Influenza hijacks myeloid cells to inflict Type-I Interferon-fueled damage in the heart

Summary

Abundant evidence has correlated influenza infection with cardiovascular disease, yet mechanisms linking infection with the heart remain poorly understood. Here, we have shown that influenza infection damaged the human and murine heart. In mice, we showed that shortly after pulmonary infection, the virus infected a circulating pro-dendritic cell 3 (pro-DC3) myeloid cell that expressed high concentrations of the chemokine receptor CCR2. The heart, which produces abundant CCL2, preferentially attracted infected pro-DC3. In the myocardium, the virus escaped pro-DC3 cells, infected cardiomyocytes, and triggered production of type-I Interferon. Engagement of the IFN-I receptor (IFNAR1) on cardiomyocytes caused tissue damage and compromised heart function. Genetically and therapeutically dampening IFNAR1 exclusively in cardiomyocytes protected the heart while preserving anti-viral immunity in the lung. Our results identify a series of host-pathogen interactions that propagate tissue damage, and uncover an axis for intervention to mitigate cardiovascular risk following viral infection.

Graphical Abstract

Introduction

Respiratory infections, including those caused by influenza A viruses (IAV) remain critical determinants of life expectancy ^1–5. Influenza alone has caused recurrent pandemics and today is responsible for an estimated 1 billion infections annually ³. Although most infections are mild and self-resolving, in certain cases influenza infections can become severe or fatal ^{3, 4}. While extensive work has uncovered local pulmonary immune axes that heighten or mitigate the risk of severe infection ⁶, less is known about the virus ’systemic effects.

Clinically, abundant data link influenza with myocarditis ⁷, heart failure (HF) ⁸, and risk of myocardial infarction (MI) ^9–11. Despite these associations, the underlying mechanisms remain unknown, though many have been proposed. On the one hand, influenza may compromise the heart and raise the risk of HF or MI by unleashing non-specific systemic inflammation, causing widespread collateral damage. On the other hand, influenza might display features of cardiotropism by targeting and replicating in the heart. The literature offers clues, but no consensus exists. In this study, we sought to investigate how IAV affects the cardiovascular system, specifically focusing on the heart, the leukocytes that seed it, and the nature of their inflammatory response.

Results

Cardiovascular disease and influenza pathogeneses accelerate each other in mice and humans

Clinical evidence shows that influenza virus (IV) infection can damage the heart ^{10, 12}. Here, we studied autopsy records of individuals who died following laboratory confirmed IV infection and found that the vast majority of individuals, irrespective of age, exhibited pronounced cardiovascular risk factors (Figure 1A), with varying magnitudes of cardiac involvement (Figure S1A). We also collected plasma from hospitalized IV⁺ individuals and evaluated acute markers of cardiomyocyte damage (troponin) and long-term deterioration of cardiac function (NT-proBNP). We found that >60% of patients exhibited signs of cardiac damage (Figure 1B). Importantly, <10% of those hospitalized from acute severe ulcerative colitis (ASUC) had elevated cardiac biomarkers (Figure S1B).

Figure 1: Cardiovascular disease and influenza infection aggravate each other.

A) Autopsy reports of individuals who succumbed confirmed influenza virus⁺ (IV⁺)

B) Troponin and NT-proBNP amounts in hospitalized IV⁺ patients

C) Troponin and NT-proBNP amounts in infected mice

D) Representative image (left) and quantification (right) of cardiomegaly

E) Representative cMRI images of end systole and end diastole (left), quantified as left ventricle ejection fraction (LVEF) on the right

F) Troponin amounts in infected mice at various PFU inocula

G) Representative and enumerated TUNEL staining images (left) and merged images (right) in uninfected and day 9 infected heart. Scale bar=100μm

H) Schema of comorbidity models employed

I) Representative cMRI images of end diastole and end systole in comorbidity models

J) LVEF in comorbidity models, expressed as a delta from the LVEF of their respective uninfected controls

K) Troponin amounts in comorbidity models, shown as a percent change from their respective uninfected controls

Data are expressed as means ± s.e.m. and each dot represents an individual patient (B) or mouse (C-J). See also Figure S1.

These pervasive clinical observations necessitated greater mechanistic insight. As such, we employed mouse models of IAV infection (H1N1 PR8) (Figure S1C–D). In line with our findings in patients, mice had significantly elevated troponin and NT-proBNP (Figure 1C), cardiomegaly (Figure 1D), and cardiac fibrosis (Figure S1E–G). Left ventricular ejection fraction (LVEF) declined (Figure 1E and Figure S1H), mimicking clinical observations ¹⁰. Moreover, damage was dose-dependent and correlated with weight loss (Figure 1F), but independent of genetic mouse background (Figure S1I). Infection caused cardiomyocyte death (Figure 1G), but largely spared other organs, with the exception of a modest decrease in colon length ¹³ (Figure S1J–K). Aging, atherosclerosis, Type 1 MI ^14–16, and Type 2 MI ¹⁷ worsened these parameters (Figure 1H–K and Figure S1L–M). Finally, optimal vaccination (trivalent inactivated IAV vaccine (TIV) followed by matching H1N1 A/New Caledonia/20/1999 infection) and/or suboptimal vaccination (TIV followed by mismatched PR8 infection) protected the heart (Figure S1N–Q). Collectively, these data show that CVD and IAV infection have bidirectional influence on severity that conventional prophylactic vaccination mitigates.

Influenza causes myocarditis but not indiscriminate tissue inflammation

As influenza is associated with myocarditis ^{10, 18, 19}, we assessed inflammation post-infection (Figure 2A). IAV infection induced robust inflammation in the lung, marked by monocytosis, neutrophilia, and macrophage accumulation (Figure S2A). However, inflammation was not confined to the lung but included the bone marrow and blood (Figure 2B–C), resolving by 28 days p.i., as well as the heart, which was highly receptive to myeloid cell infiltration. Tissue resident macrophage numbers also rose (Figure 2D and Figure S2B) and remained elevated at 28 days. Comorbidity models revealed similar but more pronounced differences in blood-derived cell influx (Figure S2C).

Figure 2: Influenza infection induces myocarditis predominated by Ly6C^hi CCR2⁺ myeloid cells.

A) Phenotyping schema to differentiate between vascular and tissue leukocytes

B) Representative flow plots (top) and quantification (bottom) of bone marrow populations

C) Representative flow plots (top) and quantification (bottom) of blood populations

D) Representative flow plots (top) and quantification (bottom) of cardiac populations

E) Quantification of populations as in D of various tissues. Cell counts are represented as a percentage change from the uninfected.

F) Representative images and quantification of cleared CCR2^RFP/+ hearts. Scale bar as indicated.

G) Parabiosis design (top) and number of Ly6C^hi monocytes in the lungs and heart

Data are expressed as means ± s.e.m. and each dot represents an individual mouse. See also Figure S2.

To test the relative organ specificity of the heart in IAV-induced extra-pulmonary inflammation, we screened 15 other tissues, noting infiltration in few, but not all, sites (Figure 2E). To investigate the spatial distribution of accumulating leukocytes in the heart, we chemically cleared hearts from healthy or day 3-infected Ccr2^Rfp/+ mice. Cells accumulated in focused aggregates in various locations, including the septum and left ventricle (Figure 2F). Finally, we wondered whether leukocyte inflammation depended on the heart’s close anatomical connection to the lung. Through parabiosis, we determined this was not the case (Figure 2G and Figure S2D–E). These results demonstrate that IAV inflames the heart via the circulation.

A granulocyte-monocyte progenitor (GMP)-independent pro-DC3 drives myocarditis

Tissue-resident macrophages are essential to homeostasis and infection ^{20, 21}. To determine whether the increase in cardiac macrophages was due to enhanced recruitment of bone marrow-derived monocytes or enhanced proliferation of embryonically-derived macrophages, we used the markers Tim4 and Lyve1 ^{22, 23}. In accordance with previous reports ²², at steady-state, the subtypes each comprised ~50% of cardiac macrophages. However, upon infection, the Tim4⁻ Lyve1⁻ population expanded while the Tim4⁺ Lyve1⁺ population remained unchanged, until 28 days post-infection when the latter population began to rise (Figure 3A), aligning with the notion that Tim4 and Lyve1 define long-lived resident macrophages. As expected, Tim4⁺ Lyve1⁺ macrophages proliferated more than Tim4⁻ Lyve1⁻, yet this was unaltered following infection (Figure S3A). Thus, IAV-induced macrophages appear to be sourced from the circulation.

Figure 3: The GMP-independent cDC3-lineage accumulates in the heart post-infection.

A) Representative flow cytometry plots and quantification of cardiac macrophage subpopulations

B) Schema of Ms4a3^TdT fate-mappers

C) TdT expression in Tim4⁻ cardiac macrophages

D) Expected versus observed frequency of TdT⁺ cardiac macrophages

E) Enumeration of CCR2-expressing cardiac macrophages over time (top) with expected versus observed frequency (bottom)

F) Representative flow cytometry plots of Ms4a3^TdT expression in Ly6C^hi cells under listed conditions

G) Subdivision of Ly6C^hi cells by CD319 and CD177 expression in the myocardium from F

H) Ms4a3^TdT positivity in the subsets of Ly6C^hi cells as quantified on the right.

I) Relative expression of listed surface markers of the indicated tissue subsets

J) Alternative gating strategy of myeloid subsets as well as expression of Ms4a3^TdT, CD319, CD177, and CD64

K) Confocal images and 3D reconstructions of infected hearts. Scale bars as indicated. For the confocal image overlap a 0.7 Gaussian Blur was applied.

Data are expressed as means ± s.e.m. and each dot represents an individual mouse. See also Figure S3.

To more precisely determine the ontogeny of accumulating cells, we employed the GMP fate mapper (Ms4a3^Cre × Rosa26^TdT; hereafter Ms4a3^TdT mice) ²⁴ (Figure 3B). We observed elevated TdT⁺ Tim4⁻ cardiac and interstitial pulmonary macrophages (Figure 3C and Figure S3B), with no changes in other tissues (Figure S3C). As cells that express(ed) Ms4a3 are permanently tagged, we anticipated strong concordance between Tim4⁻ CD64⁺ cells and accumulating TdT⁺ CD64⁺. Yet nearly 75% of accumulating Tim4⁻ CD64⁺ cells in the heart were TdT⁻ (Figure 3D). An analogous CCR2-based method for monocyte-derived macrophages ²² provided similar results (Figure 3E). To explain this, we probed the bone marrow for potential fate-mapping aberrations during infection. Yet, progenitor TdT expression in infected mice was as expected (Figure S3D) ²⁴.

We therefore sought to better characterize the ontogeny of these cells at day 3 post-infection. In the blood >90% of Ly6C^hi myeloid cells were TdT⁺, and thus classical Ly6C^hi monocytes. In the lung, most of these cells were also TdT⁺. On the other hand, ~60% of these cells in the heart were TdT⁻, and this percentage increased to >75% after infection. Using MI—a robust tissue-damaging stimulus—we found a high accumulation of TdT⁺ monocytes in the myocardium, indicating that the heart can accumulate bona fide monocytes (Figure 3F). That said, the heart was unusual in its low prevalence of TdT⁺ Ly6C^hi cells (Figure S3E). Therefore, while Ms4a3^TdT mice faithfully map GMP-derived Ly6C^hi monocytes, the heart attracted a population of GMP-independent Ly6C^hi cells.

Progenitors of the conventional dendritic cell 2 and 3 (pre-DC2 and pro-DC3) have been described within this gate ²⁵ and a recent study proposed that the surface markers CD177 and CD319 distinguish between Ly6C^hi Ms4a3^TdT+ (GMP-derived) monocytes and Ly6C^hi Ms4a3^TdT− (MDP-derived) cells ²⁶. To determine the relative contribution of these populations, we enumerated the numbers of the known monocyte and DC progenitors in the bone marrow and found notable changes among the various progenitors (Figure S3F).

As Ly6C^hi monocytes might derive from the cMoP or Ly6C^lo MDP ²⁴, and as Ly6C^hi TdT⁺ cells were a minority of total Ly6C^hi myeloid cells in the heart (Figure 3F), we speculated it was the atypical Ly6C⁺ MDP/DC lineage preferentially accumulating. Indeed, TdT⁺ Ly6C^hi cells were almost exclusively CD177⁺ while the TdT⁻ Ly6C^hi cells were mostly CD319⁺ (Figure 3G) aligning with the atypical lineage ²⁶. We confirmed this by reverse gating (Figure 3H). Consistently, the lung had a higher propensity for classical CD177⁺ TdT⁺ and CD319⁻ CD177⁻ TdT⁺ monocytes (Figure S3G). The heart had the greatest reliance on CD319⁺ Ly6C^hi myeloid cells compared to all other tissues (Figure S3H). To strengthen our characterization of this atypical cell type, we compiled surface markers and gating strategies from the two studies describing these cells ^{25, 26}, as well as included the most widely utilized DC fate-mapper (Zbtb46^TdT) and conventional CX3CR1^GFP/+ and CCR2^RFP/+ reporters ²⁷. These analyses confined the vast majority of CD319 expression in the heart to CD64-expressing Msa4a3⁻ MDP-derived pro-DC3, and not monocytes (Figure 3I–J and Figure S3I–J). Finally, confocal images and subsequent 3D reconstructions showed co-expression of CD319 with CD68 on cells interspersed with cardiomyocytes in the infected heart (Figure 3K). Whether these MDP-derived pro-DC3s give rise to conventional macrophages or whether they exclusively give rise to DC3 is unknown. However, given our lineage tracing and surface marker expression, we will refer to these cells as pro-DC3/DC3.

Infectious influenza particles seed the myocardium via pro-DC3

Next, we focused on the virus. As expected, viral titers were highest in the lung, reaching 10⁶ PFU/tissue. In accordance with previous reports ^{19, 28, 29}, the heart supported viral replication (Figure 4A and Figure S4A). We observed sporadic replication in the aorta, lung-draining medLN, and thymus, but no other tissues at day 3. Ns1 transcript amounts were equivalent in the lung, heart, and aorta, and even higher in the medLN. (Figure 4B). Cumulatively, only the lung, heart, aorta, and medLN became inflamed and supported replication (Figure 4C), suggesting a link between leukocyte migration and viral propagation. Moreover, while viral dose correlated with pulmonary viral loads, viral concentrations in the heart were relatively resistant to viral dose (Figure S4B), suggesting a maximal cardiac viral capacity.

Figure 4: pro-DC3 transport IAV to the heart via CCR2.

A) Viral titers in tissues

B) Ns1 transcript in tissues

C) Compilation of organs based on their profile at 3 days p.i. of inflammation, IAV transcript, and viral replication

D) Ns1 transcript in hearts and lungs of parabiosis mice

E) Representative flow cytometry plots and characterization of NP⁺ leukocytes

F) Schema of PB2-Cre IAV and representative cleared heart images. Scale bars as indicated.

G) Flow cytometry using the PB2-Cre as in F

H) Correlation between plasma troponin amounts and myocardial Ly6C^hi cells

I) Relative Ccl2 transcript in listed organs at homeostasis

J) Relative amount of CCL2 in various tissues/mg of tissue

K) In vivo adoptive transfer model and frequency of transferred Ly6C^hi cells in various tissues with CD319⁺ and CD177⁺ subsets

L) Frequency of transferred CCR2-deficient Ly6C^hi cells following transfer as in K

M) Blood Ly6C^hi cells, cardiac NP⁺ cells, cardiac viral loads, and troponin in Ccr2^−/− and pioglitazone-treated mice

N) Accumulation (left) and troponin amounts (right) following transfer of sorted Ly6C^hi subsets as in K

O) Correlation between plasma NT-proBNP and CCL2 in IV-infected hospitalized patients

Data are expressed as means ± s.e.m. and each dot represents an individual mouse or human (O). See also Figure S4.

In mice, viral titers dropped after day 6, prior to the onset of symptoms (Figure S4C), though individuals hospitalized with influenza may experience a prolonged period during which the virus is shed ³⁰. Accordingly, immunohistochemistry for nucleoprotein (NP) in hearts of individuals in our autopsy cohort revealed IAV-positive sections (Figure S4D), as reported during pandemics ^{10, 31}. Thus, cardiac IAV presence is conserved across mice and humans.

What makes the heart susceptible to such high concentrations of IAV transcription? We found that proximity to the lung was not the answer (Figure 4D and Figure S4E). As leukocytes may transport viruses via the circulation ^32–34, we sorted leukocyte subpopulations from the blood and hearts of infected mice and found that pro-DC3 — along with CD64⁺ macrophages in the heart — contained Ns1 (Figure S4F). This suggests that infected Ly6C^hi cells (including pro-DC3) accumulate in the myocardium where they differentiate into macrophages and/or DC3s ³².

To investigate this further, we looked for the presence of NP (i.e., translation-competent IAV virions) in leukocytes. We observed NP⁺ Ly6C^hi cells, neutrophils, and lymphocytes in lung vasculature and NP⁺ Ly6C^hi myeloid cells/macrophages in the tissue parenchyma (Figure S4G). In the vasculature, NP⁺ cells, despite constituting a minority of Ly6C^hi cells in the lung (Figure S3G), were predominantly CD319⁺ (Figure S4G). We also detected NP⁺ Ly6C^hi cells in the peripheral blood but not elsewhere (Figure S4H). In the heart vasculature and tissue, there were NP⁺ infected leukocytes (Figure S4I and Figure 4E) which, in the myocardium, were principally pro-DC3 (Figure 4E). Because we saw NP⁺ Ly6C^hi cells in the blood, we repeated parabiosis experiments and obtained similar results (Figure S4J–K). To better understand these cells ’identity and location, we used Rosa26^TdT mice infected with a PB2-Cre IAV ³⁵ to identify all infected cells. As proof-of-concept, we cleared lungs of infected mice and found extensive TdT expression in the respiratory tract and parenchyma (Figure S4L). In chemically cleared hearts, we observed foci of infected cells, the majority of which co-expressed CD319 (Figure 4F and Figure S4M). Flow cytometry confirmed these observations (Figure 4G).

The chemokine CCL2 is critical for mobilizing and recruiting CCR2⁺ cells and is particularly relevant for mobilizing monocytes ³⁶. We therefore asked whether CCL2 was important in recruiting pro-DC3. Correlation experiments confirmed that plasma troponin concentrations associated with increased cardiac Ly6C^hi cells (Figure 4H). CCR2 expression rose on blood Ly6C^hi cells following infection in vivo and in vitro (Figure S4N–O) and CD319⁺ pro-DC3s expressed the chemokine receptor higher than any other cell we screened (Figure S4P). Among the 19 tissues, the aorta and heart were most Ccl2 rich at homeostasis (Figure 4I), elevating transcript and protein abundance after infection (Figure 4J and Figure S4Q). Similar results were obtained in parabiosis mice (Figure S4R). To test whether CD319⁺ pro-DC3 preferentially accumulate in the heart due to CCL2/CCR2, we adoptively transferred the relevant cells. Bone marrow Ly6C^hi cells were purified, infected in vitro and injected into naive mice (Figure S4S) or the cells were left uninfected and injected into infected mice (Figure 4K). In both cases the heart accumulated transferred cells; the cells were predominantly CD319⁺ in infected mice (Figure S4S and Figure 4K). Transfer of CCR2-deficient cells abrogated this phenotype (Figure 4L). As CCR2-dependent inflammation correlates with tissue damage ^{37, 38}, we assessed Ccr2^−/− mice and mice treated with the PPARγ agonist pioglitazone, which inhibits CCL2 activity³⁹. Both groups had fewer blood Ly6C^hi myeloid cells compared to controls. In the heart, this led to fewer infected cells, lower cardiac viral loads, and low plasma troponin (Figure 4M). Specific transfer of CD177⁺ versus CD319⁺ Ly6C^hi cells caused greater cardiac damage and inflammation (Figure 4N). Therefore, pro-DC3 infiltrate and harm the heart via CCL2-CCR2. Finally, we found a correlation between blood CCL2 and blood NT-proBNP amounts in hospitalized influenza, but not ASUC, patients (Figure 4O and Figure 4ST), suggesting this pathway’s conserved importance in humans.

IAV replicates in cardiomyocytes due to interactions with infected myeloid cells

Previous studies detailed infection in cardiac non-leukocytes, yet the cell type principally responsible for productive PFU remains undefined ^{9, 19, 28}. We detected Ns1 in cardiomyocytes, but not endothelial cells and fibroblasts (Figure 5A), and only cardiomyocytes sustained viral replication (Figure 5B) and died (Figure S5A). Confocal microscopy and 3D reconstructions of cleared PB2-Cre infected Rosa26^TdT hearts confirmed cardiomyocyte infection (Figure 5C, Figure S5B and Video S1–2).

Figure 5: Cardiomyocytes support productive IAV replication via interactions with infiltrating infected CD319⁺ cells.

A) Ns1 transcript in sorted cardiac epithelial cells, fibroblasts, or isolated cardiomyocytes

B) Experimental design for in vitro infection with IAV of monocyte/macrophage subsets and neonatal cardiomyocytes (left) and supernatant PFU (right)

C) Representative confocal images of uninfected and day 6 infected hearts. Scale bar as indicated.

D) RNA-seq analysis of the change in population frequency and number of DEGs following infection

E) Hallmark Gene Set enrichment analyses of significantly up- or down- regulated pathways

F) Pseudotime analyses of cardiomyocyte sub-clusters

G) Cell-Chat number and intensity of interactions between clusters as whole number (left) and relative heat map (right)

H) Representative confocal images and Z stack reconstruction of CD319⁺ cells and cardiomyocytes (laminin) with co-expression of CD68. Scale bars as indicated.

I) Quantification of CD319-expressing cell area and shortest distance to cardiomyocytes

J) Representative flow cytometry plots of 7AAD⁺ (dead) Ly6C^hi cells, using differential staining for CD319, CD177

K) Schema of co-culture experiments (top left) with troponin amounts, PFU, Ns1 concentrations and relative mitochondrial mass in neonatal cardiomyocytes post-infection

Data are expressed as means ± s.e.m. In A and J, dots indicate individual mice, in B and K individual culture wells, and in I independent field of view. See also Figure S5–6.

To address how cardiomyocytes may be infected, we first performed single nucleus RNA-seq (snRNA-seq) (Figure S5C). We observed broad changes across clusters in terms of cell frequency and differentially expressed genes. Cardiomyocyte subclusters were clearly separated by cluster-defining genes (Figure S5D–E). Frequencies rose upon infection in cardiomyocyte (CM)1 (largest following infection) and CM4 (greatest relative delta in the IAV group) with contractions of CM clusters 2 (the largest at homeostasis) and 3 (Figure 5D and Figure S5D). To better understand how these populations change, we performed gene set enrichment analyses (GSEA) for each MSigDB Hallmark gene set (Figure 5E). We found a nearly ubiquitous increase of Type I IFN (α; IFN-I) and elevated GO biological processes and molecular functions in both macrophage and monocyte clusters substantially corresponded to antiviral and inflammatory responses (Figure S5F). Additionally, cardiomyocyte subsets had altered Hallmark metabolic pathways linked to heart failure and cell death (Figure 5E) ^{40, 41}. While the adult heart contains fully differentiated, non-proliferating cardiomyocytes, the spectrum of transcriptional states, the numerical variations, and the metabolic alterations following infection led us to order the cells in pseudotime (Figure 5F), which demonstrated widespread aberrations in cardiomyocytes, including a dysregulated cardiomyocyte landscape away from the largest homeostatic populations (CM2 and CM3) towards younger states (CM1 and CM4). We also performed GO pathway analyses on the top 25 genes most differentially expressed between infected and uninfected populations, noting numerous changes (Figure S5G–H). For example, CM1 showed clear signs of cardiac abnormalities and viral infection, including enriched IFN-I, abnormal myocardial fiber morphology, and apoptosis.

Next, we performed Cell Chat analyses ⁴². While total interactions and interaction strength declined upon infection (Figure 5G and Figure S5I), IAV did induce certain interactions, such as fibroblast production of colony stimulating factors (CSF)1 (Figure S5J). Notably, sender interactions from monocyte/macrophage to cardiomyocyte subpopulations increased (Figure 5G), implicating CYP and SIRP family members known for IAV pathogenesis and spread ^43–45. The monocyte cluster (which contained pro-DC3), was enriched for CD319 (Slamf7)-expressing pro-DC3, compared to CD177-expressing monocytes (Figure S5K). GO analysis further revealed that CD319⁺ cells, and not CD319⁻ myeloid cells, were enriched for cytokine responses, DC machinery, and myeloid cell differentiation (Figure S5L, M). Transcriptionally and functionally, then, MDP-dependent cells align with pro-DC3.

To investigate CD319⁺-cardiomyocyte interactions in vivo, we imaged the cells by confocal microscopy (Figure 5H). As expected, overall cardiomyocyte area decreased following infection. However, overall CD319 and CD68 area, as well as CD319/laminin overlap increased following infection (Figure 5I and Figure S5N). 3D reconstruction of Z-stacks revealed shortened distances between CD319⁺ cells and cardiomyocytes upon infection (Figure 5H–I and Figure SN). Leveraging our confocal microscopy using the PB2-Cre IAV model, we confirmed the presence of IAV^+/− CD319⁺ cells interacting with IAV⁺ cardiomyocytes (Figure 5C). Altogether, these analyses uncovered a dense interconnected network of recruited CD319⁺ pro-DC3s, cardiomyocytes, and macrophages ^46–48.

Mechanistically, our RNA-seq data also showed that macrophages expressed apoptosis-relevant genes (Figure S5F). We speculated that intercellular connections and apoptotic blebbing, as critical determinants of IAV propagation, may be responsible for viral transfer to cardiomyocytes. Flow cytometry showed more apoptosis in total macrophages as well as Ly6C^hi cells following infection (Figure S5O). The subtypes most susceptible to cell death were Tim4⁻ CD64⁺ cells and CD319⁺ pro-DC3 (Figure 5J and Figure S5P). We therefore cultured cardiomyocytes with plasma of infected mice; with infected B cells; with infected total Ly6C^hi cells, which included Ms4a3⁺ CD177⁺ monocytes and Ms4a3⁻ CD319⁺ pro-DC3; and with infected CD319⁺ pro-DC3 or CD177⁺ monocytes (Figure 5K and Figure S5Q–S). While cultures with total infected Ly6C^hi cells caused cell death and recoverable PFU, those with plasma from infected mice and infected B cells yielded no phenotype. Cell death and PFU were driven by CD319⁺ pro-DC3s and not CD319⁻ monocytes, as those co-cultures exhibited elevated troponin release (cardiomyocyte-death specific), more recoverable PFU, greater intracellular cardiomyocyte Ns1 amounts, and greater cardiomyocyte mitochondrial mass (CM1 specific enrichment of OXPHOS; Figure 5E). These findings demonstrate that CD319⁺ pro-DC3 transfer IAV to cardiomyocytes, which leads to replication, cell death, and metabolic reprogramming.

Targetable Type I IFN signaling on cardiomyocytes is responsible for IAV damage

Our results describe how influenza seeds the heart. However, the pathway responsible for the damage remains unknown. UCell score analysis on the Hallmark IFN-α pathway showed near universal enrichment, including in myeloid cells as well as CM1 and CM3 (Figure 6A). We found that both IFN-ɑ and IFN-β were produced following infection (Figure 6B and Figure S6A), as was SERPINE1, a non-canonical anti-IAV ISG ⁴⁹ (Figure 6C and Figure S6B).

Figure 6: Type I IFN signaling on cardiomyocytes causes cardiac dysfunction.

A) UCell scores for hallmark IFN-α pathway

B) Total active cardiac IFN-I using a B16 assay

C) SERPINE1 amounts in the heart p.i.

D) IFN-β amounts in hearts p.i.

E) Pulmonary and cardiac viral loads

F) Pulmonary inflammation assessed by flow cytometry

G) Troponin of infected mice

H) Representative day 9 MRI images at systole and diastole (left) used to quantify LVEF (right)

I) IFN-β concentrations in hearts at day 3 p.i.

J) Troponin of infected mice

K) Representative day 9 MRI images at systole and diastole (left) used to quantify LVEF (right)

L) Pulmonary and cardiac viral loads

M) Pulmonary inflammation assessed by flow cytometry

N) Schematic for in vivo tissue specific knockout of Ifnar1, where an AAV vector encoding Cre recombinase expressed from the cTnT promoter along with an Ifnar1-targeting gRNA is injected intracardiac into SpCas9 mice. Cardiomyocyte-specific Cre recombinase expression deletes the LSL cassette to activate SpCas9 nuclease expression, which complexes with the Ifnar1 gRNA to target and knockout the Ifnar1 gene (bottom). Validation of expression in the LV by immunofluorescence (right). Scale bar as indicated.

O) Troponin of infected mice

P) Representative day 21 MRI images at systole and diastole (left) used to quantify LVEF (right)

Q) Schematic for the design and attenuation of IFNAR1 signaling using truncated IFNAR1 mod-mRNA (left). Validation of the down regulation of IFNAR1 signaling in the hearts of vehicle or mod-RNA injected hearts that were infected (right).

R) Troponin of infected mice

S) Representative day 21 MRI images at systole and diastole (left) used to quantify LVEF (right)

Data are expressed as means ± s.e.m. and each dot represents an individual mouse. See also Figure S2.

Based on these data, we proposed that IFN-I may be directly responsible for killing infected cardiomyocytes. First, we infected whole-body IFNAR1-deficient (Ifnar1^−/−) mice. We observed similar IFN-β concentrations in Ifnar1^−/− hearts compared to controls at day 3 p.i. (Figure 6D), although Ifnar1^−/− mice had elevated viral loads in the lung, along with altered inflammatory myeloid cell dynamics (Figure 6E–F). Despite this, the heart was protected, with similar viral loads, reduced troponin, and better LVEF following infection (Figure 6G–H). These mice were also protected following Type II MI (Figure S6C). IFN-I production was dependent upon TLR3, but not cGAS or MAVS (Figure S6D). Moreover, WT but not Ifnar1^−/− mice treated with Poly (I:C) (TLR3 agonist) accumulated leukocytes in the heart and released troponin (Figure S6E), results showing that IFN-I signaling — not active viral replication per se — is responsible for cardiac sequelae. Additionally, Poly (I:C) mimicked IAV infection by enhancing Type I MI lethality, adversely affecting LVEF and augmenting inflammation at day 3 post-MI (Figure S6F–H). CD319⁺ pro-DC3 had the highest Tlr3 and TLR3 expression (Figure S5M and Figure S6I). Although TLR3-deficiency lowered IFN-β production and protected the heart, pulmonary immune responses were compromised (Figure S6J–N).

As global prevention of TLR3-dependent IFN-β induction failed as a viable target due to its effects on pulmonary responses, we sought to limit IFN-α production by targeting IFNAR1 on infiltrating cell subsets. Intracellular flow cytometry staining for IFN-α demonstrated that CD319⁺ pro-DC3 were the cytokine’s most abundant source (Figure S6O). We generated mice lacking IFNAR1, first in the entire myeloid compartment (Lysm^{Ifnar1−/−}) and then specifically in the monocyte-like lineage (Cx3cr1^{Ifnar1−/−}). Following infection, all groups had similar IFN-β amounts. Lysm^{Ifnar1−/−} mice had decreased IFN-α, driven primarily by the CX3CR1-expressing cells (Figure S6P). Yet, while in both groups IFNAR1-sufficient littermates released more troponin, this difference was mild with no effect on LVEF (Figure S6Q–R).

Thus, as targeting IFNAR1 on leukocytes did not protect the heart, we explored silencing IFNAR1 on stromal cells, generating chimeras. Stromal Ifnar1^−/− chimera had higher LVEF compared to hematopoietic Ifnar1^−/− chimera (Figure S6S). Moreover, compared to other cardiac cells, cardiomyocytes expressed higher Ifnar1, along with the ISG Ifit1, signifying more active IFN-I signaling (Figure S6T). Therefore, as a final strategy, we targeted Ifnar1 specifically on cardiomyocytes through αMHC-cre (CM^{Ifnar1−/−}). After infection, these mice had reduced troponin, better LVEF, and no off-target effects (Figure 6I–M). Taken together, genetic elimination of IFNAR1 on cardiomyocytes protects the heart.

Finally, to therapeutically target the IFN-I/IFNAR1 pathway in cardiomyocytes, we settled on two strategies. The first strategy involved genetic deletion of IFNAR1 only on cardiomyocytes via CRISPR/Cas9 targeting (Figure 6N and Figure S6U–W). Following infection, mice receiving the AAV9 vector were protected from IAV-induced cardiac sequelae (Figure 6O–P). Our second strategy used mod-mRNA technology ⁵⁰ to enable brief expression of a dominant-negative truncated IFNAR1 protein that sequesters IFN-I signaling and ISG production (Figure 6Q and Figure S6X). BMDM mod-mRNA transfection provided proof-of-concept (Figure S6Y–Z). In vivo, mice that received tIfnar1 mod-mRNA into the LV 2 hours prior to infection had a reduced ISG signature in their hearts without affecting Ifnb (Figure 6Q). Moreover, the hearts of these mice were protected (Figure 6R–S). Our results demonstrate that targeting IFNAR1 signaling specifically in cardiomyocytes is feasible and effective in preventing IAV-induced cardiac damage.

Discussion

In this study we have provided evidence that influenza damages the heart directly by exploiting circulating myeloid cells, replicating in cardiomyocytes, and engaging the IFN-I program. By applying various complementary approaches, we illustrated that pro-DC3 were the primary infected cells. Experiments in CCR2-deficient mice revealed only traces of the virus in the heart, thus suggesting that CCR2^high pro-DC3 are the essential vectors for viral cardiac seeding. By using computational analyses to infer cell-to-cell communication between myeloid cells and cardiomyocytes; confocal microscopy to visualize morphology and contact between myeloid cells and cardiomyocytes in the intact cardiac microenvironment; and direct cell transfer experiments, we showed that infected pro-DC3 can readily infect cardiomyocytes. As the virus seeds the heart via pro-DC3 and replicates productively in cardiomyocytes, it triggers an anti-viral host response dominated by IFN-I that kills cardiomyocytes and impairs heart function.

The hijacked myeloid cell that delivers influenza to the heart is poorly understood. Careful fate mapping studies have shown that circulating monocytes derive from GMP ^{51, 52}. After MI, GMP-dependent Ly6C^hi monocytes preferentially accumulate in the heart, where they participate in both inflammatory and reparative phases. Conversely, many cardiac-infiltrating cells during influenza infection (CD319⁺ GMP-independent) represent a tiny proportion of the circulating Ly6C^hi myeloid population, and, as our snRNAseq data show, resembles the recently described mouse and human-conserved, pro-DC3 ²⁵. The cells were monocyte-like insofar as they fell within monocyte gating strategies routinely performed and refined over the last 20 years; because they are GMP-independent, however, they fell outside expected monocyte ontogeny. Therefore, given their expression of well-documented DC markers, and pro-DC3 markers in particular, we elected to refer to these influenza-avid myeloid cells as pro-DC3. As cells expressing surface CCR2 at high concentrations, pro-DC3 had a competitive recruitment advantage to an organ that releases abundant CCL2. Thus, by hijacking the specialized CCR2^hi monocyte-like pro-DC3, the virus preferentially accumulated in the heart not necessarily because of intrinsic cardiotropism, but perhaps because the heart, for reasons that remain unclear, is a CCL2 super-producer.

In this study, we also investigated the impact of influenza in humans, analyzing autopsy tissue obtained from patients who died from influenza as well as plasma from hospitalized patients. In the former, we note abundant co-morbidities, the most numerous of which were cardiovascular. In the latter, we found evidence of cardiac damage that was more pronounced than in patients with ASUC, bolstering the idea that heart damage is not simply the consequence of non-specific peripheral inflammation. In aggregate, the data strengthen the correlation between influenza and cardiovascular disease. More concretely, the data suggest that the most severe manifestations of influenza infection occur in people with pre-existing cardiovascular disease, and indeed our mouse data concur. That said, the damage to the heart is not, strictly speaking, the result of infection; even very high viral titers induce low and transient cardiac infection that peaks before symptoms occur. Rather, it is the festering anti-viral IFN-I host response against cardiomyocytes that appears to be essential. Although IFN-I is necessary for controlling viral infection in the lung ^{53, 54}, in the heart it is maladaptive ⁵⁵, and such dysfunction may be exacerbated in the context of co-morbidities, thereby contributing to a feedforward loop that further damages the organ.

One implication of our study concerns the timing of a potential therapy. Given that the virus is largely cleared from the heart before symptoms are felt, any intervention directly targeting the virus is likely to fail. Yet, given the smoldering IFN-I ISG program, specific intervention targeting IFNAR1 on cardiomyocytes might work, as we demonstrated here in proof-of-concept experiments. Furthermore, as IFN-I is broadly anti-viral, our study’s insights may extend far beyond influenza. Moreover, we have shown that vaccination, even if only partially protective in preventing overt morbidity with a mismatched virus, protects from heart damage, thus extending the benefits of vaccination outside of the well-documented protection of the lung. In sum, we have shown that influenza hijacks a specific monocyte-like pro-DC3, seeds the heart, infects cardiomyocytes, and induces IFN-I production that damages the heart. Harnessing this pathway has the potential to minimize viral heart damage exacerbated by cardiovascular disease.

Limitations of the study:

Our study provides insights into influenza pathogenesis, myeloid cell biology, and the IFN-I program. However, several unanswered questions remain. 1) Influenza Pathogenesis: How exactly IAV is transferred from monocytes to cardiomyocytes is unclear. IAV may tunnel through strengthened interactions between CD319⁺ pro-DC3 and cardiomyocytes ⁴⁷ or perhaps death of CD319⁺ pro-DC3 causes endocytosis by cardiomyocytes. Also, we do not know what protective role, if any, these pro-DC3 play in the anti-IAV response. 2) Myeloid Ontogeny and Turnover: We cannot currently explain why the heart attracts pro-DC3. It is appealing to speculate that the considerable mechanical and energetic burden of the heart mandates continuous supply of these circulating myeloid cells, but further study is required. 3) Clinical Translatability: Additional work is required to optimize mod-mRNA delivery post-infection as symptoms are experienced, rather than prophylactically. Moreover, we employed intracardiac administration, which is invasive. Intravenous administration of cardiomyocyte-targeting SMRT mod-mRNA ⁵⁶ is an ongoing area of research.

Resource Availability

Lead Contact

Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, Filip K. Swirski (filip.swirski@mssm.edu).

Materials availability

This study did not generate unique reagents.

Data and code availability

Single-nuclear RNA-seq data have been deposited at GEO and are publicly available as of the date of publication. The accession number is listed in the key resources table.

Key resources table.

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
BUV737 Mouse Anti-Mouse CD45.2	BD Biosciences	612779
APC Mouse Anti-Mouse CD45.1	Biolegend	110714
BUV805 Rat Anti-CD11b	BD Biosciences	568345
BV785 Rat Anti-Mouse Ly-6C	Biolegend	128041
PerCP-eFluor710 Rat Anti-Mouse Ly-6G	eBioscience	46-9668-82
PE/Cy7 Rat Anti-Mouse CD319	Biolegend	118804
APC Rat Anti-Mouse CD319	Biolegend	152004
AlexaFluor 647 Rat Anti-Mouse CD177	BD Biosciences	566599
BUV661 Rat Anti-Mouse CD19	BD Biosciences	612971
BV605 Rat Anti-Mouse CD5	BD Biosciences	563194
APC/Fire 810 Mouse Anti-Mouse CX3CR1	Biolegend	149054
PE/Dazzle 594 Mouse Anti-Mouse CD64	Biolegend	139320
BB700 Rat Anti-Mouse Tim4	BD Biosciences	745945
eFluor450 Rat Anti-Mouse LYVE1	Invitrogen	48-0443-82
BV650 Rat Anti-Mouse CCR2	Biolegend	150613
BUV496 Rat Anti-Mouse CD115	BD Biosciences	749974
APC/Cy7 Armenian Hamster Anti-Mouse CD11c	Biolegend	117324
BV480 Rat Anti-Mouse CD117	BD Biosciences	746780
APC/Cy7 Rat Anti-Mouse Ly-6A/E	Biolegend	108126
BV650 Rat Anti-Mouse CD150	Biolegend	115932
Alexa Fluor 700 Armenian Hamster Anti-Mouse CD48	Biolegend	103426
BV421 Rat Anti-Mouse CD135	BD Biosciences	562898
FITC Rat Anti-Mouse CD34	BD Biosciences	553733
PerCP/Cyanine5.5 Rat Anti-Mouse CD16/32	Biolegend	101324
BUV805 Rat Anti-Mouse CD31	BD Biosciences	1741949
APC Feeder Cells Antibody Anti-Mouse	Miltenyi Biotec	130-12-802
PE Rat Anti-Mouse CD90.2	Biolegend	105308
PE Rat Anti-Mouse B220	Biolegend	103208
PE Rat Anti-Mouse CD11b	Biolegend	101208
PE Armenian Hamster Anti-Mouse CD11c	Biolegend	117308
PE Rat Anti-Mouse CD335	Biolegend	137604
PE Rat Anti-Mouse TER-119	Biolegend	116208
APC Rat Anti-Mouse TLR3	Biolegend	141906
FITC Mouse Anti-Influenza A Virus NP	Abcam	Ab210526
eFluor506 Fixable Viability Dye	eBioscience	65-0866-14
FITC Rat Anti-Mouse IFN-alpha	PBL Assay Science	22100-3
PE Mouse Anti-Mouse anti-STAT1	Biolegend	6866404
BV711 Rat Anti-Mouse CD45	Biolegend	103147
PE-Cy7 Mouse Anti-Mouse Ki-67	BD Biosciences	561283
Rat Anti-Mouse CD68	BioRad	MCA1957GA
Rat Anti-Mouse Laminin	Invitrogen	MA1-06100
Alexa Fluor 647 Wheat Germ Agglutinin	Invitrogen	W32466
Goat anti-Rat IgG Texas Red	ThermoFisher	T-6392
Chicken anti-Rat AlexaFluor 647	ThermoFisher	A-21472
Rabbit Anti-Mouse Laminin	Sigma	L9393
Goat Anti-Rat AlexaFluor 488	ThermoFisher	A-11006
Lycopersicon Esculentum (Tomato) Lectin (LEL, TL), DyLight 488	ThermoFisher	L32470
APC Rat Anti-Mouse CD135	BD Biosciences	560718
BV570 Mouse Anti-Mouse XCR1	Biolegend	148247
BUV615 Rat Anti-Mouse CD45RB	BD Biosciences	751202
BV750 Rat Anti-Mouse CD172a	BD Biosciences	747007
PE Mouse Anti-Mouse CD209a	Biolegend	833004
AlexaFluor 700 Rat Anti-Mouse I-A/I-E	Biolegend	107622
Rabbit anti-Mouse Alpha-actinin	Invitrogen	701914
Donkey anti-rabbit Alex Fluor 488	Invitrogen	A32790
Anti-tdTomato	Origene	TA150129
Chicken anti-Rat Alexa Fluor 647	Invitrogen	A21472
Bacterial and virus strains
Influenza A/Puerto Rico/8/34 H1N1	Schotsaert Lab	N/A
Influenza A/New Caledonia/20/1999 H1N1	Schotsaert Lab	N/A
Influenza A/Puerto Rico/8/34 PB2-Cre H1N1	Palese Lab	N/A
Biological samples
Chemicals, peptides, and recombinant proteins
Isoproterenol	Sigma	420355-100MG
FluZone Vaccine	BEI	NR-10478
Pioglitazone	Sigma	E6910-10MG
Poly I:C (LMW)	InvivoGen	Tlr-picw
IFN-Beta	R&D Systems	8234-MB-010/CF
Hoechst 33342	Invitrogen	H3570
Critical commercial assays
PE Annexin V Apoptosis Detection Kit	BD Biosciences	559763
BD Cytofix/Cytoperm Kit	BD Biosciences	554714
Foxp3 Transcription Factor Staining Buffer Set	eBioscience	00-5523-00
MitoTracker Green FM	ThermoFisher	M7514
jetMESSENGer mRNA Transfection Reagent	Polyplus	101000005
EasySep Mouse Monocyte Isolation Kit	StemCell Technologies	19861
Primary Cardiomyocyte Isolation Kit	Pierce	88281
Mouse Cardiac Troponin I Type 3 ELISA	Novus Biologics	NBP3-00456
Mouse NT-proBNP ELISA	Novus Biologics	NBP2-76775
Mouse CCL2 DuoSet ELISA	R&D Systems	DY479-05
Mouse SERPINE1 DuoSet ELISA	R&D Systems	DY3828-05
Mouse IFN-Beta ELISA High Sensitivity	PBL Assay Science	42410-1
Mouse IFN-Alpha ELISA High Sensitivity	PBL Assay Science	42115-1
Proteome Profiler Mouse XL Cytokine Array	R&D Systems	ARY028
Human Cardiac Troponin 1 Type 3 ELISA	Abcam	Ab200016
Human NT-proBNP ELISA	Abcam	Ab263877
Human CCL2 ELISA	Invitrogen	BMS281
Hydroxyproline Assay Kit	Sigma	MAK569
CytoTox 96 Non-Radioactive Cytotoxicity	Promega	G1780
Click-iT Plus TUNEL Assay Kits	ThermoFisher	CD10617
Pre-Filled 2.0mL Tubes, Zirconium Homogenizer Beads, 3.0mm Triple-Pure - High Impact	Stellar Scientific	BS-BEBU-230
RNeasy Mini Kit	Qiagen	74104
RNeasy Micro Kit	Qiagen	74004
High-Capacity cDNA Reverse Transcription Kit	ThermoFisher	4368814
PowerUp SYBR Green Master Mix	Applied Biosystems	A25742
TaqMan Fast Advanced Master Mix	Applied Biosystems	4444557
Deposited data
snRNA-seq	Gene Expression Omnibus	GSE312462
Experimental models: Cell lines
MDCK	ATCC	CCL-34
B16-Blue IFN-a/b Cells	InvivoGen	bb-ifnt1
L929	ATCC	CCL-1
Experimental models: Organisms/strains
C57BL/6J	The Jackson Laboratory	000664
BALB/cJ	The Jackson Laboratory	000651
B6129SF2/J	The Jackson Laboratory	101045
B6;129S1-Tlr3tm1Flv/J	The Jackson Laboratory	005217
B6.129S7-Ldlrtm1Her/J	The Jackson Laboratory	002207
B6.SJL-Ptprca Pepcb/BoyJ	The Jackson Laboratory	002014
C57BL/6J-Ms4a3em2(cre)Fgnx/J	The Jackson Laboratory	036382
B6.129S4-Ccr2tm1Ifc/J	The Jackson Laboratory	004999
B6.129(Cg)-Ccr2tm2.1Ifc/J	The Jackson Laboratory	017586
B6.Cg-Gt(ROSA)26Sortm14(CAG-tdTomato)Hze/J	The Jackson Laboratory	007914
B6(Cg)-Ifnar1tm1.1Ees/J	The Jackson Laboratory	028256
B6(Cg)-Ifnar1tm1.2Ees/J	The Jackson Laboratory	028288
B6.FVB-Tg(Myh6-cre)2182Mds/J	The Jackson Laboratory	011038
B6J.B6N(Cg)-Cx3cr1tm1.1(cre)Jung/J	The Jackson Laboratory	025524
B6.129P2-Lyz2tm1(cre)Ifo/J	The Jackson Laboratory	004781
B6(C)-Cgastm1d(EUCOMM)Hmgu/J	The Jackson Laboratory	026554
B6.129P2(Cg)-Cx3cr1tm1Litt/J	The Jackson Laboratory	005582
B6.Cg-Zbtb46tm3.1(cre)Mnz/J	The Jackson Laboratory	028538
B6;129-Mavstm1Zjc/J	The Jackson Laboratory	008634
Oligonucleotides
Mouse Actb (Mm00607939_s1)	ThermoFisher
Mouse Ifnb1 (Mm00439552_s1)	ThermoFisher
Mouse Irf7 (Mm00516793_g1)	ThermoFisher
Mouse Ifitm3 (Mm00847057_s1)	ThermoFisher
Mouse Ifit1 (Mm00515153_m1)	ThermoFisher
Mouse Ifnar1 (Mm00439544_m1)	ThermoFisher
Mouse Ccl2 (Mm00441242_m1)	ThermoFisher
IAV Ns1 (Forward: 5’-AGAAAGTGGVAGGCCCTCTTTGTA-3’ and Reverse: 5’-GGGCACGGTGAGCGTGAACA-3’)	ThermoFisher
Gapdh (Forward: 5’-GGTCCTCAGTGTAGCCCAAG-3’ and Reverse: 5’-AATGTGTCCGTCGTGGATCT-3’)	ThermoFisher
Recombinant DNA
Software and algorithms
GraphPad Prismv10	GraphPad Software	N/A
FlowJo v10	FlowJo	N/A
BioRender	BioRender	N/A
Imaris	Oxford Instruments	N/A
ImageJ	ImageJ	N/A
Other
High Cholesterol Diet	Research Diets	D12331

Star Methods

Experimental models and subject details

Humans

Autopsies

Unrestricted autopsies in the Mount Sinai Hospital System (spanning from 2004–2024) were screened for note of death accompanied by influenza infection (both Influenza A and B strains were included). Autopsies were performed by a certified blinded pathologist not involved in this study. Relevant IV⁺ autopsy reports were pulled and evidence of various co-morbidities was compiled by researchers (Figure 1A) and assigned a score, based on severity of each comorbidity, with higher scores falling towards the outside of the radar plots. Left ventricle tissue was collected during autopsy of all individuals, a subset were then stained with Trichrome Stain, Hematoxylin and Eosin, and/or Influenza A NP (Invitrogen #PA5–32242), by standard staining protocols by the Biorepository and Pathology Core of the Icahn School of Medicine at Mount Sinai.

Plasma Collection

Individuals presenting at the hospital with laboratory confirmed cases of IV infection (either A or B), those hospitalized with Acute Severe Ulcerative Colitis (ASUC), or healthy volunteers had their peripheral blood drawn into an EDTA tube from an antecubital vein by a trained phlebotomist, as part of standard care. Samples were de-identified. Whole blood samples were spun down (10 minutes at 4°C and 12 000 RCF) and the top layer of plasma was collected for use in subsequent assays.

Mice

All experiments were performed on sex- and age-matched mice between the ages of 5 and 12 weeks, except where indicated, such as in Figure 1I where aged mice were used. C57BL/6J, BALB/cJ, B6129SF2/J, B6;129S1-Tlr3^tm1Flv/J, B6.129S7-Ldlr^tm1Her/J, B6.SJL-Ptprc^a Pepc^b/BoyJ, C57BL/6J-Ms4a3^em2(cre)Fgnx/J, B6.129S4-Ccr2^tm1Ifc/J, B6.129(Cg)-Ccr2^tm2.1Ifc/J, B6.Cg-Gt(ROSA)26Sor^{tm14(CAG-tdTomato)Hze}/J, B6(Cg)-Ifnar1^tm1.1Ees/J, B6(Cg)-Ifnar1^tm1.2Ees/J, B6.FVB-Tg(Myh6-cre)2182Mds/J, B6J.B6N(Cg)-Cx3cr1^{tm1.1(cre)Jung}/J, B6.129P2-Lyz2^tm1(cre)Ifo/J, B6(C)-Cgas^{tm1d(EUCOMM)Hmgu}/J, B6.129P2(Cg)-Cx3cr1^tm1Litt/J, B6.Cg-Zbtb46tm3.1(cre)Mnz/J, and B6;129-Mavs^tm1Zjc/J were purchased from The Jackson Laboratory. Genotyping for each strain was performed as described on the Jackson Laboratory website. All mice were housed on a 12 h–12 h light–dark cycle at 22 °C with a humidity range of 30–70% and unrestricted access to food and water. Where appropriate, mice were randomly assigned to groups and experiments were performed in a blinded manner. All protocols were approved by the Animal Review Committee at the Icahn School of Medicine at Mount Sinai (protocols: IPROTO202200000097, PROTO202000262, and IPROTO202200000171) and complied with all relevant ethical regulations.

Viruses & Infections

All in vivo infections were performed using mouse adapted influenza A/Puerto Rico/8/34 (H1N1) virus (IAV), influenza A/New Caledonia/20/1999(H1N1), or influenza A/Puerto Rico/8/34 PB2-Cre (H1N1), kindly provided by Dr. Michael Schotsaert and Dr. Peter Palese (Icahn School of Medicine at Mount Sinai). Generally, mice were challenged intranasally (in 25μL of sterile PBS) with IAV at a sublethal dose of 35 PFU. For dosage experiments, like those in Figure 1F, inocula were as indicated and ranged from 15 to 350 000 PFU. Infections during vaccination experiments with New Caledonia used an initial inoculum of 1000 PFU. Infections with the PB2-Cre virus were with 5000 PFU. During experiments with lethal doses mice were monitored twice daily for signs of duress and weighed daily. Any mouse reaching 75% of original body weight was considered moribund and sacrificed. Viruses were propagated in eggs and titered in Madin-Darby Canine Kidney (MDCK) cells, using standard MDCK plaque assays ⁵⁷. Unknown viral titers were determined in organ homogenates (homogenized in 500 μL PBS) using standard MDCK plaque assays.

For in vitro infections, cells were seeded in tissue culture plates the day before infection, unless otherwise indicated, and infections were performed in fresh medium at a 0.25–5 multiplicity of infection (MOI) of virus, as detailed in the text or figure legends. Briefly, cells were incubated with IAV in serum-free media containing 0.1% bovine serum albumin (BSA) for 1 h, then washed with PBS and incubated with media containing 0.1% BSA and TPCK trypsin (2 μg ml⁻¹) for the indicated time periods. Viral titers were determined in cell culture supernatants as before.

Method Details

In vivo interventions, measurements, and mouse models

Type I Myocardial Infarction

Surgical Type I MI was induced by permanent ligation of the left anterior descending coronary artery (LAD) as previously described ^14–16. In brief, mice were anesthetized with isoflurane (2%/2 l O2), intubated and ventilated with an Inspira Advanced Safety Single Animal Pressure/Volume Controlled Ventilator (Harvard Apparatus). Left thoracotomy was performed in the fourth intercostal space after shaving the chest wall. The left ventricle was visualized, and the LAD was ligated with siliconized 7–0 silk suture (Ethicon) 1 mm from the Auricle inducing a “large” infarction. The chest and skin were closed with a 7–0 nylon and 5–0 proline sutures, respectively, followed by removal of air from the thorax via a pleural catheter. For sham surgery, mice underwent the exact same procedure, including anaesthesia, analgesics, thoracotomy, intubation, ventilation, and visceral cavity opening, however upon visualization of the LAD it was not ligated. All mice received 3.25 mg/kg Ethiqa XR after surgery. The procedure was performed during the light period beginning at ZT 0 by the same surgeon blinded to genotypes and experimental conditions. In some experiments, 16 hours prior to MI, mice were administered 35 PFU of IAV to constitute the infected/infarcted group or 100μg of Poly I:C. Mice were then sacrificed for downstream analyses at 3 days later for phenotypic, 21 days later for cardiac function, or monitored for survival

Type II Myocardial Infarction

Pharmacological Type II MI was induced through a single intraperitoneal administration of the β-adrenergic receptor agonist isoproterenol as previously described ¹⁷. Mice were treated with isoproterenol HCL (Sigma-Aldrich) dissolved in PBS at a dose of 160mg/kg. Injection volume was 100μL. Control mice received a similar injection composed of only PBS. In some experiments, mice were infected with 35 PFU 6 days prior to Type II MI. This group constituted the IAV + ISO group.

Murine model of vaccination

Mice were vaccinated intramuscularly in the quadriceps muscles with either 1.5 or 2μg hemagglutinin equivalent of FluZone vaccine (2005–2006 season, BEI NR-10478), depending on whether they were subsequently infected with heterotypic PR8 (suboptimal) or vaccine-matched A/New Caledonia/20/1999 (optimal) IAV, respectively. The FluZone vaccine was diluted to the appropriate concentration in PBS and unvaccinated controls received a PBS vehicle injection. 3 weeks later, mice were bled and anti-IAV antibody titers were determined to ensure efficacy of vaccination, prior to infection. Mice were then infected and monitored for morbidity, troponin release, and cardiac function at the indicated time points.

Atherosclerosis Co-morbidity Model

Ldlr^−/− were fed a high-cholesterol diet (HCD, Research Diets, D12331) for 16 weeks to induce pronounced atherosclerotic plaques prior to infection. Control mice were fed normal chow prior to infection. Mice were then infected with IAV and sacrificed 3 days later for inflammatory response analysis or at 9 days were subject to cardiac function tests.

Generation of chimeric mice

CD45.1 or Ifnar1^−/− (CD45.2) mice were lethally irradiated with 9 Gy. 16 hours post-irradiation, the BM compartment was reconstituted with 6×10⁶ nucleated cells from either CD45.1⁺ mice (Ifnar1^−/− recipient) or Ifnar1^−/− mice (CD45.1⁺ recipient). Between 10- and 12-weeks post-injection, reconstitution was validated by flow cytometry and mice were then infected for downstream assays.

Parabiosis

The procedure was conducted as previously described ⁵⁸. Age-matched female pairs of CD45.1 and CD45.2 mice were co-housed for at least two weeks prior to surgical intervention. After shaving the corresponding lateral aspects of a CD45.1 and a CD45.2 mouse, matching skin incisions were made from behind the ear to the tail of each mouse, and the subcutaneous fascia was bluntly dissected to create about 0.5 cm of free skin. The scapulae were sutured using a mono-nylon 5.0 (Ethilon), and the dorsal and ventral skins were approximated by continuous suture. Mice were joined for 3 weeks. Percent chimerism in the blood was defined for gated leukocytes as % CD45 (%CD45.1+ & %CD45.2+) in the CD45.2 mice. Then the CD45.2 mouse was infected (infected parabiont) and the CD45.1 was left uninfected (uninfected parabiont). Mice were sacrificed at the days post-infected indicated

Pioglitazone treatment

Mice were injected every day with pioglitazone (2 mg/kg in PBS) or vehicle (PBS) intravenously, starting on the day of infection (35 PFU). On day 3 post-infection, mice were euthanized, and hearts and plasma were harvested for quantification of plasma troponin and analysis of the immune response.

Poly I:C treatment

Mice were anesthetized as for infection. LMW Poly I:C (Invivogen) was resuspended in PBS and 100μg was delivered per mouse in 25μL.

Echocardiography

Transthoracic echocardiography was performed to acquire cine images of the left ventricular short and long axes using the Vevo 2100 ultrasound system (Visualsonics, equipped with a high-frequency, 30 MHz, linear array transducer). Mice were placed supine on an electrical heating pad at 37 °C under light isoflurane anesthesia (usual maintenance level 1.0% isoflurane/99% oxygen). Continual ECG monitoring was obtained via limb electrodes. The ejection fraction (EF) was calculated as percentages from the diastolic volume (EDV) and end-systolic volume (ESV) dimensions on left ventricle traces of the long-axis CINE scans. The following formula was used: $\mathrm{\%}\mathrm{E}\mathrm{F}=(\mathrm{E}\mathrm{D}\mathrm{V}-\mathrm{E}\mathrm{S}\mathrm{V})/\mathrm{E}\mathrm{D}\mathrm{V}\times 100$.

Cardiac Magnetic Resonance Imaging

Cardiac function was assessed by cardiac MRI (cMRI) using a 7.0 Tesla small animal scanner (Bruker BioSpec 70/30) equipped with a 40-mm-diameter 1H volume coil (MR Coils). Before being placed in the MRI scanner, mice were injected i.p. with 0.4 mL/kg gadolinium contrast (Gadovist, Bayer) and anesthetized with isoflurane (4% for induction, 1% for maintenance). After survey scans to plan a cardiac short-axis orientation, seven short-axis slices covering the heart from apex to base were scanned with a self-gated CINE 2D FLASH sequence. Sequence parameters were: TR = 10.5 ms, TE = 1.9 ms, flip-angle = 20o, field-of-view (FOV) = 30 × 30 mm2, slice thickness = 1 mm, acquisition matrix = 192 × 192, reconstruction matrix = 256 × 256, number of cardiac frames = 12, acquisition time = 4 min and 13 s per slice.

Heart Weight-to-Tibia Length

Hearts were carefully excised from naive controls or IAV-infected mice and weighed (heart weight). Tibiae were then removed and measured (tibia length). Data are presented as the ratio of heart weight over tibia length.

Colon Length

Entire colons were excised from healthy or post-IAV infected mice from the cecum to distal colon. Lengths were determined from the base of the cecum to the end of the distal colon as a marker for intestinal damage.

Markers of Kidney and Liver Damage

At indicated time points post-infection, mice were sacrificed and bled. Whole blood samples were incubated at room temperature for 30 minutes to allow for complete coagulation. Coagulated blood was then spun down, and the top serum layer was collected. Samples were then submitted to the Comparative Pathology Core at the Icahn School of Medicine for quantification of Albumin, BUN, ALP and Creatine Kinase in the sera.

Wet-to-dry ratio

Lungs were harvested from naive or IAV-infected mice at various time points post-infection, and blood clots were carefully removed. Then, the lungs were weighed (wet weight) and dried in an oven (56°C, 2 days; dry weight), and the dry weight was measured. Data are presented as the ratio of wet weight to dry weight.

IFNAR1-Targeting Therapies

Generation of AAV-cTnT-Cre-U6-gRNA plasmid

The AAV-cTnT-Cre-U6-[BsmBI_entry]-SpCas9_gRNA (MLE80) gRNA entry plasmid was generated by subcloning the cTnT promoter from pAAV-cTnT-iCre (Addgene plasmid 69916) and the Cre recombinase coding sequence from ITR-U6-sgRNA(backbone)-hSyn-Cre-2A-EGFP-KASH-WPRE-shortPA-ITR (Addgene plasmid 60231) into the XbaI and BglII sites of Cbh_v5 AAV-ABE C-terminal (Addgene plasmid 137178). Potential gRNAs targeting the mouse Ifnar1 gene were selected for gene knockout by comparing the highest ranked target sites from the genome-wide Brie library based on ‘Rule Set 2 ’scores ⁵⁹, and from CRISPick ⁶⁰ using the top ‘Pick Order ’gRNAs, resulting in the top gRNA spacer sequence of GATGGCGGTGACAGTAATGAC that encodes an appended 5’G for efficient transcription from the U6 promoter. The final AAV-Cre-gRNA expression plasmid was generated by digesting plasmid AAV-cTnT-Cre-U6-[BsmBI_entry]-SpCas9_gRNA (MLE80) with BsmBI (New England Biolabs), followed by annealing and ligating duplexed oligonucleotides corresponding to the spacer sequence into the digested backbone. The resulting plasmid AAV-cTnT-Cre-U6-SpCas9-gRNA-Ifnar1 (MLE105) was used as a template for AAV9 vector production (PackGene Biotech).

modRNA synthesis

All modRNA was generated by in vitro transcription of plasmid templates (GeneArt, Thermo Fisher Scientific). The open reading frame sequences used to make nGFP modRNA is: atggtgagcaagggcgaggagctgttcaccggggtggtgcccatcctggtcgagctggacggcgacgtaaacggccacaagttcagcgtgtccggcgagggcgagggcgatgccacctacggcaagctgaccctgaagttcatctgcaccaccggcaagctgcccgtgccctggcccaccctcgtgaccaccctgacctacggcgtgcagtgcttcagccgctaccccgaccacatgaagcagcacgacttcttcaagtccgccatgcccgaaggctacgtccaggagcgcaccatcttcttcaaggacgacggcaactacaagacccgcgccgaggtgaagttcgagggcgacaccctggtgaaccgcatcgagctgaagggcatcgacttcaaggaggacggcaacatcctggggcacaagctggagtacaactacaacagccacaacgtctatatcatggccgacaagcagaagaacggcatcaaggtgaacttcaagatccgccacaacatcgaggacggcagcgtgcagctcgccgaccactaccagcagaacacccccatcggcgacggccccgtgctgctgcccgacaaccactacctgagcacccagtccgccctgagcaaagaccccaacgagaagcgcgatcacatggtcctgctggagttcgtgaccgccgccgggatcactctcggcatggacgagctgtacaagggagatccaaaaaagaagagaaaggtaggcgatccaaaaaagaagagaaaggtaggtgatccaaaaaagaagagaaaggtataa and the DN-IFNAR1 sequence is atgctcgctgtcgtgggcgcggcggccctggtgctggtggccggggcgccttgggtgctaccctcagctgcaggtggagaaaatctgaaacctcctgagaatatagacgtctacattatagatgacaactacaccctaaagtggagcagccacggagagtcaatgggcagtgtgaccttttcagcagaatatcgaacaaaagacgaggcgaagtggttaaaagtgcctgaatgtcaacatactacaacgaccaagtgtgaattctctttactggacacaaatgtgtatatcaaaacacagtttcgtgtcagagcagaggaagggaacagcacatcttcgtggaatgaggttgatccgtttattccattctacacagctcacatgagccccccagaagtacgtttagaagctgaagataaagccatactagtccacatctctcctcccggacaagacgggaacatgtgggcactggagaaaccttccttcagttacaccatacgaatctggcagaagtcttccagtgacaaaaaaactattaactctacgtattatgtagaaaagataccagaactcttgccagagactacttactgtttagaagttaaagcaatacatccgtcacttaagaaacacagcaattacagcactgtgcagtgtataagcaccacagtggcaaataaaatgcctgtgccaggaaatctccaagtggatgcccaaggcaagagctatgtcctgaaatgggactacattgcgtctgcagacgtgctcttcagggcacagtggcttcctggctattcaaaaagcagttctggaagccgttcagataaatggaaaccaataccaacctgtgcaaatgtccagactacgcactgtgtcttttctcaagatactgtctacacaggaacgttctttctccatgtacaagcctcagagggaaatcacacatccttttggtctgaagagaagtttattgattctcaaaaacacattctccctcctcctccggtcattactgtcaccgccatgagtgacaccttgcttgtttatgtcaactgtcaggacagcacatgtgatggactcaattacgaaatcatcttttgggaaaacacttccaatactaagataagcatggagaaggatggcccagagttcaccctcaagaacctgcagccgctgactgtgtactgtgtccaggccagagtgctcttcagggccctgctgaataagaccagcaacttcagtgaaaagctgtgtgagaaaacacgtccaggaagtttttccacgatctggattataactggattaggtgttgtgttcttctctgtcatggtcctttatgctttgaggagcgtctggaaatacctgtgtcatgtgtgcttcccaccactcaagcctccccgcagtattgatgagtttttctctgagccgccttcaaaaaacctttga.The transcription step involved a customized ribonucleotide blend of anti-reverse cap analog; 30-O-Me-m7G(50) ppp(50)G (6 mM, TriLink Biotechnologies); guanosine triphosphate (1.5 mM, Life Technologies); adenosine triphosphate (7.5 mM, Life Technologies); cytidine triphosphate (7.5 mM, Life Technologies) and N1-methylpseudouridine-5-triphosphate (7.5 mM, TriLink Biotechnologies). Next, modRNA was purified with the Megaclear kit (Life Technologies) and treated with Antarctic Phosphatase (New England Biolabs). To eliminate any remaining impurities, modRNA was re-purified with the Megaclear kit and quantified using a Nanodrop spectrometer (Thermo Scientific). Lastly, modRNA was precipitated with ethanol and ammonium acetate and resuspended in 10 mM Tris-HCl and 1 mM EDTA.

modRNA transfection and IFN-β stimulation

BMDM were differentiated and seeded into 12 well plates at 500 000 cells/well as described above. The next day they were transfected with either control GFP modRNA or tIFNAR1 modRNA using jetMESSENGER mRNA Transfection Reagent (Polyplus), according to the instructions of the manufacturer. BMDM were transfected with 1μg of tIFNAR1 modRNA or 1μg of control GFP modRNA. In some experiments, 16 hours following transfection, BMDM were stimulated with recombinant murine IFN-β (PBL Assay) to assess activation of pSTAT1 versus PBS stimulated controls. BMDM were stimulated with 1000U/mL of IFN-β in PBS for 20 minutes. Cells were then washed, stripped and stained for pSTAT1 as detailed in the flow cytometry section.

Intracardiac injection of mRNA and AAV

Mice were prepared, anesthetized and intubated as during the Type I MI. The pericardial sac was removed, and the left ventricle was visualized as before. Injections were performed at three sites of the left ventricle totaling 10μL. Two sites were on the left and right side of the LAD and the third was in the middle of the left ventricle below the LAD branch. AAV9 injections were delivered in PBS at a concentration of 1×10¹² GC/kg. The experimental mice that received injections were Rosa26-LSL-Cas9 knock in on B6J mice (Jackson Laboratories # 026175), while B6 WT mice with no Cas9 activity injected with an equal dose of AAV9 served as controls. Injection into Rosa26-LSL-Cas9 knock in allowed for deletion of Ifnar1 and GFP expression specifically in cardiomyocytes of the left ventricle as shown in the figure. modRNA was delivered using the same strategy as AAV9 using instead a sucrose citrate buffer, containing sucrose in nuclease-free water (0.3g/ml) and citrate (0.1M pH=7; Sigma). 75μg of tIfnar1 or GFP modRNA was delivered into WT mice.

Cell Based Assays

Organ Collection and Single Cell Suspension Processing

Blood samples were collected by retro-orbital bleeding in tubes containing EDTA. Bone marrow cells were collected from the left femur by removing the epiphyses and flushing with 5mL of PBS. Marrow was pipetted up and down gently to generate a single cell suspension. Lung, heart, thymus, white adipose tissue, aorta, liver, kidney, pancreas, brain, skeletal muscle and stomach were excised, minced and digested for 45 minutes at 37°C on a shaker set to 1000 RPM in a cocktail containing 450 U/mL collagenase I, 125 U/mL collagenase XI, 60 U/mL DNase I and 60 U/mL hyaluronidase (Sigma) in PBS. Organs were then crushed through a 100μm filter and washed. Brain samples were alternatively passed through a 70-μm cell strainer and mixed with 30% Percoll layered on top of 70% Percoll. The Percoll gradient was centrifuged at 500g for 30 min with the brake off. The cell fraction was collected and washed with PBS. Peritoneal lavage cells were collected as described above. Spleens and mediastinal and inguinal lymph nodes were collected and crushed through a 100μm filter and washed. Small and large intestinal lamina propria were isolated as follows: after excision of the intestine, the Peyer’s patches were removed under a microscope, and the gut was cut open longitudinally to wash off the lumen contents in HBSS buffer. The gut was then cut into 1–2-cm pieces and subjected to 3× dissociation in EDTA-containing buffer (7.5 mM HEPES, 2% FCS, 2 mM EDTA, 10,000 U ml−1 penicillin–streptomycin, 50 μg ml−1 gentamycin in HBSS; all Thermo Fisher Scientific) in a shaker at 37 °C for 15 min. After dissociation the epithelial layer was isolated (by filtering through a mesh) and discarded and the lamina propria digested in a mixture of 11M CaCl2 (Sigma), 60 U/ml Dnase I, 31.25 μg/mL (for SI) or 62.5 μg/mL (for LI) Liberase (Roche), RPMI with L-Glutamine (Corning), 10% FCS, 5 mM Sodium pyruvate (Corning cellgro), 5 mM HEPES (1M), 1% Pen/Strep at 37°C for 40min. In all organs, erythrocytes were then lysed with RBC lysis buffer (Biolegend), washed in PBS and resuspended in MACS Buffer (0.5% BSA and 2 mM EDTA in PBS) for staining.

Flow Cytometry

Single cell suspension samples were stained in MACS buffer with antibodies at a dilution of 1:700 except where noted for 20 minutes at 4°C. Antibodies used against extracellular antigens (all from BioLegend except where indicated) were as follows: CD45.2 BUV737 (BD Biosciences), CD45.1 APC, CD11b BUV805 (BD BioSciences), Ly6C BV785, Ly6G PerCP-eFluor710 (eBioscience), CD319 PE-Cy7, CD177 AlexaFluor 647 (BD Biosciences), CD19 BUV661 (BD Biosciences), CD5 BV605 (BD Biosciences), CX3CR1 APC-Fire810, CD64 PE-Dazzle594, Tim4 BB700 (BD Biosciences), Lyve1 eFluor450, CCR2 BV650 (1:250), CD115 BUV496 (BD Biosciences), CD11c APC-Cy7, cKit BV480 (BD Biosciences), Sca-1 APC-Cy7, CD150 BV650, CD48 AlexaFluor-700, CD135 BV421 (1:350; BD Biosciences), CD34 FITC (1:350; BD Biosciences), CD16/32 PerCP-Cy5.5, CD31 BUV805, MEFSK4 APC (Miltenyi Biotec 1:100), XCR1 BV570, CCR7 BV605, CD172a BV750 (BD Biosciences), CD209 PE, or CD45RB BUV615 (BD Biosciences). Lineage in the bone marrow was CD90.2, B220, CD11b, CD11c, NKp46, Ter119 (all PE). Samples were then washed in PBS and differentiation between live and dead cells was done by staining with eFluor506 Fixable Viability LIVE/DEAD Dye (eBioscience) in PBS at 1:1000 for 5 minutes. Cells were then washed in MACS buffer and resuspended for acquisition or were further processed for intracellular staining against PE-Cy7-Ki67 (Biolegend) FITC-IAV nucleoprotein (NP) (Abcam), APC-TLR3 (Biolegend) or FITC-IFN-α (PBL Assay) at 1:50. For intracellular staining and analysis, cell fixation was performed with Foxp3/Transcription Factor Staining Buffer Set (eBioscience) for 30 min at 4°C. Cells were then washed with the buffer and intracellularly stained in the Foxp3 staining buffer for 60 min 4°C. In some experiments, BMDM were stained with antibodies against PE-pSTAT1 (BD Biosciences 1:100). In these, BMDM were permeabilized with ice cold methanol. Finally, in certain co-culture experiments, mitochondrial mass (100nM MitoTracker Green, ThermoFisher) was determined as suggested by the company. Cells were stained for extracellular antibodies as indicated above and washed. Then, cells were stained with the mitochondrial dye for 20 minutes at 37°C and immediately acquired. Data were collected on a Cytek Aurora Cytometer and analyzed using FlowJo. All experiments except those involving quantification of Annexin V and 7-AAD utilized viability dye that was applied to single viable cells. Negative cells were considered for further gating. Furthermore, all phenotyping experiments delineated between vascular and tissue leukocytes using the intravenous injection of a CD45 antibody prior to sacrifice (see next section) and then staining with an appropriate CD45.2 or CD45.1 antibody ex vivo as outlined in Figure 2A, except as otherwise noted.

Cell populations were defined using the following markers:

Bone Marrow

Lineage⁻ Sca1⁺ c-Kit⁺ (LSK): Lineage⁻ (B220, CD11b, CD5, CD11c, Ter119, CD49b) Sca1⁺ c-Kit⁺

Neutrophils: Lin⁺ Ly6G⁺ Ly6C^int

Ly6C^hi“ Monocytes”: Lin⁺ Ly6G⁻ Ly6C^hi CX3CR1⁺

Long-term hematopoietic stem cells (LT-HSC): LSK⁺ CD150⁺ CD48⁻

Short-term hematopoietic stem cells (ST-HSC): LSK⁺ CD150⁻ CD48⁻

Multipotent progenitor 3 (MPP3): LSK⁺ CD150⁻ CD48⁺ CD34⁺ CD135⁻

Common myeloid progenitor (CMP): Lin⁻ c-Kit⁺ CD16/32⁻ CD34⁺ CD135⁺ CD115⁻ Ly6C^lo

Granulocyte-macrophage progenitor (GMP): Lin⁻ c-Kit⁺ CD16/32⁺ CD34⁺ CD135⁻ CD115⁻ Ly6C^lo

Granulocyte progenitor (GP): Lin⁻ c-Kit⁺ CD16/32⁺ CD34⁺ CD135⁻ CD115⁻ Ly6C^hi

Common monocyte progenitor (cMoP): Lin⁻ c-Kit⁺ CD16/32⁺ CD34⁺ CD135⁻ CD115⁺ Ly6C^hi

Monocyte-dendritic cell progenitor Ly6C^hi (MDP Ly6C^hi): Lin⁻ c-Kit⁺ CD16/32⁻ CD34⁺ CD135⁺ CD115⁺ Ly6C^hi

Monocyte-dendritic cell progenitor Ly6C^loi (MDP Ly6C^lo): Lin⁻ c-Kit⁺ CD16/32⁻ CD34⁺ CD135⁺ CD115⁺ Ly6C^lo

Common dendritic cell progenitor (CDP): Lin⁻ c-Kit⁻ CD16/32⁻ CD135⁺ CD115⁺ CD34⁺ Ly6C^lo

Blood

Ly6C^lo Monocytes: CD11b⁺ Ly6G⁻ CD115⁺ Ly6C^lo

Ly6C^hi Monocytes: CD11b⁺ Ly6G⁻ CD115⁺ Ly6C^hi

Neutrophils: CD11b⁺ CD115⁻ Ly6G⁺

NK Cells: CD3⁻ CD19⁻ NK1.1⁺

B Cells: CD11b⁻ CD3⁻ CD19⁺

T Cells: CD11b⁻ CD19⁻ CD3⁺

Tissue

Neutrophils: CD11b⁺ CX3CR1⁻ Ly6G⁺

Ly6C^hi Monocytes: CD11b⁺ Ly6G⁻ CX3CR1⁺ Ly6C^hi CD64^+/− CD319^+/−

Macrophages (general): CD11b⁺ Ly6G⁻ Ly6C^hi/lo CX3CR1⁺ CD64⁺

Splenic macrophages: CD11b⁻ CD3⁻ CD19⁻ CD64⁺

Microglia (brain): CD45^int CD11b⁺ CD64⁺

Kupffer Cells (liver): CD64⁺ Tim4⁺

Heart

Neutrophils: CD11b⁺ CX3CR1⁻ Ly6G⁺

Ly6C^hi Monocytes: CD11b⁺ Ly6G⁻ CX3CR1⁺ Ly6C^hi CD64^+/− CD319^+/− CD177^+/− (as outlined in the text)

Cardiac Macrophages: CD11b⁺ Ly6G⁻ Ly6C^lo CX3CR1⁺ CD64⁺ Tim4^+/− Lyve1^+/− CCR2^+/− (as outlined in the text)

Fibroblasts: CD45⁻ MEFSK4⁺

Refined (Figure 3J)

Ly6C^hi Monocytes: CD11b⁺ CD64⁻ CD135⁻ CD45RB⁻ CX3CR1⁺ CD16/32⁺ Ly6C^hi

Ly6C^lo Monocytes: CD11b⁺ CD64⁻ CD135⁻ CD45RB⁻ CX3CR1⁺ CD16/32⁺ Ly6C^lo

pre-DC1: CD135⁺ XCR1⁻ CD172a⁻ Ly6C⁻ MHC-II⁺ CD11c⁺

DC1: CD135⁺ XCR1⁺ CD11c⁺

pre-DC2: CD135⁺ XCR1⁻ CD172a^int Ly6C^hi CD11c⁺

DC2: CD135⁺ XCR1⁻ CD172a^hi Ly6C^lo MHC-II^hi CD16/32^lo

pro-DC3: CD135⁺ XCR1⁻ CD172a^hi Ly6C^hi CD11c⁻

DC3: CD135⁺ XCR1⁻ CD172a^hi Ly6C^lo MHC-II^hi CD16/32^hi

Injection of antibodies for in vivo labeling of leukocytes and vasculature

Three to five minutes before euthanasia, mice were intravenously injected with 2 μg of a fluorescently labelled anti-CD45 BV711 (BioLegend) to bind all vascularly located leukocytes. In clearing experiments, 2 hours prior to sacrifice to allow for labeled vascular cells to extravasate into the tissue, 5 μg of CD319-APC antibody was injected. Mice were euthanized for subsequent histological analyses. In the same set of experiments, 5 minutes prior to euthanasia, 200μg of Lycopersicon Esculentum (Tomato) Lectin (LEL, TL), DyLight 488 (ThermoFisher) was injected for visualization of vessels to confirm extravasation.

Cell Sorting

Tissues were processed and stained extracellularly as indicated above. Indicated cell populations were sorted on a BD FACS Aria (BD Biosciences) into cell culture medium or RLT Lysis Buffer (Qiagen), as indicated in the text. Cell populations were sorted as follows: endothelial cells CD45⁻, CD31⁺; fibroblasts CD45⁻, MEFSK4⁺; Ly6C^hi“ Monocytes” CD45⁺, CD11b⁺, Ly6G⁻,Ly6C^hi; Neutrophils CD45⁺, CD11b⁺, Ly6G⁺; B cells CD45⁺, CD11b⁻, CD19⁺; T cells CD45⁺, CD11b⁻, CD3⁺; and cardiac macrophages CD45⁺, CD11b⁺, Ly6G⁻, Ly6C^lo, CX3CR1⁺, CD64⁺.

Adoptive Transfers

Adoptive transfer experiments proceeded as outlined in the corresponding schema and were composed of two types: in vitro and in vivo infections. In the in vitro experiments, monocytes were purified from the femurs and tibiae of at least 3 CD45.2 mice using the bone marrow purification kit (StemCell) and compiled. Cells were counted using a hemacytometer and evenly split amongst groups for infection with IAV (MOI=1) or mock infection. After 2 hours cells were spun down and washed to removed uninternalized virus. Cell concentrations were adjusted to 2.5×10⁵ monocytes/100μL of PBS and injected intravenously into otherwise naive CD45.1 hosts. 16 hours later mice were sacrificed and the listed organs were collected and analyzed by flow cytometry with differential staining of CD45.1 and CD45.2 cells.

For in vivo infection transfers a similar but modified strategy was employed. CD45.2 mice were infected intranasally with 35 PFU of PR8 IAV. 3 days later, bone marrow monocytes were purified from CD45.1 mice as before. Cells were counted and concentrations were adjusted to reach a final concentration of 2.5×10⁵ monocytes/100μL of PBS and injected intravenously into the infected CD45.2 hosts. 16 hours after the transfer, mice were sacrificed and the organs were collected and analyzed by flow cytometry with differential staining of CD45.1 and CD45.2 cells.

In some experiments, the in vitro infection continued for 4 hours, and a subset of cells was stained, as highlighted above, and the mean fluorescence intensity of CCR2 was delineated via flow cytometry.

In vitro assays

Adult and neonatal cardiomyocyte and non-cardiomyocyte cardiac cell culture and isolation Neonatal cardiomyocytes were generated using the Pierce Primary Cardiomyocyte Isolation Kit (ThermoFisher) according to the manufacturer. Hearts were excised from post-natal day 1–3 old pups. Cells were cultured for 5 days prior to experimentation, either for direct infection or co-culturing with infected cells. Adult cardiomyocytes were isolated using a modified Langendorff-free method, as previously described ^{61, 62}. In brief, mice were euthanized by open drop exposure to isoflurane. The chest cavity was opened to expose the heart, and the descending aorta was cut. The heart was flushed by 7 mL of EDTA buffer (130 mM NaCl; 5 mM KCl; 0.5 mM NaH₂PO₄; 10 mM HEPES; 10 mM Glucose; 10 mM Taurine; 5 mM EDTA; 15uM Blebbistatin (Selleckchem)) into the right ventricle. Ascending aorta was clamped, and the heart was transferred to a dish containing EDTA buffer, followed by flushing 10 mL EDTA buffer through the left ventricle. The heart was washed by perfusion buffer (130 mM NaCl; 5 mM KCl; 0.5 mM NaH₂PO₄; 10 mM HEPES; 10 mM Glucose; 10 mM Taurine; 1 mM MgCl₂; 15uM Blebbistatin) and digested by 30 mL digestion buffer (perfusion buffer containing 600 U/mL Collagenase II (Worthington)) into left ventricle for 20 min. Heart was gently pulled into 1 mm³ pieces and dissociated by gentle pipetting. The enzyme activity was inhibited by addition of perfusion buffer containing 5% FBS. The isolated cells were passed through a 100 μm strainer and cell suspension was left standing upright for 15min to allow the cardiomyocytes to sediment. The cardiomyocyte pellet was rinsed with perfusion buffer and subjected to two additional rounds of gravity sedimentation. For the isolation of non-myocytes, the non-myocyte-enriched supernatant was spun at 400g for 5 min. Cardiomyocyte and non-cardiomyocyte fractions were then immediately placed into RLT lysis buffer (Qiagen) for downstream qPCR.

Cell Purification and Culture

B16 cells were purchased from Invivogen and maintained in RPMI supplemented with 10% (v/v) FBS and 100 U/mL penicillin/streptomycin. MDCK cells were obtained from the American Type Culture Collection and maintained in Dulbecco’s modified Eagle medium enriched with 10% (v/v) FBS and 100 U/mL penicillin/streptomycin. Murine bone marrow-derived macrophages (BMDM) were isolated following aseptic flushing of the tibiae and femurs of 5- to ten-week-old mice. Macrophages were differentiated from bone marrow precursors for 6 d in Roswell Park Memorial Institute 1640 Medium (RPMI-1640) supplemented with 30% (v/v) L929 cell (American Type Culture Collection)-conditioned medium, 10% (v/v) fetal bovine serum (FBS), 2 mM L-glutamine, 1 mM sodium pyruvate, 1% essential and non-essential amino acids, 10 mM 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES) and 100 U/mL penicillin/streptomycin. Bone marrow monocytes were purified using the EasySep Mouse Monocyte Isolation Kit (Stem Cell Technologies) and cultured in DMEM with 10% (v/v) FBS and 100 U/mL penicillin/streptomycin. Liver Kupffer cells (CD45⁺ CD64⁺ Tim4⁺⁾ and cardiac macrophages (CD45⁺ CD64⁺ CD11b⁺) were FACS sorted and taken as cells. Peritoneal macrophages were obtained via peritoneal lavage. Briefly, mice were anesthetized and 10mL of ice cold PBS was injected into the peritoneum. The abdomen was then massaged gently to dislodge macrophages and the 10mL was aspirated back into the syringe. Lavages from four mice were combined to acquire adequate cell numbers. Cells were then left to adhere, after 1 h of adhesion, PMs were washed with PBS and placed in fresh medium. Peritoneal macrophages, cardiac macrophages and Kupffer cells were cultured in RPMI-1640 supplemented with 10% (v/v) FBS, 2 mM L-glutamine, 10 mM HEPES and 100 U/mL penicillin–streptomycin. Prior to experimentation, cells were allowed to adhere overnight in a tissue culture plate, and the next day were subsequently infected. L929 cells (ATCC) were cultured in tissue culture flasks in DMEM with 10% (v/v) FBS and 100 U/mL penicillin/streptomycin for 10 days. Supernatants were collected and filtered for use in BMDM differentiation as CSF1-containing medium. All reagents and supplements pertaining to cell culture were purchased from Gibco.

Co-Culture of Infected Leukocytes, Plasma and Neonatal Cardiomyocytes

Cells were isolated and cultured as outlined previously. Plasma was collected from Day 3 IAV-infected mice just prior to co-culture. To avoid cell lysis via complement and clotting factors in the plasma, plasma samples were diluted 1:3 in cell culture medium prior to culture with neonatal cardiomyocytes. Following this dilution, no overt cell lysis was observed. CD319⁺, CD177⁺, total Ly6C^hi cells or B cells were infected in filter-top FACS tubes at an MOI of 1 or 5 for 2 hours. Cells were spun down and washed to remove any remaining virus and to eliminate the possibility of transfer from uninternalized bound virus. 100 000–250 000 cells were then transferred per well. At the corresponding incubation times, the admixture of infected cells and neonatal cardiomyocytes were stripped with 0.25% Trypsin-EDTA, washed and stained extracellularly. The cell types were differentiated based on their expression of CD45 and the infected gate (NP⁺) was placed based on uninfected co-cultured controls. Each experiment was performed in at least triplicate. In some experiments, mitochondrial metrics were assayed as outlined in the flow cytometry section.

Protein Based Assays

Enzyme-Linked Immunosorbent Assays (ELISAs)

ELISAs for murine Cardiac Troponin I (Novus Biologics), NT-proBNP (Novus Biologics), CCL2 (R&D Systems), SERPINE1 (R&D Systems), and IFN-β and IFN-α (both PBL Assay) were performed on plasma samples or homogenized tissues, as indicated, at the indicated time points post-infection. Assays were performed according to the manufacturers ’instructions. For detection of anti-IAV antibodies in the serum post-vaccination, a modified ELISA was used. In brief, 96-well ELISA plates (Corning) were coated with 1.25 × 107 PFU per ml of ultraviolet-light-inactivated IAV in PBS and incubated at room temperature overnight. The wells were then blocked with blocking buffer (1% BSA in PBS) for 2 h at room temperature. The plates were washed, and serially diluted serum samples were added and incubated at room temperature for 2 h. After the washes, HRP-conjugated goat anti-mouse IgG was added to the wells for 2 h at room temperature. The plates were again washed and TMB substrate was added for 20 min in the dark. Reactions were then stopped with 2 N H₂SO₄ and optical density values were obtained at 450 nm. ELISAs for human Cardiac Troponin I (Abcam), NT-proBNP (Abcam), and CCL2 (Invitrogen) were performed on human plasma samples according to manufacturer’s instructions. Values falling below the minimum detection limit were considered as 0 pg/mL.

Multi cytokine/chemokine array

A membrane-based antibody array (Proteome Profiler Mouse XL Cytokine Array Kit, R&D Systems) was used according to the manufacturer’s instructions to simultaneously determine the relative concentrations of 111 mouse cytokines and chemokines in homogenized heart tissue following 9 days of infection or not. The top 20 most upregulated proteins compared to uninfected controls were then plotted.

Quantification of Cardiac Collagen

Quantification of collagen in the heart following infection was enumerated using colorimetric hydroxyproline assay (Sigma-Aldrich) according to the specifications of the manufacturers. Hydroxyproline concentrations were plotted as a surrogate for total collagen.

Total bioactive IFN-I assay

Secretion of total active IFN-I (both IFN-α and IFN-β) in the heart was assessed using the B16-Blue IFN-α/-β reporter cell line for murine samples (from InvivoGen), according to the specifications of the manufacturer.

Cell death analyses

Lactate dehydrogenase release in the cell culture supernatants of IAV-infected cells was used as a proxy for lytic cell death and quantified using the CytoTox 96 Non-Radioactive Cytotoxicity Assay (Promega), per the manufacturer’s recommendations. Dead monocyte and macrophage numbers in the heart following infection were assessed following staining for extracellular antigens for cell type delineation and then using the PE-AnnexinV and 7AAD kit (BioLegend), as recommended. These unfixed cells were then acquired immediately by flow cytometry.

Microscopy

Immunofluorescence

Mice were perfused with 10mL of PBS and then 10mL of 4% paraformaldehyde. Organs were removed and continued to be fixed in 4% paraformaldehyde overnight followed by incubation in 30% sucrose for 24 h. Tissues were imbedded in optimal cutting temperature medium (OCT, Tissue-Tek) and serial frozen sections (50μm) were prepared. Sections were blocked with 10% normal horse serum in PBS. Sections were incubated with the following primary antibodies overnight at 4 °C at 1:100: anti-CD68 (Biorad), anti-Laminin (ThermoFisher), and AlexaFluor 647 wheat germ agglutinin (WGA; ThermoFisher), according to the manufacturer’s instructions. The following secondary antibodies (all from ThermoFisher) were used at 1:200: goat anti-rat Texas Red, chicken anti-rat AlexaFluor 647 for 1 hour at room temperature. TUNEL staining was performed using the Click-iT Plus kit (ThermoFisher) against Wheat Germ Agglutinin, both as directed by the manufacturer. Images were collected on a Leica fluorescent microscope or a Keyence BZ-X microscope and analyzed with ImageJ/QuPath.

Confocal Microscopy

Cardiac sections (30 or 50μm) were incubated with the following primary antibodies overnight at 4 °C at 1:100: anti-CD68 (Biorad, #FA-11), anti-Laminin (Sigma #L9393) or 1:200: anti-CD319 (Biolegend #118806), alpha-actinin (Invitrogen #701914). The following secondary antibodies (all from ThermoFisher) were used at 1:200: goat anti-rat Alexa Fluor 488, chicken anti-rat AlexaFluor 647 for 2 hours at room temperature or 1:300 for 6 hours at room temperature donkey anti-rabbit Alex Fluor Plus 488 (Invitrogen # A32790), anti-tdTomato (Origene #TA150129), chicken anti-rat Alexa Fluor 647. CD319-APC signal from injection before sacrifice was also imaged. Sections were blocked for 2 hours with donkey serum and washed 3x with PBS between steps. Prior to imaging, DAPI or Hoescht was added at 1:1000. Confocal microscopy and image analysis with Imaris was performed at the Microscopy and Advanced Bioimaging CoRE at the Icahn School of Medicine at Mount Sinai. Z-stack images were acquired with Zen Black software on a Zeiss LSM980 microscope (Carl Zeiss Microscopy) with a 20x lens and with a Z-step size of 0.5μm and a x-y dimension of 106μm; overview images in the form of snaps at 0.6x zoom were also acquired. Image reconstruction and analysis was performed using FIJI v2.1.0 and Imaris v10.1 (Oxford Instruments). For areas and CD319⁺ counts, snaps were used and thresholds applied in ImageJ to compute the area of cardiomyocyte or CD68+ macrophages compared to the entire snap area, or manual counting of CD319⁺ monocytes was performed. Volume overlap analysis and distance from neighboring macrophages and cardiomyocytes was derived from Imaris analysis. In brief, after Z-stack conversion to Imaris files, surface reconstruction was performed. Cardiomyocytes (Laminin⁺), pro-DC3 (CD319⁺ CD68^+/−), and cardiac macrophages (CD319⁻ CD68⁺) were reconstructed as surfaces, the approximately same creation parameters being applied to all z-stacks. Multiple separate Z-stacks per mouse were analyzed. For volume overlap ratio and closest distance, CD319⁺ cells were used as the reference in Imaris. Volume overlap is defined as the overlapping volume of CD319⁺ surfaces with cardiomyocyte or CD68⁺ surfaces divided by the total volume of that particular CD319⁺ surface. Distance from cardiomyocytes or CD68⁺ macrophages was measured as the shortest distance from a particular CD319⁺ surface to the nearest cardiomyocyte surface.

Clearing of CCR2^RFP/+ and ROSA26^TdT Hearts

Hearts from mice after mock or influenza infection were taken out after perfusion with PBS/Heparin and 4% PFA. The heart were then fixed over night in 4% PFA and subsequently delipidated following the Fast 3D clear protocol ⁶³. Briefly, hearts were repeatedly washed with PBS and ddH2O after fixation and then incubated in 50%/70%/and 90% Tetrahydrofuran with 250ppm BHT and triethylamine for 1h, 1h and 12h respectively. After tissue shrinkage was observed, the hearts were washed in 70%/50% Tetrahydrofuran with 250ppm BHT and triethylamine for 1h each and then washed repeatedly in ddH2O. Afterwards, the hearts were cleared by incubating them at 37C in 50% EasyIndex (LifeCanvas Technologies) overnight and then in 100% EasyIndex at 37C overnight. After complete clearing of the tissues, they were embedded in a 2% agarose/EasyIndex gel and imaged. Imaging was performed with a LifeCanvas SmartSpim microscope, embedding the hearts in a 1.52 RT imaging solution. 1.6x, 3x and 9x lenses were used for imaging, with a Z-step size of 4μm and 2μm respectively. Alignment of the left and right light sheets was performed and focus correction for imaging depth was applied to each imaging session. 488nm and 561nm wavelength lasers were used for autofluorescence and RFP or TdT detection respectively. The raw images were destriped and stiched with LifeCanvas software and analzyed after conversion to Imaris files. MIP projections of the images are shown in the figures.

Molecular Based Assays

RNA isolation and quantitative PCR with reverse transcription

Whole organs were excised from uninfected and infected mice as indicated in the main text. Approximately 25mg of tissue was removed and placed in tubes containing 3.0mm High Impact Zirconium Beads (Stellar Scientific) with 350μL of RLT lysis buffer (Qiagen). Samples were then homogenized. Sorted cells or tissue culture cells were placed directly into lysis buffer, as indicated. Total RNA was isolated using the RNeasy Mini or Micro Kits (Qiagen) according to its instructions. The High Capacity cDNA Reverse Transcription Kit (Applied Biosystems) was used to generate cDNA from up to 1 μg of total RNA per sample. The following FAM TaqMan primers (for genes of interest) or VIC (housekeeping genes) (Applied Biosystems) were used for quantitative real-time TaqMan PCR: Actb (Mm00607939_s1), Ifnb1 (Mm00439552_s1), Irf7 (Mm00516793_g1), Ifitm3 (Mm00847057_s1), Ifit1 (Mm00515153_m1), Ifnar1 (Mm00439544_m1), and Ccl2 (Mm00441242_m1). For quantification of viral Ns1 transcripts an alternative strategy was used. Organs were processed and cDNA was generated as before. Then, qPCR was performed using BrightGreen SYBR Green (Applied Biological Materials) using primers against host Gapdh (Forward: 5’-GGTCCTCAGTGTAGCCCAAG-3 ’and Reverse: 5’-AATGTGTCCGTCGTGGATCT-3’) and IAV Ns1 (Forward: 5’-AGAAAGTGGVAGGCCCTCTTTGTA-3 ’and Reverse: 5’-GGGCACGGTGAGCGTGAACA-3’). All qPCRs were performed on a QuantStudio 3 (Applied Biosystems). Gene expression was normalized to Actb or Gapdh and quantified using the 2^−ΔCt method.

Single-nucleus RNA-Seq

Nucleus Isolation and Library Preparation

Heart tissue nuclei were isolated using the nucleus isolation kit according to the manufacturer’s instructions (10x). After nucleus counting and viability control, the library was constructed on the Chromium 10x instrument using Chromium single cell 3 ’reagent v3.0 kits, followed by sequencing on Illumina HiSeq 2500 instruments, which resulted in approximately 205–209 million reads per sample from approximately 14 000 sequenced nuclei.

Single-nuclear RNA-seq Analysis

Obtained FASTQ files for Gene Expression (GEX) libraries were used as an input in the Cellranger Count 5.0.1 pipeline with GRCh38–2020-A genome provided by 10x. Filtering, barcode counting, UMI counting, and identification of cells was also performed with this pipeline. Cells with greater than 20% hemoglobin content were filtered out.

Obtained expression data was normalized, log-transformed and projected in 2D embedding latent space following scanpy pipeline ⁶⁴. Principal component analysis was performed, and the first 50 components were used for calculating the UMAP coordinates. The Leiden algorithm was used to construct the shared nearest neighbor graph. Cluster marker genes were identified using t-test with overestimated variance following the ‘one-versus-everyone ’strategy implemented in scanpy. Clusters enriched in markers of two or more cell phenotypes were called ‘doublets ’and removed from the analysis.

Differential Gene Expression Analysis

Each cell type was analyzed separately, filtering out genes expressed in less than 5 % of cells and filtering out cells having less than 200 expressed genes. For calculating differentially expressed genes between two conditions we used either t-test with overestimated variance or Wilcoxon rank-sum test. When comparing CD319⁺ monocytes and CD319⁻ monocytes, genes expressed in less than 15% of cells were filtered out. Pathway enrichment analysis was performed using Enricher with the top 25 most up-regulated or down-regulated genes FC>2 in infected versus uninfected cells.

Pseudotime Calculations

Pseudotime analysis was performed using the scanpy implementation of pseudotime analysis. Briefly, the preprocessed single-nuclei RNA-seq dataset was used to construct a neighborhood graph based on the first 15 principal components. Diffusion pseudotime method was applied to infer the pseudotime trajectory. The root cell was identified as the most extreme diffusion component in dimension 3.

Inferred Intercellular Cell Communication

Single-cell RNA sequencing (scRNA-seq) data were analyzed using CellChat (version 2.1.0) to investigate intercellular communication patterns, as previously reported ⁴². The analysis was performed using the default parameters in CellChat. Briefly, a CellChat object was created from the snRNA-seq data, and the interaction networks were inferred based on gene expression profiles.

Heart Failure Score

To construct the ssGSEA Heart Failure score, we calculated single sample Gene Set Enrichment Analysis (ssGSEA) scores from the single-cell scaled data matrix (@scale.data slot) of the integrated data, which holds the residuals of the corrected log-normalized integrated counts, using the Gene Set Variation Analysis (GSVA, v1.32.0) package in R with default parameters and method = “ssgsea” ⁶⁵. ssGSEA is a method that allows you to summarize gene expression patterns for any desired target gene set, and for each sample, it will return a score representative of that gene set. The target gene set was derived from ⁶⁶ filtered on genes with a log2 fold change > 3 (Expr.Fold.Change > 3, n = 50 genes total)

Illustrations

Illustrations were generated with BioRender (www.biorender.com) under an academic license.

Quantification and Statistical Analysis

Results are presented as mean ± s.e.m. Statistical tests were performed using GraphPad Prism 10 and included unpaired two-tailed Student’s t-tests and nonparametric Mann-Whitney U-tests (when distribution was not Gaussian) for two groups. For multiple comparisons, non-parametric multiple-comparison tests comparing the mean rank of each group (when Gaussian distribution was not assumed), or one- or two-way ANOVA followed by Dunnett multiple-comparisons test for one-way ANOVA and Sidak multiple-comparisons test for two-way ANOVA were used. Survival curves were analysed using the log-rank (Mantel–Cox) test. Correlations were calculated using Pearson Correlation coefficients. p < 0.05 was considered to be statistically significant; *p < 0.05, **p < 0.01, ***p < 0.001, and **** p<0.0001.

LIFE SCIENCES

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Antibodies
Rabbit monoclonal anti-Snail	Cell Signaling Technology	Cat#3879S; RRID: AB_2255011
Mouse monoclonal anti-Tubulin (clone DM1A)	Sigma-Aldrich	Cat#T9026; RRID: AB_477593
Rabbit polyclonal anti-BMAL1	This paper	N/A
Bacterial and virus strains
pAAV-hSyn-DIO-hM3D(Gq)-mCherry	Krashes et al.1	Addgene AAV5; 44361-AAV5
AAV5-EF1a-DIO-hChR2(H134R)-EYFP	Hope Center Viral Vectors Core	N/A
Cowpox virus Brighton Red	BEI Resources	NR-88
Zika-SMGC-1, GENBANK: KX266255	Isolated from patient (Wang et al.2)	N/A
Staphylococcus aureus	ATCC	ATCC 29213
Streptococcus pyogenes: M1 serotype strain: strain SF370; M1 GAS	ATCC	ATCC 700294
Biological samples
Healthy adult BA9 brain tissue	University of Maryland Brain & Tissue Bank; http://medschool.umaryland.edu/btbank/	Cat#UMB1455
Human hippocampal brain blocks	New York Brain Bank	http://nybb.hs.columbia.edu/
Patient-derived xenografts (PDX)	Children’s Oncology Group Cell Culture and Xenograft Repository	http://cogcell.org/
Chemicals, peptides, and recombinant proteins
MK-2206 AKT inhibitor	Selleck Chemicals	S1078; CAS: 1032350-13-2
SB-505124	Sigma-Aldrich	S4696; CAS: 694433-59-5 (free base)
Picrotoxin	Sigma-Aldrich	P1675; CAS: 124-87-8
Human TGF-β	R&D	240-B; GenPept: P01137
Activated S6K1	Millipore	Cat#14-486
GST-BMAL1	Novus	Cat#H00000406-P01
Critical commercial assays
EasyTag EXPRESS 35S Protein Labeling Kit	PerkinElmer	NEG772014MC
CaspaseGlo 3/7	Promega	G8090
TruSeq ChIP Sample Prep Kit	Illumina	IP-202-1012
Deposited data
Raw and analyzed data	This paper	GEO: GSE63473
B-RAF RBD (apo) structure	This paper	PDB: 5J17
Human reference genome NCBI build 37, GRCh37	Genome Reference Consortium	http://www.ncbi.nlm.nih.gov/projects/genome/assembly/grc/human/
Nanog STILT inference	This paper; Mendeley Data	http://dx.doi.org/10.17632/wx6s4mj7s8.2
Affinity-based mass spectrometry performed with 57 genes	This paper; Mendeley Data	Table S8; http://dx.doi.org/10.17632/5hvpvspw82.1
Experimental models: Cell lines
Hamster: CHO cells	ATCC	CRL-11268
D. melanogaster: Cell line S2: S2-DRSC	Laboratory of Norbert Perrimon	FlyBase: FBtc0000181
Human: Passage 40 H9 ES cells	MSKCC stem cell core facility	N/A
Human: HUES 8 hESC line (NIH approval number NIHhESC-09-0021)	HSCI iPS Core	hES Cell Line: HUES-8
Experimental models: Organisms/strains
C. elegans: Strain BC4011: srl-1(s2500) II; dpy-18(e364) III; unc-46(e177)rol-3(s1040) V.	Caenorhabditis Genetics Center	WB Strain: BC4011; WormBase: WBVar00241916
D. melanogaster: RNAi of Sxl: y[1] sc[*] v[1]; P{TRiP.HMS00609}attP2	Bloomington Drosophila Stock Center	BDSC:34393; FlyBase: FBtp0064874
S. cerevisiae: Strain background: W303	ATCC	ATTC: 208353
Mouse: R6/2: B6CBA-Tg(HDexon1)62Gpb/3J	The Jackson Laboratory	JAX: 006494
Mouse: OXTRfl/fl: B6.129(SJL)-Oxtrtm1.1Wsy/J	The Jackson Laboratory	RRID: IMSR_JAX:008471
Zebrafish: Tg(Shha:GFP)t10: t10Tg	Neumann and Nuesslein-Volhard3	ZFIN: ZDB-GENO-060207-1
Arabidopsis: 35S::PIF4-YFP, BZR1-CFP	Wang et al.4	N/A
Arabidopsis: JYB1021.2: pS24(AT5G58010)::cS24:GFP(-G):NOS #1	NASC	NASC ID: N70450
Oligonucleotides
siRNA targeting sequence: PIP5K I alpha #1 : ACACAGUACUCAGUUGAUA	This paper	N/A
Primers for XX, see Table SX	This paper	N/A
Primer: GFP/YFP/CFP Forward: GCACGACTTCTTCAAGTCCGCCATGCC	This paper	N/A
Morpholino: MO-pax2a GGTCTGCTTTGCAGTGAATATCCAT	Gene Tools	ZFIN: ZDB-MRPHLNO-061106-5
ACTB (hs01060665_g1)	Life Technologies	Cat#4331182
RNA sequence: hnRNPA1_ligand: UAGGGACUUAGGGUUCUCUCUAGGGACUUAGGGUUCUCUCUAGGGA	This paper	N/A
Recombinant DNA
pLVX-Tight-Puro (TetOn)	Clonetech	Cat#632162
Plasmid: GFP-Nito	This paper	N/A
cDNA GH111110	Drosophila Genomics Resource Center	DGRC:5666; FlyBase:FBcl0130415
AAV2/1-hsyn-GCaMP6- WPRE	Chen et al.5	N/A
Mouse raptor: pLKO mouse shRNA 1 raptor	Thoreen et al.6	Addgene Plasmid #21339
Software and algorithms
ImageJ	Schneider et al.7	https://imagej.nih.gov/ij/
Bowtie2	Langmead and Salzberg8	http://bowtie-bio.sourceforge.net/bowtie2/index.shtml
Samtools	Li et al.9	http://samtools.sourceforge.net/
Weighted Maximal Information Component Analysis v0.9	Rau et al.10	https://github.com/ChristophRau/wMICA
ICS algorithm	This paper; Mendeley Data	http://dx.doi.org/10.17632/5hvpvspw82.1
Other
Sequence data, analyses, and resources related to the ultra-deep sequencing of the AML31 tumor, relapse, and matched normal	This paper	http://aml31.genome.wustl.edu
Resource website for the AML31 publication	This paper	https://github.com/chrisamiller/aml31SuppSite

PHYSICAL SCIENCES

REAGENT or RESOURCE	SOURCE	IDENTIFIER
Chemicals, peptides, and recombinant proteins
QD605 streptavidin conjugated quantum dot	Thermo Fisher Scientific	Cat#Q10101MP
Platinum black	Sigma-Aldrich	Cat#205915
Sodium formate BioUltra, ≥99.0% (NT)	Sigma-Aldrich	Cat#71359
Chloramphenicol	Sigma-Aldrich	Cat#C0378
Carbon dioxide (13C, 99%) (<2% 18O)	Cambridge Isotope Laboratories	CLM-185-5
Poly(vinylidene fluoride-co-hexafluoropropylene)	Sigma-Aldrich	427179
PTFE Hydrophilic Membrane Filters, 0.22 μm, 90 mm	Scientificfilters.com/TischScientific	SF13842
Critical commercial assays
Folic Acid (FA) ELISA kit	Alpha Diagnostic International	Cat# 0365-0B9
TMT10plex Isobaric Label Reagent Set	Thermo Fisher	A37725
Surface Plasmon Resonance CM5 kit	GE Healthcare	Cat#29104988
NanoBRET Target Engagement K-5 kit	Promega	Cat#N2500
Deposited data
B-RAF RBD (apo) structure	This paper	PDB: 5J17
Structure of compound 5	This paper; Cambridge Crystallographic Data Center	CCDC: 2016466
Code for constraints-based modeling and analysis of autotrophic E. coli	This paper	https://gitlab.com/elad.noor/sloppy/tree/master/rubisco
Software and algorithms
Gaussian09	Frish et al.1	https://gaussian.com
Python version 2.7	Python Software Foundation	https://www.python.org
ChemDraw Professional 18.0	PerkinElmer	https://www.perkinelmer.com/category/chemdraw
Weighted Maximal Information Component Analysis v0.9	Rau et al.2	https://github.com/ChristophRau/wMICA
Other
DASGIP MX4/4 Gas Mixing Module for 4 Vessels with a Mass Flow Controller	Eppendorf	Cat#76DGMX44
Agilent 1200 series HPLC	Agilent Technologies	https://www.agilent.com/en/products/liquid-chromatography
PHI Quantera II XPS	ULVAC-PHI, Inc.	https://www.ulvac-phi.com/en/products/xps/phi-quantera-ii/