Association of CD16hiCD62Llo Neutrophils with Disease Severity and Impact of SYK Inhibition in Sepsis Authors: Xianglan Yao, Heather L Teague, Lauren E Knabe, Raquel Santana da Cruz, Trenton Williams, Merte Lemma Woldehanna, Ernest Evans, Amir Hobson, Jared D Wilkinson, Cumhur Yusuf Demirkale, Steven D Nathan, Christopher King, Jeffrey R Strich Published: 2026 ABSTRACT AbstractBackgroundSepsis is a global health priority that leads to significant morbidity and mortality. Therefore, there is an urgent need to better understand sepsis pathogenesis to inform diagnosis, prognosis, and identify novel host-directed therapeutics.MethodsFlow cytometry was performed on 35 sepsis patients and 18 healthy controls to characterize neutrophils and neutrophil subsets and their association with clinical outcomes. Additionally, purified neutrophils from 20 sepsis patients were stimulated with lipopolysaccharide (LPS), with and without R406 treatment, to assess changes in neutrophil populations and the impact of spleen tyrosine kinase (SYK) inhibition.ResultsWe found that patients with sepsis have increased levels of circulating CD16hi/CD62Llo neutrophils that correlate with disease severity, including the need for intensive care unit admission, vasopressors, and mortality. This CD16hi/CD62Llo population expressed lower CD10 and CD62L and higher CD33 compared to healthy controls. Additionally, compared to healthy donors, the sepsis CD16hi/CD62Llo population had higher levels of SYK. Finally, ex vivo stimulation of sepsis neutrophils with LPS increased the percentage of CD16hi/CD62Llo population and the shift in this population could be prevented with SYK inhibition.ConclusionThese data suggest that CD16hi/CD62Llo neutrophils may contribute to sepsis pathogenesis and that SYK is a potential therapeutic target that can modulate this neutrophil population, but further mechanistic studies are warranted to establish a causal role in sepsis.Keywords: sepsis, spleen tyrosine kinase, neutrophils Introduction Sepsis is defined as a dysregulated host-response to infection and is associated with high mortality and poor long-term outcomes.1,2 Neutrophils, an important component of the innate immune response, are pivotal to pathogen clearance in sepsis. However, the overactivation of neutrophils during sepsis can propagate immune dysregulation, contributing to end-organ dysfunction.3,4 Neutrophils are a heterogenous population of innate immune cells with different phenotypic and functional characteristics.5,6 Therefore, understanding the contribution of neutrophils to sepsis pathogenesis is important to guide advances to improve clinical outcomes including mortality.7 Despite decades of research, improvements in sepsis mortality have been stagnant.8 Areas of focus to improve sepsis outcomes have been the development of better diagnostic tools, reliable prognostic factors and effective host-directed therapies. To date, the clinical diagnosis of sepsis is primarily guided by the sequential organ failure assessment (SOFA) score to identify end-organ dysfunction, which is reliant on clinical and laboratory values obtained as part of routine clinical care.9 Likewise, most prognostication of sepsis outcomes relies on similar clinical and laboratory values.10 Lastly, to date there are no FDA‑approved host-directed therapies targeting the dysregulated innate immune response that dominates sepsis.11 Spleen tyrosine kinase (SYK) is an intracellular non-receptor tyrosine kinase signaling protein that plays a vital role in several neutrophil end effector functions. SYK inhibition with fostamatinib, an FDA‑approved therapeutic for adults with chronic immune thrombocytopenia, has been studied in critically ill patients with COVID-19, and there is mounting preclinical animal model and human data that supports SYK inhibition in patients with sepsis by modulating neutrophil activation.12–15 CD16hi/CD62Llo neutrophils are a subset of neutrophils that have been identified in sepsis-associated acute respiratory distress syndrome (ARDS)17,16 cancer19,17 and cystic fibrosis.18 In this study, we evaluated this specific CD16hi/CD62Llo neutrophil population from patients with sepsis to see if this subset associates with disease severity, could be used a potential biomarker, and determine if this population is impacted by SYK inhibition. Materials and Methods Patient Enrollment This was a single-center IRB-approved cohort study undertaken at Inova Fairfax Hospital in collaboration with the National Institutes of Health (NIH) in compliance with the Declaration of Helsinki. Patients with a diagnosis of sepsis were identified at Inova Fairfax Hospital using the Sepsis-3 criteria (suspected organ dysfunction and evidence of end-organ dysfunction [sequential organ failure assessment score (SOFA) ≥ 2]) and subsequently informed consent was obtained from the patient or a legally authorized representative in those who were unable to consent as a result of their sepsis diagnosis (Inova Fairfax Hospital [IRB protocol #15-1863]). Patient blood draws were taken the day following enrollment and within 3 hours transferred to the National Institutes of Health Clinical Center for processing and downstream analyses. Age and sex-matched healthy donor (HD) blood samples (n =18) were obtained from the NIH Department of Transfusion Medicine in Bethesda, MD (IRB protocol #99-CC-0168) (Table 1). This cohort of patients was previously published in a characterization of other neutrophil subsets. Table 1Clinical Characteristics of Healthy Donors and Sepsis PatientsClinicalVariablesHD(n = 18)Sepsis(n = 35)p-valueDemographics Age (years)64 (13)65 (14)>0.9 Sex0.6 Male, n (%)10 (56)17 (49) Female, n (%)8 (44)18 (51) Race0.5 White, n (%)12 (67)19 (54) Black, n (%)3 (17)7 (20) Asian, n (%)1 (5.6)6 (17) Hispanic, n (%)1 (5.6)0 (0) Unknown, n (%)1 (5.6)3 (8.6)Hematologic Characteristics White Blood Count, 103/uL, mean (SD)--13 (8)NA Hemoglobin, g/dL, mean (SD)--9.92 (2.02)NA Platelets, /uL, mean (SD)--178 (106)NA Serum Creatinine, mg/dL, mean (SD)--1.59 (1.06)NASepsis Severity ICU StatusNA Non-ICU, n (%)--10 (29) ICU, n (%)--25 (71) SOFA Score, mean (SD)--4.4 (3.0)NA Sepsis Status (at any point)NA Sepsis, n (%)--15 (43) Septic Shock, n (%)--20 (57) Vasopressor (at time of blood draw)NA Yes, n (%)--17 (49) No, n (%)--18 (51) Mechanical VentilationNA Yes, n (%)--8 (23) No, n (%)--27 (77) Acute Kidney InjuryNA Yes, n (%)--21 (60) No, n (%)--14 (40) MortalityNA Yes, n (%)--3 (9) No, n (%)--32 (91) Length of Stay (LOS), days, mean (SD)--9.6 (6.6)NA LOS After Blood Draw, days, mean (SD)--7.1 (6.5)NA Infection Source Site of InfectionNA Skin/soft tissue, n (%)--4 (11) Intra-abdominal, n (%)--10 (29) Pneumonia, n (%)--8 (23) Urinary Tract, n (%)--6 (17) Other, n (%)--4 (11) Unknown, n (%)--3 (8.6) OrganismNA Polymicrobial, n (%)--6 (17) Gram Positive, n (%)--6 (17) Gram Negative, n (%)--6 (17) Viral, n (%)--3 (8.6) Fungal, n (%)--1 (2.9) Unknown, n (%)--13 (37)Notes: Values are reported as means ± SD for continuous variables and as n (%) for categorical variables. Statistical significance was determined by Wilcoxon rank-sum test, Pearson’s Chi-squared test or Fisher’s exact test.Abbreviations: ICU, Intensive care unit; SOFA, Sequential organ failure score; SD, Standard deviation. Clinical Characteristics of Healthy Donors and Sepsis Patients Notes: Values are reported as means ± SD for continuous variables and as n (%) for categorical variables. Statistical significance was determined by Wilcoxon rank-sum test, Pearson’s Chi-squared test or Fisher’s exact test. Abbreviations: ICU, Intensive care unit; SOFA, Sequential organ failure score; SD, Standard deviation. Flow Cytometry Analysis of Neutrophils Whole blood from sepsis patients and healthy donors was obtained in EDTA-tubes. Plasma was collected after centrifugation at 1400 rpm for seven minutes. Leukocytes were washed with Fluorescence Activated Cell Sorting (FACS) buffer after erythrocytes were lysed with ACK lysis buffer (Quality Biological, Gaithersburg, MD). One million leukocytes were aliquoted into FACS tubes; FcR block was added to leukocytes for 10 minutes (Miltenyi Biotec, Bergisch Gladbach, Germany), then the cells were stained on ice with extracellular antibodies pertaining to neutrophil maturation, degranulation, and activation for 20 minutes. Fixable Far-Red Viability Dye (Life Technologies, Eugene, Oregon) was added for 10 mins followed by washing with FACS buffer. Subsequently, cytofix/cytoperm (BD Biosciences, Franklin Lakes, NJ) was added to leukocytes for 15 minutes for fixation/permeabilization; the intracellular antibodies were then added to each sample for intracellular staining for 20 mins after washing with 1X permeabilization buffer (Please see prior publication for a list of antibodies for intracellular and extracellular staining).19 Samples were fixed in 1% paraformaldehyde after washing with 1X permeabilization and FACS buffer. Samples were acquired using a 4 laser Attune NxT flow cytometer (Life Technologies, Carlsbad, CA). Whole blood neutrophils were identified with traditional gating strategies. Following singlet gating, lymphocytes and natural killer cells were removed in the channel designated as “dump”, which also includes live/dead (viability) staining. Additionally, neutrophil subsets were gated for CD15+ (Figure 1A). Fluorescence Minus One (FMO) were used to establish gating of samples. Characterization of CD16hi/CD62Llo neutrophils amongst sepsis patients and healthy donors. (A) Gating strategy to identify neutrophils and neutrophil subsets based on CD16 and CD62L expression. (B) Bar graph comparing the percentage of each quadrant amongst healthy donors (HD). (C) Bar graph comparing the percentage of each quadrant amongst sepsis patients. (D) Bar graph comparing the percentage of Q3 CD16hi/CD62Llo neutrophils between HD and sepsis patients. Data are represented as means±SEM. Significance was determined using post-hoc Dunn’s test for pairwise comparisons after Friedman test or an Unpaired t-test and set at p<0.05*, p<0.01**, p<0.001***, p<0.0001****. Q1: quadrant 1; Q2: quadrant 2; Q3: quadrant 3; Q4: quadrant 4; HD: healthy donor. Isolation of Purified Neutrophils Purified neutrophils from 20 sepsis patient blood samples were isolated by immunomagnetic negative selection using the StemCell EasySep Direct Human Neutrophil Isolation Kit (Catalog # 19666, StemCell Technologies, Vancouver, Canada). In brief, the plasma and mononuclear cell portions were removed after density gradient centrifugation, then the RapidSpheresTM and neutrophil isolation cocktail were added to the bottom layer containing granulocytes and RBCs in round bottom polystyrene tubes, incubated for 20 min prior to diluting with Dulbecco’s PBS + 2% FBS. Following multiple rounds of immunomagnetic selection, negatively selected neutrophils were collected in X–VIVO-15 serum-free medium. Stimulation of Purified Neutrophils Purified neutrophils were preincubated for 30 minutes with or without 2 μM R406 (the in vitro active component of fostamatinib, provided by Rigel Pharmaceuticals) and stimulated with 1μg/mL of lipopolysaccharide (LPS) from Klebsiella pneumoniae (Millipore Sigma, cat# L4268) at 37°C for one hour and analyzed by flow cytometry as described above. Measurement of Biomarker Levels R-Plex and custom multiplexed U-Plex immunoassays [Meso Scale Discovery-MSD, Rockville, MD] were used to measure the plasma levels of neutrophil-associated markers [myeloperoxidase (MPO), elastase-2, lactoferrin, S100A8/MRP8, neutrophil gelatinase-associated lipocalin (NGAL), resistin], immunoregulatory cytokines [granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), interferon-gamma (IFN-γ), interleukin-1 beta (IL-1β), interleukin-6 (IL-6), interleukin-8 (IL-8), interleukin-10 (IL-10), interleukin-18 (IL-18), tumor necrosis factor-alpha (TNF-α), interleukin-1 receptor antagonist (IL-1RA), tumor necrosis factor receptor-I (TNF-RI), tumor necrosis factor receptor-II (TNF-RII)], and acute phase and endothelial activation markers [C-reactive protein (CRP), intercellular adhesion molecule-1 (ICAM-1), monocyte chemoattractant protein-1 (MCP-1), matrix metalloproteinase-9 (MMP-9), serum amyloid A (SAA), vascular cell adhesion molecule-1 (VCAM-1)]. In brief, the plates were coated with biotinylated antibodies combined with linkers and incubated overnight at 4°C. Plates were washed three times with PBS-Tween 0.05% after adding standards and diluted samples and incubated for 2 hours, followed by detection antibodies for 1 hour. A final wash step was performed before adding Read Buffer. Plates were immediately read using the Meso QuickPlex SQ 120, and concentrations were determined using Discovery Workbench 4.0 Analysis Software. Samples with undetectable biomarker levels were reported as a lower limit of detection (LLOD) for statistical purposes. Statistical Analysis Flow cytometry data from OMIQ Software was exported into Microsoft Excel and GraphPad Prism (version 10.4.1) for downstream analyses. Associations of neutrophil subpopulations with clinical variables and continuous laboratory variables were assessed using Pearson or Spearman correlations, while Pearson Point-Biserial correlation was used for dichotomous clinical variables. The significance of correlation values was determined by calculating the Pearson correlation t-statistic and using a two-tailed t-test, with p< 0.05 considered significant. Two‑group comparisons were performed using either Mann–Whitney test or the two-sample t-test. Multiple‑group comparisons were performed using either the Kruskal–Wallis test or one-way ANOVA. Following a significant overall F-test, post-hoc pairwise comparisons were conducted with appropriate multiple-testing corrections. Comparisons of two or more paired conditions were performed using the Wilcoxon signed-rank test with Friedman test, respectively. Random forest classification was implemented in R (version 4.4.3; R Core Team 2024) using the randomForest package (version 4.7–1.2). The dependent variable was vasopressor use (binary), and predictors included age, sex, white blood cell count, 24 soluble biomarkers, four CD16/CD62L subset percentages, and the mean fluorescence intensity of ten Q3 CD16hi/CD62Llo markers. Results Study Population A total of 35 sepsis patients and 18 healthy donors were included in this study. Baseline demographics and characteristics have been previously published (Table 1). The mean ages of sepsis patients and healthy donors were 66 and 64 years old, respectively, with similar distributions of sex and race. Intra-abdominal (29%) and pneumonia (23%) were the most common sites of infection, with 51% of infections being culture-positive for bacteria and 37% culture‑negative, while 8.6% were viral and 2.9% were fungal. Septic shock occurred in 57% of the patients, the mean SOFA score was 4.4, and in-hospital mortality was 9%. CD16hi/CD62Llo Neutrophils are Elevated in Patients with Sepsis We divided the CD15+ neutrophils into four subpopulations based on CD16 and CD62L expression and compared them as a proportion of the total circulating CD45+ leukocytes (Figure 1A). Among healthy donors, the largest population was the Q2 CD16hi/CD62Lhi neutrophils (Q2 = 48%) with very low levels of the other three subsets of neutrophils: Q1 CD16lo/CD62Lhi (5.1%), Q3 CD16hi/CD62Llo (7.4%), and Q4 CD16lo/CD62Llo (2%) (Figure 1B). A different distribution of neutrophil subsets was identified when evaluating sepsis patients, in which there was a decrease in the Q2 CD16hi/CD62Lhi subpopulation to 36% and an increase in the Q3 CD16hi/CD62Llo to 22.2% (Figure 1C). When comparing the percentage of the Q3 CD16hi/CD62Llo subset between HD and sepsis patients, we found a statistically significant increase in the Q3 CD16hi/CD62Llo subset in sepsis patients compared to HD (p<0.01) (Figure 1D). CD16hi/CD62Llo Neutrophils Associate with Disease Severity In sepsis patients, the percentage of Q3 CD16hi/CD62Llo neutrophils positively correlated with ICU admission (R=0.35, p=0.04), need for vasopressors (R=0.41, p=0.01) and mortality (R=0.39, p=0.02). Interestingly, the percentage of Q2 CD16hi/CD62Lhi, which decreased in sepsis patients compared to HDs, negatively correlated with septic shock (R=−0.42, p=0.01), need for ICU (R=−0.38, p=0.02) and vasopressors (R=−0.35, p=0.04) (Figure 2A). We further found an association with Q3 CD16hi/CD62Llo neutrophils and white blood cell count (R=0.51, p<0.01) (Figure 2B). When we compared biomarkers and clinical data in this cohort between those that survived and those that died, the latter group had higher SOFA scores (mean, SD; 8.0 [2.87] vs 4.03 [1.73], p<0.05), and higher white blood cell counts (mean, SD; 25 [12] vs 12 [7], p<0.05). Additionally, compared with patients that survived, the percentage of Q3 CD16hi/CD62Llo neutrophils was higher in those that died (mean, SD; 49 [21] vs 20 [20], p<0.05) (Table 2). Lastly, to identify if any of the CD16/CD62L neutrophil subpopulations or soluble biomarkers of interest can predict disease severity, we performed a random forest model. NGAL and MPO and the proportion of Q2 CD16hi/CD62Lhi and Q3 CD16hi/CD62Llo neutrophils were the most important biological variables to predict the need for vasopressor administration (Figure 2C). Additionally, to confirm our random forest model, we performed a multivariable regression analysis with adjustment for age and sex. This ranking demonstrated similar results with the proportion of Q3 CD16hi/CD62Llo neutrophils, Q2 CD16hi/CD62Lhi neutrophils, NGAL and MPO among the top 8 biomarkers that predict the need for vasopressors. Association of neutrophil subpopulations with clinical outcomes. (A) Heat map representing the correlation matrix comparing each quadrant with clinical outcomes (red and blue) with the adjacent blue and white matrix showing the corresponding p-values. (B) Correlation of Q3 CD16hi/CD62Llo neutrophils with WBC. (C) Feature plot of biomarkers and neutrophil populations that distinguish patients on vasopressors. Data are represented as correlation coefficients. Significance was determined using a Pearson two-tailed test. Q1: quadrant 1; Q2: quadrant 2; Q3: quadrant 3; Q4: quadrant 4; HD: healthy donor; ICU: intensive care unit; AKI: acute kidney injury. Table 2Clinical and Immunological Data in Those Who Lived versus DiedCharacteristicMortalityNo N = 32aMortalityYes N = 3ap-valuebSepsis Severity ICU Status0.5 Non-ICU, n (%)22/32 (69%)3/3 (100%) ICU, n (%)10/32 (31%)0/3 (0%) SOFA Score, mean (SD)4.03 (2.87)8.00 (1.73)0.03 Septic Shock, n (%)17/32 (53%)3/3 (100%)0.2 Mechanical Ventilation, n (%)6/32 (19%)2/3 (67%)0.12 Vasopressor (at time of blood draw), n (%)14/32 (44%)3/3 (100%)0.10 Length of Stay (LOS), days, mean (SD)9.6 (6.5)10.0 (9.2)>0.9 Age (years)64 (14)75 (6)0.10 Sex>0.9 Female, n (%)16/32 (50%)2/3 (67%) Male, n (%)16/32 (50%)1/3 (33%)Hematologic Characteristics White Blood Count, 103/uL, mean (SD)12 (7)25 (12)0.04 % Q1, mean (SD)12 (19)1 (0)0.2 % Q2, mean (SD)37 (25)27 (25)0.5 % Q3, mean (SD)20 (20)49 (21)0.04 % Q4, mean (SD)7 (10)10 (15)0.8Notes: an/N (%); Mean (SD). bFisher’s exact test; Wilcoxon rank-sum test. Values in bold indicate statistical significance. Clinical and Immunological Data in Those Who Lived versus Died Notes: an/N (%); Mean (SD). bFisher’s exact test; Wilcoxon rank-sum test. Values in bold indicate statistical significance. Characterization of CD16hi/CD62Llo Sepsis Neutrophils versus Healthy Donors and Other Sepsis Neutrophil Populations Compared to Q3 CD16hi/CD62Llo population in HDs, sepsis patients had lower CD10 (p<0.0001) and higher CD33 (p<0.05), indicating that the sepsis Q3 neutrophils are more immature. Additionally, they had decreased CD62L (p<0.0001) suggesting increased activation, and lower MPO (p<0.01), a marker of degranulation (Figure 3A). We next compared each feature across all four sepsis subpopulations (Figure 3B). We found that Q3 CD16hi/CD62Llo had the highest expression of the activation marker CD11b across all the sepsis quadrants, which was statistically higher than Q1 (p<0.001) and Q4 (p<0.0001). Interestingly, when directly comparing Q3 to Q2, Q3 CD16hi/CD62Llo had increased MPO (p<0.001) and lower CD62L (p<0.0001). Comparison of individual features amongst neutrophils subpopulations. (A) Individual bar graphs comparing the expression (MFI) of each feature in Q3 CD16hi/CD62Llo neutrophils between HD and sepsis patients. (B) Individual bar graphs comparing the expression (MFI) of each feature in Q3 CD16hi/CD62Llo neutrophils across all quadrants from sepsis patients. Data are represented as means±SEM. Significance was determined using a Mann–Whitney test, Unpaired t-test or a post-hoc Conover’s tests with Bonferroni correction following Friedman test and set at p<0.05*, p<0.01**, p<0.001***, p<0.0001****. Q1: quadrant 1; Q2: quadrant 2; Q3: quadrant 3; Q4: quadrant 4; HD: healthy donor; MFI: mean fluorescence intensity. Ex vivo Stimulation of Sepsis Neutrophils with LPS Stimulation of purified sepsis neutrophils with LPS (n=20) for one hour ex vivo significantly increased the percentage of the Q3 CD16hi/CD62Llo subset from 48% to 80% (p=0.0001), while significantly decreasing the percentage of the Q2 CD16hi/CD62Lhi subset from 37% to 10% (p=0.002) (Figure 4A and B). After noting that LPS stimulation of sepsis neutrophils was able to increase the percentage of Q3 CD16hi/CD62Llo neutrophils, we next wanted to evaluate changes in surface and intracellular markers on the Q3 CD16hi/CD62Llo subset of neutrophils following LPS stimulation. We observed that LPS simulation led to an up‑regulation in CD11b (p<0.0001), CD63 (p<0.01), CD10 (p<0.0001), CD33 (p<0.0001), and CD15 (p<0.0001) and a decrease in CD62L (p<0.05) (Figure 4C). Characterization of the impact of LPS stimulation on Q3 CD16hi/CD62Llo neutrophils. (A) Bar graph comparing the percentage of Q3 CD16hi/CD62Llo neutrophils from sepsis patients before and after LPS stimulation. (B) Bar graph comparing the percentage of Q2 CD16hi/CD62Lhi neutrophils from sepsis patients before and after LPS stimulation. (C) Bar graph comparing relative expression (MFI) of features among Q3 CD16hi/CD62Llo neutrophils from sepsis patients before and after LPS stimulation. Data are represented as means±SEM. Significance was determined using a Mann–Whitney test or a Wilcoxon signed‑rank test and set at p<0.05*, p<0.01**, p<0.001***, p<0.0001****. Abbreviations: NS, non-stimulated; LPS, Lipopolysaccharide. Spleen Tyrosine Kinase Expression is Increased in CD16hi/CD62Llo Sepsis Neutrophils and Associates with Mortality To determine novel therapeutic targets, we evaluated SYK expression in the different neutrophil populations. Compared to HDs, the Q3 CD16hi/CD62Llo population in sepsis patients had increased SYK expression (p<0.01) (Figure 5A). However, when we compared SYK expression across the sepsis neutrophil subsets, we found the highest SYK expression in Q1 CD16lo/CD62Lhi compared to Q2 CD16hi/CD62Lhi (p=0.002), Q3 CD16hi/CD62Llo (p<0.0001), and Q4 CD16lo/CD62Llo (p=0.002) neutrophils (Figure 5B). Interestingly, when we evaluated expression of each individual feature in the Q3 CD16hi/CD62Llo population we found a strong association with SYK expression in Q3 sepsis neutrophils with mortality (p<0.001). Characterization of the impact of LPS stimulation and SYK inhibition on Q3 CD16hi/CD62Llo neutrophils. (A) Bar graph comparing SYK expression in Q3 CD16hi/CD62Llo neutrophils from healthy donors versus sepsis patients. (B) Bar graph comparing SYK expression among neutrophils from sepsis patients across each quadrant. (C) Bar graph comparing the percentage of Q3 CD16hi/CD62Llo neutrophils from sepsis with LPS stimulation and SYK inhibition. (D) Individual bar graphs comparing the expression (MFI) of each feature among Q3 CD16hi/CD62Llo neutrophils from sepsis with LPS stimulation and SYK inhibition. Data are represented as means±SEM. Significance was determined using a Mann–Whitney test, a Wilcoxon test, or Tukey’s HSD test following one-way ANOVA and set at p<0.05*, p<0.01**, p<0.001***, p<0.0001****. Abbreviations: NS, non-stimulated; LPS, Lipopolysaccharide; SYK, spleen tyrosine kinase; MFI, mean fluorescence intensity. Spleen Tyrosine Kinase Inhibition Modulates CD16hi/CD62Llo Neutrophils Noting that LPS stimulation led to an increase in Q3 CD16hi/CD62Llo neutrophils, and regulated markers of neutrophil activation, we next sought to determine the impact of SYK inhibition. At the population level, pre-incubation of sepsis neutrophils with R406, the SYK inhibitor, prior to LPS stimulation, lowered the percentage of Q3 CD16hi/CD62Llo neutrophils from 80% to 69% (p<0.05) (Figure 5C). We next looked at the impact of SYK inhibition on each of the individual markers and found that at the individual feature level, R406 was able to inhibit the upregulation of CD11b (p<0.001), MPO (p<0.001), CD10 (p<0.001), CD33 (p=0.001), and CD15 (p=0.01) in sepsis Q3 CD16hi/CD62Llo neutrophils (Figure 5D). Discussion Sepsis is a significant global health priority that urgently requires a better understanding of pathogenesis, diagnosis, risk stratification, and treatment. In this study, we identify a subset of neutrophils (CD16hi/CD62Llo) that are increased in patients with sepsis and associate with clinical outcomes of disease severity, including the need for ICU admission, vasopressors, and mortality. Additionally, we found that SYK inhibition with R406, the in vitro active component of fostamatinib, was able to inhibit the generation of this neutrophil subset, indicating that it may be a potential therapeutic target in patients with sepsis. CD16hi/CD62Llo neutrophils have been characterized in sepsis‑associated acute respiratory distress syndrome (ARDS), trauma, and cancer.16,17,20 In sepsis‑associated ARDS, higher levels of CD16hi/CD62Llo neutrophils were associated with an increased risk of 28-day mortality and were identified to be an independent risk factor for the development of infectious complications.16 Our study confirms the finding that CD16hi/CD62Llo neutrophils in patients with sepsis are associated with increased mortality, potentially indicating that this subset of neutrophils can be used to stratify patients for mortality risk. CD16hi/CD62Llo are rapidly recruited from the bone marrow following intravenous administration of LPS to healthy volunteers and have been described to have dysfunctional phenotypes.21 The CD16hi/CD62Llo (described as CD16bright/CD62Ldim) neutrophil phenotype has also been described as immunosuppressive based on their ability to prevent T-cell activation and proliferation in a MAC-1 (CD11b/CD18)-dependent) mechanism.22 While we do not have evidence that the neutrophil population can functionally suppress T-cells, the Q3 CD16hi/CD62Llo population identified by our group expresses high levels of CD11b. Additionally, CD16bright/CD62Ldim neutrophils isolated from healthy volunteers and challenged with LPS have decreased antimicrobial capacity as measured by their ability to prevent bacterial proliferation or time to unrestrained growth compared to CD16dimCD62Lbright neutrophils.23 While no defect in phagocytosis was found in the CD16bright/CD62Ldim neutrophils, a higher pH in their phagosomes and low concentrations of hypochlorous acid were measured, leading to impaired bacterial killing. Trauma patients who have neutrophils with decreased acidification in the days following the trauma, have been shown to have an increased risk of infection, providing another potential link in this neutrophil population between pH levels and the immunosuppressive phenotype.24 Lastly, CD16dimCD62Lbright have been shown to have decreased adhesion to endothelium under flow conditions, highlighting further how this subset of neutrophils propagates to the dysregulated innate immune response during sepsis.25 In addition, to understand cellular phenotypes that associate with disease severity, there is great interest in finding treatable traits that may be amenable to a targeted host-directed therapy. In this study, we found that SYK was elevated in the CD16hi/CD62Llo neutrophil population in sepsis patients compared to HDs, indicating that SYK may be a therapeutic target that can be modulated to improve outcomes of patients with sepsis. Further evidence for this comes from the ability of R406 to inhibit the generation of CD16hi/CD62Llo neutrophils following ex vivo stimulation with LPS. Prior work from our group has shown that SYK expression is increased in neutrophils in a nonhuman primate model of sepsis and from patients with sepsis and associates with disease severity.19,26 In a study in which healthy donor neutrophils were stimulated with LPS, we found that SYK inhibition was able to modulate important end effector functions of neutrophils, including the release of neutrophil extracellular traps, reactive oxygen species, degranulation, and the adherence of neutrophils to endothelial cells.15 Furthermore, we have demonstrated that the activation of neutrophils, as measured by the regulation of extracellular neutrophil markers, from HDs and sepsis patients, can be modulated with SYK inhibition. In this study, we demonstrate that SYK inhibition can prevent the appearance of cellular populations that are associated with disease severity, including death, in sepsis patients. Given this neutrophil population is associated with immunosuppression and the development of infectious complications, mechanistically, R406 may be able to not only inhibit the acute inflammation associated with sepsis but also modulate the concomitant immunosuppression, ultimately restoring neutrophil homeostasis. Further evidence that SYK inhibition can restore neutrophil homeostasis comes from previous COVID-19 studies, where fostamatinib was able to decrease the activation and proportion of circulating neutrophils from patients, thereby allowing the restorations of immunosuppressive phenotypes of circulating HLA-DR negative monocytes.15,27–29 Our study does have limitations. First, the population was drawn from a single center within a small number of deaths, and therefore its generalizability to other patient cohorts needs to be confirmed. Larger multicenter validation cohorts are needed. Additionally, we were unable to purify the CD16hi/CD62Llo subset to assess other functional characteristics of this population that may be different from the CD16hi/CD62Llo neutrophils from HD, other sepsis neutrophils subsets, or CD16hi/CD62Llo neutrophils from other diseases. In conclusion, we found that in patients with sepsis, CD16hi/CD62Llo neutrophils are associated with disease severity, including mortality, indicating that these neutrophils may aid in sepsis risk stratification. Additionally, these CD16hi/CD62Llo neutrophils express high levels of SYK, and their activation can be modulated with SYK inhibition, providing further evidence that SYK may be a therapeutic target to restrain the dysregulated innate immune response that defines sepsis. However, given the exploratory nature of this study, further mechanistic studies are needed to confirm the direct contribution of CD16hi/CD62Llo neutrophils to sepsis pathogenesis and their role as a biomarker. Funding Statement This work has been funded in part by the Intramural Research Program of the National Institutes of Health Clinical Center and National Heart Lung and Blood Institute. Disclaimer This research was supported in part by the Intramural Research Program of the National Institutes of Health (NIH). The contributions of the NIH authors are considered works of the United States Government. The findings and conclusions presented in this paper are those of the authors and do not necessarily reflect the views of the NIH or the US Department of Health and Human Services. Data Sharing Statement Data available upon request by Email to the corresponding author. Ethics Approval Due to realignment in research infrastructure and rapidly growing research program, the Inova Fairfax Hospital Investigational Review Board (IRB) outsourced IRB activities to Western IRB (WIRB). All patients were enrolled under protocol #15–1863. Author Contributions Xianglan Yao: Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Validation, Writing - original draft, Writing - reviewing and editing. Heather L Teague: Conceptualization, Data curation, Formal Analysis, Investigation, Methodology, Validation, Visualization, Writing – reviewing and editing. Lauren E Knabe: Investigation, Writing - reviewing and editing. Raquel Santana da Cruz: Investigation, Writing – reviewing and editing. Trenton Williams: Formal Analysis, Investigation, Writing – reviewing and editing. Merte Lemma Woldehanna: Investigation, Writing – reviewing and editing. Ernest Evans: Investigation, Writing – reviewing and editing. Amir Hobson: Investigation, Writing – reviewing and editing. Jared D Wilkinson: Investigation, Writing – reviewing and editing. Cumhur Yusuf Demirkale: Investigation, Writing – reviewing and editing. Steven D Nathan: Funding acquisition, Investigation, Resources, Writing – reviewing and editing. Christopher King: Investigation, Writing – reviewing and editing. Jeffrey R Strich: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing – original draft, Writing – reviewing and editing. All authors have given final approval of the version to be published, have agreed on the journal to which the article has been submitted, and have agreed to be accountable for all aspects of the work. Disclosure Dr. Christopher King reports Honoraria from United Therapeutics, outside the submitted work. Dr. Jeffrey Strich reports having a material transfer agreement with Rigel Pharmaceutical to obtain R406. The authors declare that there are no other conflicts of interest in this study. References Associated Data Data Availability Statement Data available upon request by Email to the corresponding author.