key: cord-1029701-av5g0r92
authors: Fahlberg, MD; Blair, RV; Doyle-Meyers, LA; Midkiff, CC; Zenere, G; Russell-Lodrigue, KE; Monjure, CJ; Haupt, EH; Penney, TP; Lehmicke, G; Threeton, BM; Golden, N; Datta, PK; Roy, CJ; Bohm, RP; Maness, NJ; Fischer, T; Rappaport, J; Vaccari, M
title: Cellular events of acute, resolving or progressive COVID-19 in SARS-CoV-2 infected non-human primates
date: 2020-10-16
journal: bioRxiv
DOI: 10.1101/2020.07.21.213777
sha: c58feb15274155eea054143856ec55edfc43f744
doc_id: 1029701
cord_uid: av5g0r92

We investigated the immune events following SARS-CoV-2 infection, from the acute inflammatory state up to four weeks post infection, in non-human primates (NHP) with heterogeneous pulmonary pathology. The acute phase was characterized by a robust and rapid migration of monocytes expressing CD16 from the blood and concomitant increase in CD16+ macrophages in the lungs. We identified two subsets of interstitial macrophages (HLA-DR+ CD206–), a transitional CD11c+ CD16+ cell population that was directly associated with IL-6 levels in plasma, and one long lasting CD11b+ CD16+ cell population. Strikingly, levels of monocytes were a correlate of viral replication in bronchial brushes and we discovered TARC (CCL17) as a new potential mediator of myeloid recruitment to the lungs. Worse disease outcomes were associated with high levels of cell infiltration in lungs including CD11b+ CD16hi macrophages and CD11b+ neutrophils. Accumulation of macrophages was long-lasting and detectable even in animals with mild or no signs of disease. Interestingly, animals with anti-inflammatory responses including high IL-10:IL-6 and kynurenine to tryptophan ratios had less signs of disease. Our results unravel cellular mechanisms of COVID-19 and suggest that NHP may be appropriate models to test immune therapies.

presents with highly variable outcomes, and patients that develop symptoms exhibit a wide range 48 of disease, spanning from mild (fever, cough, shortness of breath) to severe (dyspnea, 49 pneumonia, rapidly progressing radiographic changes) 1 . It is currently estimated that up to 15% 50 of COVID-19 patients progress to acute respiratory distress syndrome (ARDS) 2 , which is the 51 major cause of death among fatal SARS-CoV-2 cases. Studies elucidating molecular and 52 immunological details of SARS-CoV-2 infection are underway, and many of the reported 53 clinical observations point to the disease being at least partly the result of an excessive host 54 response aimed to clear the virus but instead contributing to disease development 3 . This 55 hypothesis is supported by studies on the closely related coronavirus SARS-CoV-1 4-6 , and by 56 clinical reports showing elevated levels of the proinflammatory cytokine IL-6 in SARS-CoV-2 57 infected patients, and profound dysregulation of the myeloid cell compartment, particularly 58 among those with severe symptoms 3,7-9 . 59 60 Interestingly, the increase in IL-6 coincides with the upregulation of chemokines responsible for 61 myelopoiesis and monocyte recruitment to the lungs 3,10 suggesting that peripheral inflammatory 62 monocytes and tissue macrophages may play a role in the cytokine storm seen in severe COVID-63 19 patients 11 . The underlying inflammatory state and increased myelopoiesis seen in older 64 individuals may be responsible for the high mortality rate seen in this group 12 . A better with no or mild signs, lower IL-10:IL-6 and kynurenine /tryptophan ratios (anti-inflammatory: 96 pro-inflammatory) cytokines in plasma 21 . By investigating temporal and spatial changes in 97 immune cells our study links together findings in both NHP and human patients and unravels 98 new cellular events of COVID-19 22,23 . Further, we show that NHP are highly relevant models to 99 study COVID-19 immunopathology and suggest that they may be appropriate models to test 100 immune therapies and vaccines for SARS-CoV-2. 101 102

Increased levels of monocytes and chemokines after infection is associated with viral load 105 and disease severity. Four adult RM and four adult AGM were exposed to SARS-CoV-2 by 106 either aerosol or mucosal challenge including buccal, intranasal, intratracheal, and conjunctival 107 (multi-route) exposure (Supplementary Table 1 ). The study concluded at 4 weeks post 108 exposure (Fig. 1A) 18 . All eight animals had detectable viral load by RT-PCR on nasal and 109 pharyngeal swabs and bronchial brushes by week 1 (Fig. 1B, Supplementary Table 2 ). Viral 110 RNA was also detected in pharyngeal, nasal, buccal, rectal and vaginal swabs. No differences in 111 viral load (VL) were observed between the different routes of exposure or between the two 112 species at week 1 (Supplementary Table 2 Table 2 ). Similar to humans, the course of disease varied widely across animals, as detailed in 116 Table 1 . Of the four AGM enrolled, two animals developed severe respiratory signs and were 117 euthanized on days 8 (NC34) and 22 (NC33) post-infection, following established endpoint 118 criteria for NHP study protocols, and as we previously reported 18 . Of the two remaining AGM, 119 NC38 had multifocal mild to moderate interstitial pneumonia scattered throughout all lung lobes 120 and NC40 was mostly asymptomatic and had scant inflammation in all lung lobes at necropsy. 121

Three of the four RM (GH99, HD09, and FR04) developed pneumonia characterized by a 122 granulomatous to pyogranulomatous inflammatory response. This was severe in GH99, mild in 123 HD09, and minimal in FR04. Histopathology of the fourth RM (HB37) revealed a lymphocytic 124 vasculitis and proliferative vasculopathy in the right middle lung lobe. 125

Clinical signs observed during the course of the study included mild tachypnea (day 11) and 126 cough (day 15) in GH99 (Supplementary Table 3 ). Intermittent mild to moderate pyrexia was 127 observed in HB37 (day 1-21 post infection). Radiographic changes included a nodule in the right 128 caudal lung lobe (day 11) in GH99, and HD09 also developed a mild, focal increased opacity in 129 the ventral lung field at day 11 (data not shown).Viral loads in the bronchial alveolar brushes at 130 all time points, temperature, pulse oxygen levels (SpO2) and ranking did not differ between 131 female (n = 3) and males (n = 5) or between animals exposed to different routes of infection 132 (multi route n = 4 and aerosol n = 4), possibly due to the small number of animals per group 133 (Supplementary Table 2 Table 5) . 139 140 Interestingly, virus-driven immunological changes were remarkably similar between species at 141 this time ( Fig. 1 and Supplementary Fig. 1) . To investigate overall changes driven by viral 142 replication in both groups, we combined their immunological analysis, however at the same time 143

we maintained different specifications for each species (teal = RM and magenta = AGM). Data 144 and statistics are also shown separately per group in Supplementary Figure 1 . At 1 week after 145 infection both AGM and RM had an increase in the absolute number and frequency of 146 monocytes in the blood (P = 0.065, and P = 0.02 respectively, by the Tukey multiple comparison 147 test, n = 8) compared to baseline levels ( Fig. 1D and 1E) . This change was significant in AGM 148 alone (absolute number: P = 0.0006, n = 4), and was also observed in all RM, particularly GH99 149 ( Supplementary Fig. 1C -1F ). Of note, the increase in absolute number of monocytes remain 150 significant when GH99 was removed from the statistical analysis (P = 0.00002, Supplementary 151 Fig. 1H) . Interestingly, the absolute number of monocytes at week 1 was 153 strongly positively associated with the viral load levels in the bronchial brush samples (P = 154 0.002, R = 0.93 by the Spearman test) when all the animals were combined (Fig. 1F ) (AGM: R = 155 0.8; R: R = 1, not significant in separated groups, Supplementary Fig. 1I and 1J) , and remained 156 significant when GH99 was removed from the statistical analysis (Supplementary Fig. 1K) . 157

We then used the CD14 and CD16 markers to discriminate between three circulating monocytes 158 subtypes: classical (CD14hi CD16low), intermediate (CD14hi CD16hi) and non-classical 159 (CD14low CD16hi) within the total (CD45, live , HLA-DR+ CD16hi) monocyte population by 160 flow cytometry (Fig. 1G -1I and Supplementary Fig. 2 ). Subsets were gated as shown in 161

Supplementary Figure 2A , using a strategy that excludes CD16+ NK cells that are negative for 162 HLA-DR 24 . At week 1, the frequency of classical monocytes was significantly increased in 163 blood, as seen in humans 25 , and the non-classical population was profoundly depleted (increased 164 levels baseline versus week 1: classical P = 0.007 and decreased non-classical P = 0.005 in all 165 animals combined, n = 4) ( Fig. 1H and 1I, and Supplementary Fig. 2B -2E ). As we observed 166 for the frequency, the absolute counts of classical and non-classical monocytes at week 1 were 167 also significantly increased and decreased, respectively, compared to the baseline values 168 (Supplementary Figure 2F -G) . Interestingly, pre infection levels were never completely 169 restored even after 3 weeks post-infection (week 1 vs. week 3: classical P = 0.03 and non-170 classical P = 0.03), as the number of circulating CD16hi monocytes increased overtime (non-171 classical week 1 versus week 3: P = 0.0019). These results suggest emerging myelopoiesis and 172 rapid and robust recruitment of patrolling monocytes in tissues following infection in both 173 species. We did not observe changes in the frequency or number of intermediate monocytes 174 Fig. 2H and 2I ). In line with what we observed for the total monocyte count, at 175

week 1 the counts for the subsets of monocytes were positively associated with viral replication 176 detected at the bronchial alveolar sites (classical: R = 0.98; P = 0.06, non-classical: R = 0.96; P = 177 0.03 and intermediate R = 0.94; P = 0.06, by the Pearson test), further suggesting that the robust 178 myelopoiesis is therefore driven by the virus in the lung. Within the total monocytes, the 179 frequency of non-classical monocytes was significantly associated with lower viral loads 180 (classical: R = -0.9867; P = 0.0133, by the Pearson Test, adjusted for multiple comparison P = 181 0.07). It is important to note that none of these analyses were significant when the values were 182 adjusted for multiple comparisons (Supplementary Table 6) . 183

Recent studies in humans have described an increase in the frequency of CD14 positive cells 184 expressing low levels of HLA-DR in COVID-19 patients. These cells phenotypically resemble 185 suppressive subsets of myeloid derived suppressor cells (MDSCs) 8 thus they may decrease 186 disease driven inflammation. We and others have previously identified MDSCs in blood of 187 monkeys by flow cytometry as HLA-DR low CD14 hi cells within the CD45/live population 26, 27 . 188 Some of the animals in our study showed increased MDSCs frequency at week 3 compared to 189 baseline levels, however these changes were not significant (Supplementary Fig. 2L ). We 190 measured Arginase 1 activity as a functional assay to assess the function of immunosuppressive 191 myeloid cells at the same time points (baseline and week 3) and we did not observe significant 192 changes in plasma overtime (Supplementary Fig. 2M ) 28 . 193 194 We analyzed a group of eight chemokines known to be involved in the mobilization of innate 195 cells, including monocytes, from the blood to the lung: IP-10 (CXCL-10), MCP-1 (CCL2), influx of myeloid HLA-DR+ CD206-CD163-cells ( Fig. 2A and 2B, Supplementary Fig. 4C ). 223 Levels of AM similar to pre-infection were restored at week 2 in all but one animal (GH99), and 224 a concomitant increase of CD86 mean fluorescent intensity (MFI) in this population was 225 observed suggesting an M1 activation state 33 (Supplementary Fig. 4D ). An increase was also 226 seen in the frequency of CD206+ CD163-that have been previously described as endothelial 227 cells which facilitate lymphocyte trafficking and they are detectable in BAL during inflammatory 228 events (Fig. 2C) 34 . 229 T-distributed stochastic neighbor embedding (tSNE) analysis revealed changes in the phenotype 230 of CD16, CD11b, and CD11c positive populations most prominently at 1-2 weeks post infection 231 ( Fig. 2D -2F) . In contrast to what we observed in the blood, there was an increase of CD16+ 232 HLA-DR+ cells in BAL over time (Fig. 2G, and Supplementary Fig. 4E and 4F) . We observed 233 a transient increase in the level of CD11c+ CD16+ myelomonocytic cells, which function has 234 been described as patrolling, and they participate in rapid tissue invasion and monocyte survival 235 (baseline vs. week 1: P = 0.015) (Fig. 2H) 35, 36 . The frequency of CD11c+ macrophages in BAL 236 at week 3 was strongly associated with the levels of the proinflammatory IL-6 cytokine in the 237 plasma at the same time point (week 3: P = 0.003, R = 0.96, Spearman test) (Supplementary 238 Fig. 5) . The CD11b marker is required for monocyte migration to sites of inflammation 37 , and is 239 highly expressed on myeloid cells recruited from the blood and differentiating into 240 macrophages 38 . We also observed an increase in the percentage of CD11b+ CD16+ HLA-DR+ 241 CD206-myelomonocytic cells in BAL at week 2 and 3 following infection (Fig. 2I) Figure 2) , with the exception of NC33 and NC34. These two 248 animals had higher levels of IL-6 at necropsy compared to the other six monkeys, and NC34 had 249 symptoms similar to ARDS 18 (Fig. 3) . Overall, infiltration of macrophages was observed in the 250 lungs of SARS-CoV-2 infected AGM and RM at necropsy and was more robust in animals with 251 severe disease (Fig. 3A) . We further characterized macrophages population in the lung by 252 staining with the AM marker CD206 (in blue) and with CD11b integrin (in green) and CD16 253 (red) (Fig. 3B) . Low numbers of macrophages were seen in the lungs of a non-infected control 254 RM, scattered throughout the interstitium and alveolar spaces, as shown by the arrows. In the 255 control this population was bi-morphic and composed of CD11b negative CD16+ CD206-256 (arrow) in the alveolar septa or CD16+ CD206+ cells within the alveolar spaces (arrowhead) 257 ( Fig. 3B) . In all animals that were infected we could detect CD11b+ CD16+ macrophages. were also high in NC33 ( Figure 4C ). Accordingly, neutrophils and lymphocytes could be seen in 290 animals with severe disease but not in animals with moderate disease (Supplementary Fig. 7) . 291 NC34 (necropsy day 8 post infection; rank = 1, severe) had lymphocytic aggregates that were 292 perivascular and intermixed with neutrophils and histiocytes. In NC38 (necropsy day 28, rank 5, 293 moderate) lymphocytic aggregates are scattered within the alveolar septa and predominately 294 mixed with histiocytes and rare neutrophils. 295 296

Infiltration of lymphocytes was observed in the lungs of SARS-CoV-2 infected AGM and RM at 298 necropsy and was more robust in animals with severe disease (Fig. 5A) . We first analyzed the 299 kinetics of innate NK cells subsets, that we defined as HLA-DR-CD3-cells expressing the 300 NKG2a receptors, which are known to traffic to the lung 45 ( Fig. 5B -D, and Supplementary 301 overlapping with spatial regions corresponding to both CD4+ T and CD8+ T cells ( Fig. 5G and 318 Supplementary Fig. 9A ). In fact, the frequency of PD-1+ CD4+ T cells was elevated in the 319 blood of both RM and AGM (Fig. 5H, and 5I , respectively). During the first week following the 320 infection we observed an increase in IP-10 and IFN-g (Supplementary Fig. 9B ), however we 321 also observed a switch toward Th2 type responses characterized by the increase of Th2 cytokines 322 IL-5 and IL-13 over time (Supplementary Fig. 9B ). Accordingly, a gradual increase of Th2 type 323 cells (defined as CXCR3-CCR6 -CD4+ T cells) 24 was seen in the blood within the PD-1 324 positive population (Fig. 5J) . The increase of PD-1+ Th2 cells was significantly and positively 325 associated with IL-4 cytokine levels (P = 0.016, R = 0.94, Spearman test) (Supplementary Fig.  326 9C). Moreover, we observed an increased level of PD-1+ CD8+ T cells in blood and BAL (BAL: 327 baseline vs. week 2, P = 0.015; baseline vs. week 3, P = 0.03) (data not shown and Fig. 5K) . 328 Animals divided by species are shown in Supplementary Fig. 9D and 9E . Associations between 329 viral replication and the frequency of PD1+ CD4 T cells in blood and PD1+ CD8 T cells in BAL 330 at week 3 were found (Supplementary Fig. 9G-I) , however, they were most likely driven by 331 the AGM but not RM which had no detectable viral replication in bronchial brushes at that 332 specific time point. 333

The ratio between the anti-inflammatory and proinflammatory responses is associated with 335 disease progression. The pro-inflammatory cytokine IL-6 has been described as a strong 336 predictor to disease progression in humans 3,7 . We measured IL-6 and IL-10 in the plasma of all 337 the animals (Fig. 6A) . While IL-6 levels alone did not associate with disease severity in NHPs, 338 the ratio of IL-10 to IL-6 levels measured in plasma associated with both the ranking of disease 339 severity and with the pathological score that was obtained by analyzing the extension of the 340 inflammation and edema of the lungs (R = 0.76 P = 0.037; and R = -0.76, P = 0.03, by the 341 Spearman correlation test, all animals) ( Fig. 6B and C) . To investigate further, we also measured 342 the levels of tryptophan and kynurenine in plasma at baseline, week 1, week 3 and at necropsy 343 (Supplementary Figure 10) . Kynurenine was elevated and significantly increased at week 3 344 compared to baseline levels (P= 0.013 adj.). The kynurenine (Kyn)/ tryptophan (Tryp) ratio has 345 been used as a measurement of the indoleamine 2,3-dioxygenase (IDO) activity and is often 346 associated with T regulatory cell function (21) . . In line with the observed association for the IL-347 10:IL-6 ratio, the Kyn /Tryp ratio was also negatively correlated with the pathology score (1 = 348 mild -18 = severe) suggesting that animals with higher levels of immune suppression had less 349 inflammation in the lung and better disease outcome (Figure 6D) . to the lung 2) a gradual switch from a type 1 to type 2 response, and 3) a "make it or break it" 366 phase with either an increase in anti-inflammatory cytokines IL-10 and regulatory cell subsets 367 with suppressive activity, or IL-6, a pro-inflammatory cytokine associated with disease 368 progression in humans, as summarized in Figure 6E . Table 1) . 459

The multi route exposure was given to four animals, one adult RM male (GH99, 14 years old), 460 one adult RM female (HD09, 13 years old) and two AGM, one aged male (NC40, 16 years old, 461 approximately) and one aged female (NC33, 16 years old, approximately). An additional two 462 adult males RM (FR04 and HB37, 15 and 13 years old, respectively) and one aged male and 463 female AGM: NC34 and NC38 (16 years old, approximately) were exposed by aerosol 464 Heatmaps were generated using the 'pheatmap' package in R 60,61 . Data were normalized by 532 dividing raw values at week 1 by baseline values for each animal, followed by the application of 533 log2. Values below the limit of detection were replaced with the lowest limit of detection value 534 based on the standard curve for each run, or with the lowest value detected during the run, 535 whichever was smaller. Bubble plots were generated using the 'ggplot2' package in R, using the 536 same normalized data shown in the heatmap 59 . Scatter plots were drawn using raw data points 537 and display Pearson's correlation coefficients and a 95% confidence interval. samples were thawed on ice and, in order to deplete the urea, 40 μl of plasma were loaded in an 579 Amicon®Ultra 10K centrifugal filter (UFC501096 EMD Millipore), diluted with pure water to 580 500 μl, and centrifuged at 13,000 × g for 30 min at 4°C. Following centrifugation, the eluted 581 solution was discarded. Filtered samples were then diluted with pure water to 500 μl, and 582 centrifuged at 13,000 × g for 30 min at 4°C. At the end of centrifugation, the remaining volume 583 of each sample was measured, and ultra-pure water was added to reach a final volume of 40 μl. 584

Each sample was loaded into 2 wells of a 96-well plate (20 μl/well), representing the sample well 585 and the sample blank well, and 20 μl/well of ultra-pure water were added to each well. Together 586 with samples, the plate was loaded with urea standard and water as positive and negative 587 controls, respectively. Samples were loaded in singlicate, whereas controls were loaded in 588 duplicate. Ten microliters of 5X substrate buffer, composed of Arginine Buffer and Mn Solution, 589 were added to the wells except for sample blank wells, and they were incubated for 120 min at Flow Cytometry Core for their help. We thank Angela Birnbaum for reviewing and optimizing 624 all technical SOPs and overseeing the safety of this study. 625 We would also like to thank Fast Grant Funding for COVID-19 Science for partially funding this 626 work and the NIH for supporting this work through the TNPRC base grant (P51 OD011104 59). 627 628

All data will be made available upon acceptance. 

Severe reduction in SpO2, moderate tachypnea, and severe hypothermia just prior to euthanasia for severe respiratory distress at 8 days post infection. Less than 24 hours earlier, the only abnormality was a moderate reduction in SpO2. Radiography appeared unremarkable until euthanasia when moderate to severe alveolar pattern with lobar sign and air bronchograms in right lung fields was observed.

Bronchointerstitial lung pattern in all lobes that was most severe in the right caudal lung lobe. Microscopically all lung lobes exhibited similar changes with the right caudal and right accessory lung lobe being the most severely affected. The lung lesions were comprised of diffuse alveolar damage with hyaline membrane formation, type II pneumocyte hyperplasia, and rare multinucleated giant syncytia. Some regions showed early evidence of organizing pneumonia and fibrosis.

Severe reduction in SpO2, severe tachypnea, and severe hypothermia just prior to euthanasia for severe respiratory distress at 22 days post infection. Less than 24 hours earlier, the only abnormality was a mild reduction in SpO2. Radiography appeared unremarkable until euthanasia when moderate to severe alveolar pattern in right middle/caudal lung fields and mild alveolar pattern left caudal lung fields were observed.

Severe diffuse alveolar damage of the right lower lung lobe. Other lobes exhibited suppurative alveolar infiltrate with lesser fibrin and edema. The right lung lobe contains peracute to acute lesions of necrosis, fibrin, and hemorrhage with minimal cellular infiltrate.

14 2

Transient cough beginning on 15 days post infection. Radiography showed significant changes at 11 days post infection which improved through 27 days post infection.

Severe bronchopneumonia of the right lower lung with granulomatous to pyogranulomatous inflammatory response.

Mild interstitial pneumonia with low numbers of multinucleated giant cells and rare atypical pneumocyte hyperplasia.

Mild focal increase opacity ventral lung field at 11 days post infection.

Minimal to mild inflammation scattered throughout the lungs. Proliferative lymphocytic vasculitis and proliferative vasculopathy in the right middle lung lobe. Scattered interstitial inflammation throughout the remaining lung lobes. Tregs?

Clinical Characteristics of Coronavirus Disease 2019 in China. N Engl 651

Clinical characteristics of hospitalized patients with SARS-CoV-2 653 infection: A single arm meta-analysis

Imbalanced Host Response to SARS-CoV-2 Drives Development 655 of COVID-19

Forty years with coronaviruses

Primary severe acute respiratory syndrome coronavirus infection limits 658 replication but not lung inflammation upon homologous rechallenge

A decade after SARS: strategies for 661 controlling emerging coronaviruses

Clinical predictors of mortality due 663 to COVID-19 based on an analysis of data of 150 patients from Wuhan, China. Intensive 664

Elevated Calprotectin and Abnormal Myeloid Cell Subsets Discriminate 666 Severe from Mild COVID-19

Severe COVID-19 Is Marked by a Dysregulated Myeloid 668 Cell Compartment

Transcriptomic characteristics of bronchoalveolar lavage fluid and 670 peripheral blood mononuclear cells in COVID-19 patients

Pathogenic T cells and inflammatory monocytes incite inflammatory 673 storm in severe COVID-19 patients

Immunosenescence 675 and human vaccine immune responses

Animal models for SARS and MERS coronaviruses

Replication of SARS coronavirus administered into the respiratory 679 tract of African Green, rhesus and cynomolgus monkeys

SARS-CoV-2 infection protects against rechallenge in rhesus 681 macaques

SARS-CoV-2 infection leads to acute infection with dynamic cellular 683 and inflammatory flux in the lung that varies across nonhuman primate species

DNA vaccine protection against SARS-CoV-2 in rhesus macaques

ARDS and Cytokine Storm in SARS-CoV-2 Infected Caribbean 688 Vervets

Respiratory disease in rhesus macaques inoculated with SARS-CoV-690 2

Diverse macrophage populations mediate 692 acute lung inflammation and resolution

The combined effects of tryptophan starvation and tryptophan 695 catabolites down-regulate T cell receptor zeta-chain and induce a regulatory phenotype in 696 naive T cells

A single-cell atlas of the peripheral immune response in patients with 698 severe COVID-19

Breadth of concomitant immune responses prior to patient recovery: 700 a case report of non-severe COVID-19

HIV vaccine candidate activation of hypoxia and the inflammasome in 702 CD14

Immune cell profiling of COVID-19 patients in the recovery stage by 704 single-cell sequencing

Myeloid Cell Crosstalk Regulates the Efficacy of the 706 DNA/ALVAC/gp120 HIV Vaccine Candidate

Rhesus Macaque Myeloid-Derived Suppressor Cells Demonstrate T Cell 708

Inhibitory Functions and Are Transiently Increased after Vaccination

L-arginine metabolism 711 in myeloid cells controls T-lymphocyte functions

A degradatory fate for CCR4 suggests a primary role in Th2 713 inflammation

Plasma IP-10 and MCP-3 levels are highly associated with disease 715 severity and predict the progression of COVID-19

Severe COVID-19 and aging: are monocytes the key? Geroscience (2020)

In vivo characterization of alveolar and interstitial lung macrophages in 718 rhesus macaques: implications for understanding lung disease in humans

Pulmonary 721 macrophage subpopulations in the induction and resolution of acute lung injury

Mannose receptor is a novel ligand for L-selectin and mediates 724 lymphocyte binding to lymphatic endothelium

Monitoring of blood vessels and tissues by a population of monocytes 726 with patrolling behavior

CX3CR1 is required for monocyte homeostasis and atherogenesis by 728 promoting cell survival

Regulated expression of integrins and other 730 adhesion molecules during differentiation of monocytes into macrophages

Ly6G-immature myeloid cells recruited in response to Salmonella enterica serovar 734 Typhimurium infection exhibit protective and immunosuppressive properties

Single-cell landscape of bronchoalveolar immune cells in patients with 737 COVID-19

739 Local activation of nonspecific defense against a respiratory model infection by 740 application of interferon-gamma: comparison between rat alveolar and interstitial lung 741 macrophages

Human neutrophils in the saga of 743 cellular heterogeneity: insights and open questions

Combinatorial Single-Cell Analyses of Granulocyte-Monocyte Progenitor 745

Heterogeneity Reveals an Early Uni-potent Neutrophil Progenitor

Heterogeneity of neutrophils

Expression of 750 myeloperoxidase (MPO) by neutrophils is necessary for their activation by anti-751 neutrophil cytoplasm autoantibodies (ANCA) against MPO

Natural Killer Cells in the Lungs

757 47. National Research Project for SARS, B.G. The involvement of natural killer cells in the 758 pathogenesis of severe acute respiratory syndrome

Immunologic perturbations in severe COVID-19/SARS-CoV-2 761 infection

Non-classical tissue monocytes and two functionally distinct populations 763 of interstitial macrophages populate the mouse lung

Regulation of pulmonary fibrosis by chemokine receptor CXCR3

Cxcr3 and its ligand CXCL10 are expressed by inflammatory cells 767 infiltrating lung allografts and mediate chemotaxis of T cells at sites of rejection

The NKG2A immune checkpoint -a new direction in 770 cancer immunotherapy

T cell responses to whole SARS coronavirus in humans

CD11b+ myeloid cells are the key mediators of Th2 cell homing into 774 the airway in allergic inflammation

Compartmentalized production of CCL17 in vivo: strong inducibility in 776 peripheral dendritic cells contrasts selective absence from the spleen

Establishment of an African green monkey model for COVID-19

Profiling serum cytokines in COVID-19 patients reveals IL-6 and IL-10 781 are disease severity predictors

Age-Associated Pathology in Rhesus Macaques (Macaca mulatta)

Elegant Graphics for Data Analysis

Pretty Heatmaps

R: A language and environment for statistical computing

Econometric Tools for Performance 790 and Risk Analysis