Effect of the stiffness of one-layer protein-based microcapsules on dendritic cell uptake and endocytic mechanism

Abstract

Microcapsules are one of the most promising microscale drug carriers due to their facile fabrication, excellent deformability, and high efficacy in drug storage and delivery. Understanding the effects of their physicochemical properties (size, shape, rigidity, charge, surface chemistry, etc.) on both in vitro and in vivo performance is not only highly significant and interesting but also very challenging. Stiffness, an important design parameter, has been extensively explored in recent years, but how the rigidity of particles influences cellular internalization and uptake mechanisms remains controversial. Here, one-layered lysozyme-based microcapsules with well-controlled stiffness (modulus ranging from 3.49 ± 0.18 MPa to 26.14 ± 1.09 MPa) were prepared and used to investigate the effect of stiffness on the uptake process in dendritic cells and the underlying mechanism. The cellular uptake process and endocytic mechanism were investigated with laser scanning confocal microscopy, mechanism inhibitors, and pathway-specific antibody staining. Our data demonstrated that the stiffness of protein-based microcapsules could be a strong regulator of intracellular uptake and endocytic kinetics but had no obvious effect on the endocytic mechanism. We believe our results will provide a basic understanding of the intracellular uptake process of microcapsules and the endocytic mechanism and inspire strategies for the further design of potential drug delivery microcarriers.

---

1. Introduction

Recently, increasing attention has been shown to drug carriers due to their ability to improve poor pharmacokinetics and reduce the side effects of traditional therapeutic agents (e.g., nonspecific targeting, poor physiological stability, low endocytosis efficiency) as well as to contribute to new treatment strategies for many diseases previously thought to be incurable.^1–4 Specifically, targeting drug delivery to diseased tissues is critical for efficiently treating various diseases, and this targeting could be regulated by the physicochemical properties of the carriers, such as size,^5,6 shape,^7,8 elasticity,^9,10 charge,^11–13 chemical composition,^14–16 surface–ligand arrangement,^17–19 surface chemistry,^20,21 and other properties.^22–25 Understanding the effects of these physicochemical properties both in vitro (endocytic mechanism, internalization kinetics, etc.) and in vivo (circulation time, biodistribution, tissue targeting, etc.) is highly significant and expected to improve treatment outcomes; however, it is also very challenging because of inevitable changes in other parameters when one becomes the subject of focus.

Compared to the abovementioned physicochemical properties, the stiffness of particles is recognized as a critical parameter and has been extensively explored in recent years.^26–28 Although many reports have demonstrated that the stiffness of particles affects their interactions with cells, there is a lack of consensus about how stiffness influences cellular internalization and uptake mechanisms. Generally, increasing the stiffness of particles enhances their cellular uptake.^29–32 Based on the simulation of molecular dynamics, the reason for this phenomenon may be that compared to stiff nanoparticles, flexible polymeric nanoparticles need to overcome higher binding energy to undergo morphological deformation during cellular internalization.³³ For example, Yi and coworkers verified that soft particles are more difficult for membranes to engulf for cellular uptake than stiff particles.⁹ Conversely, Liu et al. reported that the uptake rate of soft submicron hydrogel particles (Young's moduli: 15–35 kPa) by HepG2 cells was higher than that of stiff particles (75–160 kPa).³⁴ In particular, Banquy et al. showed that more nanoparticles with intermediate elasticity (Young's modulus: 30–140 kPa) were internalized by RAW 264.7 murine macrophages than other softer and stiffer nanoparticles.³⁵ In addition, Guo et al. showed that in vitro neoplastic and nonneoplastic cells showed more efficient uptake of soft nanolipogels (<1.6 MPa) than the elastic ones (>13.8 MPa), and in vivo, significantly more soft nanolipogels accumulated in tumours.⁵⁸ However, Hui et al. found that in vitro the uptake of soft silica nanocapsules (704 kPa) by macrophages was 3 times lower than that of stiff nanocapsules (9.7 GPa), whereas PEGylated nanocapsules were taken up by cancer cells independently of stiffness. Moreover, in vivo experiments confirmed that soft nanocapsules accumulated more efficiently in tumour cells than stiff nanocapsules due to the high splenic clearance of stiff nanocapsules.⁵⁹ Obviously, the impacts of stiffness on cellular interactions in vitro and biological systems in vivo are far from being understood due to the different cells and materials applied in different investigations, making comparisons difficult.

Microcapsules are emerging as one of the most promising microscale drug carriers due to their facile fabrication, high efficacy in drug storage and delivery,^36–38 and excellent deformability for passing through narrow blood vessels, which are important considerations when seeking to increase circulation time and target treatments to various diseased tissues. Capsules with various properties were coupled with applicable polymers, ranging from natural to synthetic, from low-fouling to active-targeting, from neutral to negatively charged, and from deformable and soft to stiff.³⁹ Among all the properties, the effect of microcapsule stiffness on uptake by cells has seen meaningful experimental progress. Hartmann et al. observed that soft microcapsules entered lysosomes more rapidly than stiff microcapsules.²⁹ Caruso's group stated that soft microcapsules favoured faster and more efficient cellular uptake because of their deformable shape and contracted volume.⁴⁰ The internalization process of polyelectrolyte microcapsules by mammalian cells was pioneered by Gil et al. through nonspecific adsorption and formation of phagocytic cup steps.⁴¹ Although some unique and interesting phenomena have been reported, the relationship between stiffness and the internalization process and mechanism remains unclear.

Protein-based microcapsules have received attention as effective delivery carriers owing to their abundant bioactivities, including their targeting ability and their use as medicine on their own. Additionally, proteins with well-positioned groups are highly reproducible building blocks for producing nano- and micro-objects with well-defined functionalities. However, effect of the stiffness of protein-based microcapsules on cellular uptake and the endocytosis mechanism has rarely been reported due to the lack of a method for generating protein capsules with controllable properties. Considering these issues, herein, we proposed one-layer protein-based microcapsules⁴² with well-defined size and chemical components as well as controllable stiffness to investigate the effect of stiffness on the uptake and intracellular mechanism of dendritic cells (DCs, which form large cytoplasmic protrusions that engulf microcapsules, leading to cellular uptake⁴³). Our findings demonstrated that the stiffness of protein-based microcapsules could be a strong regulator of intracellular uptake and endocytic kinetics, but no effect on the endocytic mechanism was observed. We believe our results will provide a basic understanding of the intracellular uptake process and endocytic mechanism of microcapsules and inspire strategies for the further design of potential drug delivery microcarriers.

2. Experimental section

2.1 Materials

Analytical grade calcium chloride (CaCl₂), sodium carbonate (Na₂CO₃), sodium hydroxide (NaOH), sodium dihydrogen phosphate (NaH₂PO₄), disodium hydrogen phosphate dodecahydrate (Na₂HPO₄·12H₂O), sodium bicarbonate (NaHCO₃), dimethylsulfoxide (DMSO), ethylenediaminetetraacetic acid (EDTA) and tannic acid (TA) were purchased from Aladdin (China). Lysozyme (twice crystalline, from chicken egg white, ≥90%) was purchased from Sinopharm Chemical Reagent Co., Ltd (China).  4-(4,6-Dimethoxy-1,3,5-triazin-2-yl)-4-methyl morpholinium chloride (DMT-MM), fluorescein isothiocyanate (FITC) and FITC-dextran with various average molecular weights (0.3, 40, 400, and 2000 kDa) were purchased from Sigma-Aldrich (USA). FITC-lysozyme was prepared as reported.⁴⁴ A dialysis membrane (MWCO: 3500 Da) was purchased from Shanghai Yuanye Bio-Technology Co., Ltd (China). Ultrapure water was purified using a Millipore Simplicity 185 purification unit (Milli-Q water, resistivity ∼18.2 MΩ cm).

Dulbecco's modified Eagle's medium (DMEM), foetal bovine serum (FBS), antibiotics, wheat germ agglutinin (WGA), and 4′-6-diamidino-2-phenylindole (DAPI) were purchased from Thermo Scientific (Shanghai, China). DCs (ATCC number 14990) were obtained from American Type Culture Collection (ATCC, USA). Cell Counting Kit-8 (CCK-8) reagent was purchased from Dojindo (China). Cytochalasin D, amiloride, chlorpromazine, and filipin III were purchased from Sigma-Aldrich. Rabbit anti-LAMP 1 antibody (ab62562), rabbit anti-caveolin-1 antibody (ab2910), and rabbit anti-clathrin heavy chain antibody (ab21679) were purchased from Abcam (UK). An immunofluorescent staining kit with Alexa Fluor 647-labelled goat anti-rabbit IgG (H + L) was purchased from Beyotime Biotechnology (China). Four percent paraformaldehyde was purchased from Solarbio (China).

2.2. Preparation of TA-CaCO₃ microparticles

The preparation of CaCO₃ microparticles was reported in our previous study.^45,46 With a similar procedure, TA-CaCO₃ microparticles were fabricated in the presence of TA (1 mL of desired concentration) while mixing the CaCl₂ solution (2 mL of 0.5 M) and the Na₂CO₃ solution (3 mL of 0.33 M). Finally, a series of TA-CaCO₃ microparticles was prepared with different final TA concentrations (0, 0.5, 2.0, and 10.0 mg mL⁻¹) and the microparticles were denoted TA-0/CaCO₃, TA-5/CaCO₃, TA-20/CaCO₃, and TA-100/CaCO₃, respectively.

2.3. Loading lysozyme into TA-CaCO₃ particles

For the quantification of lysozyme loading, diluted FITC-lysozyme standards with known concentrations were investigated by UV spectrophotometry. The standard curve of absorbance values was plotted against the FITC-lysozyme concentration, as shown in Fig. S8c.†

In the next experiment, weighed particles (20 mg, TA-CaCO₃) were dispersed in 1 mL of H₂O by ultrasonication for several seconds, the desired lysozyme solution (2 mg mL⁻¹) was added, and the mixture was shaken for 200 min. The particles were centrifuged and washed three times with H₂O at 800g for 5 min. The loading efficiency (LE, %) of FITC-lysozyme was calculated using the following equation:⁴⁶

2.4. Generation of lysozyme capsule crosslinking with DMT-MM

FITC-lysozyme-loaded TA-CaCO₃ (20 mg) was redispersed into 1 mL of DMT-MM solution (0.3 mg mL⁻¹) and shaken for more than 3 h. Then, after centrifuging and washing with water, the particles were incubated with 2 mL of EDTA solution (0.2 M, pH ≥ 7.4) for 12 h. Lysozyme capsules were generated and collected after centrifuging at 1200g for 10 min and washing twice with water.

2.5. Permeability tests

The dispersed lysozyme capsules (1.28 × 10⁶ capsules per mL) were mixed with an equal volume of 1 mg mL⁻¹ FITC (0.39 kDa) or FITC-dextran (4, 40, 400, and 2000 kDa) solution and incubated under absolute dark conditions for 30 min. LSCM images of the capsules were taken. To determine the authentic permeability of capsules, 100 capsules were assessed in each group, and then, the percent (%) was plotted against the molecular weight of FITC-dextran.

2.6. Thickness tests

Lysozyme capsules of different thicknesses were prepared in solutions with different pH values, and the thicknesses were measured by atomic force microscopy (AFM). First, a drop of each capsule suspension was added to a polished silicon wafer and dried overnight. Then, AFM images and corresponding height profiles of all the capsules prepared under the indicated pH conditions were obtained. Finally, the thickness was plotted against different pH values (mean ± S.D., N = 20).

2.7. Mechanical tests

AFM of lysozyme capsules was performed in water with a Bruker Dimension Icon atomic force microscope (USA). In this experiment, an SNL-10 model probe (resonance frequency 56 kHz) was selected. The probe and cantilevers were cleaned with 30 vol% isopropanol, water, and plasma treatment to remove any contaminating material. First, the mechanical properties mode of the Dimension Icon AFM was used for calibration, and sapphire was used as the calibration sample. Second, with the calibration operation method for solid material indentation under normal circumstances, the cantilever deflection sensitivity was marked. Third, the cantilever deflection sensitivity was raised, the probe was thermally tuned, and the elastic coefficient (0.24 N m⁻¹) of the cantilever beam was measured. Fourth, the standard sample for measuring the tip radius was scanned in peak force QNM in air mode, and the experimental probe radius was obtained through offline image analysis software. Finally, using the peak force tapping technology of the peak force QNM mode, the force–distance curves of lysozyme capsules were quickly obtained. Prior to that measurement, PEI and PAA for electrostatic immobilization of the capsules, in turn, were adsorbed on cleaned glass substrates. Through the rapid analysis of each force–distance curve data to generate force-deformation (F − δ) data, the elastic modulus diagram was obtained. Four points of each area and 50 detection points of 10 capsules were selected randomly to analyse the modulus of each sample using the NanoScope Analysis software.

2.8. Cytotoxicity assay

Bone marrow DCs (DC 2.4, 1 × 10⁴ cells per mL) were incubated in 96-well plates for 24 h and then cultured for another 48 h or 72 h with fresh media containing a different number of capsules, which had been pretreated with antibiotic solution and high-purity sterilized water in turn. The capsule-to-cell ratios were set to 1 : 0, 1 : 10, 1 : 25, 1 : 50, and 1 : 100. Afterwards, the cells were lightly washed with PBS to remove the capsules. Finally, the cells were further incubated for 2 h to calculate the relative cell viabilities using the CCK-8 assay. All the experiments were performed in septuplicate, and the relative percentage of cell viability (%) was normalized to that of the untreated control cells.

Similarly, the cytotoxicities of the inhibitors, including cytochalasin D, amiloride, chlorpromazine, and filipin III, at the indicated concentrations in DCs were assessed over 48 h.

2.9. Intracellular uptake experiments

DCs (3 × 10⁴ cells per well) were cultured in 8-well plates for 24 h and further incubated with capsules (pretreated with antibiotic solution and high-purity sterilized water in turn, capsule-to-cell ratio of 100 : 1) for another 48 h.

For the quantitative cellular uptake assay, cells were incubated with paraformaldehyde (4%) for 15 min at 37 °C. After washing twice with PBS, the cells were stained with WGA (20 μg mL⁻¹) for 15 min, followed by DAPI (50 μM) for 20 min at room temperature. Finally, cellular uptake was recorded by a laser scanning confocal microscope (LSCM) (Nikon, JPN). Each sample was run in quintuplicate. The statistical analysis of capsule uptake by cells (N = 200) was performed, and the percentage (%) of quantitative uptake was plotted against the range of pH values.

For the study of the colocalization of endocytic pathways, cellular structures were stained with either antibodies or markers of known intracellular compartments, and the corresponding targets were phagolysosomes (lysosomal-associated membrane protein 1, LAMP1), caveolin-coated lipid rafts (caveolin 1), and clathrin-coated vesicles (clathrin).⁴⁷ Cells were incubated with paraformaldehyde (4%) for 15 min at 37 °C. After washing twice with PBS, the cells were permeabilized in a mixed solution of glycin (5 mg mL⁻¹) and saponin (0.5 mg mL⁻¹) for 5 min and blocked with bovine serum albumin solution (20 mg mL⁻¹) for 30 min. Then, the cells were treated with primary antibodies for 1 h, followed by fluorophore-conjugated secondary antibodies for another 1 h. Cytoskeleton staining with phalloidin–tetramethylrhodamine was performed together with secondary antibody treatment. Finally, the cell nucleus was stained with DAPI for 5 min, and the cells were mounted and imaged by LSCM.

For the chemical inhibition of uptake, DCs (3 × 10⁴ cells per well) were subcultured in fresh growth media (100 μL) without antibiotics and supplemented with 10 μL of cytochalasin D (20 μM), amiloride (50 μM), chlorpromazine (10 μM), or filipin III (1 μg mL⁻¹) for 24 h prior to the addition of FITC-labelled lysozyme capsules.⁴⁸ Then, the cells were incubated for an additional 48 h with gentle shaking and washed twice with PBS. Finally, the cell membranes and nuclei were stained with WGA (20 μg mL⁻¹) for 15 min, followed by DAPI (50 μM) for 20 min at room temperature, and the cells were mounted and imaged by LSCM. The data are presented as average values with standard deviations (mean ± S.D., N = 5). Prior to the experiment, the suitable lower doses of each inhibitor were investigated to determine their optimal concentrations, and the details are shown in Fig. S13.†

2.10. Characterization

The morphologies of the samples were observed by a field emission scanning electron microscope (FESEM, SU8010 HITACHI, JPN) at a voltage of 3 keV. The composition of the samples was measured by energy dispersive spectroscopy (EDS, IXRF SDD3060-A5501, JPN) for 15 min. The surface morphologies and stiffness of all the lysozyme capsules were characterized by atomic force microscopy (AFM, Bruker Dimension Icon, USA). All of the above samples were prepared by applying a drop of the particle or capsule suspension to a polished silicon wafer and dried overnight. Infrared spectra of samples were tested by a Fourier transform infrared spectrometer (FT-IR, Bruker TENSOR II, USA). The BET specific surface area and the BJH mean pore diameter of TA-CaCO₃ microparticles were measured by a N₂ isothermal adsorption–desorption instrument (Micromeritics 3Flex, USA) in the pressure range of 0.05–1.00. The zeta potentials of TA-CaCO₃ particles were determined at pH 7.0 by dynamic light scattering (Nano 4e, Malvern, UK). Confocal micrographs of FITC-lysozyme-loaded TA-CaCO₃ microparticles and FITC-lysozyme capsules were observed with laser scanning confocal scanning microscopy (LSCM, Nikon Al apparatus, JPN). UV-Vis absorption spectra were measured by UV spectroscopy (Lambda 25, PerkinElmer, USA). The data were analysed using one-way ANOVA (*p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001).

3. Results and discussion

3.1. Fabrication of hollow lysozyme capsules

The lysozyme capsules were prepared using our previously described method,⁴² as shown in Fig. S1a,† where single-step adsorption of lysozyme (a cartoon image of a lysozyme molecule is shown in Fig. S1b†) onto tannic acid (molecular structure is shown in Fig. S1c†)-doped mesoporous templates (TA-CaCO₃) was followed by chemical crosslinking with 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methyl morpholinium chloride (DMT-MM, molecular structure shown in Fig. S1d†). Upon the removal of the supporting templates, the protein films retained their original shape and sufficient integrity. All of the lysozyme capsules were separately characterized by SEM, AFM, and elemental mapping analysis (Fig. 1a, b and S2†). The SEM image of lysozyme capsules showed a collapsed morphology (Fig. 1a). The details of the morphology observed by AFM showed similar wrinkles and folds (Fig. 1b). In Fig. 1e, the energy-dispersive X-ray (EDX) mapping images of lysozyme capsules showed that all elemental distribution (N, O, and C) patterns matched well with the SEM images and overlap maps, except the Ca pattern, which strongly demonstrated that the templates were removed completely. To further confirm the generation of hollow capsules, green circles (in Fig. 1c) with fluorescein isothiocyanate-labelled lysozyme (FITC-lysozyme) were observed by laser scanning confocal microscopy (LSCM). It was the typical fluorescence profile, in which distances from the outer surface to the inner part and fluorescence intensity immediately decreased. Furthermore, FITC-lysozyme capsules were incubated with aqueous solution (pH 7.4) for three months at 4 °C and still maintained their overall structure, as shown in Fig. S3;† these results demonstrated that the capsules were relatively stable. The results indicated the successful generation of lysozyme capsules exhibiting a morphology similar to that reported for protein capsules. Moreover, in the FT-IR spectrum of lysozyme capsules, compared with TA, lysozyme, and a mixture of TA and lysozyme (Fig. S4†), the amide I and II bands (1667 and 1530 cm⁻¹) shifted to lower wavenumbers, confirming that the cross-linking reaction occurred between lysozymes. Moreover, the characteristic peaks of TA (1720 and 1191 cm⁻¹) concurrently disappeared, suggesting that all TA was removed with the template of CaCO₃. Thus, lysozyme capsules were successfully generated by the cross-linking of proteins. As shown in Fig. 1f and g, the permeability of lysozyme capsules with FITC-dextrans was dependent on molecular weight (M_w); the permeability ranged from 100% at M_w ≤ 4 kDa to 0% at M_w ≥ 2000 kDa, and partial permeability was observed at 4 kDa < M_w < 2000 kDa. In detail, the percentages of partially permeable capsules were approximately 89% and 36% at M_w values of 40 and 400 kDa, respectively. These results indicated that the generated capsules had good integrity but probably possessed different densities and thicknesses of lysozyme membranes.

Fig. 1 (a) SEM, (b) AFM, and (c) LSCM images, (d) fluorescent profile, and (e) EDX elemental mapping images of lysozyme capsules. (f) LSCM images of lysozyme capsules incubated with FITC (0.39 kDa) and FITC-dextran (4, 40, 400, or 2000 kDa). (g) The percentage of permeable capsules pretreated with FITC-dextran of different molecular weights. The dependence of lysozyme adsorption on amounts of doped TA at pH 7.0, 8.0, 9.5, and 11.0, respectively.

3.2. Effect of doped TA on lysozyme loading

In the process of capsule preparation, because doped TA is the major cause of lysozyme deposition, the amount of doped TA should regulate lysozyme adsorption. In a previous publication,⁴² introducing TA into porous CaCO₃ microparticles resulted in a large change in their ability to interact with lysozyme. Therefore, it is key that more TA is doped into TA-CaCO₃ microparticles during their synthesis. However, at concentrations over 15 mg mL⁻¹, it is difficult to form large numbers of integrated microparticles. For example, broken particles prepared with 20 mg mL⁻¹ TA-200/CaCO₃ were observed everywhere in the images (Fig. S5a and b†). Herein, CaCO₃ microparticles were synthesized in the presence of TA at the indicated concentrations (0, 0.5, 2.0, 5.0, and 15.0 mg mL⁻¹, denoted as TA-0/CaCO₃, TA-5/CaCO₃, TA-20/CaCO₃, TA-50/CaCO₃, and TA-150/CaCO₃, respectively). The results showed that doping TA did not affect the spherical morphology and overall size of TA-CaCO₃ microparticles with different concentrations of doped TA, and the details of the TA-50/CaCO₃ particles are shown in Fig. S6a–c.† In Fig. S6d,† the zeta potential (−18.6 mV) of the TA-50/CaCO₃ particles at pH 7.0 was completely reversed compared to that of TA-0/CaCO₃ (6 mV),⁴⁹ which was one reason that the particles could load lysozyme, i.e., a positively charged protein, at this pH value. Furthermore, the SEM images of the surface morphology of a single TA-CaCO₃ microparticle shown in Fig. 2a revealed that the microparticles were generated from the nanoparticle assembly and that the nanoscale size obviously decreased with increasing TA doping concentration. For the N₂ adsorption–desorption isotherms and pore size distribution of TA-CaCO₃ (Fig. S7a and b†), the channels of all microparticles were mesoporous and single, and the dependence of pore size distributions on the concentration of doped TA is shown in Fig. 2b. Obviously, the increase in concentration minimized the pore size. The average pore size and BET surface area were further analysed. In Fig. 2b, the pore size decreased sharply with the initial increase in TA concentration, then slowly plateaued, and reached approximately 12.2 nm in TA-150/CaCO₃ microparticles, which is 2.9-fold smaller than the 35.0 nm pore size of TA-0/CaCO₃ microparticles. The BET surface area showed a contradictory trend and finally reached 26.0 m² g⁻¹ in TA-150/CaCO₃ microparticles, which was 3.8 times higher than the 6.8 m² g⁻¹ of TA-0/CaCO₃ microparticles.

Fig. 2 (a) SEM images of the surface topography of single TA-CaCO₃ microparticles with different concentrations of TA (0, 0.5, 2, 5, or 15 mg mL⁻¹). All scale bars are 200 nm. (b) The surface areas and average pore sizes of TA-CaCO₃ microparticles vs. the concentration of doped TA. (Particle: 20 mg; lysozyme: 2 mg, in 1 mL buffer). (c) The effect of lysozyme adsorption on particles. The dependence of lysozyme adsorption on amounts of doped TA at pH 7.0, 8.0, 9.5, and 11.0, respectively.

In the following analyses, pH was considered another key factor that affects the interaction between lysozyme and TA-CaCO₃ particles because all proteins are amphoteric electrolytes. Therefore, the effect of the amount of lysozyme absorbed on the particles was evaluated with UV-vis. In the quantitative test of protein absorbed into particles, because the particles were subjected to a long period of absorption in protein solution, the absorption intensity at low wavelengths was influenced by the released TA (Fig. S8a†). Here, we used FITC-lysozyme to investigate the absorption intensity of the characteristic peak at 494 nm, as shown in Fig. S8b†. A good standard curve was plotted, as shown in Fig. S8c.† To reach the maximum loaded amount at each applied concentration of TA, a high concentration (2 mg mL⁻¹) of lysozyme at the desired pH (7.0, 8.0, 9.5 and 11.0) was used for these experiments, and the final amount of loaded lysozyme was calculated (detailed information is provided in Experimental section 2.3). As presented in Fig. 2c, for all particles of the same batch at the same doped TA, the maximum amount of loaded lysozyme almost increased with increasing pH value. The detailed SEM images are shown in Fig. S9.† Between polyphenol and protein, there were multiple interactions, including hydrogen bonding interactions, hydrophobic interaction, and Schiff-base formation.^50,51 Here, the major driving force of protein deposition could be Schiff-base formation between TA and lysozyme. Under alkaline conditions, the phenol groups of TA can be oxidized into benzoquinone groups, which have high possibility to form Schiff-base bonding with the amino groups of lysozyme. After more oxidization with increasing pH value, more lysozyme molecules covalently combined with TA possessing more benzoquinone groups on the surface of particles. In particular, at pH 8.5, there was a slight decrease in the amount of protein deposits. Because the isoelectric point (IEP) of CaCO₃ is pH 8.5,⁵² there was not only an interaction between the protein and TA of the CaCO₃ surface but also an electrostatic interaction between the amphoteric protein and CaCO₃ with a weakly positive charge. At pH 8.0 (approximately the IEP of CaCO₃), considering that the net charge of CaCO₃ approached zero, there was no interaction between the CaCO₃ surface and lysozyme, and the adsorption of lysozyme decreased compared with that at pH 7.0. Generally, at the same pH, the maximum loaded amount had almost no effect on the initial increase in the concentration of TA and increased sharply with the increase in the concentration of TA (2.0 mg mL⁻¹). Only with the concentration of TA at 5.0 mg mL⁻¹ (that is, TA-50/CaCO₃) was there a distinct division of the loaded amount for the particles generated under high pH conditions, indicating that we can obviously control the amount of lysozyme on the particle.

3.3. Control of capsule thickness

To further investigate the regulation of capsule thickness, we selected TA-50/CaCO₃ as the template and prepared lysozyme-based capsules by absorbing lysozyme under the indicated pH conditions (7.0, 8.0, 9.5, and 11.0). The thicknesses measured by AFM are shown in Fig. 3a. In general, the thickness value increased significantly with the pH value besides a slight decrease at pH 8.0, which was consistent with the result on amounts of loaded lysozyme. Obviously, the loaded amounts were unaffected by the process of lysozyme cross-linking and removing CaCO₃ core. Fig. 3b and c shows the corresponding AFM images and the AFM height map. At low pH, the hollow, thin-walled microcapsules were covered with wrinkles, while at high pH, the microcapsules had smooth surfaces because the thicker walls could not collapse. The thickness ranged from 32.1 ± 8.9 nm to 73.45 ± 9.5 nm, suggesting that there was a significant difference over a wide range of capsule thicknesses when the pH was controlled in the capsule preparation process. In addition, all the capsule sizes were 3–5 microns (Fig. S10†) and closer, which is crucial for further research on the effect of individual capsule wall thicknesses on cells. This result is essential for exploring the effect of the single-factor wall thickness of the capsule on cells.

Fig. 3 (a) Film thicknesses of lysozyme capsules prepared under different pH conditions as measured by AFM (mean ± S.D., N = 20). **p < 0.01 and ***p < 0.001. Representative AFM images (b) and corresponding height profiles (c) of lysozyme capsules prepared under the indicated pH conditions. Scale bars are 1 μm.

3.4. Capsule uptake in DCs

Because DCs have evolved a particular nature to internalize and remove viruses, bacteria, and even microsized exogenous particles or abnormal cells, they are considered as the proper cell line for studying the uptake capability and efficiency of protein microcapsules with different thicknesses. First, using the CCK-8 assay, the cytotoxicity of lysozyme capsules in DCs was investigated after incubation at various cell-to-capsule ratios (from 1 : 0 to 100 : 1) for 48 h or 72 h at 37 °C (Fig. S11†). As expected, lysozyme capsules did not cause marked cytotoxicity in DCs, which was consistent with the cytotoxicity of gelatine capsules in DCs.⁴²

To visualize the cellular uptake of lysozyme capsules and the intracellular distribution, the capsules were monitored by LSCM after treating immature DCs with FITC-lysozyme capsules after 48 h. In Fig. 4a, the 3D LSCM image clearly showed the specific location of the capsule that had entered the cell, and this uptake did not interfere with cell viability.

Fig. 4 (a) 3D LCSM image of capsule uptake in DCs. (b) Number of capsules taken up per DC. ***p < 0.001, and ****p < 0.0001. (c) LCSM image of capsule uptake in DCs at pH values of 7.0, 8.0, 9.5, and 11.0. Scale bars are 50 μm.

Furthermore, we evaluated the number of capsules prepared under the indicated pH conditions (7.0, 8.0, 9.5, and 11.0) that entered the cells with LSCM images. As shown in Fig. 4b, the average number of capsules taken up per cell showed the same correlation with pH value as the thickness data described above. The range of uptake per cell could be controlled from 1.59 (pH 8.0) to 6.37 (pH 11.0). In detail, LCSM images of capsule uptake into DCs at pH values of 7.0, 8.0, 9.5 and 11.0 are shown in Fig. 4c. For the same size lysozyme capsule (Fig. S10†), the results showed that the uptake of capsules was more dependent on thickness.

3.5. The modulus and uptake ratio of lysozyme capsules

We proposed a possible mechanism for the thickness-dependent uptake of lysozyme capsules, and this mechanism involves the modulus of the capsules. AFM force spectroscopy measurements were performed in water (Fig. 5a), and the details are provided in Experimental section 2.7. The modulus of the spherical capsules could be estimated for thin-walled spherical shells, and their modulus images and corresponding sections of lysozyme capsules prepared under pH conditions of 7.0, 8.0, 9.5, and 11.0 are shown in Fig. S12.† In Fig. 5b, the average modulus of lysozyme capsules was calculated. At least 50 detection points of 10 single capsules were analysed in each sample, and the values were estimated to be 10.58 ± 3.22 MPa at pH 7.0, 3.49 ± 0.18 MPa at pH 8.0, 18.77 ± 1.85 MPa at pH 9.5, and 26.14 ± 1.09 MPa at pH 11.0. All of the results suggested that the average modulus of the capsules also showed remarkable pH dependence, which is consistent with the thickness of the capsules. That is, the thicker the lysozyme film of the capsule, the stiffer the capsule in water.

Fig. 5 (a) Schematic representation of the elastic modulus of immobilized lysozyme capsules with the AFM technique. (b) The average modulus of lysozyme capsules prepared at pH 7.0, 8.0, 9.5, and 11.0. ***p < 0.001 and ****p < 0.0001.

3.6. Uptake pathway of lysozyme capsules

The uptake pathway is important to further understand the significance of capsules. To clearly elucidate the endocytic mechanism, two methods were used.⁴⁷ One was to label the involved proteins produced in the cell during capsule uptake into DCs and observe the endocytosis pathway directly by LSCM images. The other was to pretreat DCs with chemical inhibitors after the uptake of lysozyme capsules and demonstrate the main endocytosis pathway indirectly by observing larger changes in the number of capsules taken up before and after inhibition.

Usually, cells take in micrometre-sized particles through phagocytosis or macropinocytosis. Phagocytosis is mainly used to remove invading pathogens, dead cells, and cell debris. Phagocytosis is carried out by membrane protrusions that surround particles. Additionally, macropinocytosis is used to engulf substantial amounts of extracellular fluid or particles through plasma membrane ruffling involving actin aggregation.

The direct method was used by observing the location of FITC-lysozyme capsules and DC endocytic markers involved in uptake pathways, such as lysosomal-associated membrane protein 1 (LAMP 1), caveolin, and clathrin; the details are shown in Table S1.† LAMP1 is a marker of phagolysosomes at earlier stages of phagocytosis, and caveolin and clathrin, in turn, are markers for caveolin- and clathrin-mediated endocytosis. In Fig. 6, the uptake of lysozyme-based capsules into DCs and their intracellular localization were studied in LSCM images. Colocalization of the capsules (green) with LAMP1 (red), caveolin (red), and clathrin (red) was assessed to determine the intracellular pathway. Only large red areas can be clearly seen in the upper row of Fig. 6, with no evidence in the lower two rows, suggesting that phagocytosis rather than caveolin- and clathrin-mediated endocytosis was the uptake pathway.

Fig. 6 Colocalization of FITC-lysozyme capsules (green) treated at pH 7.0, 8.0, 9.5, and 11.0 and DCs marked with LAMP1 (red, upper two rows), caveolin (red, middle two rows) and clathrin (red, lower two rows) in LSCM. The cell membranes and nuclei were stained with WGA (adjusted greyscale to aquamarine) and DAPI (blue), respectively. All scale bars are 20 μm.

Furthermore, we investigated the cellular uptake of these capsules in the presence of cytochalasin D, amiloride, chlorpromazine, and filipin III, which are known chemical inhibitors of phagocytosis, micropinocytosis, caveolae-mediated endocytosis, and clathrin-mediated endocytosis, respectively.⁵³ Generally, low concentration of the inhibitors and short time of their incubation with cells indicate that their inhibitory effect can be low at 2–4 h; otherwise, high concentration and long incubation time will cause cytotoxicity. Therefore, the study of capsule uptake by DCs had to be conducted at suitable lower doses, and the details are shown in Fig. S13 and Table S2.† Capsule uptake was dependent on energy, which was verified by a previous report at 4 °C.⁴⁷ As shown in Fig. 7a, the inhibition of phagocytosis by cytochalasin D and micropinocytosis by amiloride decreased the cellular uptake of capsules, whereas inhibition of caveolin-mediated endocytosis by filipin III and clathrin-mediated endocytosis by chlorpromazine did not influence capsule uptake, indicating that caveolae and clathrin were not involved in capsule uptake. In detail, the inhibition of actin polymerization by cytochalasin D resulted in a 90% decrease in capsule uptake,⁵⁴ indicating that microtubules may control the cellular uptake of capsules. Inhibition of Na⁺/H⁺ exchanger isoform 1 by amiloride resulted in a 60–75% decrease in capsule uptake,⁵⁵ demonstrating that microtubules may also be involved in the cellular uptake of lysozyme capsules. Specifically, capsule uptake was not affected by chlorpromazine, a potent inhibitor of clathrin assembly, or filipin III, a potent inhibitor of caveolin assembly,^56,57 suggesting that the inhibition of clathrin did not play an important role in capsule uptake in the groups with thin capsules prepared at pH 7.0 and 8.0. For these two groups treated with chlorpromazine and filipin III, the total inhibition of capsule uptake was critically related to the thickness of the capsules. This result may have occurred due to the increasingly weak inhibition and capsule uptake over a long time. In summary, in light of the results described above that the uptake pathway does not involve caveolin- and clathrin-mediated endocytosis, cellular uptake of protein-based capsules was completely inhibited by the phagocytosis and micropinocytosis inhibitors cytochalasin D and amiloride, indicating that lysozyme capsules enter DCs via phagocytosis and macropinosomes.

Fig. 7 (a) Percentages of capsule uptake in DCs exposed to different inhibitors (cytochalasin D, amiloride, chlorpromazine, and filipin III) compared to control treatment (0.9% saline). (b) LSCM images of DCs following incubation with lysozyme capsules for 48 h after pretreatment with endocytosis inhibitors or 0.9% saline. All scale bars are 20 μm.

4. Conclusions

In summary, we successfully generated one-layer lysozyme-based microcapsules with well-controllable stiffness, and the effects of the microcapsule stiffness on intracellular uptake by DCs and their endocytic mechanism were revealed. The stiffness, with a modulus ranging from 3.49 ± 0.18 MPa to 26.14 ± 1.09 MPa, was regulated by the thickness of lysozyme deposited onto TA-doped CaCO₃ templates, which can be influenced by both the pH during which lysozyme is deposited and the doped amount of TA. The number of intracellular microcapsules increased with increasing microcapsule stiffness, and a faster endocytic process was observed for stiffer microcapsules. However, the endocytic pathway into DCs did not exhibit obvious differences between these microcapsules, and all of them followed phagocytosis and macropinosomes. These findings provide a fundamental understanding of the intracellular uptake process and endocytic mechanism of microcapsules and inspire strategies for the future design of high-efficiency antigen/drug delivery microcarriers.

Conflicts of interest

The authors declare that they have no conflicts of interest.

Acknowledgements

This work was supported by start-up funding from the Wenzhou Institute of UCAS (No. WIUCASQD2019009).

Notes and references

1. W. Arap, R. Pasqualini and E. Ruoslahti, Science, 1998, 279, 377–380.
2. A. M. Wu and P. D. Senter, Nat. Biotechnol., 2005, 23, 1137–1146.
3. F. Danhier, O. Feron and V. Preat, J. Controlled Release, 2010, 148, 135–146.
4. C. Li, J. Wang, Y. Wang, H. Gao, G. Wei, Y. Huang, H. Yu, Y. Gan, Y. Wang, L. Mei, H. Chen, H. Hu, Z. Zhang and Y. Jin, Acta Pharm. Sin. B, 2019, 9, 1145–1162.
5. S. Reichert, P. Welker, M. Calderon, J. Khandare, D. Mangoldt, K. Licha, R. K. Kainthan, D. E. Brooks and R. Haag, Small, 2011, 7, 820–829.
6. K. Nam, T. Kim, Y. M. Kim, K. Yang, D. Choe, L. B. Mensah, K. Y. Choi and Y. H. Roh, Chem. Commun., 2019, 55, 4905–4908.
7. Y. Geng, P. Dalhaimer, S. Cai, R. Tsai, M. Tewari, T. Minko and D. E. Discher, Nat. Nanotechnol., 2007, 2, 249–255.
8. P. Chaturbedy, M. Kumar, K. Salikolimi, S. Das, S. H. Sinha, S. Chatterjee, B. S. Suma, T. K. Kundu and M. Eswaramoorthy, J. Controlled Release, 2015, 217, 151–159.
9. X. Yi, X. Shi and H. Gao, Phys. Rev. Lett., 2011, 107, 098101.
10. Z. Guan, L. Q. Wang and J. P. Lin, Biomacromolecules, 2017, 18, 797–807.
11. P. Chen, Z. Huang, J. Liang, T. Cui, X. Zhang, B. Miao and L. T. Yan, ACS Nano, 2016, 10, 11541–11547.
12. C. A. Fromen, G. R. Robbins, T. W. Shen, M. P. Kai, J. P. Ting and J. M. DeSimone, Proc. Natl. Acad. Sci. U. S. A., 2015, 112, 488–493.
13. M. A. Shahbazi, T. D. Fernandez, E. M. Makila, X. Le Guevel, C. Mayorga, M. H. Kaasalainen, J. J. Salonen, J. T. Hirvonen and H. A. Santos, Biomaterials, 2014, 35, 9224–9235.
14. L. C. Glangchai, M. Caldorera-Moore, L. Shi and K. Roy, J. Controlled Release, 2008, 125, 263–272.
15. M. P. Lokugamage, C. D. Sago, Z. B. Can, B. R. Krupczak and J. E. Dahlman, Adv. Mater., 2019, 31, 1902251.
16. C. D. Sago, M. P. Lokugamage, F. Z. Islam, B. R. Krupczak, M. Sato and J. E. Dahlman, J. Am. Chem. Soc., 2018, 140, 17095–17105.
17. A. Bandyopadhyay, R. L. Fine, S. Demento, L. K. Bockenstedt and T. M. Fahmy, Biomaterials, 2011, 32, 3094–3105.
18. H. M. Ding and Y. Q. Ma, Biomaterials, 2012, 33, 5798–5802.
19. N. W. Schmidt, F. Jin, R. Lande, T. Curk, W. Xian, C. Lee, L. Frasca, D. Frenkel, J. Dobnikar, M. Gilliet and G. C. Wong, Nat. Mater., 2015, 14, 696–700.
20. Y. Li, X. Li, Z. Li and H. Gao, Nanoscale, 2012, 4, 3768–3775.
21. M. Venturoli, M. M. Sperotto, M. Kranenburg and B. Smit, Phys. Rep., 2006, 437, 1–54.
22. C. Bao, B. Liu, B. Li, J. J. Chai, L. W. Zhang, L. L. Jiao, D. Li, Z. Q. Yu, F. Z. Ren, X. H. Shi and Y. Li, Nano Lett., 2020, 20, 1352–1361.
23. B. Chen, W. Dai, B. He, H. Zhang, X. Wang, Y. Wang and Q. Zhang, Theranostics, 2017, 7, 538–558.
24. J. Liu, L. Miao, J. Sui, Y. Hao and G. Huang, Asian J. Pharm. Sci., 2020, 15, 576–590.
25. R. Toy and K. Roy, Bioeng. Transl. Med., 2016, 1, 47–62.
26. X. Li, H. Tang, X. Huang, M. Li, Z. Jiang, H. He and Z. Zhou, Macromol. Biosci., 2019, 19, 1900105.
27. R. Rajamanickam, K. Kwon and G. Tae, Mater. Sci. Eng., C, 2020, 111, 110789.
28. W. Yu, W. Zhang, Y. Chen, X. Song, W. Tong, Z. Mao and C. Gao, J. Colloid Interface Sci., 2016, 465, 149–157.
29. R. Hartmann, M. Weidenbach, M. Neubauer, A. Fery and W. J. Parak, Angew. Chem., Int. Ed., 2015, 54, 1365–1368.
30. Q. Wei, C. Huang, Y. Zhang, T. Zhao, P. Zhao, P. Butler and S. Zhang, Adv. Mater., 2018, 30, 1707464.
31. S. Cao, R. Tang, G. Sudlow, Z. Wang, K.-K. Liu, J. Luan, S. Tadepalli, A. Seth, S. Achilefu and S. Singamaneni, ACS Appl. Mater. Interfaces, 2019, 11, 5499–5508.
32. J. P. Best, Y. Yan and F. Caruso, Adv. Healthcare Mater., 2012, 1, 35–47.
33. J. Sun, L. Zhang, J. Wang, Q. Feng, D. Liu, Q. Yin, D. Xu, Y. Wei, B. Ding, X. Shi and X. Jiang, Adv. Mater., 2015, 27, 1402–1407.
34. W. Liu, X. Zhou, Z. Mao, D. Yu, B. Wang and C. Gao, Soft Matter, 2012, 8, 9235–9245.
35. X. Banquy, F. Suarez, A. Argaw, J.-M. Rabanel, P. Grutter, J.-F. Bouchard, P. Hildgen and S. Giasson, Soft Matter, 2009, 5, 3984–3991.
36. S. De Koker, L. J. De Cock, P. Rivera-Gil, W. J. Parak, R. Auzély Velty, C. Vervaet, J. P. Remon, J. Grooten and B. G. De Geest, Adv. Drug Delivery Rev., 2011, 63, 748–761.
37. R. R. Costa and J. F. Mano, Chem. Soc. Rev., 2014, 43, 3453–3479.
38. M. Delcea, H. Möhwald and A. G. Skirtach, Adv. Drug Delivery Rev., 2011, 63, 730–747.
39. H. L. Sun, J. W. Cui, Y. Ju, X. Chen, E. H. H. Wong, J. Tran, G. G. Qiao and F. Caruso, Bioconjugate Chem., 2017, 28, 1859–1865.
40. H. Sun, E. H. H. Wong, Y. Yan, J. Cui, Q. Dai, J. Guo, G. G. Qiao and F. Caruso, Chem. Sci., 2015, 6, 3505–3514.
41. L. Kastl, D. Sasse, V. Wulf, R. Hartmann, J. Mircheski, C. Ranke, S. Carregal-Romero, J. A. Martínez-López, R. Fernández-Chacón, W. J. Parak, H.-P. Elsasser and P. Rivera_Gil, ACS Nano, 2013, 7, 6605–6618.
42. P. Shi, J. Qin, X. Wu, L. Wang, T. Zhang, D. Yang, X. Zan and D. Appelhans, ACS Appl. Mater. Interfaces, 2019, 11, 39209–39218.
43. S. De Koker, B. G. De Geest, S. K. Singh, R. De Rycke, T. Naessens, Y. Van Kooyk, J. Demeester, S. C. De Smedt and J. Grooten, Angew. Chem., Int. Ed., 2009, 48, 8485–8489.
44. P. Shi, J. Qin, J. Hu, Y. Bai and X. Zan, Colloids Surf., B, 2019, 175, 184–194.
45. M. L. De Temmerman, J. Demeester, F. De Vos and S. C. De Smedt, Biomacromolecules, 2011, 12, 1283–1289.
46. L. N. Hassani, F. Hindre, T. Beuvier, B. Calvignac, N. Lautram, A. Gibaud and F. Boury, J. Mater. Chem. B, 2013, 1, 4011–4019.
47. L. Kastl, D. Sasse, V. Wulf, R. Hartmann, J. Mircheski, C. Ranke, S. Carregal-Romero, J. A. Martínez-López, R. Fernández-Chacón, W. J. Parak, H.-P. Elsasser and P. Rivera-Gil, ACS Nano, 2013, 7, 6605–6618.
48. A. N. Zelikin, K. Breheney, R. Robert, E. Tjipto and K. Wark, Biomacromolecules, 2010, 11, 2123–2129.
49. P. Shi, S. Luo, B. Voit, D. Appelhans and X. Zan, J. Mater. Chem. B, 2018, 6, 4205–4215.
50. Y. Jia and J. Li, Chem. Rev., 2015, 115, 1597–1621.
51. J. Qin, S. Wei, H. Xu, L. Wang, P. Hao, B. Peng, Y. Zhang, P. Shi and X. Zan, Adv. Mater. Interfaces, 2020, 7, 2000329.
52. L. N. Hassani, F. Hindré, T. Beuvier, B. Calvignac, N. Lautram, A. Gibaud and F. Boury, J. Mater. Chem. B, 2013, 1, 4011–4019.
53. A. I. Ivanov, Pharmacological Inhibition of Endocytic Pathways: Is It Specific Enough to Be Useful?, in Exocytosis and Endocytosis, ed. A. I. Ivanov, Humana Press, Totowa, NJ, 2008, pp. 15–33.
54. O. V. Vieira, R. J. Botelho and S. Grinstein, Biochem. J., 2002, 366, 689–704.
55. J. S. Wadia, R. V. Stan and S. F. Dowdy, Nat. Med., 2004, 10, 310–315.
56. S. K. Lai, K. Hida, S. T. Man, C. Chen, C. Machamer, T. A. Schroer and J. Hanes, Biomaterials, 2007, 28, 2876–2884.
57. J. Panyam, W.-Z. Zhou, S. Prabha, S. K. Sahoo and V. Labhasetwar, FASEB J., 2002, 16, 1217–1226.
58. P. Guo, D. Liu, K. Subramanyam, B. Wang, J. Yang, J. Huang, D. T. Auguste and M. A. Moses, Nat. Commun., 2018, 9, 1–9.
59. Y. Hui, D. Wibowo, Y. Liu, R. Ran, H. Wang, A. Seth, A. P. J. Middelberg and C. Zhao, ACS Nano, 2018, 12, 2846–2857.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d1bm01448j

---