Cargo-less nanoparticles program innate immune cell responses to Toll-like receptor activation

Abstract

Developing biomaterials to control the responsiveness of innate immune cells represents a clinically relevant approach to treat diseases with an underlying inflammatory basis, such as sepsis. Sepsis can involve activation of Toll-like receptor (TLR) signaling, which activates numerous inflammatory pathways. The breadth of this inflammation has limited the efficacy of pharmacological interventions that target a single molecular pathway. Here, we developed cargo-less particles as a single-agent, multi-target platform to elicit broad anti-inflammatory action against innate immune cells challenged by multiple TLR agonists. The particles, prepared from poly(lactic-co-glycolic acid) (PLGA) and poly(lactic acid) (PLA), displayed potent molecular weight-, polymer composition-, and charge-dependent immunomodulatory properties, including downregulation of TLR-induced costimulatory molecule expression and cytokine secretion. Particles prepared using the anionic surfactant poly(ethylene-alt-maleic acid) (PEMA) significantly blunted the responses of antigen presenting cells to TLR4 (lipopolysaccharide) and TLR9 (CpG-ODN) agonists, demonstrating broad inhibitory activity to both extracellular and intracellular TLR ligands. Interestingly, particles prepared using poly(vinyl alcohol) (PVA), a neutrally-charged surfactant, only marginally inhibited inflammatory cytokine secretions. The biochemical pathways modulated by particles were investigated using TRanscriptional Activity CEll aRrays (TRACER), which implicated IRF1, STAT1, and AP-1 in the mechanism of action for PLA-PEMA particles. Using an LPS-induced endotoxemia mouse model, administration of PLA-PEMA particles prior to or following a lethal challenge resulted in significantly improved mean survival. Cargo-less particles affect multiple biological pathways involved in the development of inflammatory responses by innate immune cells and represent a potentially promising therapeutic strategy to treat severe inflammation.

Graphical Abstract

Introduction

Sepsis affects more than 30 million people each year, resulting in over 6 million deaths globally [1]. Although many efforts have been devoted to finding effective treatments, the sepsis-related mortality rate remains alarmingly high (15% to 50%). This life-threatening condition is caused by an exaggerated host response to infection and subsequent development of immunological deficits [2]. Antibiotics may be effective in resolving the bacterial burden, yet this approach is inadequate to stop or reverse the resulting inflammatory cascade [3, 4]. Many pre-clinical and clinical strategies to manage sepsis with pharmaceutic and biological agents have not been successful [5, 6]. These have included blood factors (activated protein C), inhibitors (e.g. of platelet activating factor, bradykinin, and cyclo-oxygenase), blockade of cytokines (e.g. TNF-α and IL-1), and Toll-like receptor antagonists [7–9]. These therapies have been hypothesized to have failed because they target a single molecular pathway and do not address redundancies in pathways associated with immune activation [8]. Therefore, a clinical intervention that broadly influences the multiple inflammatory pathways activated during the development of sepsis is critically needed.

The endotoxemia that accompanies Gram-negative bacterial infection is precipitated by innate immune cells in response to lipopolysaccharide (LPS), a toll-like receptor (TLR) 4 agonist [10]. During bacterial growth, LPS is continually shed into the bloodstream and the bactericidal activities of antibiotics administered during infection have been associated with increased LPS release and adverse outcomes such as mortality [11]. The innate immune system mediates inflammatory signaling through pathogen recognition receptors (PRRs). TLRs are a family of PRRs that recognize specific components of microorganisms called pathogen-associated molecular patterns (PAMPs). Engagement of LPS with TLR4 through the TLR4/MD-2 complex leads to the initiation of downstream signaling cascades that result in the activation of transcription factors including nuclear factor κB (NF-κB), activator protein 1 (AP-1), mitogen-activated protein kinase (MAPK), and the subsequent production and secretion of cytokines, chemokines, and nitric oxide [12–15]. The inflammatory response may aid in the clearance of pathogens; however, excessive or prolonged exposure to LPS stimulation can result in a cytokine storm leading to a collapse of cardiovascular function, multiple organ failure, or death [16, 17].

Modulation of innate immune cell responses using nanoparticles is the basis for new therapies to modulate inflammatory cells and suppress their range of inflammatory functions including trafficking, cell-cell interaction, and paracrine signaling [18–20]. Previous studies demonstrated that intravenous infusion of negatively-charged nanoparticles prepared from polystyrene or poly(lactic-co-glycolic acid) (PLGA) were internalized by inflammatory monocytes and neutrophils, altering their trafficking away from inflammatory foci to the spleen [21–24]. Others have investigated the impact of nanoparticles on treating endotoxin-mediated inflammation, models of polymicrobial sepsis, or systemic bacteremia [25–32]. For example, Spence et al. engineered 150 nm Siglec-targeted nanoparticles to abrogate LPS-induced inflammation in peritoneal macrophages, human monocyte-derived macrophages, and human monocytes [30]. Further, the targeting of the particles towards Siglec-E was shown to be necessary as non-targeted particles were unable to improve survival in mouse models of LPS-induced endotoxemia or polymicrobial sepsis. Other in vitro studies have corroborated the immunomodulatory potential of polymeric nanoparticles to alter the maturation level and inflammatory cytokine secretion of the cells under LPS stimulation [25, 33]. Alternative approaches such as coating PLGA particles with macrophage membranes to the scavenge LPS and inflammatory cytokines have proven efficacious in a mouse model of bacteremia [31]. However, the complexity associated with manufacturing, scale up, and characterizing these particles may hinder the clinical translation of this technology. Therefore, a strategy that is simple, clinically translatable, and mitigates the inflammatory damage by innate immune cells has the potential to dramatically affect the management of sepsis.

This report describes the tunable immunomodulatory properties of PLGA and poly(lactic acid) (PLA) particles and their ability to program anti-inflammatory cell responses to inhibit TLR-mediated innate immune cell activation. Several particle formulations were evaluated using in vitro and ex vivo assays to establish fundamental and functional relationships between particle properties and regulation of inflammatory responses induced by LPS (TLR4) and unmethylated CpG oligodeoxynucleotides (CpG-ODN) (TLR9). The dynamic regulation of gene expression in macrophages resulting from particle treatment was evaluated using TRanscriptional Activity CEll aRray (TRACER) technology where the activity of over 60 transcription factors was investigated. Subsequently, the efficacy of particles in mice was evaluated in both prophylactic and therapeutic treatment models of LPS-induced endotoxemia, a well-established model of sepsis that recreates the activation of immune cells through TLR signaling [34]. The investigation of particle treatment in this model provides insight into modulation of the endotoxin-mediated contributions of septic inflammation. Cargo-less particles represent a tunable biomaterial-based platform and potentially promising single-agent, multi-target treatment to inhibit the broad and deleterious inflammatory responses that accompany severe inflammation induced by TLR stimulation.

Materials and Methods

Materials

Acid-terminated poly(D,L-lactide-co-glycolide) (PLGA), of low inherent viscosity (low molecular weight; PLGA_Lo) in hexafluoro-2-propanol ~ 0.17 dL/g (approx. 4,200 g/mol) and high inherent viscosity (high molecular weight; PLGA_Hi) in hexafluoro-2-propanol ~ 0.66 dL/g (approx. 43,500 g/mol) monomer ratios 50:50 and acid terminated poly(D, L-lactide) (PLA) of low inherent viscosity (low molecular weight) in hexafluoro-2-propanol ~ 0.21 dL/g (approx. 11,700 g/mol) were purchased from Lactel Absorbable Polymers (Birmingham, AL). Poly(ethylene-alt-maleic anhydride) (PEMA) was purchased from Polysciences, Inc. (Warrington, PA). Poly(vinyl alcohol) (PVA, MW 30,000-70,000), β-mercaptoethanol, and LPS from Escherichia Coli serotype O111:B4 were obtained from Sigma-Aldrich (St. Louis, MO). CpG-ODN 1668 was obtained from Invivogen (San Diego, CA).

Particle Preparation and Characterization

PLGA_Lo, PLGA_Hi and PLA particles were prepared by the oil-in-water (o/w) emulsion solvent evaporation (SE) technique as previously described in publications [35, 36]. Briefly, 400 mg of the acid-terminated polymer was dissolved in 2 mL of dichloromethane (DCM) and to this 10 mL of 1% PEMA (or 8 mL of 2% PVA) was added and sonicated at 100% amplitude for 30 sec using a Cole-Parmer Ultrasonic processor (Model XPS130). The resulting o/w emulsion was then added to 200 mL of magnetically stirred 0.5% PEMA (or 0.5% PVA) overnight until all the DCM evaporated. The particles were then collected by centrifugation at 11,000 x g for 20 min at 4°C and washed with 40 mL of 0.1M sodium bicarbonat e buffer. The centrifugation and washing steps were repeated two more times with a final wash using MilliQ water. A mixture of sucrose and mannitol were added to the particle suspension as cryoprotectants to achieve a final concentration of 4% and 3% w/v, respectively. The particles were then frozen at −80°C and lyophilized for 48 h before use. The size and zeta potential of the particles were determined by dynamic light scattering (DLS) by mixing 10 μL of a 10 mg/mL particle solution into 990 μL of MilliQ water using a Malvern Zetasizer ZSP. Cy5.5-labeled PLA particles were prepared by incorporating 1% w/w of PLA-Cy5.5 into particles as previously described [37]. The stability of the particles prior to administration has been previously evaluated by performing an in-use stability test. No, or minimal aggregation of particles was observed for over an hour prior to use as determined by DLS [38].

Mice

Female C57BL/6J (6-8 weeks old) were purchased from The Jackson Laboratories (Bar Harbor, ME). The mice were housed under specific pathogen-free conditions in the University of Michigan Unit for Laboratory Animal Medicine and all mice procedures and experiments were compliant to the protocols of the University of Michigan Animal Care and Use Committee.

Antibodies and Flow Cytometry

All antibodies were purchased from BioLegend (San Diego, CA). Cell staining was conducted according to BioLegend protocols. Flow cytometric data were collected using a Beckman Coulter CytoFLEX S Research Flow Cytometer or a Becton Dickinson LSR II flow cytometer. Analysis was performed using FlowJo software (FlowJo, San Jose, CA) or FCS Express 6 (De Novo, Glendale, CA). FcR blocking was performed with anti-CD16/32 antibody prior to staining with various combinations of the following antibodies: anti-CD80 (clone 16-10A1), anti-CD86 (clone GL-1), anti-MARCO (clone PLK-1), anti-MHCII (clone M5/114.15.2), anti-CD45 (clone 30-F11), anti-CD11b (clone M1/70), anti-F4/80 (clone BM8), anti-CD11c (clone N418), anti-CD45R (B220) (clone RA3-6B2), anti-Ly6C (clone HK1.4), anti-Ly6G (clone 1A8). Viability was assessed with 4’,6-Diamidino-2-Phenylindole, Dilactate (DAPI) exclusion dye (Biolegend).

Antigen Presenting Cells

Bone marrow from the tibia and femurs of C57BL/6J mice were harvested to obtain a primary population of antigen presenting cells (APCs). Cell media consisted of RPMI 1640 supplemented with L-glutamine (Life Technologies, Carlsbad, CA), penicillin (100 units/mL), streptomycin (100 μg/mL), 10% heat-inactivated fetal bovine serum (FBS) (Invitrogen Corporation, Carlsbad, CA). For bone marrow-derived macrophages (BMMØ) the media was further supplemented with 20% L929 (ATCC, Manassas, VA) cell-conditioned media on days 0, 3, 6, and 8, and for bone marrow-derived dendritic cells (BMDCs) 50 mM β-mercaptoethanol and 20 ng/mL of granulocyte-macrophage colony-stimulating factor (GM-CSF) (Peprotech, Rocky Hill, NJ) were added. The macrophages were removed using Versene (ThermoFisher, Waltham, MA) treatment and the dendritic cells were loosely adherent. The cell numbers and viability were determined using Trypan blue solution and Countess™ Automated Cell Counter (Life Technologies, Carlsbad, CA).

Detection of Cytokine Production by ELISA

Enzyme-linked immunosorbent assays (ELISA) (R&D Systems, Minneapolis, MN) were used to measure murine interleukin-6 (IL-6), tumor necrosis factor α (TNF-α), and monocyte chemotactic protein-1 (MCP-1) in the cell culture supernatants and were performed by the University of Michigan Cancer Center Immunology Core.

Particle-Cell Association Studies

Day-8 BMMØs were seeded in sterile 24-well plates at a concentration of 8 × 10⁴ cells/well and then treated with 20 μg/mL of Cy5.5-labeled particle formulations (PLGA_Lo-PEMA, PLGA_Lo-PVA, PLGA_Hi-PEMA, PLGA_Hi-PVA, PLA-PVA and PLA-PEMA for 8, 6, 4, 2, 1, and 0.5 h. All treated wells were washed twice with PBS to remove excess particles and replenished with 500 μL of fresh sterile PBS and scraped using a blunt 1000 μL pipette tip. The macrophages were collected by centrifugation and stained for viability using 4’,6-diamidino-2-phenylindole dihydrochloride (DAPI) dye. Flow cytometry was used to measure Cy5.5 signal on viable cells.

Cytokine Secretion by Raw 264.7 Macrophages and BMDCs

Raw Macrophages 264.7 were obtained from American Type Culture Collection (ATCC, Manassas, VA). In order to evaluate the pro-inflammatory cytokine production, Raw macrophages (1 × 10⁵ cells/well) and BMDCs (1 × 10⁵ cells/well) were seeded in sterile 24-well plates and treated with 300 μg/mL of the 6 different particle formulations at 37°C and 5% CO₂ for 3 h. Excess particles were removed by washing twice with PBS followed by replacing with complete medium containing 100 ng/mL of either LPS or CpG-ODN. After 48 h, cell culture supernatants were analyzed by ELISA for IL-6, MCP-1, and TNF-α.

BMMØ Cell Surface Molecule Expression by Flow Cytometry and Cytokine Production by ELISA

Day-8 BMMØs were seeded in sterile 24-well plates at a concentration of 1.25 × 10⁵ cells/well and then treated with 300 μg/mL of the 6 different particle formulations. Three hours following particle incubations, all treated wells were washed to remove excess particles and then treated with 100 ng/mL of either LPS or CpG-ODN. 48 h post LPS/CpG-ODN treatment the cell culture supernatants were analyzed by ELISA for IL-6, MCP-1 and TNF-α. The LPS treated cells were collected and co-stimulatory molecules CD80 and CD86, MHC II, and MARCO scavenger receptor were measured using flow cytometry.

Ex vivo Splenocyte Cytokine Secretion and Cellular Interactions

C57BL/6J mouse splenocytes were processed into single-cell suspensions and erythrocytes lysed. The splenocytes were seeded in sterile 48-well tissue culture plates at a concentration of 2 × 10⁶ cells/well and then treated with 100 μg/mL of the 6 different particle formulations. Three hours following particle incubations, all treated wells were incubated with 100 ng/mL of either LPS or CpG-ODN. 48 h post LPS/CpG-ODN treatment the cell culture supernatants were analyzed by ELISA for IL-6, MCP-1 and TNF-α.

Splenocytes were similarly analyzed from mice injected with PLA-PEMA and PLA-PVA particles. Female C57BL/6J mice were intravenously injected with vehicle (PBS) or PLA-PEMA (2 mg) or PLA-PVA (2 mg) particles for one day or four consecutive days. Splenocytes were harvested from the spleen the following day and the erythrocytes were lysed. The cells were seeded in sterile 48 well plates at a concentration of 2 × 10⁶ cells/well and incubated with 100 ng/mL of either LPS or CpG-ODN. At 48 h post LPS/CpG-ODN treatment, the cell culture supernatants were analyzed by ELISA for IL-6, MCP-1 and TNF-α.

For ex vivo cell interaction experiments, splenocytes were seeded in sterile 24-well tissue culture plates at a concentration of 2 × 10⁶ cells/well and then treated with 10 μg/mL of either PLA-PEMA-Cy5.5 or PLA-PVA-Cy5.5 particle formulations. Three hours following particle-Cy5.5 incubations, cells were collected, co-stained for innate immune cells, and measured using flow cytometry. Leukocytes were identified as CD45⁺ (% of live cells). Myeloid cells were identified as CD45⁺/CD11b⁺ (% of live cells); Macrophages were identified as CD45⁺/F4/80⁺ (% of live cells); Dendritic cells were identified as CD45⁺/CD11c⁺ (% of live cells); B cells were identified as CD45⁺/CD45R(B220)⁺ (% of live cells); Monocytes were identified as CD45⁺/CD11b⁺/Ly6C⁺ (% of live cells); Neutrophils were identified as CD45⁺/CD11b⁺/Ly6G⁺ (% of live cells).

PLA Particle TRACER

The effects of PLA particles on LPS (100 ng/mL) activated transcription factors (TF) was studied using the dynamic TRanscriptional Activity CEll aRray (TRACER) [39, 40]. The TF reporter constructs, delivered in self-inactivating lentivirus, consist of specific TF response elements cloned upstream of a minimal thymidine kinase promoter that causes the expression of firefly luciferase (Fluc). The methods for designing reporter construction and reporter binding specificity have been described in detail previously [40, 41]. Day-6 BMMØs were batch transduced with a single lentiviral reporter construct at a multiplicity of infection (MOI) of 80 and seeded in black 96-well plates at a seeding density of 4 × 10⁴ cells/well and cultured for a minimum of 48 h. After 48 h, fresh media was exchanged and supplemented with 1 mM D-Luciferin (Promega Corporation, Fitchburgh, WI) to enable the measurement of Fluc activity by bioluminescence imaging (Perkin Elmer IVIS Spectrum, Waltham, MA). PLA-PEMA and PLA-PVA particles were added to the BMMØs at a concentration of 300 μg/mL for 3 h and the unbound particles were removed by washing with sterile DPBS twice. Fresh macrophage media (containing 100 ng/mL LPS and 1 mM D-Luciferin) was added post-treatment and the bioluminescence intensity (BLI) was measured at several time points (0, 1, 2, 3, 4, 6, 8, 24, and 48 h).

For all treatment groups (cell control, LPS-treated, and PLA particle-treated) each TF was examined in triplicate and the data was pooled from all individual experiments and normalized using methodology previously described [40, 42]. Briefly, for each timepoint the background signal measured in a non-transduced cell control was subtracted from each BLI measurement and then normalized to the corresponding treatment’s minimal TA control reporter to give the log2 fold change in BLI. Heatmaps were generated by averaging the replicate log2 fold change for each reporter condition and time point. Statistical and partial least squares discriminate analysis have been detailed previously [43].

LPS-Induced Endotoxemia Model

In the prophylactic treatment model of LPS-induced endotoxemia, female C57BL/6J mice (6-8 weeks) (n=7-8) were intraperitoneally injected with 2 mg of PLA-PEMA particles one hour prior to intraperitoneal LPS injection (20 mg/kg). In the therapeutic model, mice (n=13-14) were intraperitoneally injected with LPS (20 mg/kg) and, 30 min later, were intraperitoneally injected with 2 mg of PLA-PEMA particles. Mice were monitored for a period of 7 days compliant to the protocols set forth by the University of Michigan Animal Care and Use Committee. Mice were euthanized immediately at a humane endpoint noted by acute loss of function and non-sensitivity to touch.

Statistical Analyses

Results are reported as mean ± standard deviation (SD). Student’s t-test was used to determine the significance of parametric data between two groups. Significant differences between cytokine expressions were determined by one-way ANOVA along with Tukey’s multiple comparison test. Mouse survival was reported using a Kaplan-Meier survival curve and statistical significance of mouse survival was determined with a log-rank (Mantel-Cox) X² test. In all cases unless otherwise noted, p < 0.05 was considered to be statistically significant.

Results

Preparation and characterization of immunomodulatory particles

Six particle formulations were fabricated to investigate the relationship between physicochemical properties and modulation of inflammatory responses. These particles were prepared using a single emulsion-solvent evaporation method with three polymers (high molecular weight PLGA_Hi, low molecular weight PLGA_Lo, and a single composition of PLA) and two different types of surfactants (poly(ethylene-alt-maleic acid) (PEMA) and poly(vinyl alcohol) (PVA)) (Figure 1A). All particles displayed similar sizes between 350 and 500 nm with low polydispersity. The zeta potential of each formulation varied as a consequence of surfactant selection, where particles prepared using PEMA were approximately −40 to −50 mV and particles prepared using PVA were approximately −20 mV (Figure 1B).

Figure 1:

Synthesis and characterization of immunomodulatory particles. (A) Particles were formulated from polymers with various molecular weights (Low or High), compositions (glycolic acid (GA) and lactic acid (LA)), or surfactants (PVA or PEMA). (B) Size and zeta potential measurements for the particles evaluated in this study. (C) The kinetics of particle interactions with BMMØs are dependent on emulsion surfactant. Cy5.5-labeled particles were administered to BMMØs at 300 μg/mL. Particle association was determined by flow cytometry. Statistical differences between zeta potentials of PEMA and PVA formulations were determined by onetailed student’s t test (****, p <0.0001). Error bars represent standard deviation (n=3) and are representative of 10 (B) and 2 (C) independent experiments.

Differential cell association kinetics were observed between bone marrow-derived macrophages (BMMØs) and fluorescently (Cy5.5) labeled PLGA_Hi-PEMA, PLGA_Hi-PVA, PLGA_Lo-PEMA, PLGA_Lo-PVA, PLA-PEMA, and PLA-PVA particle types. The particles formulated with PEMA surfactant rapidly associated with BMMØs, achieving 100% cellular interaction within 1-2 h (Figure 1C). In contrast, the particles formulated with PVA surfactant gradually associated with BMMØs and their cellular interactions plateaued between 4-6 h. By 6 h, each particle type achieved maximal association with BMMØs. The viability of BMMØs was not significantly altered by particle incubation (Figure S1). Collectively, particles prepared using PEMA possessed a more negative surface charge and associated with cells more rapidly than particles prepared with PVA.

Particle properties differentially inhibit inflammatory cytokine secretion by innate immune cells stimulated with extracellular (LPS) and intracellular (CpG-ODN) TLR ligands

BMMØs, Raw 264.7, and bone marrow-derived dendritic cells (BMDCs) were subsequently investigated, following treatment with the multiple particle formulations, for their secretion of inflammatory cytokines induced by a TLR4 ligand, LPS (Figure 2A). The cells were pretreated with particles prior to the addition of LPS and the initiation of cytokine secretion to better distinguish between the ability of the various particles to influence cellular phenotype and cytokine effector potential. Proinflammatory cytokines, such as IL-6, MCP-1, and TNF-α [17], are important in the development of the cytokine storm and were differentially inhibited by particles in all cells tested. BMMØs and Raw 264.7 cells followed similar trends with three general responses (Figure 2B and S2). First, treatment with PLGA_Lo-PVA particles did not affect the high levels of cytokine secretion observed in the LPS-only positive control. Second, intermediate levels of cytokine suppression were observed for cells treated with PLGA_Hi-PVA, PLA-PVA, or PLGA_Lo-PEMA particles. Third, nearly complete inhibition of proinflammatory cytokine secretion was observed in cells treated with PLGA_Hi-PEMA or PLA-PEMA. A similar trend was observed in BMDCs, except PLGA_Lo-PEMA particles were more effective than in BMMØs and Raw 264.7 cultures (Figure S3). Independent of the cell type tested, particles formulated with PEMA, PLGA_Hi, or PLA tended to more effectively inhibit the secretion of inflammatory cytokines induced by LPS compared to those formulated with PVA or PLGA_Lo.

Figure 2:

Particles modulate innate inflammatory responses, and the bioactivity is dependent on the physicochemical properties of the particles. (A) Schematic of cytokine inhibition assay where BMMØs were treated with particles, washed, and stimulated with TLR agonists. (B) Suppression of inflammatory cytokine production in particle-treated BMMØs (300 μg/mL) stimulated with 100 ng/mL LPS. (C) The particle-induced suppression of BMMØ cytokine production are particle- and LPS-concentration dependent. (D) BMMØs were evaluated by flow cytometry for surface marker expression of live cells. Downregulation of cell surface markers induced by particles is polymer- and surfactant-dependent. Cytokines were measured by ELISA. Statistical differences were determined by one-way ANOVA with Tukey’s multiple comparisons test. In each data set, all groups were compared to each other and matching letters indicate no significant difference (p>0.05). For all non-matching letters, the groups were statistically different (p<0.05). Error bars represent standard deviation (n=3) and are representative of 5 (B,D) and 2 (C) independent experiments.

The particle-induced inhibition of cytokine secretions was then tested in the context of TLR9 ligand stimulation (CpG-ODN). Distinct from TLR4, which is expressed on the cell membrane, TLR9 is expressed intracellularly in the endosomal compartment and detects foreign DNA with unmethylated CpG dinucleotides [44]. Strikingly, the treatment of CpG-ODN-stimulated BMMØs with any PEMA-formulated particle resulted in a complete inhibition of inflammatory cytokine secretion. Particles prepared with PVA displayed, at most, a 30% reduction in cytokine production relative to controls (Figure S4). In Raw 264.7 cells, IL-6 was significantly reduced in all particle-treated conditions, whereas MCP-1 and TNF-α were most affected in conditions treated with PLA-PEMA (Figure S5).

The two particle types that provided the greatest level of cytokine inhibition under TLR stimulation, PLGA_Hi-PEMA and PLA-PEMA, were then tested over a range of particle concentrations and a greater concentration of LPS to discriminate between their similar activity (Figure 2C). Interestingly, the higher concentration of LPS (200 ng/mL versus 100 ng/mL) abolished the ability of PLGA_Hi-PEMA particles to inhibit inflammatory cytokine production in BMMØs, yet PLA-PEMA particles remained highly effective at concentrations similar to those used in the initial bioactivity screening studies. Importantly, the results of these studies demonstrated the broad inhibitory activities enabled by particles to modify the responsiveness of innate immune cells to multiple TLR agonists acting through distinct biochemical mechanisms.

PLA-PEMA particles downregulate macrophage MHC, co-stimulatory molecules, and MARCO under LPS stimulation

Particle-induced BMMØ cell surface molecule expression was measured under LPS stimulation in vitro. Major histocompatibility complex (MHC) II, costimulatory molecules (CD80 and CD86), and the scavenger receptor MARCO were used to assess levels of cell maturation and the anti-inflammatory effects of various particle types (Figure 2D). The PLGA_Lo-PEMA particle-treated cells did not induce appreciable changes in cell surface molecules, yet the high molecular weight PLGA_Hi-PEMA particles induced slight downregulation of all surface markers assessed. Cells treated with PLA-PEMA particles dramatically reduced the expression of all surface molecules evaluated. Particles prepared using PVA as the surfactant did not induce any significant changes compared to the LPS-treated control. Notably, the downregulation of these cell surface molecules was correlated with the reduced cytokine secretion by BMMØs (Figure 2B). Among the treatments containing PEMA-formulated particles, the distinct differences in surface marker expression affected by PLA-PEMA particles compared to PLGA_Hi-PEMA or PLGA_Lo-PEMA particles implies a role for polymer composition in the mechanism of particle-induced immune cell modulation [24].

Particles formulated using PVA have limited anti-inflammatory properties

The profound ability of PEMA particles to inhibit inflammatory cytokine production in vitro compared to PVA particles motivated the investigation of surfactant effects on cytokine production by immune cells. The relative impact of PEMA and PVA surfactants on the antiinflammatory properties of particles was tested by preparing particles with constant PEMA content and increasing the percentage of PVA incorporated during particle formation (Figure 3A). The size range of the particles was approximately 400-700 nm and the zeta potentials for particles containing PEMA were highly negative and less than −30 mV for particles containing PEMA (Figure 3B). The immunomodulatory properties of the various particles were measured by BMMØ cytokine secretion as depicted in Figure 2A. Strikingly, even the subtle incorporation of 0.1% PVA to 1% PEMA negatively affected the ability of particles to suppress IL-6 production (Figure 3C) and further increasing the PVA:PEMA ratio resulted in markedly less inhibition of inflammatory cytokines. The results indicated that minimal incorporation of PVA into particle formulations could dramatically alter immune responses.

Figure 3:

Incorporation of PVA into particle formulations abrogates their ability to modulate inflammatory responses. (A) Schematic of particle formulation, where an increasing amount of PVA was incorporated into the emulsion conditions while maintaining a consistent concentration of PEMA. (B) Size and zeta potential measurements for the particles evaluated in this study. (C) Suppression of IL-6 production in particle-treated BMMØs stimulated with 100 ng/mL LPS. (D) BMMØ secretion of IL-6 was measured after incubation with soluble PEMA or PVA (0.1 μg/mL, 1 μg/mL, and 10 μg/mL) and stimulation with 100 μg/mL LPS. Statistical differences were determined by one-way ANOVA with Tukey’s multiple comparisons test. In each panel, all groups were compared to each other and matching letters indicate no significant difference (p>0.05). For all non-matching letters, the groups were statistically different (p<0.05). Error bars represent standard deviation (n=3) and are representative of 2 independent experiments.

BMMØs were then incubated with soluble PEMA and PVA to assess the relative impact of the surfactant on LPS-induced cytokine production. Three surfactant concentrations were evaluated (0.1 μg/mL, 1 μg/mL, and 10 μ/mL), which correspond to the percent of residual surfactants typically associated with particles (1-5% w/w) [45]. BMMØs challenged with LPS exhibited no significant reduction in inflammatory cytokine production for either surfactant tested (Figure 3D). However, an increase in IL-6 secretion was observed for cells treated with PEMA. The uninhibited cytokine secretion profile of cells treated with soluble surfactants indicated that the unique combination of the surfactants with the particles was responsible for the reduced sensitivity of cells to TLR stimulation.

Particles inhibit inflammatory cytokine production by splenocytes when stimulated with LPS or CpG-ODN

The in vitro evaluation of particle immunomodulatory properties was extended to a culture of splenocytes to test responses within a diverse cell population that is more representative of an in vivo setting. The cell composition of splenocytes was evaluated using flow cytometry and the major innate immune cell types detected were B cells (45%) and myeloid cells (6%) (Figure S6A). Other cell types, including macrophages, dendritic cells, neutrophils, and monocytes, were present at less than 3% of cells detected. Splenocytes were isolated and incubated ex vivo with particles and stimulated with LPS (Figure 4A) or CpG-ODN (Figure 4B). In the LPS-stimulated groups, PLA-PEMA particles had the greatest level of IL-6 inhibition, whereas other groups did not affect its production significantly. The secretion of MCP-1 and TNF-α were more responsive to particle treatment, and cytokine production was most significantly reduced in splenocytes treated with PLA-PEMA particles. In splenocytes treated with CpG-ODN, the IL-6 cytokine responses were suppressed by approximately 50-60% by all particle groups tested. For the production of MCP-1, PLA-PEMA particles were the most effective at suppressing cytokine production. Interestingly, TNF-α was equally suppressed by all particle formulations prepared using PEMA, whereas minimal or insignificant suppression was achieved using particles formulated with PVA. These results suggest that the administration of particles in vivo could potentially attenuate TLR-mediated inflammatory responses in a heterogeneous cell population.

Figure 4:

Particles modulate inflammatory cytokine secretion by particles in splenocytes stimulated with LPS or CpG-ODN. (A) Splenocytes were isolated from C57BL/6 mice and plated at concentration of 2 × 10⁶ cells/well (48 well format). Particles were added to the splenocytes at 100 μg/mL for 3 h prior to the addition of LPS (100 ng/mL) or (B) CpG-ODN (100 ng/mL). The cytokines in the supernatants were measured by ELISA after 2 days of culture. Statistical differences were determined by one-way ANOVA with Tukey’s multiple comparisons test. In each panel, all groups were compared to each other and matching letters indicate no significant difference (p>0.05). For all non-matching letters, the groups were statistically different (p<0.05).Error bars represent standard deviation (n=3) and represent 2 independent experiments.

Intravenous injection of PLA-PEMA particles attenuates inflammatory cytokine production in splenocytes challenged with LPS or CpG-ODN

The immunomodulatory properties of particles in vivo were next assessed by intravenous particle administration followed by ex vivo splenocyte culture and TLR stimulation. PLA-PVA or PLA-PEMA particles were injected into naive C57BL/6 mice for either one or four consecutive days (Figure 5A), which we have previously reported leads to particle accumulation within the spleen [24, 37]. Following the final particle injection, isolated splenocytes were incubated with either LPS (Figure 5B) or CpG-ODN (Figure 5C) for assessment of inflammatory cytokine secretion. PLA-PVA particles did not alter IL-6, MCP-1, and TNF-α responses after 1 injection, but increased the cytokine secretion, relative to controls, after 4 consecutive days of injections. Importantly, PLA-PEMA particles significantly reduced the production of each cytokine when administered as either a single dose or multiple doses. In general, splenocytes from mice receiving 4 doses of PLA-PEMA particles demonstrated a further reduction in cytokine production when stimulated with LPS or CpG-ODN. Similar trends in particle-induced modulation of inflammatory cytokine secretion was observed for both LPS and CpG-ODN-treated splenocytes.

Figure 5:

Intravenous administration of PLA particles to mice, as a single injection or as multiple daily injections, alters the ex vivo cytokine responses of splenocytes to LPS and CpG-ODN stimulation. (A) Schematic of dosing and experimental design to evaluate inflammatory cytokine secretions by splenocytes. (B) Cytokine secretions of splenocytes obtained from PLA-PVA or PLA-PEMA particle-treated mice that were stimulated ex vivo with LPS (100 ng/mL) or (C) CpG-ODN (100 ng/mL). Cytokines were measured by ELISA after 2 days of culture. Statistical differences were determined by one-way ANOVA with Tukey’s multiple comparisons test. In each panel, all groups were compared to each other and matching letters indicate no significant difference (p>0.05). For all non-matching letters, the groups were statistically different (p<0.05). Error bars represent standard deviation (n=3) and represent 2 independent experiments.

Splenocytes were treated with PLA particles in vitro to identify differences in cellular association that may account for the trends in cytokine secretions in Figures 4 and 5. The composition of cells did not vary as a result of either particle treatment compared to control (Figure S6A). PLA-PEMA particles were associated with approximately 6% of the CD45⁺ splenocytes, a 4-fold greater interaction than the PLA-PVA particles exhibited (Figure S6B). This increase in overall association by PLA-PEMA particles was accompanied by a greater distribution across different cell types (Figure S6C). The PLA-PEMA particles were measured in greater than 10% of myeloid cells, macrophages, DCs, B cells, neutrophils, and monocytes, whereas the PLA-PVA particles were measured in only myeloid, macrophages, and DCs. In summary, these studies showed that PLA-PEMA particles associated with a greater number and breadth of splenocytes, suggesting that the immunomodulatory properties may be related to or directly attributed to the extent of their interaction with innate immune cells.

Particle design differentially influences transcription factor activity

The differential responses to particles were subsequently investigated at the level of signaling pathway and transcription factor activation using the bioluminescence-based transcription factor (TF) activity reporter assay, TRACER (Figure 6A) [46]. TRACER enables real-time assessment of the signaling pathways in live cells and enables tracking of the differential TF activity induced by various particle treatments. The dynamic expression of over 60 transcription factors was measured at 10 timepoints in BMMØs treated with either no particles or particles (PLA-PVA or PLA-PEMA) for 3 h prior to the addition of LPS at time 0 h. Treatment with both PLA particle types resulted in changes to TF activity compared to the LPS-only condition. Significant downregulation of the activity of canonical early-acting LPS-induced factors NF-κB, IRF1, and STAT-1 was observed for the PLA-PEMA particles but not for the PLA-PVA particles (Figure 6A). Overall, treatment with PLA-PEMA particles broadly downregulated the LPS-induced TF activity of BMMØs (Figure 6B). Treatment with PLA-PEMA particles resulted in 5 TFs upregulated, 41 downregulated, and 7 that were upregulated and downregulated at distinct time points during the experiment. Cells treated with PLA-PVA particles had a less pronounced effect on the response of BMMØs to LPS stimulation, with 6 TFs upregulated, 19 downregulated, and 3 that were both upregulated and downregulated over the course of the experiment.

Figure 6:

Particle treatment results in deviation of LPS-induced transcription factor activity, which was measured using TRanscriptional Activity CEll aRray (TRACER). BMMØs were transduced with bioluminescent transcription factor reporters. The cells were treated with particles for 3 h, or untreated, prior to stimulation with LPS (100 ng/mL). (A) Normalized TF activity of NF-κB, IRF1, and STAT1. (B) Heatmap of dynamic TF profiles for BMMØs treated with PLA-PVA or PLA-PEMA particles and stimulated with LPS. The heatmap values indicate fold-change relative to LPS-only control. (C) PLS-DA analysis distinguishing the phenotypic transcription factor activity in treated BMMØs (line represents 95% confidence interval). (D) Variable importance in projection (VIP) scores indicating the transcription factors that were significant in discriminating the BMMØ phenotypes. Error bars represent standard deviation (n=3) and data are representative of 10 independent experiments (A). B and C represent data combined from 3 independent experiments.

Partial least squares discriminate analysis (PLS-DA) was used to identify distinct phenotypic TF activity signatures derived from TRACER (Figure 6C). The top 14 differentially activated TFs were used to discriminate between phenotypes. Pretreatment with PLA-PEMA particles resulted in discrete skewing of the BMMØ phenotype away from the LPS-induced inflammatory phenotype towards the untreated control in latent variable 1 (LV1), whereas PLA-PVA led to a partial deviation on LV1. Loadings in latent variable 2 (LV2) further separated the PLA-PEMA condition from the LPS-only and untreated controls. In this model, the four highest scoring factors by variable importance in projection (VIP) score were immune factors IRF1, STAT-1, and AP1, as well as the pluripotency factor POU5F1 (Figure 6D). These TFs were upregulated by LPS, were significantly inhibited by PLA-PEMA particles, and were less affected by PLA-PVA.

Local delivery of PLA-PEMA particles improves survival in a lethal LPS-induced endotoxemia mouse model

The effectiveness of PLA-PEMA particles to improve survival during acute inflammation was evaluated using prophylactic and therapeutic treatments in a mouse model of lethal LPS-induced endotoxemia. Here, both particles and LPS were administered intraperitoneally to measure the immunomodulatory effects on immune cells apart from their influence on cell trafficking which is associated with intravenous delivery [21, 22, 24]. Prophylactic treatment of endotoxemia may have clinical relevance as a pre-emptive intervention to mitigate inflammation resulting from the surge in systemic LPS release following antibiotic administration in the context of Gram-negative bacteremia [11]. In the prophylactic model, a single dose of PLA-PEMA particles was intraperitoneally administered to C57BL/6 mice 1 h prior to LPS challenge. In the PBS-treated control group, all mice succumbed to the endotoxin challenge (n=7) within 2 days, whereas, mice treated with PLA-PEMA particles experienced a 62.5% survival (n=8) (Figure 7A). The therapeutic use of particles was then tested to evaluate the clinical scenario where particle treatment is used as an intervention for patients experiencing early endotoxemia. In this therapeutic model, a lethal LPS challenge was initiated 30 min prior to intraperitoneal administration of PLA-PEMA particles. Similar to the prophylactic model, PLA-PEMA particle treatment significantly improved survival, although to a lesser extent. In the PBS-treated control group, all mice died within 24 h (n=13) and in the PLA-PEMA particle-treated group, 14.3% of mice survived (n=14) (Figure 7B). These results indicate that PLA-PEMA particles could be a protective intervention in endotoxin-induced inflammation.

Figure 7:

PLA-PEMA particles significantly improve survival in lethal models of LPS-induced endotoxemia. (A) C57BL/6 mice were either treated with a single intraperitoneal injection of 2 mg of PLA-PEMA particles 1 h prior to intraperitoneal injection of 20 mg/kg LPS (n = 7-8/group) or (B) 30 min following injection of 20 mg/kg LPS (n = 13-14/group). Survival was tracked for 7 days. Statistical differences were determined by Mantel-Cox log-rank test where p<0.05 was considered significant. Data are representative of 3 independent experiments.

Discussion

The development of effective treatments for severe inflammation, which must overcome redundant and broad mechanisms associated with the inflammatory response, may limit organ damage and reduce mortality rates in life-threatening conditions such as sepsis. Numerous failures of clinical trials utilizing single-agent, single-target treatment strategies have provided evidence that a multifaceted approach is needed to address the breadth of inflammatory stimuli associated with sepsis [8, 47, 48]. The objective of this investigation was to focus on modulating the initial acute hyperinflammatory response generated by innate immune cells as a consequence of endotoxemia, referred to as cytokine storm [17], and to develop a single-agent, multi-target therapeutic strategy that could broadly and simultaneously reduce the production of inflammatory cytokines induced by TLR stimulation. In particular, two TLR agonists, LPS and CpG-ODN, were used to evaluate two disparate mechanisms of immune activation.

Particles represent a promising therapeutic platform for their ability to associate with and influence innate immune cells [18]. This investigation tested the hypothesis that delivery of cargo-less particles to innate immune cells would directly inhibit their responses to inflammatory stimuli by altering their ability to respond to multiple TLR agonists. This approach is distinct from previous reports where cargo-less particles have been utilized to indirectly affect inflammation by redirecting and sequestering inflammatory cells in the spleen [21–24]. Here, a set of 500 nm particles were generated from biocompatible and biodegradable polyesters, namely PLGA and PLA. Variants of particles were prepared by changing the polymer molecular weight or the surfactant (PVA or PEMA) used during the particle preparation (Figure 1). Notably, these particle modifications (composition, molecular weight, and surface charge/chemistry) each contributed to the alterations observed in the inflammatory responses of innate immune cells under TLR stimulation (Figure 2). Overall, the PLA-PEMA particles were most effective and induced an almost complete inhibition of inflammatory cytokines. Particles prepared from PLGA_Lo-PVA were immunologically inert or contributed towards an increased immune response to TLR stimulation. These simple particle designs may overcome disadvantages of pharmaceutical agents, which act systemically on singular molecular pathways. These particles are drug-free carriers that affect a broad range of immune cell inflammatory cytokines, costimulatory molecules, and trafficking to sites of inflammation [21]. Since these particle formulations do not require further processing, surface functionalization (e.g. (peptide, protein, small molecule), or complicated membrane coating procedures) to achieve their immunomodulatory properties, their translation to the clinic may be more accessible [26, 28, 31, 32, 49].

PLGA and PLA particles demonstrated broad anti-inflammatory bioactivity by tempering the inflammatory response to TLR activation. Three points in the LPS signaling cascade provide opportunities where the particles may impart their influence: sequestering LPS, antagonizing the extracellular signaling of LPS, and inducing a cellular phenotype that is resistant to LPS signaling. In the first case, investigators have reported strategies for binding and sequestering LPS using particles coated in macrophage membranes and soluble high-density lipoprotein (HDL) to prevent downstream signaling of LPS [31, 50]. However, in this study, the presence of particles in the medium was not necessary to produce the inhibitory effect in vitro, which makes this sequestration mechanism unlikely to play a role here. Other studies have demonstrated that negatively charged molecules are able to inhibit LPS activation of innate immune cells by interfering with signaling in the LPS-TLR4 cascade. Under normal circumstances, LPS is shuttled by LPS binding protein (LBP) to CD14 with subsequent passage to the TLR4/MD-2 receptor complex [51]. Mueller et al. showed that negatively charged phospholipids antagonistically block LPS from interacting with LPS binding protein (LBP), preventing the delivery of LPS to membrane-bound CD14 of mononuclear cells and macrophages [52]. Similarly, the negatively charged pulmonary surfactants phosphatidylinositol (PI) and palmitoyl-oleoyl-phosphatidylglycerol (POPG) inhibited the membrane signaling activity of LPS by antagonizing the signaling between LPS and CD14 and between LPS and LBP, CD14, and MD-2 respectively [51]. Negatively charged PLGA and PLA particles have the potential to provide a similar antagonistic influence on the extracellular signaling of the LPS-TLR4 cascade as these anionic phospholipids. Alternatively, or perhaps concurrently, these particles may be influencing the phenotype of monocytic cells, impairing their ability to respond to LPS. Allen et al. reported a particle-induced phenotypic resistance to LPS maturation by PLGA microparticles, which was attributed to the influence of immunomodulatory lactic acid derived from PLGA hydrolysis [25]. This response may explain the enhanced inhibitory effect observed by the all-lactic acid composition of PLA particles compared to the 50:50 lactic acid and glycolic acid of the PLGA particles in this study. Lactic acid, has been shown to delay LPS-induced genes, reduce NF-κB nuclear localization, and modify effector proteins such as TNF-α and IL-23 or chemokines such as MCP-1 and CCL7 [53]. In agreement with these findings, PLGA and PLA microparticles affected the maturation status of BMDCs and reduced the secretion of IL-12 when challenged with LPS [25]. While the cellular mechanisms impacted by particles have not yet been comprehensively identified, this study showed the influence of biomaterials to induce an LPS-resistant immunomodulatory phenotype of immune cells.

The highly negative particles prepared using the anionic surfactant PEMA were significantly more effective at inhibiting inflammatory cytokine secretions induced by LPS than those prepared using PVA (Figures 2,4). Against TLR agonist challenges with LPS (TLR4) and CpG-ODN (TLR9), PLGA and PLA particles formulated with PEMA modulated the secretion of sepsis-related cytokines IL-6, MCP-1, and TNF-α by multiple cell types (BMDCs, BMMØs, and in the diverse context of a mixed splenocyte culture) (Figures 2,4) [54–56]. The immunomodulatory effect of highly negative PEMA-formulated particles may result from an interference of the LPS-TLR signaling cascade, as described earlier, or may result from their enhanced cellular association and biodistribution. Interestingly, PLA particles formulated with PEMA had a 4-fold greater association with leukocytes and were measured with a greater diversity of innate immune cell types of the spleen (Figure S6). In splenocytes, the immunomodulatory effects of the particle formulations correlated with the overall trends observed in the monocultures of macrophages, dendritic cells, and Raw macrophages. Additional findings by Saito et al. showed that PLA particles have a higher association rate with monocytes and neutrophils in vivo than PLGA particles [21, 24]. These data support that the immunomodulatory influence of particles may be enhanced by increasing the association of particles with cells and the distribution of particles across a diversity of cell types.

The particle-induced downregulation of macrophage costimulatory molecules and inflammatory cytokines was associated with transcriptional changes that suggest an inhibition of MyD88- and TRIF-dependent TLR4 signaling pathways [57]. The MyD88-dependent pathway is fast-acting and initiates the translocation of NF-κB to the nucleus by deactivating the inhibitory action of IκB. This rapid signaling of NF-κB is associated with the production of inflammatory cytokines. These kinetics were observed in TRACER assays where LPS treatment induced a surge of NF-κB activity after 2 h (maximum at 4 h) and subsequent production of IL-6 and TNF-α. In support of this mechanism, treatment with PLA-PEMA, and to a lesser extent PLA-PVA, reduced the activity of NF-κB which was accompanied by an inhibition, or partial inhibition, of these cytokines, respectively (Figures 2,6). This inhibition of NF-κB signaling was observed in another study using PLGA-PVA microparticles, and was attributed to the interference of TAK1 and IKKβ phosphorylation, both kinases in the MyD88 cascade [25]. In parallel, the TRIF-dependent pathway is MyD88-independent and utilizes IRF3 to induce IFNβ protein expression which acts by autocrine and paracrine signaling to upregulate inflammatory genes via IRF1 and STAT1 [58]. In alignment with these signaling mechanisms, the PLA-PVA particles did not significantly modify the activity of IRF1 and STAT1 signaling, and correspondingly were less effective at reducing cytokine expression and were unable to alter costimulatory molecule expression on BMMØs (Figure 2). In contrast, PLA-PEMA particles reduced IRF1 and STAT1 signaling and costimulatory molecules were correspondingly decreased in BMMØs challenged with LPS. Together, the transcriptional activity indicates that PLA-PEMA particles inhibit signaling by the MyD88 and TRIF-dependent signaling pathways, whereas the PLA-PVA particles did not seem to influence the TRIF pathway.

The time of intervention for PLA-PEMA particles determined the extent of survival benefit in lethal LPS-induced endotoxemia mouse models (Figure 7). Clinical studies have shown that, despite a lowering in endotoxin and inflammatory cytokine levels within hours following antibiotic administration, the acute release of inflammatory mediators in the bloodstream has potentially lethal effects on the body’s ability to recover from septic shock [59]. In accordance with the presented in vitro results, mice treated prophylactically with particles followed by LPS challenge showed the most improved survival. Mice treated with particles following LPS challenge showed improved survival, yet to a lesser extent, which suggests a therapeutic window for particle intervention. The kinetics of transcriptional activity obtained from TRACER assays may inform why particles administered after induction of endotoxemia were less effective. The LPS-induced transcription factors, such as NF-κB, IRF1, and STAT1 act rapidly to initiate a cascade of inflammatory signals and phenotypic changes. This rapid transcriptional response, coupled with the in vivo endotoxemia data, suggests that the current version of particles may be acting to inhibit immune activation, but are less able to reverse previously initiated inflammatory signaling events. Thus, there appears to be a window of opportunity for effective treatment against severe TLR-induced inflammation [31]. Although LPS-induced endotoxemia has been used to model sepsis, this model is limited in that it does not capture the contribution of host-pathogen interactions to the disease severity [34]. Additional studies comparing different particle formulations, LPS challenge doses, administration schemes, and alternative models of sepsis may provide greater insights into the mechanisms of protection and opportunities for improving this particle-based intervention.

In this report, PLGA and PLA particles have demonstrated inherent immunomodulatory properties that are physicochemical property-dependent and their mechanism of TLR signaling inhibition by programming innate immune cells is broad acting. Collectively, particles interact with the immune cells central to the development sepsis and interfere with multiple differentially acting TLR agonists to inhibit an array of inflammatory outputs. The redundancy of cytokine release that is characteristic of sepsis have limited the success of strategies that specifically target cytokines or TLRs in humans [17] and is why particles may be more effective if they modulate multiple TLR pathways. The proposed mechanism of particle-based immunomodulation likely shares similarities with tolerogenic immune-modifying particles, where negatively-charged PLGA particles downregulate costimulatory molecule expression on innate immune cells [18, 33, 35–37, 60, 61]. The results presented here support the need for further investigation into the inherent immunomodulatory properties of polymer-based particles as they have the potential to clinically address a wide variety of human diseases with aberrant TLR activation.