Engineering a biocompatible chitosan/MOF aerogel for hyperbilirubinemia treatment in liver failure

Abstract

Liver failure, marked by a rapid or progressive decline in liver function, leads to complications such as hyperbilirubinemia and infections, contributing to approximately 2 million global deaths each year. Current hemoperfusion methods for hyperbilirubinemia treatment are limited by their low adsorption selectivity and inadequate biosafety, necessitating better treatments. In this study, a chitosan-based HKUST-1 aerogel (HC) was developed as a highly selective, biosafe, and antibacterial adsorbent. The integration of HKUST-1 with chitosan formed an aerogel that achieved high bilirubin adsorption in vitro along with significant biosafety, anticoagulant properties, and antibacterial activity against Staphylococcus aureus and Escherichia coli. In an animal model of hyperbilirubinemia, the HC aerogel demonstrated 42.4% bilirubin clearance from whole blood. These findings suggest that the HKUST-1/chitosan aerogel is a promising candidate for liver failure treatment, effectively combining bilirubin adsorption with infection control in complex blood environments.

---

1. Introduction

Liver failure represents a critical medical condition characterized by rapid deterioration of hepatic function, invariably accompanied by life-threatening complications such as hyperbilirubinemia and systemic infections.^1–3 This dual pathology creates a vicious cycle, while impaired bilirubin metabolism leads to neurotoxicity and multi-organ damage,⁴ concurrent infections further exacerbate systemic inflammation and liver injury, resulting in mortality rates exceeding 90% in acute cases.^5,6 The clinical management of this syndemic presents a formidable challenge, as conventional therapies often fail to address both components simultaneously.

While hemoperfusion (HP) remains the standard intervention for hyperbilirubinemia,⁷ current adsorbents (e.g., carbon,⁸ resins,⁹ and nanomaterials^10–12) suffer from three key limitations: low selectivity: inefficient toxin removal in complex blood environments; poor biocompatibility: material shedding risks thrombocytopenia and secondary toxicity;¹³ and single-functionality: inability to concurrently address infections.

Their minimal therapeutic effects also render them ineffective in treating complex conditions involving elevated bilirubin levels and infections. Therefore, by integrating detoxification and infection control, our objective is to directly target the dual-pathology challenge in liver failure management, offering a transformative strategy to improve patient outcomes.

Metal–organic frameworks (MOFs) are advanced porous materials with unsaturated coordination sites.^14,15 HKUST-1 and its derivatives have been used as adsorbents for tetracycline, oxytetracycline, and rare ions.^16–18 Their high adsorption capacities and multiple active sites indicate their potential application in the adsorption of in vivo toxins. Additionally, HKUST-1 has been used as a vascular coating material that helps reduce thrombus formation,¹⁹ demonstrating good blood compatibility. Furthermore, HKUST-1 exhibits excellent antibacterial properties, as its antimicrobial metal ions and linkers can eliminate bacteria,^20,21 making it suitable for the treatment of hyperbilirubinemia combined with infections (such as liver failure). However, when used directly in powder form for blood perfusion, there is a risk of embolising small distal blood vessels, which limits its application. Chitosan (CS) is a cationic biopolymer derived from the partial deacetylation of chitin and has numerous applications in biomedicine such as wound healing, drug delivery, and tissue engineering.^22,23 Therefore, we selected CS as a carrier for HKUST-1 for HP applications.

In this study, we developed a chitosan-based HKUST-1 aerogel (HC) with high efficiency, biosafety, and antibacterial properties to treat liver failure (hyperbilirubinemia combined with infection) (Fig. 1). HKUST-1 was successfully incorporated into chitosan to prepare an aerogel with a graded porous structure, and it's in vitro adsorption capacity, biosafety, and antibacterial properties were verified. The effectiveness and safety of this material were further validated in an animal model of hyperbilirubinemia, demonstrating its strong adsorption capacity and biosafety. We believe that this multifunctional HKUST-1/CS aerogel can serve as an effective material for treating liver failure.

Fig. 1 Synthesis and blood perfusion application of HKUST-1/CS aerogels.

2. Experimental section

2.1. Materials

The chemical reagents used are as follows: Cu(NO₃)₂·3H₂O (CAS: 10031-43-2) from Aladdin; trimeric acid (CAS: 554-95-0) from Angene Chemical; anhydrous ethanol and acetic acid from Sinopsin Group; chitosan (CAS: 9012-76-4, 90% degree of deacetylation, low viscosity <200 mPa s, M_w = 4.06 × 10⁵ g mol⁻¹, Macklin); 50% glutaraldehyde (CAS: 111-30-8); bovine serum albumin (Biosharp); phosphate-buffer saline (PBS); cell counting kit 8 (CCK-8) reagent; and bilirubin (Solarbio).

2.2. Preparation of HKUST-1

HKUST-1 cells were prepared as previously described.²⁴ Firstly, 3.2 g of Cu(NO₃)₂·3H₂O was dissolved in 48 mL of pure water, followed by the addition of 1.2 g of PVP and stirring for 10 min. Subsequently, 1.52 g of H₃BTC was dissolved in a mixture of 48 mL ethanol and 48 mL dimethylformamide and combined with the previous mixture. The resulting mixture was stirred at room temperature for 10 min before being transferred to a Teflon-lined autoclave for recrystallisation at 120 °C for 12 h. The blue crystals formed upon cooling were washed multiple times with dimethylformamide (DMF) and ethanol solution, dried at 80 °C for 8 h, and finally stored in an oven at 60 °C.

2.3. Preparation of HKUST-1/CS aerogels

First, CS (2.0 g) was dissolved in 198 g of 1 wt% acetic acid at room temperature for 4 h to prepare a 1 wt% CS solution. Then, 22.2, 50, and 85.7 mg of HKUST-1 powders were dispersed in pure water (500 μL) for 10 min by ultrasonic treatment to obtain various HKUST-1 dispersions. Finally, the dispersion was slowly dropped into a 20 g CS solution, then 60 μL GA was added and further stirred at room temperature for 30 min. HKUST-1/CS aerogels were obtained by placing the mixed solution in a 200 × 200 mm cylindrical mould by the solution pouring method, and thermal fixing at 60 °C for 1 h for a thorough cross-linking. Finally, the aerogels were obtained by freeze-drying at −30 °C. For ease of expression, HC1, HC2, and HC3 represent the following mass ratio of HKUST-1: 10 wt%; 20 wt%; and 30 wt%, respectively.

2.4. Characterisation of HKUST-1/CS aerogels

The surface morphology of the aerogel was observed using field emission scanning electron microscopy (Zeiss SIGMA, Zeiss, Germany). After chilling the sample with liquid nitrogen, a complete 2 mm thin slice was cut with a blade, coated with a thin layer of gold, and then sent into the scanning electron microscopy (SEM) chamber. The infrared spectra of the aerogels were obtained using a Fourier-transform infrared spectrometer (FTIR5700, Thermo Electron Corporation, USA). The crystal structure of the aerogel spheres was analysed using an X-ray diffractometer (SmartLab SE, Japan) with CuKα as the target, operating at 40 kV and 30 mA, with a 2θ scanning range of 5–60° at a scanning rate of 10° min⁻¹. The specific surface areas and pore size distributions of the aerogels were measured using a Brunauer–Emmett–Teller surface area and pore size analyser (TriStar II 3020, Micromeritics, USA). Thermogravimetric analysis was performed using a thermogravimetric analyser (TGA2/DSC3, Mettler Toledo, Switzerland) from 25 °C to 800 °C under N₂ gas at a heating rate of 10 °C min⁻¹ and 60 mL min⁻¹. X-ray photoelectron spectroscopy (XPS) analysis was conducted using an ESCALAB 250Xi spectrometer, and data analysis was performed using the XPSpeak software package. The mechanical properties of the cylindrical aerogel specimens were investigated using a universal material-testing machine (Mark-10). Standard specimens with a diameter of 18.1 ± 0.2 mm and a height of 6.7 ± 0.1 mm were subjected to quasi-static compression testing under a constant loading rate of 1 mm min⁻¹. Inductively coupled plasma-atomic emission spectroscopy (ICP-OES, Agilent 5110) was used to analyse the aerogels and supernatant copper ion concentration.

2.5. Bilirubin adsorption experiment

The HC3 aerogels (5 mg) were immersed in an aqueous solution of bilirubin (200 mg L⁻¹, 10 mL) to study the adsorption kinetics. The samples were then stirred at room temperature in the dark. At set time points (0.5, 1, 1.5, 2, 2.5, 3, and 3.5 h), the concentration of the bilirubin solution after adsorption was determined at 438 nm using a UV-VIS spectrophotometer. The adsorption quantity q_e was determined using the following equation.where q_e (mg g⁻¹) is the mass of bilirubin adsorbed per unit mass of the adsorbent and m (g) is the adsorbent mass. C₀ (mg L⁻¹) and C_i (mg L⁻¹) are the initial and final concentrations of the bilirubin solution, respectively, and V_i (L) is the volume of the bilirubin solution.

2.6. Isothermal adsorption

Similarly, the sample was added to the bilirubin solution at room temperature and in the dark for 3 h to reach equilibrium, with initial concentrations of 50, 100, 200, 300, 400, and 500 mg L⁻¹ to obtain the adsorption isotherms of the sample.

2.7. Coagulation assay

To investigate the anticoagulant effect, aerogels (400 μg) were incubated in 2 mL of fresh sodium citrate-treated non-coagulated blood, gently agitated at 37 °C for 2 h. The resulting mixture was centrifuged at 3000 rpm for 10 min to separate the plasma. The supernatant was analysed using a blood coagulation analyser (RAC-1830, Sevier, Wuhan, China) to evaluate the prothrombin time (PT), activated partial thromboplastin time (APTT), thrombin time (TT), and fibrinogen (FIB) levels. Anticoagulated plasma was used as the control group, and all experiments were conducted in triplicate to ensure robustness and reliability.

2.8. Haemolysis assay

The rabbit blood was anticoagulated with heparin and centrifuged at 2000 rpm for 10 min. The lower layer, which contained red blood cells, was washed three times with PBS. The red blood cells were then resuspended in 2% PBS, with PBS and deionised water used as negative and positive controls, respectively. Samples weighing 400 μg were immersed in 2 mL of the red blood cell–PBS solution and incubated at 37 °C for 2 h. After centrifugation at 1000 rpm for 10 min, the solution was observed using a digital camera. The absorbance of the supernatant obtained after centrifugation was analysed at 545 nm using a UV spectrophotometer to calculate the haemolysis rate given bywhere A_s, A_p, and A_n represent the absorbance values of the test sample, positive control, and negative control, respectively.

2.9. Cytotoxicity assay

UV-disinfected mixed aerogels were added to the RPIM-1640 complete medium (Wuhan procell) in advance, and the concentration was 200 μg mL⁻¹ for 24 h, after which the extraction solution was obtained. L929 cells were seeded in 96-well plates at a density of 1 × 10⁴ cells per well and incubated for 12 h. Then, 200 μL of the extract was added to each well and cultured at 37 °C under a 5% CO₂ atmosphere. The medium was changed every 48 h. The cytotoxicity of the aerogel complex was assessed using CCK-8, cell viability was assessed using a live/dead cell assay kit, and cell status was observed by inverted fluorescence microscopy.

2.10. Antimicrobial activity

In vitro antimicrobial activity was evaluated as previously described.²⁵Escherichia coli (E. coli) and Staphylococcus aureus (S. aureus) were cultured in AGAR/broth medium at 37 °C, and the organisms were diluted to 5 × 10⁵ CFU mL⁻¹ in PBS solution. Centrifugal tubes of CS and HC3 were prepared with concentrations of 200–400 μg mL⁻¹, with the control group at 0 μg mL⁻¹. A total of 100 μL of the microbial suspension was added to each centrifuge tube and incubated for 6 h in a constant temperature culture shaker, and then 100 μL of the mixed droplet sample was added to the prepared AGAR plate and incubated for 24 h at 37 °C. Finally, the AGAR plates were photographed, and the colony counts were determined using ImageJ software. The antibacterial rate was quantified as follows.where N_c is the control group count and N_a is the colony count after co-culture with aerogels.

2.11. Microorganism morphology assay

The control group was cultured in material-free medium, while the experimental group was incubated with 300 μg mL⁻¹ HC3 in bacterial suspension for 12 h, followed by centrifugation (10 000 rpm, 5 min) to discard the supernatant and collect bacterial pellets for subsequent experiments. Under 4 °C, the sample was soaked in 2.5% neutral glutaraldehyde solution for 12 h, then dehydrated in ethanol of different concentrations (20%, 40%, 60%, 80%, and 100%) for 10 min.²⁶ This was followed by another 10 min treatment with tert-butyl alcohol. The specimens were dried in a freeze dryer and microorganisms were observed using a scanning electron microscope (Zeiss SIGMA).

2.12. Establishment of a high bilirubin model in rabbits

All experimental animals were cared for in accordance with the ARRIVE guidelines and in strict compliance with the National Research Council's Guide for the Care and Use of Laboratory Animals. The Animal Care and Use Committee of Wuhan University approved all animal experimental procedures. The bile duct ligation model in rodents is well established to simulate hyperbilirubinemia in patients with liver failure.²⁷ Anaesthesia was administered via the marginal ear vein. The anaesthetic used was 1% sodium pentobarbital, and heparinisation of rabbit blood was performed through the marginal ear vein at a dose of 300 IU per kg. Prior to surgery, the animals were fasted for 12 h. A midline abdominal incision was made in the rabbits to expose and ligate the common bile duct (approximately 2.5–3.0 mm in diameter) in the duodenal bulb. The distal and proximal ends were ligated and severed bilaterally. Total bilirubin levels were measured using a fully automated biochemical analyser (Sevier Company). The HC3 adsorbent was packed into the adsorption column at a dosage of 150 mg kg⁻¹. Subsequently, a tube with a length of 25 cm and diameter of 5 mm was connected, and the temperature was controlled at 36.3 °C using an external temperature control device. The tube was then connected to a peristaltic pump (Langer Pump, BT100-2J) and the pump speed was set to 4 r per s. Prior to perfusion, the cells were rinsed with heparin.

2.13. Statistical analysis

All experiments were repeated at least three times, and the data are expressed as mean and standard deviation. The statistical significance of differences between groups was analysed using a two-tailed Student's t-test and one-way analysis of variance. Statistical analyses were performed using SPSS 19 software, with p < 0.05 (*) as the level of significance.

3. Results and discussion

3.1. Structural and chemical characterisation of HKUST-1/CS aerogels

The morphology of the prepared aerogels was characterised by SEM. The surface and cross-section of the CS aerogel presented an irregular lamellar structure (Fig. 2a), and the stacked lamellae formed loose pores, which are large enough for the liquid to flow quickly. The octahedral shape of HKUST-1 (Fig. 2b) was consistent with the originally synthesised form (Fig. S1, ESI†), indicating that the incorporation of MOFs did not collapse the aerogel because of the dissolution of chitosan.

Fig. 2 Structural and chemical characterisation of HKUST-1/CS aerogels. (a) and (b) SEM of chitosan and HKUST-1/CS aerogels cross-section. (c) SEM images and corresponding EDS images of HKUST-1/CS aerogels. (d) XRD patterns of CS, HKUST-1, and HC3. (e) and (f) FT-IR and TG analysis of CS, HKUST-1, and HC3. (g) Chemical composition of CS and HC3 as characterised by XPS. (h) N₂ adsorption isotherms of HKUST-1 and HC3. (i) Pore size distribution of HC3.

As shown in Fig. 2b, HKUST-1 is uniformly dispersed across the aerogel surface. Energy dispersive X-ray spectroscopy (EDS) analysis of the aerogel cross-section (Fig. 2c) confirmed the presence of the characteristic Cu element from HKUST-1 along with the N-phase of chitosan, further verifying the successful incorporation of the MOF into the chitosan matrix. The Cu content on the HC3 surface was determined to be 23 wt% by EDS (Fig. S2, ESI†). Inductively coupled plasma mass spectrometry of the entire aerogel revealed a bulk Cu content of 61 270.56 ppm HC3 (Table S1, ESI†).

HC3 was further analysed using XRD (Fig. 2d). CS had a typical non-diffraction peak near 2θ = 21.1°, indicating its low crystallinity.²⁸ The characteristic peaks of HKUST-1 appeared at 6.7°, 9.6°, 11.6°, 19.0°, and 20.2°, corresponding to its (200), (220), (222), (440), and (600)²⁹ planes, respectively. The XRD pattern of HC3 contained the characteristic peaks of CS and HKUST-1, demonstrating the successful incorporation of HKUST-1 crystals into the chitosan aerogel.

The successful synthesis of both the MOF and chitosan/MOF composite aerogels was further confirmed by infrared spectroscopy. As shown in Fig. 2e, the FTIR spectrum of HKUST-1 exhibited the characteristic peaks of MOFs. The absorption peak at 1622 cm⁻¹ is attributed to the asymmetric stretching vibration of carboxylate groups (–COOH), indicating the coordination between organic ligands and metal centres. The symmetric stretching vibration of carboxylate groups (–COOH) appears at 1557 cm⁻¹, further verifying the formation of metal–ligand coordination bonds.^30,31 The peak at 1443 cm⁻¹ can be assigned to either C=C stretching vibrations of benzene rings or C–H in-plane bending vibrations, reflecting the characteristic vibrational modes of the conjugated aromatic system.³² Additionally, the characteristic peaks at 718 and 499 cm⁻¹ correspond to Cu–O stretching vibrations,³³ which collectively demonstrate the existence of metal–ligand coordination bonds in the HKUST-1 framework. For chitosan, the broad peak in the 3450–3200 cm⁻¹ range is attributed to O–H and N–H stretching vibrations,³⁴ reflecting hydrogen bonding interactions between the hydroxyl and amino groups. The absorption peak near 2865 cm⁻¹ corresponds to aliphatic C–H stretching vibrations.³⁵ The FT-IR spectrum of the HC3 composite retains the characteristic peaks of HKUST-1, including Cu–O bonds (718 and 499 cm⁻¹) and carboxylate vibrations (1622 and 1557 cm⁻¹), confirming the intact MOF structure. Meanwhile, the presence of CS characteristic peaks (O–H/N–H [3450–3200 cm⁻¹], C–H [2865 cm⁻¹], and amide I band [1611 cm⁻¹]) demonstrates the successful incorporation of chitosan as the matrix. Furthermore, the appearance of copper in the XPS spectrum of HC3 (Fig. 2f) further confirms the successful synthesis of the composite material.

Thermogravimetric analysis (Fig. 2g) revealed two weight loss stages: moisture loss (20–170 °C) and degradation of chitosan/decomposition of H3BTC (170–350 °C).¹⁸ HC3 exhibits excellent thermal stability and remains intact at temperatures above those of the human body.

N₂ adsorption/desorption experiments were performed to characterise the porosity and specific surface area of MOF aerogels. As shown in Fig. 2h, the specific surface area of HKUST-1 was 654.2 m² g⁻¹, which followed a Type I isotherm.³⁶ The specific surface area of HC3, based on chitosan loaded with HKUST-1, reached 107.4 m² g⁻¹, which is three times that of previously reported CS/graphene composites.³⁷ The average pore size is 10.9 nm (Fig. 2i), which corresponds to a type-IV isotherm. This structure provided sufficient channels for bilirubin binding. The pore distribution within the first 3 nm was comparable to that of pure HKUST-1 (Fig. S3, ESI†), indicating that the pore structure remained intact after the incorporation of chitosan. This effectively ensured mass diffusion and transfer, thereby promoting the adsorption of bilirubin. These results demonstrate the successful synthesis of porous HC aerogels, which provide the basis for efficient bilirubin adsorption. The mechanical stability of aerogels plays a crucial role in the adsorption process. The mechanical properties were further characterised using loading and compression tests. As shown in Fig. S4(a) (ESI†), both pure CS and HC3 aerogels exhibited negligible deformation under a 250 g weight load, demonstrating that their rigid structure effectively maintains morphological integrity in blood environments to enable adsorption functionality.^38,39 The compressive stress–strain curves (Fig. S4b and c, ESI†) revealed that at 30% compressive strain, the CS aerogel sustained 0.055 MPa stress, whereas the MOF-loaded HC3 aerogel retained 0.052 MPa stress under identical strain conditions (a mere 5.5% reduction). These comparative results confirmed that MOF loading did not considerably compromise the mechanical properties, with the HC3 aerogels demonstrating exceptional mechanical stability.

3.2. Adsorption capacity of HC3 for bilirubin

To address the challenges of recyclability and biotoxicity associated with pure MOFs,^40–42 a low-loading MOF/chitosan composite aerogel (MOF content <30 wt%) was used to balance safety and adsorption performance. At a fixed bilirubin concentration (300 mg g⁻¹), a systematic comparison of the adsorption capacities of CS, HC1, HC2, and HC3 revealed a significant positive correlation between the adsorption capacity and MOF loading. HC3 (30 wt% MOF loading) demonstrated optimal performance (Fig. S5, ESI†), achieving a bilirubin adsorption capacity of 210.35 ± 5 mg g⁻¹, which is 6.7-fold higher than that of pure chitosan (31.61 ± 8 mg g⁻¹). Further adsorption kinetics experiments elucidated the mechanism: HC3 exhibited rapid adsorption within 0.5–3.5 h, reaching 164.5 mg g⁻¹ within 30 min and equilibrium (q_e = 173.01 mg g⁻¹) within 1 h; whereas CS required >2 h to attain a lower q_e of 43.17 mg g⁻¹ (Fig. 3a). Pseudo-second-order kinetic analysis (Fig. 3b, c and Table S2, ESI†) showed that HC3 exhibited superior fitting (R² = 0.999) and a higher rate constant (K₂ = 0.0384 g mg⁻¹ h⁻¹) compared to that of CS (R² = 0.9284, K₂ = 0.0205 g mg⁻¹ h⁻¹), indicating that MOF incorporation enhanced adsorption efficiency via chemisorption, with a 53.4% increase in K₂ and a 69.7% improvement in q_e.⁴³

Fig. 3 Bilirubin adsorption curve. (a) Adsorption kinetics of HC3 and CS on bilirubin adsorption. (b) and (c) Linear fitting curves of (b) primary kinetic and (c) secondary kinetic models of HC3 and CS. (d) Nonlinear Langmuir adsorption isotherm and nonlinear Freundlich adsorption isotherm of HC3 for bilirubin adsorption. (e) Freundlich adsorption isotherm. (f) Langmuir adsorption isotherm.

The bilirubin isothermal adsorption curve of HC3 (Fig. 3d) demonstrated exceptional concentration adaptability, with q_e increasing from 42.1 to 241.0 mg g⁻¹ as the initial bilirubin concentration increases. Langmuir and Freundlich model fittings (Fig. 3e, f and Table S3, ESI†) revealed that the HC3 adsorption data aligned better with the Langmuir model (R² = 0.998) than CS (R² = 0.89), yielding a theoretical maximum adsorption capacity (q_max) of 561.79 mg g⁻¹, which is 15.5-fold higher than that of CS (36.76 mg g⁻¹). Freundlich analysis further indicated multilayer adsorption characteristics for HC3 (1/n = 0.69, n > 1),⁴⁴ contrasting with the near-monolayer behaviour of CS (1/n = 0.22, n ≈ 1).⁴⁵ This hierarchical adsorption mechanism arises from the pore structure synergy: MOF micropores (0.4–2.0 nm) provide high-density active sites;⁴⁶ while chitosan macropores facilitate molecular diffusion,⁴⁷ collectively affirming HC3 as a robust bilirubin adsorbent.

3.3. HC3 interaction with bilirubin

FT-IR spectroscopy revealed a distinct characteristic peak at 422.8 cm⁻¹ (Fig. 4a) after bilirubin adsorption, which is attributed to the asymmetric stretching vibration of the Cu–N bond,^48,49 indicating coordination between the pyrrolic nitrogen of bilirubin and the HC3 moiety.

Fig. 4 Proposed bilirubin adsorption mechanism of HC3. (a) FT-IR spectra of HC3 before and after bilirubin adsorption. (b) XPS survey scan. High-resolution XPS spectra of (c) Cu 2p, (d) N 1s, (e) O 1s, and (f) C 1s.

XPS characterisation further corroborated this coordination mechanism (Fig. 4c). The Cu 2p_3/2 binding energy decreased from 933.9 to 933.4 eV (ΔE = 0.5 eV), reflecting a significant increase in the electron density of Cu²⁺. This phenomenon originates from the donation of lone pair electrons from pyrrolic nitrogen to the metal centre, inducing a contraction of the ionic radius and an increase in the coordination number (transitioning from a four-coordinate tetrahedral configuration to a five-coordinate square-planar geometry).⁵⁰ Concurrently, the N 1s binding energy shifted upward to 402.3 eV (ΔE = 2.7 eV), corresponding to a partial loss of electron density at the nitrogen atom (Fig. 4d). This observation aligns with the electron transfer process where lone pair electrons become confined within the vacant d-orbitals of Cu²⁺ upon coordination, consistent with predictions from ligand field effect theory. Additionally, the decreased binding energies of O 1s (532.8 to 531.2 eV) and C 1s (288.0 to 286.4 eV; Fig. 4e and f) suggest enhanced electron density at oxygen and carbon atoms, further confirming the chemical interactions (e.g., hydrogen bonding or covalent bond formation) between HC3 and bilirubin. These spectroscopic datasets collectively establish a coherent evidence chain: FT-IR reveals molecular vibrational signatures of Cu–N bonding, while XPS deciphers electronic redistribution (electron density gain at Cu and modulation at N/O/C sites), synergistically confirming that the adsorption process is governed by Cu–N coordination bonds, accompanied by complementary chemical interactions at oxygen and carbon sites.

3.4. Biosafety evaluation of HC3

Biosafety is a prerequisite for the safe application of materials for blood perfusion. The haemolysis rates of CS, HC1, HC2, and HC3 aerogels and HKUST-1 were 0.69%, 1.07%, 1.25%, 2.71%, and 63.00%, respectively (Fig. 5a). The introduction of chitosan significantly reduced the haemolysis of MOFs, and the haemolysis rate of HC3 aerogels was also below 5% (ISO safety standard).⁵¹ Adsorption of albumin by the material may cause pore obstruction,⁵² thereby reducing bilirubin adsorption. Second, a decrease in protein levels in the body may cause an imbalance in the internal environmental homeostasis and systemic oedema.⁵³ Therefore, the protein adsorption must be verified. Using 40 g L⁻¹ protein stock solution as the control group, CS, HC1, HC2, HC3, and HKUST-1 were co-cultured in 40 g L⁻¹ PBS. No significant change in albumin concentration was observed (Fig. 5b). Thus, HC3 has good blood compatibility and can be used for whole-blood perfusion. In this study, the copper ion concentration in the supernatant after 2 h of in vitro simulated adsorption was calculated using the equation.

Fig. 5 Biocompatibility evaluation. (a) Haemolysis rate test. (b) Albumin adsorption assay. (c) Cell proliferation after 1 and 2 d of culture material extraction. (d) Fluorescence micrographs of AM/PI-stained cells (green: live cells, red: dead cells). (e) Apoptotic effects of different extracts (CS and HC3) on L929 cells. (f) Coagulation function (mean ± SD, n = 3 for each, *p < 0.05, **p < 0.01).

ICP-MS showed that the maximum release of HC3 was 0.03795 mg L⁻¹ (Fig. S7, ESI†), which was significantly lower than that of the normal human blood copper concentration range (0.7–1.42 mg L⁻¹).⁵⁴ This indicates that even at the highest concentration, the released Cu remained far below normal physiological levels, suggesting low toxicity risk.

To observe the cytotoxicity of HC3, the cytotoxicity of MOF aerogels was explored in mouse fibroblasts (L929) in vitro. HC3 cells exhibited a cell survival rate of 98.03% after 2 d, as determined by the CCK-8 assay (Fig. 5c). The live/dead cell viability assay results (Fig. 5d) showed that the number of cells increased with an increase in the number of culture days, and cell proliferation was normal. To observe cell survival, flow cytometry was used to detect apoptosis. The results showed that, compared with the control (0.48% and 6.82%) and CS groups (0.81% and 8.29%), the proportions of early and late apoptotic cells in L929 cells incubated with HC3 were only 1.66% and 10.2%, respectively (Fig. 5e), indicating that HC3 did not induce apoptosis in most of the L929 cells. The coagulation function of the material has an explanatory value in ensuring HP. Coagulation tests showed that HC3 aerogels exhibited no significant differences in PT, international normalised ratio, TT, and FIB values (Fig. 5f), but exhibited an increase in APTT. The modest APTT prolongation induced by HC3 likely reflects its surface-mediated anticoagulant properties, which are advantageous for preventing thrombosis in perfusion systems without requiring systemic heparinization (as self-anticoagulant hemoperfusion adsorbents⁵⁵). In conclusion, HC3 does not induce apoptosis and exhibits mild anticoagulant properties, making it advantageous for HP.

3.5. Antibacterial activity of HC3

In patients with liver failure, hyperbilirubinemia and infection form a vicious cycle that exacerbates each other. Liver failure considerably increases the risk of infection through Kupffer cell dysfunction, reduced complement synthesis, and intestinal barrier (due to portal hypertension-induced endotoxin translocation and bile acid deficiency-induced dysbiosis).⁵⁶ Hyperbilirubinemia further suppresses neutrophil phagocytic function,⁵⁷ induces T lymphocyte apoptosis,⁵⁸ and activates the TLR4/NF-κB pathway, aggravating the inflammatory storm. Meanwhile, infection worsens the hepatic microcirculatory dysfunction and oxidative stress by releasing pro-inflammatory cytokines (TNF-α and IL-6),⁵⁹ further increasing treatment difficulty. Therefore, combining blood purification therapy (such as HP) with antimicrobial/anti-infective treatment can break the ‘liver injury–hyperbilirubinemia–infection’ vicious cycle, thereby improving patient prognosis.

The antibacterial effects of CS and HC3 aerogels against the liver failure-associated pathogens S. aureus and E. coli were evaluated using the plate counting method (Fig. 6a). The results revealed a concentration-dependent antibacterial effect of CS: as shown in Fig. 6b, the antibacterial rate against E. coli increased progressively with concentration (200 μg mL⁻¹–27.3%; 300 μg mL⁻¹–30.7%; and 400 μg mL⁻¹–51.6%), whereas Fig. 6c demonstrates more significant inhibition against S. aureus (200 μg mL⁻¹–21.4%; 300 μg mL⁻¹–46.8%; and 400 μg mL⁻¹–70.1%).

Fig. 6 In vitro antibacterial evaluation. (a) Antibacterial coated plates of CS and HC3 against different concentrations of S. aureus. (b) and (c) Inhibition rate against (b) E. coli and (c) S. aureus. (d) SEM images of S. aureus and E. coli (mean ± SD, n = 3 for each group, *p < 0.05, **p < 0.01).

The synergistic antibacterial enhancement of HC3 was demonstrated by its significantly superior inhibitory effects against both bacterial strains compared with CS at equivalent concentrations. HC3 exhibited antibacterial rates against E. coli ranging from 76.7% at 200 μg mL⁻¹ to 99.6% at 400 μg mL⁻¹, while achieving 98.1% inhibition against S. aureus even at 200 μg mL⁻¹ (Fig. 6b and c). SEM observations (Fig. 6d) further elucidated the bactericidal mechanism of HC3. Untreated S. aureus displayed a smooth spherical morphology, and E. coli maintained intact rod-shaped structures, whereas HC3-treated bacteria exhibited surface wrinkling, pore formation, and structural collapse, confirming its potent antibacterial action. These results demonstrate that the HC3 aerogel possesses broad-spectrum and highly effective antimicrobial activity against both Gram-positive (e.g., S. aureus) and Gram-negative (e.g., E. coli) bacteria commonly encountered in patients with liver failure. Combined with its bilirubin-eliminating capacity, HC3 significantly reduces the risk of secondary infections during HP, offering an innovative therapeutic strategy for complex clinical management.

3.6. Bilirubin adsorption in rabbit models of HC3

In vivo animal experiments can simulate the dynamic environment of human blood flow velocity, pressure, and haemodynamic,⁶⁰ allowing the evaluation of the material adsorption efficiency under real blood flow conditions. A common bile duct ligation (BDL) model was employed to assess the efficacy and safety of bilirubin-adsorbing materials (Fig. S8, ESI†). Seven days post-BDL, rabbit serum total bilirubin (TBIL) levels peaked at 39.8 μmol L⁻¹ (Fig. 7a), after which the HC3 aerogel was used for in vivo perfusion therapy. During perfusion, the TBIL concentration rapidly decreased within 0.5 h, reaching adsorption equilibrium at 1 h and declining to 22.9 μmol L⁻¹ after 2 h (p < 0.01), achieving a clearance rate of 42.4%. This performance is not inferior to the clearances of UIO-66 reported in the literature.⁶¹

Fig. 7 Animal blood perfusion evaluation. (a) Dynamic monitoring of TBIL, IDBIL, and DBIL. (Mean ± SD, n = 5, **p < 0.01). (b) Dynamic protein monitoring. (c) Inorganic ion detection. (d) Blood cell detection. (e) Histological examination of the heart, liver, spleen, lungs, and kidneys after HC3 aerogel perfusion in rabbits.

Synchronous monitoring revealed that indirect bilirubin (IDBIL) decreased from 23.6 to 11.2 μmol L⁻¹, demonstrating a significantly greater reduction in IDBIL (52.5%) compared to that of direct bilirubin (27.7% reduction). This indicates that the superior selective clearance capability of HC3 for IDBIL is likely attributable to its surface coordination sites and a hydrophobic microenvironment, which exhibit a strong affinity for lipid-soluble IDBIL.⁶² These results confirm that the HC3 aerogel can effectively eliminate multiple bilirubin species in complex physiological environments. Notably, its exceptional IDBIL clearance capacity provides a novel adsorption strategy that combines broad-spectrum efficacy and molecular specificity for the treatment of liver failure-associated hyperbilirubinemia.

Considering the complexity of blood components and potential risks of nonspecific adsorption, this study systematically evaluated the impact of the HC3 aerogel on key blood constituents. As shown in Fig. 7b, HC3 perfusion caused no significant fluctuations in concentrations of inorganic ions (Na⁺, K⁺, Ca²⁺; p > 0.05), demonstrating effective avoidance of ion exchange interference through material design. Albumin and total protein, which are critical components for maintaining plasma colloid osmotic pressure,⁶³ may cause oedema or material pore biofouling upon abnormal adsorption, subsequently hindering bilirubin diffusion.⁶⁴ Quantitative analysis (Fig. 7c) revealed stable levels of total protein (41.33 to 40.28 mg L⁻¹) and albumin (26.91 to 26.34 mg L⁻¹) after 2 h perfusion (p > 0.05), confirming the superior selective adsorption capability of HC3. Further haematological analysis (Fig. 7d) showed that all parameters (erythrocyte/leukocyte counts and neutrophil/lymphocyte ratios) remained within normal ranges without statistical differences, excluding the risks of acute haemolysis, immune activation, or inflammatory responses.^65,66

For a comprehensive biosafety assessment, histopathological examinations of Hematoxylin and eosin (H&E)-stained sections were performed on major organs (heart, liver, spleen, lungs, and kidneys; Fig. 7e). The results demonstrated that: (1) the control group exhibited normal physiological structures; (2) BDL group presented mild portal fibrosis with neutrophil infiltration and focal hepatocyte ballooning degeneration (cytoplasmic vacuolisation), indicating minor cholestatic injury; and (3) the HC3-treated group showed no additional pathological changes (necrosis, fibrosis, or haemorrhage), with tissue architectures indistinguishable from controls. Collectively, these data and the H&E findings prove that HC3 achieves efficient bilirubin clearance through precise molecular recognition mechanisms, while maximally preserving blood homeostasis, thereby providing critical safety evidence for HP therapy.

4. Conclusion

This study successfully developed a bifunctional chitosan/HKUST-1-based aerogel (HC3) that achieves efficient bilirubin adsorption in vitro through Cu–N coordination, while demonstrating broad-spectrum antimicrobial activity against liver failure-associated pathogens with a maximum inhibition rate of 99.6%. In a rabbit model of BDL, HC3-mediated whole blood perfusion achieved 42.4% bilirubin clearance, with selective removal of lipid-soluble indirect bilirubin while maintaining blood homeostasis. Its hierarchical porous structure, controlled copper ion release, and mechanical stability address the limitations of traditional adsorbents with respect to low selectivity and high biotoxicity, offering an innovative therapeutic strategy that integrates detoxification and anti-infection functions for liver failure complicated by infection.

However, this study still presents several limitations that warrant discussion. To mitigate environmental hazards, we implemented a vacuum recovery system for DMF (85 °C/−0.095 MPa) during HKUST-1 synthesis. Nevertheless, the development of truly sustainable solvents (e.g., water/ethanol systems) remains imperative to advance green fabrication protocols. While the study has convincingly demonstrated bilirubin clearance performance through well-characterized Cu–N coordination mechanisms (as evidenced by XPS/FTIR analyses), we acknowledge that molecular dynamics (MD) simulations could provide crucial atomic-scale insights into adsorption dynamics. However, conducting accurate MD simulations of our system presents notable technical challenges, including: (1) the current lack of appropriate force field parameters for modeling the hybrid organic–inorganic interface of HC3, and (2) substantial computational limitations in accurately simulating the complex blood–material interaction environment.

Notwithstanding these limitations, this dual-functional therapeutic strategy—which simultaneously addresses bilirubin removal and infection control—represents a significant advancement toward the clinical translation of next-generation blood purification systems. The approach holds particular promise for improving treatment outcomes in liver failure patients, where concurrent management of hyperbilirubinemia and infection remains a critical unmet clinical need.

Author contributions

Zhiwen Hu: writing the original draft, software, methodology, data curation, and formal analysis. Fangjiu He: validation, software, and methodology. Shao rui Long: software and validation. Xiaowen Shi: investigation, supervision, and writing – review, and editing. Qifa Ye: investigation, supervision, and writing, review, and editing. Zibiao Zhong: supervision, project administration, funding acquisition, conceptualisation, investigation, resources, validation, and writing – review, and editing.

Conflicts of interest

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

Data availability

All experimental data and analytical results from this study are provided in the ESI.† Raw data are available from the corresponding author upon reasonable request.

Acknowledgements

This project was financially sponsored by the National Natural Science Foundation of China (No. 82370758 and 82470685); the Health Commission of Hubei Province Scientific Research Project; and the Hubei Provincial Elite Program – Technological Innovation Team. The authors thank the School of Resource and Environmental Sciences, Wuhan University, for their experimental support.

References

1. A. Cheung, S. Tanna and M. G. Ison, Infectious complications in critically ill liver failure patients, Semin. Respir. Crit. Care Med., 2018, 39, 578–587,  DOI:10.1055/s-0038-1673657.
2. S. Uemura, R. Higuchi, T. Yazawa, W. Izumo, T. Otsubo and M. Yamamoto, Level of total bilirubin in the bile of the future remnant liver of patients with obstructive jaundice undergoing hepatectomy predicts postoperative liver failure, J. Hepatobiliary Pancreat Sci., 2020, 27, 614–621,  DOI:10.1002/jhbp.784.
3. L. Yang, T. Wu, J. Li and J. Li, Bacterial infections in acute-on-chronic liver failure, Semin. Liver Dis., 2018, 38, 121–133,  DOI:10.1055/s-0038-1657751.
4. S. Yin, Y. Xu, Z. Wang, Z. Wei, T. Xu, W. Zhao and C. Zhao, Molecularly-imprinted hydrogel beads via self-sacrificing micro-reactors as safe and selective bilirubin adsorbents, J. Mater. Chem. B, 2022, 10, 2534–2543,  10.1039/D1TB01895G.
5. R. Manikat, A. Ahmed and D. Kim, Current epidemiology of chronic liver disease, Gastroenterol. Rep., 2024, 12, goae069,  DOI:10.1093/gastro/goae069.
6. H. Devarbhavi, S. K. Asrani, J. P. Arab, Y. A. Nartey, E. Pose and P. S. Kamath, Global burden of liver disease: 2023 update, J. Hepatol., 2023, 79, 516–537,  DOI:10.1016/j.jhep.2023.03.017.
7. M. Sartori, A. Sharma, M. Neri, F. Garzotto, F. Nalesso, D. Giavarina, M. Zancato and C. Ronco, New option for the treatment of hyperbilirubinemia: In vitro direct hemoperfusion with the lixelle S-35, Int. J. Artif. Organs, 2014, 37, 816–823,  DOI:10.5301/ijao.5000363.
8. Z. Juan, Z. Yongke, Z. Yaping, Z. Qingyun and W. Siyu, Influence of the acid groups on activated carbon on the adsorption of bilirubin, Mater. Chem. Phys., 2021, 260, 124133,  DOI:10.1016/j.matchemphys.2020.124133.
9. M. Marcello and C. Ronco, Bilirubin Adsorption with DPMAS: Mechanism of Action and Efficacy of Anion Exchange Resin, Adsorption: The New Frontier in Extracorporeal Blood Purification, Books Gateway, Karger Publishers, https://karger.com/books/book/5532/chapter-abstract/12977938/Bilirubin-Adsorption-with-DPMAS-Mechanism-of?redirectedFrom=fulltext, (accessed May 28, 2024).
10. L. Qiao, Y. Li, Y. Liu, Y. Wang and K. Du, High-strength, blood-compatible, and high-capacity bilirubin adsorbent based on cellulose-assisted high-quality dispersion of carbon nanotubes, J. Chromatogr. A, 1634, 2020, 461659,  DOI:10.1016/j.chroma.2020.461659.
11. X. Song, X. Huang, Z. Li, Z. Li, K. Wu, Y. Jiao and C. Zhou, Construction of blood compatible chitin/graphene oxide composite aerogel beads for the adsorption of bilirubin, Carbohydr. Polym., 2019, 207, 704–712,  DOI:10.1016/j.carbpol.2018.12.005.
12. A. S. Timin, A. V. Solomonov, I. I. Musabirov, S. N. Sergeev, S. P. Ivanov and E. V. Rumyantsev, Characterisation and evaluation of silica particles coated by PVP and albumin for effective bilirubin removal, J. Sol-Gel Sci. Technol., 2015, 74, 187–198,  DOI:10.1007/s10971-014-3595-y.
13. K. E. Hagstam, L. E. Larsson and H. Thysell, Experimental studies on charcoal hemoperfusion in phenobarbital intoxication and uraemia, including histopathologic findings, Acta Med. Scand., 1966, 180, 593–603,  DOI:10.1111/j.0954-6820.1966.tb02875.x.
14. S. Kitagawa, R. Kitaura and S. Noro, Functional porous coordination polymers, Angew. Chem., Int. Ed., 2004, 43, 2334–2375,  DOI:10.1002/anie.200300610.
15. G. Férey and C. Serre, Large breathing effects in three-dimensional porous hybrid matter: facts, analyses, rules, and consequences, Chem. Soc. Rev., 2009, 38, 1380–1399,  10.1039/B804302G.
16. R. Han, M. Zhao, X. Li, S. Cui and J. Yang, N-doped regular octahedron MOF-199 derived porous carbon for ultra-efficient adsorption of oxytetracycline, Sep. Purif. Technol., 2022, 302, 121960,  DOI:10.1016/j.seppur.2022.121960.
17. J. Pan, X. Bai, Y. Li, B. Yang, P. Yang, F. Yu and J. Ma, HKUST-1 derived carbon adsorbents for tetracycline removal with excellent adsorption performance, Environ. Res., 2022, 205, 112425,  DOI:10.1016/j.envres.2021.112425.
18. P. Goyal, A. Paruthi, D. Menon, R. Behara, A. Jaiswal, K. V. A. Kumar, V. Krishnan and S. K. Misra, Fe doped bimetallic HKUST-1 MOF with enhanced water stability for trapping Pb(II) with high adsorption capacity, Chem. Eng. J., 2022, 430, 133088,  DOI:10.1016/j.cej.2021.133088.
19. Y. Zhang, Q. Zhao, W. Li, J. Liu, J. Chen, Y. Fan and Y. Weng, Substrate independent liquid phase epitaxy of HKUST-1 as anticoagulant and antimicrobial coating, Adv. Mater. Interfaces, 2020, 7, 1902011,  DOI:10.1002/admi.201902011.
20. Y. Zhu, D. Yang, J. Li, Z. Yue, J. Zhou and X. Wang, The preparation of ultrathin and porous electrospinning membranes of HKUST-1/PLA with good antibacterial and filtration performances, J. Porous Mater., 2023, 30, 1011–1019,  DOI:10.1007/s10934-022-01394-z.
21. H. Dong, Y. He, C. Fan, Z. Zhu, C. Zhang, X. Liu, K. Qian and T. Tang, Encapsulation of imazalil in HKUST-1 with versatile antimicrobial activity, Nanomaterials, 2022, 12, 3879,  DOI:10.3390/nano12213879.
22. C. Jiang, X. Wang, G. Wang, C. Hao, X. Li and T. Li, Adsorption performance of a polysaccharide composite hydrogel based on crosslinked glucan/chitosan for heavy metal ions, Composites, Part B, 2019, 169, 45–54,  DOI:10.1016/j.compositesb.2019.03.082.
23. D. Nataraj, S. Sakkara, M. Meghwal and N. Reddy, Crosslinked chitosan films with controllable properties for commercial applications, Int. J. Biol. Macromol., 2018, 120, 1256–1264,  DOI:10.1016/j.ijbiomac.2018.08.187.
24. X. Ma, L. Wang, H. Wang, J. Deng, Y. Song, Q. Li, X. Li and A. M. Dietrich, Insights into metal-organic frameworks HKUST-1 adsorption performance for natural organic matter removal from aqueous solution, J. Hazard. Mater., 2022, 424, 126918,  DOI:10.1016/j.jhazmat.2021.126918.
25. T. Xia, F. Xie, X. Bian, Z. Chen, S. Zhang, Z. Fang, Q. Ye, J. Cai and Y. Wang, Ultrabroad-spectrum, multidrug resistant bacteria-killing, and biocompatible quaternised chitin derivative for infected wound healing, Mater. Sci. Eng., C, 2021, 126, 112177,  DOI:10.1016/j.msec.2021.112177.
26. X. Qiu, L. Nie, P. Liu, X. Xiong, F. Chen, X. Liu, P. Bu, B. Zhou, M. Tan, F. Zhan, X. Xiao, Q. Feng and K. Cai, From haemostasis to proliferation: Accelerating the infected wound healing through a comprehensive repair strategy based on GA/OKGM hydrogel loaded with MXene@TiO2 nanosheets, Biomaterials, 2024, 308, 122548,  DOI:10.1016/j.biomaterials.2024.122548.
27. Q. Guo, Z. Wu, K. Wang, J. Shi, M. Wei, B. Lu, Z. Huang and L. Ji, Forsythiaside-A improved bile-duct-ligation-induced liver fibrosis in mice: The involvement of alleviating mitochondrial damage and ferroptosis in hepatocytes via activating Nrf2, Free Radical Biol. Med., 2024, 222, 27–40,  DOI:10.1016/j.freeradbiomed.2024.05.042.
28. L. Ren, X. Yan, J. Zhou, J. Tong and X. Su, Influence of chitosan concentration on mechanical and barrier properties of corn starch/chitosan films, Int. J. Biol. Macromol., 2017, 105, 1636–1643,  DOI:10.1016/j.ijbiomac.2017.02.008.
29. B. S. Sran, S. Kamalakannan, J. W. Hwang, J. W. Yoon, K. H. Cho, D. Jo, H. C. Ham, S.-K. Lee and U.-H. Lee, Optimised separation of C₂H₂ from binary mixtures of C₂H₂/CO₂ and C₂H₂/C₂H₄ by coordinative copper (I) metal nodes modified from HKUST-1, Chem. Eng. J., 2024, 491, 152034,  DOI:10.1016/j.cej.2024.152034.
30. X. Ma, L. Wang, H. Wang, J. Deng, Y. Song, Q. Li, X. Li and A. M. Dietrich, Insights into metal-organic frameworks HKUST-1 adsorption performance for natural organic matter removal from aqueous solution, J. Hazard. Mater., 2022, 424, 126918,  DOI:10.1016/j.jhazmat.2021.126918.
31. S. Li, L. Huang, B. Jia, X. Feng, Y. Cao, Y. Chen and Y. Bin, Effect and mechanism of inorganic anions on the adsorption of Cd²⁺ on two-dimensional copper-based metal–organic framework, Inorg. Chem. Commun., 2024, 159, 111819,  DOI:10.1016/j.inoche.2023.111819.
32. K. A. Lin and Y. T. Hsieh, Copper-based metal organic framework (MOF), HKUST-1, as an efficient adsorbent to remove p-nitrophenol from water, J. Taiwan Inst. Chem. Eng., 2015, 50, 223–228,  DOI:10.1016/j.jtice.2014.12.008.
33. Y. Chen, O. Sakata, Y. Nanba, L. S. R. Kumara, A. Yang, C. Song, M. Koyama, G. Li, H. Kobayashi and H. Kitagawa, Electronic origin of hydrogen storage in MOF-covered palladium nanocubes investigated by synchrotron X-rays, Commun. Chem., 2018, 1, 61,  DOI:10.1038/s42004-018-0058-3.
34. P. Ngamsurach, N. Namwongsa and P. Praipipat, Synthesis of powdered and beaded chitosan materials modified with ZnO for removing lead (II) ions, Sci. Rep., 2022, 12, 17184,  DOI:10.1038/s41598-022-22182-4.
35. S. Sheshmani and M. Mardali, Harnessing the synergistic potential of ZnS nanoparticle-interfacing chitosan for enhanced photocatalytic degradation in aqueous media and textile wastewater, J. Polym. Environ., 2024, 32, 5783–5805,  DOI:10.1007/s10924-024-03307-4.
36. B.-C. Li, J.-Y. Lin, J. Lee, E. Kwon, B. X. Thanh, X. Duan, H. H. Chen, H. Yang and K.-Y. A. Lin, Size-controlled nanoscale octahedral HKUST-1 as an enhanced catalyst for oxidative conversion of vanillic alcohol: The mediating effect of polyvinylpyrrolidone, Colloids Surf., A, 2021, 631, 127639,  DOI:10.1016/j.colsurfa.2021.127639.
37. Z. Li, X. Song, S. Cui, Y. Jiao and C. Zhou, Fabrication of macroporous reduced graphene oxide composite aerogels reinforced with chitosan for high bilirubin adsorption, RSC Adv., 2018, 8, 8338–8348,  10.1039/C8RA00358K.
38. J. Hu, H. Gao, X. Wang and B. Tan, High-capacity volumetric methane storage in hyper-cross-linked porous polymers via flexibility engineering of building units, Adv. Mater., 2025, 2418005,  DOI:10.1002/adma.202418005.
39. J. Li, Y. Xi, H. Shao, Q. Ma, X. Dong and X. Li, Ambient-dried biomass 3D porous aerogel with ‘brick-mortar-binder’ structure for the determination and adsorption of Cr(VI), Chem. Eng. J., 2025, 509, 161359,  DOI:10.1016/j.cej.2025.161359.
40. M. M. Ahmed, Y. Z. Shen, Z. Wang, J. Li, J. Du, S. Azat and Q. Xu, MOFs and their derivatives: Functionalisation strategy and applications as sensitive electrode materials for environmental EDCs analysis, Coord. Chem. Rev., 2025, 531, 216504,  DOI:10.1016/j.ccr.2025.216504.
41. P. Wiśniewska, J. Haponiuk, M. R. Saeb, N. Rabiee and S. A. Bencherif, Mitigating metal-organic framework (MOF) toxicity for biomedical applications, Chem. Eng. J., 2023, 471, 144400,  DOI:10.1016/j.cej.2023.144400.
42. J. Feng, Y. Tan, L. Sun, Q. Liang, T. Jiang and Z. Li, Fe-based metal-organic frameworks modified with carboxymethyl cellulose for targeted pesticide delivery and reducing biotoxicity, Int. J. Biol. Macromol., 2024, 282, 137285,  DOI:10.1016/j.ijbiomac.2024.137285.
43. K. Baskaran, M. Ali, B. J. Riley, I. Zharov and K. Carlson, Evaluating the physisorption and chemisorption of iodine on bismuth-functionalised carbon foams, ACS Mater. Lett., 2022, 4, 1780–1786,  DOI:10.1021/acsmaterialslett.2c00466.
44. H. Sampei, K. Saegusa, K. Chishima, T. Higo, S. Tanaka, Y. Yayama, M. Nakamura, K. Kimura and Y. Sekine, Quantum annealing boosts prediction of multimolecular adsorption on solid surfaces avoiding combinatorial explosion, JACS Au, 2023, 3, 991–996,  DOI:10.1021/jacsau.3c00018.
45. J. C. A. Braun, C. E. Borba, M. Godinho, D. Perondi, J. M. Schontag and B. M. Wenzel, Phosphorus adsorption in Fe-loaded activated carbon: Two-site monolayer equilibrium model and phenomenological kinetic description, Chem. Eng. J., 2019, 361, 751–763,  DOI:10.1016/j.cej.2018.12.073.
46. X.-S. Wang, J. Liang, L. Li, Z.-J. Lin, P. P. Bag, S.-Y. Gao, Y.-B. Huang and R. Cao, An anion metal–organic framework with Lewis basic sites-rich toward charge-exclusive cationic dyes separation and size-selective catalytic reaction, Inorg. Chem., 2016, 55, 2641–2649,  DOI:10.1021/acs.inorgchem.6b00019.
47. L. F. Hartje, B. Munsky, T. W. Ni, C. J. Ackerson and C. D. Snow, Adsorption-coupled diffusion of gold nanoclusters within a large-pore protein crystal scaffold, J. Phys. Chem. B, 2017, 121, 7652–7659,  DOI:10.1021/acs.jpcb.7b03999.
48. H. Mohabey and H. Mohabey, IR Spectra, magnetic, and thermal studies of copper (II) complex of N-hydroxy –N-(4-chloro) phenyl N’(4-fluoro) phenyl benzamidine hydrochloride, Mater. Sci. Res. India, 2014, 11, 63–65.
49. M. M. Campos-Vallette, R. E. Clavijo, F. Mendizabal, W. Zamudio, R. Baraona and G. Diaz, Infrared spectrum of the bis-(1,10-phenanthroline) Cu(I) and Cu(II) perchlorate complexes, Vib. Spectrosc., 1996, 12, 37–44,  DOI:10.1016/0924-2031(96)00012-4.
50. I. Persson, P. Persson, M. Sandström and A.-S. Ullström, Structure of Jahn–Teller distorted solvated copper(II) ions in solution, and in solids with apparently regular octahedral coordination geometry, J. Chem. Soc., Dalton Trans., 2002, 7, 1256–1265,  10.1039/B200698G.
51. S. Wu, B. Duan, X. Zeng, A. Lu, X. Xu, Y. Wang, Q. Ye and L. Zhang, Construction of blood compatible lysine-immobilised chitin/carbon nanotube microspheres and potential applications for blood purified therapy, J. Mater. Chem. B, 2017, 5, 2952–2963,  10.1039/c7tb00101k.
52. Q. Li, H. Guo, J. Yang, W. Zhao, Y. Zhu, X. Sui, T. Xu, J. Zhang and L. Zhang, MOF-based antibiofouling hemoadsorbent for highly efficient removal of protein-bound bilirubin, Langmuir, 2020, 36, 8753–8763,  DOI:10.1021/acs.langmuir.0c01047.
53. B. A. Harms, G. C. Kramer, B. I. Bodai and R. H. Demling, Effect of hypoproteinaemia on pulmonary and soft tissue edema formation, Crit. Care Med., 1981, 9, 503.
54. S. Jäger, M. Cabral, J. F. Kopp, P. Hoffmann, E. Ng, J. B. Whitfield, A. P. Morris, L. Lind, T. Schwerdtle and M. B. Schulze, Blood copper and risk of cardiometabolic diseases: a Mendelian randomisation study, Hum. Mol. Genet., 2022, 31, 783–791,  DOI:10.1093/hmg/ddab275.
55. Ye Yang, et al., Construction of Kevlar nanofiber/graphene oxide composite beads as safe, self-anticoagulant, and highly efficient hemoperfusion adsorbents, J. Mater. Chem. B, 2020, 8.9, 1960–1970,  10.1039/C9TB02789K.
56. J. Pohl, D. Aretakis, F. Tacke, C. Engelmann and M. Sigal, Role of intestinal barrier disruption to acute-on-chronic liver failure, Semin. Liver Dis., 2025, 45, 52–64,  DOI:10.1055/a-2516-2361.
57. T. Arai, Y. Yoshikai, J. Kamiya, M. Nagino, K. Uesaka, N. Yuasa, K. Oda, T. Sano and Y. Nimura, Bilirubin impairs bactericidal activity of neutrophils through an antioxidant mechanism in vitro, J. Surg. Res., 2001, 96, 107–113,  DOI:10.1006/jsre.2000.6061.
58. N. M. Khan and T. B. Poduval, Immunomodulatory and immunotoxic effects of bilirubin: molecular mechanisms, J. Leukocyte Biol., 2011, 90, 997–1015,  DOI:10.1189/jlb.0211070.
59. H. Huang, S. Li, Y. Zhang, C. He and Z. Hua, Microglial priming in bilirubin-induced neurotoxicity, Neurotoxic. Res., 2023, 41, 338–348,  DOI:10.1007/s12640-023-00643-6.
60. W. J. Jung, Y.-I. Roh, H. Im, Y. Lee, D. Im, K.-C. Cha and S. O. Hwang, Efficacy of cardiopulmonary resuscitation using automatic compression—defibrillation apparatus: An animal study and a manikin-based simulation study, J. Clin. Med., 2023, 12, 5333,  DOI:10.3390/jcm12165333.
61. N. Gan, Q. Sun, X. Peng, P. Ai, D. Wu, B. Yi, H. Xia, X. Wang and H. Li, MOFs-alginate/polyacrylic acid/poly (ethylene imine) heparin-mimicking beads as a novel hemoadsorbent for bilirubin removal in vitro and vivo models, Int. J. Biol. Macromol., 2023, 235, 123868,  DOI:10.1016/j.ijbiomac.2023.123868.
62. R. Jin, C. Zhao, Y. Song, X. Qiu, C. Li and Y. Zhao, Competitive adsorption of sulfamethoxazole and bisphenol A on magnetic biochar: Mechanism and site energy distribution, Environ. Pollut., 2023, 329, 121662,  DOI:10.1016/j.envpol.2023.121662.
63. S. A. Barclay and D. Bennett, The direct measurement of plasma colloid osmotic pressure is superior to colloid osmotic pressure derived from albumin or total protein, Intensive Care Med., 1987, 13, 114–118,  DOI:10.1007/BF00254796.
64. Q. Li, H. Guo, J. Yang, W. Zhao, Y. Zhu, X. Sui, T. Xu, J. Zhang and L. Zhang, MOF-based antibiofouling hemoadsorbent for highly efficient removal of protein-bound bilirubin, Langmuir, 2020, 36, 8753–8763,  DOI:10.1021/acs.langmuir.0c01047.
65. M. M. Kulkarni, B. Popovic, A. L. Nolfi, C. D. Skillen and B. N. Brown, Distinct impacts of aging on the immune responses to extracellular matrix-based versus synthetic biomaterials, Biomaterials, 2025, 320, 123204,  DOI:10.1016/j.biomaterials.2025.123204.
66. Y. Lin, X. Lv, C. Sun, Y. Sun, M. Yang, D. Ma, W. Jing, Y. Zhao, Y. Cheng, H. Xuan and L. Han, TRIM50 promotes NLRP3 inflammasome activation by directly inducing NLRP3 oligomerisation, EMBO Rep., 2022, 23, e54569,  DOI:10.15252/embr.202154569.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5tb01037c

‡ These authors contributed equally to this work.

---