www.nature.com/cddis ARTICLE OPEN Differential kinetics of splenic CD169+ macrophage death is one underlying cause of virus infection fate regulation Valentina Casella1, Eva Domenjo-Vila 1, Anna Esteve-Codina 2,3, Mireia Pedragosa1, Paula Cebollada Rica 1, Enric Vidal 4,5, ✉ ✉ Ivan de la Rubia3,6, Cristina López-Rodríguez7, Gennady Bocharov 8,9, Jordi Argilaguet1,4,5 and Andreas Meyerhans 1,10 © The Author(s) 2023 Acute infection and chronic infection are the two most common fates of pathogenic virus infections. While several factors that contribute to these fates are described, the critical control points and the mechanisms that underlie infection fate regulation are incompletely understood. Using the acute and chronic lymphocytic choriomeningitis virus (LCMV) infection model of mice, we find that the early dynamic pattern of the IFN-I response is a differentiating trait between both infection fates. Acute-infected mice generate a 2-wave IFN-I response while chronic-infected mice generate only a 1-wave response. The underlying cause is a temporal difference in CD8 T cell-mediated killing of splenic marginal zone CD169+ macrophages. It occurs later in acute infection and thus enables CD169+ marginal zone macrophages to produce the 2nd IFN-I wave. This is required for subsequent immune events including induction of inflammatory macrophages, generation of effector CD8+ T cells and virus clearance. Importantly, these 1234567890();,: benefits come at a cost for the host in the form of spleen fibrosis. Due to an earlier marginal zone destruction, these ordered immune events are deregulated in chronic infection. Our findings demonstrate the critical importance of kinetically well- coordinated sequential immune events for acute infection control and highlights that it may come at a cost for the host organism. Cell Death and Disease (2023)14:838 ; https://doi.org/10.1038/s41419-023-06374-y INTRODUCTION genetic make-up and immune competence that affect how a virus The fate of a virus infection is determined by the dynamic infection evolves [11, 12]. Conceptually, infection outcomes are interplay between the expanding virus and the concomitantly based on the two simple principles “numbers game” and “immune induced immune response. It can be categorized as “acute” or geography” [13]. These terms refer to the kinetics of virus growth “chronic” depending on the temporal virus-host relationship [1]. In and immune reactions, and the spatial organization of the immune humans, acute infections are commonly resolved within a few response in relation to viral entry sites and tissue tropism [14]. weeks. Once the infection is under control, tissue damage has to Together they determine cytokine gradients and immune cell be repaired and immune homeostasis re-established. In contrast, homing patterns across tissues that ultimately regulate infection chronic infections are not resolved and emerge when innate and fates and pathogenicity [15–17]. However, while both principles adaptive immune responses are not sufficient to eliminate the provide a rationale for a dynamic view on virus infection regulation, invading virus during the primary infection phase. In this case, the they do not specify when and how a particular infection fate is host immune response has to adapt to the continuous presence of mechanistically determined. Here we use the LCMV-Doc mouse the invading virus and downregulate immune effector mechan- model of acute and chronic virus infection to characterize the isms to avoid immunopathology. This is achieved by activating spatiotemporal events underlying both infection courses. We show suppressive activities including T cell exhaustion [2–5], the that the decisive event for the infection fates is the different generation of monocyte-derived suppressor cells (MDSCs) [6, 7] destruction kinetics of CD169+ marginal zone macrophages. These and regulatory cell subsets [8]. cells are key coordinators of adaptive immune responses. Their Virus and host factors contribute to infection fate decisions and presence is required during a specific time window and ensures the physiological infection outcomes. Virus factors include host cell efficient virus control in an acute infection. Together our results tropism, infection dose, virus immune evasion mechanisms and the highlight the critical importance of kinetic coordination of innate capacity to latently infect cells [9, 10]. From the host, it is age, and adaptive effector mechanisms for pathogen control. 1 Infection Biology Laboratory, Department of Medicine and Life Sciences (MELIS), Universitat Pompeu Fabra, 08003 Barcelona, Spain. 2CNAG-CRG, Centre for Genomic Regulation (CRG), Barcelona Institute of Science and Technology, 08028 Barcelona, Spain. 3Universitat Pompeu Fabra (UPF), 08003 Barcelona, Spain. 4Unitat mixta d’Investigació IRTA-UAB en Sanitat Animal. Centre de Recerca en Sanitat Animal (CReSA), Campus de la Universitat Autònoma de Barcelona (UAB), 08193 Bellaterra, Catalonia, Spain. 5IRTA, Programa de Sanitat Animal, Centre de Recerca en Sanitat Animal (CReSA), Campus de la Universitat Autònoma de Barcelona (UAB), 08193 Bellaterra, Catalonia, Spain. 6EMBL Australia Partner Laboratory Network at the Australian National University, Acton, Canberra, ACT 2601, Australia. 7Immunology Unit, Department of Medicine and Life Sciences (MELIS), Universitat Pompeu Fabra, 08003 Barcelona, Spain. 8Marchuk Institute of Numerical Mathematics, Russian Academy of Sciences, 119333 Moscow, Russia. 9Sechenov First Moscow State Medical University, 119991 Moscow, Russia. 10Institució Catalana de Recerca i Estudis Avançats (ICREA), 08010 Barcelona, Spain. ✉email: jordi.argilaguet@irta.cat; andreas.meyerhans@upf.edu Edited by Professor Yufang Shi Received: 29 August 2023 Revised: 29 November 2023 Accepted: 5 December 2023 Official journal of CDDpress V. Casella et al. 2 RESULTS differentiating trait between both infection fates has not been Differential kinetics of type I IFN genes during acute and described. To validate the second wave of IFN-I at the transcript chronic LCMV infections and protein level, we quantified the transcripts of Ifnb1 (Fig. 1d) To identify early differentiating traits of acute and chronic virus and of the IRG Mx1 (Supplementary Fig. S3b) in spleens from infections, we used the well-established lymphocytic choriome- additional infected mice by quantitative PCR (qPCR) at days 3, 4 ningitis virus (LCMV) mice infection model. It is the “gold- and 5 p.i., and the IFNB1 protein in serum every 12 h from day 0 to standard” for the analysis of interactions between a non- day 6 post acute and post chronic infection by an IFNB ELISA (Fig. cytopathic virus with its host organism and enables the establish- 1e). Overall, transcript and protein data are in good agreement. ment of acute and chronic infections in two ways [18]: (i) the use Splenic transcript and serum IFNB1 protein levels peak at day 2 of high dose infections with the virus strains LCMV-Arm and and day 1 in acute and chronic infection, respectively. Peak IFNB1 LCMV-Cl13 that differ in their replication rate and target cell levels are about 15-fold higher in chronic infection, similar to what infectivity [19], and (ii) low or high dose infection with LCMV-Doc has been previously described [25]. While IFNB1 then continuously [20]. Since we were interested in differentiating traits of the host declines in chronic infection, a second IFNB1 wave is observed in responses, we chose the latter option that excludes differences in acute infection from days 3.5 to 6 (Fig. 1e). Since these early viral properties. C57BL/6J mice were infected with a low-dose differences in the IFN-I response may contribute to the regulation (2 ×102 plaque-forming units (PFU)) or a high-dose (2 ×106 PFU) of of the infection fate, we subsequently focused on the character- LCMV-Doc to establish an acute or chronic infection, respectively. ization of the second wave of IFN-I and its downstream effects. A detailed description of virus kinetics and T cell responses during both infections has been given previously [21, 22]. To identify CD169+ marginal zone macrophages produce the second early global differences between both infection fates, time- wave of IFN-I expression during acute LCMV infection resolved splenic transcriptomes were determined by RNAseq Acute and chronic LCMV infections lead to an early depletion of every 24 h from day 0 to day 7, and at days 9 and 31 post-infection pDCs and a shift of the main IFN-I-producing cell types from pDCs (p.i.). We previously used a similar approach to determine to conventional dendritic cells (cDCs) and macrophages including differences in late host responses [21]. Taking non-infected mice CD169+ splenic marginal zone macrophages [24, 26, 27]. To as a reference point, 15919 differentially expressed genes determine the source of the second wave of IFN-I during acute between acute and chronic infections were identified (Fig. 1a, LCMV infection, we analyzed the expression of Ifnb1 in different left panel). To narrow down the analyses to early antiviral innate cell subsets by qPCR at day 5 p.i. (Fig. 2a). Ifnb1 expression was immune responses, the differentially expressed genes that were significantly upregulated in macrophages but not cDCs in acute regulated by type I interferons (IFN-I) were selected through the infected mice. Only basal levels of Ifnb1 were observed in interferome database (http://www.interferome.org). This led to macrophages from chronic infected mice (Fig. 2b). Since the IFN-I 2825 differentially expressed interferon-regulated genes (IRGs; Fig. response in antigen-presenting cells including macrophages has 1a, right panel). The number of differentially expressed total genes to be balanced by negative feedback regulation for enforced virus and IRGs during chronic infection remained roughly constant over replication and efficient activation of the adaptive immune time, reaching maximum levels already at d1 p.i. In contrast, response [28], we tested for the presence of IFN-I-induced antiviral transcriptomic changes in acute LCMV-infected mice showed a Mx1 and the IFN-I receptor signaling suppressor Usp18. Acute biphasic dynamics characterized by an initial peak at d2 p.i., infected mice showed elevated levels of Mx1 and a concomitant followed by a gradual increase reaching the same level as in upregulation of Usp18 in macrophages (Fig. 2c). Conversely, chronic infection by d7 p.i. (Fig. 1a). To further characterize the macrophages from chronically infected mice showed only basal IFN-I response behavior in acute and chronic infections, a Mx1 or Usp18 expression (Fig. 2c). Thus, splenic macrophages of correlation matrix of the differentially expressed IRGs was acute infected mice have a balanced IFN-I signature. To further generated (Fig. 1b). IRGs showed well-defined clusters of co- define the macrophage subset that produces the second IFN-I expression during acute infection while displayed a more wave, splenic CD169+ marginal zone macrophages from acute heterogeneous pattern during chronic infection, indicating a and chronic infected mice were enriched by magnetic beads and differential regulation of IFN-I responses during both infection sorted by flow cytometry. Levels of Ifnb1, Mx1 and Usp18 fates. transcripts were higher in cells from acute infected mice than in To explore splenic transcriptome dynamics in acute and chronic cells from chronic infected mice (Fig. 2d and Supplementary Fig. infections, two signed coexpression networks were constructed S4a), suggesting that CD169+ macrophages are the source of the from all differentially expressed genes via weighted gene second IFN-I wave. To further confirm this, acute infected CD169- correlation network analysis (WGCNA) [23]. This methodology DTR transgenic mice were treated with diphtheria toxin (DT) at allows to cluster genes with similar patterns of expression and day 3 p.i. to deplete CD169+ macrophages (Fig. 2e, f and enables the definition of modules of highly coexpressed genes Supplementary Fig. S4f). This significantly decreased the (Supplementary Fig. S1a) [23]. Twenty and twenty-one modules IFNB1 serum levels at day 5 p.i. (Fig. 2g). Together this from acute and chronic infections were identified, respectively, demonstrates that CD169+ macrophages are the main producers and color-coded for identification purposes (Supplementary Fig. of the second IFN-I wave in acute LCMV infection. S1b). Two modules, the acute-brown module of acute infection and chronic-brown of chronic infection, were enriched in the gene Early CD169+ marginal zone macrophage depletion in ontology (GO) term “cellular response to interferon-beta” (Sup- chronic infection is mediated by CD8+ T cells plementary Fig. S2). These two modules contained Ifnb1 and Splenic marginal zone cells are known to be depleted during an seven Ifna genes (Fig. 1c) together with several IRGs (Supplemen- acute LCMV infection [29]. Since the accumulated virus load at tary Fig. S3a). Importantly, the module eigengenes (Fig. 1c, brown early time points during a chronic infection is increased over an lines) and the individual IFN-I genes (Fig. 1c, black lines) showed acute infection [21], we hypothesized that this may translate to an different expression kinetics in acute and chronic infections. Acute earlier elimination of IFN-I-producing CD169+ marginal zone infected mice showed two IFN-I gene expression peaks at d2 and macrophages. To test this, LCMV-infected CD169+ macrophages d5 p.i., while chronically-infected mice had only a single peak at and total CD169+ macrophages from spleens of acute and d1 p.i. chronic infected mice at days 3 and 5 p.i. were quantified by flow The first wave of IFN-I expression in acute and chronic infection cytometry (Fig. 3a–e). Frequencies of LCMV-NP-positive CD169+ is known to be produced from plasmacytoid dendritic cells (pDCs) macrophages increased to about 30% in acute infected mice and [24], however the second IFN-I expression wave as a to about 90% in chronic infected mice from d3 to d5 p.i., Cell Death and Disease (2023)14:838 V. Casella et al. 3 Fig. 1 Differential type I IFN responses in acute and chronic LCMV infections. a Numbers of upregulated and downregulated total and interferon-regulated genes that were differentially expressed in spleens at the indicated time points after acute and chronic LCMV infections. The transcriptome of naive mice is used as reference. b Pearson’s pairwise correlation matrix showing IRGs expression patterns in acute and chronic infections. The color scale indicates the degree of correlation. c Eigengene expression profiles from acute-brown and chronic-brown modules are represented (brown lines) together with expression kinetics of IFN-I genes contained in the same modules (black lines). d qPCR of Ifnb1 from spleens of acute- (black) and chronic-(red)infected mice at the indicated time points post-infection. The relative gene expression level was normalized to Gapdh. e ELISA of IFNB1 from serum of acute- and chronic-infected mice at the indicated time points post-infection. For each group and time point the mean ± SEM is shown. *p ≤ 0.05; unpaired two-tailed t-test. Cell Death and Disease (2023)14:838 V. Casella et al. 4 Fig. 2 Analysis of type I IFN response at day 5 post- acute and -chronic LCMV infections. a Schematic representation of LCMV infections and cell sorting. b Ifnb1 expression levels measured by qPCR using RNA extracted from sorted macrophages (macros) and conventional dendritic cells (cDCs) of naive or infected mice. c qPCR of Mx1 and Usp18 from macrophages sorted from acute- and chronic-infected mice. d Ifnb1 expression levels in splenic CD169+ macrophages obtained at day 5 p.i. To obtain sufficient number of cells mice were previously injected in vivo with a biotin-labeled anti-CD169 antibody, and CD169+ macrophages cells were isolated ex vivo by magnetic separation using anti-biotin microbeads, and further purified by sorting after staining with Streptavidin-Phycoerythrin. The relative gene expression levels were normalized to Gapdh. e Schematic representation of diphtheria toxin (DT) treatment in acutely LCMV-infected CD169DTR mice. Flow cytometry quantification of splenic CD169+ macrophages (f) and ELISA quantification of IFNB1 in serum (g) from CD169DTR mice untreated (DT−) or treated (DT+) with diphtheria toxin (DT), at day 5 post-infection. For each group the mean ± SEM is shown; significance is determined via one-way ANOVA or unpaired two-tailed t-test; ns not significant; *p ≤ 0.05; **p ≤ 0.01; ****p ≤ 0.0001. respectively (Fig. 3b). The higher infection frequency was directly Moreover, T cells were increased after NK cell depletion linked to the loss of this cell population in chronic infection (Supplementary Fig. S4e), confirming earlier observations and (Fig. 3c), regardless of the size of the spleen that was increased at the role of NK cells in limiting CD8+ T cell-mediated lymphatic this time point (splenomegaly; unpublished observation). Next, to tissue damage [33]. assess the cause of CD169+ macrophage loss, we depleted CD8+ To visualize LCMV distribution, splenic CD169+ marginal zone cytotoxic T cells or NK cells which are known to be the two main macrophage destruction within the lymphatic tissue, and their cell types killing LCMV-infected cells in vivo [30, 31] and which rescue by CD8+ T cell depletion, spleens of acute and chronic appear earlier in chronic LCMV infection than in acute infection infected mice with and without anti-CD8α antibody treatment [2, 32]. For this, we treated mice with specific antibodies directed were analyzed by fluorescence microscopy at days 3, 5 and 8 (Fig. against NK cells or CD8+ T cells as outlined in Fig. 3a. Anti-CD8α 3h, i). LCMV was detected in the tissue sections with the NP- antibody but not anti-NK1.1 antibody treatment prevented the specific VL4 antibody. At d3 p.i., LCMV was still confined to specific loss of CD169+ marginal zone macrophages (Fig. 3f, g), loci in acute infection while it was already spread throughout the demonstrating that CD8+ T cells are the culprit. In contrast, marginal zone in chronic infection (Fig. 3h). This increased early anti-NK1.1 antibody treatment further reduced the absolute LCMV dissemination in chronic infection was reflected in an earlier number of CD169+ macrophages suggesting that NK cells did destruction of CD169+ splenic marginal zone macrophages which not play a significant role in macrophage killing (Fig. 3g). were already absent at d3 while still present at d3 and d5 but Cell Death and Disease (2023)14:838 V. Casella et al. 5 absent at d8 in acute infection (Fig. 3i). Depletion of the CD169+ The second wave of type I IFN in acute infected mice is macrophages was inhibited by anti-CD8α antibody treatment (Fig. required to induce inflammatory monocytes/macrophages 3i) demonstrating the importance of CD8+ T cells in CD169+ and virus-specific CD8 T cells macrophage elimination. Thus, the splenic architecture of acute Co-expressed genes within WGCNA-derived modules belong to and chronic infected mice is disrupted with a different kinetics. functionally related biological processes [34]. To determine the Cell Death and Disease (2023)14:838 V. Casella et al. 6 Fig. 3 Temporal changes of the splenic marginal zone in acute versus chronic LCMV infection. a Schematic representation of timepoints of spleen sampling in acute and chronic LCMV infected mice and of anti-CD8α or anti-NK1.1 treatments in chronically LCMV-infected C57BL/6J mice. Frequencies of splenic LCMV-infected (VL4+) CD169+ macrophages (b) and of total splenic CD169+ macrophages (c) obtained by flow cytometry at days 3 and 5 post-infection from acute and chronic LCMV-infected mice. Absolute numbers of LCMV-infected (VL4+) CD169+ macrophages (d) and of total CD169+ macrophages (e) from days 3 and 5 post acute and chronic LCMV-infected mice. Frequencies (f) and absolute numbers (g) of splenic CD169+ macrophages at day 5 post-chronic LCMV infection in untreated and antibody-treated (anti-CD8α or anti-NK1.1) mice. For each group and time-point the mean ± SEM is shown; significance is determined via unpaired two-tailed t-test; *p ≤ 0.05; **p ≤ 0.01; ****p ≤ 0.0001. Immunofluorescence of spleen sections at the indicated time points after LCMV-infection from untreated or anti- CD8α antibody (aCD8) treated mice. Spleen sections were stained for VL4 (LCMV NP; red) and CD3 (T cells; blue) (h) and for CD169 (CD169+ Macrophages; green), CD3 (T cells; blue) and B220 (B cells; red) (i). One representative section of 3 mice per group is shown. Images were taken at 20X magnification, the scale bar indicates 300 µm. functional role of the second IFN-I wave in acute infection, we infected C57BL/6J WT mice and depleted CD169+ macrophages therefore analyzed the genes that are coexpressed with the IFN-I in acute infected CD169DTR mice at days 3–5 p.i. (Fig. 5a). Both genes of the acute-brown module. Interestingly, there were treatments resulted in (i) decreased functionality (Fig. 5b, c) and several genes associated with inflammation that were not numbers of LCMV-specific CD8 T cells (Fig. 5e, f) at days 9 and 15 coexpressed with IFN-I genes of the chronic-brown module p.i., and (ii) increased virus titers (Fig. 5d). Importantly, mice during chronic infection (Fig. 4a). This suggested that the second lacking either IFNAR signaling of the second IFN-I wave or IFN-I wave is specifically linked to an inflammatory response that CD169+ macrophages, became chronically infected with LCMV as occurs during acute infection. Indeed, our previous work shown by viremia beyond d15 (Fig. 5d) and signs of CD8+ T cell demonstrated that monocytes/macrophages from day 6 post- exhaustion as evidenced by the increase of PD1+Tim3+ cells (Fig. acute LCMV infection but not from chronic infection displayed a 5g, h). Together these data highlight the impact of the CD169+ pro-inflammatory profile [21]. In line with this, flow cytometry macrophage destruction kinetics on immune response regulation. analysis (see Supplementary Fig. S5) demonstrated the induction In acute infection, the delayed CD8 T cell-mediated killing of of IL1β-producing macrophages at day 6 after acute LCMV CD169+ marginal zone macrophages is instrumental for the infection that was significantly reduced in chronic infection (Fig. induction of a second IFN-I wave which coordinates an optimally- 4b, d). This difference in macrophage phenotypes can be timed sequence of innate and adaptive immune events. In chronic visualized by t-Distributed Stochastic Neighbour Embedding (t- infection, the early CD8 T cell-mediated killing of CD169+ SNE) analysis that reveals IL1β+ expression as a differentiating marginal zone macrophages eliminates the primary source of feature between acute and chronic infection (Fig. 4c). To the second IFN-I wave and thus is a key regulator for this infection investigate whether the second wave of IFN-I expression is outcome. required for the induction of this pro-inflammatory response, we abrogated IFN-I signaling by anti-type I IFN receptor (aIFNAR) Early IFN-I kinetics determines the appearance of fibrosis in treatment in WT C57BL/6J mice or depleted IFN-I-producing lymphatic tissue CD169+ macrophages by diphtheria toxin (DT) treatment in Lymphoid tissue fibrosis is a typical feature of chronic infections CD169-DTR transgenic mice [35], respectively (Fig. 4d). Both such as HIV and is intimately linked to chronic immune activation treatments resulted in a significant reduction of IL1β-producing and inflammation within the affected tissues [37, 38]. In our macrophages at day 6 p.i. (Fig. 4e). previous study, we showed that also during acute infection, To further characterize the spectrum of cytokines, beyond IL1β, lymphoid tissue fibrosis occurs and persists even when the virus is affected by the second wave of IFN-I signaling, mRNA from spleen already well-controlled [21]. Notably, acute LCMV infection of acute LCMV-infected mice with or without IFNAR blockage and induces a strong cytotoxic T-lymphocyte (CTL) response [39] of chronic LCMV-infected mice at day 6 p.i. were isolated and which results in varying degrees of immunopathology during CTL- transcriptomes determined by RNAseq. DE genes from the mediated killing of virus-infected cells [40]. However, no previous transcriptomes of aIFNAR-treated and untreated acute infected work ever characterized the histopathological costs of a well- mice were enriched in the GO terms “monocyte differentiation”, coordinated immune response against LCMV. This prompted us to “chemokine-mediated signaling pathway”, “chemotaxis”, and investigate the role of the IFN-I-mediated CTL response in the “inflammatory response”, thus highlighting the role of the second development of fibrosis during acute LCMV infection. For this we IFN-I wave in modulating the inflammatory environment (Fig. 4f). either blocked IFNAR signaling by anti-IFNAR antibodies or This becomes apparent in the heatmap of the most differentially depleted CD8+ T cells at the peak of the CTL response [21] as expressed genes taken from the GO terms above (Fig. 4g). Genes depicted in Fig. 6a. Importantly, blockage of signaling of the related to chemokine receptors and proinflammatory cytokines/ second IFN-I wave entirely prevented spleen fibrosis at d15 p.i. chemokines are upregulated in acute infected mice, while (Fig. 6b, c). Similarly, anti-CD8 antibody treatment strongly downregulated in aIFNAR-treated mice. Anti-inflammatory genes reduced the occurrence of fibrosis at d30 p.i. These findings instead follow the opposite trend, they are downregulated in indicate that the generation of fibrotic tissue in spleens from acute acute-infected mice and upregulated in aIFNAR-treated mice. infected mice is a consequence of the IFN-I-dependent antiviral Finally, aIFNAR-treated mice show a transcription pattern that is CD8 T cell response that is required to resolve the infection. Taken closer to chronic infected mice than to acute infected mice. together, our findings demonstrate that the significant immune- Together this demonstrates that the second IFN-I wave produced related tissue damage observed in mice after acute LCMV by CD169+ marginal zone macrophages is required for the infection is the cost for effective virus control. inflammatory macrophage response at d6 post acute infection, and lack of this, due to prior CD169+ macrophage destruction, results in the significantly reduced inflammatory response in DISCUSSION chronic infection. How early virus-host interaction events and their dynamics IFNAR signaling in CD8 T cells is critical for the generation of influence virus infection fates remains elusive. Here we demon- effector and memory cells [36]. To investigate the influence of IFN- strate that acute and chronic LCMV infection outcomes are the I produced by CD169+ marginal zone macrophages on the virus- result of a distinct kinetics of lymphatic tissue alterations that specific CD8 T cell response, we blocked IFNAR signaling in acute manifest themselves as differences in type I IFN responses. We Cell Death and Disease (2023)14:838 V. Casella et al. 7 unmask the spatio-temporal immune events that cause these coordinated sequence of innate and adaptive immune events for differences and link the changes in the IFN-I response to virus infection control. subsequent characteristics of the different infection outcomes. Our systems-based approach to analyze acute and chronic Together, this illustrates the importance of a temporally well- LCMV infection fates in mice enabled mechanistic and Cell Death and Disease (2023)14:838 V. Casella et al. 8 Fig. 4 Characterization of the type I IFN-dependent inflammatory macrophage response in acute LCMV infection. a RNAseq-derived expression kinetics of genes related to pro-inflammatory macrophages in spleens from acute (black lines) and chronic (red lines) infected mice. b Representative flow cytometry plots showing IL1β-producing Ly6Chi macrophages. c T-distributed stochastic neighbor-embedding (tSNE) representation of macrophages from naïve, acute and chronic infected mice at day 6 post-infection. Distribution and relative expression of IL1β is shown in the merged plot. d Schematic representation of anti-IFNAR antibody and diphtheria toxin (DT) treatments in acutely LCMV- infected WT and CD169DTR mice, respectively. e Percentages of IL1β-producing macrophages in the spleen at d0 (naïve) or d6 post-acute or chronic infection in untreated (aIFNAR−/DT−) or treated (aIFNAR+/DT+) mice. For each group, the mean of n = 6 to 13 mice, representative of two independent experiments, is shown. ****p ≤ 0.0001; unpaired two-tailed t-test. f Enriched GO terms (obtained from DAVID) on genes differentially expressed in spleens of aIFNAR-treated versus untreated mice at day 6 post-acute LCMV infection. g Heatmap illustrating the relative expression of genes in spleens from acute, acute aIFNAR-treated and chronic-infected mice at day 6 post-infection. conceptional advances that go beyond current knowledge. First, albeit at later times. Indeed, this differential kinetics of spleen while IFN-I is a well-established major player in the early antiviral disruption perfectly fits to the described earlier appearance and response against virus infection [41], we show here that acute and expansion of CTLs in chronic LCMV Doc-infected mice [2]. chronic LCMV infections fundamentally differ in their IFN-I Together this indicates that the decision point for the infection responses: IFN-I genes are expressed in two waves in acute fate was entirely determined by a kinetic feature of the virus- infection but have only one wave in chronic infection. Second, induced immune response. while it is known that CD169+ macrophages contribute to IFN-I CD169+ marginal zone macrophages play a central role in production during LCMV infection [27], we show here that the IFN- coordinating the early immune events required to control a virus I expression by splenic CD169+ macrophages during acute infection. They are crucial for the filtering and retention of infection corresponds to the 2nd IFN-I wave and has a polyfunc- pathogens from blood [49], participate in the distribution and tional role in bridging innate and adaptive immunity. Third, from presentation of antigens to B cells [50], collaborate with cross- the pioneer work of the Zinkernagel group, it has been widely presenting DCs in the activation of CD8 T cells [51] and produce accepted that CD8 T cells contribute to spleen architecture IFN-I when productively infected with viruses such as Herpes disruption during LCMV infections [42]. Here we describe that CD8 Simplex Virus or LCMV [27, 52]. The prompt disappearance of T cell-mediated killing of CD169+ macrophages in chronic CD169+ macrophages and subsequent lack of IFN-I production infection dampens the production of the second wave of IFN-I. observed in chronic LCMV-infected mice represents the loss of a This results in the appearance of chronic infection-specific critical component of the coordinated antiviral immune response. immune features such as the lack of inflammatory macrophages Indeed, the disruption of this regulatory mechanism during acute and the generation of exhausted T cells. Forth, we previously infection through IFNAR blockage or the early depletion of showed that LCMV infection can induce lymphatic tissue fibrosis CD169+ macrophages in CD169DTR mice resulted in the [21]. Fibrosis is a common feature also in human infections like induction of exhausted T cells and the subsequent development those with HIV-1 and is associated with a drop in immune of a chronic infection. Together, our results further highlight the functioning [43]. Here we demonstrate that spleen fibrosis in essential role of marginal zone macrophages in coordinating LCMV acute-infected mice is a consequence of the IFN-I- innate and adaptive immune responses for efficient virus control. dependent antiviral CD8 T cell response that is required to resolve We have previously shown that acutely LCMV-infected mice the infection. Thus, fibrosis is ‘the price’ the host has to pay for develop spleen fibrosis which is maintained even when the virus is effective virus control. already well controlled [21]. Here we demonstrate that these The crosstalk of viruses with the IFN-I response of the host histological changes are a consequence of the killing of infected organism can determine infection outcomes. Many viruses encode cells by cytotoxic CD8 T cells, which in turn are induced by the IFN antagonists that suppress or delay IFN-I expression and second wave of IFN-I production by CD169+ macrophages. Thus, provide viruses with a kinetic advantage over host-mediated lymphoid tissue fibrosis is ‘the price’ the host has to pay for the restriction mechanisms [44, 45]. In case virus expansion and successful clearance of the virus. However, to what extent fibrosis restriction mechanisms become very unbalanced, virus infection induced after acute virus infection affects organ functionality and may be highly pathogenic. Remarkable examples of this are therefore the capacity of the immune system to properly respond infections with the 1997 Hongkong Flu strain that carried a potent to subsequent challenges has still to be determined. Previous IFN antagonistic NS1 protein [46] or infections of host organisms studies have shown that fibrosis footprints in lymphatic tissues that have weakened IFN-I responses due to autoantibodies [47] or derived from infections are linked to a significant impairment of genome variation within IFN-inducible restriction factors [48]. responses towards vaccines [53] and subsequent other infections However, in our model system of acute and chronic LCMV [54]. Moreover, lymphatic tissue fibrosis in chronic HIV and SIV infections, the same LCMV strain was used and thus, different infections impairs immune cell interactions [43, 55–57]. None- antagonistic viral protein functions were not operative. Rather it theless, fibrosis during acute or chronic virus infections develops was the system’s intrinsic kinetics in the destruction of infected in the distinct physiological contexts of either fully functional or marginal zone CD169+ macrophages by CD8 T cells that caused a immune-compromised environments, respectively, and thus may differential IFN-I expression. These cells were the dominant source differ in their immune modulating properties. of the second IFN-I wave at day 5 in low-dose acute infection, In conclusion, our systems immunology-based analysis of acute while in high-dose chronic infection CD169+ macrophages were and chronic LCMV infections identifies a missing regulatory already disappearing by day 3. Thus, the 2-wave behavior of IFN-I element in the coordination of systemic virus-host interactions: during acute infection is a reflection of IFN-I production mediated the induction of a second IFN-I peak in acute infection and its lack by different cell subsets at different timepoints, first from pDCs, in chronic infection. This helps to better appreciate the complexity then from cDCs and macrophages [27]. Hence, the subsequent of the tightly regulated IFN-I responses early during a virus IFN-I-dependent immune components characterized in acute infection, and reveals several IFN-I-dependent immune events that infection, namely proinflammatory macrophages and effector influence infection fates. Moreover, the systemic approach used CD8 T cells, did not arise in chronic infection [21]. Importantly, the illustrates the importance of analyzing immune responses as a loss of the splenic marginal zone was not a specific event during whole, considering the relationship among all its components. chronic infection, since it also happened in acutely infected mice Only by integrating “numbers game” and “immune geography”, Cell Death and Disease (2023)14:838 V. Casella et al. 9 Fig. 5 Characterization of type I IFN-dependent CTLs response in acute LCMV infection. a Schematic representation of anti-IFNAR antibody and diphtheria toxin (DT) treatments in acutely LCMV-infected WT and CD169DTR mice, respectively. b Representative flow cytometry plots showing gp33-responsive IFNγ-producing CD8+ T cells. Frequency of gp33-responsive IFNγ-producing CD8+ T cells (c) and viral loads (d) in WT mice untreated (aIFNAR−) or treated (aIFNAR+) with anti-IFNAR antibody, and in CD169DTR mice untreated (DT−) or treated (DT+) with diphtheria toxin (DT) at days 9 and 15 post-infection. Frequencies (e) and absolute numbers (f) of gp33-tet+ CD8+ T cells after anti-IFNAR antibody treatment at day 9 post-infection. g Frequency of Tim3+PD1+ exhausted CD8+ T cells in WT mice after anti-IFNAR antibody treatment, and in CD169DTR mice after DT treatment at day 9 post-infection. h tSNE representation of CD8 T cells from untreated (aIFNAR−) and treated (aIFNAR+) infected mice at day 9 post-infection. Distribution and relative expression of selected markers in T cells are shown. For each group, the mean of n = 4 to 8 mice, is shown. *p ≤ 0.05; **p ≤ 0.01; ***p ≤ 0.001; ****p ≤ 0.0001; unpaired two-tailed t-test. Cell Death and Disease (2023)14:838 V. Casella et al. 10 Fig. 6 The resolution of LCMV acute infection is associated to CTL-mediated spleen fibrosis. a Schematic representation of anti-IFNAR (aIFNAR) and anti-CD8α (aCD8) antibody treatments in acutely LCMV-infected WT mice. Spleen fibrosis score (b) and representative images of Masson’s Trichrome Stained spleen sections (c) used to score the fibrosis in each group on days 15 and 30 post-infection; the black arrows indicate deposits of collagen fibers (in blue – Masson trichrome stain) characteristic of fibrosis. Data shown are the mean of n = 6 to 9 mice. ****p ≤ 0.0001; unpaired two-tailed t-test. i.e. the kinetics of virus replication and immune responses at (clone 2.43; Bio X Cell) respectively at days -1 and 2 p.i. and at days 7 and spatial resolution, will we fully understand virus-host battles and 10 p.i. For depletion of NK cells, chronic LCMV-infected mice were i.p. their outcomes [14]. Finally, our study not only provides a unique injected with 75 μg of anti-NK1.1 antibody (clone PK136; Bio X Cell) at days IRG transcriptome data set during acute and chronic virus -3 and -2 p.i. infections but also suggests directions for immunotherapeutic strategies to avoid chronicity. An inspection of the results in Flow cytometry and sorting human infections is clearly indicated. Spleens were processed into a single-cell suspension and stained as previously described [21]. For staining and sorting of CD169+ macro- phages, spleens were previously enzymatically digested with a mix of Liberase-DNAse (0.16 mg/ml; 0.2 mg/ml) (Roche) for 30 min. Splenocytes METHODS were stained for viability with either Live/Dead Fixable Violet Cell Stain Mice (ThermoFisher Scientific) or Fixable Viability Stain 620 (BD biosciences) or C57BL/6J mice (RRID:IMSR_JAX:000664) were purchased from Charles River Fixable Viability Stain 780 (BD biosciences). After washing, cells were Laboratories and CD169-DTR transgenic mice, generated in the RIKEN BRC incubated with Fc block (BD Biosciences) to block non-antigen-specific laboratory (Dr. Kenji Kohno and Dr. Masato Tanaka) through the National binding of immunoglobulins to Fc receptors and then stained with BioResource Project of the MEXT/AMED (Japan), were a kind gift from fluorochrome-labeled antibodies targeting surface or intracellular proteins. Andres Hidalgo (Madrid, Spain). All mice were bred and maintained under For gp33-tetramer stainings, biotinylated MHC class I monomer (MBL specific pathogen-free conditions at in-house facilities. The number of International) was tetramerized by adding PE-conjugated Streptavidin over animals for each experiment was determined based on previous 10-time intervals and then added together with extracellular antibodies. experience with the model system. For detection of IFNɣ-producing T cells, intracellular cytokine staining was performed after 5 h ex vivo stimulation with GP33 (1 μg/ml) in the presence of Brefeldin A (BFA, Sigma). To detect IL-1b-producing macrophages, Virus infections spleens were harvested in media containing BFA and antibody staining LCMV strain Docile (LCMVDoc) was used for mouse infections. The virus was was performed without additional stimulation. After surface antibody grown, stored and quantified according to previously published methods staining, cells were fixed with 2% Formaldehyde, permeabilized in Perm/ [21]. Mice were infected by intraperitoneal (i.p.) injection of either 2 ×102 or Wash buffer (PBS 1% FCS, NaN3 0,1%, Saponin 0,1%) and then stained with 2 ×106 plaque-forming units (pfu) of LCMVDOC to respectively induce an intracellular antibodies for IFNɣ, IL-1b and VL-4. Flow cytometry samples acute or a chronic infection. were acquired on a SP6800 Spectral (Sony) or an Aurora (Cytek) analyzer. FACS data were analyzed using FlowJo 10.1 software (Tree Star Inc). In vivo treatments with antibodies FACSAria II Sorter (BD Biosciences) was used to sort cells in RLT buffer To block the IFN-I signaling at the desired timepoint, C57BL/6J mice were (Qiagen) with sorting purity >95%. intraperitoneally (i.p.) inoculated with 500 µg of anti-IFNAR (aIFNAR) antibody (clone MAR1-5A3; Leinco Technologies) or 500 µg of a mouse Magnetic-activated cell sorting IgG1 isotype control (clone MOPC21; Bio X Cell) on days 3 and 4.5 post- To sort splenic CD169+ macrophages, mice were injected with a biotin- infection. In order to block the proinflammatory response at day 6 p.i. anti-CD169 antibody (36 μg; Abcam) and 10 min later, spleens were harvested and digested with a Liberase-DNAse mix. CD169+ cells were In vivo cell depletion stained with anti-biotin microbeads (Miltenyi) for 30 min and purified by In order to deplete CD169+ macrophages at day 5 p.i., diphtheria toxin magnetic-activated cell sorting using LS columns (Miltenyi). CD169+ cells (DT; Sigma-Aldrich) was i.p. injected into CD169DTR transgenic mice were additionally sorted by flow cytometry after staining with streptavidin- (30 μg/kg) at day 3 post-acute infection. Physiological serum was used as a phycoerythrin (Thermo Fisher) and CD11b. Sorted cells were collected in control vehicle. CD8 T cell depletion in acute and chronic LCMV infection RLT buffer (Qiagen) and kept on ice. Sorting purity was >95% for all was performed by i.p. administration of 200 μg of anti-CD8α antibody populations. Cell Death and Disease (2023)14:838 V. Casella et al. 11 Histological staining sequenced (single strand, 50 nts) on an Illumina HiSeq2000 sequencer to Spleen samples were fixed overnight with 4% buffered formaldehyde and obtain 50–60 million reads per sample. Bioinformatic analysis, carried out paraffin-embedded. For semi quantification of fibrosis, Masson’s Trichrome according to previously published methods [21]. The total number of was performed on 3 µm thick spleen cuts that revealed collagen fibers in mapped genes used to perform further analysis was similar at all time blue. A set of values was defined from 0 to 10: a score of 0 represents points. normality; a score of 10 would correspond to a spleen in which the parenchyma has totally been replaced by connective tissue. Quantification and statistical analysis Graphs were compiled and statistical analyses were performed by Prism Immunofluorescence (GraphPad software). Statistical significance was evaluated with the Freshly removed spleens were fixed for 24 h in Cytofix buffer (BD), and unpaired t-test when comparing two groups and one-way ANOVA multiple dehydrated for 12 h in 30% sucrose at 4 °C. Next, samples were OCT- comparison test when comparing more than two groups. Non-significant embedded (Tissue-TEK) in cryomolds and stored at −80 °C before differences were indicated as “ns”. P-values below 0.05 were considered proceeding with sectioning. Serial tissue sections of 10 µm in thickness significant and were indicated by asterisks: *p < 0.05; **p < 0.01; were obtained using Leica Cryostat and mounted on Superfrost Plus slides ***p < 0.001; ****p < 0.0001. (Fisher Scientific) and frozen at −80 °C. Spleen sections were rehydrated in 0.1 M Tris buffer (1 M Tris powder in deionized water), blocked for 2 h in standard blocking buffer (1% Bovine serum albumin, 0.3% Triton 100X in Reporting summary Further information on research design is available in the Nature Research 0.1 M Tris buffer; Sigma-Aldrich) and incubated overnight at 4 °C with Reporting Summary linked to this article. primary and/or conjugated antibodies. CD169+ splenic marginal zone macrophages were visualized by staining with CD169 AF647 (MOMA-1; clone 3D6.112; Biolegend). B cells and T cells were identified using antibodies against B220 (rat anti-mouse clone RA3-6B2; Biolegend) and DATA AVAILABILITY CD3 (hamster anti-mouse clone 145-2C11; Biolegend). For unconjugated All the data and information concerning this study will be made available upon primary antibodies, species-specific secondary antibody coupled to Alexa request. Fluor 488 (goat anti-hamster; Biolegend), or Alexa Fluor 555 (goat anti-rat; Thermofisher) fluorochromes were incubated for 8 h at 4 °C. Sections were washed, DAPI-stained (4’,6’-diamino-2-phenylindoledihydrochloride; REFERENCES Sigma) to visualize nuclei and finally mounted using Vectashield mounting 1. Virgin HW, Wherry EJ, Ahmed R. Redefining chronic viral infection. Cell. media (Vector laboratories). Images were captured using a Leica SP5 2009;138:30–50. Inverted Confocal microscope and analyzed with ImageJ software. 2. Moskophidis D, Lechner F, Pircher H, Zinkernagel RM. Virus persistence in acutely infected immunocompetent mice by exhaustion of antiviral cytotoxic effector Virus load quantification T cells. Nature. 1993;362:758–61. Viral titers from spleens of infected mice were determined by focus 3. Zajac AJ, Blattman JN, Murali-Krishna K, Sourdive DJD, Suresh M, Altman JD, et al. forming assay as previously described [58]. Briefly, spleens were frozen at Viral immune evasion due to persistence of activated T cells without effector −80 °C right after collection. Tissue was mechanically disrupted and function. J Exp Med. 1998;188:2205–13. resuspended in 500 μl of DMEM 2%FBS. One in ten-fold dilutions were 4. Wherry EJ, Blattman JN, Murali-Krishna K, Most R, van der, Ahmed R. Viral per- prepared and overlaid onto MC57 cell monolayers in 24-well plates. After sistence alters CD8 T-cell immunodominance and tissue distribution and results 48 h of incubation at 37 °C 5% CO2, staining was performed using in distinct stages of functional impairment. J Virol. 2003;77:4911–27. monoclonal rat anti-LCMV antibody (VL-4) for 1 h, followed by Peroxidase 5. Barber DL, Wherry EJ, Masopust D, Zhu B, Allison JP, Sharpe AH, et al. Restoring Anti-Rat IgG Polyclonal Ab (Jackson ImmunoResearch). Plaques were function in exhausted CD8 T cells during chronic viral infection. Nature. visualized using DAB Peroxidase substrate kit (Vector Laboratories). 2006;439:682–7. 6. Cai W, Qin A, Guo P, Yan D, Hu F, Yang Q, et al. Clinical significance and functional studies of myeloid-derived suppressor cells in chronic hepatitis C patients. J Clin ELISA Immunol. 2013;33:798–808. Mouse blood was collected from the submandibular vein and clotted at 7. Norris BA, Uebelhoer LS, Nakaya HI, Price AA, Grakoui A, Pulendran B. Chronic but room temperature for 20 min before centrifugation at 2000 × g for 10 min. not acute virus infection induces sustained expansion of myeloid suppressor cell The serum was collected and stored at −80 °C. VeriKine-HS Mouse numbers that inhibit viral-specific T cell immunity. Immunity. 2013;38:309–21. Interferon-Beta ELISA Kit (Pbl Assay Science) was used to determine IFNB1 8. Wilson EB, Kidani Y, Elsaesser H, Barnard J, Raff L, Karp CL, et al. Emergence of levels in serum, according to the manufacturer’s instructions. distinct multiarmed immunoregulatory antigen-presenting cells during persistent viral infection. Cell Host Microbe. 2012;11:481–91. 9. Alcami A, Koszinowski UH, Alcami A, Koszinowski UH. Viral mechanisms of Quantitative real-time PCR immune evasion. Trends Microbiol. 2000;8:410–8. Total RNA was extracted from whole spleens or from sorted cells, using 10. Asabe S, Wieland SF, Chattopadhyay PK, Roederer M, Engle RE, Purcell RH, et al. respectively Qiagen RNeasy Mini Kit and Qiagen RNeasy Micro Kit. Quality The size of the viral inoculum contributes to the outcome of hepatitis B virus and concentration of RNA were determined by an Agilent Bioanalyzer. RNA infection. J Virol. 2009;83:9652–62. from whole spleens was reverse-transcribed to cDNA using SuperScript III 11. Rouse BT, Sehrawat S. Immunity and immunopathology to viruses: what decides Reverse Transcriptase (ThermoFisher). Quantitative real-time PCR (qPCR) the outcome? Nat Rev Immunol. 2010;10:514–26. was then performed using SYBR select master mix (ThermoFisher) via the 12. Smith N, Possémé C, Bondet V, Sugrue J, Townsend L, Charbit B, et al. Defective Quantstudio 12 K flex (ThermoFisher). RNA from sorted cells, was directly activation and regulation of type I interferon immunity is associated with used to perform qPCR with a one-step Quantitect SYBR green kit (Qiagen) increasing COVID-19 severity. Nat Commun. 2022;13:7254. via the Lightcycler 480 II (Roche) according to manufacturer’s instructions. 13. Zinkernagel RM, Hengartner H, Stitz L. On the role of viruses in the evolution of Primers for all genes were designed using the program Primer Express 3.0 immune responses. Brit Med Bull. 1985;41:92–7. (Applied Biosystems) and ordered from Biomers. Gene expression was 14. Bocharov G, Casella V, Argilaguet J, Grebennikov D, Güerri-Fernandez R, Ludewig normalized to that of Gapdh and compared for the study groups. B, et al. Numbers game and immune geography as determinants of coronavirus pathogenicity. Front Cell Infect Microbiol. 2020;10:559209. RNA-seq library preparation, sequencing and bioinformatic 15. Zhang W, Jang S, Jonsson CB, Allen LJS. Models of cytokine dynamics in the analysis inflammatory response of viral zoonotic infectious diseases. Math Med Biol. Total RNA from spleens was isolated from two mice/ group/ time point and 2019;36:269–95. submitted for sequencing to the Genomics Unit of the Centre for Genomic 16. Boby N, Srivastav A, Srivastav SK, Pahar B. Divergent cytokine and chemokine Regulation (CRG, PRBB). Sequencing libraries were obtained after removing responses at early acute simian immunodeficiency virus infection correlated with virus ribosomal RNA by a Ribo-Zero kit (Illumina). cDNA was synthesized and replication and CD4 T cell loss in a rhesus macaque model. Vaccines. 2023;11:264. tagged by addition of barcoded Truseq adapters. Libraries were quantified 17. Grebennikov D, Karsonova A, Loguinova M, Casella V, Meyerhans A, Bocharov G. using the KAPA Library Quantification Kit (KapaBiosystems) prior to Predicting the kinetic coordination of immune response dynamics in SARS-CoV-2 amplification with Illumina’s cBot. Four libraries were pooled and infection: implications for disease pathogenesis. Mathematics. 2022;10:3154. Cell Death and Disease (2023)14:838 V. Casella et al. 12 18. Suprunenko T, Hofer MJ. Complexities of type I interferon biology: lessons from T cells in human immunodeficiency virus type 1 infection. Clin Vaccine Immunol. LCMV. Viruses. 2019;11:172. 2006;13:556–60. 19. Ahmed R, Salmi A, Butler LD, Chiller JM, Oldstone MB. Selection of genetic var- 44. Channappanavar R. Interferons in coronavirus pathogenesis: the good, the bad, iants of lymphocytic choriomeningitis virus in spleens of persistently infected and the ugly. Cell Host Microbe. 2022;30:427–30. mice. Role in suppression of cytotoxic T lymphocyte response and viral persis- 45. Haller O, Kochs G, Weber F. The interferon response circuit: induction and sup- tence. J Exp Med. 1984;160:521–40. pression by pathogenic viruses. Virology. 2006;344:119–30. 20. Ou R, Zhou S, Huang L, Moskophidis D. Critical role for alpha/beta and gamma 46. Palese P, Basler CF, García-Sastre A. The makings of a killer. Nat Med. interferons in persistence of lymphocytic choriomeningitis virus by clonal 2002;8:927–8. exhaustion of cytotoxic T cells. J Virol. 2001;75:8407–23. 47. Puel A, Bastard P, Bustamante J, Casanova JL. Human autoantibodies underlying 21. Argilaguet J, Pedragosa M, Esteve-Codina A, Riera G, Vidal E, Peligero-Cruz C, et al. infectious diseases. J Exp Med. 2022;219:e20211387. Systems analysis reveals complex biological processes during virus infection fate 48. Chen Y, Graf L, Chen T, Liao Q, Bai T, Petric PP, et al. Rare variant MX1 alleles decisions. Genome Res. 2019;29:907–19. increase human susceptibility to zoonotic H7N9 influenza virus. Science. 22. Domenjo-Vila E, Casella V, Iwabuchi R, Fossum E, Pedragosa M, Castellví Q, et al. 2021;373:918–22. XCR1+ DCs are critical for T cell-mediated immunotherapy of chronic viral 49. Junt T, Scandella E, Ludewig B. Form follows function: lymphoid tissue micro- infections. Cell Rep. 2023;42:112123. architecture in antimicrobial immune defence. Nat Rev Immunol. 2008;8:764–75. 23. Zhang B, Horvath S. A general framework for weighted gene co-expression 50. Junt T, Moseman EA, Iannacone M, Massberg S, Lang PA, Boes M, et al. Sub- network analysis. Stat Appl Genet Mol. 2005;4:Article17. capsular sinus macrophages in lymph nodes clear lymph-borne viruses and 24. Wang Y, Swiecki M, Cella M, Alber G, Schreiber RD, Gilfillan S, et al. Timing and present them to antiviral B cells. Nature. 2007;450:110–4. magnitude of type I interferon responses by distinct sensors impact CD8 T cell 51. Dinther, van D, Veninga H, Iborra S, Borg EGF, Hoogterp L, et al. Functional CD169 exhaustion and chronic viral infection. Cell Host Microbe. 2012;11:631–42. on macrophages mediates interaction with dendritic cells for CD8+ T cell cross- 25. Teijaro JR, Ng C, Lee AM, Sullivan BM, Sheehan KCF, Welch M, et al. Persistent priming. Cell Rep. 2018;22:1484–95. LCMV infection is controlled by blockade of type I interferon signaling. Science. 52. Eloranta ML, Alm GV. Splenic marginal metallophilic macrophages and marginal 2013;340:207–11. zone macrophages are the major interferon‐α/β producers in mice upon intra- 26. Zuniga E, Macal M, Lewis GM, Harker JA. Innate and adaptive immune regulation venous challenge with herpes simplex virus. Scand J Immunol. 1999;49:391–4. during chronic viral infections. Ann Rev Virol. 2014;2:1–25. 53. Kityo C, Makamdop KN, Rothenberger M, Chipman JG, Hoskuldsson T, Beilman 27. Shaabani N, Duhan V, Khairnar V, Gassa A, Ferrer-Tur R, Häussinger D, et al. GJ, et al. Lymphoid tissue fibrosis is associated with impaired vaccine responses. J CD169+ macrophages regulate PD-L1 expression via type I interferon and Clin Invest. 2018;128:2763–73. thereby prevent severe immunopathology after LCMV infection. Cell Death Dis. 54. Makris S, Winde CM, de, Horsnell HL, Cantoral-Rebordinos JA, Finlay RE, et al. 2016;7:e2446. Immune function and dysfunction are determined by lymphoid tissue efficacy. 28. Honke N, Shaabani N, Cadeddu G, Sorg UR, Zhang DE, Trilling M, et al. Enforced Dis Model Mech. 2022;15:dmm049256. viral replication activates adaptive immunity and is essential for the control of a 55. Estes JD, Wietgrefe S, Schacker T, Southern P, Beilman G, Reilly C, et al. Simian cytopathic virus. Nat Immunol. 2012;13:51–7. immunodeficiency virus-induced lymphatic tissue fibrosis is mediated by trans- 29. Scandella E, Bolinger B, Lattmann E, Miller S, Favre S, Littman DR, et al. forming growth factor β1-positive regulatory T cells and begins in early infection. Restoration of lymphoid organ integrity through the interaction of lymphoid J Infect Dis. 2007;195:551–61. tissue–inducer cells with stroma of the T cell zone. Nat Immunol. 2008;9:667–75. 56. Schacker TW, Nguyen PL, Beilman GJ, Wolinsky S, Larson M, Reilly C, et al. Col- 30. Welsh R Jr. Cytotoxic cells induced during lymphocytic choriomeningitis virus lagen deposition in HIV-1 infected lymphatic tissues and T cell homeostasis. J Clin infection of mice. I. Characterization of natural killer cell induction. J Exp Med. Invest. 2002;110:1133–9. 1978;148:163–81. 57. Zeng M, Smith AJ, Wietgrefe SW, Southern PJ, Schacker TW, Reilly CS, et al. 31. Kyburz D, Speiser DE, Battegay M, Hengartner H, Zinkernagel RM. Lysis of infected Cumulative mechanisms of lymphoid tissue fibrosis and T cell depletion in HIV-1 cells in vivo by antiviral cytolytic T cells demonstrated by release of cell internal and SIV infections. J Clin Invest. 2011;121:998–1008. viral proteins. Eur J Immunol. 1993;23:1540–5. 58. Battegay M, Cooper S, Althage A, Bänziger J, Hengartner H, Zinkernagel RM. 32. Lang PA, Lang KS, Xu HC, Grusdat M, Parish IA, Recher M, et al. Natural killer cell Quantification of lymphocytic choriomeningitis virus with an immunological activation enhances immune pathology and promotes chronic infection by lim- focus assay in 24- or 96-well plates. J Virol Methods. 1991;33:191–8. iting CD8+ T-cell immunity. Proc National Acad Sci. 2012;109:1210–5. 33. Waggoner SN, Cornberg M, Selin LK, Welsh RM. Natural killer cells act as rheostats modulating antiviral T cells. Nature. 2012;481:394–8. 34. Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network ACKNOWLEDGEMENTS analysis. Bmc Bioinformatics. 2008;9:559. We thank the Flow Cytometry and Genomic Core Facilities (Universitat Pompeu 35. Miyake Y, Asano K, Kaise H, Uemura M, Nakayama M, Tanaka M. Critical role of Fabra) and the Advanced Light Microscopy Unit (ALMU-CRG) for excellent technical support, Monica Perez (CReSA-IRTA-UAB) for helping with the histology samples and macrophages in the marginal zone in the suppression of immune responses to Hector Huerga Encabo (Immunology Unit, Universitat Pompeu Fabra) for fruitful apoptotic cell–associated antigens. J Clin Invest. 2007;117:2268–78. 36. Kolumam GA, Thomas S, Thompson LJ, Sprent J, Murali-Krishna K. Type I inter- discussions and ideas. We further thank Dr. Kenji Kohno and Dr. Masato Tanaka for ferons act directly on CD8 T cells to allow clonal expansion and memory for- allowing us to use their CD169-DTR transgenic mice which we obtained through Dr. mation in response to viral infection. J Exp Med. 2005;202:637–50. Andres Hidalgo. The graphical abstract was created using BioRender.com platform. 37. Samal J, Kelly S, Na-Shatal A, Elhakiem A, Das A, Ding M, et al. Human immu- This work was supported by grants from the Spanish Ministry of Science and Innovation grant No. PID2019-106323RB-I00 AEI//10.13039/501100011033 and nodeficiency virus infection induces lymphoid fibrosis in the BM-liver-thymus- PID2022-141395OB-I00, the “Unidad de Excelencia María de Maeztu”, funded by spleen humanized mouse model. Jci Insight. 2018;3:e120430. the AEI (CEX2018-000792-M), the Russian Science Foundation grant No. 23-11-00116 38. Sanchez JL, Hunt PW, Reilly CS, Hatano H, Beilman GJ, Khoruts A, et al. Lymphoid fibrosis occurs in long-term nonprogressors and persists with antiretroviral and the AEC funded by ISCIII /MINECO (PT17/0009/0019) and co-funded by FEDER. therapy but may be reversible with curative interventions. J Infect Dis. 2015;211:1068–75. 39. Fuller MJ, Khanolkar A, Tebo AE, Zajac AJ. Maintenance, loss, and resurgence of T AUTHOR CONTRIBUTIONS cell responses during acute, protracted, and chronic viral infections. J Immunol. Conceptualization, JA, GB and AM; Investigation, VC, ED-V, MP and PCR; Bioinformatic 2004;172:4204–14. analysis, AE-C, IdlR; Resources, CL-R; Writing, VC, JA and AM; Visualization, VC; Project 40. Bocharov G, Klenerman P, Ehl S. Modelling the dynamics of LCMV infection in Administration, JA; Funding acquisition, JA, GB and AM. mice: II. Compartmental structure and immunopathology. J Theor Biol. 2003;221:349–78. 41. McNab F, Mayer-Barber K, Sher A, Wack A, O’Garra A. Type I interferons in infectious disease. Nat Rev Immunol. 2015;15:87–103. COMPETING INTERESTS 42. Odermatt B, Eppler M, Leist TP, Hengartner H, Zinkernagel RM. Virus-triggered The authors declare no competing interests. acquired immunodeficiency by cytotoxic T-cell-dependent destruction of antigen-presenting cells and lymph follicle structure. Proc Natl Acad Sci. 1991;88:8252–6. ETHICS APPROVAL 43. Schacker TW, Brenchley JM, Beilman GJ, Reilly C, Pambuccian SE, Taylor J, et al. The experimental procedures in mice were carried out in compliance with the Lymphatic tissue fibrosis is associated with reduced numbers of naïve CD4 + guidelines from Generalitat de Catalunya and were approved by the ethical Cell Death and Disease (2023)14:838 V. Casella et al. 13 committees for animal experimentation at Parc de Recerca Biomèdica de Barcelona Open Access This article is licensed under a Creative Commons (CEEA-PRBB, Spain). Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative ADDITIONAL INFORMATION Commons license, and indicate if changes were made. The images or other third party Supplementary information The online version contains supplementary material material in this article are included in the article’s Creative Commons license, unless available at https://doi.org/10.1038/s41419-023-06374-y. indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory Correspondence and requests for materials should be addressed to Jordi Argilaguet regulation or exceeds the permitted use, you will need to obtain permission directly or Andreas Meyerhans. from the copyright holder. To view a copy of this license, visit http:// creativecommons.org/licenses/by/4.0/. Reprints and permission information is available at http://www.nature.com/ reprints © The Author(s) 2023 Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Cell Death and Disease (2023)14:838