A mixture of poly (ethylene glycol) (5.5 g, 1,500 Da, Sigma-Aldrich, United Kingdom), glycolide (3.08 g, Fisher Scientific Ltd., United Kingdom) and D,L-Lactide (9.57 g, Fisher Scientific Ltd., United Kingdom) was dried under vacuum for 1 h at 100°C. The temperature was then increased to 150°C followed by the addition of Tin (II) 2-ethylhexanoate (29 μL, Sigma-Aldrich, United Kingdom) in a dropwise manner. The reaction was carried out for 5 h under dry nitrogen. After which the reaction mixture was cooled to room temperature and subsequently poured into 250 mL of dichloromethane (DCM) until complete dissolution. The solution was then purified by pouring into 250 mL of ice-cold diethyl ether. The precipitation was collected and the purification process was repeated for a total number of 3 times. The purified polymer was then collected and residual solvent evaporated under vacuum overnight. The product was characterised using GPC (Shimadzu, Japan) and NMR (Bruker 400 MHz). ¹H NMR (400 MHz, CDCl₃) δ 5.27 (s, 1H), 5.23–5.07 (m, 5H), 4.89–4.60 (m, 3H), 4.36–4.20 (m, 1H), 3.60 (s, 20H), 3.44 (q, J = 7.0 Hz, 5H), 1.65–1.40 (m, 15H), 1.16 (t, J = 7.0 Hz, 8H). The resultant triblock co-polymer had Mw 6,090 Da, Mn 4,940 Da, and PDI 1.23.

Microparticles were fabricated using a double emulsion method as previously described (White et al., 2013). In brief, a mixture of PLGA (85:15, 55 kDa, IV 0.47 dL/g, Evonik Industries AG) with various amounts of TB at a total mass of 1 g was dissolved in 5 mL of DCM. An aqueous solution of human serum albumin (HSA, 100 mg mL⁻¹) or a mixture of HSA (95 mg mL⁻¹, Sigma-Aldrich, United Kingdom) and IL-4 (5 mg mL⁻¹, Peprotech, United Kingdom), was prepared. The use of HSA has previously been shown to stabilise cytokines during the encapsulation process and enhance overall encapsulation efficiency (Pean et al., 1998; White et al., 2013). One hundred microlitres of protein solution was added to the polymer solution and homogenised using a Silverson L5M homogeniser (Silverson Machines, United Kingdom) at 4,000 rpm for 2 min to form the primary water-in-oil emulsion. The emulsion was then poured into 200 mL of 0.3% (w/v) poly (vinyl alcohol)_(aq) (13,000–23,000 Da, 87%–89% hydrolysed, Sigma-Aldrich, United Kingdom) and homogenised at 9,000 rpm for 2 min. The product was then stirred at 300 rpm on a stirrer plate for a minimum of 4 h to facilitate solvent evaporation. The microparticles were then washed with copious amounts of water followed by centrifugation at 2,100 × g after each wash. The final product was then lyophilised using a CoolSafe 4–15 L freeze dryer (LaboGene, Denmark). The size distribution of microparticles was determined using laser diffraction particle size analysis (LA-960, HORIBA Scientific Ltd.). Qualitative images of the microparticles were obtained by sputter coating with gold at 5 nm thickness using Balzers SCD 030 sputter coater (Balzers, Liechtenstein) and imaging using scanning electron microscopy (FEI XL30, Hillsboro, OR).

Ten milligrams of microparticles were dissolved in 750 µL of dimethyl sulfoxide, followed by addition of 2.15 mL of 0.02% (w/v) sodium dodecyl sulfate in 200 mM sodium hydroxide solution for 1 h. The concentration of protein encapsulated was assessed using a micro BCA™ protein assay (Fisher Scientific Ltd., United Kingdom) according to manufacturer’s instructions. Standard curves were obtained using HSA in concentrations ranging from 30 to 2 μg mL⁻¹.

Twenty-five milligrams of microparticles were suspended in 1.5 mL of complete culture media, described in Section 2.6. The samples were incubated at 37°C on a rocker continuously at 60 rpm. At appropriate time intervals, the samples were briefly centrifuged, supernatant was collected and replaced with fresh media. Functional release of IL-4 and HSA were measured using ELISA kits (DY204 and DY1455, R&D Systems, United Kingdom) according to manufacturer’s instructions.

Four commonly used drug release kinetic models were selected to analyse the controlled release kinetics: zero order, first order, Higuchi and Ritger-Peppas (Peppas and Narasimhan, 2014; Li et al., 2021; Trucillo, 2022). The correlation coefficient values, R², were used to select the model that best fitted the release profile.

Human peripheral blood monocyte cells (THP-1, ATCC^® TIB-202™, ATCC, United Kingdom) were cultured as previously described and used within 20 population doublings upon receipt (Huleihel et al., 2017). Cultures were expanded under an atmosphere of 5% CO₂ at 37°C in Roswell Park Memorial Institute 1,640 medium (RPMI-1640, 52400025, Thermo Fisher Scientific, United Kingdom) supplemented with 10% fetal bovine serum (Sigma-Aldrich, United Kingdom) and ß-mercaptoethanol (Fisher Scientific Ltd., United Kingdom) to a final concentration of 0.05 mM.

To differentiate THP-1 into functional macrophages, THP-1 cells were plated at a density of 3 × 10⁵ cells cm⁻² in complete RPMI-1640 medium (as described above) supplemented with 300 nM of phorbol 12-myristate 13-acetate (PMA, Sigma-Aldrich, United Kingdom) for 24 h. The culture was then replaced with complete RPMI-1640 medium, with absence of PMA, for 48 h.

Functional macrophages were polarised into pro-inflammatory-like or pro-immunoregulatory-like phenotypes by direct treatment of cytokines. Pro-inflammatory-like phenotype was induced in complete RPMI-1640 medium supplemented with 100 ng mL⁻¹ of lipopolysaccharide (LPS, Sigma-Aldrich, United Kingdom) and 20 ng mL⁻¹ interferon-γ (IFN-γ, Peprotech, United Kingdom) for up to 72 h. Pro-immunoregulatory-like phenotype was induced in complete RPMI-1640 medium supplemented with 20 ng mL⁻¹ IL-4 for up to 72 h.

In a 1.5 mL microcentrifuge tube, microparticles of dosage adjusted to 20 ng mL⁻¹ were suspended in complete RPMI-1640 medium. In a 24-well plate, suspensions of microparticles were co-cultured either in direct contact or indirectly with functional macrophages through transwell plate inserts (membrane size of 1 μm, 83.3932.101, Sarstedt, Germany) to result in IL-4 dosage equivalent to exogenous administration of IL-4. At each time interval, media was removed and stored at −80°C until cytokine secretion analysis was undertaken withELISAs. Separately, cells were lysed at each time interval with 150 µL of Buffer RLT Plus provided with RNeasy Plus Mini Kit (74134, Qiagen, United States) and stored at −80°C until gene expression analysis was undertaken.

Cytokine secretion of macrophages cultured as described above (Section 2.7) was determined using ELISA. The pro-inflammatory-like phenotype was determined bydetection of functional human TNF-α (DY210, R&D Systems, United Kingdom) according to manufacturer’s instructions. The pro-immunomodulatory-like phenotype was indicated by measuring functional human TGF-β1 (DY240, R&D Systems, United Kingdom) and CCL18 (DY394, R&D Systems, United Kingdom), according to manufacturer’s instructions.

Gene expression profiles of macrophages cultured as described above (Section 2.7) were determined using RT-qPCR. Total RNA was isolated using RNeasy plus mini kit (74134, Qiagen, United States of America) according to manufacturer’s instructions. Nucleic acid concentrations were measured using a Nanodrop (Labtech, United Kingdom). Complementary deoxyribonucleic acid (cDNA) was then synthesized immediately with 45 ng μL⁻¹ of purified RNA using High-Capacity cDNA Reverse Transcription Kit (4368814, Applied Biosystems, United Kingdom) with RNase Inhibitor (10615995, Applied Biosystems, United Kingdom) according to manufacturer’s instructions.

The synthesized cDNA samples were diluted 1:10 with nuclease-free water. To each well of a 384-well qPCR reaction plate, 5 µL of PowerUp SYBR Green Master Mix (15350929, Applied Biosystems, United Kingdom), 0.5 µL of selected primers, 2.5 µL of nuclease-free water and 2 µL of cDNA were added to yield in a 10 µL reaction volume. The reaction plate was sealed and qPCR was conducted using a Bio-Rad CFX384 Touch Real-Time PCR Detection System for 40 cycles at 60 °C. Primers used for qPCR were GAPDH (NM_002046.7, QT00079247, Qiagen, United States), CXCL2 (NM_002089.4, QT00013104, Qiagen, United States), MRC1 (NM_002438, QT00012810, Qiagen, United States), CCL22 (NM_002990.5, QT00089817, Qiagen, United States of America), IL-1β (NM_000576.3, Forward (5′-3′): CTA​AAC​AGA​TGA​AGT​GCT​CC; Reverse (5′-3′): GGT​CAT​TCT​CCT​GGA​AGG, Sigma-Aldrich, United Kingdom).

Relative expression was normalised to GAPDH and expression in functional macrophages using ΔΔCt method.

Statistical analyses were performed using one-way ANOVA using Origin software. Where indicated n represents the number of independent experiments per condition performed in triplicate and N represents the batches of microparticles fabricated. Statistical analysis of two batches of IL-4/HSA microparticle cumulative release best-fit kinetics were performed with a two-sample t-test using Origin software.

Preliminary formulation screening of microparticle release profiles were fabricated from PLGA85:15 with 0%, 20% and 40% TB, encapsulated with lysozyme/HSA at 1:9 ratio was used as model protein (Supplementary Figure S1). PLGA85:15 with 40% TB showed a sustained, linear release profile and this was chosen to fabricate microparticles encapsulated with HSA and IL-4. The morphology and size distribution of the microparticles were characterized qualitatively by scanning electron microscopy and quantitatively by laser-diffraction size distribution (Figure 1). Microparticles displayed an overall smooth and non-porous surface morphology with a bell-curved distribution of microparticle diameter. Microparticles with IL-4 and HSA (IL-4/HSA), or HSA alone did not exhibit apparent differences in morphology or size distribution. The mean diameter of microparticles without protein (i.e., blank), encapsulated with HSA and encapsulated with IL-4/HSA was 18.01 ± 7.01 µm, 16.79 ± 6.06 µm and 20.23 ± 7.84 µm (mean ± s. d.), respectively. Microparticles encapsulated with IL-4/HSA yielded a mean efficiency of 68.43% ± 6.69% encapsulation.

Microparticles encapsulated with HSA exhibited controlled linear release kinetics (Figure 2; Table 1). Microparticles encapsulated with HSA alone had a burst release of 12.49 µg per mg of particles within the first 24 h followed by sustained release for up to 28 days. HSA encapsulated microparticles displayed a release profile that best-fitted zero-order kinetics suggesting that this formulation has the potential to provide linear drug release. Therefore, this formulation was used to fabricate IL-4 incorporated microparticles.

Two separate batches of IL-4/HSA microparticles were fabricated and release kinetics were examined (N1 and N2) (Table 1; Figure 3A). The functional release of IL-4 over 28 days fitted best with first order kinetics. Microparticles fabricated with IL-4 (N1) displayed an initial burst release of 85.13 ng per mg of particles within the first 24 h followed by a sustained total release of 95.81 ng of IL-4 per mg of particles over the course of 2–7 days (Figure 3A). Microparticles fabricated with IL-4 (N2) displayed an initial burst release of 168.73 ng per mg of particles within 24 h followed by a sustained total release of 125.6 ng of IL-4 per mg of particles over the course of 2–7 days. It is worth noting functional IL-4 release kinetics over both 7 days and 28 days were best fitted with first order kinetics (Supplementary Table S1; Supplementary Figure S2A).

Functional release of HSA from IL-4/HSA microparticles displayed similar release kinetics as IL-4 for both N1 and N2 batches (Table 1; Figure 3B). However, HSA release over 28 days fitted best with the Ritger-Peppas model. Although HSA release from IL-4/HSA microparticles for 7 days were best-fitted with first order kinetics (Supplementary Table S1; Supplementary Figure S2B), microparticles had continued to release HSA for 28 days.

THP-1 differentiated functional macrophages were treated exogenously with LPS and IFN-γ for up to 72 h for modulation towards a pro-inflammatory-like phenotype (M(LPS/IFN-γ)) or treated with IL-4 for up to 72 h to modulate towards a pro-immunomodulatory-like phenotype (M(IL-4)). Microparticles with no protein (blank) and microparticles with HSA alone were used as controls. Microparticles formulations containing no protein (blank), HSA, and IL-4/HSA were cultured in direct contact (d) with macrophages or indirectly (id) through transwells plate inserts.

Cells exogenously treated with pro-inflammatory factors (M(LPS/IFN-γ)) displayed an increased secretion in TNF-α at both 24 h and 72 h (Figure 4A). In contrast, cells treated with pro-immunomodulatory factors (M(IL-4)) and blank, HSA and IL-4/HSA microparticles secreted significantly lower levels of TNF-α.

IL-4/HSA microparticles co-cultured directly or indirectly with cells exhibited increased levels of CCL18 and TGF-β1 secretion, similar to M(IL-4) polarised candidates (Figures 4B, C). M(LPS/IFN-γ) polarised candidates displayed a reduction of TGF-β1 secretion upon 72 h polarisation compared to M0 (Figure 4C). Functional macrophages in direct or indirect contact with both blank and HSA-alone MPs demonstrated no apparent level of expression of both pro-inflammatory-like (TNF-α) and pro-immunoregulatory-like (TGF-β1 and CCL18) cytokine secretion in comparison to functional macrophages.

THP-1 differentiated macrophages treated with pro-inflammatory factors showed upregulation in expression of both IL-1β and CXCL2 whereas when exposed to IL-4/HSA encapsulated microparticles and M(IL-4), there was a downregulation in expression of IL-1β and CXCL2 (Figures 5A, B). The results were consistent with inflammatory cytokine (TNF-α) expressions measured using an ELISA assay. MRC1 and CCL22 were selected as pro-immunoregulatory markers pro-inflammatory shifted cells (M(LPS/IFN-γ)) showed downregulation in expression of MRC1 and CCL22 whereas upregulation was observed in those of exogenously pro-immunoregulatory shifted M(IL-4) cells and those exposed to IL-4/HSA encapsulated microparticles.

Both blank MPs and HSA-alone MPs either directly or indirectly co-cultured with functional macrophages showed no significant regulatory differences in any markers examined, indicating that MPs and the release of HSA have no effects on macrophage polarisation. IL-4/HSA encapsulated MPs when co-cultured with functional macrophages directly or indirectly separated by a transwell membrane showed downregulation of IL-1β and CXCL2 and upregulation of MRC1 and CCL22 (Figures 5C, D), compared to functional macrophages. There was no significant difference in gene expression when functional macrophages were exposed to IL-4 exogenously or IL-4/HSA microparticles.