TNF Superfamily Member 14 Drives Post-Influenza Depletion of Alveolar Macrophages Enabling Secondary Pneumococcal Pneumonia

TNF Superfamily Member 14 Drives Post-Influenza Depletion of Alveolar Macrophages Enabling Secondary Pneumococcal Pneumonia Christina Malainou, … , Ulrich Matt, Susanne Herold J Clin Invest. 2025. https://doi.org/10.1172/JCI185390. In-Press Preview Secondary bacterial infection, often caused by Streptococcus pneumoniae (Spn), is one of the most frequent and severe complications of influenza A virus (IAV)-induced pneumonia. Phenotyping of the pulmonary immune cell landscape after IAV infection revealed a substantial depletion of the tissue-resident alveolar macrophage (TR-AM) population at day 7, which was associated with increased susceptibility to Spn outgrowth. To elucidate the molecular mechanisms underlying TR-AM depletion, and to define putative targets for treatment, we combined single-cell transcriptomics and cell-specific PCR profiling in an unbiased manner, using in vivo models of IAV infection and IAV/Spn co-infection. The TNF superfamily 14 (TNFSF14) ligand-receptor axis was revealed as the driving force behind post-influenza TR-AM death during the early infection phase, enabling the transition to pneumococcal pneumonia, while intrapulmonary transfer of genetically modified TR-AMs and antibody-mediated neutralization of specific pathway components alleviated disease severity. With a mainly neutrophilic expression and a high abundance in the bronchoalveolar fluid (BALF) of patients with severe virus-induced ARDS, TNFSF14 emerged as a key determinant of virus-driven lung injury. Targeting the TNFSF14mediated intercellular communication network in the virus-infected lung can, therefore, improve host defense, minimizing the risk of subsequent bacterial pneumonia, and ameliorating disease outcome. Research Cell biology Infectious disease Find the latest version: https://jci.me/185390/pdf 1 TNF Superfamily Member 14 Drives Post-Influenza Depletion of Alveolar Macrophages Enabling Secondary Pneumococcal Pneumonia Christina Malainou1,2,3, Christin Peteranderl1, Maximiliano Ruben Ferrero2,3,4,5, Ana Ivonne Vazquez-Armendariz1,6, Ioannis Alexopoulos1,2,3, Katharina Franz1,2,3, Klara Knippenberg1,2,3, Julian Better1,2,3,7, Mohammad Estiri1,2,3, Cheng-Yu Wu3,7, Hendrik Schultheis3,5, Judith Bushe8, Maria-Luisa del Rio9, Jose Ignacio Rodriguez-Barbosa9, Klaus Pfeffer10, Stefan Günther3,5, Mario Looso3,5, Achim Dieter Gruber8, István Vadász2,3,7, Ulrich Matt1,2,3, Susanne Herold1,2,3 1 Department of Medicine V, Internal Medicine, Infectious Diseases and Infection Control, Universities of Giessen and Marburg Lung Center (UGMLC), member of the German Center for Lung Research (DZL), member of the German Center for Infection Research (DZIF), Justus Liebig University Giessen, Giessen, Germany. 2 Institute of Lung Health (ILH), Justus Liebig University Giessen, Giessen, Germany. 3 Excellence Cluster Cardio-Pulmonary Institute (CPI), Hessen, Germany. 4 Biomedicine Research Institute of Buenos Aires - CONICET-Partner Institute of the Max Planck Society (IBioBA-MPSP) Buenos Aires, Argentina. 5 Max-Planck-Institute for Heart and Lung Research, Bad Nauheim, Germany. 6 University of Bonn, Transdisciplinary Research Area Life and Health, Organoid Biology, Life & Medical Sciences Institute, Bonn, Germany. 7 Department of Internal Medicine II, Universities of Giessen and Marburg Lung Center, University Hospital Giessen, Justus Liebig University, Member of the German Center for Lung Research (DZL), Giessen, Germany. 8 Department of Veterinary Pathology, Free University Berlin, Berlin, Germany. 2 9 Transplantation Immunobiology Section, Institute of Molecular Biology, Genomics and Proteomics (INBIOMIC), University of León, León, Spain. 10 Institute of Medical Microbiology and Hospital Hygiene, Heinrich Heine University Düsseldorf, Düsseldorf, Germany. Address correspondence to: Susanne Herold, Department of Medicine V, Internal Medicine, Infectious Diseases and Infection Control, University Hospital Giessen and Marburg, Justus-Liebig-University, Klinikstrasse 33, D-35392 Giessen, Germany. Phone: +49-(0)641-985-57061; E-mail: Susann[email protected]e. The authors have declared that no conflict of interest exists. Key words: influenza A virus, pneumonia, alveolar macrophages, TNF superfamily, Streptococcus pneumoniae 3 Abstract 1 Secondary bacterial infection, often caused by Streptococcus pneumoniae (Spn), is 2 one of the most frequent and severe complications of influenza A virus (IAV)-induced 3 pneumonia. Phenotyping of the pulmonary immune cell landscape after IAV infection 4 revealed a substantial depletion of the tissue-resident alveolar macrophage (TR-AM) 5 population at day 7, which was associated with increased susceptibility to Spn 6 outgrowth. To elucidate the molecular mechanisms underlying TR-AM depletion, and 7 to define putative targets for treatment, we combined single-cell transcriptomics and 8 cell-specific PCR profiling in an unbiased manner, using in vivo models of IAV infection 9 and IAV/Spn co-infection. The TNF superfamily 14 (TNFSF14) ligand-receptor axis 10 was revealed as the driving force behind post-influenza TR-AM death during the early 11 infection phase, enabling the transition to pneumococcal pneumonia, while 12 intrapulmonary transfer of genetically modified TR-AMs and antibody-mediated 13 neutralization of specific pathway components alleviated disease severity. With a 14 mainly neutrophilic expression and a high abundance in the bronchoalveolar fluid 15 (BALF) of patients with severe virus-induced ARDS, TNFSF14 emerged as a key 16 determinant of virus-driven lung injury. Targeting the TNFSF14-mediated intercellular 17 communication network in the virus-infected lung can, therefore, improve host defense, 18 minimizing the risk of subsequent bacterial pneumonia, and ameliorating disease 19 outcome. 20 21 4 Introduction 22 Bacterial pneumonia, often caused by Spn, is one of the most common complications 23 of primary IAV infection, increasing the risk of death, intensive care unit (ICU) 24 admission, and requirement for mechanical ventilation (1). While strengthening host 25 defense offers a potential alternative to antibiotics amid globally rising resistance, 26 progress is limited by poor understanding of the immune mechanisms behind severe 27 influenza and the transition to post-viral bacterial pneumonia. Among these, virus28 induced depletion of the TR-AM pool is considered a key factor in promoting secondary 29 bacterial pneumonia, alongside epithelial damage, influx of pro-inflammatory cells, and 30 impaired mechanical clearance (2, 3). TR-AM numbers remain relatively unchanged 31 during homeostasis, with the main function of the cells being surfactant clearance and 32 containment of minor infections (4). This tolerogenic programming, however, can be 33 overridden by abundant viral presence, leading to a pro-inflammatory phenotypic 34 switch, including extensive cytokine release and phagocytosis of viral particles and 35 apoptotic cells (4, 5). TR-AM loss, often observed after severe infection and 36 notoriously known as the ´TR-AM disappearance reaction´ (6, 7), dramatically 37 increases IAV-associated mortality (8, 9), the specific pathomechanisms behind it 38 remain, however, elusive. 39 The TNFSF involves a variety of structurally homologous ligands with multiple 40 functions during development, homeostasis, and tissue response to injury (10). 41 TNFSF14 or LIGHT [homologous to lymphotoxins, exhibits inducible expression and 42 competes with Herpes Simplex Virus (HSV) glycoprotein D for herpes virus entry 43 mediator (HVEM), a receptor expressed by T-lymphocytes] is widely expressed on 44 cells of the hematopoietic compartment (11, 12). In the lung, TNFSF14 has been 45 associated with airway remodeling in asthma, idiopathic pulmonary fibrosis, systemic 46 sclerosis models (13, 14), and more recently, with disease severity in covid-19 (15, 47 5 16). Still, little is known regarding the pathomechanistic role of TNFSF14 in virus48 induced pneumonia. 49 TNFSF14 binds to three different receptors; type I transmembrane lymphotoxin beta 50 receptor (LTbR), HVEM, also known as TNF receptor superfamily 14 (TNFRSF14), 51 and decoy receptor 3 (DcR3), which is only found in primates (10). TNFSF14 receptors 52 have a broad distribution on immune, stromal, and parenchymal cells (17), and 53 orchestrate distinct intracellular pathways. The outcome of TNFSF14 crosslinking to 54 its receptors, therefore, heavily relies on disease context and microenvironmental cues 55 (10). Here, we sought to investigate the molecular mechanisms of post-influenza TR56 AM loss and its consequences for host defense in a model of IAV and IAV/Spn 57 infection, aiming at identifying any putative targets for immune-based pneumonia 58 treatment options. 59 60 6 Results 61 Severe IAV infection increases susceptibility to secondary pneumococcal 62 infection 63 To elucidate the pathomechanisms behind TR-AM death after severe IAV infection and 64 its effect on the establishment of secondary bacterial pneumonia, we established a 65 robust co-infection model (Figure 1A). Disease severity after viral, bacterial, or co66 infection was shown to vary in a pathogenand infection dose-dependent manner. 67 Orotratracheal (o.t.) infection of C57BL/6 wild-type (wt) mice with 500 foci-forming units 68 (ffu) IAV A/PR/8/34 decreased mouse survival by 50% 14 days after infection, whereas 69 intranasal infection with 2000 colony-forming units (cfu) Spn Serotype 3 (PN36 70 NCTC7978) did not affect survival (Figure 1B) or weight loss (Figure 1C). However, 71 IAV infection seven days prior to pneumococcal infection caused massive leukocyte 72 infiltration (Figure 1, D-G) and a 100% lethal outcome (Figure 1B). Upon use of lower 73 IAV and Spn doses (250ffu/20cfu on day 7 post-IAV infection, pi), average survival was 74 calculated at 37.5% (Figure 1B). Despite the low infection doses, bacterial load in the 75 BALF of previously IAV-infected mice was remarkably high 48h after pneumococcal 76 infection, whereas PBS-pretreated mice completely cleared the infection (Figure 1H), 77 suggesting an IAV-associated impaired immune response against invading Spn. We, 78 therefore, characterized the leukocyte landscape of the IAV-infected lung, as distinct 79 immune cell populations and their interactions can differentially affect post-influenza 80 bacterial clearance (3, 18, 19). Flow cytometry profiling of BALF leukocytes (gating 81 strategy depicted in Supplemental Figure 1A) revealed cell-specific kinetics over the 82 infection course (Figure 1I). Of note, BALF TR-AM numbers, which started significantly 83 declining on day 3 pi, were almost completely depleted between 7-11 days pi (Figure 84 1J). Similar results were shown for lung-tissue leukocytes (gating strategy in 85 Supplemental Figure 1B), which could not be acquired through BAL due to their sessile 86 7 nature (Supplemental Figure 2A) (20) or extra-alveolar location (Supplemental Figure 87 2, B-M). Upon co-infection, BALF bacterial outgrowth was observed 24h after 88 pneumococcal infection (Figure 1K), coinciding with the period of maximum TR-AM 89 depletion, despite the abundant presence of bone marrow-derived macrophages 90 (BMDM). TR-AMs presented higher Spn phagocytosis capacity (Figure 1L) and no 91 inferiority in killing capacity to the infection-driven pro-inflammatory BMDM (Figure 92 1M), highlighting the importance of TR-AM preservation for maintaining intact host 93 defense. To address this, we sought to identify the molecular underpinnings of post94 influenza TR-AM death. 95 96 Post-influenza TR-AM death involves the activation of caspase-8 97 Direct viral infection can lead to epithelial cell apoptosis in IAV-induced pneumonia (21, 98 22), posing the question whether this also drives post-IAV TR-AM depletion. Flow 99 cytometry analysis revealed only a small number of IAV-infected TR-AMs (quantified 100 by virus hemagglutinin (HA) expression) with no significant increase over the infection 101 course (Figure 2A). The majority of HA-negative cells had been depleted by day 7 pi 102 (Figure 2A), implying the involvement of a different mechanism with a much higher 103 impact. In accordance with that, when naïve TR-AMs were ex vivo treated with virus104 and cell-free BALF from IAV-infected mice (iBALF) from day 7 pi (day of maximum 105 depletion), we observed a significant decrease in TR-AM survival (Figure 2B) and an 106 increase in caspase-3/7 (Figure 2C) and caspase-8 activity (Figure 2D). Transcriptome 107 analysis of flow-sorted HA-negative TR-AMs on days 3 and 7 pi based on a cell-death 108 gene array revealed an upregulation of multiple apoptosis-related genes, such as Bax, 109 Cd40lg, and Cflar, and necrosis-related genes, including Bmf, Commd4, Defb1, and 110 Parp1 (Figure 2E and Supplemental Material, Cell death arrays wt data). Fold changes 111 of upregulated genes did not differ significantly from baseline, as cell death is mainly 112 8 regulated on a (post-)translational level (23). Concomitant flow cytometry analysis, 113 however, revealed a remarkable increase in apoptotic TR-AMs over the infection 114 course (Figure 2F, gating strategy in Supplemental Figure 3A). As a result, we raised 115 the question whether apoptosis inhibition would improve TR-AM survival. Following 116 pre-incubation with a non-toxic concentration (Supplemental Figure 3, B-C) of 50µM of 117 a specific caspase-3 (Z-DEVD-FMK) or caspase-8 inhibitor (Z-IETD-FMK), naïve TR118 AMs were treated with iBALF. Whereas caspase-3 inhibition only showed a negligible 119 protective effect, TR-AM death was completely abrogated in the caspase-8 inhibition 120 group (Figure 2G). When mice were treated with daily subcutaneous (s.c.) injections 121 of the caspase-8 inhibitor (schematic of experimental layout in Figure 2H), an 122 attenuated weight loss was observed up to day 7 pi (Figure 2I). Caspase-8 inhibition 123 fully protected the TR-AM pool on day 3 pi (Figure 2J), without affecting viral titers 124 (Figure 2K), and significantly mitigated TR-AM loss on day 7 pi (Figure 2L). Caspase125 8 is a known orchestrator of cell death, typically activated upon the crosslinking of a 126 soluble ligand to a death receptor (24), which, together with the primarily virus127 independent TR-AM apoptosis, hinted at a soluble ligand as a driver of TR-AM death. 128 129 IAV pneumonia sensitizes TR-AMs to TNFSF14 ligation 130 Death-inducing members of the TNFSF have been associated with promoting alveolar 131 epithelial cell death and driving post-IAV lung injury (18, 25, 26). As such, we 132 hypothesized that a TNFSF member could be involved in post-IAV TR-AM death and 133 analyzed gene expression patterns of receptors and ligands belonging to the TNFSF 134 signaling network in flow-sorted, HA-negative, TR-AMs from mock-infected and 135 infected mice on days 3 and 7 pi. TNFRSF14, a receptor for TNFSF14, showed a 136 significant upregulation at both time points (Figure 3A and Supplemental material, TNF 137 signaling arrays wt data). TNFRSF14 demonstrated a significant increase in mRNA 138 15 Discussion 287 Lower respiratory tract infections (LRTIs) are a leading global cause of death, with IAV 288 infection playing a major role due to a variety of potential complications, most notably 289 secondary bacterial infections, which greatly increase the risk of respiratory failure and 290 ICU admission, and overall mortality (21, 33, 34). With no causative pharmacological 291 treatment for pneumonia-related lung injury, research has focused on understanding 292 the mechanisms behind severe IAV pneumonia and the transition to post-influenza 293 secondary infection. Proposed mechanisms include bacterial dissemination due to 294 IAV-associated epithelial cell death, fibrin deposition, impaired mechanical clearance, 295 microbial dysbiosis, and interferon-driven suppression of phagocyte function (2, 35, 296 36). As the lungs' first line of defense, IAV-induced TR-AM depletion is a critical step 297 in compromising host immunity. Patient and animal studies have demonstrated that 298 severe viral infections drive TR-AM depletion and niche replenishment by BMDM, with 299 the depletion phase aligning with peak susceptibility to bacterial infection (3, 37-39), 300 yet the involved pathways remain poorly understood. Here, we identified TNFSF14 as 301 a driver of TR-AM loss during IAV pneumonia. 302 In the first week post-infection, TR-AM numbers progressively declined, while other 303 leukocytes gradually entered the alveoli in response to viral infection. TR-AM fate after 304 acute infection is dictated by cell death-inducing mechanisms, impaired self-renewal 305 capacity, and loss of pro-survival signals from the injured neighboring epithelium (6, 306 7). The extent of TR-AM depletion and the intensity of the inflammatory response 307 shape the composition and (re)programming of lung-resident cells after infection (40). 308 Our own previous data indicates that partial TR-AM loss enables the recruitment of 309 circulating BMDM, which are essential for post-viral repair through their transitioning 310 into pro-homeostatic phenotypes (41). Co-existence of newly recruited BMDM and 311 surviving original TR-AMs is the outcome of a balanced immune response, which 312 16 culminates in BMDM-orchestrated tissue repair and return to homeostasis, assisted by 313 the tolerogenic functions of TR-AMs, aimed at restricting epithelial damage (42, 43). 314 Infection severity determines the extent of TR-AM depletion, with a dramatic loss upon 315 severe IAV pneumonia, as demonstrated in our model. Recruitment of pro316 inflammatory immune cells and chemokine abundance contribute to viral clearance, 317 but can also escalate to a dysbalanced immune response (44). The highly pro318 inflammatory programming of BMDM can aggravate local injury and promote aberrant 319 lung remodeling (18, 28), while dysregulated neutrophil migration and activation 320 positively correlate with disease severity and poor patient outcomes (45, 46). Following 321 TR-AM depletion, early recruitment of professional phagocytes such as BMDM and 322 neutrophils failed to control bacterial spread, leading to dramatic bacterial outgrowth 323 within 24h of Spn infection. This aligns with prior studies demonstrating high 324 susceptibility to secondary pneumococcal infection 5-7 days after IAV infection, 325 coinciding with the TR-AM depletion phase (3). IFN-γ, which is profusely released in 326 the alveoli as part of the antiviral response, heavily impairs TR-AM antibacterial 327 properties, as it downregulates the macrophage receptor with collagenous structure 328 scavenger receptor (MARCO) on TR-AM surface, one of the key elements in TR-AM 329 antibacterial response (47, 48). Defective chemokine production by macrophages, as 330 observed in severe IAV infection and sepsis models, further aggravates disease 331 outcome (49, 50). The near-complete TR-AM loss within a microenvironment of 332 exuberant death-inducing signals in our model highlights an additional important step 333 towards the establishment of lethal post-viral pneumococcal pneumonia. Though a 334 considerable advantage in terms of antibacterial properties has been described for 335 infection-experienced BMDM and newly originating, BMDM-derived, TR-AMs after IAV 336 infection (19), pneumococcal infection in that model was performed weeks after the 337 initial viral hit, as opposed to the window of increased host vulnerability during the acute 338 17 infection phase described in our study. At this point, TR-AMs showed superior 339 phagocytic capacity and similar killing capacity to BMDM. Thus, preserving TR-AMs 340 early on, may provide critical protection until re-establishment of a fully functional 341 resident macrophage niche, including infection-trained BMDM, has been completed. 342 The remarkable TR-AM loss in C57BL/6 wild-type mice in our study differs from 343 previously published data, where mouse genetic strain determined TR-AM survival, 344 with BALB/c mice exhibiting a drastic TR-AM reduction, as opposed to C57BL/6 mice, 345 which maintained TR-AMs of an altered phenotype (51). In this study by Califano et 346 al., a relatively low dose of IAV PR8 was administered intranasally, whereas we 347 administered a high viral dose orotracheally, aiming at inducing severe pneumonia. 348 Animal strain and administration route for the infection may, therefore, depict a 349 limitation of our study, as results may differ for different in vivo models. 350 Transcriptomic and flow cytometry analyses revealed apoptosis as the primary cause 351 of post-influenza TR-AM death, largely independent of direct viral infection, suggesting 352 the involvement of a death-inducing ligand. While apoptosis promotes early viral 353 spread (22, 52), it also limits infection through elimination of infected cells (53, 54). 354 Leukocyteand virus-driven alveolar epithelial cell apoptosis, however, compromises 355 the gas-blood barrier and impairs gas exchange (25, 28). Apoptosis inhibition can, 356 therefore, influence infection outcome. To compensate for any effect of caspase 357 inhibition on early virus propagation, we began our treatment on day 2 pi and observed 358 no significant difference in viral titers on day 3 pi, the peak of viral replication in this 359 model (28). Caspase-8 inhibition was chosen based on our in vitro data, which showed 360 a clear advantage over caspase-3 inhibition after TR-AM iBALF treatment. This initially 361 surprising result could be potentially explained through PANoptosis as the joint result 362 of pyroptosis, apoptosis, and necroptosis. This would permit cell death via caspase-3363 independent apoptosis or pyroptosis. Caspase-8 is involved in both these pathways 364 18 (55-58) and is currently the only known programmed cell death (PCD) member that 365 connects all PCD pathways. This can explain the complete abrogation of TR-AM death 366 on day 3 pi, the reduced weight loss, and the improved TR-AM survival upon caspase367 8 inhibition. Nevertheless, the PANoptosis concept suggests that a single PCD 368 component cannot individually rescue cells once PANoptosis has been initiated, which 369 might explain why TR-AM loss was not completely prevented on day 7 pi, when death370 inducing signals are highly abundant (59-61). 371 Stochastic interrogation of TNFSF members revealed significant upregulation of 372 TNFSF14 in infected mouse lungs, with high soluble TNFSF14 levels also found in 373 BALF from patients with severe virus-induced ARDS. Previous studies on severe viral 374 pneumonia and sepsis positively linked elevated BALF/serum TNFSF14 levels to 375 disease severity (15, 16, 62, 63). Depending on cell type, pathogen interaction, and 376 receptor availability, TNFSF14 can influence cell survival, profile (re)programming, 377 immune response establishment, and infection memory (10, 64). TNFSF14 has been 378 previously described as a determinant of macrophage survival, phenotype, and 379 antibacterial properties (65-67), however, extensive studies regarding post-influenza 380 TR-AM death are lacking. In our study, TNFSF14 deletion or blockade preserved TR381 AMs and improved survival and weight loss during co-infection. These benefits were 382 not solely due to reduced bacterial burden but likely stemmed from enhanced tissue 383 repair and accelerated return to homeostasis due to improved TR-AM survival. 384 Previous work from our lab has highlighted TR-AMs as drivers of epithelial repair (41) 385 and mitigators of lung inflammation, even at the cost of bacterial clearance (43). 386 Further experiments would be required to fully address the role of TNFSF14 on 387 pathogen resistance and tolerance in the context of co-infection beyond the lung388 confined effect on TR-AM survival. 389 19 Post-influenza TNFSF14-induced TR-AM death was cell-specific, with no significant 390 differences in other leukocytes (except neutrophils on day 3 pi) or in endothelial, 391 mesenchymal, or epithelial cells between wt and Tnfsf14 ⁻ / ⁻ mice. TNFSF14 treatment 392 did not worsen virus-induced death in alveolar epithelial cells, suggesting TNFSF14 is 393 not a strong driver of post-influenza distal epithelial cell apoptosis. We identified 394 neutrophils as the main leukocyte source of TNFSF14, which is in accordance with 395 previously published data (68, 69). TNFSF14 has been shown to play an instrumental 396 role in NK and T cell activation and expansion (11, 17, 70) and DC maturation (71) and 397 may thus serve as an intermediate between the acute and adaptive immune response. 398 Aberrant release due to dysregulated neutrophil activation could offer an alternative 399 explanation for the abundant TNFSF14 presence upon severe infection. This is in 400 accordance with literature, as high circulating or organ-specific TNFSF14 levels have 401 been positively correlated with highly inflammatory states (13, 15, 72-74) and blocking 402 of the ligand was shown to limit inflammation and attenuate organ injury (75). 403 TNFRSF14 and LTbR, the two competitor TNFSF14 receptors, followed distinct 404 kinetics in terms of transcriptional regulation and protein expression in TR-AMs during 405 the infection course. Attenuated TR-AM loss after IAV infection of Ltbr-/- mice, 406 compared to Tnfrsf14-/- mice, and improved survival after intrapulmonary transfer of 407 Ltbr-/- TR-AMs to co-infected wt mice, demonstrated a stronger impact on TR-AM death 408 for LTbR. Given the more prominent TR-AM preservation in Tnfsf14-/- mice compared 409 to Ltbr-/- mice, we hypothesize that cell death could also be initiated through ligation of 410 TNFSF14 to TNFRSF14, potentially through activation of a co-receptor, as TNFRSF14 411 lacks a pro-death domain (10, 76). It should be noted, however, that engagement of 412 LTbR by TNFSF14 on macrophages is not merely confined to apoptosis induction. 413 Transforming growth factor-beta (TGF-β) can be secreted upon crosslinking (77), 414 which has been shown to drive an immunoparalysis state in the aftermath of infection, 415 20 further enhancing secondary infection susceptibility (78). LTbR-TNFSF14 interaction 416 on the endothelium alters microvasculature structure, which can in turn favor the 417 recruitment of immune cells (79). The intricate nature of TNFSF14-TNFRSF14/LTbR 418 interactions thus points at a multitude of potential roles for the signaling axis in the 419 context of influenza, besides TR-AM depletion. Nevertheless, with clinical trials in the 420 context of virus-induced pneumonia and systemic inflammation already revealing 421 beneficial safety profiles (63, 80, 81), therapeutic interventions disrupting TNFSF14422 initiated intercellular pathways to preserve TR-AM function appear as promising 423 approaches for improving host defense in the context of IAV pneumonia. 424 425 21 Methods 426 Sex as a biological variable. Sex was not considered as a biological variable for patient 427 samples. Both male and female mice were used for all studies. 428 Mice. Wt C57BL/6 mice were purchased from Charles River Laboratories. Tnfsf14-/- 429 (82), Tnfrsf14-/- (83), and Ltbr-/- (84) mice were a gift from Prof. Klaus Pfeffer (Heinrich 430 Heine University Düsseldorf, Düsseldorf, Germany). Tnfsf10-/- (85) mice were obtained 431 from AMGen. All mice were bred under specific-pathogen-free conditions (SPF) and 432 infected at 10-12 weeks of age. 433 In vivo infection. For in vivo IAV infection experiments, mice were orotracheally 434 inoculated with 250-1000ffu of A/Puerto Rico/8/1934 (PR8, H1N1) influenza virus. 435 Control groups were inoculated with sterile PBS-/-. For co-infection experiments, mice 436 were i.n. infected with 20cfu Spn [serotype 3, strain PN36 (NCTC 7978), provided by 437 the group of M. Witzenrath, Department of Infectious Diseases and Pulmonary 438 Medicine, Charité, University Medicine Berlin, Berlin, Germany] 7 days after IAV 439 infection. 440 In vivo treatment. For apoptosis inhibition, wt mice were infected with 500ffu (day 7 441 experiments) or 1000ffu IAV (day 3 experiments) and treated with s.c. injections of 442 10mg/kg of a specific caspase-8 inhibitor (Z-IETD-FMK, R&D Systems), or a DMSO 443 control. For day 3 experiments, treatment involved a single injection on day 2 pi, 444 whereas daily injections were applied days 2-6 pi for analysis on day 7 pi. Neutrophil 445 depletion was performed through the i.p. application of 200µg anti-1A8 antibody 446 (InVivoPlus rat anti-mouse Ly6G, cat. BP0075-1, BioXCell) or an anti-2A3 isotype 447 control (InVivoPlusTM rat IgG2a isotype control, anti-trinitrophenol, cat. BP0089, 448 BioXCell) diluted in sterile PBS in mice infected with 500ffu IAV on days -1, 1, 3, and 449 5 pi. TNFSF14 neutralization was achieved with a mouse anti-mouse LIGHT blocking 450 antibody (clone 3D11, IgG2b, k, isotype control mouse IgG2b, clone 27-35, 451 22 BioLegend), kindly provided by Prof. José Ignacio Rodríguez Barbosa and Prof. Maria452 Luisa del Rio (INBIOMIC, University of León, León, Spain). A single i.p. injection of 453 500µg of antibody or isotype control was performed two days after IAV infection with 454 500ffu. For in vivo treatment with rTNFSF14, 10µg of carrier-free mouse rTNFSF14 455 (cat. 1794-LT, R&D Systems) diluted in 0.03mL PBS-/- were orotracheally administered 456 to IAV-infected mice on days 1 and 2 pi for analysis on day 3 pi. 457 Adoptive TR-AM transfer. Murine TR-AM were obtained from the BALF of naïve wt, 458 Tnfrsf14-/-, and Ltbr-/- mice, as previously described (25). Adoptive transfer of 400,000 459 TR-AM per mouse was performed on day 3 after infection of wt mice with 250ffu IAV, 460 with an engraftment efficiency of 14-20% calculated 24h later, at a time point where 461 over 70% of the original TR-AM pool was still detectable in the BALF of infected mice 462 (data not shown). Secondary pneumococcal infection was performed four days later. 463 TR-AM isolation and cell culture for ex vivo treatment. Following BALF extraction from 464 naïve mice, cells were resuspended in full TR-AM medium (RPMI-1640/2% fetal 465 bovine serum (FBS)/2.5% HEPES/1% L-glutamine/1% penicillin/streptomycin). TR466 AMs were seeded at a density of 10-50,000 cells/well on a 96-well plate. 467 Colorimetric viability assay for ex vivo treated TR-AMs. Primary TR-AMs isolated from 468 the BALF of naïve wt mice and BALF TR-AMs from control patients who underwent 469 routine bronchoscopy for diagnostic purposes and revealed normal BALF cellularity 470 were treated with 0.1mL TR-AM medium containing 10% iBALF or rTNFSF14 for 24h. 471 For caspase inhibition experiments, cells were pre-incubated in 50µM of a specific 472 caspase-3 (Z-DEVD-FMK, cat. FMK004, R&D Systems) or caspase-8 inhibitor (Z473 IETD-FMK, cat. FMK007, R&D Systems) for 3h prior to BALF treatment. Viability was 474 assessed via colorimetric assay (Cell Counting Kit-8, cat. 96992, Sigma Aldrich), as 475 per the manufacturer´s instructions, and was considered proportional to the measured 476 light absorbance. Absorbance was measured in an iMark microplate reader (Bio-Rad). 477 23 Caspase-3/7 and caspase-8 activity. Wt, Tnfrsf14-/-, and Ltbr-/- BALF TR-AMs were 478 treated with 0.1mL TR-AM medium containing 10% day 0 or day 7 iBALF from wt or 479 Tnfsf14-/- mice for 24h. Lyophilized Caspase-Glo® 3/7 or Caspase-Glo® 8 substrate 480 (Promega) was resuspended in 10mL luciferase-containing Caspase-Glo® buffer 481 (Promega) and added á 0.1mL/well to the cells. After 1h incubation at RT, cells were 482 transferred to a black 96-well plate for luminescence detection using a 520/25 filter in 483 an FLx800 fluorescence reader (BioTek Instruments). For ligand blocking experiments, 484 iBALF had been previously incubated with 1µg/mL of the mouse anti-mouse TNFSF14 485 antibody or an isotype control for 1h at 4°C prior to iBALF treatment. For TNFSF14 486 receptor blocking experiments, TR-AMs were pretreated with 1µg/mL for 1h at 37°C, 487 CO2, prior to iBALF treatment, to achieve receptor saturation. Antibodies included 488 Armenian hamster anti-mouse anti-TNFRSF14 (CD270 (HVEM) monoclonal antibody, 489 LH1, functional grade, cat. 16-5962-85, eBioscienceTM), Armenian hamster anti-mouse 490 isotype control (cat. 16-4888-85, eBioscienceTM), rat anti-mouse anti-LTbR (clone 4H8491 WH2, developed by the laboratory of Dr. Carl F. Ware, marketed by AdipoGen Life 492 Sciences, and kindly provided by Prof. José Ignacio Rodríguez Barbosa and Prof. 493 Maria-Luisa del Rio, University of León, Spain), and rat anti-mouse IgG2a isotype 494 control (cat. AG-35B-0002-C050, AdipoGen). To dissect the roles of soluble and 495 transmembrane TNFSF14 on TR-AM apoptosis, flow-sorted day 3 TR-AMs were either 496 treated with iBALF or co-cultured with neutrophils at a 1:5 ratio for 24h. Marimastat 497 (cat. M2699, Sigma-Aldrich) was added at a concentration of 10µM to prevent 498 TNFSF14 shedding. 499 Spn load after in vivo infection. Two days after IAV infection and 6-72h after Spn 500 infection, BALF, lungs, and spleens from co-infected mice were harvested and 501 homogenized. A series of inoculum dilutions in NaCl was prepared for each sample in 502 1:10 dilution steps. For each dilution step, 4 x 0.01mL inoculum were pipetted on a 503 24 blood agar plate and stored at 37°C overnight. Bacterial load was calculated by 504 counting the average number of separately grown colonies, multiplied by 10number of 505 dilution step*100 (= number of colonies in 1mL). 506 Ex vivo phagocytosis and killing assay. Wt and Ltbr-/- naïve BALF TR-AMs were 507 isolated as previously described. Cells were seeded á 100,000 cells/well on a 96-well 508 round-bottom plate and ex vivo infected with Spn at an multiplicity of infection (MOI) 509 1000 for 10min (37°C). Cells were vigorously washed 5 times in ice-cold PBS to 510 remove any extracellular bacteria and lysed in water. An inoculum dilution series of cell 511 lysates was pipetted on blood agar plates as described above. Total colony count on 512 the following day depicted phagocytosed bacteria. For flow cytometry-based 513 comparison of day 8 TR-AMs and BMDM Spn uptake and killing, 100,000 cells of whole 514 BALF cell samples were ex vivo infected with Spn at an MOI 100 for 10min (t0). Cells 515 were washed three times with ice-cold PBS and were either fixed and permeabilized 516 using the eBioscience™ Foxp3/Transcription Factor Staining Buffer Set (cat. 00-5523517 00, Invitrogen), as per the manufacturer´s instructions, or returned to the incubator for 518 an additional 30min (t1) in sterile medium, after which the same procedure was 519 performed. Staining was performed in two steps, starting with an anti-Spn antibody 520 (rabbit, cat. PA17259, Invitrogen) or a rabbit IgG isotype control (cat. ab172730, 521 abcam) at a concentration of 40µg/mL for 1h at RT, followed by leukocyte surface 522 staining containing a secondary donkey anti-rabbit IgG (H+L) Alexa Fluor™ 555 523 antibody (cat. A-31572, Invitrogen). Non-infected samples were used as 524 autofluorescence controls. Phagocytosis capacity was reflected in the percentage of 525 Spn+ cells at (t0), killing capacity at (t1) was determined for each macrophage 526 population as follows: percent killing = 100 − [(%Spn+ cells at t1/%Spn+ cells at t0) × 527 100]. 528 31 10. Ward-Kavanagh LK, et al. The TNF Receptor Superfamily in Co-stimulating and 631 Co-inhibitory Responses. Immunity. 2016;44(5):1005-19. 632 11. del Rio ML, et al. Immunotherapeutic targeting of LIGHT/LTbetaR/HVEM 633 pathway fully recapitulates the reduced cytotoxic phenotype of LIGHT-deficient 634 T cells. MAbs. 2016;8(3):478-90. 635 12. Otterdal K, et al. Platelet-derived LIGHT induces inflammatory responses in 636 endothelial cells and monocytes. Blood. 2006;108(3):928-35. 637 13. Mehta AK, et al. Tumor necrosis factor family member LIGHT acts with IL-1beta 638 and TGF-beta to promote airway remodeling during rhinovirus infection. Allergy. 639 2018;73(7):1415-24. 640 14. Doherty TA, et al. The tumor necrosis factor family member LIGHT is a target 641 for asthmatic airway remodeling. Nature medicine. 2011;17(5):596-603. 642 15. Arunachalam PS, et al. Systems biological assessment of immunity to mild 643 versus severe COVID-19 infection in humans. Science. 2020;369(6508):1210644 20. 645 16. Perlin DS, et al. Levels of the TNF-Related Cytokine LIGHT Increase in 646 Hospitalized COVID-19 Patients with Cytokine Release Syndrome and ARDS. 647 mSphere. 2020;5(4). 648 17. Seo GY, et al. LIGHT-HVEM Signaling in Innate Lymphoid Cell Subsets 649 Protects Against Enteric Bacterial Infection. Cell Host Microbe. 2018;24(2):249650 60 e4. 651 18. Ellis GT, et al. TRAIL+ monocytes and monocyte-related cells cause lung 652 damage and thereby increase susceptibility to influenza-Streptococcus 653 pneumoniae coinfection. EMBO Rep. 2015;16(9):1203-18. 654 19. Aegerter H, et al. Influenza-induced monocyte-derived alveolar macrophages 655 confer prolonged antibacterial protection. Nat Immunol. 2020;21(2):145-57. 656 32 20. Westphalen K, et al. Sessile alveolar macrophages communicate with alveolar 657 epithelium to modulate immunity. Nature. 2014;506(7489):503-6. 658 21. Peteranderl C, et al. Human Influenza Virus Infections. Semin Respir Crit Care 659 Med. 2016;37(4):487-500. 660 22. Herold S, et al. Apoptosis signaling in influenza virus propagation, innate host 661 defense, and lung injury. J Leukoc Biol. 2012;92(1):75-82. 662 23. D'Arcy MS. Cell death: a review of the major forms of apoptosis, necrosis and 663 autophagy. Cell Biol Int. 2019;43(6):582-92. 664 24. Orning P, and Lien E. Multiple roles of caspase-8 in cell death, inflammation, 665 and innate immunity. J Leukoc Biol. 2021;109(1):121-41. 666 25. Hogner K, et al. Macrophage-expressed IFN-beta contributes to apoptotic 667 alveolar epithelial cell injury in severe influenza virus pneumonia. PLoS Pathog. 668 2013;9(2):e1003188. 669 26. Peteranderl C, et al. Macrophage-epithelial paracrine crosstalk inhibits lung 670 edema clearance during influenza infection. J Clin Invest. 2016;126(4):1566-80. 671 27. Skeate JG, et al. TNFSF14: LIGHTing the Way for Effective Cancer 672 Immunotherapy. Front Immunol. 2020;11:922. 673 28. Herold S, et al. Lung epithelial apoptosis in influenza virus pneumonia: the role 674 of macrophage-expressed TNF-related apoptosis-inducing ligand. J Exp Med. 675 2008;205(13):3065-77. 676 29. Celik S, et al. Proinflammatory and prothrombotic effects on human vascular 677 endothelial cells of immune-cell-derived LIGHT. Eur J Med Res. 678 2009;14(4):147-56. 679 30. Davis RWt, et al. Luminol Chemiluminescence Reports Photodynamic Therapy680 Generated Neutrophil Activity In Vivo and Serves as a Biomarker of Therapeutic 681 Efficacy. Photochem Photobiol. 2019;95(1):430-8. 682 33 31. Narasaraju T, et al. Excessive neutrophils and neutrophil extracellular traps 683 contribute to acute lung injury of influenza pneumonitis. Am J Pathol. 684 2011;179(1):199-210. 685 32. Reilly EC, et al. The Effects of Acute Neutrophil Depletion on Resolution of 686 Acute Influenza Infection, Establishment of Tissue Resident Memory (TRM), 687 and Heterosubtypic Immunity. PLoS One. 2016;11(10):e0164247. 688 33. Okahashi N, et al. Secondary streptococcal infection following influenza. 689 Microbiol Immunol. 2022;66(6):253-63. 690 34. Lim YK, et al. Impact of bacterial and viral coinfection in community-acquired 691 pneumonia in adults. Diagn Microbiol Infect Dis. 2019;94(1):50-4. 692 35. Connolly E, and Hussell T. The Impact of Type 1 Interferons on Alveolar 693 Macrophage Tolerance and Implications for Host Susceptibility to Secondary 694 Bacterial Pneumonia. Front Immunol. 2020;11:495. 695 36. McKelvey M, et al. IL-10 Counteracts IFN-gamma to Alleviate Acute Lung Injury 696 in a Viral-Bacterial Superinfection Model. Am J Respir Cell Mol Biol. 697 2024;71(1):110-20. 698 37. Cole SL, et al. M1-like monocytes are a major immunological determinant of 699 severity in previously healthy adults with life-threatening influenza. JCI Insight. 700 2017;2(7):e91868. 701 38. Liao M, et al. Single-cell landscape of bronchoalveolar immune cells in patients 702 with COVID-19. Nat Med. 2020;26(6):842-4. 703 39. Smith AM, and Smith AP. A Critical, Nonlinear Threshold Dictates Bacterial 704 Invasion and Initial Kinetics During Influenza. Sci Rep. 2016;6:38703. 705 40. Martin FP, et al. Alveolar Macrophages: Adaptation to Their Anatomic Niche 706 during and after Inflammation. Cells. 2021;10(10). 707 34 41. Pervizaj-Oruqaj L, et al. Alveolar macrophage-expressed Plet1 is a driver of 708 lung epithelial repair after viral pneumonia. Nat Commun. 2024;15(1):87. 709 42. Uderhardt S, et al. Resident Macrophages Cloak Tissue Microlesions to Prevent 710 Neutrophil-Driven Inflammatory Damage. Cell. 2019;177(3):541-55 e17. 711 43. Better J, Estiri M, et al. Cell type-specific efferocytosis determines functional 712 plasticity of alveolar macrophages. Sci Immunol. 2025;10(107):eadl3852. 713 44. Lamichhane PP, and Samarasinghe AE. The Role of Innate Leukocytes during 714 Influenza Virus Infection. J Immunol Res. 2019;2019:8028725. 715 45. Tang BM, et al. Neutrophils-related host factors associated with severe disease 716 and fatality in patients with influenza infection. Nat Commun. 2019;10(1):3422. 717 46. Brandes M, et al. A systems analysis identifies a feedforward inflammatory 718 circuit leading to lethal influenza infection. Cell. 2013;154(1):197-212. 719 47. Sun K, and Metzger DW. Inhibition of pulmonary antibacterial defense by 720 interferon-gamma during recovery from influenza infection. Nat Med. 721 2008;14(5):558-64. 722 48. Wu M, et al. Immunomodulators targeting MARCO expression improve 723 resistance to postinfluenza bacterial pneumonia. Am J Physiol Lung Cell Mol 724 Physiol. 2017;313(1):L138-L53. 725 49. Shahangian A, et al. Type I IFNs mediate development of postinfluenza 726 bacterial pneumonia in mice. J Clin Invest. 2009;119(7):1910-20. 727 50. Deng JC, et al. Sepsis-induced suppression of lung innate immunity is mediated 728 by IRAK-M. J Clin Invest. 2006;116(9):2532-42. 729 51. Califano D, et al. Effects of Influenza on Alveolar Macrophage Viability Are 730 Dependent on Mouse Genetic Strain. J Immunol. 2018;201(1):134-44. 731 52. Wurzer WJ, et al. Caspase 3 activation is essential for efficient influenza virus 732 propagation. EMBO J. 2003;22(11):2717-28. 733 35 53. Peteranderl C, and Herold S. The Impact of the Interferon/TNF-Related 734 Apoptosis-Inducing Ligand Signaling Axis on Disease Progression in 735 Respiratory Viral Infection and Beyond. Front Immunol. 2017;8:313. 736 54. FitzGerald ES, et al. Competitive Cell Death Interactions in Pulmonary Infection: 737 Host Modulation Versus Pathogen Manipulation. Front Immunol. 2020;11:814. 738 55. Bhadra K. A Mini Review on Molecules Inducing Caspase-Independent Cell 739 Death: A New Route to Cancer Therapy. Molecules. 2022;27(19). 740 56. Sarhan J, et al. Caspase-8 induces cleavage of gasdermin D to elicit pyroptosis 741 during Yersinia infection. Proc Natl Acad Sci U S A. 2018;115(46):E10888-E97. 742 57. Hughes SA, et al. Caspase-8-driven apoptotic and pyroptotic crosstalk causes 743 cell death and IL-1beta release in X-linked inhibitor of apoptosis (XIAP) 744 deficiency. EMBO J. 2023;42(5):e110468. 745 58. Pang J, and Vince JE. The role of caspase-8 in inflammatory signalling and 746 pyroptotic cell death. Semin Immunol. 2023;70:101832. 747 59. Sauler M, et al. Cell Death in the Lung: The Apoptosis-Necroptosis Axis. Annu 748 Rev Physiol. 2019;81:375-402. 749 60. Gullett JM, et al. It's All in the PAN: Crosstalk, Plasticity, Redundancies, 750 Switches, and Interconnectedness Encompassed by PANoptosis Underlying 751 the Totality of Cell Death-Associated Biological Effects. Cells. 2022;11(9). 752 61. Wang Y, and Kanneganti TD. From pyroptosis, apoptosis and necroptosis to 753 PANoptosis: A mechanistic compendium of programmed cell death pathways. 754 Comput Struct Biotechnol J. 2021;19:4641-57. 755 62. Haljasmagi L, et al. Longitudinal proteomic profiling reveals increased early 756 inflammation and sustained apoptosis proteins in severe COVID-19. Sci Rep. 757 2020;10(1):20533. 758 36 63. Qu HQ, et al. Circulating LIGHT (TNFSF14) and Interleukin-18 Levels in Sepsis759 Induced Multi-Organ Injuries. Biomedicines. 2022;10(2). 760 64. Heo SK, et al. LIGHT (TNFSF14) Increases the Survival and Proliferation of 761 Human Bone Marrow-Derived Mesenchymal Stem Cells. PLoS One. 762 2016;11(11):e0166589. 763 65. Petreaca ML, et al. Vascular endothelial growth factor promotes macrophage 764 apoptosis through stimulation of tumor necrosis factor superfamily member 14 765 (TNFSF14/LIGHT). Wound Repair Regen. 2008;16(5):602-14. 766 66. Heo SK, et al. LIGHT enhances the bactericidal activity of human monocytes 767 and neutrophils via HVEM. J Leukoc Biol. 2006;79(2):330-8. 768 67. Yuan X, et al. LIGHT is increased in patients with coronary disease and 769 regulates inflammatory response and lipid metabolism in oxLDL-induced THP770 1 macrophages. Biochem Biophys Res Commun. 2017;490(3):732-8. 771 68. Conlon TM, et al. Inhibition of LTbetaR signalling activates WNT-induced 772 regeneration in lung. Nature. 2020;588(7836):151-6. 773 69. Zhang J, et al. Two waves of pro-inflammatory factors are released during the 774 influenza A virus (IAV)-driven pulmonary immunopathogenesis. PLoS Pathog. 775 2020;16(2):e1008334. 776 70. Desai P, et al. The TNF Superfamily Molecule LIGHT Promotes the Generation 777 of Circulating and Lung-Resident Memory CD8 T Cells following an Acute 778 Respiratory Virus Infection. J Immunol. 2018;200(8):2894-904. 779 71. Holmes TD, et al. Licensed human natural killer cells aid dendritic cell 780 maturation via TNFSF14/LIGHT. Proc Natl Acad Sci U S A. 781 2014;111(52):E5688-96. 782 37 72. Otterdal K, et al. Increased Serum Levels of LIGHT/TNFSF14 in Nonalcoholic 783 Fatty Liver Disease: Possible Role in Hepatic Inflammation. Clin Transl 784 Gastroenterol. 2015;6(7):e95. 785 73. Hsu CY, et al. Circulating TNFSF14 (Tumor Necrosis Factor Superfamily 14) 786 Predicts Clinical Outcome in Patients With Stable Coronary Artery Disease. 787 Arterioscler Thromb Vasc Biol. 2019;39(6):1240-52. 788 74. Manresa MC, et al. A Deficiency in the Cytokine TNFSF14/LIGHT Limits 789 Inflammation and Remodeling in Murine Eosinophilic Esophagitis. J Immunol. 790 2022. 791 75. Ware CF, et al. Realigning the LIGHT signaling network to control dysregulated 792 inflammation. J Exp Med. 2022;219(7). 793 76. Shui JW, and Kronenberg M. HVEM is a TNF Receptor with Multiple Regulatory 794 Roles in the Mucosal Immune System. Immune Netw. 2014;14(2):67-72. 795 77. Liang QS, et al. Splenectomy improves liver fibrosis via tumor necrosis factor 796 superfamily 14 (LIGHT) through the JNK/TGF-beta1 signaling pathway. Exp Mol 797 Med. 2021;53(3):393-406. 798 78. Roquilly A, et al. Local Modulation of Antigen-Presenting Cell Development after 799 Resolution of Pneumonia Induces Long-Term Susceptibility to Secondary 800 Infections. Immunity. 2017;47(1):135-47 e5. 801 79. Ramachandran M, et al. Tailoring vascular phenotype through AAV therapy 802 promotes anti-tumor immunity in glioma. Cancer Cell. 2023;41(6):1134-51 e10. 803 80. Perlin DS, et al. Randomized, double-blind, controlled trial of human anti-LIGHT 804 monoclonal antibody in COVID-19 acute respiratory distress syndrome. J Clin 805 Invest. 2022;132(3). 806 38 81. Zhang M, et al. A Randomized Phase 1 Study to Assess the Safety and 807 Pharmacokinetics of the Subcutaneously Injected Anti-LIGHT Antibody, 808 SAR252067. Clin Pharmacol Drug Dev. 2017;6(3):292-301. 809 82. Scheu S, et al. Targeted disruption of LIGHT causes defects in costimulatory T 810 cell activation and reveals cooperation with lymphotoxin beta in mesenteric 811 lymph node genesis. J Exp Med. 2002;195(12):1613-24. 812 83. Wang Y, et al. The role of herpesvirus entry mediator as a negative regulator of 813 T cell-mediated responses. J Clin Invest. 2005;115(3):711-7. 814 84. Futterer A, et al. The lymphotoxin beta receptor controls organogenesis and 815 affinity maturation in peripheral lymphoid tissues. Immunity. 1998;9(1):59-70. 816 85. Cretney E, et al. Increased susceptibility to tumor initiation and metastasis in 817 TNF-related apoptosis-inducing ligand-deficient mice. J Immunol. 818 2002;168(3):1356-61. 819 820 39 Figures 821 822 Figure 1. IAV infection increases susceptibility to secondary pneumococcal 823 infection. (A) Schematic representation of the co-infection model, with pneumococcal 824 infection taking place 7 days after IAV infection. (B-C) Survival (B) and weight loss (C) 825 40 after IAV/Spn co-infection of wt mice (n=6-8, data pooled from five independent 826 experiments). (D-G) Representative histological images of mock-infected (D), Spn827 infected (E), IAV-infected (F), or mice infected with IAV 7 days prior to Spn infection 828 (G), lungs harvested ten days after IAV infection. Scale bar set at 100µm, data pooled 829 from two independent experiments. (H) Bacterial load in the BALF of IAVor mock830 infected mice nine days after IAV infection and 48h after Spn infection (mean ± SEM, 831 n=6-9, data representative of three independent experiments). (I) Leukocyte 832 populations including neutrophils (n=7-9), BMDM (n=8-9), NK cells (n=7-9), T cells 833 (n=3-9), NK1.1+ NKT cells (n=3-9), and B cells (n=5-9), in the BALF of IAV-infected 834 mice 0-14 days pi (mean ± SEM, data pooled from sixteen independent experiments). 835 (J) TR-AM population during the IAV infection course (means ± SEM, n=3-10, data 836 pooled from six independent experiments). (K) BALF bacterial load 6-72h after 837 pneumococcal superinfection performed seven days post-IAV infection (mean ± SEM, 838 n=3-9 per time point, data pooled from three independent experiments). (L-M) Spn 839 phagocytosis capacity (L) depicted as %Spn+ cells 10min (t0) after infection and killing 840 capacity (M, % killing at t1 over t0) for day 8 TR-AMs and BMDM, n=5, data 841 representative of three independent experiments. Significance was determined by log842 rank (Mantel-Cox) test, unpaired 2-tailed t-test, and by 1-way ANOVA with Tukey’s 843 posthoc test; *p<0.05, **p<0.01, ***p <0.001, ****p <0.0001. 844 47 genes on day 3 pi (C) and day 7 pi (D) over mock-infected wt and Tnfsf14-/- TR-AMs, 921 n=3-5, data pooled from five different experiments. Wt data extracted from the data set 922 presented in Figure 2E. (E) Caspase-3/7 activity after TR-AM treatment with wt and 923 Tnfsf14-/- iBALF. Graph represents means ± SEM, n=4 per condition, data 924 representative of three independent experiments. (F) Body weight of wt and Tnfsf14-/- 925 mice over the IAV infection course. Graph represents means ± SEM of weight at each 926 time point, n=12-13, data pooled from six independent experiments, wt controls 927 including data presented in Figure 3M). (G-H) BALF (G) and lung tissue (H) TR-AMs 928 on day 7 pi after anti-TNFSF14 treatment on day 2 pi. Graph represents means ± SEM, 929 n=9-13, data pooled from four independent experiments. (I) Body weight of anti930 TNFSF14 and isotype-treated wt mice after IAV infection. Graph represents means ± 931 SEM of weight at each time point, n=10-11, data pooled from four independent 932 experiments. (J) Caspase-3/7 activity after 24h iBALF treatment of anti-TNFSF14 pre933 treated naïve TR-AMs. Data pooled from six independent experiments (means ± SEM, 934 n=16). Significance was determined by unpaired 2-tailed t-test, 1-way, or 2-way 935 ANOVA with Tukey’s posthoc test; *p<0.05, **p<0.01, ***p <0.001, ****p <0.0001. 936 937 48 938 Figure 6. Neutrophils are the main cellular source of TNFSF14 during IAV 939 infection. (A) TNFSF14 expression on leukocytes, epithelial, endothelial cells, and 940 MC in the lungs of non-infected and day 7 mice, n=4-5, data pooled from two 941 independent experiments. (B) TNFSF14 expression in different lung regions of mock942 49 and IAV-infected wt mice on day 7 pi (n=3-4, data representative of two independent 943 experiments). Scale bar set at 25µm. (C) Serum TNFSF14 on days 0, 3, and 7 pi (n=3944 5, data pooled from three independent experiments). (D-E) Leukocyte scRNA-Seq 945 analysis on day 3 (D) and day 7 pi (E), n=4, data from two independent experiments. 946 Monocyte and neutrophil clusters on day 3 pi were characterized according to the gene 947 signature of the top 5 uniquely expressed genes per cluster. Dot plots depict Tnfsf14, 948 Ltbr, and Tnfrsf14 expression. (F) qPCR analysis for Tnfsf14 expression in blood and 949 BALF neutrophils (n=4, data pooled from two independent experiments). (G) qPCR 950 analysis for TNFSF14 expression in BALF neutrophils from patients with severe IAV951 induced ARDS compared to non-IAV controls (n=3-4). (H) TNFSF14 expression on 952 BALF leukocytes on day 3 pi and bone marrow-derived (BM) neutrophils from non953 infected mice, n=3-10, data pooled from two independent experiments. (I-J) qPCR 954 analysis in lung tissue, n=7-9 (I), and BALF ELISA, n=9-13 (J), for TNFSF14 955 expression on day 7 pi after neutrophil depletion. Data pooled from four and five 956 independent experiments, respectively. (K-L) BALF (K) and lung (L) TR-AMs on day 957 7 pi after neutrophil depletion (n=8, data pooled from four and five independent 958 experiments, respectively). Graphs represent means ± SEM. Significance was 959 determined by unpaired 2-tailed t-test, 1-way, or 2-way ANOVA; *p<0.05, **p<0.01, 960 ***p <0.001, ****p <0.0001. 961 962 50 963 Figure 7. Severity of post-influenza pneumococcal pneumonia is attenuated in 964 the absence of TNFSF14. (A-B) Survival (A) and weight loss (B) after IAV/Spn co965 infection of wt and Tnfsf14-/- mice (means ± SEM, n=17-18, data pooled from five 966 different experiments). (C-D) Bacterial burden in the BALF (C) and in the lungs (D) of 967 wt and Tnfsf14-/- mice 9 days after IAV infection and 48h after Spn infection (means ± 968 51 SEM, n=10, data pooled from eight independent experiments). (E) Total BALF TR-AM 969 numbers of wt and Tnfsf14-/- mice 9 days after IAV/ 48h after Spn infection (means ± 970 SEM, n=7-8, data pooled from five different experiments). (F-G) Survival (F) and weight 971 loss (G) following IAV/Spn co-infection and TNFSF14 blocking on day 2 pi (means ± 972 SEM, n=12-16, data pooled from five independent experiments). (H) Schematics of 973 experimental layout for IAV/Spn co-infection with adoptive transfer (a.t.) of naïve wt, 974 Tnfrsf14-/-, or Ltbr-/- TR-AMs on day 3 pi. (I) Survival of wt mice upon IAV/Spn co975 infection and TR-AM a.t. (means ± SEM, n=4-12, data pooled from five independent 976 experiments). (J) Bacterial load in lysed wt and Ltbr-/- naïve TR-AMs after ex vivo Spn 977 infection (means ± SEM, n=12, data pooled from four independent experiments). 978 Significance was determined by log-rank (Mantel-Cox) test, unpaired 2-tailed t-test, 1979 way and 2-way ANOVA with Tukey’s posthoc test; *p<0.05, **p<0.01. (K) Proposed 980 hypothesis: Severe IAV-induced pneumonia is characterized by massive leukocyte 981 recruitment, including neutrophils. Once in the alveoli, neutrophils start releasing 982 TNFSF14, which is sensed by TR-AMs through ligation to surfaced-expressed 983 receptors TNFRSF14 and LTbR, culminating in TR-AM death, which increases host 984 susceptibility to post-influenza pneumococcal pneumonia. AEC: alveolar epithelial 985 cells. 986 987 52 988 Supplemental Figure 1. Gating strategy for the identification of different immune 989 cell populations in murine BALF (A) and lung tissue (B) per flow cytometry 990 analysis. Representative plots following multicolor staining of BALF and lung-tissue 991 immune cells, as described in the ´Supplemental Methods´ document. Gating 992 53 strategies were set according to the appropriate isotype and fluorescence minus one 993 (FMO) controls. 994 995 54 996 Supplemental Figure 2. Immune cell kinetics in the lung tissue of IAV-infected 997 mice over the infection course. Following BALF extraction, lungs of wild-type mice 998 infected with 500ffu IAV were harvested at different time points. Immune cell 999 populations were identified per flow cytometry analysis, including sessile TR-AMs 1000 (n=6-11) (A), neutrophils (n=5-10) (B), BMDM (n=5-10) (C), T cells (n=4-10) (D), B 1001 55 cells (n=3-7) (E), CD11bresident DCs, (n=6-7) (F), eosinophils (n=6-7) (G), NK cells 1002 (n=5-10) (H), CD11b+ DCs (n=6-7) (I), pDCs (n=6-7) (J), IM (n=6-7) (K), NK1.1+ NKT 1003 cells (n=5-10) (L), and Ly6Cresident monocytes (n=6-7) (M). Data shown pooled from 1004 nine independent experiments, mean ± SEM is depicted. Significance was determined 1005 by 1-way ANOVA with Tukey’s posthoc test; *p<0.05, **p<0.01, ***p <0.001, 1006 ****p <0.0001. (p)DCs: (plasmacytoid) dendritic cells, IM: interstitial macrophages, NK: 1007 natural killer. 1008 1009 1010 56 1011 Supplemental Figure 3. TR-AM apoptosis can be attenuated through the use of 1012 a caspase inhibitor. (A) Gating strategy for the quantification of apoptotic TR-AMs in 1013 the BALF of IAV-infected mice on days 0, 3, and 7 pi. (B-C) TR-AM viability following 1014 treatment with 0.5μM staurosporine and different doses of a caspase-3 (B) or a 1015 caspase-8 (C) inhibitor, values depicted as % survival of control. DMSO-treated cells 1016 set at 100%. Graphs represent means ± SEM, n=4 per condition. Data pooled from 1017 three independent experiments. Significance was determined by 2-way ANOVA with 1018 Tukey’s posthoc test; *p<0.05. 1019 1020 63 1079 Supplemental Figure 7. Loss of TNFSF14 does not impact neutrophil influx, 1080 spleen bacterial burden, or TR-AM phagocytosis capacity after IAV/Spn co1081 infection. (A) Spn burden in the spleens of wt and Tnfsf14-/- mice 9 days after IAV 1082 infection and 48h after Spn infection (means ± SEM, n=10, data pooled from eight 1083 different experiments). (B) Neutrophils depicted as percentage of total live cells in the 1084 BALF of wt and Tnfsf14-/- mice at the same time point (means ± SEM, n=7-8, data 1085 pooled from five independent experiments). (C-D) Phagocytosis capacity of wt and 1086 Tnfsf14-/- day 7 TR-AMs depicted as FITC signal normalized to non-infected TR-AMs 1087 (C) and percentage of FITC+ TR-AMs (phagocytic cells, D), following ex vivo 1088 incubation with pHrodoTM Green Escherichia coli BioParticlesTM (means ± SEM, n=31089 4, data representative of three independent experiments). Significance was 1090 determined by unpaired 2-tailed t-test. 1091