of NO and, by 141 extension, global levels of protein S-nitrosylation are tightly controlled by the enzyme 142 S-nitrosoglutathione reductase (Gsnor). Gsnor catalyzes the metabolism of S-143 nitrosoglutathione (GSNO), the most abundant low-molecular-weight SNO and a 144 major biological reservoir of NO bioactivity. Consequently, mutants with no GSNOR 145 activity provide a powerful tool to investigate the physiological consequences of 146 elevated global S-nitrosylation14, 15. 147 While NO has been implicated in D. melanogaster immunity16, its specific molecular 148 targets and precise regulatory functions, particularly within the complex proteolytic 149 cascades of the Toll pathway, have remained largely undefined. Work in other 150 organisms, such as plants, has revealed that loss of Gsnor can paradoxically either 151 enhance or compromise disease resistance, suggesting its role is highly context-152 dependent14, 17. This ambiguity highlights a significant gap in our understanding of 153 how redox signalling is integrated with innate immune activation. Here, we 154 hypothesize that Gsnor-mediated control of NO homeostasis serves as a critical 155 regulatory checkpoint for the D. melanogaster Toll pathway. We propose that under 156 conditions of Gsnor deficiency, excessive S-nitrosylation directly targets and inhibits a 157 key component of the Toll-activating protease cascade. This study sought to identify 158 this molecular target, elucidate the mechanism of inhibition, and explore the broader 159 physiological and evolutionary implications of this novel regulatory axis. 160 161

162 Loss of GSNOR Function Compromises Immunity to Toll-Dependent Pathogens 163 To investigate whether S-nitrosylation has a role in the immune response of D. 164 melanogaster adults, we reduced Gsnor activity using mutations in the formaldehyde 165 dehydrogenase gene fdh (CG6958) that codes for the enzyme with Gsnor activity18, 19. 166 At the start of this investigation there were no null alleles of fdh available so we 167 generated flies heterozygous for deficiencies (deletions) Df(3R)Exel7305 (Df7305) and 168 Df(3R)Exel7306 (Df7306)20 that have breakpoints that overlap within fdh (Fig. 1A). 169 These flies have no intact copy of fdh (Fig. 1B) and have reduced Gsnor activity that is 170 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint restored if Gsnor is expressed from an fdh transgene using the GAL4-UAS system21 171 (Fig. 1C). Females trans-heterozygous for these overlapping fdh deletions have 172 increased levels of nitrite, a proxy NO (Fig. 1D) and also show reduced survival as 173 compared with wild-type (Oregon R) after being fed a 5% sucrose solution containing 174 5mM sodium nitroprusside (SNP) a phenotype that is also rescued by expression of 175 Gsnor from an fdh transgene (Fig. 1E). We have since used CRISPR/Cas922 to replace 176 the coding sequence of fdh with the 3xP3-DsRed reporter23 resulting in the knockout 177 of the gene. Three independent lines were generated, fdhcp1, fdhcp2, and fdhcp3 (Fig. 178 S1A). Flies homozygous for these alleles have no intact copy of fdh (Fig. S1B), have 179 no detectable Gsnor activity (Fig. S1C) and are sensitive to SNP (Fig. S1D). 180 Having established these genetic tools, we challenged the fdh mutants with fungal 181 pathogen Beauveria bassiana, the Gram-positive bacterium Staphylococcus aureus or 182 the Gram-negative bacterium Escherichia coli. Wild type D. melanogaster females 183 have been reported by others to be more sensitive to B. bassiana than males24. We 184 have confirmed this (Fig. S2A), and the experiments reported here were performed 185 with flies of a single sex. 186 Flies either heterozygous for deficiencies Df7305 and Df7306 (Df7305/Df7306) (Fig. 187 2A) or homozygous for the CRISPR mutation fdhcp2 (Fig. S1E) are more susceptible 188 than wild type to challenge with B. bassiana and S. aureus (Fig. 2B) and in both cases 189 this is rescued by expression of Gsnor from an fdh transgene (Fig. S2A and 2B) 190 suggesting that Gsnor is required for the activation of the Toll pathway in flies. 191 Activation of the Toll pathway results in increased expression of the anti-microbial 192 peptides Drosomycin (Drs) and Metchnikowin (Mtk)3, 25, 26 and as expected 193 accumulation of transcripts coding for these peptides increases after infection of wild 194 type flies with B. bassiana but this is reduced in fdh mutants (Fig. 2D and 2E). 195 In contrast to their response to challenge by B. bassiana or S. aureus, Df7305/Df7306 196 flies are no more sensitive than wild type to septic infection with the Gram-negative 197 bacterium E. coli (Fig. 2C) and much less so than flies with mutations in either IMD or 198 rel (relish) that code for proteins in the IMD pathway suggesting that Gsnor is not 199 required for the activation of the IMD pathway. However, it has been suggested that 200 NO is required for IMD activity as inhibition of nitric oxide synthesis increases the 201 sensitivity of D. melanogaster larvae and adults to infection with the Gram-negative 202 bacterium Erwinia carotovora carotovora 16. 203 204 Genetic Epistasis Places GSNOR Function Within the Persephone "Danger-Sensing" 205 Branch 206 The Toll pathway can be activated by PAMP recognition7, 9, 27, 28, 29 or by sensing 207 pathogen-derived proteolytic activity via Persephone (Psh)7, 8, 9, 10, 30. To pinpoint 208 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint where Gsnor-mediated regulation occurs within this architecture, we performed 209 genetic epistasis analysis using mutants in the gene necrotic (nec). Nec is a serine 210 protease inhibitor (serpin) that negatively regulates the Toll cascade, and nec loss-of-211 function mutants exhibit constitutive Toll pathway activation, resulting in 212 spontaneous melanisation and reduced survival10, 31, 32. This phenotype is known to 213 be entirely dependent on a functional Psh protease10. 214 Remarkably, we have found that the Nec phenotype is also suppressed if Gsnor 215 activity is reduced as flies with both nec and fdh mutations show increased survival 216 and reduced melanisation as compared with flies that are nec mutant but fdh+ (Fig. 217 3A, 3B and Fig. S2D). Since Nec inhibits the cascade downstream of Psh activation, 218 the most parsimonious model is that excessive S-nitrosylation directly inhibits Psh or 219 prevents its activation, thereby blocking the signalling cascade even when the 220 upstream serpin brake is removed. 221 To further investigate whether the PAMP-recognition or danger-sensing branch was 222 affected, we challenged flies with heat-killed pathogens. Unlike infection with live 223 microbes, exposure to heat-inactivated B. bassiana (Fig. S2C) or S. aureus (Fig. S2B) 224 caused little to no mortality in either wild-type or fdh mutant flies. This result 225 demonstrates that the hyper-susceptibility of fdh mutants is not due to a defect in 226 the recognition of structural PAMPs on the pathogen surface but is instead 227 dependent on the presence of active virulence factors, such as secreted proteases, 228 produced by live, metabolically active pathogens. Collectively, these genetic data 229 strongly focussed our investigation onto the Psh-mediated danger-sensing branch of 230 the Toll pathway as the locus of NO-mediated regulation. 231 232 Persephone Is a Direct Target of S-nitrosylation in vivo 233 The genetic evidence strongly implicated the Psh protease as the key node for NO-234 mediated regulation. To determine if this genetic link reflects a direct biochemical 235 interaction, we linked a human influenza haemagglutinin peptide (HA) to the C-236 terminus of Psh and expressed the tagged protein (Psh-HA) in fdh+/+ flies using the 237 Gal4/UAS system21. Expression of Psh-HA was confirmed by Western blotting with 238 anti-HA antibody (Fig. 3C) and its S-nitrosylation was investigated using the Biotin 239 Switch Technique (BST) that replaces SNO in S-nitrosylated proteins with biotin33, 34. 240 Following application of the BST to proteins extracts of wild type flies biotinylated 241 proteins were pulled down with streptavidin coated beads and screened for 242 biotinylated Psh-HA in Western Blots probed with anti-HA antibody. The results 243 show that Psh-HA can be S-nitrosylated in vivo (Fig. 3D), providing direct biochemical 244 evidence that Psh is not merely downstream of an NO-sensitive process but is itself a 245 bona fide substrate for S-nitrosylation within the fly. This finding forges a crucial 246 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint molecular link between the whole-organism immune phenotype and the central 247 hypothesis of direct regulation by a post-translational modification. 248 S-nitrosylation Impedes the Proteolytic Activation of pro-Persephone 249 Having established that Psh can be S-nitrosylated, we next sought to determine the 250 functional consequence of this modification. Previous study demonstrated B. 251 bassiana effector protease PR1 cleaves pro-Psh in vitro between residues 115 and 252 116 and Psh is further processed in vivo by the endogenous cathepsin 26-29-p8. We 253 hypothesized that S-nitrosylation might inhibit the Toll pathway by preventing this 254 critical activation step. To test this, we monitored the processing of Psh-HA in wild-255 type (fdh+/+) and GSNOR-deficient (fdhcp2/cp2) flies following infection with B. 256 bassiana. 257 In protein extracts from uninfected flies of either genotype, Psh-HA was detected 258 exclusively as the full-length pro-protein. However, upon infection with B. bassiana, a 259 significant portion of pro-Psh-HA in wild-type flies was cleaved, yielding a smaller, 260 faster-migrating band corresponding to the cleaved Psh-HA. This infection-dependent 261 processing was dramatically impaired in the fdhcp2/cp2 mutants. In these flies, the 262 amount of cleaved Psh-HA generated post-infection was markedly reduced compared 263 to that seen in wild-type flies (Fig. 3E). 264 This observation was further supported by measurements of NO levels during 265 infection. In wild-type flies, the endogenous Gsnor activity appeared sufficient to 266 buffer any immune-induced NO production, as systemic nitrite levels remained stable 267 after infection. In contrast, in fdh mutants, nitrite levels, already elevated at baseline, 268 increased further following infection, indicating that Gsnor is essential for managing 269 the NO surge during an immune challenge (Fig. 1D). These results provide a potential 270 mechanistic explanation for the observed immune deficiency: pathogen infections 271 induce NO production, the loss of Gsnor leads to an uncontrolled increase in total S-272 nitrosylation, which in turn prevents the proteolytic activation of the pro-Psh 273 zymogen, thereby shutting down the danger-sensing arm of the Toll pathway at its 274 point of initiation. 275 276

277 This study identifies a novel redox-based brake on the D. melanogaster Toll pathway, 278 demonstrating that excessive S-nitrosylation prevents the proteolytic activation of 279 the danger-sensing protease Persephone. We propose a model where Gsnor 280 functions as a homeostatic rheostat; by metabolizing GSNO, it keeps cellular S-281 nitrosylation in check, permitting the efficient cleavage of pro-Psh required for a 282 robust immune response to fungal and Gram-positive bacterial pathogens. In Gsnor-283 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint deficient flies, uncontrolled S-nitrosylation of Psh blocks this activation step, leading 284 increased sensitivity of flies to infection by fungal and Gram-positive bacterial 285 pathogens. 286 Our finding that fdh mutants are hyper-susceptible to the entomopathogenic fungus 287 B. bassiana35, 36, 37, 38 provides proof-of-concept for a novel biocontrol strategy. The 288 efficacy of mycoinsecticides is often limited by the host's immune response39, and 289 the strategy of combining microbial agents with stressors that weaken the host is a 290 validated approach in pest management. For instance, the combination of B. 291 bassiana with the insect growth regulator lufenuron, which inhibits chitin synthesis, 292 was significantly more effective at controlling fruit fly larvae than the fungus alone40. 293 We propose a similar synergistic strategy where co-application of B. bassiana with 294 chemical inhibitor of Gsnor could create a potent "one-two punch". While Gsnor is 295 evolutionarily conserved, insect-selective Gsnor inhibitors could be deployed as 296 adjuvants to the pathogen, simultaneously exposing insect to infection while 297 compromising a key component of the insect's defence, offering a rational path 298 towards more effective and sustainable pest management solutions. 299 The regulation of self-amplifying serine protease cascades is fundamental for many 300 biological processes from insect immunity to mammalian blood clotting41, 42. Our 301 discovery that S-nitrosylation plays a key role in regulating a serine protease cascade 302 in D, melanogaster may represent a conserved regulatory principle. Unlike 303 irreversible serpin-based inhibition, regulation by a diffusible gas like NO provides a 304 rapid, transient, and spatially localized feedback mechanism. This suggests that a 305 "redox-protease" regulatory module may be a convergent evolutionary solution for 306 dynamic control of proteolytic cascades, complementing traditional inhibitory 307 mechanisms. 308 The versatility of NO as a signalling molecule is underscored by its opposing roles in 309 D. melanogaster immunity. While our data show excessive NO is detrimental to the 310 Toll pathway, it is required for the IMD pathway's defence against Gram-negative 311 bacteria¹⁶. This apparent paradox is resolved if the functional outcome is dictated by 312 the specific molecular targets modified in each pathway. The inhibitory S-313 nitrosylation of pro-Psh in the Toll pathway contrasts with the presumably activating 314 modifications of yet-unidentified targets in the IMD pathway. This allows a single 315 molecule to differentially tune distinct arms of the immune system, suggesting NO 316 may act as a master arbiter of immune balance, preventing immunopathology by 317 dampening one response while another is engaged. 318 In conclusion, we have defined a novel regulatory axis in which the Gsnor-NO system 319 controls Toll signalling through the direct S-nitrosylation of Persephone. This work 320 not only reveals a new layer of immune homeostasis but also highlights a clear 321 translational potential. 322 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint 323

324 We thank Prof. Matt Tinsley (University of Stirling) for providing Beauveria bassiana. 325 We thank Prof. Bruno Lemaitre (École Polytechnique Fédérale de Lausanne, 326 Switzerland) for providing the Relish, Spätzle, and imd fly lines. We thank Prof. Garry 327 Blakely (University of Edinburgh) for providing Escherichia coli MG1655. We are 328 grateful to the staff of the School of Biological Sciences media and wash-up service 329 for the supply of Drosophila media and clean glassware. This work was financially 330 supported by the Development and Promotion of Science and Technology Talents 331 Project (DPST) and the Institute for the Promotion of Teaching Science and 332 Technology (IPST), Thailand, and by the Darwin Trust of the University of Edinburgh. 333 334

335 1. Bernheim A, Cury J, Poirier EZ. The immune modules conserved across the 336 tree of life: Towards a definition of ancestral immunity. PLoS Biol 22, 337 e3002717 (2024). 338 339 2. Buchon N, Silverman N, Cherry S. Immunity in Drosophila melanogaster--from 340 microbial recognition to whole-organism physiology. Nat Rev Immunol 14, 341 796-810 (2014). 342 343 3. Lemaitre B, Nicolas E, Michaut L, Reichhart JM, Hoffmann JA. The 344 dorsoventral regulatory gene cassette spätzle/Toll/cactus controls the potent 345 antifungal response in Drosophila adults. Cell 86, 973-983 (1996). 346 347 4. Lemaitre B, et al. A recessive mutation, immune deficiency (imd), defines two 348 distinct control pathways in the Drosophila host defense. Proc Natl Acad Sci U 349 S A 92, 9465-9469 (1995). 350 351 5. Lemaitre B, Hoffmann J. The host defense of Drosophila melanogaster. Annu 352 Rev Immunol 25, 697-743 (2007). 353 354 6. Hoffmann JA, Reichhart JM. Drosophila innate immunity: an evolutionary 355 perspective. Nat Immunol 3, 121-126 (2002). 356 357 7. Gottar M, et al. Dual detection of fungal infections in Drosophila via 358 recognition of glucans and sensing of virulence factors. Cell 127, 1425-1437 359 (2006). 360 361 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint 8. Issa N, et al. The Circulating Protease Persephone Is an Immune Sensor for 362 Microbial Proteolytic Activities Upstream of the Drosophila Toll Pathway. Mol 363 Cell 69, 539-550.e536 (2018). 364 365 9. El Chamy L, Leclerc V, Caldelari I, Reichhart JM. Sensing of 'danger signals' and 366 pathogen-associated molecular patterns defines binary signaling pathways 367 'upstream' of Toll. Nat Immunol 9, 1165-1170 (2008). 368 369 10. Ligoxygakis P , Pelte N, Hoffmann JA, Reichhart JM. Activation of Drosophila 370 Toll during fungal infection by a blood serine protease. Science 297, 114-116 371 (2002). 372 373 11. Coleman JW. Nitric oxide in immunity and inflammation. Int 374 Immunopharmacol 1, 1397-1406 (2001). 375 376 12. Lundberg JO, Weitzberg E. Nitric oxide signaling in health and disease. Cell 377 185, 2853-2878 (2022). 378 379 13. Hess DT, Matsumoto A, Kim SO, Marshall HE, Stamler JS. Protein S-380 nitrosylation: purview and parameters. Nat Rev Mol Cell Biol 6, 150-166 381 (2005). 382 383 14. Feechan A, Kwon E, Yun BW , Wang Y , Pallas JA, Loake GJ. A central role for S-384 nitrosothiols in plant disease resistance. Proc Natl Acad Sci U S A 102, 8054-385 8059 (2005). 386 387 15. Liu L, Hausladen A, Zeng M, Que L, Heitman J, Stamler JS. A metabolic enzyme 388 for S-nitrosothiol conserved from bacteria to humans. Nature 410, 490-494 389 (2001). 390 391 16. Foley E, O'Farrell PH. Nitric oxide contributes to induction of innate immune 392 responses to gram-negative bacteria in Drosophila. Genes Dev 17, 115-125 393 (2003). 394 395 17. Yun BW, et al. S-nitrosylation of NADPH oxidase regulates cell death in plant 396 immunity. Nature 478, 264-268 (2011). 397 398 18. Hou Q, et al. Nitric oxide metabolism controlled by formaldehyde 399 dehydrogenase (fdh, homolog of mammalian GSNOR) plays a crucial role in 400 visual pattern memory in Drosophila. Nitric Oxide 24, 17-24 (2011). 401 402 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint 19. Dijkers PF, O'Farrell PH. Dissection of a hypoxia-induced, nitric oxide-mediated 403 signaling cascade. Mol Biol Cell 20, 4083-4090 (2009). 404 405 20. Parks AL, et al. Systematic generation of high-resolution deletion coverage of 406 the Drosophila melanogaster genome. Nat Genet 36, 288-292 (2004). 407 408 21. Brand AH, Perrimon N. Targeted gene expression as a means of altering cell 409 fates and generating dominant phenotypes. Development 118, 401-415 410 (1993). 411 412 22. Gratz SJ, Rubinstein CD, Harrison MM, Wildonger J, O'Connor-Giles KM. 413 CRISPR-Cas9 Genome Editing in Drosophila. Curr Protoc Mol Biol 111, 414 31.32.31-31.32.20 (2015). 415 416 23. Pavlopoulos A, Averof M. Establishing genetic transformation for comparative 417 developmental studies in the crustacean Parhyale hawaiensis. Proc Natl Acad 418 Sci U S A 102, 7888-7893 (2005). 419 420 24. Taylor K, Kimbrell DA. Host immune response and differential survival of the 421 sexes in Drosophila. Fly (Austin) 1, 197-204 (2007). 422 423 25. Lemaitre B, Reichhart JM, Hoffmann JA. Drosophila host defense: differential 424 induction of antimicrobial peptide genes after infection by various classes of 425 microorganisms. Proc Natl Acad Sci U S A 94, 14614-14619 (1997). 426 427 26. Levashina EA, Ohresser S, Bulet P , Reichhart JM, Hetru C, Hoffmann JA. 428 Metchnikowin, a novel immune-inducible proline-rich peptide from 429 Drosophila with antibacterial and antifungal properties. Eur J Biochem 233, 430 694-700 (1995). 431 432 27. Buchon N, et al. A single modular serine protease integrates signals from 433 pattern-recognition receptors upstream of the Drosophila Toll pathway. Proc 434 Natl Acad Sci U S A 106, 12442-12447 (2009). 435 436 28. Pili-Floury S, et al. In vivo RNA interference analysis reveals an unexpected 437 role for GNBP1 in the defense against Gram-positive bacterial infection in 438 Drosophila adults. J Biol Chem 279, 12848-12853 (2004). 439 440 29. Gobert V, et al. Dual activation of the Drosophila toll pathway by two pattern 441 recognition receptors. Science 302, 2126-2130 (2003). 442 443 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint 30. Dudzic JP , Hanson MA, Iatsenko I, Kondo S, Lemaitre B. More Than Black or 444 White: Melanization and Toll Share Regulatory Serine Proteases in Drosophila. 445 Cell Rep 27, 1050-1061.e1053 (2019). 446 447 31. Green C, Levashina E, McKimmie C, Dafforn T, Reichhart JM, Gubb D. The 448 necrotic gene in Drosophila corresponds to one of a cluster of three serpin 449 transcripts mapping at 43A1.2. Genetics 156, 1117-1127 (2000). 450 451 32. Levashina EA, et al. Constitutive activation of toll-mediated antifungal defense 452 in serpin-deficient Drosophila. Science 285, 1917-1919 (1999). 453 454 33. Jaffrey SR, Snyder SH. The biotin switch method for the detection of S-455 nitrosylated proteins. Sci STKE 2001, pl1 (2001). 456 457 34. Forrester MT, Foster MW , Benhar M, Stamler JS. Detection of protein S-458 nitrosylation with the biotin-switch technique. Free Radic Biol Med 46, 119-459 126 (2009). 460 461 35. Dong Y , Morton JC, Jr., Ramirez JL, Souza-Neto JA, Dimopoulos G. The 462 entomopathogenic fungus Beauveria bassiana activate toll and JAK-STAT 463 pathway-controlled effector genes and anti-dengue activity in Aedes aegypti. 464 Insect Biochem Mol Biol 42, 126-132 (2012). 465 466 36. Lee JY , Woo RM, Choi CJ, Shin TY , Gwak WS, Woo SD. Beauveria bassiana for 467 the simultaneous control of Aedes albopictus and Culex pipiens mosquito 468 adults shows high conidia persistence and productivity. AMB Express 9, 206 469 (2019). 470 471 37. Al Khoury C, Nemer N, Bernigaud C, Fischer K, Guillot J. First evidence of the 472 activity of an entomopathogenic fungus against the eggs of Sarcoptes scabiei. 473 Vet Parasitol 298, 109553 (2021). 474 475 38. Darbro JM, Graham RI, Kay BH, Ryan PA, Thomas MB. Evaluation of 476 entomopathogenic fungi as potential biological control agents of the dengue 477 mosquito, Aedes aegypti (Diptera: Culicidae). Biocontrol Science and 478 Technology 21, 1027-1047 (2011). 479 480 39. Wang Y , Cui C, Wang G, Li Y , Wang S. Insects defend against fungal infection by 481 employing microRNAs to silence virulence-related genes. Proc Natl Acad Sci U 482 S A 118, (2021). 483 484 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint 40. Ndii A, Rahardjo B, Himawan T. The Combination of Entomopathogenic 485 Fungus of Beauveria bassiana (Balls) Vuill. with the Insect Growth Regulator 486 (IGR) of Lufenuron Against Reproductive of Bactrocera carambolae Fruit Flies 487 (Diptera: Tephritidae). The Journal of Experimental Life Sciences 6, 25-28 488 (2016). 489 41. Loof TG, et al. Coagulation, an ancestral serine protease cascade, exerts a 490 novel function in early immune defense. Blood 118, 2589-2598 (2011). 491 492 42. Krem MM, Di Cera E. Evolution of enzyme cascades from embryonic 493 development to blood coagulation. Trends Biochem Sci 27, 67-74 (2002). 494 495

496 Drosophila culture 497 Drosophila lines were maintained at 22˚ C on yeast cornmeal agar medium (YCMA) (1L 498 H2O, glucose 78.5g, maize meal 71.5g, Yeast 50g, agar 10.7g, Nipagen 2.7g, propionic 499 acid 3.25ml). The genotype and origin of the strains of D. melanogaster used in the 500 experiments described here are listed in Table S1. Where the flies used in an 501 experiment are the progeny of a cross between two strains the figure indicates the 502 genotype of the female parent first. The mutations that were not generated during 503 this work were obtained from the sources indicated in Table S1. 504 505 Gsnor enzyme assay 506 The activity of Gsnor was assayed spectrophotometrically by measuring the rate of 507 NADH oxidation in the presence of S-nitrosoglutathione (GSNO). Proteins were 508 extracted from flies in HE buffer (25 mM HEPES, 1 mM EDTA pH7.7) supplemented 509 with the protease inhibitors [ 50μg/ml N-tosyl-L-phenylalaninyl-chloromethylketone 510 (TPCK), 50μg/ml N-alpha-tosyl-L-lysinyl-chloromethylketon (TLCK) and 0.5 mM 511 phenylmethanesulfonyl fluoride (PMSF)] and quantified using the Bradford assay 512 (Bradford, 1976). For Gsnor activity assays 75μg of protein was incubated in 1ml of 513 HE buffer with the addition of 350μM NADH and 350μM GSNO and NADH oxidation 514 was measured by following the absorbance at 340nm. Zero time readings were 515 taken immediately after the addition of GSNO. 516 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint 517 Measurement of nitrite in flies 518 Female flies were snap-frozen in liquid nitrogen and homogenised in 20μL of 519 extraction buffer (50mM Tris-HCl pH 7.5, 150mM NaCl, 5mM EDTA, 0.1% Triton X-520 100) per fly using a disposable plastic grinder. Samples were centrifuged at 521 13,000rpm at 4° C for 15 minutes and the supernatant was transferred to a fresh pre-522 chilled tube. Fifty microlitres of the resulting supernatant were dispensed, in 523 triplicate, into a flat-bottom 96-well enzymatic assay plate, after which 50 μL of 524 Sulfanilamide Solution from the Griess Reagent System (Promega, G2930) were 525 added. Plates were incubated at room temperature for 10 min, protected from light, 526 before the sequential addition of 50μL N-1-naphthylethylenediamine 527 dihydrochloride Solution. After a further 10 min incubation in the dark, the azo 528 chromophore was quantified at 540 nm using a microplate reader within 30 min of 529 colour development. Sodium nitrite standards (0–100 μM, prepared in homogenate 530 buffer) were processed in parallel, and sample nitrite concentrations were 531 interpolated from the standard curve generated on the same plate. 532 533 Sodium nitroprusside (SNP) sensitivity assay 534 Female flies aged between 3 and 6 days were transferred to empty vials and starved 535 for two hours at 25° C and then transferred to vials containing a cotton roll soaked in 536 5% sucrose solution supplemented with 5 mM SNP. The vials were kept at 25° C and 537 the number of survivors was recorded one day after the treatment. 538 539 Septic infection with bacteria 540 Staphylococcus aureus NCTC 8325 (https://www.culturecollections.org.uk) was used 541 to test sensitivity to Gram positive bacteria and Escherichia coli strain MG16551 was 542 used to test sensitivity of Gram negative bacteria. Bacteria were grown on LB (10g/L 543 Foremedium Tryptone, 5g/L Foremedium Yeast extract, 10g/L NaCl) agar (15g/L) 544 plates and for each experiment a single colony was transferred to 5ml LB and grown 545 overnight at 37° C shaking at 200rpm and then the culture was centrifuged. Flies 546 were anaesthetised with CO2 or by chilling on ice and their abdomen was pierced 547 with a tungsten needle that had been dipped in the bacterial pellet. After infection 548 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint the flies were transferred to vials containing YCMA and kept at 29˚C. The proportion 549 of flies surviving was calculated at the time points shown in the relevant figures with 550 any flies that died within two hours excluded from the total number at the start of 551 the experiment. 552 553 Assaying sensitivity to infection with B. bassiana 554 Beauveria bassiana was grown at 29˚C on potato dextrose agar (39g PDA (Sigma 555 P2182) in 1L of water) supplemented with 50μg/ml chloramphenicol. Once the 556 mycelium had covered the plate spore formation was induced by protecting the 557 plates from light and leaving them in a fume hood until completely desiccated. The 558 dried plates were then stored in the dark at 4˚C. For infection assays twenty 3-7 559 days old flies were anaesthetized by chilling on ice and transferred to a 2ml 560 microcentrifuge tube containing B. bassiana with mycelium and spores scraped from 561 an area of about 1cm2 of a dried plate. The tubes were then shaken gently by hand 562 for 2 minutes after which the flies were transferred to vials containing YCMA, 563 maintained at 29˚C and the number of surviving flies was recorded over time. Flies 564 that died within two hours of infection were excluded when calculating survival 565 rates. 566 567 Genomic DNA extraction from Drosophila 568 Thirty flies were frozen at -70° C and then homogenised with a disposable plastic 569 grinder in 400μL of 100mM Tris-HCl pH7.5, 100mM EDTA, 100mM NaCl, 0.5% SDS 570 (Sodium Dodecyl Sulphate). The homogenate was incubated at 65˚C for 30 minutes 571 and then 800μL of 1.67M C2H3O2K, 4M LiCl were added to the tubes, followed by 572 incubation on ice for 10 minutes and a subsequent centrifugation at 14,000rpm for 573 15 minutes. Equal volumes of the supernatant were transferred to two clean tubes, 574 and DNA was precipitated by adding 700μL isopropanol followed by centrifugation at 575 14,000rpm for 15 minutes. The pellet was then dissolved in 100μL sterile distilled 576 water. 577 578 Detection of fdh sequences by PCR 579 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint Genomic DNA of flies heterozygous for the overlapping deletions Df7305 and Df7306 580 was amplified by PCR to confirm that they do not contain an intact fdh. The reaction 581 mix contained 1 x Ex Taq Buffer (TakaraBio), 0.2mM dNTPs, 0.2µM of each of primer 582 1 (5‘-AATAAACCATACTGCAAAGATGTCTGCTAC-3´ and primer 2 (5‘-583 TGCAGCTGAGACGG-3´ ) (Fig. 1A), 5%v/v genomic DNA, and 0.05unit/µL TaKaRa Ex 584 Taq DNA polymerase. The DNA was denatured at 94˚C for 1 minute followed by 5 585 cycles at 94˚C for 0.5 minute, 60˚C for 0.5 minute and 72˚C for 2.5 minutes, and then 586 25 cycles of 94˚C for 0.5 minute and 68˚C for 2.5 minutes. Amplified DNA was 587 analysed by agarose gel electrophoresis (1% agarose in 1 x Tris-acetate-EDTA buffer 588 [40 mM Tris, 20 mM Acetic acid and 1 mM EDTA]). Gels were stained with 0.5μg/ml 589 ethidium bromide and the DNA visualized using an ultraviolet imaging system 590 591 Total RNA extraction 592 Five adult female flies were frozen in liquid nitrogen and then homogenised in 500μL 593 of TRizol (Invitrogen) using a disposable plastic grinder. The homogenate was left at 594 room temperature for 5 minutes followed by centrifugation at 12,000rpm for 10 595 minutes at 4˚C. 180μL of the supernatant were transferred to a new microcentrifuge 596 tube and 60μL of chloroform was added. The homogenate and chloroform were 597 mixed vigorously by hand followed by incubation at room temperature for 3 minutes 598 and centrifugation at 10,000rpm for 15 minutes at 4˚C. About 80μL of the upper 599 phase was transferred to a new microcentrifuge tube and 100μL of isopropanol was 600 added followed by incubation at room temperature for 5 minutes and centrifugation 601 at 12,000rpm for 10 minutes at 4˚C. The supernatant was replaced by 600μL of 75% 602 ethanol followed by centrifugation at 2,200g for 5 minutes at 4˚C. After removing the 603 supernatant, the RNA pellet was dried in a laminar flow cabinet and then dissolved in 604 55μL of diethylpyrocarbonate (DEPC) treated water. 605 606 RNA extraction and RT-PCR 607 Twenty five 3-4 days old flies were frozen in liquid nitrogen and subsequently ground 608 in 500µL of TRizol (Invitrogen) using a bead mill (Qiagen). The homogenate was left 609 at room temperature for 5 minutes before centrifugation at 5,600g for 10 minutes at 610 4˚C. The supernatant was transferred to a new microcentrifuge tube and 100µL of 611 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint chloroform was added. The contents were mixed vigorously by hand and incubated 612 at room temperature for 3 minutes before centrifugation at 10,000g for 15 minutes 613 at 4˚C. The upper phase was transferred to a new microcentrifuge tube and 250µL of 614 isopropanol was added followed by incubation at room temperature for 10 minutes 615 and centrifugation at 12,000g for 10 minutes at 4˚C. The supernatant was replaced 616 with 0.5mL of 75 % ethanol followed by centrifugation at 7,500xg for 5 minutes at 617 4˚C. The supernatant was removed, and the RNA pellet was dried in a laminar flow 618 cabinet and then dissolved in 50µL of diethylpyrocarbonate (DEPC) treated water. 619 620 Quantitative RT-PCR 621 Expression of the anti-microbial peptide encoding genes Drosomycin (Drs) and 622 Metchnikowin (Mtk) was quantified by reverse transcript polymerase chain reaction 623 (RT PCR) using SYBR Green I Master Mix (Thermo Fisher Scientific) in a LightCycler 624 480 system (Roche). Relative gene expression levels were quantified using the 2-∆∆Ct 625 method2 the housekeeping gene Rp49 (Ribosomal protein 49) serving as an internal 626 control for normalisation. The primers and conditions used to quantify the 627 expression of these genes have been described previously3,4 The primers used were 628 5´-AGATCGTGAAGAAGCGCACCAAG-3´ and 5´-CACCAGGAACTTCTTGAATCCGG-3´ for 629 Rp49, 5‘-CGTGAGAACCTTTTCCAATTATGATG-3´ and 5´-TGGTGGAGTTGGGCTTCATG-630 3´ for Drs, and 5-GATGCAACTTAATCTTGGAGCG-3´ and 5´-631 TTAATAAATTGGACCCGGTCTTGGTTGG-3´ for Mtk. 632 633 Total protein extraction and quantification 634 Flies were homogenised in 20μL of extraction buffer (50mM Tris-HCl pH 7.5, 150mM 635 NaCl, 5mM EDTA, 0.1% Triton X-100) per fly freshly supplemented with protease 636 inhibitors (50μg/ml TPCK 50 μg/ml TLCK and 0.5mM PMSF). Samples were 637 centrifuged at 13,000rpm at 4° C for 15 minutes and the supernatant was transferred 638 to a fresh pre-chilled tube. Protein concentrations were measured using the 639 Bradford Assay5. 640 641 SDS-Poly-Acrylamide Gel Electrophoresis (SDS-PAGE) 642 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint Protein samples were mixed with a 4x stock of sample buffer to a final concentration 643 of 50mM Tris-HCl pH 6.8, 2% SDS, 0.02% bromophenol blue and 10% glycerol either 644 with or without 50mM dithiothreitol. Samples were heated at 85°C for 10 min 645 before separation by polyacrylamide gel electrophoresis. Gels were washed in H2O 646 before being incubated in staining solution (0.25% Coomassie Brilliant Blue R, 40% 647 methanol, 7% acetic acid) for between 30 min and one hour. Gels were de-stained 648 overnight in de-staining solution (40% methanol, 10% acetic acid) and then 649 photographed 650 651 Western Blots 652 Proteins were transferred to nitrocellulose membranes using a Bio-Rad Trans Blot® 653 system either overnight at a constant voltage of 20V or for 2-3 hours at 90V. Proteins 654 were visualised on the membranes with Ponceau S (0.1% Ponceau S, 5% acetic acid) 655 for 1 min and then rinsed with H2O before being photographed. Membranes were 656 de-stained with PBS-T (137mM NaCl, 2.7mM KCl, 10mM Na2HPO4, 1.8mM KH2PO4, 657 0.1% Tween-20) and blocked for 1 hour at room temperature with 5% dried skimmed 658 milk in PBS-T before incubation with primary antibodies either overnight at 4˚ C or at 659 room temperature for 1-2 hours. After washing to remove excess primary antibody 660 the membrane was incubated for 1 hour at room temperature with the appropriate 661 secondary antibody coupled to horseradish peroxidase (HRP). SuperSignal West 662 Pico/Dura Chemiluminescent Substrate (Thermo Scientific) was added to the 663 membranes and labelled proteins were detected with X-ray film. All antibodies were 664 diluted in 5% dried skimmed milk in PBS-T. The primary antibodies were mouse anti-665 HA (Roche, clone 12CA5) and HRP conjugated goat anti-Biotin (Cell Signalling). The 666 secondary antibody for detection of HA antigen was HRP linked goat anti-mouse IgG 667 HRP (Cell Signalling #7075). 668 669 Expression of Gsnor from fdh transgenes and expressing Psh-HA 670 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint Gsnor was expressed in flies using the UAS/GAL4 system6 to drive expression of an 671 fdh transgene linked to the GAL4 UAS binding site. GAL4 protein was expressed from 672 either the ubi (ubiquitin) or act5C (Actin 5C) promoter. 673 674 The coding sequence of psh was amplified using Phusion high-fidelity polymerase 675 (New England Biolabs, UK) from freshly synthetized cDNA of wild type Oregon- R 676 flies. The primers (Forward: 5´-CACCATGCCATTGAAGTGGTC-3´ and Reverse: 5´-677 TTACTTCACCCGATTGTCCGG-3´) were designed to add the nucleotides CACC at the 678 5´end of the coding strand to allow TOPO® cloning (Life Technologies). The PCR 679 products were gel-purified and cloned into the pENTRTM/D-TOPO® vector according 680 to the manufacturers’ instructions, transfected into E.coli and plated on LB agar 681 containing 50µg/ml kanamycin. 682 683 Plasmid DNA was isolated from single colonies and sequenced. Inserts from positive 684 constructs were transferred by Gateway® cloning (LR reactions following 685 manufacturer’s instructions - Life Technologies) into pUASt-HA to generate pUASt-686 Psh-HA7 for expression of C-terminal HA-tagged protein in D. melanogaster. 687 Recombinant clones were selected on LB agar containing 50µg/ml ampicillin and 688 were confirmed by sequencing. pUASt-Psh-HA constructs were purified using the 689 QIAfilter plasmid midi kit (Qiagen), in accordance with the manufacturer’s 690 instructions. DNA quality and concentrations were measured using a NanoDrop 691 spectrophotometer ND 1000, and 50μg of each construct was sent to Genetic 692 Services Inc for integration at the attP40 φC31 integration site on the 2nd 693 chromosome8. 694 695 Detection of S-nitrosylated PSH-HA 696 The Biotin Switch Technique (BST)9,10 was used to detect S-nitrosylation of Psh-HA. 697 Flies expressing Psh-HA were homogenised in extraction buffer (100mM HEPES 698 pH7.8, 1mM EDTA, 0.1mM Neocuproine, 0.5% Triton X-100, 50μg/ml TPCK, 50μg/ml 699 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint TLCK and 0.5 mM PMSF. Samples were then centrifuged at 13,000rpm at 4°C for 15 700 minutes. Free thiols on proteins in the supernatant were blocked in an equal volume 701 of blocking buffer (250mM HEPES pH7.8, 1mM EDTA, 0.1 mM Neocuproine, 5% (w/v) 702 SDS, 50mM N-Ethylmaleimide) for 30 minutes at 55oC. The blocking buffer was 703 removed by precipitating proteins with two volumes of cold acetone. After 20 704 minutes at -20oC, samples were centrifuged at 15,000rpm for five minutes at 4oC. 705 The pellet was washed three times with 70% acetone and resuspended in 85μL of 706 250mM HEPES pH7.8, 1 mM EDTA, 0.1 mM Neocuproine, 1% (w/v) SDS). S-707 nitrosothiols were reduced by adding sodium ascorbate to a final concentration of 708 25mM and the newly free thiols were biotinylated by adding biotin-N-[6-709 (biotinamido)hexyl]-3′-(2′-pyridyldithio)-propionamide) to a final concentration of 710 0.4mM placed on a rocker plate for one hour. Ascorbate was omitted from samples 711 used as negative controls. Proteins were then precipitated and washed with acetone 712 as described above. 713 714 For detection of biotinylated Psh-HA the pellet was dissolved in 300μL of 25mM 715 HEPES, pH7.8, 1mM EDTA, 0.1mM Neocuproine, 1% (w/v) SDS to which 20µL of 716 streptavidin beads in 100µL of 25 mM HEPES, 1mM EDTA, 0.1mM Neocuproine, 717 100mM NaCl, 0.5% Triton X-100, were added and incubated at 4°C on a rocking plate 718 overnight. Next morning samples were washed five times with 500µL of 25mM 719 HEPES, 1 mM EDTA, 0.1 mM Neocuproine, 600mM NaCl, 0.5% Triton X-100 and 720 resuspended in 20µL of elution buffer (25mM HEPES, 1mM EDTA, 0.1mM 721 Neocuproine, 1% v/v β-mercaptoethanol). After 30 minutes at room temperature the 722 beads were pelleted by centrifugation at maximum speed for one minute at room 723 temperature and 30µl of the supernatant was loaded on a SDS-PAGE and biotinylated 724 Psh-HA was detected in a Western blot using anti-HA antibody. 725 726 CRISPR/Cas9 deletion of fdh 727 The fdh gene was replaced with DNA coding for the visible marker DsRed using 728 CRISPR/Cas9 to cleave DNA on either side of fdh and a plasmid with DsRed flanked 729 by sequences from either side of fdh as the template to allow homology-directed 730 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint repair (Fig S1A). The “CRISPR Optimal Target Finder” web tool 731 (https://flycrispr.org/target-finder/)11 was used to find targets for CRISPR/Cas9 732 cleavage on either side of fdh. The gRNA target sequences identified were 733 Chromosome 3R nucleotides 10870883 to 10870905 to the left of fdh and 734 nucleotides 10872425 to 10872446 to the right of fdh. Two homology arms adjacent 735 to the gRNA target sequences and flanking fdh were amplified by PCR using primers 736 with restriction sites allowing the products to be inserted on either side of the 737 3xDsRed coding DNA in the plasmid pHD-DsRed-attP-w+ (Addgene #80898 donated 738 by K. O’Connor-Giles) to give the plasmid pHD-DsRed-attP-fdh. The primers, with 739 restriction sites underlined, for amplifying the sequence to the left of fdh were 5´-740 CACTGCAGCGTATCTCTACGGATATCC-3´ and 5´-741 CAAGATCTTTGGGGGTCGGATTACTGTC-3´ , and for amplifying the sequence on the 742 right of fdh were 5´-ATACATATGAAGCTCACCGGGACTCAG-3´ and 5´-743 GATAGAATTCACACTGACGATGTGATCCACATAG-3´ . 744 745 A plasmid expressing gRNAs to direct cleavage to the left and right of fdh was 746 constructed from pCFD4-U6:1_U6:3tandemgRNAs (Addgene #49411 donated by S. 747 Bullock) as described by Port et al12. Primers containing gRNA sequences for cleavage 748 on either side of fdh were used to amplify a fragment that was then cloned into BbsI 749 cut pCFD4 by homology directed cloning (Gibson Assembly® Cloning Kit – New 750 England Biolabs. Primers for this PCR amplification, with the gRNA sequence 751 underlined, were 5´-752 TATATAGGAAAGATATCCGGGTGAACTTCGACATAAGAGTATCTTCATTGGTTTTAGAGCTA753 GAAATAGCAAG-3´ , and 5´-754 ATTTTAACTTGCTATTTCTAGCTCTAAAACTAAAAGCTCACCGGGACTCCGACGTTAAATTGA755 AAATAGGTC-3´ ). The resulting plasmid pCFD4-gRNAs expresses one gRNA from the 756 U6:1 promoter and the other from the U6:3 promoter. 757 758 Both constructs pCFD4-gRNAs and pHD-DsRed-attP-fdh were purified and sent to the 759 Genetic services Inc, where they were co-injected into embryos of D. melanogaster 760 strain BDSC #51323 (y[1],{vas-Cas9}ZH-2A w1118/FM7) that expresses Cas9 protein 761 under the control of the vasa promoter. Transgenic flies expressing DsRed were 762 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint crossed with a balancer strain to construct stocks homozygous for the fdh deletion 763 and to remove the chromosomes carrying vas-Cas9. The replacement of fdh by 764 DsRed was confirmed by PCR and sequencing of genomic DNA using primers fwd - 5´-765 CTTGGAGCCGTACTGGAACTG 3´ and rev - 5‘-GCCTCCGATTTGGTTTGTTG-3‘as shown 766 in Fig S1B. 767 768 Figure legends 769 Figure 1. (A) Map of fdh. Deficiencies Df7305 and Df7306 overlap in region 86C7 of 770 Chromosome 3 with breakpoints within the gene fdh that is transcribed from right to 771 left in this diagram. The coordinates refer to bases in the D. melanogaster genomic 772 sequence and are taken from FlyBase1. The 5’ and 3’ untranslated regions of fdh are 773 shown in grey and the coding regions in orange with introns indicated by a thin line. 774 The diagram is not drawn to scale. (B) Flies heterozygous for Df7305 and Df7306 775 lack an intact copy of fdh. DNA was purified from flies with the genotypes indicated 776 and was amplified by PCR using primers 1 and 2 shown in (A). (C) Flies heterozygous 777 for Df7305 and Df7306 have greatly reduced Gsnor activity. Gsnor activity in protein 778 extracts from flies of the genotypes indicated was assayed as described in the 779

and Methods. GSNO was added at time 0 and the absorbance at 340nm 780 was measured at ten minutes intervals thereafter. (D) Infection by B. bassiana 781 increases nitrite in flies heterozygous for Df7306 and Df7305. The level of nitrite in 782 extracts of wild-type (Oregon R) or heterozygous Df7305/Df7306 female flies is 783 shown either with or without infection with B. bassiana. The bars show the mean ± 784 S.E of the results of three independent experiments and the “p” value for the 785 differences in nitrite levels between the two genotypes was calculated by one-way 786 ANOVA and Tukey HSD tests, * indicates p ≤ 0.05 while ** indicates p ≤ 0.01. (E) 787 Flies with reduced Gsnor show increased sensitivity to SNP . The SNP sensitivity of 788 groups of 15 female flies aged for 4 to 7 days was tested as described in the Materials 789 and Methods. The number of survivors recorded one day after exposure to SNP is 790 shown for flies of the genotypes indicated. The data show the mean ±SE from three 791 vials. The “p” value for the differences between the different genotypes was 792 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint calculated using a one-way ANOVA (F3,8 = 56.68 p< 0.001) and a Tukey HSD test, * 793 indicates p ≤ 0.05 while ** indicates p ≤ 0.01. 794 795 Figure 2. (A) Susceptibility of flies to B. bassiana. Groups of 15 male flies aged for 3 796 to 6 days, were infected with B. bassiana as described in the Materials and Methods. 797 The number of flies surviving after infection was recorded every 24 hours and the 798 survivors were transferred to a fresh vial every other day. Flies trans-heterozygous 799 for the fdh deficiencies Df7305 and Df7306 were more sensitive to infection by B. 800 bassiana than wild type (Oregon R). The graphs show the mean ± S.E of three 801 independent experiments. (B) Susceptibility of flies to infection with S. aureus. 802 Groups of 30 female flies aged for 3 to 6 days were infected with a needle coated in 803 S. aureus as described in the Materials and Methods, transferred to vials containing 804 YCMA and kept at 29°C and the number of survivors was recorded every 24 hours for 805 five days. The graphs show the mean ± S.E of three independent experiments. (C) 806 Susceptibility of flies to infection with E. coli. About 30 female flies of the genotypes 807 indicated were aged for 3 to 4 days then infected with E. coli MG1655 as derscribed 808 in the Material and Methods. Flies homozygous for mutant alleles of imd (immune 809 deficiency) or rel (relish) were used as controls to indicate the response of flies with 810 defects in the IMD pathway. (D) Activation of Drs and (E) Mtk is reduced in fdh 811 mutant flies. RNA was extracted from flies of the genotypes shown either without 812 infection or 24 hours after infection with B. bassiana. The levels of Drs and Mtk 813 relative to Rp49 RNA were measured by Quantitative RT PCR as described in the 814

and Methods. These are shown as the mean ± S.E of the ratio of Drs and 815 Mtk to Rp49 RNAs from three experiments. Differences in gene expression between 816 mutant fly lines were analysed using ANOVA followed by Fisher's LSD (Least 817 Significant Difference) test. Different letters indicate statistically significant 818 differences between groups (p < 0.05). 819 820 Figure 3. (A) Survival of necrotic mutant flies is suppressed by reduction in Gsnor. 821 Total survivors were counted one day post emergence. The percentage of survivors 822 was calculated by dividing the number of survivors by the total number of flies. The 823 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint data show the mean ±SE from three repeated experiments with 24-40 flies per 824 group. (B) Melanisation of necrotic mutant flies is suppressed by reduction in 825 Gsnor. Melanised files were counted one day post emergence. The percentage of 826 melanised flies was calculated by dividing number of melanised flies by the total 827 number of flies. The data show the mean ±SE from three repeated experiments with 828 24-40 flies per group. (C) Expression of HA tagged PSH in D. melanogaster. Total 829 protein was extracted from wild type (Oregon R) and flies expressing Psh-HA using 830 the UAS/GAL4 system with GAL4 expressed from the Ubiquitin promoter. Proteins 831 were separated by SDS PAGE, transferred to a nitrocellulose membrane and HA 832 tagged protein was detected with anti-HA antibody as described in the Materials and 833 Methods. (D) Persephone is S-nitrosylated in vivo. Total protein extracted from 10 834 female flies expressing Psh-HA was submitted to the BST followed by streptavidin 835 pulldown and Western blotting with anti-HA antibody. 25mM sodium ascorbate was 836 used in the BST to reduce S-nitrosothiols and was omitted from the reaction to 837 provide a negative control. Sample aliquots were taken before the pulldown step 838 and used in parallel western blots to confirm that similar amounts of total protein 839 had been loaded in each lane. (E) Cleavage of Psh is inhibited if Gsnor is reduced. 840 Groups of 20 female flies, were infected by B. bassiana as described in the Material 841 and Methods. Total protein was extracted from fdh-/- flies expressing Psh-HA and 842 fdh+/+ flies expression Psh-HA 3 days post B. bassiana inoculation. Proteins were 843 separated by SDS PAGE, transferred to a nitrocellulose membrane and HA tagged 844 protein was detected with anti-HA antibody as described in the Materials and 845 Methods. Cleavage of Psh-HA (red arrow) was observed in fdh+/+, Psh-HA flies 846 infected by B. bassiana, while reduced cleavage was observed in fdh-/-, Psh-HA flies 847 infected by B. bassiana. No cleavage was observed in non-infected flies. Blots are 848 cropped for clarity. Some lanes were omitted compared with the original blot; no 849 other image processing was performed. 850 851 Figure S1. (A) Strategy for replacing fdh coding sequence with DsRed. The method 852 used is described in Materials and Methods. The chromosomal region containing fdh 853 gene and flanking genes is diagrammed in the top line. gRNAs expressed from 854 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint plasmid pCFD4-gRNA direct Cas9 nuclease to cut intergenic DNA on either side of 855 fdh. The resulting gap is repaired by the homology directed repair pathway using the 856 pHD-DsRed-attP plasmid as the homology template thereby replacing fdh with 3xP3-857 DsRed. This diagram is not drawn to scale. (B) Confirmation that DsRed is 858 integrated in place of fdh. PCR using genomic DNA from flies homozygous for 859 mutations fdhcp1, fdhcp2, fdhcp3 and Oregon R flies as template and the primers P5’-F 860 and P5’-R (Fig. S1A) confirmed the integration of DsRed in the mutant alleles. (C) 861 Flies homozygous for CRISPR replacement of fdh have reduced Gsnor activity. 862 Protein extracts from flies homozygous for fdhcp1, fdhcp2 or fdhcp3 were assayed for 863 Gsnor activity as described in the Materials and Methods. Protein from flies 864 heterozygous for Df7305 and Df7306 was included for comparison, with protein from 865 Oregon R flies used as a positive control and a reaction without protein as negative 866 control. (D) Flies homozygous for CRISPR generated mutations of fdh show 867 increased sensitivity to SNP . The SNP sensitivity of groups of 20 female flies aged for 868 3 to 6 days was tested as described in the Materials and Methods. The data 869 represent the mean ±SE of three independent experiments and the results were 870 analysed by one-way ANOVA and Tukey HSD tests, * indicates p<0.05 and ** 871 indicates p<0.01. (E). Flies homozygous for CRISPR generated mutations of fdh 872 show increased susceptibility to B. bassiana. Groups of 15-20 flies (3-7 days old) of 873 the indicated genotypes were infected with B. bassiana as described in Materials and 874 Methods. Each experiment used three vials of 15-20 flies each (45-60 flies total per 875 genotype per experiment). Survival was monitored daily for 14 days post-infection. 876 The survival curves show pooled data from three independent experiments (total n = 877 135-180 flies per genotype). 878 879 Figure S2. (A) Female flies are more susceptible than males to infection with B. 880 bassiana. Groups of 15 male or female wild type (Oregon R) flies, aged for 3 to 6 881 days, were tested for their sensitivity to B. bassiana infection as described in the 882

and Methods. The figure shows the mean ±SE of three repetitions. (B) 883 Flies with reduced Gsnor are not affected by exposure to heat killed S. aureus. 884 Groups of 15 female flies, aged for 3 to 6 days were pierced with a needle coated in 885 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint heat-killed S. aureus, placed in vials of fresh YCMA, incubated at 29˚ C and the number 886 of survivors was recorded every 24 hours for five days. The figure shows the mean 887 ±SE of three independent experiments. (C) Flies with reduced Gsnor are not 888 affected by exposure to heat killed B. bassiana. Flies were exposed to heat killed 889 spores of B. bassiana as described for live spores in the Materials and Methods and 890 the percentage of surviving flies recorded over time. The figure shows the mean ±SE 891 of three independent experiments. (D) Melanisation of flies due to nec mutations is 892 suppressed in flies with reduced Gsnor activity. Each panel shows a 3 to 5 days old 893 female fly of the genotypes indicated. 894 Table S1 895 Mutation Source rel[E20] Gift of B. Lemaitre spz[rm7] Gift of B. Lemaitre imd Gift of B Lemaitre Df(3)Exel7305 BDSC Stock #7956 Df(3)Exel7306 BDSC Stock #7957 nec[2] BDSC Stock #9151 nec[10] BDSC Stock #4288 Act5C-gal4 BDSC Stock #3954 Ubi-Gal4 BDSC Stock #32551 896 Table S1: The sources of mutant alleles used in these experiments. 897 898 899 900 901 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint 902 Figure 1 903 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint 904 Figure 2 905 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint 906 Figure 3907 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint 908 Figure S1 909 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint 910 Figure S2 911 .CC-BY-NC-ND 4.0 International licenseperpetuity. It is made available under a preprint (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in The copyright holder for thisthis version posted January 26, 2026. ; https://doi.org/10.64898/2026.01.23.701332doi: bioRxiv preprint