Dual-function oral nanotherapeutic mitigates sepsis-like multi-organ failure by targeting inflammatory and fibrotic pathways

Abstract

Sepsis, a life-threatening condition, often leads to multi-organ failure and has limited treatment options. We developed a novel dual-function nanoparticle, P2Ns-NAR(NAR), which utilizes naringenin (NAR) as both a targeting ligand for gut folate receptors and an encapsulated therapeutic agent to overcome its poor oral bioavailability. Here, we investigated the efficacy of this oral formulation in a mouse model of lipopolysaccharide-induced sepsis. We observed a significant reduction in the mRNA expression of pro-inflammatory (Tlr4, NF-κB, and IL-18), apoptotic (p53, and Fas), fibrosis (TGFβ1 and Smad3) and inflammasome-related (P2x7, gasdermin D, Nlrp3, Caspase 1, Nek7) markers. Histological analyses showed a prevention of tissue injury in the lungs, liver, kidney, heart, brain, intestines, and spleen. Additionally, Masson's trichrome staining revealed a remarkable reduction in collagen deposition, a hallmark of fibrosis, across multiple organs such as the lungs, liver, kidney, and heart. Our findings establish this dual-function nanoparticle platform as a highly effective oral therapy to prevent multi-organ failure.

Graphical abstract

Highlights

• •P2Ns-NAR(NAR) is a dual-function nanoparticle that uses naringenin (NAR) as a gut-targeting ligand and sepsis therapeutic.
• •The nanoparticle reduced gene expressions linked to inflammation, cell death, fibrosis, and inflammasomes.
• •It prevented tissue injury and decreased collagen deposition in lungs, liver, kidneys, and heart.
• •This nanoparticle platform is a highly effective oral therapy for preventing multi-organ failure in sepsis.

1. Introduction

Sepsis is a life-threatening condition caused by a dysregulated host response to infection, often leading to multi-organ failure with limited treatment options [1]. It is triggered by pathogens or bacterial toxins like lipopolysaccharide (LPS), a major component of the outer membrane of gram-negative bacteria such as Escherichia coli (E coli) [2,3]. The global burden of sepsis is rising, contributing to high rates of morbidity and mortality, especially in immuno-compromise individuals [4]. While inflammation is a vital defense mechanism against infection, a dysregulated immune response to LPS can lead to systemic inflammation and widespread damage to multiple organs, including the lungs, liver, kidneys, heart, brain, and intestines [[5], [6], [7]]. In addition, the impact of LPS-induced signaling extends beyond typical sepsis and immune dysregulation, playing a role in various diseases [8]. This is especially evident in recent reports demonstrating that low concentrations of LPS can perpetuate low-grade inflammation in several chronic conditions [9,10]. As LPS can initiate multiple metabolic pathways involving self-sustaining loops of ROS, NOS, and related metabolites, it creates persistent inflammatory damage, leading to a high demand for treatments that not only repair tissues but also restore their integrity. However, current therapies rely mainly on broad-spectrum antibiotics or steroids, which are not suitable for long-term use and may contribute to drug resistance and immunocompromise [11]. Therefore, researchers are increasingly turning toward natural compounds with established anti-inflammatory potential.

Naringenin (NAR), a flavonoid found in citrus fruits, has shown significant promise in combating inflammation-driven conditions by downregulating pro-inflammatory markers. However, its therapeutic potential is severely limited by poor oral bioavailability, primarily due to low solubility and permeability [12]. To overcome this, our laboratory engineered a novel oral drug delivery system, a dual-functional nanoparticle, P2Ns-NAR(NAR). This nanoparticle utilizes naringenin as both a targeting ligand for gut folate receptors (FR) and an encapsulated therapeutic agent, thereby enhancing its therapeutic efficacy and bioavailability [13], and also demonstrated that P2Ns-NAR(NAR) has remarkable efficacy in a cisplatin-induced acute kidney injury model [14]. Building on these findings, we now investigate the efficacy of this oral formulation in a mouse model of LPS-induced sepsis. Our goal is to determine if a single low dose of P2Ns-NAR(NAR) can provide superior therapeutic effects by simultaneously targeting the key inflammatory, apoptotic, and fibrotic pathways central to sepsis pathogenesis. Our findings show that this dual-function nanoparticle effectively inhibited pro-inflammatory, apoptotic markers, and reversed inflammation and tissue injury in lungs, liver, kidney, heart, brain, intestines, and spleen. This strategy offers a promising new approach to prevent multi-organ failure.

2. Results

2.1. Synthesis and formulation

Initially, ʟ-lactide was subjected to ring-opening polymerization with a tin catalyst, using dihydroxyl polyethylene glycol (PEG) as the initiator to synthesize a triblock copolymer, prepolymer (Fig. 1A and S1). The number-average molecular weight (M_n) of the prepolymer was calculated to be 2780 g/mol by proton nuclear magnetic resonance (¹H NMR) spectroscopy and 6430 g/mol by gel permeation chromatography (GPC) (Fig. S2 and S3). The prepolymer was subsequently combined with a dianhydride spacer, resulting in P2s that feature several carboxylic acid functionalities [13]. Titration experiment of P2s with methanolic KOH using phenolphthalein as an indicator indicated the presence of five carboxylic groups per mole of P2s. GPC analysis revealed that the molecular weight of P2s is estimated at M_n = 9860 g/mol, with a dispersity (Đ) of 1.49. NAR was then subjected to Steglich esterification with β-Boc alanine to yield a mono-substituted intermediate, which, upon Boc deprotection produced the primary amine derivative of NAR (Fig. S4a and S4b). This amine-functionalized NAR was subsequently conjugated to the carboxylic acid groups of P2s (Fig. 1A) [13]. The effective attachment of the NAR to P2s was validated by the widening of the aromatic proton peaks of NAR (Fig. S5). To verify NAR-conjugation efficiency, the integral values of the characteristic proton peaks corresponding to the PLA and PEG segments in P2s (Fig. S2b) were kept fixed and used as internal references for comparison with P2s–NAR. Subsequently, the characteristic aromatic proton peaks of NAR were integrated (Fig. S5). Quantitative analysis of the integration ratios between the aromatic protons of NAR and the PLA methyl (1.52–1.40 ppm) and methylene protons (5.24–5.16 ppm) demonstrated the incorporation of two NAR moieties per polymer chain. Further validation of the conjugation success was provided by GPC, which indicated an increase in M_n to 11,840 g/mol (Fig. S3).

Fig. 1.

Schematic representation of NAR conjugated polymer and characterization of resulting nanoparticles. (A) Synthetic scheme to conjugate NAR to P2s. (B) SEM micrographs of P2Ns(NAR) and P2Ns-NAR(NAR), Scale bar = 2 μm. (C) Tabulated particle sizes, zeta potentials, and entrapment efficiencies of P2Ns(NAR) and P2Ns-NAR(NAR) (n = 3). (D) Tabulated survival of mice 24 h post-LPS injection. (E-H) Plasma cytokines levels of (E) INF-γ and (F) MIP-1α; liver enzymes (G) ALP and (H) Amylase (n = 3 mice/group). (I-J) Tissue concentration levels of NAR in (I) Lungs, (J) Liver, and (K) Kidney (n = 3). Data presented as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001. One-way ANOVA followed by Tukey's multiple comparisons test and unpaired t-test. Comparisons between G2 and G5 are shown in Fig. 1. For Fig. 1(I–K), ⁺p < 0.05, ⁺⁺p < 0.01 One-way ANOVA followed by Tukey's multiple comparisons compared to G3. Additional statistical comparisons are summarized in Table S1.

The NAR-loaded formulations were prepared from unfunctionalized P2s and P2s-NAR using the oil-in-water (O/W) emulsification method. Scanning electron microscopy (SEM) revealed that the nanoparticles exhibited a spherical shape for both P2Ns(NAR) and P2Ns-NAR(NAR), with an average particle size of approximately 200 nm (Fig. 1B and S6), confirmed by DLS, suggesting that our formulations are suitable for oral administration [15,16]. Furthermore, each formulation exhibited a ζ-potential greater than ±20 mV (Fig. 1C), indicating that the formulations were in stable colloidal states [17,18]. The high-performance liquid chromatography (HPLC) determined that the drug entrapment efficiencies for P2Ns(NAR) and P2Ns-NAR(NAR) were approximately 74% and 93%, respectively (Fig. 1C). For oral drug delivery applications, nanoparticle stability within the gastrointestinal environment is critical to ensure effective therapeutic performance. Our laboratory and others previously demonstrated that polyester-based oral delivery systems possess strong mechanical stability and exhibit minimal cargo leakage during gastrointestinal transit [[19], [20], [21], [22]]. To assess the stability of as-synthesized nanoparticles under physiologically relevant conditions, the nanoparticles were incubated in simulated gastric fluid (SGF, pH 1.2) for 2 h. Subsequent analysis by DLS, HPLC, and SEM imaging confirmed that the particles maintained their structural integrity and cargo loading without significant degradation, with only 2% NAR leaching after 2h (Fig. S6 c-f).

2.2. The effect of P2Ns-NAR (NAR) on plasma cytokines and liver enzymes

A schematic diagram of the study plan is shown in Fig. S7. 24 h after LPS injection, we found that 50% and 25% of mice in LPS control (G2) and unformulated NAR (G3), respectively were dead, however, mice in normal control (G1), P2Ns(NAR) (G4), and P2Ns-NAR(NAR) (G5) achieved 100% survival (Fig. 1D). LPS administration successfully induced an inflammatory response in mice, as evidenced by a significant increase in plasma levels of interferon gamma (IFN-γ) and macrophage inflammatory protein-1 alpha (MIP-1α). Treatment with NAR mitigated this inflammatory response, with varying degrees of effectiveness depending on the specific formulation. Treatment with G5 proved to be the most effective treatment, significantly reducing IFN-γ levels (p < 0.0001) and MIP-1α levels (p < 0.01) compared to G2 (Fig. 1E and F). G3 was also effective, significantly reducing both IFN-γ (p < 0.0001) and MIP-1α (p < 0.05). G4 showed a partial effect, significantly reducing IFN-γ (p < 0.0001) but failing to produce a significant reduction in MIP-1α. A broader analysis confirmed this robust effect, showing that G5 significantly suppressed a wide panel of inflammatory markers, including IL-1β, IL-6, TNF-α, KC, MIG, RANTES, IP-10, and MCP-1 (Fig. S8). In addition, hepatic enzymes alkaline phosphatase (ALP) and amylase were increased by LPS, however, these enzymes were significantly reduced by treatment with G5 (Fig. 1G and H). The LC–MS quantification in lung tissue indicates higher bioavailability of NAR in G5 compared with G3 and G4 (Fig. 1I). Although these differences were not statistically significant, they suggest that comparable tissue exposure can be achieved even at a fourfold lower dose. In the liver, NAR levels were higher in G3 than in G4 and G5, indicating greater metabolic processing in G3, while no significant difference was observed between G4 and G5 (Fig. 1K). In the kidney, despite the fourfold higher dose in G3, NAR levels were comparable to those in G5 (Fig. 1J). Given the known kidney-protective effects of flavonoids, these findings suggest that similar therapeutic benefits can be achieved at lower doses using our formulation. Additional analysis of our data normality and robustness, revealing a high treatment-related effect of IFN-γ (η² = 0.93), MIP-1α (η² = 0.84), ALP (η² = 0.94), amylase (η² = 0.73), and other cytokines are shown in Fig. S9.

2.3. P2Ns-NAR(NAR) ameliorates LPS-induced inflammation and tissue damage in lungs

Herein, LPS injection induced inflammation by upregulation of mRNA expression of Tlr4, NF-κB, and IL-18 (Fig. 2A–C). Also, Nlrp3 inflammasome-related genes P2x7 and gasdermin D (Gsdmd), apoptosis markers p53 and Fas that amplify inflammation, and fibrosis marker TGF-β1 were significantly increased in G2 (Fig. 2D–H). However, G5 significantly reduced mRNA expression of Tlr4, NF-κB, IL-18, P2x7, Gsdmd, p53, Fas, and TGF-β1 compared to G2 (Fig. 2A–H). Further analysis assessing the robustness of our data revealed large treatment-related effects for NF-κB (η² = 0.91), IL-18 (η² = 0.80), p53 (η² = 0.89), Fas (η² = 0.86) and TGF-β1 (η² = 0.96) as shown in Fig. S10. Also, histological analysis using H&E staining revealed that LPS injection led to immune cell infiltration and increased thickness of the alveolar and alveolar capillary walls (Fig. 2I). Furthermore, Masson's trichrome staining showed a marked increase in collagen fibers in the lung interstitial spaces in G2 (Fig. 2J) while immunofluorescence (IF) staining showed remarkable increase in phosphorylated NF-κB (NF-κB p65) (2K); however, G5 significantly reduced collagen production and deposition by 2-fold (Fig. 2L) and NF-κB p65 by 3-fold (2M).

Fig. 2.

P2Ns-NAR(NAR) ameliorates LPS-induced inflammation and tissue damage in lungs.

mRNA expression of (A) Tlr4, (B) NF-κB, (C) IL-18, (D) P2x7, (E) Gsdmd, (F) p53, (G) Fas, and (H) TGF-β1. n = 3 with 12 technical replicates across all groups. (I) Representative sections of H&E staining show thickened alveolar wall thickness (black arrow) and alveolar capillary wall (yellow arrow). Scale bar = 200 μm. n = 4 mice/group. (J) Masson's trichrome staining illustrating inflammation and collagen deposition (blue-stained areas) in lungs interstitial space of LPS-induced sepsis mice. Scale bar = 200 μm. n = 4 mice/group. (K) Immunofluorescence staining of NF-κB p65. Scale bar = 500 μm, n = 4 mice/group. (L) Quantification of Masson's trichrome staining in all groups. n = 8-12 images/group. (M) Quantification of mean fluorescence intensity of NF-κB p65. n = 4 mice, 8 images/group. Data presented as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 by One-way ANOVA followed by Tukey's multiple comparisons test and unpaired t-test. Comparisons between G2 and G5 are shown in Fig. 2. Additional statistical comparisons are shown in Table S2.

2.4. P2Ns-NAR(NAR) suppresses the formation of Nlrp3 inflammasome in the lungs

Assessing both gene and protein expression levels of key components of the Nlrp3 inflammasome, particularly Nlrp3, Caspase 1 (Casp1), and Nek7, is of high importance in understanding the mechanistic processes underlying the innate immune system's response to pathogens. As indicated in Fig. 3A–C, mRNA expression levels of Nlrp3, Casp1, and Nek7 were markedly upregulated in the diseased group without treatment. G5 demonstrated a significant reduction in mRNA expression of Nlrp3 by 3-fold and both Casp1 and Nek7 by 2-fold (Fig. 3A–C). Further comparison between the groups revealed that the reduction in mRNA expression of Nlrp3 was significantly lower in G5 than in G3 (p < 0.05). Moreover, a broader analysis showed a large effect-size of the mRNA expression data for Nlrp3 (η² = 0.93), Casp1 (η² = 0.79), and Nek7 (η² = 0.85) (Fig. S11). Immunofluorescence staining of lungs showed strong fluorescence of antibodies against Nlrp3, Casp1 p20 (activated form of Casp1), and Nek7 in G2 (Fig. 3D–F). Quantification of the fluorescence intensity, which depicts protein levels, showed that mice responded to the treatment with a significant reduction in the signal intensity of Nlrp3 and Casp1 p20 by 3-fold, and Nek7 by 7-fold compared to G2 (Fig. 3G–I). Moreover, the signal intensity of Nlrp3 and Casp1 p20 was significantly lower in G5 than in G3 (p < 0.001), as well as in G4 (p < 0.01). Additional analysis of fluorescence intensity showing 4′,6-diamidino-2-phenylindole (DAPI), each Nlrp3, Casp1 p20, and Nek7 antibody, and their corresponding merged fluorescence intensities can be found in Fig. S12–S14.

Fig. 3.

P2Ns-NAR(NAR) inhibits the activation of Nlrp3 inflammasome in the lungs. (A-C) mRNA expression of (A) Nlrp3, (B) Casp1, and (C) Nek7. n = 3 with 12 technical replicates in all groups. (D-F) Immunofluorescence staining of (D) Nlrp3, (E) Casp1 p20, and (F) Nek7 showing alveolar epithelial cells (insert). Scale bar = 200 μm. n = 4 mice/group. (G-I) Quantification of fluorescence intensity of (G) Nlrp3, (H) Casp1 p20, and (I) Nek7. n = 8-12 images/group. Data presented as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 by One-way ANOVA followed by Tukey's multiple comparisons test and unpaired t-test. Comparisons between G2 and G5 are shown in Fig. 3. Additional statistical comparisons are summarized in Table S3.

2.5. P2Ns-NAR (NAR) mitigates LPS-induced inflammation and tissue damage in liver

Analyses of liver markers show that mice stimulated with LPS exhibited upregulation of mRNA expression levels of inflammatory, apoptotic markers, fibrosis markers, as well as components of Nlrp3 inflammasome, which were significantly reduced by our treatment regimen. Specifically, G5 significantly reduced mRNA expression of Tlr4, NF-κB, Nek7, P2x7, Gsdmd, p53, Fas, TGF-β1, and Smad3 (Fig. 4A–I) compared to G2. Importantly, G5 significantly reduced NF-κB expression (p < 0.05) compared to G4. Histological analyses showed that IP injection of LPS elicited immune cell infiltration and scarring in the periportal region of the liver (Fig. 4J) accompanied with increased deposition of collagen fibers (Fig. 4K) in G2. These trends were markedly reversed in G5. Additional analysis on the mRNA expression revealed that NF-κB (η² = 0.95), Nek7 (η² = 0.75), P2x7 (η² = 0.73), Gsdmd (η² = 0.84), p53 (η² = 0.84), TGF-β1(η² = 0.92), and Smad3 (η² = 0.73) exhibited large treatment-related effects while Tlr4 (η² = 0.65) and Fas Tlr4 (η² = 0.67) showed moderate size effects (Fig. S15). Quantification of blue-stained collagen fibers confirmed that LPS-induced fibrosis was significantly reduced following treatment with G5 (p < 0.0001) compared to G2 (Fig. 4L), and by p < 0.0001 and p < 0.001 compared to G3 and G4, respectively.

Fig. 4.

P2Ns-NAR(NAR) mitigates LPS-induced inflammation, tissue damage, and fibrosis in liver. (A-I) mRNA expression of (A) Tlr4, (B) NF-κB, (C) Nek7 (D) P2x7, (E) Gsdmd, (F) p53, (G) Fas, and (H) TGF-β1, and (I) Smad3. n = 3 with 12 technical replicates across the groups. (J) Representative section of H&E staining showing cells infiltration (arrow) and infiltration and scarring (insert). Scale bar = 100 μm. (K) Representative section Masson's trichrome staining illustrating collagen deposition in the liver interstitium of LPS-induced sepsis mice. Scale bar = 100 μm. n = 4 mice/group. (L) Quantification of Masson's trichrome staining in all groups. n = 8-12 images/group. Data presented as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 by One-way ANOVA followed by Tukey's multiple comparisons test and unpaired t-test. Comparisons between G2 and G5 are shown in Fig. 4. Additional statistical comparisons are summarized in Table S4.

2.6. P2Ns-NAR(NAR) decreases LPS-induced inflammation and tissue damage in kidney

In the kidney, mRNA expression of inflammatory markers, apoptotic markers, and histological analyses were carried out. Our results show that mRNA levels of pro-inflammatory markers (Tlr4 and NF-κB) and components of Nlrp3 inflammasome (Nlrp3 and Nek7) were upregulated in G2. Similarly, mRNA expression of P2x7 receptor and p53 were upregulated in G2. However, G5 significantly reduced mRNA expression levels of Tlr4, NF-κB, Nlrp3, Nek7, and p53 compared to G2 (Fig. 5A–E). Additional analyses of data normality showing treatment-related effects and biological relevance of Tlr4 (η² = 0.80), NF-κB (η² = 0.93), Nlrp3 (η² = 0.71), Nek7 (η² = 0.91), and p53 (η² = 0.81) are shown in Fig. S16. Histological analyses using H&E staining of the kidney show infiltration of immune cells, renal tubular vacuolation, and sloughing of tubular epithelial cells in proximal tubules (Fig. 5F). Furthermore, Masson's trichrome staining showed the production and deposition collagen fibers within the renal interstitium was markedly high in G2 (Fig. 5G). Quantification of areas of fibrosis showed that collagen fiber deposition observed in G2 was reduced by G5 by 2-fold (Fig. 5H). Notably, G5 exhibited a significant anti-fibrotic effect by reducing collagen aggregation by p < 0.0001 and p < 0.05 compared with G3 and G4, respectively.

Fig. 5.

P2Ns-NAR(NAR) decreases LPS-induced inflammation and tissue damage in the kidney. (A-E) mRNA expression of (A) Tlr4, (B) NF-κB, (C), Nlrp3, (D) Nek7, and (E) p53. n = 3 with 12 technical replicates across the groups. (F) Representative section of H&E staining showing immune cell infiltration (black arrow, insert) and tubular cells detaching by forming vacuoles (yellow arrow) in the kidney tubulointerstitial area. Scale bar = 50 μm. n = 4 mice/group. (G) Representative section of Masson trichrome staining showing collagen deposition. Scale bar = 50 μm. n = 4 mice/group (H) Quantification of Masson's trichrome staining in all groups. n = 8-12 images/group. Data presented as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 by One-way ANOVA followed by Tukey's multiple comparisons test and unpaired t-test. Comparisons between G2 and G5 are shown in Fig. 5. Additional statistical comparisons across the groups are shown in Table S5.

2.7. P2Ns-NAR(NAR) mitigates LPS-induced inflammation and tissue damage in the heart

As depicted in Fig. 6, mRNA expression levels of Tlr4, NF-κB, Nek7, P2x7, and p53 were upregulated in G2. Notably, treatment with G5 significantly reduced the expression of these genes compared to G2 (Fig. 6A–E). Furthermore, an expanded analysis of the data showed strong effect size outcomes in Tlr4 (η² = 0.88), NF-κB (η² = 0.86), Nek7 (η² = 0.90), P2x7 (η² = 0.93), and p53 (η² = 0.83) as exhibited in Fig. S17. Histological analyses of H&E staining showed an increased number of noticeable nuclear enlargements of cardiomyocytes in G2, which was reduced by G5 (Fig. 6F). In addition, Masson's trichrome staining revealed massive collagen fiber deposition (Fig. 6G), and quantification analysis showed that G5 significantly reduced collagen production and deposition observed in G2 (Fig. 6F).

Fig. 6.

P2Ns-NAR(NAR) mitigates LPS-induced inflammation and tissue damage in heart. (A-E) mRNA expression of (A) Tlr4, (B) NF-κB, (C) Nek7 (D) P2x7, and (E) p53. n = 3 with 12 technical replicates in all groups. (F) Representative histological section of H&E staining showing notable enlarged nuclei of cardiomyocytes (arrow). Scale bar = 20 μm. n = 4 mice/group. G) Representative histological section of Masson's trichrome staining showing collagen deposition in the heart. Scale bar = 50 μm. n = 4 mice/group. (H) Quantification of Masson's trichrome staining. n = 8-12 images/group. Data presented as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 by One-way ANOVA followed by Tukey's multiple comparisons test and unpaired t-test. Comparisons between G2 and G5 are shown in Fig. 6. Additional statistical comparisons are shown in Table S6.

2.8. P2Ns-NAR(NAR) alleviates tissue damage induced by LPS in brain and intestines

In addition to the lungs, liver, kidney, and heart, H&E staining was also performed for the brain and intestines to assess the systemic consequences of the induced inflammatory response, as these are critical organs susceptible to inflammation-induced tissue damage and neuronal or intestinal barrier dysfunction. In the brain cortex, and hippocampal CA3 and hilus regions, no damaged cells were observed in G1. However, in G2, noticeable brain injuries were observed, with normal spherical neurons adopting an amoeboid shape. This alteration was less pronounced in G3 and G4. Importantly, G5 markedly improved neuronal morphology, with spherical cells and distinct nuclei evident in the cortex, as well as in the hippocampal CA3 and hilus regions (Fig. 7A–C).

Fig. 7.

P2Ns-NAR(NAR) alleviates tissue damage induced by LPS in brain and intestines. (A-C) Representative H&E staining sections showing (A) brain cortex, hippocampal (B) CA3 and (C) hilus regions. G1: Normal spherical brain cells with a visibly clear nucleus; G2: more damaged brain cells taking amoeboid shapes (arrow); G3 & G4: a reduced number of damaged cells; G5: little to no cells damaged. Scale bars = 50 μm. n = 4 mice/group. (D) Representative H&E staining sections of ileum showing villi (black arrow) and crypt depth (yellow arrow). Scale bars = 100 μm. n = 4 mice/group. (E) Representative H&E staining sections of colon showing muscularis mucosa (arrow). Scale bars = 100 μm. n = 4 mice/group. (F-G) Measurement of villus height (F) and crypt depth (G) in all groups. Scale bars = 50 μm (brain) and 100 μm (intestines). n = 6-8 images/group. Data presented as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 by One-way ANOVA followed by Tukey's multiple comparisons test and unpaired t-test. Comparisons between G2 and G5 are shown in Fig. 7. Additional statistical comparisons are summarized in Table S7.

In the small intestine (ileum), there were no observable changes in shape, villi length, and crypt depth in G1. However, in G2 we observed disturbed arrangement and disruption of villi integrity as exhibited by irregular shape, decreased villi height, and increased crypt depth (Fig. 7D), which were confirmed by corresponding measures of villi height and crypt depth (Fig. 7F and G). Villi heights showed significant improvement in G5 compared with G2 (Fig. 7D and F), and crypt depth showed a significant decrease by G5 compared to G2 (Fig. 7D and G). Further analysis of the data showed effect size of η² = 0.71 and η² = 0.35 for villi height and crypt depth, respectively (Fig. S18). Analysis of the colon further supported these findings. The G1 had an intact muscularis mucosa, whereas G2 showed extensive disruption (Fig. 7E). This damage was improved in G3, G4, and G5, with G5 demonstrating the most pronounced restoration (Fig. 7E).

2.9. P2Ns-NAR(NAR) alleviates LPS-induced inflammation and histological damage in spleen

To evaluate its central role in immune response to LPS-induced sepsis, spleen weight, mRNA expression, and histological staining were carried out. As indicated in Fig. 8, weight, mRNA expression levels of Nlrp3, IL-18, and P2x7 were increased in G2. However, G5 treatment significantly decreased weight and gene expressions compared to G2 (Fig. 8A–D). Further, an expanded analysis of the data revealed strong effect size outcomes in weight (η2 = 0.80), Nlrp3 (η2 = 0.81), IL-18 (η2 = 0.72), and P2x7 (η2 = 0.68), as shown in Fig. S19. Furthermore, histological analysis of H&E staining showed notable enlargement of white pulp area relative to red pulp area in G2, which was remarkedly reduced by G5 (Fig. 8E). In addition, quantification of the white pulp area showed a significant reduction in G5 (Fig. 8F).

Fig. 8.

P2Ns-NAR(NAR) alleviates LPS-induced inflammation and histological damage in spleen. (A-E) Spleen weight (A), mRNA expression of (B) Nlrp3, (C) IL-18, (D) P2x7. n = 3 with 12 technical replicates in all groups. (E) Representative histological section of H&E staining showing increased white pulp area (arrow). Scale bar = 200 μm. n = 4 mice/group. (F) Quantification of white pulp area. n = 8 images/group). Data presented as mean ± SEM. ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and ∗∗∗∗p < 0.0001 by One-way ANOVA followed by Tukey's multiple comparisons test and unpaired t-test. Comparisons between G2 and G5 are shown in Fig. 8. Additional statistical comparisons are shown in Table S8.

3. Discussion

Our study demonstrates that P2Ns-NAR(NAR) significantly reduced inflammation and protected against multiorgan injury in a model of sepsis. These findings align with the modern understanding of sepsis, which has shifted from a purely inflammatory disorder to one of host-immune dysregulation [23].

Our research provides compelling evidence that P2Ns-NAR(NAR) represents a significant step toward this goal. We demonstrated its therapeutic efficacy at a dose of just 10 mg/kg, which is a remarkable 90% dose reduction compared to the 100 mg/kg dose of free NAR often required in other studies [24]. Furthermore, this dose was 75% and 50% lower than plain NAR (40 mg/kg) and P2Ns(NAR) (20 mg/kg), respectively, in our own experiments. This remarkable efficiency addresses the low water solubility and poor bioavailability that limit NAR's clinical use. By improving the therapeutic index, our nanoparticle system offers a path to greater patient safety and more effective treatment. A key innovation of our system is the dual role of NAR, which acts both as a therapeutic agent and as a FR-targeting ligand on the nanoparticle surface for enhanced delivery.

The success of P2Ns-NAR(NAR) is rooted in its ability to modulate the overactive immune response at a fundamental level. Our findings confirm that it effectively suppresses the Tlr4/NF-κB signaling pathway, which initiates the inflammatory cascade [[25], [26], [27]]. More critically, we demonstrated that our formulation significantly inhibited the activation of the Nlrp3 inflammasome, a key driver of organ damage and the hallmark of severe sepsis [28]. We found that our formulation suppressed key steps in inflammasome activation, including Nek7 binding and the priming of P2x7 receptors [29]. In our study, LPS injection significantly increased the expression of P2x7, Nek7, and Casp1 in multiple organs, findings that are consistent with the activation and colocalization of key Nlrp3 inflammasome components [30]. Importantly, P2Ns-NAR(NAR) treatment markedly reduced the expression of these markers and reversed the activation of the inflammasome, as confirmed by both mRNA and immunofluorescence analyses.

Consistent with this suppression, we observed a significant reduction in inflammatory, apoptotic, and fibrotic markers such as IL-18, Gsdmd, p53, Fas, TGF-β1, and Smad3 following P2Ns-NAR(NAR) treatment. Thus, our treatment not only downregulates inflammation but also inhibits apoptotic and fibrotic pathways that plague multiple organs in sepsis [[31], [32], [33]]. This provides a clear mechanistic link between our formulation and its therapeutic effect, suggesting it can prevent the widespread tissue injury that leads to high mortality and the long-term disabilities of post-sepsis syndrome. Our results show that P2Ns-NAR(NAR) can remarkably ameliorate alveolar-capillary thickening, hepatic necrosis, and other organ injuries by inhibiting the expression of key components of the Nlrp3 inflammasome [34,35].

The therapeutic potential of P2Ns-NAR(NAR) extends beyond sepsis. The inflammatory pathways we successfully targeted are implicated in a range of acute inflammatory syndromes. For example, conditions like phosgene-induced acute lung injury (P-ALI) share key pathogenic features with our sepsis model, including Nlrp3 inflammasome activation [[36], [37], [38]]. The ability of P2Ns-NAR(NAR) to ameliorate fibrosis and suppress the Nlrp3 inflammasome suggests it could be a promising treatment for other Nlrp3-mediated inflammatory conditions, representing a broader application of our technology.

In conclusion, our study successfully establishes the P2Ns-NAR(NAR) nanoparticle system as a highly effective oral therapy for preventing multi-organ failure in sepsis. P2Ns-NAR(NAR) overcomes the limitations of traditional treatments by delivering a powerful, low-dose treatment for sepsis, requiring 75% less NAR than a standard oral dose, 40 mg/kg used in this study. This research not only confirms the efficacy of immunomodulation as a viable therapeutic strategy but also offers a promising drug delivery platform with the potential to significantly reduce mortality and morbidity in sepsis and other chronic inflammatory diseases.

4. Methods

4.1. Synthetic procedures

Synthesis of prepolymer (Triblock copolymer): A triblock copolymer was synthesized by combining dried PEG₄₀₀ (3.87 g, 9.67 mmol), ʟ-lactide (25.1 g, 174.14 mmol), and stannous octoate (29 mg, 0.1% w/w) in a 250 mL three-neck round-bottom flask. Toluene (100 mL) was added as the solvent. The reaction was conducted under a nitrogen atmosphere. The mixture was heated to 115 °C and stirred for 24 h. After the reaction, the solvent was removed under reduced pressure. The remaining solid was dissolved in 30 mL of dichloromethane (DCM) and precipitated in 300 mL of cold diethyl ether. The polymer was isolated by centrifugation. This reprecipitation process was repeated two more times to ensure high purity, resulting in 19.7 g of white polymer (Fig. S1). Theoretical: M_n = 3000 g/mol; ¹H NMR: M_n = 2780 g/mol; GPC: M_n = 6430 g/mol, Ɖ = 1.07 (Fig. S2 and S3), respectively.

Synthesis of P2s: To synthesize the polymer (P2s), a mixture of prepolymer (19.0 g, 2.95 mmol), triethylamine (598 mg, 819 μL, 5.90 mmol), cyclohexanetetra-carboxylic dianhydride (HCDA) (662.4 mg, 2.95 mmol), and stannous octoate (20.3 mg, 0.1% w/w) it was added to a flask containing toluene as the solvent. The reaction vessel was kept under continuous nitrogen atmosphere and heated to 115 °C for 20 h. After the reaction, the solvent was removed under reduced pressure. The resulting material was purified by dissolving in 30 mL of DCM and precipitating in 300 mL of cold diethyl ether. The polymer was isolated by centrifugation, and this purification step was repeated two additional times to ensure purity. This process yielded 16.5 g of a white polymer (Fig. S2). GPC: M_n = 9860 g/mol, Ɖ = 1.49 (Fig. S3).

Synthesis of NAR-Ala-Boc: To synthesize NAR-Ala-Boc, β-Boc-alanine (500 mg, 2.64 mmol) was first dissolved in 7 mL of anhydrous N, N-dimethylformamide (DMF). The solution was then chilled to 0 °C under a nitrogen atmosphere. Next, the coupling agents, dicyclohexyl carbodiimide (DCC) (654 mg, 3.17 mmol) and 4-dimethylaminopyridine (DMAP) (194 mg, 1.58 mmol) were introduced, and the mixture was stirred for 30 min. After this, naringenin (863 mg, 3.17 mmol) was added, and the reaction mixture was left to stir overnight for 18 h at room temperature. After the reaction was complete, the product was isolated by precipitating the mixture with 50 mL of ice-cold water. The solid precipitate was collected by filtration, dried, and further purified using column chromatography with a chloroform/methanol (90:10 v/v) mixture as the mobile phase. The final product, a white powder weighing 620 mg, was obtained after drying the organic phase over anhydrous Na₂SO₄ and removing the solvent under reduced pressure (Fig. S4a). ¹H NMR (500 MHz, DMSO-d₆) δ: 12.1 (s, 1H, HO-C=C-C=O), 10.8 (s, 1H, HO-C=C-OH), 7.6 (d, 2H, -C-C=CH-, -C-C-CH=C), 7.2 (d, 2H, -O-C=CH-, -O-C-CH = ), 7.0 (t, 1H, -NH-), 5.9 (dd, 2H, HO-C-CH = , HO-C=CH-), 5.6 (dd, 1H, -O-CH-), 3.4-3.2 (m, 2H, -NH-CH₂-), 3.4-3.2 (m, 1H, -C-CH_a-C=O), 2.8-2.7 (dd, 1H, -C-CH_b-C=O), 2.8-2.7 (t, 2H, -NH-CH₂-CH₂-), 1.4 (s, 9H, -C-(CH₃)₃) (Fig. S4b).

Synthesis of NAR-Ala-NH₂: The synthesis of NAR-Ala-NH₂ began by dissolving NAR-Ala-Boc (500 mg, 1.12 mmol) in a 10 mL solvent mixture composed of DCM (7 mL) and methanol (3 mL). The solution was cooled to 0 °C and purged with nitrogen. Next, trifluoroacetic acid (TFA, 3.5 mL; ∼30% v/v) was added, and the reaction was stirred for 2 h under an inert environment. The solvents were then removed under reduced pressure, leaving a crude oil. To purify the product, the crude oil was treated with ice-cold water, which caused the product to precipitate as an off-white solid. The solid was collected by centrifugation, dried, and the final yield was 370 mg. ¹H NMR (500 MHz, DMSO-d₆) δ: 12.1 (s, 1H, HO-C=C-C=O), 10.9 (bs, 1H, HO-C=C-OH), 7.9 (s, 2H, -NH₂), 7.6 (d, 2H, -C-C=CH-, -C-C-CH = ), 7.2 (d, 2H, -O-C=CH-, -O-C-CH=C), 5.9 (dd, 2H, HO-C-CH = , OH-C=CH-), 5.6 (dd, 1H, -O-CH-), 3.5-3.3 (m, 1H, -C-CHa-C=O), 3.20-3.1 (m, 2H, NH₂-CH₂-), 3.0 (t, 2H, NH₂-CH₂-CH₂-), 2.8 (dd, 1H, -C-CH_b-C=O)) (Fig. S4b).

Synthesis of P2s-NAR: To synthesize P2s-NAR, P2s (5.00 g, 0.507 mmol) and 1-ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) (632 mg, 3.29 mmol) were dissolved in a mixture of 20 mL anhydrous DMF and 10 mL anhydrous DCM. The solution was cooled to 0 °C and stirred under a nitrogen atmosphere for 30 min. Afterward, a separate solution of NAR-Ala-NH₂ (1.13 g, 3.29 mmol) and N,N-Diisopropylethylamine (DIEA) (575 μL, 3.296 mmol) in 5 mL of anhydrous DMF was added dropwise to the reaction mixture. The combined solution was then allowed to stir overnight for 18 h at room temperature. Following this, DCM was removed under reduced pressure. The polymer was precipitated by adding the reaction mixture to cold water and isolated by centrifugation. To remove unreacted NAR-Ala-NH₂, the polymer was washed three times by dissolving it in a 1:1 mixture of DCM and methanol, followed by reprecipitation in cold diethyl ether, yielding an off-white polymer (4.5 g) (Fig. S5). GPC: M_n = 11840 g/mol, Ɖ = 1.58 (Fig. S3).

Formulation of NAR-laden nanoparticles: NAR-laden P2Ns and P2Ns-NAR nanoparticles were prepared via a single oil-in-water emulsion method. For the preparation of P2Ns-NAR: briefly, 150 mg P2s-NAR and 45 mg corresponding P2s were dissolved in 8 mL ethyl acetate. Separately, 15 mg of NAR was dissolved in 1.5 mL of ethyl acetate and 0.5 mL of methanol. After 1 h of stirring, NAR solution was added to P2s-NAR solution, and the mixture was divided into three equal parts. Simultaneously, 150 mg of polyvinyl alcohol (PVA) (1% w/v) was dissolved in 15 mL of water under stirring at 1000 rpm for 1 h and then divided into three separate 5 mL portions. The polymeric solutions were added to the PVA solution and stirred for 7 min, followed by homogenization for 7 min at 15,000 rpm. The resulting three nano-emulsion were then slowly added to 60 mL of water and stirred for 4 h to facilitate the diffusion and evaporation of the organic solvent. The nanoparticles were isolated by centrifugation at 15,000 g for 30 min at 4 °C. The supernatant was discarded, and the resulting pellet was redispersed in 12 mL of a 6% w/v sucrose solution. The suspended nanoparticles were aliquoted into vials, frozen at −80 °C for 30 min, and lyophilized under vacuum at 0.02 mbar for 48 h at −30 °C, followed by a secondary drying cycle at 0.02 mbar for 24 h at 20 °C. The prepared formulations were characterized using DLS and FE-SEM. The entrapment efficiency of NAR in the formulations was determined by HPLC using the following equation:

$$\text{Entrapment}\text{efficiency}(\%)=\frac{AmountofNARinnanoparticles}{AmountofNARinthefeed}\times 100\%$$

Animal study and treatment plan. 9-week-old forty male C57BL/6J mice (25 ± 1.4 g, source: The Jackson Laboratory) were divided into five groups for this study, n = 8. All animal procedures were approved by The University of Alabama Institutional Animal Care and Use Committee (IACUC number: 23-01-6233). All mice had free access to food and water ad libitum. Sepsis was induced by intraperitoneal injection (IP) of LPS (15 mg/kg) from Escherichia coli O111:B4 (source: Sigma-Aldrich, CAS number: 93572-42-0), dissolved in sterile 0.9% normal saline (Medline Industries). Group (G) 1- normal control, while LPS-induced sepsis mice were divided into 4 groups: G2-untreated, G3- Naringenin (NAR 40 mg/kg), G4- P2Ns (NAR) (20 mg/kg), and G5- P2Ns-NAR(NAR) (10 mg/kg). Treatments were administered by oral gavage (single dose) 30 min after LPS injection, and all mice were humanely sacrificed 24 h post-LPS. The study plan is briefly described in Fig. S7. Immediately following sacrifice, blood was collected by cardiac puncture into heparinized tubes and centrifuged at 3000 g and 4 °C for 30 min to separate plasma. Lungs, liver, kidney, heart, intestine, spleen, and brain were collected, snap-frozen, and stored at - 80 °C and/or fixed in formalin until further analysis.

Bioplex cytokines: A bioplex mouse cytokine/chemokine panel I (catalog number: MCYTOMAG-70K) was used to quantify plasma biomarkers on Bio- Plex 200 system according to manufacturer's instructions.

Liver enzyme assay: Tissue extraction reagent II (FNN0081, ThermoFisher Scientific) was supplemented with protease inhibitor at 1 μl inhibitor to 100 μl of the reagent. 10 μl of the cocktail per 1 mg of liver tissue was homogenized and centrifuged at 10000 rpm for 30 min at 4 °C. The supernatant was harvested and assayed for alkaline phosphatase and amylase using an automated chemistry analyzer (Vet Axcel, Alfa Wassermann).

Real-time polymerase chain reaction (PCR): 30 mg of each tissue was homogenized in Qiazol reagent (catalog number: 79306, Qiagen) followed by RNA isolation using RNeasy plus universal mini kit (catalog number: 73404, Qiagen). Take3 (BioTek) was used to quantify RNA at 280/260 nm and measurements were obtained with BioTek's citation 5. Subsequently, cDNA was prepared from the quantified RNA samples using BIO-RAD iScript Advace cDNA kit for RT-qPCR (catalog number: #1725038) and gene amplification was carried out using BIO-RAD PowerUp™ SYBR™ Green Master Mix in a CFX Opus 384 RT-PCR system. β-actin was used as housekeeping gene for normalizing ΔCt values, where Ct is the threshold cycle and fold change between groups was expressed as 2^−ΔΔCt. Although 4 mice/group with 16 technical replicates were evaluated for mRNA expressions, 3 mice/group with 12 technical replicates were used for statistical comparisons across all the groups. Primers sequences are listed in Table S8.

Histology: Tissue samples from lungs, liver, kidney, heart, intestine, and brain were initially fixed in 10% buffer-neutralized formalin at room temperature. These samples then underwent processing and 5 μm thick sections were taken from paraffin-embedded blocks. A customized program was used, involving heating at 60 °C for 1.5 h, followed by deparaffinization in xylene and rehydration through a graded series of ethanol (100%, 95%, 70%) and three washes in PBS. Subsequent staining procedures were performed using a Leica Histocore Spectra ST stainer. Hematoxylin and Eosin (H&E) staining was applied to all six tissue types: lungs, liver, kidney, heart, intestines, and brain. Spleen tissues were fixed in 4% paraformaldehyde (PFA) for 24 h, washed in PBS and embedded in OCT medium from Leica Biosystems (FSC 22 Blue, Cat#: NC0278476) and snap-frozen in liquid nitrogen. 5 μm thick sections were taken using Cryostat microtome (Leica CM 1860 UV) at −20 °C and stained with H&E after been dehydrated in 100%, 95%, and 70% ethanol. In addition, to assess collagen deposition, Masson's Trichrome (MT) staining was conducted on lungs, liver, kidney, and heart.

Immunofluorescence: Lung sections were incubated in antigen retrieval (citrate buffer, pH = 6) for 45 min and washed 3 times with PBS. The sections were then blocked with 1.5% horse serum for 2 h and washed with PBS. For phosphorylated NF-κB (p65), sections were incubated in Tris HCL buffer (pH 10) antigen retrieval in microwave for 4 min, washed 3 times with PBS, blocked with UltraCruz blocking reagent (Cat#: sc-516214) for 30 min, and then washed with PBS. Thereafter, sections were stained with a recombinant rabbit monoclonal antibody against Nlrp3 (invitrogen, Cat#: MA5-32255, RRID: AB_2809541), Casp1 p20 (cleaved Asp296) polyclonal antibody (invitrogen, Cat#: PA5-99390, RRID: AB_2818323), and Nek7 polyclonal antibody (invitrogen, Cat#: PA5-101861, RRID: AB_2851293) at 1:100 dilution in 1.5% horse serum in PBS, while NF-κB p65 was incubated with mouse monoclonal antibody (SantaCruz, Cat#: A-12 sc: 514451, RRID: AB_2891257) at 1:25 dilution overnight at 4 °C followed by secondary antibody [goat anti-rabbit IgG Alexa Flour-488 (Nlrp3 and Nek7) and 594 (Casp1 and NF-κB p65)] at 1:1000 dilution in PBS for 2 h. Sections were then washed 3 times with PBS and counterstained with DAPI (1:1000 dilution) for 5 min and subsequently washed with PBS one time before sections were cover-slipped with mounting media containing DAPI (Vectorshield, H-1800, Vector laboratories).

After staining, all slides were visualized/scanned under LSM-900 Zeiss/Axioscan 7 confocal microscope and images were processed with Zen blue 3.3 software and quantified with ImageJ software.

Statistical analysis. One-way ANOVA and Tukey's posthoc tests performed to compare groups, considering P < 0.05 as significant. Further, unpaired t-test was carried out for comparison between two groups. Statistical analysis was performed using GraphPad prism (9.3.1) and all data were presented as mean ± SEM.

CRediT authorship contribution statement

James Asenso: Writing – original draft, Methodology, Investigation, Formal analysis. Neha Choudhury: Writing – original draft, Methodology, Investigation, Formal analysis. Raghu Ganugula: Writing – review & editing, Validation, Supervision, Methodology, Investigation, Funding acquisition, Formal analysis. Tyler Posey: Writing – review & editing, Investigation. Meenakshi Arora: Writing – review & editing, Supervision, Project administration, Methodology, Investigation, Funding acquisition. M.N.V. Ravi Kumar: Writing – review & editing, Supervision, Project administration, Methodology, Funding acquisition, Conceptualization.

Ethics approval and consent to participate

All animal procedures were approved by The University of Alabama Institutional Animal Care and Use Committee (IACUC number: 23-01-6233).

Declaration of competing interest

The authors declare no competing financial interests or personal relationships that could have appeared to influence the work reported in this article.

Abbreviations used:

Casp1

Caspase 1

GPC

Gel permeation chromatography

Gsdmd

Gasdermin D

HPLC

High-performance liquid chromatography

IFN-γ

Interferon gamma

IL

Interleukin

IP-10

Interferon gamma-induced protein 10

LPS

Lipopolysaccharide

MCP-1

Monocyte Chemoattractant Protein-1

MIG

Monokine induced by gamma interferon

MIP-1α

Macrophage inflammatory protein-1 alpha

Nek7

NIMA related kinase 7

Nlrp3

NOD-, LRR- and pyrin domain-containing protein 3

PEG

Polyethylene glycol

RANTES

Regulated upon activation, normal T cell expressed and secreted

Tlr4

Toll-like receptor 4

TNF-α

Tumor necrosis factor alpha

Appendix A. Supplementary data

The following is the Supplementary data to this article:

Details on polymer characterization techniques, including GPC, HPLC, and LC-MS/MS; instrumental parameters; cytokine analysis data; and additional in vivo immunofluorescence staining images of lung tissues.