Streptococcus pyogenes Infection and CD4⁺ T Cells Crosstalk Promote an Antigen-Presenting Cell-Like Phenotype in Neutrophils

Abstract

Introduction

Neutrophils are the most abundant leukocytes in human blood and a key component of host defense against Streptococcus pyogenes (group A streptococci [GAS]). They are rapidly recruited to the site of infection, where they mediate phagocytosis, degranulation, and release of neutrophil extracellular traps, thereby influencing both bacterial clearance and tissue damage. Traditionally considered short-lived effector cells, neutrophils are increasingly recognized for their immunomodulatory functions, including regulation of adaptive immunity.

Methods

Human neutrophils were infected with GAS wild-type 5448 or isogenic mutants lacking streptolysin O (Δslo) or streptolysin S (ΔsagA) and analyzed either after infection alone or after infection followed by co-culture with autologous CD4⁺ T cells. Neutrophil activation and T cell responses were evaluated via flow cytometry.

Results

GAS infection robustly activated neutrophils in a dose-dependent manner, driving expansion of the CD15^bright/CD66b^bright population and heparin-binding-protein release. Infection prompted neutrophils to acquire antigen-presenting cell (APC)-like characteristics, including HLA-DR, CD40, and CD86 expression, particularly when exposed to autologous CD4⁺ T cells. In 5448ΔsagA infections, neutrophils showed lower expression of costimulatory markers, and bacteria were more susceptible to intracellular killing. In co-cultures, CD4⁺ T cells were partially activated, as indicated by CD25 upregulation and Th1- and Th17-cell-associated cytokine release.

Conclusion

GAS-infected neutrophils acquire an APC-like phenotype and partially modulate CD4⁺ T cell activation, revealing a previously unrecognized role for infected neutrophils in shaping adaptive immunity in streptococcal infections.

Plain Language Summary

Neutrophils are among the first immune cells to respond to bacterial infections and are best known for directly killing invading microbes. In this study, we show that during streptococcal infection, neutrophils also influence how other immune cells respond. Infected neutrophils become strongly activated and adopt features that allow them to communicate with T cells, thereby shaping immune responses beyond their classical antimicrobial role. These findings highlight an unexpected way in which neutrophils can help regulate immune reactions during infection, with potential consequences for disease severity and immune balance.

Introduction

Neutrophils are the most abundant leukocytes in circulation and play a crucial role in the innate immune defense. They are among the first recruited responders at sites of tissue damage or infection, rapidly migrating in response to inflammatory signals [1]. Equipped with a broad array of pattern recognition receptors, including Toll-like receptors (TLRs), neutrophils recognize diverse microbial and danger signals [2, 3]. Activated neutrophils deploy multiple antimicrobial strategies, including phagocytosis with granule-mediated killing, degranulation of antimicrobial effector molecules, and the extrusion of neutrophil extracellular traps (NETs) [4–6].

Beyond the direct innate effector functions, neutrophils are increasingly recognized as potential modulators of adaptive immunity. In BCG infections, it was shown that they transport antigens to lymphoid tissues [7]. In co-cultures, they modulate the activity of professional antigen-presenting cells (APCs), such as dendritic cells (DCs), which in turn upregulate surface expression of CD86 and major histocompatibility complex class II (MHC II, HLA-DR) [8]. Similarly, extracellular vesicles released by Mycobacterium tuberculosis (Mtb)-infected neutrophils promote increased CD86 and MHC II expression in monocyte-derived DCs [9]. HLA-DR provides the primary activation signal to T cells through interaction with the T cell receptor (TCR). Full T cell activation requires a secondary costimulatory signal, which is mediated by engagement of CD40 or the B7 receptors CD80 and CD86 on the APCs with CD40 ligand (CD40L) or CD28 on the T cells, respectively [10, 11]. In healthy individuals, circulating human neutrophils typically lack expression of HLA-DR and classical costimulatory molecules [12]. However, exposure to cytokines such as granulocyte-macrophage colony-stimulating factor (GM-CSF) and interferon γ (IFN-γ) can induce expression of HLA-DR as well as B7 receptors [13, 14]. Once expressed, these molecules enable neutrophils to present antigens, including tetanus toxoid or Staphylococcus aureus superantigens to CD4⁺ T cells [13, 14]. Recent work has also shown that neutrophils can upregulate HLA-DR and costimulatory molecules when exposed simultaneously to antigen and cognate memory CD4⁺ T cells [15]. Furthermore, the study reported that APC-like neutrophils were able to induce proliferation of antigen-specific CD4⁺ T cells upon co-culture [15]. Furthermore, neutrophils are no longer viewed as a homogeneous population. Multiple subsets with distinct immunomodulatory functions toward CD4⁺ T cells have been identified. Neutrophils expressing MHC II retain classical antimicrobial effector activities while also exerting regulatory effects on CD4⁺ T cells, including suppression of T cell activation, as demonstrated in cattle and mice [16]. Functional diversity among neutrophils is further linked to their maturation status. Mature CD10⁺ neutrophils display suppressive effects on CD4⁺ T cells, whereas CD10⁻ neutrophils promote T cell proliferation and cytokine production [17]. In addition, neutrophils have been broadly categorized into pro- and anti-inflammatory phenotypes [18]. In tuberculosis, pro-inflammatory N1-like neutrophils correlate with effective immunity against Mtb, while low responders predominantly exhibit anti-inflammatory N2-like neutrophils [18]. Together, these findings highlight the remarkable plasticity of neutrophils in shaping adaptive immune responses.

S. pyogenes (group A streptococci [GAS]) causes a wide spectrum of diseases, ranging from mild infections such as pharyngitis and impetigo to severe invasive manifestations, including bacteremia, necrotizing soft tissue infections (NSTIs), and streptococcal toxic shock syndrome [19]. Recent reports from several countries indicated a significant increase in invasive GAS (iGAS) infections, especially among children and the elderly [20, 21]. iGAS infections progress rapidly, with high mortality and systemic complications driven by the interplay of dysregulated host inflammatory responses and bacterial virulence factors [22, 23].

Neutrophil manipulation is central to GAS pathogenesis. GAS can evade neutrophil responses by degrading chemoattractants through virulence factors such as the cell envelope proteinase SpyCEP [24] and streptococcal C5a peptidase (ScpA) [25]. In addition, streptolysin S (SLS) was shown to block early neutrophil recruitment to the site of infection, thereby promoting immune evasion and bacterial dissemination [26]. Conversely, GAS also trigger robust neutrophil activation, leading to degranulation and the release of pro-inflammatory mediators [27, 28]. Streptococcal M protein and phosphoglycerate kinase (PGK) have been identified as potent inducers of neutrophil activation and degranulation [29, 30]. Streptolysin O (SLO) exerts opposing effects on neutrophils, suppressing or promoting degranulation and NET formation depending on the local environment [31, 32]. Patient studies showed that neutrophil-derived molecules released during degranulation correlate with sepsis severity [33] and tissue pathology in GAS NSTIs [34–36], underscoring the pathogenic impact of streptococcal modulation of neutrophil function.

NSTIs caused by GAS are characterized by a massive neutrophil influx into infected tissues, accompanied by the presence of T cells [28, 34, 37]. While neutrophils are traditionally regarded as short-lived phagocytes specialized in microbial killing [4–6, 38], their overwhelming abundance in NSTIs raises the question of whether they may also contribute to antigen presentation. In this context, neutrophils often vastly outnumber professional APCs such as DCs, suggesting that they could exert a substantial influence on local T cell responses within a confined tissue environment. Investigating the potential role of neutrophils as APCs in streptococcal NSTIs may, therefore, uncover previously unrecognized mechanisms of host-pathogen interaction in these life-threatening infections.

Methods

Bacterial Strains

The following GAS strains were used: 5448 [39], 5448Δslo [40], and 5448ΔsagA [41]. The allelic exchange of the target genes was previously validated, and potential off-target mutations were ruled out [40, 41]. All strains were cultivated in Todd-Hewitt broth (Carl Roth) supplemented with 1.5% (w/v) yeast extract (Carl Roth) at 37°C. The growth and colony-forming unit (CFU) counts of the overnight cultures were evaluated and showed no differences between the strains (online suppl. Fig. S1a, b; for all online suppl. material, see https://doi.org/10.1159/000551693).

Isolation and Infection of Human Neutrophils

Human neutrophils were isolated from healthy donors using a density gradient centrifugation on PolymorphPrep (Axis Shield). For each experiment, neutrophils were isolated on different days from independent donor sets, with no more than two donors processed per day. Granulocyte purity (99.9 ± 0.1%), eosinophil contamination (3.2 ± 0.7%), and post-isolation viability (99.5 ± 0.2%) were assessed by flow cytometry. After isolation, neutrophils were suspended in RPMI 1640 medium supplemented with 10 mm L-glutamine (Gibco), 25 mmol/L HEPES (Cytiva), and 5% fetal calf serum. Neutrophils (5 × 10⁵) were infected with bacteria at a multiplicity of infection (MOI) of 0.1/1/10 for 30 min or 1 h. Infections were carried out in 24-well plates using 1 mL of supplemented RPMI 1640 per well. The bacterial inoculum was determined based on CFU counts from overnight cultures. After infection, neutrophils were directly prepared for flow cytometry. Unstimulated cells and cells stimulated with 100 nm phorbol-12-myristate-13-acetate (PMA; Carl Roth) were used as negative and positive controls, respectively.

To assess the kinetics of intracellular bacterial survival, 5 × 10⁵ neutrophils were infected at MOI 1. After 30 min, antibiotics (100 µg/mL streptomycin and 100 IU/mL penicillin G [Hyclone]) were added to kill extracellular bacteria. After an additional 30 or 90 min of incubation, the cells were lysed, and the intracellular bacteria were plated on blood agar plates (Oxoid, Basingstoke, UK).

Human Resistin and Human Heparin-Binding Protein Enzyme-Linked Immunosorbent Assay

Heparin-binding protein (HBP) and resistin concentrations in supernatants of neutrophils infected for 1 h were analyzed via commercially available ELISAs (HBP from Axis Shield; resistin from BioVendor) according to the manufacturer’s instructions.

Co-Culture of Neutrophils with Autologous CD4⁺ T Cells

Autologous CD4⁺ T cells were isolated using the EasySep Direct Human CD4⁺ T Cell Isolation Kit (Stemcell). Purity (94.6 ± 1.4%) and viability (98.5 ± 0.7%) after isolation were assessed via flow cytometry. Neutrophils were infected at MOI 1 as described above, and antibiotics were added for 1 h to kill extracellular bacteria. Next, half of the medium was removed, and fresh medium containing autologous CD4⁺ T cells, was added at a neutrophil:CD4⁺ T cell ratio of 1:5. After 22 h, neutrophils were prepared for flow cytometry analysis.

For the assessment of T cell activation, CD4⁺ T cells were labeled with 5 µm Tag-it Violet proliferation dye (BioLegend) according to manufacturer’s instructions prior to co-cultivation with neutrophils. Labeled CD4⁺ T cells were incubated at a neutrophil:CD4⁺ T cell ratio of 1:5 for 48 h or 96 h in 2 mL supplemented RPMI 1640. Unstimulated cells and cells stimulated with 2% Phytohemagglutinin M (PHA-M; Capricorn) were used as negative and positive controls, respectively.

Flow Cytometry

Neutrophils and T cells were labeled using the Zombie Aqua Fixable Viability Kit (BioLegend) to stain dead cells. Human TruStain FcX (BioLegend) was used according to the manufacturer’s instructions to block unspecific binding of immunoglobulins to Fc receptors. Cells were incubated with fluorochrome-labeled monoclonal antibodies for 30 min at 4°C in the dark and washed between each step. The following antibodies were used for neutrophil analyses (target, clone, fluorochrome, all BioLegend): cluster of differentiation (CD) 15 (W6D3, Alexa Fluor 700), CD66b (G10F5, Pacific Blue), CD193 (5E8, APC), CD45 (HI30, BV605), CD40 (5C30, APC), CD80 (2D10, BV711), CD86 (BU63, PE/Cyanine7) and HLA-DR (L243, FITC). In addition to CD45, the following antibodies were used for T cell analyses: CD3 (HIT3a, PerCP/Cyanine5.5), CD4 (SK3, Alexa Fluor 700), CD25 (BC96, APC), CD69 (FN50, KIRAVIA Blue 520), CD134 (ACT35, PE-Fire 810), CD137 (4B4-1, PE). The respective gating strategies are shown in online supplementary Figures S2a, b and S6a. Data were acquired using a FACSAria III flow cytometer and FACS DIVA 8.0 Software (both BD Biosciences, San Jose, CA, USA) and analyzed using FCS Express 7 Software (De Novo Software).

Cytokine Measurements

Cytokine concentrations in cell culture supernatants of neutrophil-CD4⁺ T cell co-cultures were measured via LEGENDPlex human inflammation panel 1 (13-plex) kit (BioLegend) according to the manufacturer’s instructions. Data were acquired with a FACSAria III flow cytometer using FACS DIVA Software (both BD Bioscience) and analyzed using LEGENDPlex software (BioLegend).

Statistics

Friedman tests with Dunn’s post hoc were applied to paired non-normally distributed data, including neutrophil surface marker expression, 24 h cytokine secretion, and intracellular bacterial survival. Non-paired, non-normal data were analyzed using Kruskal-Wallis tests with Dunn’s post hoc. Statistics were performed using GraphPad Prism version 8. A p value <0.05 was considered significant.

Results

GAS Infections Activate Neutrophils

To assess neutrophil degranulation, hereafter referred to as activation, following GAS 5448 infection, two approaches were used: (i) quantification of HBP and resistin in culture supernatants (Fig. 1a, b) and (ii) analysis of surface granule markers CD15 and CD66b by flow cytometry (Fig. 1c; online suppl. Fig. S3a–c, e). Neutrophils were infected at different MOIs (0.1, 1, 10) for 30 min or 1 h. For the initial assessment of neutrophil activation by flow cytometry, both viable and dead cells were included (Fig. 1c; online suppl. Fig. S3a, b, e). To control for potential lytic effects, isogenic mutants lacking either SLO (5448Δslo) or SLS (5448ΔsagA) were also used. Infections for 30 min caused only minor neutrophil activation, regardless of the MOI used (online suppl. Fig. S3e). Infections for 1 h induced the strongest neutrophil activation at MOI 10, as reflected by robust HBP release (Fig. 1a) and expansion of the CD15^bright/CD66b^bright population (Fig. 1c). In contrast, low-dose infections (MOI 0.1) failed to activate neutrophils (Fig. 1c). MOI 1 infections resulted in moderate neutrophil activation (Fig. 1c) but did not trigger detectable HBP or resitin release (Fig. 1a, b). Analysis restricted to viable CD15^bright/CD66b^bright neutrophils revealed similar levels of activation across all three strains. Notably, infections at an MOI 1 for 1 h resulted in the highest proportion of viable, activated neutrophils (online suppl. Fig. 3c). Overall, resistin release was not observed under any infection condition (Fig. 1b). These findings suggest that all three strains effectively induce neutrophil activation. However, all GAS infections were also associated with dose-dependent cytotoxic effects on neutrophils, which were not mitigated by the use of 5448Δslo or 5448ΔsagA mutants (Fig. 1d; online suppl. Fig. S3d). Based on these observations, all subsequent neutrophil infections were performed at MOI 1 to balance effective activation with optimal cell viability for downstream analyses.

Fig. 1.

Activation of neutrophils in response to GAS infections. Neutrophils were infected for 60 min with the indicated GAS strains at MOI 0.1, 1, or 10. Culture supernatants were analyzed for HBP (a) and resistin (b). Cell-associated activation (c) and viability (d) were assessed via flow cytometry. PMA-stimulated and unstimulated cells served as positive and negative controls, respectively. Representative histograms with numbers indicating marker-positive cells are shown in the left panels of (c) and (d). Each dot represents one independent experiment with cells from one donor (n = 10). Horizontal lines denote median values. The level of significance was determined using the Kruskal-Wallis test with Dunn’s posttest. p values indicate comparisons to the unstimulated control. Unstim, unstimulated; FMO, fluorescence minus one; MOI, multiplicity of infection.

GAS-Infected Neutrophils Acquire APC-like Features in Co-Culture with CD4⁺ T Cells

Since neutrophils and T cells are both present as infiltrating leukocytes in GAS NSTI patient biopsies [35–37], we investigated whether they can mutually influence each other in GAS infections. Therefore, neutrophils were first infected with GAS for 1 h, after which extracellular bacteria were eliminated using antibiotics for an additional hour. As T cells were shown to be required for the induction of APC-like features in neutrophils [15], autologous CD4⁺ T cells were added, and cells were co-incubated for 22 h in the continued presence of antibiotics.

Neutrophil viability showed a modest reduction as compared to the initial 1 h infections (Fig. 1d; online suppl. Fig. S4a), with a more pronounced decline observed during co-culture with CD4⁺ T cells. In co-cultures, however, 5448- and 5448Δslo-infected neutrophils displayed a marked increase in HLA-DR surface expression, which was even further elevated by the presence of T cells (Fig. 2a, b). CD40 expression remained low in neutrophils infected alone but was upregulated in all conditions in the presence of CD4⁺ T cells, except for 5448ΔsagA infections (Fig. 2c). Analyses of B7 receptors showed that CD80 expression remained at baseline levels across all conditions (online suppl. Fig. S4b). In contrast, CD86 expression was upregulated in response to 5448 and 5448Δslo infections (Fig. 2d). Increased frequencies of CD86 on infected neutrophils persisted following co-culture with CD4⁺ T cells (Fig. 2d). This effect was absent in 5448ΔsagA infections (median values are provided in online suppl. Table 1).

Fig. 2.

GAS-infected neutrophils acquire APC-like features. Neutrophils were infected with the indicated strains, followed by antibiotic treatment to eliminate extracellular bacteria, and subsequently incubated alone (−) or with autologous CD4⁺ T cells (+). After 24 h, surface expression of HLA-DR (a, b), CD40 (a, c), and CD86 (a, d), as well as co-expression of HLA-DR and CD86 (e), was analyzed via flow cytometry. Representative histograms with numbers indicating marker-positive cells are shown in (a). Each dot represents one independent experiment with cells from one donor (n = 10). Lines connect paired measurements with cells from the same donor and bars denote mean values. The level of significance was determined using the Friedman test with Dunn’s posttest. Unstim, unstimulated; FMO, fluorescence minus one.

Co-expression of HLA-DR with the costimulatory molecules is a prerequisite for the productive activation of T cells [10, 11]. Therefore, HLA-DR expression together with CD40, CD80, or CD86 was assessed (Fig. 2e; online suppl. Fig. S4c, d; online suppl. Table 1). Infections with 5448 and 5448Δslo led to a significant increase in the frequencies of CD86⁺/HLA-DR⁺ neutrophils following single infections, which was maintained in co-cultures with CD4⁺ T cells (Fig. 2e). In co-cultures, elevated proportions of CD40⁺/HLA-DR⁺ cells were also observed, whereas no increase in CD80⁺/HLA-DR⁺ neutrophils was detected (online suppl. Fig. S4c, d; online suppl. Table 1). In contrast, 5448ΔsagA infections did not promote expansion of double-positive neutrophils (Fig. 2e; online suppl. Fig. S4c, d; online suppl. Table 1). These findings demonstrate that a subset of neutrophils acquires APC-like features during GAS infection, and that SLS contributes to this phenotype.

Although some professional APCs are able to present exogenous antigens directly, phagocytosis and intracellular processing remain critical steps in the antigen presentation pathway [42]. To assess whether differences in bacterial uptake could explain the lack of an APC-like phenotype in 5448ΔsagA-infected neutrophils, bacterial uptake and intracellular killing were determined at two time points (Fig. 3). After 60 min, bacterial counts were comparable across all infections. One hour later, a marked decline in intracellular bacteria was observed. However, 5448ΔsagA was more efficiently eliminated as compared to wild-type infections. These findings indicate that the reduced expression of APC markers in 5448ΔsagA-infected neutrophils is not linked to impaired phagocytosis.

Fig. 3.

Intracellular survival of GAS in neutrophils. Neutrophils were infected with the indicated strains at MOI 1. After 30 min, extracellular bacteria were eliminated by adding antibiotics. Intracellular bacterial survival was quantified by CFU enumeration at 60 and 120 min post-infection. Each dot represents an independent experiment using neutrophils from one donor (n = 7). Horizontal lines denote median values. The level of significance was determined using the Friedman test with Dunn’s posttest. CFU, colony-forming units.

Efficient antigen presentation requires a third signal, in addition to TCR engagement and costimulatory signals, consisting of cytokines that direct CD4⁺ T cell polarization [43]. To investigate how APC-like neutrophils influence T cell responses through cytokine production and how CD4⁺ T cells and bacterial infection jointly shape the cytokine environment and neutrophil APC-like features, cytokine concentrations were measured in cell culture supernatants after 24 h of co-culture. For most cytokines, increased concentrations were observed under at least one condition, either CD4⁺ T cell co-culture or infection. IFN-α2, IL-10, IL-12p70, IL-17A, and IL-33 were undetectable in all samples (online suppl. Fig. S5a–e). The presence of MCP-1 and IL-18 was strictly dependent on CD4⁺ T cell co-culture and unaffected by GAS infection (Fig. 4d, g). Conversely, infection with the three GAS strains induced robust TNF-α, IL-6, and IL-8 secretion in the absence of T cells, which was further enhanced by CD4⁺ T cell co-culture (Fig. 4c, e, f). For IL-1β, IFN-γ, and IL-23, both GAS infection and T cell co-culture were required for detectable secretion as either condition alone yielded concentrations near the detection limit (Fig. 4a, b, h). Across all cytokines, the three GAS strains elicited comparable responses, indicating that SLS and SLO did not significantly affect cytokine secretion by neutrophils or CD4⁺ T cells under these conditions, and that reduced APC-marker expression during 5448ΔsagA infection did not impair cytokine production.

Fig. 4.

Co-culture of CD4⁺ T cell with infected neutrophils induces differential cytokine secretion. Neutrophils were infected with the indicated strains, followed by antibiotic treatment to eliminate extracellular bacteria, and subsequently incubated for 22 h alone (−) or with autologous CD4⁺ T cells (+). Concentrations of IL-1β (a), IFN-γ (b), TNF-α (c), MCP-1 (d), IL-6 (e), IL-8 (f), IL-18 (g), and IL-23 (h) were measured in culture supernatants. Each dot represents one independent experiment with cells from one donor (n = 10). Dashed lines indicate the detection limit. The level of significance was determined using the Friedman test with Dunn’s posttest. unstim, unstimulated.

GAS-Infected Neutrophils Partially Induce CD4⁺ T Cell Activation

Given that neutrophils expressed HLA-DR and costimulatory molecules, and that various cytokines were detected in supernatants from infected neutrophils co-cultured with CD4⁺ T cells, their capacity to activate autologous CD4⁺ T cells was subsequently assessed. To evaluate the overall ability of neutrophils to present GAS antigens, infections with strain 5448 were analyzed. CD4⁺ T cell activation was evaluated by monitoring CD25, CD69, CD134, and CD137 expression as well as proliferation (Fig. 5a–d; online suppl. Fig. S7a–e; median values are provided in online suppl. Table 2). Considering the different activation kinetics of naive and memory T cells [44], incubation periods of 48 h and 96 h were used. No upregulation of CD69, CD134, CD137, or proliferation was detected at either time point (Fig. 5a, c, d; online suppl. Fig. S7b–e). In contrast, frequencies and expression of CD25⁺ cells increased after 96 h in response to 5448-infected neutrophils (Fig. 5b; online suppl. Fig. S7a). However, while a proportion of neutrophils remained viable at 48 h, almost all neutrophils were dead by 96 h (online suppl. Fig. S6b). In contrast, CD4⁺ T cells maintained high viability throughout the entire time course (online suppl. Fig. S6c).

Fig. 5.

CD4⁺ T cell responses to GAS 5448-infected neutrophils. Autologous CD4⁺ T cells were co-cultured with 5448-infected neutrophils for 48 h or 96 h. Frequencies of CD25⁺ (a, b), CD69⁺ (a, c), and CD134⁺ (a, d) CD4⁺ T cells were quantified via flow cytometry. CD4⁺ T cells co-cultured with uninfected neutrophils served as the negative control, while CD4⁺ T cells co-cultured with uninfected neutrophils and stimulated with PHA served as the positive control. Representative histograms with numbers indicating marker-positive cells are shown in a. Each dot represents an individual donor (n = 10). Lines connect paired measurements with cells from the same donor. The level of significance was determined using the Kruskal-Wallis test with Dunn’s posttest. FMO, fluorescence minus one; unstim, unstimulated; PHA, phytohemagglutinin M-type.

Activated CD4⁺ T cells are defined not only by upregulation of surface activation markers, but also by their distinct cytokine profile [43]. Accordingly, cytokine concentrations in supernatants from co-cultures were measured at 48 and 96 h. Of the 13 cytokines analyzed, nine were detectable under at least one experimental condition (Fig. 6a–i; online suppl. Fig. S8a–d). Furthermore, the secretion occurred almost exclusively in infections (Fig. 6a–i). IL-1β reached the highest concentration at 48 h but fell below the detection limit by 96 h (Fig. 6a). In contrast, IFN-γ and TNF-α were elevated at 48 h and remained high at 96 h (Fig. 6b, c). IL-6, IL-8, IL-12p70, and IL-23 were consistently elevated at both time points (Fig. 6d, e, g, i), while IL-17A and IL-10 were only detected at 96 h (Fig. 6f, h). Taken together, these results indicate that GAS-infected neutrophils can induce functional activation of CD4⁺ T cells, as evidenced by robust cytokine production, even in the absence of pronounced T cell proliferation.

Fig. 6.

Cytokine profile of co-cultures of CD4⁺ T cell and GAS-infected neutrophils. Autologous CD4⁺ T cells were co-cultured with 5448-infected neutrophils for 48 h or 96 h. Concentrations of IL-1β (a), IFN-γ (b), TNF-α (c), IL-6 (d), IL-8 (e), IL-10 (f), IL-12p70 (g), IL-17A (h), and IL-23 (i) were measured in culture supernatants. CD4⁺ T cells co-cultured with uninfected neutrophils served as the negative control. Each dot represents one independent experiment with cells from one donor (n ≥ 9). Dashed lines indicate the detection limit. The level of significance was determined using the Kruskal-Wallis test with Dunn’s posttest. unstim, unstimulated.

Discussion

Neutrophils play a central role in GAS infections, contributing not only to pathogen clearance but also to hyper-inflammatory responses and tissue pathology. Upon recruitment to the site of infection, neutrophils combat bacteria through phagocytosis, degranulation, and the release of NETs [4–6]. Recent studies have highlighted that neutrophils can also perform functions beyond their classical roles, notably including antigen presentation [15, 45]. While GAS-mediated modulation of neutrophil effector functions has been studied extensively, investigations into the induction of APC-like characteristics in neutrophils remain limited. Given the prominent presence of neutrophils and T cells at the site of infection in GAS NSTIs [35, 37], the capacity of GAS to induce antigen-presenting features in neutrophils was investigated. GAS infections activated neutrophils and, in the presence of CD4⁺ T cells, promoted the expression of APC-like markers, with SLS contributing to this phenotype.

Neutrophil activation was evaluated by analyzing the surface expression of CD66b and CD15 alongside the assessment of neutrophil degranulation markers HBP and resistin. Although secretion of resistin remained at the base levels, HBP release was significantly elevated following infection with all three strains at highest MOI tested. Both HBP and resistin are stored in azurophilic granules; however, the differential secretion likely reflects the additional presence of HBP in secretory vesicles, which are released early during neutrophil extravasation [46]. This suggests that neutrophils did not undergo full degranulation as granule release occurs sequentially [47]. Nevertheless, flow cytometry analyses revealed that even at lower MOI 1, a substantial proportion of neutrophils exhibited activation in response to GAS infections, consistent with previous reports on GAS-mediated neutrophil activation [48]. Furthermore, neither SLO nor SLS appeared to affect neutrophil activation. SLO has been reported to either suppress or enhance pro-inflammatory neutrophil responses depending on the environmental context [31, 32]. Under experimental conditions used, the 1-hour infection period likely limited any detectable influence of SLO on neutrophil activation or viability. Likewise, our findings on SLS are consistent with previous reports showing no significant impact on the viability of human neutrophils [49], in contrast to its reported effects on murine neutrophils [50].

The process of antigen presentation involves several coordinated steps. The initial steps include the uptake of extracellular antigens and their digestion into peptide fragments [42]. Granule compartments in neutrophils are specialized for rapid microbial killing [51, 52], which may lead to accelerated degradation of phagocytosed material and thus limit effective antigen processing. Nevertheless, despite the high lysosomal activity and rapid degradation of phagocytosed material, neutrophils are still capable of generating antigenic peptides [53]. In our experimental setup, neutrophils ingested all GAS strains with comparable efficiency, while intracellular killing was slightly more effective against 5448ΔsagA. This observation is consistent with previous reports showing that SLS contributes to resistance against phagocytic killing [41].

Another essential step in antigen presentation is the loading of peptides onto MHC II molecules and their subsequent display on the cell surface. Unlike professional APCs, which constitutively express MHC II and can load antigens via both surface binding and intracellular proteolysis [54], neutrophils were long thought to require prior stimulation to upregulate MHC II. Cytokines such as granulocyte-macrophage colony-stimulating factor, IFN-γ, IL-3, and TNF-α have been shown to provide such signals [13, 14, 55]. In our experimental setup, neutrophils readily expressed HLA-DR on their surface in response to wild-type and 5448Δslo infections without prior cytokine stimulation. Moreover, co-culture with autologous CD4⁺ T cells further enhanced HLA-DR expression under these conditions. Furthermore, effective antigen presentation requires the expression of costimulatory molecules such as CD40 and B7 family receptors. Notably, CD40 expression was induced solely by the presence of CD4⁺ T cells, whereas CD86 expression depended primarily on infection. These differences highlight the distinct regulatory mechanisms governing the surface expression of costimulatory molecules. The E3 ubiquitin ligase MARCH-I, among other regulators, has been shown to control HLA-DR and CD86 by maintaining their surface levels on neutrophils at negligible amounts [56]. Importantly, MARCH-I expression decreases following TLR stimulation, underscoring the specific impact of bacterial infections on the regulation of CD86 and HLA-DR [57].

Our analyses revealed that GAS SLS specifically contributes to the regulation of HLA-DR during infection. Notably, apoptotic neutrophils have been documented to exhibit elevated levels of APC-specific markers [58]. Although our analyses focused on viable neutrophils, the presence of pre-apoptotic cells that had not yet undergone complete membrane disruption cannot be fully excluded. SLS, however, is known not only for its cytotoxic potential but also for exerting cell type-dependent activating effects independent of lysis [26, 59, 60]. Thus, the observed neutrophil phenotype may reflect receptor-mediated interactions, suggesting a yet unidentified mechanism of SLS-driven regulation, while reduced apoptosis in the absence of SLS may also contribute.

The secretion of cytokines by APCs is essential for directing CD4⁺ T cell differentiation into distinct effector subsets [43]. Infection with GAS triggered the release of pro-inflammatory cytokines, including IL-6, IL-8, and TNF-α. Most other cytokines, however, were detected when CD4⁺ T cells were present. These findings suggest that a subset of neutrophils undergoes a phenotypic shift toward an APC-like state. Many of the cytokines detected are classically associated with APC function and play key roles in shaping CD4⁺ T cell polarization. For example, IL-12 and IFN-γ promote Th1 differentiation [61, 62], whereas IL-1β, IL-6, and IL-23 support Th17 polarization [63]. The 24-h time point is likely too early to observe substantial effector cytokine production, even from memory CD4⁺ T cells [64, 65]. Thus, these cytokines are likely predominantly produced by neutrophils, except for IFN-γ, which can be rapidly secreted by activated memory CD4⁺ T cells and, therefore, cannot be conclusively assigned to a specific cellular source [65].

Both SLO and SLS have been reported to modulate cytokine production in other human cell types. SLO, for instance, can reduce IL-1β and IL-18 production in moDCs, whereas SLS induces pro-inflammatory cytokine expression in THP-1 cells [66, 67]. Under the experimental conditions used here, however, neither cytolysin had a measurable effect on neutrophil mediated cytokine secretion. Prior reports of reduced IL-8 production in hypervirulent covR/S mutants suggest that deletion of a single cytolysin may only have a limited impact on neutrophil cytokine output, whereas coordinated upregulation of both SLO and SLS in covR/S-deficient strains could alter neutrophil responses [66, 68]. Overall, these data indicate that while SLO or SLS alone do not significantly influence cytokine patterns, interactions with CD4⁺ T cells are the dominant factor driving the acquisition of an APC-like, immunomodulatory cytokine profile in neutrophils.

Although neutrophils acquired APC-like features, the resulting CD4⁺ T cell response remained incomplete. Only CD25 was upregulated, while other activation markers and proliferation were not induced. Given the known kinetics of early (CD69, CD137) and late (CD134, CD25) markers [69], the incubation period was sufficient to capture activation, making a kinetic delay unlikely as an explanation. Nevertheless, co-culture with GAS-infected neutrophils induced the secretion of canonical CD4⁺ T cell effector cytokines, including IFN-γ and IL-17A. Although neutrophils can produce IFN-γ, its sustained elevation at later time points is likely attributed to CD4⁺ T cells, consistent with the pronounced loss of neutrophil viability. The absence of proliferation and the lack of antigen-specific activation markers, such as CD134 and CD137, suggest that the observed cytokine production reflects either incomplete antigen-dependent activation or antigen-independent bystander activation [70]. Bystander activation can occur independently of the TCR or via cross-reactive TCR signaling [71]. This activation requires inflammatory cues, such as TLR engagement or exposure to pro-inflammatory cytokines, which can induce an innate-like activation state in CD4⁺ T cells [71]. Activated CD4⁺ T cells can then produce canonical effector cytokines, including IFN-γ and IL-17, characteristic of Th1 and Th17 subsets, respectively. While Th1 cells are critical for defense against intracellular pathogens, Th17 cells play a key role in combating extracellular pathogens and maintaining mucosal immunity [61, 72]. In mouse models, IFN-γ was found to provide protection against GAS-mediated skin infection [73]. Th17 cells, through IL-17 secretion, were shown to be important for preventing systemic dissemination of GAS following upper respiratory tract infections, although they appear less protective in skin infections [74]. Alongside Th1- and Th17-associated cytokines, IL-10 became detectable at later time points. While often linked to regulatory T cells, Th1 cells can also produce IL-10 as an intrinsic feedback mechanism to limit excessive inflammation [75]. In general, the frequency of GAS-specific CD4⁺ T cells likely varies between donors, which may influence donor-specific responses. Nevertheless, given the high prevalence of non-invasive GAS infections [76], some donors probably harbored GAS-specific memory CD4⁺ T cells from prior exposure.

Neutrophils are also known to suppress T cell responses, mediated by factors such as arginase-1 or PD-L1/PD-1 interactions [77–79]. PD-L1, a member of the B7 receptor family, engages PD-1 on CD4⁺ T cells, and under conditions of weak CD28 co-stimulation, this interaction can inhibit T cell proliferation and cytokine production [78]. In a mouse sepsis model, high mobility group box-1 protein (HMGB1), which also accumulates in tissues of GAS NSTI patients, upregulated PD-L1 on neutrophils, leading to T cell apoptosis and immunosuppression [35, 80]. In addition to antigen-dependent activation, bystander activation can also be inhibited by the PD-L1/PD-1 axis [81]. Further studies are required to clarify the role of PD-L1/PD-1 in modulating antigen-dependent CD4⁺ T cell activation by GAS-infected neutrophils.

In general, several factors should be considered when interpreting these findings. The experiments were performed in vitro, which does not fully recapitulate the complexity of GAS NSTIs, where neutrophils, T cells, and other immune and tissue-resident cells interact within a highly pro-inflammatory environment. The limited lifespan of neutrophils in co-culture restricted the observation period, which may have prevented detection of long-term or more nuanced T cell responses. Reduced neutrophil viability likely reflected the culture conditions rather than a direct effect of CD4⁺ T cells, consistent with previous reports. While the study did not comprehensively assess T cell phenotypes or suppressive pathways, the observed cytokine responses indicate meaningful activation of CD4⁺ T cells. Importantly, these limitations do not affect the central conclusion that GAS infection directs neutrophils toward an APC-like phenotype with a capacity to modulate adaptive immune responses.

In conclusion, this study revealed that neutrophils promote effector functions of CD4⁺ T cells during GAS infection. The findings also show that neutrophils act not only as effector cells but also as modulators of adaptive immunity. These insights expand our understanding of host-pathogen interactions in invasive GAS infections.

Acknowledgments

We thank all volunteers for blood donation. J. Schoon and S. Kreutzer are acknowledged for expert organization of donation. We would also like to thank Constantin Klein for expert technical assistance.

Statement of Ethics

Blood samples from healthy volunteers were used in this study. Donors were individuals well acquainted with the research conducted and written informed consent was obtained. The Ethical Research Committee at the University Medicine Greifswald approved the study (BB 006/18a). All experiments were carried out in accordance with the approved guidelines.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This research is supported by the German Research Foundation (DFG; grant No. 492903360 to N.S.). The funder had no role in the design, data collection, data analysis, and reporting of this study.

Author Contributions

Conceptualization, data analysis and writing: A.R. and N.S. Experimental work: A.R. Project administration and supervision: N.S.

Funding Statement

This research is supported by the German Research Foundation (DFG; grant No. 492903360 to N.S.). The funder had no role in the design, data collection, data analysis, and reporting of this study.

Data Availability Statement

All data associated with this study are presented in the manuscript and online supplementary material. Further inquiries can be directed to the corresponding author.