Functionalized core-shell nanogel scavenger for immune modulation therapy in sepsis

Abstract

Sepsis is a complex, life-threatening hyperinflammatory syndrome associated with organ failure and high mortality due to lack of effective treatment options. Here we report a core-shell hydrogel nanoparticle with the core functionalized with telodendrimer (TD) nanotrap (NT) to control hyperinflammation in sepsis. The combination of multi-valent charged and hydrophobic moieties in TD enables effective binding with biomolecules in NT. The higher crosslinking in the shell structure of nanogel excludes the abundant large serum proteins and allows for size-selectivity in scavenging the medium-sized septic molecules (10–30 kDa), e.g., lipopolysaccharides (LPS, a potent endotoxin in sepsis), thus reducing cytokine production. At the same time, the core-shell TD NT nanogel captures the over-flowing proinflammatory cytokines effectively both in vitro and in vivo from biological fluids to further control hyperinflammation. Intraperitoneal injection of core-shell TD NT nanogel effectively attenuates NF-κB activation and cytokine production in LPS-induced septic mouse models. These results indicate the potential applications of the injectable TD NT core-shell nanogel to attenuate local or systemic inflammation.

Graphical Abstract

A core-shell nanogel with size exclusive property has been developed to immobilize a versatile telodendrimer nanotrap in the core for selective scavenge of the small-medium sized stimulating and signaling molecules in inflammation. The nanotrap captures both LPS and proinflammatory cytokines for effective control of peritonitis in vivo induced by LPS.

1. Introduction

Severe sepsis and septic shock are commonly associated with the patients in the intensive care unit (ICU) with high mortality. Sepsis is a complex inflammatory disease mediated by dysregulated innate immune reactions.^[1] Despite many efforts devoted to advance the ICU care of sepsis patients, it’s still extremely challenging to treat and the unacceptably high mortality rate remains unchanged in the past decades.^[2] Lipopolysaccharides (LPS), also called endotoxins, the structurally and functionally essential components of the outer cell wall of Gram-negative bacteria, play an important role in sepsis pathogenesis.^[3] LPS can be recognized by immune cells as a pathogen-associated molecular pattern (PAMP) and stimulate the host systemic immune response to secret massive proinflammatory cytokines, which may lead to organ dysfunction and failure.^[4] Neutralization of endotoxin presents a promising approach for inflammation control in sepsis treatments.^[5] The stem of immunogenic LPS is lipid A, a glycolipid component composing of a hydrophilic bis-phosphorylated diglucosamine backbone and a hydrophobic domain with six or seven acyl chains. The anionic and amphiphilic feature of lipid A is an attractive target for molecular design of LPS-binder to sequester the toxin.^[6]Anti-LPS therapies via cationic amphiphilic small molecules,^{[3, 7]} peptides,^[8] and proteins,^[9] have been developed, but have yet to show efficacy in reducing the mortality of sepsis. Simultaneous attenuation of both immune stimulating molecules, e.g., LPS and overflowing cytokines is critical for effective sepsis treatment.

Interestingly, most proinflammatory cytokines are negatively charged proteins and have small to medium molecular weights (less than 30 kDa) compared to abundant serum proteins.^[10] Yang et al. developed an injectable cationic hydrogel which exhibited a high binding capacity for proinflammatory molecules and thus reduced both local and systemic inflammation.^[11] In our previous study, we have developed a size exclusive hydrogel resin functionalized with telodendrimer nanotrap (TD NT), which exhibits high binding affinity to LPS and pro-inflammatory cytokines with both charge and size selectivity. TD NT hydrogel resin has demonstrated 100% survival when combined with antibiotics in a severe sepsis mouse model.^[12] This TD NT hydrogel resins, with size 200–500 µm, are readily incorporated into the standard clinical hemoperfusion therapy for sepsis treatment. Alternatively, the development of small-sized injectable nanoparticles with the similar selectivity and affinity to the broad spectrum of septic molecules is promising to attenuate systemic and local inflammation in situ for easy clinical administration and possibly improved efficacy in sepsis treatment.

Core-shell nanogels have been extensively studied for biomedical applications due to the engineerable particle properties and functionality.^[13] Different from solid nanoparticles, the swelling hydrogel structure in nanogel allows drug molecules or proteins to diffuse in and out for controlled drug release.^[14] In addition, the core can be functionalized with a ligand to capture specific analytes for biomedical detection.^[15] The biocompatible shell can be designed to shield the core and payload for in vivo applications. The structure and density of the shell can be controlled to alter porosity and permeability.^[16] Using precipitation polymerization, the parameters including size, crosslinking density, and the incorporation of functional groups in both core and shell compartments can be precisely controlled.^[17]

In this study, we created core-shell hydrogel nanoparticles with the core functionalized with a versatile hybrid telodendrimer nanotrap with multiple charge-hydrophobic moieties for septic molecules scavenging, e.g., LPS and proinflammatory cytokines. Poly(N-isopropylacrylamide-co-acrylic acid-co-2-methacryloyloxyethyl phosphorylcholine) (poly(NIPAm-co-AAc-co-MPC)) was applied to synthesize the core with carboxylic groups for nanotrap conjugation. Hydrophilic zwitterionic shell with higher crosslinking degree was further polymerized on the core nanoparticle using poly(N-isopropylacrylamide-co-2-methacryloyloxyethyl phosphorylcholine) (poly(NIPAm-co-MPC)) to selectively exclude the abundant large serum proteins by size exclusive effects. These hydrophilic nanogels have low cell toxicity and effectively sequester bacterial endotoxin and proinflammatory cytokines both in vitro and in vivo, demonstrating effective inflammation control in mouse model challenged with LPS. This core-shell nanogel is promising to serve as a novel injectable therapeutic modality to attenuate hyperinflammation for bacteremia sepsis treatment.

2. Results and Discussion

2.1. Synthesis of nanotraps

Our previous studies have demonstrated that negatively charged biomolecules such as LPS and proinflammatory cytokines can be effectively captured by telodendrimers (TD) comprising positive charges and hydrophobic moieties.^[18] Here we designed oligomers with multiple hydrophobic heptadecanoic acid (C17) moieties and positive charged moieties, such as arginine (Arg) or spermine (Spm) on the periphery of TD for efficient binding with the negatively charged proteins and LPS (Figure 1a). C17 can fit into diverse hydrophobic domains in protein with the flexible chain. Positively charged groups on the oligomers can interact with the phosphate groups on LPS. The oligomers were synthesized by solid phase peptide synthesis (SPPS) as previous description (Figure S1).^[19] Intermediates in the synthetic process can be cleaved from the resin for structural characterization by MALDI-TOF MS (Figure S2). The chemical composition of the final oligomers Arg₄C17₂ and Spm₂C17₂ were confirmed by ¹H NMR (Figure S3). The oligomers show high binding affinity with LPS isolated from E. coli (Figure 1b). In contrast, the PMB-LPS complex has moderate micromolar binding affinity and dissociates in electrophoresis.^[20]

Figure 1. Schematic illustration of the synthesis of core-shell nanogels.

(a) Structure of nanotraps; (b) Agarose gel electrophoresis profiles reveal the formation of FITC-LPS-nanotraps complex, as indicated by the lost mobility in migration; (c) Synthesis of core and core-shell nanogels with nanotraps; (d) Hydrodynamic sizes and zeta-potentials of core and core-shell nanogels measured by DLS (The data are presented as mean ± SD, n=3); The morphology of (e) C(Spm₂C17₂), (f) CS(Spm₂C17₂), (g) C(Arg₄C17₂) and (h) CS(Arg₄C17₂) by TEM imaging. Scale bar: 200 nm.

2.2. Hydrogel nanoparticle synthesis and characterization

The core and core-shell nanogels were prepared by precipitation polymerization in aqueous solution with N, N’-methylenebis(acrylamide) (BIS) as crosslinker (Figure 1c). NIPAm-based particles were chosen because of its unique thermal-sensitive property, as the monomer is readily soluble under the reaction conditions, while the newly formed polymers undergo a phase separation and aggregate into dense polymer globules during polymerization at high temperature.^[21] Meanwhile, it is feasible to control the particle size and the effective porosity by changing the percentage of cross-linking agent and temperature.^[22] Sodium dodecyl sulfate (SDS) is the surfactant used in the nanogel syntheses, which can stabilize the growing nuclei against aggregation early in the reaction.^[23] Zwitterionic monomer 2-methacryloyloxyethyl phosphorylcholine (MPC) was also applied to increase the hydrophilicity of the nanogel. Acrylic acid (AAc, 0.1%−2%) with carboxyl acid group was incorporated into the nanogel as anchors to introduce the nanotrap into the particles. During the shell coating step, the core particles serve as nuclei to graft/adhere the newly polymerized polymer chains. The permeability of the shell can be altered by adjusting the thickness and the crosslinking degree of the shell. We synthesized a core with low crosslinking degree 2% and size around 230 nm by dynamic light scattering (DLS) in solution. After covering the core particle with the shell, the size of the nanogel increased around 30 nm (Figure 1d). The higher crosslinking degree (5%) was applied in the shell structure to establish the size exclusive effects. We immobilized the nanotraps by conjugating amino groups in the oligomers to the NHS/EDC pre-activated carboxylic groups in the core particles (Figure 1c). The introduction of the nanotraps didn’t change the sizes of the nanogel (Figure S5) due to low percentage of AAc introduced to ensure efficient molecular diffusion. The morphology of the nanogel was revealed under TEM (Figure 1e-h and S4). The conjugation of nanotraps didn’t change the morphology of the nanogel. The covering of shell can increase the size of the nanogel from 150±15 nm (C(Spm₂C17₂)) and 147±12 nm (C(Arg₄C17₂)) to 173±14 nm (CS(Spm₂C17₂)) and 162±11 nm (CS(Arg₄C17₂)). Although the core-shell structures were barely recognizable under TEM due to the similar chemical components, the zeta potential of nanogel significantly decreases to neutral after shell coating on the positively charged core particles with nanotrap conjugated (Figure 1d).

2.3. Optimization of the properties of core-shell nanogel

In a core-shell architecture, the properties of the core and shell can be tailored separately for specific application. In this study, we delicately design the core for effective binding of small sized cytokines and LPS; thus, a loose network in the core (2% crosslinking degree) is favorable for analytes diffusion and capture. The shell with higher crosslinking degree (5%) was then separately coated onto the core, which has a denser network than the core to create size-exclusive effects to exclude the essential large serum proteins.

Before the nanotrap conjugation, the blank pNIPAM core and core-shell nanogels were tested for bioscavenging, which didn’t exhibit any effective capture for both proteins (BSA and α-LA) and LPS (Figure S7). In comparison, the loading capacity of the core nanogel can be optimized by both density and valency of the introduced TD nanotrap. The density can be adjusted by the percentage of AAc in the core nanogel for nanotrap conjugation via amide bond formation. We first prepared nanogel with a AAc density of 2% for nanotrap immobilization. As shown on agarose gel, fluorescently labeled LPS can be completely encapsulated into the nanogel at LPS/particle mass ratio of 1/100 (Figure 2a). However, we observed the reduced particle sizes of nanogel from 295 nm to 260 nm (Figure S6) after nanotraps conjugation at 2% density. It is due to the aggregation of closely adjacent nanotraps at high density resulting in the contraction of hydrogel network, which may hinder the kinetics for analytes to penetrate and capture. Therefore, we prepared a series of nanogel with reduced AAc density of 0.1%, 0.2% and 0.6%, in the core respectively. As expected, less impact on particle sizes was observed after nanotraps conjugation at the reduced density (Figure S5, S6). More importantly, the reduced density didn’t sacrifice the capacity of nanogel for LPS capture (Figure 2a). All the core nanogel with mono-valent nanotrap (Arg₂C17 and SpmC17) at density from 0.1% to 2% can completely absorb LPS at 100/1 mass ratio, which correspond to the molar ratios of nanotrap to LPS ranging from 10:1 to 200:1. The higher nanotrap density in nanogel will self-assemble into nano-domain similar to micelles formed by telodendrimers in solution,^[18b] which may not be accessible for LPS or protein binding due to the restricted mobility in hydrogel network, thus resulting in low LPS binding capacity even at high 2% AAc density. It is interesting that fluorescent signal of FITC-LPS was enhanced significantly after loading in the nanogel with high density of nanotraps, due to the compact hydrophobic environments, which was consistent with the nanotrap with the increased number of C17 (Figure 2b). In order to further improve LPS capturing efficiency, we increase the valency of the nanotrap by increasing binding affinity. As expected, the doubling of valence of both charge and hydrophobic moieties Arg₄C17₂ in nanotraps significantly increases the loading capacity of LPS in the core of nanogel from 100/1 to 25/1 (nanogel/LPS mass ratio) as compared to the monomeric Arg₂C17 (Figure 2b). Similarly, the increased valency of nanotraps with spermine as the positive charges also captures LPS more effectively at a 25:1 mass ratio and 0.1% nanotrap density.

Figure 2. Size-exclusive property and LPS capture in nanogel.

Agarose gel electrophoresis showing dose dependent efficacy for LPS capturing (a) by the core nanogels with arginine- or spermine-containg nanotraps at different density (0.1%, 0.2%, 0.6% and 2%); (b) or by the core nanogels with nanotraps at different valency (mono- and bi-valent); (c) the LPS capture in the core and core-shell nanogels with bivalent nanotrap at 0.1% density in comparison with nanogel conjugated with polymyxin B (PMB) at 25:1 or 50:1 mass ratio; (d) selective absorption of small and negatively charged model biomolecules, LPS or α-LA with the exclusion of BSA by size exclusive effect in nanogels with arginine-containing bivalent nanotraps: the core or core-shell nanogel were incubated with LPS/α-LA, BSA or the mixture of LPS/α-LA and BSA (1:1 molar ratio) before loaded on agarose gel ectrophoresis; (e) TNF-α and IL-6 level in RAW 264.7 cell culture medium with overnight stimulation with LPS (50 ng/mL), nanogel (10 μg/mL), LPS/PMB (1:20 mass ratio) or LPS/nanogel (pre-incubated for overnight at mass ratios of 1:50, 1:100, 1:200), respectively. (n = 3, mean ± SD. Statistical significance was measured by one-way ANOVA).

After core optimization for effective LPS adsorption, it is important to create selectivity for analytes capture from bio-fluids. A biocompatible and size-exclusive shell layer on the functional core may exclude various proteins abundant in blood, e.g., serum albumin and antibodies. A similar chemical component poly(NIPAm-co-MPC) except for AAc was polymerized on the surface of core nanogel via precipitation polymerization. Meanwhile, 25% MPC were incorporated to increase the hydrophilicity of surface chemistry and reduce nonspecific binding. In addition, the crosslinking degree was increased to 5% in order to create size exclusive effects. Fortunately, the coverage of a shell layer did not influence the adsorption of small proteins in the core. As shown in figure 2c, the core-shell nanogels with the optimized density (0.1%) and valency (bivalent) of nanotrap in the core exhibited the same loading capacity for LPS at 25:1 mass ratio.

Polymyxin B (PMB), a LPS-binding antibiotic, was also conjugated in the core of nanogel through the same procedure for LPS binding.^[24]As comparison, it showed low binding affinity to LPS as indicated by fluorescent polarization (FP) spectrometry studies with minimum noticeable changes in FP reading (Figure S8). Accordingly, PMB containing nanogel is also not effective for LPS capture at 50/1 mass ratio as shown in figure 2c. Further, we incubated FITC-LPS with the core or core-shell nanogels containing Arg₄C17₂ or Spm₂C17₂ in the presence of four-times more PMB in mass relative to nanogel for competing LPS binding. As results, LPS can be captured effectively in both core and core-shell nanogels as shown in agarose gel electrophoresis assay (Figure S9), confirming effective LPS binding in nanogel.

To determine the size-selective effect of core-shell nanogels, a mixture of FITC labeled LPS or α-LA and Rhodamine B labeled BSA (1:1molar ratio) was incubated with either core or core-shell nanogels for 2 h and characterized with agarose electrophoresis (Figure 2d). The core nanogels effectively bind to both larger BSA (67 kDa) and smaller protein (α-LA) and LPS. In contrast, the binding of larger BSA were significantly decreased after incubation with core-shell nanogels, while LPS and α-LA can still be effectively captured (Figure 2d). This result indicates that the coverage of a higher crosslinking shell on the core particles can effectively create a size-exclusive effect to exclude large proteins, such as albumin, but remains effective for the binding of broad range of small negatively charged proteins and endotoxin.

LPS, shedding from gram-negative bacteria, can effectively stimulate immune cells to secrete inflammatory mediators, such as tumor necrosis factor-α (TNF-α) and interleukin-6 (IL-6). These proinflammatory cytokines are very potent in mediating inflammatory reactions to control infections, but can also cause tissue damage when dysregulated in endotoxemia, thus, can be used as biomarkers for prognosis of sepsis.^[25] We co-incubated LPS (50 ng/mL)^[26] with our core or core-shell nanogels containing nanotraps of Spm₂C17₂ or Arg₄C17₂ before added into culture medium for macrophage-like RAW 264.7 cells incubation in comparison with free LPS stimulation. PMB is a potent LPS-binding antibiotic and was used as an antagonist for LPS in immune cell stimulation. In addition, PMB was also introduced in the core of the nanogel to compare with our nanotrap nanogels. After overnight incubation, cell culture medium was collected, and the supernatant was obtained for ELISA analysis for TNF-α and IL-6 production (Figure 2e). As expected, free LPS induced abundant cytokine productions for both TNF-α and IL-6, which can be effectively attenuated by free PMB. The blank nanogels are not immunogenic and didn’t induce cytokine production (Figure 2e). Nanogels functionalized with PMB also exhibited dose-dependent inhibition of TNF-α and IL-6 production significantly compared with free LPS group. However, the nanotrap-functionalized nanogels with both core and core-shell structures exhibited superior inhibitory effects for cytokine productions, which is correlated with the stronger LPS binding in nanotrap than PMB. Interesting, the functionalized core nanogel had a reduced efficacy with the increased nanogel/LPS mass ratio from 100:1 to 200:1 for both nanotrap structures of Spm₂C17₂ and Arg₄C17₂, indicating the surface binding of LPS on the core nanogel, which may still present to immune cells for immune stimulation. In contrast, the core-shell nanogels with inert surface exhibited dose-dependent efficacy for LPS attenuation.

2.4. Biocompatibility and cell uptake

NIPAm containing particles have been extensively studied for biotechnological applications with respect to their thermo-responsivity.^[27] Both core and core-shell nanogels exhibit good hemocompatibility when incubated with red blood cells for 0.5 h and 4 h at a concentration up to 500 µg/mL, without significant hemolytic activities observed (Figure 3a). Blank core and core-shell nanogel nanoparticles have been tested to be non-toxic up to 1 mg/mL in cell cultures with both immune cells and normal fibroblast cells (Figure S10). After TD conjugation, nanogels still exhibit good biocompatibility in cell culture with immune cells (Figure 3b), e.g., RAW264.7 (murine macrophage) and THP-1 (human monocyte). Interestingly, the core and core-shell nanogel with nanotrap Arg₄C17₂ show better biocompatibility than nanogel with Spm₂C17₂. The cells show 100% survival at concentration at 250 µg/mL after 72 h incubation with C(Arg₄C17₂) and CS(Arg₄C17₂).

Figure 3. In vitro cytotoxicity characterization.

(a) In vitro hemolytic activity of nanogels after incubation with red blood cells at different concentrations for 0.5, 4 and 24 h, and (b) cell viability of RAW 264.7 and HTP-1 cells after 72 h incubation with nanogels C(Spm₂C17₂), CS(Spm₂C17₂), C(Arg₄C17₂) and CS(Arg₄C17₂), respectively at different concentrations. The data was represented as the mean ± SD based on three separate experiments.

As nanotrap is designed to scavenge circulating septic molecules and cytokines, the phagocytosis of nanogel is preferred to be low to ensure longer circulation and effective immune modulation. It has been demonstrated that nanoparticles with antifouling zwitterionic group on the surface have long circulation time in human blood with the reduced nonspecific uptake by macrophages. Thus, we incorporated 25% of zwitterionic monomer MPC into the shell. Immune cell uptake was evaluated in cell culture of both murine macrophage RAW 264.7 and human monocyte THP-1 cells to compare the stealth properties of the core and core-shell nanogels. Nanogels were preincubated with Cy3-α-LA at mass ratio 50:1 for 30 min for complete adsorption as shown in agarose gel electrophoresis (Figure S11), then added into the cell culture medium (final concentration of α-LA is 5 μg/mL). The cells were fixed, and nuclei were stained with DAPI after 30 min incubation for microscope observation. As shown in figure 4, the core nanogels with Spm₂C17₂ or Arg₄C17₂ nanotraps have effective uptake by both immune cells, due to the positive charges as shown in zeta-potential analysis in figure 1. The core-shell nanogels significantly reduced the cell uptake, especially for CS(Arg₄C17₂) with negligible cell uptake, which was confirmed by quantitative fluorescent intensity analysis on cell signals (Figure S12). It was also observed that cell morphology changes significantly after nanogel incubation, except for CS(Arg₄C17₂), which is consistent with the cell viability assays (Figure 3b). Thus, we focused on the core-shell nanogel CS(Arg₄C17₂) in comparison with the core nanogel in the following in vitro and in vivo studies.

Figure 4. Reduced cell uptake of the core-shell nanogels.

The fluorescent microscopic images of (a) RAW264.7 and (b) THP-1 cells after incubation with the core and core-shell nanogel encapsulated with Cy3-α-LA at a concentration of 250:5 μg/mL for 30 min at 37 °C: cell nucleus was stained into blue with DAPI, red color is Cy3-α-LA. Scale bar: 25 μm.

2.5. LPS attenuation in vivo

To determine the LPS attenuation capability of nanogel in vivo, we apply wild type BALB/c mice to evaluate the efficacy of nanogel for inhibiting cytokine production in vivo after LPS i.p. injection. LPS was pre-incubated with the core or core-shell nanogels at mass ratio of 1/100 and then i.p. injected into the mice at a LPS dose of 0.1 mg/kg.^[28] The mice were sacrificed 2 h post-injection, and blood plasma and intraperitoneal lavage were collected for cytokine analysis. Figure 5a-b shows the levels of the pro-inflammatory cytokines, TNF-α and IL-6, in the plasma and lavage. A significant increase in the pro-inflammatory cytokines was observed in free LPS group compared to the control PBS group. Blank core and core-shell nanogels didn’t induce significant cytokine production. The LPS/nanogels groups exhibited a significant reduction in cytokine production for both TNF-α and IL-6, especially in the plasma, compared to the LPS group. Consistent with the ex vivo cell culture study, the functionalized core nanogel has less efficacy in attenuating LPS in cytokine stimulation compared to the core-shell nanogel. It showed that the empty nanogels are not immunogenic. Interestingly, free LPS didn’t induced higher cytokine production in the intraperitoneal lavage than in the blood plasma, which may be due to the quick diffusion of LPS into systemic circulation stimulating white blood cells (WBC) for cytokine production. The significantly reduced cytokine production in the blood by core-shell nanogel indicates effective LPS attenuation. At the same time, we analyzed the blood counts in mice after sacrificed (Figure 5c-d), showing significantly reduced WBC and neutrophil counts in free LPS group, which were effectively attenuated by core-shell nanotrap, whereas no significant attenuation was observed in mice treated with the nanotrap core structure. In conclusion, the introduction of an inert shell can effectively shelter the septic molecules adsorbed in the core of nanogel to effectively attenuate the LPS stimulation of inflammation in vivo.

Figure 5. LPS attenuation via core-shell nanogels in vivo.

The level of cytokines TNF-α (a) and IL-6 (b) in the plasma and peritoneal lavage fluids of BALB/c mice were analyzed at 2 h after the i.p. injection of LPS (0.1 mg/kg), nanogel with (Arg₄C17₂) moieties (10 mg/kg) or nanogel/LPS (0.1 mg/kg/10 mg/kg, nanogels were pre-incubated with LPS for overnight before i.p. injection); The numbers of white blood cells (c) and neutrophil (d) in these mice were analyzed 2 h after treatments; Real-time in vivo bioluminescent imaging (e) and the quantitative bioluminescent analysis (f) of the HLL NF-κB reporter mice were recorded at time 0, 2, 6, 24 h after i.p. injection of PBS, CS(Arg₄C17₂) (10 mg/kg), LPS (0.1 mg/kg) or LPS/CS(Arg₄C17₂) (0.1mg/kg/10mg/kg, pre-incubated overnight), respectively. Bioluminescence was acquired by in vivo imaging system (IVIS) 10 minute after i.p. injection of luciferin solution (200µL/20g mice, dose of 150mg/kg) (n = 3–5, mean ± SD. One-way ANOVA has been applied for the comparison between multiple groups).

LPS binds to TLR-4 receptor triggering intracellular signaling pathways, mainly through nuclear factor-κB (NF-κB) activation to promote inflammatory cell gene expression.^[29] Thus, we apply HIV-LTR/Luciferase (HLL) mice as luciferase reporter for NF-κB activation and inflammatory responses to LPS stimulation with/without nanogels. HLL mice were imaged at t=0 for baseline and at 2, 6, 24 h after i.p. injection of PBS, CS(Arg₄C17₂), LPS or LPS/CS(Arg₄C17₂), respectively (Figure 5e, f). As shown in the bioluminescent imaging in figure 5e, LPS-treated mice exhibited significantly increased luminescent signal in the peritoneal space at 2 h post injection and decreased to almost baseline after 6 h to 24 h, which is correlated with the peak of LPS-induced inflammation reported in the literature.^[30] However, the signal remained low in mice treated with nanogel-treated LPS solution throughout the observation, indicating the attenuated activity of LPS. Blank nanogel injection didn’t induce any inflammatory signals as with PBS injection (Figure 5f).

In order to mimic clinical disease treatment, we applied LPS and nanogel separately through i.p. injection to determine further if nanogel binding with LPS in vivo inhibits cytokine production for inflammation control. We first i.p. injected CS(Arg₄C17₂) nanogel (20mg/kg) into HLL NF-κB reporter mice (n=3) with gentle massage to let nanogel homogeneously distributed in the peritoneal cavity. Then, LPS (0.1mg/kg) was i.p. injected 5 min later. PBS was used to replace either LPS or nanogel to generate either negative or positive control for inflammation induction. As shown in figure 6a, the separately administrated core-shell nanogel effectively attenuated the LPS-induced NF-κB activation as indicated by the reduced bioluminescent signals in the abdominal area in LPS-challenged mice treated with nanogel in comparison to the LPS-challenged mice. The abdominal bioluminescent signal was quantified and normalized by the signal before LPS injection (T0) as shown in figure 6b. Further, the area under curve (AUC) of the relative bioluminescent signal were calculated for each group of mice to reflect the relative overall inflammation level within 24 h (Figure 6c). The AUC analysis revealed the same trend of the reduced inflammation by the separately injected nanotrap for effective in vivo LPS attenuation.

Figure 6. In situ LPS attenuation via core-shell nanogels in vivo.

Real-time in vivo imaging (a) and the relative bioluminescent flux normalized by the background signal before treatment (t=0) (b) and the area under curve (AUC) of relative flux (c) for LPS-induced peritoneal inflammation in HLL mice treated, respectively, with LPS (0.1 mg/kg), CS(Arg₄C17₂) (20 mg/kg) or CS(Arg₄C17₂)/LPS (0.1mg/kg / 20mg/kg, LPS was i.p. injected 5 min after nanogel i.p. injection) at 0, 2, 6, 24 h. Basal bioluminescence level was acquired by IVIS. Luciferin solution (150 mg/kg) was i.p. injected 10 minute before the abdominal bioluminescence acquiring (The data are presented as mean ± SD, n=3).

2.6. Cytokine removal from biological fluids.

The management of hyperinflammation is crucial for the treatment of sepsis. Inflammatory mediators, such as TNF-α and IL-6, are produced by immune cells in response of infection, which stimulates strong immune reactions and even hyperinflammation if released into the circulation. Most of the proinflammatory cytokines have small molecular weight (10–30 kDa) and are negatively charged.^[12] We speculated the nanogels may be able to absorb these cytokines, which has a similar molecular weight and negative charges like LPS. Thus, we collected several cytokine containing solutions for incubation with nanogels to test their efficacy for cytokine adsorption. First, pure TNF-α and IL-6 solutions was incubated with the core-shell nanogel with (Arg₄C17₂) nanotrap and were effectively removed in a dose-dependent manner with >90% efficiency at a mass ratio of 200:1 evaluated by ELISA (Figure S13).

Cell culture medium of RAW264.7 cells after LPS stimulation (50 ng/mL) for 24 h was harvested and incubated with nanogels to characterize efficiency for cytokine adsorption. Significant cytokines, e.g., TNF-α and IL-6, were produced in the cell culture medium after LPS stimulation (Figure 7a, b). Both core and core-shell nanogels with positively charged (Arg₄C17₂) or (Spm₂C17₂) nanotraps can effectively adsorb both TNF-α and IL-6 in a dose dependent manner as measured by ELISA assays. The core-shell nanogels exhibited better efficacy than the core nanogels. In contrast, nanogel with PMB or free PMB didn’t show any capacity for the adsorption of these two cytokines (Figure 7a, b). Further, the plasma and peritoneal lavage fluid collected from septic mice 24 h after the cecum ligation and puncture (CLP) surgery for incubation with different concentrations of core and core-shell nanogels, respectively. After overnight incubation, TNF-α and IL-6 level in the fluids were analyzed by ELISA. Similarly, both core and core-shell nanogels can scavenge TNF-α and IL-6 production in a dose-dependent manner, which was significantly better than the PMB-containing nanogel (Figure 7c-f). The adsorption of cytokines in both cell culture media, and septic blood and body fluids was more effective after incubation with core-shell nanogels compared to the core nanogels, which is expected due to the size excusive effects of shell layer preventing the competition of large abundant serum proteins.

Figure 7. The management of hyperinflammation with nanogels in vitro.

The removal of cytokines TNF-α and IL-6 by core and core-shell nanogel with nanotraps in compare with the nanogel with PMB in (a, b) RAW264.7 cell culture medium after LPS stimulation (50 ng/mL) for 24h; (c, d) Sepsis mice plasma and (e, f) peritoneal lavage collected from mice 24 h after the CLP surgery (n = 3, mean ± SD. Statistical significance was measured by one-way ANOVA).

2.7. Attenuation of septic molecules in human blood

To determine the potential efficacy of the nanogel in treating human sepsis patients, we collected minimum amount of blood from surgical sepsis patients and pooled eight de-identified patient blood together to get an average cytokine level for nanogel incubation. The cytokine concentrations in plasma were analyzed via ELISA with or without nanogel incubation (0.2 mg/mL) overnight. As shown in figure 8a-b, there is significant decrease of both TNF-α and IL-6 after incubation with both CS(Arg₄C17₂) and CS(Spm₂C17₂). Further, fresh blood was collected from healthy volunteer and doped with LPS (1 ng/mL), or LPS pre-incubated or freshly added together with CS(Arg₄C17₂) or empty CS nanogel at a mass ratio of 1:200. The activation of neutrophils in the blood sample were evaluated via an FDA approved endotoxin activity assay (EAA) based on the chemiluminescent measurement of the redox species produced mainly by neutrophils after phagocytosis of the immune complex of LPS with the added murine anti-LPS IgM antibody. EAA is correlated with both LPS concentration and neutrophil activity to diagnose the severity of sepsis in the clinic (EAA range: 0–1, normal level <0.3; medium risk: 0.3–0.6; severe sepsis: >0.6). Given the same healthy blood applied in this study, the EAA score only reflects the concentration of the free LPS that forms the complex with IgM. As shown in figure 8c, 1 ng/mL LPS induces significant EA score to ~ 0.6, which mimics relatively severe sepsis. With pre-incubation of LPS with the non-functionalized nanogel, EAA score was barely decreased. In comparison, the CS(Arg₄C17₂) (0.2 µg/mL) significantly reduced EAA to safe level <0.3, irrespective of pre-incubation or added concurrently into blood, indicating effective LPS attenuation.

Figure 8. Cytokine scavenging and LPS adsorption in human blood:

(a) TNF-α and (b) IL-6 can be effectively absorbed by the functionalized core-shell nanogel from septic patients after overnight incubation at 0.2 mg/mL; (C) Endotoxin activity in stimulating neutrophil for redox production was applied to evaluate the efficiency of nanogel (0.2 µg/mL) for attenuation of LPS doped (1 ng/mL) in healthy human blood. (n = 3, mean ± SD. Statistical significance was measured by one-way ANOVA. *: p<0.05; **: p<0.01; ***: p<0.001; ****: p< 0.0001)

3. Conclusion

In the present study, we have developed a core-shell nanogel with size-exclusive effect for selective adsorption of small-medium sized septic molecules for effective immune modulation. The introduction of dendritic binding clusters combining both positive charges and hydrophobic moieties in the core of nanogel enables the effective absorption of LPS and pro-inflammatory cytokines. The shell layer with higher crosslinking degree on the surface prevents the large and abundant serum proteins to compete for binding. These nanogels show high efficacy for LPS attenuation both in vitro and in vivo in mouse model, showing protective effect for LPS induced inflammation in vivo. In addition, nanogel shows effective adsorption of both cytokines and LPS from human blood for clinical translation. In this context, the results indicated core-shell nanogels to be a promising therapeutic agent to treat endotoxemia and sepsis, which will be explored in our future study.

4. Experimental Section/Methods

4.1. Materials

N-Isopropylacrylamide (NIPAm, TCI, +99%) was recrystallized by hexane and dried in vacuum before use. Acrylic acid (AAc, +99%), heptadecanoic acid (C17, +98%), 2-methacryloyloxyethyl phosphorylcholine (MPC, +96%) were purchased from TCI. N, N’-methylenebis(acrylamide) (BIS, +99%), di-tert-butyl decarbonate (Boc₂O, +98%) and sodium dodecyl sulfate (SDS, +98%) were obtained from Alfa Aesar. tetra-Boc-spermine-5-carboxylic Acid, Rink amide-MBHA resin (HCRAm 04–1-1) was ordered from Nankai HECHENG S&T Co., Ltd (Tianjin, China). (Fmoc)-Lys(Boc)-OH, (Fmoc)-Lys(Fmoc)-OH, (Fmoc)-Lys(Dde)-OH, trifluoroacetic acid (TFA), N-(3-Dimethylaminopropyl)-N’-ethylcarbodiimide hydrochloride (EDC HCl) were obtained from Chem-Impex International, Inc. (Wood Dale, IL). (Fmoc)-Arg(Pbf)-OH was purchased from AnaSpec Inc. (San Jose, CA). N, N’-diisopropylcarbodiimide (DIC), N-hydroxybenzotriazole (HOBt), succinic anhydride, 4-dimethylaminopyridine (DMAP), N, N-dimethylformamide, anhydrous (DMF, 99.8%) chloroform (CHCl₃), methylene chloride (DCM), methanol (MeOH), dimethyl sulfoxide (DMSO), potassium persulfate (KPS) and N-hydroxysuccinimide (NHS, +98%) were received from Acros Organics (Belglum, NJ). Tert-Butyl bromoacetate, N-hydroxysuccinimide (NHS), triethylamine (TEA), polymyxin B (PMB), hydrazine hydrate, triisopropylsilane (TIS), α-lactalbumin (α-LA from bovine milk), and LPS from Escherichia coli (L4130) were purchased from Sigma-Aldrich (St. Louis, MO). Tetrazolium compound [3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, MTS] and phenazine methosulfate were purchased from Promega (Madison, WI). ELISA kits were purchased from Invitrogen for direct use (e.g., Mouse IL-6 Cat. #: 88–7064-88, Mouse TNF-α: Cat. #: 88–7324-88, Human IL-6 Cat. #: 88–7066-88 and Human TNF-α: Cat. #: 88–7346-88).

4.2. Synthesis of nanotraps

Oligomers bearing both positive charge groups and hydrophobic groups were formulated by solid phase peptide synthesis (SPPS) following a published procedure.^[19] Starting from Rink amide resin (0.5 mmol/g), Fmoc-Lys(Fmoc)-OH, Fmoc-Lys(Dde)-OH, and Fmoc-Arg(Pbf)-OH were coupled sequentially following the standard peptide synthesis procedures. DIC and HOBt were used as catalytic coupling reagents. All the coupling reactants were in three-fold excess to amine functional group on the resin. Fmoc protecting group was removed by the treatment of resin with 20% 4-methylpiperidine solution in DMF for 30 min. De-Dde was carried out in 2% hydrazine DMF solution for 5 min and three times. The completion of the reaction was monitored by the ninhydrin test and confirmed by MALDI-TOF MS of the cleaved compound. After completion, residual reactants were removed by filtration under vacuum and washed with copious solvents of DMF, DCM, and MeOH. The cleavage of dendrons from rink resin was conducted in TFA/TIS/H₂O (95/2.5/2.5, v/v/v) cocktail. The final oligomers were purified by precipitation with cold ethyl ether followed by dialysis against water.

4.3. Synthesis of core nanogel

Poly(NIPAm-co-AAc-co-MPC) nanogel was prepared by free radical precipitation polymerization. N-isopropylacrylamide (NIPAm, 900 mg, 7.95 mmol), acrylic acid (AAc, 0.64 mg, 0.0088 mmol), 2-methacryloyloxyethyl phosphorylcholine (MPC, 260 g, 0.88 mmol), N, N’-methylene bisacrylamide (BIS, 27.2 mg, 0.18 mmol) and sodium dodecyl sulfate (SDS, 30 mg) were dissolved in 100 mL of DI water in a two-neck round-bottom flask. The solution was purged with nitrogen for 1 h at room temperature, at medium stirring rate, and then heated to 70 ^oC. Potassium persulfate (KPS, 23.5 mg, 0.088 mol) was dissolved in 1 mL of DI water and purged with nitrogen for 10 min and then added to the solution to initiate the polymerization. The reaction was maintained at 70 ^oC under nitrogen for 30 min. The reaction mixture was then cooled, transferred into a pre-washed dialysis tube with molecular cutoff ~3.5kDa and dialyzed for 2 days.

4.4. Synthesis of core-shell nanogel

The core-shell nanogel were synthesized via a two-stage seeded polymerization method with the core particle as seeds. 50 mg of poly(NIPAm-co-AAc-co-MPC) was dissolved in 10 mL DI water and sonicated to ensure complete solubilization. The particle suspension was purged with nitrogen for 1 h and heated at 70 ^oC. NIPAm (20 mg, 0.18 mmol), MPC (17 mg, 0.059 mmol), BIS (1.7 mg, 0.011 mmol), and KPS (0.6 mg, 0.0022 mmol) were dissolved in 2 mL of water and purged with nitrogen for 10 min. 0.3 mL of this solution was added to the core nanoparticle suspension, and the remaining 1.7 mL of solution was added in aliquots of 0.3 mL every 5 min. The reaction was allowed to proceed at a temperature of 70 °C for 3 h. Poly(NIPAm-co-AAc-co-MPC) core-poly(NIPAm-co-MPC) shell particles were dialysis with molecular cutoff ~3.5kDa for 2 days.

4.5. Functionalization of core and core-shell nanoparticles

The nanotraps containing amine group were conjugated by condensation to the carboxylic group of AAc present in the poly(NIPAm-co-AAc-co-MPC) core nanogel. A preliminary activation of the carboxylic group present in the nanoparticles was performed. Briefly, 10 mL core or core-shell nanogel suspensions (5mg/mL) were added 0.84 mg (0.0044 mmol) of EDC·HCl and 0.51 mg (0.0044 mmol) of NHS. The reaction was held at room temperature and stirred for 2 h. Then, an appropriate amount of nanotraps (nanotrap/AAc = 10/1, molar ratio) was dissolved in 1 mL of water and added to the reaction mixture with triethylamine; the reaction was held at room temperature at medium stirring rate for 24 h. In order to remove the unreacted small molecules, the core-shell nanoparticles were dialysis with molecular cutoff ~3.5kDa for 3 days.

4.6. Electrophoresis assays

The loading capacities of the nanogel to LPS, α-LA or BSA were investigated using electrophoresis assay. The electrophoresis was carried out in 1.5% agarose gel (Tris-borate-EDTA (TBE) buffer) at constant current of 35 mA for 30 min. The gel was imaged by a Bio-Rad Universal Hood II Imager (Bio-Rad Laboratories, Inc.) under SYBR Green modes or photographed under UV illumination.

4.7. Fluorescent polarization assays

The binding of nanogel to LPS were evaluated and compared with polymyxin B (PMB) by fluorescence polarization (FP) using Multi-Mode Microplate Reader (SynergyTM 2, Biotek, VT) equipped with dichroic mirror (510 nm) and polarizing filter. The measurements were carried out on black flat-bottom 96-well plates (Nunclon™ Surface, Roskilde, Denmark). LPS-FITC with different ratios of nanogel or PMB were incubate in the 96-well plate for 1h. The FP of LPS-FITC was recorded at excitation and emission filter of 485/20 nm and 528/20 nm, respectively. The experiments were performed in triplicate.

4.8. Hemocompatibility assay

Fresh human blood was collected from healthy adult volunteers used in the study. The study was approved by the SUNY Upstate Institutional Review Board (IRB # 754811). Fresh blood was diluted into 5 mL of 20 mM EDTA PBS. Red blood cells (RBCs) were separated by centrifugation at 3,000 rpm for 10 min and then washed three times with 10 mL of PBS. The RBCs were re-suspended in 20 mL of PBS and 200 μL of RBC solution was added into each well of a 96-well plate. The nanogel solutions were added into the RBCs suspensions to final concentrations of 10, 100, and 500 μg/mL and incubated at 37 °C. At determined time (0.5 h, 4 h and 24 h), the mixtures were centrifuged at 3,000 rpm for 3 min, and then the hemoglobin in the supernatant was determined by measuring the UV-Vis absorbance at 540 nm (NanoDrop 2000c spectrophotometer, Thermo Scientific). The RBCs incubated with Triton-100 (2%) and PBS were used as positive and negative controls, respectively. The hemolysis ratio of RBCs was calculated using the following formula: Hemolysis% = (OD_sample - OD_PBS) / (OD_triton - OD_PBS) × 100%. All hemolysis experiments were carried out in triplicates.

4.9. Cell culture and cell viability assay

Murine macrophage-like RAW 264.7 cells, human fibroblast HFF-1 cells and human monocytic THP-1 cells were purchased from American Type Culture Collection (Manassas, VA). RAW 264.7 and HFF-1 cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) medium, THP-1 cells were cultured in Roswell Park Memorial Institute (RPMI) 1640 medium, supplemented with10% fetal bovine serum (FBS), 100 U/mL penicillin G and 100 mg/mL streptomycin at 37 °C in a humidified incubator with 5% CO₂. The cytotoxicities of nanogel with or without nanotraps were studied by measuring cell viability via MTS assays. Cells were seeded at a cell density of 3×10³ (RAW 264.7, HFF-1) and 8×10³ (THP-1) cells per well in 96-well plates. After overnight incubation, nanogels with different concentrations were added in each well to treat the cells. After 72 h incubation, CellTiter 96 AQ_ueous MTS reagent was added to each well according to the manufacturer’s instructions. The cell viability was determined by measuring the absorbance at 490 nm using a microplate reader (BioTek Synergy H1) with the untreated cells served as negative controls. Results were obtained as the average cell viability of triplicate experiments calculated by a formula of [(OD_treat - OD_blank) / (OD_control - OD_blank) × 100%].

4.10. Cellular uptake of nanogels by macrophages

The cellular uptake of core and core-shell nanogels were determined by fluorescence microscopy. α-LA was used as a model protein, which was chemically labeled with fluorescent dye of Cy-3. Raw 264.7 and THP-1 cells (1×10⁴ cells/well) were plated 96-well plates in the cell culture medium and incubated overnight. The Cy3-α-LA loaded nanogels (mass ratio 1:50) were added into the cell culture medium at a final concentration of Cy3-α-LA 5 μg/mL. After half hour incubation, the cell culture medium was removed. The cells were fixed by 4% paraformaldehyde and the nuclei were stained by 4’,6-diamidino-2-phenylindole (DAPI). Finally, the cells were imaged by microscope.

4.11. LPS attenuation in vitro

Raw 264.7 Cells were plated in 96-well plates at a density of 2×10⁴ cells per well. Stock solution of LPS (50 μg/mL) was pretreated with nanogels/PMB at different mass ratio for overnight incubation before being added into macrophage cell culture. The untreated stock LPS solution was directly added to the cell culture to a final LPS concentration of 50 ng/mL as a control for cytokine production comparison. After overnight incubation, cell medium was collected, and the supernatant was obtained by centrifugation. The level of TNF-α and IL-6 production were assessed using the commercial ELISA Kit.

4.12. LPS attenuation in vivo

BALB/c mice (7–8 weeks) purchased from Charles River (USA) were maintained under pathogen-free conditions (22 ± 2 °C and 60% air humidity,12 h light/dark cycle) according to the AAALAC (Association for Assessment and Accreditation of Laboratory Animal Care) guidelines and were allowed to acclimatize for at least 4 days before any experiments. All animal experiments were performed in compliance with the institutional guidelines and according to the protocol approved by the Committee for the Humane Use of Animals of State University of New York Upstate Medical University (IACUC # 437).

BALB/c mice were randomly divided into six groups (n=5–7). Group 1: Sham administered with a single intraperitoneal (i.p.) dose of PBS; Group 2,3: Sham associated with core or core-shell nanogel administered i.p. at the dose of 10 mg/kg; Group 4: LPS i.p. administered at the dose 0.1 mg/kg; Group 5, 6: LPS associated with nanogels (pre-incubated for overnight), with a single i.p. administration. Mice were sacrificed after 2 h. The blood was extracted, the peritoneal cavity was rinsed with 600 μL of PBS to collect peritoneal lavage. The cytokine level in plasma and peritoneal lavage were evaluated by ELISA.

The NF-κB reporter HLL (HIV-long terminal repeat/luciferase) mice colony were originally provided by Dr. Timothy Blackwell - Vanderbilt University Medical Center (also available in the Jackson laboratory # 027547)^[31] and maintained by our laboratory staff. Briefly, we maintained the colony by crossing hemizygous mice to their wildtype siblings, or to C57BL/6J. The offspring had both hemizygous mice and wildtype mice with a theoretical ratio of 1:1. Their genotypes were confirmed by the presence of NF-κB-induced bioluminescence by injection of luciferin (200µL/20g mice, 150mg/kg) when the mice were 3–4 weeks old at the steady state. 8–12-week-old HLL mice were used for in vivo bioluminescence studies. One day before the procedure, the abdominal area of the mice was shaved for optimal imaging acquisition. To obtain a basal bioluminescence level of each experimental mice, luciferin solution was i.p. injected, 10 minute later the abdominal bioluminescence was acquired using in vivo imaging system (IVIS) while the mice were under anesthesia through isoflurane inhalation. At the procedure day, LPS (Pseudomonas aeruginosa, dose of 0.1mg/kg) pre-incubated with core-shell nanogel (dose of 20mg/kg) or PBS was i.p. injected into mice. Controls were injected with PBS or nanogel in PBS solution. The abdominal bioluminescence signals were acquired using IVIS 50 (PerkinElmer) 2 hours, 6 hours and 24 hours post injection.

4.13. Cytokines removal from biological fluids

Raw 264.7 Cells were plated in 96-well plates at a density of 2×10⁴ cells per well. Stock solution of LPS was directly added to the cell culture to a final LPS concentration of 50 ng/mL. After 24h coincubation, the cell culture medium was harvested. Septic mice were induced by the cecum ligation and puncture (CLP) procedure as previously described.^[12] 24 h after CLP, the sepsis mice plasma and peritoneal lavage were collected for cytokines removal study. To determine nanogels binding with cytokines, including IL-6 and TNF-α, 100 μL of nanogel samples mixed with IL-6 or TNF-α were incubated at 37 °C for overnight. Following the incubation, cytokine concentrations in the supernatant were quantified by using ELISA. The cytokine removal capability was tested by adding core or core-shell nanogel with different concentration into LPS-challenged cell culture medium, sepsis mice plasma and peritoneal lavage, and sepsis human patient plasma, respectively. After overnight incubation at 37 °C, cytokine levels from cell culture medium, plasma and peritoneal lavage were measured by ELISA. All experiments were performed in triplicate. Clinical surgical sepsis patient blood was obtained in EDTA tube and de-identified for this study under a protocol approved by SUNY Upstate Institutional Review Board (IRB # 1321635–7]). Plasma was isolated for incubation with nanogel, and cytokine concentrations were analyzed by ELISA before and after incubation.

4.14. Nanogel prevents LPS induced inflammation in vivo

NF-κB reporter HLL mice were used for in vivo bioluminescence studies. 200µL (20mg/kg) core-shell nanogel in PBS solution was i.p. injected into mice, followed by a gentle rub around the shaved abdominal area. PBS with same volume was injected as controls. 5 minutes later, 50µL LPS (Pseudomonas aeruginosa, dose of 0.1mg/kg) in PBS solution was i.p. injected into the mice. The abdominal bioluminescence signals were acquired using IVIS at different time points (2, 6 and 24 h) post injection.

4.15. LPS attenuation of nanogel by endotoxin activity assay (EAA)

LPS (1 ng/mL) with or without the overnight preincubation with CS(Arg₄C17₂) (200 ng/mL) were added into 1mL of healthy human blood. LPS preincubated with PBS or core-shell without nanotrap (CS(COOH), 200 ng/mL) were added into the blood as controls. The endotoxin activity in each sample were evaluated using the EAA kit (Spectral Medical INC.).

4.16. Statistical analysis

All data points referred to the mean ± standard deviation (SD) and were based on at least three separate experiments (n=3). All statistical tests were performed by GraphPad Prism using one-way analyses of variance (ANOVAs) for two-group and multiple-group analyses. Statistical significance was represented as *: p<0.05; **: p<0.01; ***: p<0.001; ****: p< 0.0001.

Data availability

The authors declare that all data supporting this study are available in the paper or supplementary information. Additional relevant information is available from the authors upon reasonable request.