Abstract

Background

   Pulmonary fibrosis, a major complication of severe COVID-19 and
   post-acute sequelae of SARS-CoV-2 infection (PASC), is driven by
   excessive neutrophil activation and the formation of neutrophil
   extracellular trap (NET).

Results

   This study presents a sequential nanoparticle-based therapy combining
   DNase-I-loaded polydopamine nanoparticles (DNase-I@PDA NPs) with
   Sivelestat-encapsulated PLGA nanoparticles (Siv@PLGA NPs) to target
   both NETs and neutrophil elastase (NE) activity. DNase-I@PDA NPs were
   aerosolized to the lungs, facilitating NET clearance and reducing the
   fibrotic microenvironment, followed by intravenous administration of
   Siv@PLGA NPs to inhibit NE activity and prevent neutrophil
   hyperactivation. In a murine model of lipopolysaccharide (LPS)-induced
   pulmonary fibrosis, this dual approach significantly decreased fibrotic
   lesions, collagen deposition, and myofibroblast activation. Notably,
   treatment with the nanoparticles led to substantial improvements in
   pulmonary function. In neutrophils isolated from COVID-19 patients, the
   combined nanoparticle therapy reduced circulating cell-free DNA, NET,
   NE, and myeloperoxidase (MPO) levels, while enhancing neutrophil
   viability and reducing inflammatory responses.

Conclusions

   These findings highlight the efficacy of DNase-I@PDA NPs and Siv@PLGA
   NPs in addressing both acute inflammation and chronic fibrosis by
   simultaneously targeting NET formation and neutrophil hyperactivation.
   This dual nanoparticle therapy represents a promising clinical strategy
   for treating COVID-19-associated pulmonary complications, including
   PASC, by preventing long-term fibrotic progression and promoting lung
   recovery.

Graphical abstract

   [52]graphic file with name 12951_2025_3421_Figa_HTML.jpg

Supplementary Information

   The online version contains supplementary material available at
   10.1186/s12951-025-03421-y.

   Keywords: Pulmonary fibrosis, Neutrophil extracellular traps,
   Aerosolized drug delivery, Chronic inflammation, Post-acute sequelae of
   SARS-CoV-2 (PASC)

Background

   Despite advances in treatment strategies through continuous endeavors
   after the outbreak of COVID-19, plethora of patients being declared
   fully cured are still suffering from the silent yet devastating
   aftermath of acute lung injury referred to as post-acute sequelae of
   SARS-CoV-2 infection (PASC) [[53]1]. PASC is being regarded as symptoms
   triggered by etiologically unknown SARS-CoV-2-driven chronic
   inflammation including dyspnea, coughing, thoracic discomfort, fatigue
   and impaired pulmonary function [[54]2]. Inasmuch as the precise
   mechanisms underlying PASC remain elusive, making effective treatment
   is daunting. As a result, clinicians often struggle with the
   intricacies of prescribing and resort to recommending symptomatic
   relief medications rather than eliminating the fundamental cause
   altogether. Moreover, this contributes to prolonged discomfort and pain
   for the patient, further adversely affecting quality of life. Thus,
   four years after the beginning of the COVID-19 pandemic, given that we
   now know the chance in which PASC occurs, it would be prudent to make
   prior consideration of long-term therapeutic approaches immediately
   after the infection to forestall the occurrence of PASC.

   Among the myriad clinical manifestations of PASC, pulmonary fibrosis
   stands out as a particularly severe and prevalent outcome driven by
   alveolar damage and the accumulation of collagen-producing fibroblasts,
   which result in excessive extracellular matrix deposition and
   irreversible impairment of pulmonary function [[55]3, [56]4]. While the
   acute inflammatory response is essential for pathogen clearance, in
   severe viral infections, it can trigger unchecked inflammation and
   immune dysregulation, culminating in chronic immune activation [[57]5].
   This prolonged immune state is characterized by sustained production of
   inflammatory cytokines and the infiltration of immune cells into lung
   tissue, precipitating extensive tissue damage and promoting
   extracellular matrix deposition, thus accelerating the progression of
   fibrosis [[58]6].

   Recent studies have proposed that NET may influence fibrotic responses
   in the lung and contribute to the development of long-term pulmonary
   complications following viral infection, including those observed in
   PASC [[59]7]. NET may activate signaling pathways in fibroblasts and
   epithelial cells, including TGF-β1/Smad2, β-catenin, and ERK1/2, which
   have been implicated in fibroblast-to-myofibroblast transition and
   extracellular matrix production. Suppression of miR-7, a regulator of
   TGF-β signaling, has also been observed in NET-rich environments,
   further enhancing Smad2 phosphorylation and promoting
   epithelial-to-mesenchymal transition (EMT) [[60]8]. Additionally, NET
   components such as oxidized DNA and histones can act as
   damage-associated molecular patterns (DAMPs), stimulating TLR9–NF-κB
   and cGAS–STING pathways, which are associated with chronic inflammation
   and immune-mediated activation of fibrotic pathways[[61]7, [62]9,
   [63]10].

   As part of the systems responsible for regulating NET turnover, DNase-I
   has been implicated in facilitating the breakdown of extracellular DNA
   [[64]11]. However, under fibrotic conditions, DNase-I activity appears
   to be reduced, which may delay NET clearance and result in their
   sustained accumulation in the lung [[65]7]. Persistent NET burden in
   this context has been linked to ongoing epithelial injury, inflammatory
   cell recruitment, and pro-fibrotic signaling, suggesting a potential
   self-reinforcing mechanism in fibrotic progression. Insights from these
   studies support the relevance of targeting NET stability and clearance
   as a potential therapeutic approach to limit inflammation-associated
   fibrosis in post-viral lung disease.

   In this study, we put forth a novel therapeutic strategy combining
   localized and systemic targeting aimed at curbing neutrophil
   hyperactivation and aberrant NET formation during the early phase of
   infection, with the intent of forestalling the progression of PASC.
   Specifically, we explored the combined use of DNase-I and Sivelestat,
   an established therapeutic regimen for pulmonary fibrosis that has
   shown inhibitory effects on NET formation in preclinical models and
   alleviating lung inflammation [[66]12]. Our approach was tailored to
   optimize both the timing and route of administration, ensuring maximal
   therapeutic effect within the inflammatory milieu that characterizes
   the early post-infection phase. Aerosolized DNase-I is initially
   administered to selectively target regions of the lung predisposed to
   fibrosis, thereby mitigating the excessive formation of NETs and the
   accumulation of circulating free DNA (cfDNA), both of which are
   strongly implicated in amplifying inflammatory cascades and fostering a
   pro-fibrotic microenvironment during the early stages of infection
   [[67]13, [68]14]. Following this, intravenous administration of
   Sivelestat, a neutrophil elastase (NE) inhibitor, modulates peripheral
   neutrophil activity to reduce excessive neutrophil activation. By
   integrating both localized and systemic interventions, this strategy
   aims to interrupt the pathological cascade that drives inflammation and
   tissue damage, thereby offering a comprehensive approach to mitigating
   the progression of lung fibrosis in severe infectious diseases and
   PASC.

   To address the limitations of DNase-I and Sivelestat, particularly
   their short half-lives and suboptimal pharmacokinetics, we developed
   nanoparticle formulations to extend their therapeutic efficacy and
   enhance targeted delivery. DNase-I, while effective in reducing NET
   formation, is rapidly degraded in vivo, limiting its sustained impact
   in modulating the pro-fibrotic microenvironment. Similarly, Sivelestat
   requires precise delivery to neutrophils to effectively mitigate
   neutrophil hyperactivation and systemic inflammation. To overcome these
   challenges, we fabricated DNase-I-loaded polydopamine nanoparticles
   (DNase-I@PDA NPs) and Sivelestat-encapsulated poly(lactic-co-glycolic
   acid) nanoparticles (Siv@PLGA NPs). In the case of Siv@PLGA NPs, we
   further enhanced targeting precision by conjugating F(ab’)[2] fragments
   of Ly6g antibodies, which specifically bind to neutrophils, ensuring
   that the therapeutic agents are delivered directly to the sites of
   neutrophil activation. This design not only extends the duration of
   drug action but also minimizes off-target effects, improving overall
   treatment specificity.

   We assessed the therapeutic efficacy of these nanoparticle formulations
   in both animal models and blood specimens from COVID-19 patients,
   showing reduction in neutrophil-driven inflammation and prevent
   pulmonary fibrosis. Our findings suggest that by improving the
   pharmacokinetics and targeting precision of DNase-I and Sivelestat,
   these nanoparticle-based therapies can offer a more effective and
   sustained intervention against the long-term sequelae of COVID-19,
   particularly in mitigating PASC-associated pulmonary fibrosis. This
   approach could pave the way for more refined therapeutic strategies in
   the management of post-infection complications.

Methods

Preparation and characterization of DNase-I@PDA NPs

   To fabricate DNase-I@PDA NPs, PDA NPs were first synthesized following
   a previously reported protocol [[69]15]. Briefly, 180 mg of dopamine
   hydrochloride (Sigma-Aldrich, USA) was completely dissolved in 90 mL of
   deionized water (DW). Subsequently, 760 μL of 1 N NaOH was added to the
   solution under vigorous stirring at 50 °C, leading to a gradual color
   change to dark brown. After 12 h, the PDA NPs were purified by
   centrifugation at 17,000 rpm and washed three times with DW, followed
   by lyophilization. To fabricate DNase-I@PDA NPs, 50 mg of lyophilized
   PDA was re-suspended in 10 mL of Tris buffer (10 mM, pH 8.5) containing
   25 mg of 4-arm amine-terminated PEG (Sunbio, Korea) and 25 mg of
   DNase-I (Roche, Swiss), and incubated under stirring at 4 °C for 4 h.
   Purification was performed as previously mentioned, involving
   centrifugation and three washes with DW, followed by lyophilization.
   The size, morphology, and zeta potential of the NPs were analyzed by
   using DLS (Zetasizer pro, Malvern Panalytical, UK) and a scanning
   electron microscope (SEM) Dimension ICON, Bruker, USA). The enzymatic
   activity of DNase-I@PDA NPs were confirmed via degradation of
   polymerized salmon sperm DNA (Sigma-Aldrich, USA) by gel
   electrophoresis [[70]16]. For this, 0.1–200 µg of DNase-I@PDA NPs were
   incubated with 2 µg of salmon sperm DNA for 10 min at 37 °C, followed
   by agarose 1% gel electrophoresis.

Preparation of neutrophil targeted Siv@PLGA NPs

   To fabricate neutrophil-targeted Sivelestat loaded PLGA nanoparticles
   (i.e., Neutrophil- targeted Siv@PLGA NPs), Siv@PLGA was first prepared
   using conventional emulsification methods [[71]17, [72]18]. Briefly, 30
   mg of PLGA (LG 90:10, Polysciences, USA) and 10 mg of
   PLGA-PEG-Maleimide (LG 50:50, PolySciTech, USA) were dissolved in 2 mL
   of dichloromethane (DCM). Sivelestat sodium tetrahydrate (TCI, Japan)
   at a concentration of 100 mg/mL in DMSO was added to the PLGA solution
   at 30% of the PLGA weight. The resulting polymer solution was sonicated
   for 10 min at 20% amplitude (140 W) using a 1 sec on-1 sec off cycle
   (Qsonica, USA). The resulting emulsion was then stirred at room
   temperature under vacuum for 1 h to completely evaporate the DCM.
   Siv@PLGA NPs were then collected by centrifugation at 17,000 rpm,
   washed three times with DW, and lyophilized. Next, to enable neutrophil
   targeting, the F(ab’)[2] fragment of the anti-mouse Ly6G antibody
   (BioXcell, USA) was conjugated to the NPs surface. To generate the
   F(ab’)[2] fragment, the antibody was cleaved using IdeS (Promega, USA)
   and confirmed via non-reducing SDS-PAGE. The F(ab’)[2] fragments were
   then reduced with 2.5 μL of 10 mM dithiothreitol (DTT) (Sigma-Aldrich,
   USA) per 100 μg of antibodies to obtain free thiol groups in the hinge
   region. The free DTT was removed from the F(ab’)[2] fragments by using
   7 kDa MWCO desalting columns (Thermo Scientific, USA). The previously
   prepared NPs were re-suspended in phosphate-buffered saline (PBS,
   pH7.4) to a concentration of 8 mg/mL. To each milligram of particles,
   25 μg of F(ab’)[2] was added and incubated at room temperature for 1 h.
   The mixture was then spun down at 17,000 rpm and washed with PBS to
   obtain the neutrophil-targeted Siv@PLGA NPs.

Characterization of neutrophil targeted Siv@PLGA NPs

   The NPs size, zeta potential, and morphology were analyzed as
   previously mentioned using DLS and SEM. In vitro drug release was
   assessed by reverse-phase high-performance liquid chromatography
   (HPLC). Siv@PLGA was dispersed in PBS at a concentration of 2.5 mg/mL,
   and 2 mL of this dispersed polymer solution was dialyzed using
   SnakeSkin^® dialysis tubing (MWCO: 3500, Thermo Fisher Scientific, USA)
   against 23 mL of PBS at 37 °C with gentle shaking. At each time point,
   the PBS outside the tubing was collected and replaced with fresh PBS.
   The collected PBS was centrifuged at 17,000 rpm for 10 min, and the
   supernatant was analyzed by HPLC. The HPLC analysis was conducted using
   a 1260 Infinity II (Agilent, USA), with chromatographic separation
   achieved on a C18 reverse-phase column (4.6 × 150 mm, 5 µm particle
   size, Agilent, USA). The mobile phase comprised solvent A (0.1% Formic
   acid in DW) and solvent B (Acetonitrile), with a gradient elution
   applied as follows: 0–8 min, 20% B to 80% B; 8–12 min, 20% B. The flow
   rate was 1.0 mL/min, the column temperature was maintained at 30 °C,
   and detection was performed at 230 nm with an injection volume of 20
   µL. The drug loading was measured using the same method after
   dissolving Siv@PLGA NPs in a 1:1 (v/v) mixture of DCM and
   dimethylformamide.

   The conjugation efficiency of the F(ab’)[2] fragment on the NPs surface
   was quantified using a BCA assay (Micro BCA Protein Assay Kit, Thermo
   Fisher Scientific, USA). According to the instructions, a working
   reagent (WR) was prepared, and 100 μL of a 2 mg/mL dispersion of
   Neutrophil-targeted Siv@PLGA NPs were aliquoted into a 96-well plate.
   An equal volume of WR was added, and the reaction was conducted at 37
   °C for 2 h. To establish a standard curve, an equal volume of bare
   Siv@PLGA (without antibody conjugation) was used in the wells, and
   F(ab’)[2] fragments were added in amounts ranging from 0 to 5 μg.

   To evaluate in vitro neutrophil targeting efficiency,
   neutrophil-targeted PLGA NPs labeled with the lipophilic carbocyanine
   dye, 1,1′-Dioctadecyl-3,3,3′,3′-Tetramethylindodicarbocyanine (DiD,
   Biotium, USA), were co-incubated with splenocytes isolated from murine
   spleen, followed by flow cytometry analysis. Mouse splenocytes were
   seeded at 0.5 × 10^6 cells in PBS into a 96-well plate, and treated
   with 20 µg of either F(ab’)[2] of anti-Ly6G-conjugated PLGA NPs or bare
   PLGA NPs. The incubation was performed at 37 °C with shaking for 2 h.
   The cells were washed in PBS after co-incubation. Subsequently, the
   cells were labeled with fluorescence-conjugated antibodies including
   Zombie Aqua (Biolegend, USA) for Live/Dead staining, PE-CD45
   (Biolegend, USA), APC/Cy7-Ly6g/6c (Biolegend, USA), and BV421-CD11b
   (Biolegend, USA). To evaluate non-specific targeting efficiency, the
   fluorescence of the NPs was detected in T cells and macrophages. Each
   cell type was labeled with Zombie Violet (Biolegend, USA), BV780-CD45
   (Biolegend, USA), FITC-TCRβ (Biolegend, USA), and PE/Cy5-F4/80
   (Biolegend, USA), respectively.

Animals and husbandry

   5-week-old female ICR mice (Orient Bio, Republic of Korea) were used
   after a week of adaptation period. Under controlled temperature of
   20–25 °C and humidity of 40–45%, five mice were housed per cage on a
   12-h light, 12-h dark cycle. The mice were provided with a standard
   rodent pellet diet and had unrestricted access to water. All mice were
   handled in accordance with the Guidelines for the Care and Use of
   Laboratory Animals set forth by Sungkyunkwan University (IACUC No.:
   202111291).

Aerosolized Lipopolysaccharide (LPS)-induced pulmonary fibrosis mouse model

   An aerosol of LPS (E. coli O111, Sigma, USA), or same amount of PBS as
   a control, was intratracheally administered to anesthetized mice using
   a microsprayer. Each mouse was treated with 3 mg/kg LPS three times
   with intermissions of 12 h. Description for treatment of DNase I-PDA@NP
   and Siv@PLGA NP in mouse model.

Quantification of DNase-I using Enzyme-Linked Immunosorbent Assay (ELISA) in
murine plasma and bronchoalveolar lavage fluid (BALF)

   To obtain plasma, collected whole blood were centrifuged at 2000×g for
   20 min. Supernatant of bronchoalveolar lavage fluid BALF was obtained
   after centrifuged at 1000×g for 10 min. Samples were coated on
   immunoplate and incubated overnight at 4 °C. DNase-I antibody
   (sc-376207, Santa Cruz, USA) was 1:1000 diluted, and loaded. The plate
   was incubated for 2 h at RT. 1:5000 diluted HRP-linked anti-mouse
   secondary antibody (Thermo Scientific, USA) was used as detection
   antibody. o-Phenylenediamine(OPD) was loaded as substrate, followed by
   being detected at 492 nm using Tecan Spark microplate reader (Tecan,
   Austria GmbH, Austria).

Quantification of plasma cell-free DNA (cfDNA)

   To isolate cell-free DNA (cfDNA from the plasma of SARS-CoV-2-infected
   patients or from mouse plasma, the Qiagen QIAamp DNA Mini Blood Mini
   Kit was employed. This procedure was carried out in strict accordance
   with the manufacturer's instructions provided by Qiagen (Valencia, CA,
   USA). Prior to the isolation process, the plasma samples underwent
   centrifugation at 16,800×g for 10 min to remove cellular debris and
   ensure a clean sample for cfDNA extraction. Following isolation, the
   purified cfDNA was quantified using NanoDrop spectrophotometry.

Method

   Complete Blood Count-For hematological analysis, white blood cell
   (WBC), neutrophil, and monocyte counts were measured using the Mindray
   BC-5000 vet automated hematology analyzer (Mindray Bio-Medical
   Electronics Co., Ltd., Shenzhen, China) at the Chiral Material Core
   Facility Center of Sungkyunkwan University. Prior to analysis,
   murine whole blood was collected into tubes preloaded with an
   appropriate volume of K[2]-EDTA solution to prevent coagulation. To
   maintain data consistency, daily quality assurance procedures were
   implemented using reference controls recommended by the manufacturer.

In vivo biodistribution

   The mice received aerosol administration of fluorescence-conjugated
   DNase-I@PDA NPs. The other group was administered
   fluorescence-conjugated Siv@PLGA NPs via intravenous (i.v.) injection.
   At specified intervals (6-, 12-, and 24-h post-administration), each
   group of mice was anesthetized and imaged in vivo using the
   Fluorescence In Vivo Imaging System (FOBI) from NEO Science (Republic
   of Korea). Following imaging, mice were sacrificed, and images of
   harvested organs were taken by the FOBI system.

Cytokine array

   Cytokine profiles were evaluated using a Cytokine Antibody Array Kit
   (Full Moon Bio, USA). Blood samples were collected from pulmonary
   fibrosis (PF) mice, and plasma was isolated and stored at − 80 °C until
   analysis. Following the manufacturer’s instructions, the antibody array
   membranes were blocked and subsequently incubated overnight at 4 °C
   with diluted plasma samples. After thorough washing, biotin-conjugated
   detection antibodies were applied, followed by streptavidin-HRP.
   Chemiluminescent detection was performed using a substrate, and signals
   were captured via a digital imaging system. Quantitative analysis was
   conducted to identify and quantify cytokines involved in the
   inflammatory response.

   Cytokine expression values were comprehensively analyzed using R
   statistical software. Clustering analysis was performed using the
   k-means clustering method in R. The gene sets associated with each
   cluster were input for pathway enrichment analysis using Metascape
   ([73]https://metascape.org), and pathways with an adjusted p-value <
   0.01 were selected. Additionally, genes enriched in LPS and free drugs,
   but not in NP, were identified by applying a significant threshold of
   Log[2]FC < − 1 when comparing NP to LPS and free drugs. Visualization
   of cytokine expression levels and principle component analysis (PCA)
   analysis was conducted using the ComplexHeatmap and ggplot2 packages in
   R.

Hematoxylin and eosin staining

   Lungs from each mouse were collected, rinsed with PBS, and immersed in
   4% paraformaldehyde (PFA) solution (Biosesang, Republic of Korea) at pH
   7.4 for fixation. The fixation process was conducted at 4 ºC for 12 h.
   After fixation, the lung tissues underwent dehydration through a series
   of sucrose solutions before being embedded in Optimal Cutting
   Temperature (OCT) compound and stored at − 80 ºC until sectioning. The
   samples were then sliced into 10 μm sections and mounted onto slides.
   Subsequently, the slides were fixed in 4% PFA solution and stained with
   hematoxylin (Sigma, USA). To remove excess hematoxylin, the slides were
   briefly rinsed in distilled water before being counterstained with
   eosin (Sigma, USA). Following staining, the slides were treated with a
   series of ethanol and xylene solutions, then cover-slipped. The images
   were obtained by light microscopic analysis (DM500, Leica, Germany).

Orange G staining

   Processing of lung tissue samples was performed in accordance with the
   previously established protocol. After fixation in 4% PFA solution, the
   samples were thoroughly washed with PBS to remove excess fixative
   before proceeding to the staining process.

   The lung tissue sections were then stained with hematoxylin (Sigma,
   USA) to visualize cellular structures and orange G-6 (Yeong Dong
   Diagnostics, Korea). Following staining, the slides underwent a series
   of graded ethanol washes to remove excess stain and were subsequently
   dehydrated using xylene to prepare them for mounting. Finally, the
   stained slides were cover-slipped using a mounting medium and analyzed
   under a light microscope to assess histological features.

Trichrome staining

   The samples were cross-sectioned into 10 μm thick slices, which were
   then fixed in 4% PFA for 30 min. After fixation, the slides were
   sequentially stained by immersing them in Mordant in Bouin’s Solution
   at 60 °C for 1 h, followed by Weigert’s working hematoxylin for 7 min,
   Phosphotungstic/Phosphomolybdic Acid Solution for 10 min, and 1% acetic
   acid for 1 min. Post-staining, the slides underwent dehydration with
   two changes of 95% ethanol and two changes of 100% ethanol (2 min per
   change). The samples were then cleared with three changes of xylene (5
   min per change) and cover slipped using permount mounting solution.
   Finally, light microscopic analysis of the lung specimens was performed
   under blinded conditions to evaluate the extent of fibrosis.

Hydroxyproline assay

   Hydroxyproline levels in lung tissues were assessed using a
   commercially available kit (MAK008, Sigma-Aldrich). The procedure was
   carried out following the detailed instructions provided by the
   manufacturer.

Immunohistochemistry

   For immunohistochemistry, frozen lung sections were initially fixed in
   4% PFA solution and then washed three times with PBS. The sections were
   permeabilized using 0.05% Triton-X solution and blocked overnight at 4
   °C in 1% BSA in PBS. The following day, the samples were incubated for
   12 h at 4 °C with anti-Cit-H3 (ab5103, Abcam) and anti-NE (ab68672,
   Abcam) in BSA solution in PBS, gradually increasing concentrations up
   to 2%. After each step, the slides were washed three times with PBS to
   remove nonspecific binding. Subsequently, the sections were incubated
   at room temperature for 1 h with secondary antibodies: goat anti-rabbit
   Alexa Fluor 594 and 350. The slides were then washed, incubated with
   Sytox Green (S7020, Invitrogen) for 10 min, washed with 1X PBST,
   mounted with Dako mounting medium, and imaged using fluorescence
   microscopy (Leica, Germany).

   Immunohistochemistry for α-smooth muscle actin (α-SMA) and Collagen I
   was performed following the same protocol as described above. For
   primary antibody staining, sections were incubated overnight at 4 °C
   with mouse anti-Collagen I (Ab34710, Abcam) and anti-α-SMA (ab7817,
   Abcam) antibodies diluted in blocking solution. After extensive washing
   with PBS, sections were incubated with secondary antibodies (goat
   anti-mouse Alexa Fluor 594, goat anti-rabbit Alexa Fluor 488).

Lung function assessment

   The flexiVent system (SCIREQ Scientific Respiratory Equipment, Canada)
   was utilized to assess various lung function parameters in mice,
   including Elastance, Compliance, Total Lung Capacity, Inspiratory
   Capacity, Tissue Elasticity, Quasi-Static Compliance, and the
   Pressure-Volume Curve. To prepare the mice for these measurements, they
   were administered an injection of pancuronium bromide at a dose of 0.8
   mg/kg. This compound serves as a skeletal muscle relaxant, ensuring
   that the mice remained immobile during the procedure. Following
   euthanasia, the mice underwent orotracheal intubation using an 18 G, 30
   mm catheter, which allowed for precise placement on the flexiVent
   system. The mice were then mechanically ventilated with forced
   oscillation to simulate breathing conditions. Each parameter
   measurement was conducted three times per mouse to ensure accuracy and
   reproducibility. The collected data were then recorded and calculated
   to provide comprehensive insights into the respiratory mechanics of the
   mice.

Body weight and survival rate

   The survival rate and body weight of LPS-treated mice were monitored
   over a 480-h period following NP treatment. Body weights were measured
   daily using a precision balance and recorded in grams, while survival
   rates were assessed at regular intervals throughout the study. Mice
   were carefully monitored for any signs of morbidity or distress, and
   humane endpoints were established following institutional guidelines.
   Survival data were represented as percentages and plotted using
   Kaplan–Meier survival curves. Body weight changes were calculated
   relative to baseline measurements and presented as absolute values (in
   grams).

Sample collection and of COVID-19 patients’ blood

   Blood samples were collected from patients admitted to Yeungnam
   University Medical Center after being diagnosed with SARS-CoV-2 at a
   public health center in Daegu, Republic of Korea. Patients were
   eligible if they were over 18 years old and had an acute medical
   condition with at least one of the following: fever (tympanic
   temperature ≥ 38 °C at nurse triage), suspected systemic infection, two
   or more systemic inflammatory response syndrome (SIRS) criteria,
   hypotension (systolic blood pressure < 90 mmHg), or shock. Healthy
   volunteers were used as a control group. Clinical data were gathered
   for all participants, and the study was conducted following the
   approval of the Institutional Review Board of Yeungnam University
   Hospital (YUH 2020-03-057 and 2020-05-031-001). Complete blood counts
   were meticulously analyzed using venous blood samples collected within
   24 h of the patients'admission to the hospital. Specifically, the
   neutrophil count was measured as part of this analysis. The assessment
   was performed with the aid of a Sysmex XE-2100 Automated Hematology
   System (TOA Medical Electronics, Japan).

NET ELISA

   NETs were generated from freshly isolated neutrophils at a
   concentration of 1 × 10^5 cells. To induce the formation of NETs, the
   neutrophils were stimulated with phorbol-myristate acetate (PMA)
   (Sigma-Aldrich, USA) at a concentration of 25 nM. In parallel, control
   samples were treated with media composed of RPMI 1640, which was
   supplemented with glutamine, penicillin, and streptomycin. Following
   the stimulation, the production of NETs was analyzed using a
   fluorometric technique, as outlined in previously established
   protocols. The quantification of NET production was expressed in terms
   of arbitrary fluorescent units (AFUs), providing a standardized measure
   of the extent of NET formation under these experimental conditions.

MPO ELISA

   Plasma samples were thoroughly analyzed to assess the extent of granule
   matrix protein release that occurs during the degranulation process in
   peripheral blood mononuclear cells (PBMCs) from both
   SARS-CoV-2-infected patients and mice. To achieve this, a specialized
   human MPO ELISA kit (BMS2038INST, Invitrogen, USA) was employed for
   human samples, while a corresponding mouse MPO ELISA kit (MBS700747,
   MyBioSource, Canada) was utilized for the murine samples.

ELISA for DNase-I activity

   Plasma samples were diluted at a ratio of 1:50 and subjected to
   analysis using a digestion buffer that had been spiked with
   double-stranded DNA at a concentration of 1 µg/mL. The samples were
   then stained with PicoGreen, a fluorescent nucleic acid stain provided
   by Invitrogen (MA, USA), following the detailed instructions outlined
   in the manufacturer’s protocol. After the staining process, the samples
   were incubated at 37 °C for a duration of 5 h. Following incubation,
   the reduction in PicoGreen staining was assessed by measuring the
   fluorescence emission (Em) using a fluorometer. This reduction in
   fluorescence provided a quantitative measure of the changes in DNA
   content within the plasma samples under the experimental conditions.

Statistical analysis and software

   Each experiment was conducted independently a minimum of three times.
   Statistical significance was assessed via the unpaired t-test, and all
   analyses were carried out using GraphPad Prism software. Results are
   presented as mean ± SEM, with a significance threshold of P < 0.05.
   P-values and additional experimental details are outlined in the figure
   legends. R statistical software (version 4.3.2) was used for
   comprehensive analysis of cytokine array data. Visualization of
   cytokine expression levels and PCA analysis were performed using the
   ComplexHeatmap and ggplot2 packages in R.

Results

NPs preparation and characterization

   The PDA NPs were synthesized through the oxidative polymerization of
   dopamine hydrochloride in an alkaline aqueous solution. This process
   was initiated by adjusting the pH with NaOH, followed by auto-oxidation
   under ambient air conditions, as illustrated in Scheme 1 (Scheme 1)
   [[74]15]. Subsequently, following our previously established method
   [[75]19–[76]21], DNase-I and PEG were co-immobilized onto the surface
   of PDA NPs through a single-step procedure. This process exploits the
   reactivity of catechol and quinone groups on the PDA surface. The
   primary amine groups of DNase-I and the terminal amine groups of PEG
   formed covalent bonds with the PDA surface via Schiff base formation
   and Michael addition reactions, resulting in stable surface
   functionalization. Additionally, neutrophil-targeted Siv@PLGA NPs were
   prepared using a conventional emulsification method and subsequently
   conjugated with F(ab’)[2] fragments of an antibody specific for
   neutrophils [[77]22]. The morphology of the particles, as observed via
   SEM, was uniformly spherical (Fig. [78]1A). Additionally, the sizes of
   the PDA and DNase-I@PDA NPs, measured by DLS, were approximately 199
   and 184 nm, respectively, with minimal size variation attributable to
   the binding of PEG and DNase-I to the surface (Fig. [79]1B). The sizes
   of PLGA and Siv@PLGA NPs were found to be 274 and 273 nm, respectively.
   This minimal difference in size indicates that the incorporation of
   Sivelestat into the PLGA nanoparticles did not significantly alter
   their dimensions. The zeta potentials of PLGA and Siv@PLGA NPs were
   −11.13 and − 11.87 mV, respectively, showing minimal variation (Fig.
   [80]1C). In contrast, PDA and DNase-I@PDA NPs exhibited zeta potentials
   of − 11.33 mV and − 17.92 mV, respectively, with DNase-I@PDA NPs
   displaying approximately 36% more negative charge. This increased
   negative charge is due to the binding of negatively charged DNase-I to
   the surface of PDA NPs [[81]23, [82]24]. The release profile of
   Sivelestat from Siv@PLGA NPs was quantified over time using HPLC (Fig.
   [83]1D). The drug loading achieved was 163 µg of Sivelestat per mg of
   PLGA NPs, yielding an encapsulation efficiency of 54%. As shown in Fig
   [84]1D, Sivelestat demonstrated a sustained release from the PLGA NPs
   over a 24-h period. The DNA degradation enzymatic activity of
   DNase-I@PDA NPs were evaluated using agarose gel electrophoresis (Fig.
   S1). While 2 U of free DNase-I completely degraded 2 µg of DNA,
   DNase-I@PDA NPs achieved complete degradation of the same amount of DNA
   at concentrations above 5.0 µg. These results demonstrate that the
   DNase-I immobilized on the PDA NPs surface retains its effective DNA
   degradation capability.

Fig. 1.

   [85]Fig. 1
   [86]Open in a new tab

   Characterization of NPs. A SEM image (n = 3, Scale bar = 500 nm) B Size
   and C Zeta potential (n = 3) of PDA, DNase-I@PDA, PLGA, and Siv@PLGA
   NPs. D Drug release profile of Siv@PLGA NPs (n = 4). E In vitro
   targeting efficiency of Neutrophil targeted PLGA NPs (n = 5).
   Statistical analysis was performed using a two-tailed unpaired t-test.
   Data are presented as mean ± SEM. * P < 0.05, ** P < 0.01, *** P <
   0.001, **** P < 0.0001 (v.s. LPS)

   Neutrophil-targeted Siv@PLGA NPs were prepared by conjugating the
   F(ab’)[2] fragment of the anti-Ly6g antibody, which targets
   neutrophils, to the Siv@PLGA NPs. To achieve this, the anti-Ly6g
   antibody is cleaved into F(ab’)[2] and Fc fragments using the protease
   IdeS [[87]17, [88]22]. The cleavage of the antibody was confirmed by
   non-reducing SDS–PAGE gel stained with Coomassie Brilliant Blue, where
   bands corresponding to F(ab’)[2] and Fc were observed at 115 and 25
   kDa, respectively (Fig. S2). After cleavage, the hinge region of the
   F(ab’)[2] fragment is reduced with DTT to generate thiol groups, which
   then conjugate with the maleimide groups on the particle surface. The
   conjugation efficiency of the antibody to the particle surface was
   subsequently verified using a BCA assay. It was determined that 17.5 µg
   of antibody was bound per 1 mg of NPs, corresponding to 70% of the
   antibody used in the reaction being conjugated to the particle surface.

   The targeting efficiency by the antibody conjugation to surface of PLGA
   NPs was assessed in vitro using flow cytometry with fluorescently
   labeled PLGA NPs in mouse splenocytes (Fig. [89]1E). The intensity
   shifts of DiD was compared between cells treated with DiD-loaded
   neutrophil-targeted PLGA NPs and cells treated with bare PLGA NPs, with
   untreated cells as baseline. In the case of target cells, neutrophils,
   the shift in DiD signal increased by 19% when treated with
   neutrophil-targeted PLGA NPs, which is approximately five times higher
   than that observed with DiD-loaded bare PLGA NPs. This indicates that
   the conjugation of antibodies to the surface of PLGA NPs enhances the
   targeting efficiency towards neutrophils. Additionally, to evaluate
   non-specific binding of the PLGA NPs, DiD signal shifts were examined
   in non-target cells, such as T cells and macrophages. The results
   showed that neutrophil-targeted PLGA NPs exhibited a slightly higher
   signal shift (approximately 2%) compared to bare PLGA NPs in both T
   cells and macrophages, indicating that non-specific binding of the
   neutrophil-targeted PLGA NPs is minimized.

   Next, we analyzed time-dependent degradation of DNase-I@PDA NP in
   murine serum to evaluate its enzymatic persistence. The results showed
   that DNase-I levels peaked at 6 h post-treatment, suggesting that the
   nanoparticle formulation effectively maintained DNase-I activity during
   the early phase after administration. After reaching its peak, DNase-I
   levels steadily decreased. A significant drop was detected at 12 h
   post-treatment, which indicates continuous enzymatic degradation. By 24
   h following treatment, DNase-I levels were no longer detectable within
   this timeframe. This observation demonstrates that the
   nanoparticle-conjugated DNase-I had undergone complete degradation.
   (Fig. S3).

   These findings imply that DNase-I@PDA NPs remain active for a limited
   but sufficient duration to exert their enzymatic function before being
   naturally degraded, aligning with the expected pharmacokinetics of
   enzyme-based therapies. The observed degradation pattern suggests that
   DNase-I@PDA NPs provide a therapeutic window that allows for effective
   NET degradation while minimizing the risk of prolonged systemic
   exposure, which could potentially reduce off-target effects.

   To assess the therapeutic potential of the nanoparticle formulations
   under inflammatory conditions resembling infection-induced lung injury,
   we established an in vitro transwell co-culture system that models the
   crosstalk between endothelial and alveolar compartments with a focus on
   neutrophil-mediated responses. Human endothelial cells (hy926) were
   seeded in the upper chamber, while human lung fibroblasts (MRC5) were
   placed in the lower chamber. LPS was applied to the entire system to
   induce a pro-inflammatory environment (Fig. S4 A). Following 12 h of
   LPS exposure, free DNase-I or DNase-I@PDA NPs were administered into
   the lower chamber to simulate alveolar delivery of therapeutics. One
   hour later, neutrophils were introduced into the upper chamber
   alongside either free Sivelestat or Siv@PLGA NPs, modeling systemic
   drug administration and neutrophil recruitment from circulation into
   inflamed lung tissue.

   Under these conditions, neutrophils were anticipated to become
   activated and migrate toward the lower compartment in response to
   LPS-driven cues. Notably, treatment with nanoparticle formulations
   resulted in a measurable reduction in neutrophil infiltration (Fig.
   S4B), suggesting a potential role in tempering activation or migratory
   responsiveness. A concurrent decrease in cfDNA levels within the upper
   chamber supernatant over time (Fig. S4 C) further implied suppression
   of NET formation, which has been implicated in amplifying fibroblast
   activation through extracellular signaling. To better understand the
   functional consequence of this modulation, we examined the early
   fibrotic response in fibroblasts by measuring α-SMA levels in MRC5
   lysates. Western blot analysis revealed that, in contrast to untreated
   conditions where α-SMA expression was elevated at early time points
   following LPS and neutrophil exposure, NP-treated groups did not
   exhibit this early induction (Fig. S4D). This early suppression may
   reflect a favorable alteration in the cellular
   microenvironment—potentially linked to reduced exposure to
   neutrophil-derived signals such as NETs or inflammatory proteases.
   Building on these in vitro findings, we proceeded to evaluate the
   therapeutic efficacy of the nanoparticle formulations in vivo to
   further validate their impact under physiologically relevant
   conditions.

Amelioration of acute immune responses in LPS-induced lung injury through
early administration of DNase-I

   In previous study, we demonstrated that the activity of DNase-I may
   decrease in the immediate aftermath of COVID-19 infection, potentially
   contributing the complications such as the formation of NETs and
   eventual increase of disease severity [[90]19]. Based on these
   observations, we hypothesized that early intratracheal administration
   of DNase-I@PDA NPs could compensate for reduced DNase-I activity and
   attenuate the rapid accumulation of NET-derived DNA within the
   pulmonary system. This strategy was anticipated to prevent excessive
   inflammatory responses that contribute to acute lung injury.
   Additionally, intravenous administration of Siv@PLGA NPs, by inhibiting
   NE activity, was designed to suppress peripheral neutrophil
   hyperactivation, thereby preventing the sustained inflammatory
   escalation despite neutrophil infiltration into lung tissue (Fig.
   [91]2A).

Fig. 2.

   [92]Fig. 2
   [93]Open in a new tab

   Early administration of DNase-I ameliorates acute immune response in
   the case of COVID-19 infection. A Schematic of DNase-I@PDA NPs and
   Siv@PLGA NPs administration. B Schematic of LPS injection and sacrifice
   schedule. Mice were intratracheally (i.t.) administered three times
   with LPS (3 mg/kg) and sacrificed at every 24 h interval. C Schematic
   of LPS injection and treatment schedule. Mice were intratracheally
   (i.t.) administered three times with LPS (3 mg/kg) and were
   intratracheally (i.t.) administered twice with free DNase-I (D group)
   or DNase-I @NP (D@P group) at 24 h after the third injection of LPS.
   Mice were sacrificed at 12 h after the final administration of free
   DNase-I or DNase-I@NP at every 24 h interval. D Concentration of
   DNase-I in blood and BALF. E Neutrophil count in blood and BALF. F
   Immune cell populations in BALF at 72 h post-administration. G Immune
   cell populations in BALF at 96 h post-administration. Statistical
   analysis was performed using a two-tailed unpaired t-test. Data are
   presented as mean ± SEM. * P < 0.05, **P < 0.01, ***P < 0.001, ****P <
   0.0001 (v.s. LPS), ^##P < 0.01, ^###P < 0.001 (v.s. D + S)

   To confirm the validity, we performed an investigation to determine
   whether aerosol injection of free DNase-I (D group) or DNase-I@PDA NPs
   (D@P group) decrease activation of inflammation mediators in pulmonary
   fibrosis mouse model induced by serial intratracheal injections of LPS.
   We assessed the levels of DNase-I and various immune factors in an
   animal model at early, intermediate, and late stages, which we
   hypothesized correspond to the development of post-infectious pulmonary
   fibrosis (Fig. [94]2B, [95]C). Regarding DNase-I levels, the results
   indicated a significant decrease in DNase-I concentration in both
   murine plasma and BALF 48 h after direct intratracheal administration
   of LPS (Fig. [96]2D). Additionally, we noted an increase in the
   neutrophil population in the blood (Fig. S5), along with a decrease in
   monocyte infiltration into the BALF (Fig. [97]2E). This observation
   implies a shift in immune cell dynamics throughout the inflammatory
   response. By 72 and 96 h post-LPS injection, DNase-I levels in plasma
   showed a significant rebound compared to the 48 h time point (Fig. S6),
   mirroring our previous observations [[98]19] of elevated DNase-I in
   severe COVID-19 patients. DNase-I levels in the BALF remained reduced
   throughout the 96 h period, despite the systemic rebound observed in
   the plasma (Fig. [99]2F, [100]G). However, it is worth noting that
   treatment with free DNase-I and DNase-I@PDA NPs effectively restored
   DNase-I levels in the BALF (Fig [101]2F, [102]G). DNase-I@PDA NPs
   achieved a more sustained elevation of DNase-I levels compared to free
   DNase-I, likely due to the enhanced release profile of DNase-I.
   Moreover, a marked reduction in blood neutrophil levels and monocyte
   infiltration into the BALF was observed in the D and D@P group.
   Immunohistochemical (IHC) analysis of monocytes also showed that the
   LPS group exhibited a greater presence of monocytes beyond the vascular
   boundary. In comparison, monocyte infiltration appeared reduced in both
   the D and D@P-treated groups. Such histological evidence is consistent
   with the observed decrease in monocyte levels within the BALF and
   supports the possibility that the treatment may help temper
   monocyte-associated inflammatory activity (Fig. S7). These findings
   suggest that early administration of DNase-I, particularly in its
   nanoparticle formulation, may help restore DNase-I levels in the
   context of pulmonary fibrosis and influence immune cell dynamics,
   including neutrophil and monocyte populations, which play important
   roles in the inflammatory cascade associated with fibrotic progression.

Pulmonary retention and coadministration of DNase-I@PDA NPs and Siv@PLGA NPs

   Following the demonstrated efficacy of DNase-I formulations in
   attenuating acute lung inflammation, we sought to explore the
   synergistic therapeutic effects of co-administration of DNase-I@PDA NPs
   and Siv@PLGA NPs in vivo. Our primary objective was to assess pulmonary
   retention, systemic distribution, and potency of these nanoparticles to
   exert both localized and systemic anti-inflammatory effects,
   particularly in the setting of persistent inflammation and fibrosis.

   To visualize nanoparticle biodistribution and pulmonary retention,
   fluorescence(Cy7)-conjugated DNase-I@PDA NPs and Siv@PLGA NPs were
   administered according to the outlined experimental scheme (Fig.
   [103]3A). Fluorescence imaging of lung tissues revealed sustained
   pulmonary localization of aerosol-administered DNase-I@PDA NPs, with a
   gradual decrease in fluorescence intensity over 24 h
   post-administration. This extended retention supports the hypothesis
   that inhalation delivery ensures effective deposition and persistence
   of DNase-I@PDA NPs within pulmonary tissue (Fig. [104]3A), enhancing
   their therapeutic window in mitigating early inflammatory processes.
   This result is also consistent with the previously conducted in vitro
   DNase-I degradation experiment (Fig. S3).

Fig. 3.

   [105]Fig. 3
   [106]Open in a new tab

   Coadministration of DNase-I and Sivelestat nanoparticle and drug
   release. A Ex vivo imaging of dissected organs after the i.v.
   administration of Cy7- Siv@PLGA NP and the i.t. administration of Cy7-
   DNase-I PDA@NP. B Schematic of LPS injection and treatment schedule.
   Mice were intratracheally (i.t.) administered three times with LPS (3
   mg/kg). DNase-I PDA@NP was administrated intratracheally and Siv@PLGA
   NP was administrated intravenously, alternating 12 h apart. Mice were
   sacrificed at 48 h after the final administration of Siv@PLGA NP. C
   Heatmap illustrating the top 50 cytokines identified from a cytokine
   array analysis across the treatment conditions: LPS, DNase-I combined
   with Sivelestat (D + S), and each nanoparticle treatment (NP). The
   color scale denotes relative expression levels, with red indicating
   upregulation. Pathway enrichment analysis revealed significant
   involvement of inflammatory and immune-related pathways, including
   cytokine-cytokine receptor interaction, chemokine signaling, and the
   MAPK signaling pathway. D Neutrophil counts, E NET ratio, F DNase-I
   activity, G Concentration of NE, H MPO activity, and I cfDNA in murine
   BALF. Statistical analysis was performed using a two-tailed unpaired
   t-test. Data are presented as mean ± SEM. *P < 0.05, **P < 0.01, ***P <
   0.001, ****P < 0.0001 (v.s. LPS), ^# P < 0.05, ^## P < 0.01, ^### P <
   0.001 (v.s. D + S)

   Siv@PLGA NPs, delivered intravenously, were primarily expected to
   target peripheral neutrophils and inhibit NET formation upon migration
   into pulmonary tissues (Fig. [107]1E). While hepatic accumulation is an
   expected outcome due to the liver’s role in nanoparticle clearance,
   efficient entry to the lungs was also successfully achieved (Fig.
   [108]3A). Fluorescence tracking showed that NP-bound neutrophils
   migrated to the lungs within 6 h of administration, suggesting
   effective neutrophil targeting. The continued presence of Siv@PLGA NPs
   in peripheral tissues indicates that they may contribute to regulating
   systemic neutrophil activity, which could help limit excessive NET
   formation and reduce inflammation in the lungs.

   Cytokine profiling and pathway analysis in LPS-induced lung fibrosis
   models developed according to the scheme (Fig. [109]3B) showed that
   treatment with DNase-I@PDA NPs and Siv@PLGA NPs (NP group)
   significantly modulated inflammatory and fibrotic responses compared to
   the LPS and DNase-I and Sivelestat (D+S) groups (Fig. S8). PCA
   identified distinct cytokine expression patterns, with NP-treated
   animals showing enrichment in cluster 4, indicating a unique
   immunomodulatory effect (Fig. S9). Pathway analysis of differentially
   expressed cytokines indicated a marked suppression of proinflammatory
   and profibrotic pathways, including extracellular matrix organization,
   cell migration, as well as the PI3 K-Akt, NF-κB, and MAPK signaling
   pathways. Notably, NP treatment suppressed key chemotaxis-related
   cytokines such as MCP-2, GRO, PF-4, and LD78beta, critical for
   neutrophil recruitment and NET formation. This NET inhibition, along
   with the regulation of cytokine-cytokine receptor interactions and
   chemokine signaling, reflects the potential anti-inflammatory and
   anti-fibrotic effects of DNase-I@PDA NPs and Siv@PLGA NPs. This outcome
   underscores their potential role in controlling fibrosis and
   facilitating lung tissue recovery in fibrotic conditions (Fig.
   [110]3C).

   Pertaining to the impact of the nanoparticles on neutrophil activation
   and the inflammatory mediators released by neutrophils, the combined
   administration of DNase-I@PDA NPs and Siv@PLGA NPs was associated with
   a marked reduction in neutrophil infiltration in the BALF (Fig.
   [111]3D). This reduction was accompanied by a significant decrease in
   NET formation in the lungs (Fig. [112]3E), demonstrating the efficacy
   of the treatment in suppressing excessive neutrophil activity.
   Significantly, DNase-I enzymatic activity remained stable across the
   treatment period, confirming that nanoparticle formulation did not
   impair its functionality (Fig. [113]3F). Co-administration with
   Siv@PLGA NPs led to a further reduction in NE and MPO levels, which may
   have contributed to NET inhibition and enhanced the anti-inflammatory
   effect. Since NE and MPO are associated with extracellular matrix
   degradation and fibrosis progression, this reduction may have played a
   role in modulating these processes (Fig. [114]3G, [115]H). The combined
   treatment also led to a decrease in circulating cfDNA levels, which are
   considered markers of immune dysregulation and have been linked to
   chronic inflammation and fibrosis (Fig. [116]3I and Figs. S10, 11). The
   reduction in NE and MPO levels, along with the decrease in cfDNA in
   both plasma (Fig. S12) and BALF, indicates that co-administration may
   help regulate acute inflammatory responses and limit pathways
   associated with fibrotic progression.

Relieved pathological phenotypes of pulmonary fibrosis in vivo

   To evaluate the therapeutic efficacy of DNase-I@PDA NPs and Siv@PLGA
   NPs in the context of pulmonary fibrosis, we conducted detailed
   histopathological and functional assessments. Histological analysis
   using H&E staining revealed that NP-treated mice exhibited significant
   restoration of alveolar structure and a marked reduction in pneumonic
   damage compared to untreated LPS-exposed mice (Fig. [117]4A). Orange G
   staining revealed pronounced extracellular matrix deposition in
   LPS-treated lungs, with NP demonstrating superior efficacy in reducing
   matrix accumulation compared to D+S group, as evidenced by histological
   patterns most closely resembling PBS controls. (Fig. [118]4B). Cit-H3
   and NE, well-established markers of NETs, were analyzed in pulmonary
   vascular tissue to assess the efficacy of nanoparticles in inhibiting
   NET accumulation and their potential role in preserving vascular
   integrity via IHC staining. In LPS group, prominent Cit-H3 and NE
   signals were observed within the vasculature, indicating persistent
   intravascular NET accumulation, inefficient NET clearance, and
   potential endothelial dysfunction driven by neutrophil-mediated
   cytotoxicity. This accumulation may indicate sustained neutrophil
   activation and prolonged inflammatory signaling, contributing to
   endothelial damage. In contrast, D+S and NP groups exhibited markedly
   reduced Cit-H3 and NE deposition in the vasculature, suggesting that
   the treatment effectively mitigated neutrophil hyperactivation and
   excessive NET release. This reduction in intravascular NET burden
   implies improved NET clearance mechanisms and decreased
   neutrophil-induced endothelial injury. These findings highlight the
   therapeutic potential of DNase-I@PDA NPs and Siv@PLGA NPs to attenuate
   neutrophil-driven vascular inflammation and improve (Fig. [119]4C).
   Trichrome staining further confirmed a reduction in collagen
   deposition, with both the area and intensity of collagen expression
   significantly diminished in NP-treated group, indicating that the
   treatment effectively limited collagen accumulation, a hallmark of
   fibrosis (Fig. [120]4D). Another IHC analysis demonstrated reductions
   in the expression of collagen I and α-SMA in the lungs of NP-treated
   mice. The decreased collagen I level suggests reduced vascular
   inflammation and fibrotic remodeling, helping to preserve lung
   architecture by preventing vascular occlusion. Additionally,
   suppression of α-SMA expression indicates that NP treatment effectively
   inhibited myofibroblast activation, a critical process in extracellular
   matrix remodeling and tissue stiffening (Fig. [121]4E). The reduction
   in collagen I and α-SMA expression was also quantitatively confirmed
   through real-time PCR analysis (Fig. S13). To further corroborate these
   findings, we measured hydroxyproline levels as a proxy for collagen
   synthesis. Hydroxyproline levels were significantly lower in the
   NP-treated group, suggesting reduced fibrotic activity and a decreased
   risk of progressive fibrosis (Fig. [122]4F). These results suggest that
   the treatment mitigates the pathological remodeling associated with
   fibrosis by maintaining structural integrity within the lungs.

Fig. 4.

   [123]Fig. 4
   [124]Open in a new tab

   Improved respiratory function by administrating of DNase-I@PDA NPs and
   Siv@PLGA NPs. A H&E analysis of lung tissue from LPS-induced pulmonary
   fibrosis mouse model. (Scale bar: 200 μm). B Orange G staining images.
   (Scale bar: 200 μm). C Immunohistochemistry (IHC) analysis (Cit-H3, NE,
   and Sytox-Green™) of lung sections of LPS-induced pulmonary fibrosis
   mouse model. (Scale bar: 40 μm). D Trichrome stained images of murine
   lung tissue. (Scale bar: 200 μm). E IHC analysis of fibrotic markers
   (collagen I and α-SMA). (Scale bar: 100 μm). F Measurement of
   hydroxyproline content. G Inspiratory capacity, lung resistance,
   compliance, and elastance ascertained by FlexiVent forced oscillation
   technique at endpoint. H PV curve. I Body weight of LPS-induced
   pulmonary fibrosis mouse model. J Survival rate of LPS-induced
   pulmonary fibrosis mouse model. Statistical analysis was performed
   using a two-tailed unpaired t-test. Data are presented as mean ± SEM.
   *P < 0.05, **P < 0.01, ****P < 0.0001 (v.s. LPS), ^##P < 0.01, ^####P <
   0.0001 (v.s. D + S)

   Pulmonary function assessments were conducted to investigate the
   physiological impact of NP treatment on respiratory mechanics in the
   LPS-induced lung injury model. LPS-treated mice exhibited a reduction
   in inspiratory capacity (IC), along with increased elastance (E) and
   decreased compliance (C), indicating impaired lung distensibility and
   increased tissue stiffness—features commonly associated with fibrotic
   remodeling (Fig. [125]4G). A reduction in IC reflects a limitation in
   the volume of air drawn into the lungs during inspiration. Elevated
   elastance and reduced compliance further support the presence of
   restricted mechanical expansion, as these measures describe the lung’s
   resistance to inflation and its ability to stretch under pressure.
   These functional impairments were further reflected in the
   pressure–volume (PV) curves, which were shifted downward in the LPS
   group, indicating reduced lung inflation capacity under applied
   pressure (Fig. [126]4H). Subsequent administrations of NPs were
   associated with trends toward improvement, as indicated by a modest
   recovery in IC and gradual normalization of E and C parameters (Fig.
   [127]4G). Correspondingly, The PV curves of NP-treated mice showed an
   upward shift compared to those of the untreated group (Fig. [128]4H).
   Together, these results provide evidence that NP administration
   contributed to the recovery of pulmonary mechanics by mitigating the
   functional consequences of inflammation-induced fibrotic injury.

   In terms of overall physiological outcomes, NP-treated mice showed
   significant improvements in body weight maintenance and survival rates
   compared to untreated LPS-exposed controls. Mice in the NP-treated
   group experienced less weight loss and demonstrated a substantial
   increase in survival rates, underscoring the effectiveness of the
   treatment in mitigating the detrimental effects of LPS-induced lung
   injury and fibrosis (Fig. [129]4I, J). These results collectively
   demonstrate the potential of DNase-I@PDA NPs and Siv@PLGA NPs to
   attenuate both acute and chronic pathological features of pulmonary
   fibrosis, restoring lung structure and function while improving
   survival outcomes in treated animals.

Therapeutic efficacy of nanoparticles on neutrophils of COVID 19 patients

   After confirming the in vivo restorative effects of the
   co-administration of DNase-I@PDA NPs and Siv@PLGA NPs, we extended our
   investigation to assess the therapeutic potential of these
   nanoparticles on neutrophils isolated from COVID-19 patients. Our
   results indicated that both free drug formulations and
   drug-encapsulated NPs contributed to a reduction in circulating cfDNA
   levels, which have been associated with inflammatory responses and
   immune dysregulation in COVID-19 patients (Fig. [130]5A). When
   DNase-I@PDA NPs were introduced to plasma samples from COVID-19
   patients, we observed a notable enhancement in DNase-I enzymatic
   activity, suggesting that the nanoparticle formulation improved the
   degradation of extracellular DNA and enhanced the overall therapeutic
   potential in reducing cfDNA levels (Fig. [131]5B). Further analysis of
   neutrophil samples revealed that treatment with DNase-I@PDA NPs and
   Siv@PLGA NPs led to a substantial reduction in NET formation, as well
   as significant decreases in NE and MPO activity (Fig. [132]5C–E). These
   findings indicate that formulations of the nanoparticles are highly
   effective at suppressing neutrophil hyperactivation and limiting the
   downstream enzymatic processes that contribute to pulmonary tissue
   damage and inflammation. In addition, NP treatment markedly improved
   neutrophil viability (Fig. [133]5F), demonstrating the ability of the
   nanoparticles to stabilize immune cell function under inflammatory
   conditions. This improvement in neutrophil stability is particularly
   important in promoting immune homeostasis in the context of severe
   viral infections such as COVID-19. Collectively, these results suggest
   that DNase-I@PDA NPs and Siv@PLGA NPs exhibit strong therapeutic
   potential in mitigating the harmful effects of NETosis and reducing the
   risk of long-term fibrotic complications in COVID-19 patients.

Fig. 5.

   [134]Fig. 5
   [135]Open in a new tab

   Evaluation of NETosis following coadministration of free drugs and NPs
   in the blood of SARS-CoV-2 patients. A Circulating cfDNA. B DNase-I
   enzymatic activity. C NET formation index in PBMCs. D NE activity. E
   MPO activity. F Viability of the neutrophils. Statistical analysis was
   performed using a two-tailed unpaired t-test. Data are presented as
   mean ± SEM. *P < 0.05, ****P < 0.0001 (v.s. PBS), ^#P < 0.05, ^####P <
   0.0001 (v.s. D + S)

Discussion

   In this study, we demonstrated that the sequential co-administration of
   DNase-I@PDA NPs followed by Siv@PLGA NPs represents a potent
   therapeutic strategy for alleviating pulmonary inflammation and
   fibrosis. Our findings underscore the importance of early intervention,
   beginning with the observation that DNase-I levels significantly
   decrease immediately after infection, thereby promoting the formation
   of NETs and contributing to excessive inflammation. Early
   supplementation of DNase-I@PDA NPs supported extracellular DNA
   clearance in the lungs, leading to a reduction in NET formation and the
   fibrotic microenvironment. Subsequently, the administration of Siv@PLGA
   NPs targeted NE via the neutrophil-specific Ly6g antibody’s Fab region,
   contributing to the modulation of NE activity and an extended
   therapeutic effect. This two-step approach showed the potential to
   regulate neutrophil hyperactivation and limit the pathological
   progression toward pulmonary fibrosis (Figs. [136]1 and [137]2).

   The treatment was associated with a reduction in the expression of
   immune mediators such as NETs, NE, and MPO, which are produced by
   activated neutrophils and contribute to both acute and chronic
   inflammatory states (Fig. [138]3). In an in vivo fibrosis model, the
   sequential administrations of NPs were observed to reduce fibrotic
   lesions, suggesting a potential protective effect against tissue
   remodeling and injury (Fig. [139]4). The therapeutic effects were not
   limited to animal models, as the nanoparticle regimen also showed
   beneficial outcomes in neutrophils isolated from patients with COVID-19
   (Fig. [140]5). These results broaden the potential clinical relevance
   of this nanoparticle-based dual approach beyond preclinical models and
   suggest its possible utility in managing COVID-19-related pulmonary
   complications, including PASC.

   Monotherapy with DNase-I or Sivelestat has shown limited efficacy in
   addressing the multifaceted nature of pulmonary inflammation and
   fibrosis, particularly in the context of chronic conditions like lung
   fibrosis. Based on our findings, a sequential, multi-targeted approach
   appears to be more effective in modulating the progression of fibrosis.
   The dual mechanism of action provided by DNase-I@PDA NPs and Siv@PLGA
   NPs offers complementary mechanisms of action that engage distinct
   aspects of the inflammatory cascade. First, DNase-I@PDA NPs,
   administered directly to the lungs through aerosol delivery, facilitate
   the breakdown of extracellular DNA and NETs, which are known to
   contribute to sustained inflammation by forming structural and
   immunological barriers. Next, systemically delivered Siv@PLGA NPs
   target peripheral neutrophils, limiting their activation and reducing
   their subsequent infiltration into the lungs, which may help attenuate
   ongoing inflammatory responses even after neutrophil recruitment has
   occurred.

   This combined strategy contributes to the attenuation of acute
   inflammation and also helps limit long-term tissue remodeling and
   fibrosis associated with sustained neutrophil activation. By targeting
   both the upstream generation of NETs and the downstream effects of
   neutrophil hyperactivation, the approach offers a broader means of
   addressing pulmonary injury in both acute and chronic phases. Notably,
   the sequential administration of DNase-I and Sivelestat appeared to
   enhance therapeutic outcomes beyond those observed with either agent
   alone, providing both early and prolonged modulation of inflammatory
   and fibrotic responses. From a clinical perspective, Sivelestat,
   although only approved in select regions (China, Japan, and Korea) for
   treating pulmonary dysfunction [[141]25], has demonstrated potent
   anti-inflammatory properties, supporting its broader applicability in
   conditions such as PASC. Given that neutrophils play a pivotal role in
   the prolonged inflammatory response observed in PASC [[142]26], our
   sequential approach of DNase-I followed by Sivelestat is designed to
   directly engage this pathophysiological mechanism. The initial
   clearance of NETs by DNase-I, followed by the inhibition of NE activity
   by Sivelestat, contributes to reducing inflammation and limiting
   further tissue injury. This, in turn, lowers the likelihood of fibrotic
   progression and supports the maintenance of pulmonary function over
   time. Additionally, the combined delivery strategy offers a targeted
   means of addressing both the clinical manifestations and the underlying
   drivers of chronic pulmonary fibrosis associated with PASC.

   In this context, the dual approach offers a cohesive therapeutic
   framework that has the potential to address both the acute inflammatory
   response and the subsequent fibrotic remodeling observed in
   COVID-19-related pulmonary complications. By sequentially targeting key
   neutrophil-driven mechanisms, the co-administrations of these
   nanoparticle formulations present a therapeutically relevant strategy
   that may be applicable not only to PASC but also to other fibrotic lung
   conditions characterized by persistent inflammation. Taken together,
   these findings provide a basis for further exploration of
   neutrophil-targeted nanotherapies in the broader context of chronic
   pulmonary disease management.

Conclusions

   COVID-19-associated pulmonary fibrosis remains a severe long-term
   complication, urgently requiring effective treatments. This study
   presents an innovative nanoparticle-based therapy targeting both
   neutrophil activation and NETs) formation. Sequential administration of
   DNase-I-loaded polydopamine nanoparticles and Sivelestat-encapsulated
   PLGA nanoparticles significantly reduced pulmonary fibrosis by clearing
   NETs and inhibiting NE. This approach addresses both acute inflammation
   and chronic fibrosis, improving lung function. Our findings open new
   possibilities for treating various lung diseases, including PASC, and
   represent a significant milestone in translating nanomedicine to
   clinical applications. This dual-action nanoparticle strategy offers a
   promising path to prevent long-term fibrotic progression and promote
   lung recovery in COVID-19 patients and beyond.

Supplementary Information

   [143]Supplementary Material 1.^ (2.8MB, docx)

Acknowledgements