Mannose/stearyl chloride doubly functionalized polyethylenimine as a nucleic acid vaccine carrier to promote macrophage uptake

Abstract

Transmembrane transport remains a significant challenge for nucleic acid vaccine vectors. Promoting the ability of immune cells, such as macrophages, to capture foreign stimuli is also an effective approach to improving cross-presentation. In addition, polyethyleneimine (PEI) has gained attention in the field of nucleic acid vaccine carriers due to its excellent gene transfection efficiency and unique proton buffering effect. However, although high molecular weight PEI exhibits high efficiency, its high-density positive charges make it highly toxic, which limits its application. In this study, mannose/stearyl chloride functionalized polyethylenimine (SA-Man-PEI) was prepared by functionalizing PEI (molecular weight of 25 kDa) with mannose with immunomodulatory and phagocyte targeting effects, and an alkyl hydrophobic chain segment, which could easily promote cell uptake. Moreover, the functionalized-PEI retains a strong proton buffering effect, which helps the carrier escape from the lysosome. The particle sizes of the composite particles formed by SA-Man-PEI and ovalbumin (OVA) were below 200 nm, with good storage stability at both 4 °C and 37 °C. At a drug concentration of 2 μg/mL, the cell survival rate of functionalized-PEI was 19.2% higher than that of unfunctionalized PEI. In vitro macrophage endocytosis experiments showed that SA-Man-PEI could significantly enhance the macrophage uptake of composite particles, compared to unfunctionalized PEI or single-functionalized PEI. This study offers a new approach for developing PEI as a nucleic acid vaccine carrier, which could simultaneously enhance cell targeting and promote cell uptake.

1. Introduction

mRNA delivery carriers have garnered attention due to their simple design and manufacturing, inherent immunogenicity, rapid mass production, and low risk of insertion mutations (Pardi et al., 2018; Maruggi et al., 2019; Van Hoecke & Roose, 2019; Sahin et al., 2020; Xu et al., 2023). However, due to its large size (10⁴–10⁶ Da) and negative charge, mRNA cannot pass through the anionic lipid bilayer of the cell membrane, resulting in low transmembrane efficiency. Additionally, it is often engulfed by innate immune system cells and degraded by nucleases in the body (Chaudhary et al., 2021). Therefore, in vivo applications require mRNA delivery carriers that transfect immune cells without causing toxicity or unnecessary immunogenicity (Brito et al., 2014; Matsui et al., 2015; McKinlay et al., 2017). At present, many studies are focusing on developing innovative new materials as mRNA delivery carriers (Barbier et al., 2022; Ganley et al., 2023; Gote et al., 2023; You et al., 2023; Bitounis et al., 2024; Zhang et al., 2024).

There are several types of delivery carriers in the drug delivery field, including lipid nanoparticles (Hou et al., 2021; Tenchov et al., 2021; Leo et al., 2022; De Leo et al., 2023; Zhang et al., 2024), inorganic nanomaterials (Delong & Curtis, 2017; Sun et al., 2023), polymers, polymer nanoparticles (Ulkoski et al., 2019; Park et al., 2022), and polymer vesicles (Luo et al., 2023; Pan et al., 2024). However, some drug carriers have limitations, including targeting difficulties due to the large size of lipid nanoparticles and the permeability issues of polymer vesicles (Choi et al., 2023; Zhu et al., 2023). Polymer nanoparticle carriers, being compact, can bind drugs to nanoparticles by adsorption, encapsulation, or covalent binding. These nanoparticles can carry substances such as proteins, nucleic acids, and drugs that need to be transported across cell membranes. As a result, researchers continue to explore whether there are better potential applications of polymer nanoparticles in drug delivery carriers (Ishaqat & Herrmann, 2021; Ben-Akiva et al., 2023; Cui et al., 2023; Jiang & Xu, 2023; Yang et al., 2023; Sun et al., 2024).

There are two main approaches to ensuring that nanoparticles avoid lysosomal degradation: one is to bypass the lysosome through pit-dependent endocytosis, and the other is to modify the nanoparticles so they enter the lysosome and disrupt its structure to successfully facilitate escape. The proton sponge effect is a classical theory of lysosomal escape (Vermeulen et al., 2018), with representative substances including polycationic polymers with buffering effects, such as polyethyleneimine (PEI), polyamide, and chitosan.

The cationic PEI chemical structure contains numerous primary and tertiary amines, allowing for its structural modification. Li et al. (Liu et al., 2010) studied the synthesis of a novel star-shaped EAP-PEI copolymer (EAPP) structure, using eight-arm polyethylene glycol (EAP) and PEI as the basic units, and further coupled with MC11 to form the EAPP-MC11 carrier to enhance tumor targeting while exhibiting low cytotoxicity and high targeting efficacy. In addition, the primary amines in PEI provide strong buffer capacity in acidic environments, making PEI a notable candidate for carrier systems. In recent years, PEI has been commercialized as a gene transfection reagent, and an increasing number of studies have shown that PEI, as a nucleic acid vaccine carrier, can enhance the anti-infection and anti-tumor effects of traditional vaccines (Wegmann et al., 2012; Sheppard et al., 2014). Forest et al. (Forrest et al., 2003) synthesized a cross-linked PEI with PEI and diacrylate, which was biodegradable and nontoxic, and found it to be 2 to 16 times more efficient in mediating gene expression than 2.5 kDa PEI. However, it has also been reported that high-density positively charged polycationic materials have certain toxicity issues, potentially causing apoptosis, cell necrosis, and acute inflammation (Liu et al., 2010; Xin et al., 2017; Schulze et al., 2018; Vermeulen et al., 2018). Therefore, it is necessary to modify the cationic polymer to reduce its toxicity.

As a glucotrophic nutrient, mannose has been shown to regulate the immune system, as well as inhibit tumor production and metastasis (Guo et al., 2015; Yang et al., 2018; Jatczak-Pawlik et al., 2021). The presence of mannose receptors on the surfaces of macrophages can promote antigen capture and play a targeting role. The long-chain alkyl hydrophobic segments in stearyl chloride can increase the lipid solubility of the molecule and enhance cellular endocytosis (Forrest et al., 2004; Alshamsan et al., 2009; Liu et al., 2010). In this work, mannose and stearoyl chloride functionalized polyethyleneimine (SA-Man-PEI) were designed as delivery carriers. Using mannose and stearyl chloride to react with primary or secondary amines in the PEI molecular structure, the cytotoxicity of PEI can be reduced, while cell uptake capacity and transmembrane efficiency can be improved. The particle size, storage stability, and proton buffer capacity of SA-Man-PEI combined with ovalbumin (OVA) were studied using OVA as the model. Additionally, the cytotoxicity and intracellular delivery ability of polycationic carriers were investigated.

2. Experiment

2.1. Materials

Polyethyleneimine (PEI, M_w=25 kDa) was obtained from BASF Co., Ltd. Mannose, stearyl chloride, and triethylamine were purchased from Shanghai Aladdin Biochemical Technology Co., Ltd. Chloroform was supplied by the Tianjin Damao Chemical Reagent Factory. Ovalbumin (OVA, chicken egg white) was purchased from Sigma-Aldrich Co., Ltd. Macrophages (CP-M128) were obtained from Wuhan Pricella Biotechnology Co., Ltd. Fluorescein isothiocyanate ovalbumin (FITC-OVA) was purchased from Beijing Suolaibao Technology Co., Ltd. Fetal bovine serum was purchased from Lanzhou Roya Bio-Technology Co. Ltd.

2.2. Preparation of polyethylenimine mRNA carrier and composites

2.2.1. Preparation of Man-PEI

First, 10 g of PEI and 80 g of deionized water were added to a three-necked flask equipped with a mechanical stirrer and thermometer, increasing the temperature of the reaction system to 60 °C, and then stirred to enable dissolution. A total of 1.05 g of mannose was dissolved, and 10 g of deionized water was slowly added to the system, while the temperature was raised to 85 °C. After 1 h, the reaction solution was collected and dialyzed with a dialysis bag with an interception molecular weight of 3.5 kDa, and the water was changed three times every 12 h. After the dialysis, it was freeze-dried using a freeze-vacuum drying oven to obtain a light-yellow viscous substance, which consisted of mannose-functionalized polyethyleneimine (Man-PEI).

The molar ratios of mannose to the primary amino group in PEI during the reaction were 1:5, 1:10, 1:15, 1:20, and 1:25, and denoted as Man-PEI_a, Man-PEI_b, Man-PEI_c, Man-PEI_d, and Man-PEI_e, respectively.

2.2.2. Preparation of SA-PEI

First, 12.5 g of PEI dissolved in 140 g of chloroform and 0.07 g of acid-binding agent triethylamine were added to a three-necked flask equipped with a mechanical stirrer and thermometer. The system was subjected to an ice bath and nitrogen atmosphere at 5 °C. A total of 0.22 g of stearyl chloride was added to 10 g trichloromethane, which was slowly added to the flask with a constant pressure drip funnel for 30 min, and the reaction was mechanically stirred at 5 °C for 20 h. After the reaction was completed, we collected and washed the reactant twice with isopropyl ether precipitation. The precipitate was then centrifuged and dried under vacuum at 25 °C for 12 h to obtain an opaque white viscous substance, which was stearoyl chloride functionalized polyethyleneimine (SA-PEI).

The molar ratios of stearyl chloride to the primary amino group in PEI during the reaction were 1:100, 2:100, 3:100, 4:100, and 5:100, respectively, and were denoted SA-PEI_a, SA-PEI_b, SA-PEI_c, SA-PEI_d, and SA-PEI_e.

2.2.3. Preparation of SA-Man-PEI

First, 10 g SA-PEI and 80 g deionized water were added to a three-necked flask equipped with a mechanical stirrer and a thermometer, and the system was heated to 60 °C for stirring and dissolution. Subsequently, 1.05 g of mannose was added into 10 g of deionized water and stirred until dissolved. The mannose solution was then slowly added to the flask, and the water bath was heated to 85 °C and stirred for 1 h. After the reaction, the reaction solution was collected and dialysis bags with an interception molecular weight of 3.5 kDa were used for dialysis. The water in the dialysis system was changed every 12 h and dialysis was performed three times. After dialysis, the material was freeze-dried using a vacuum drying oven to obtain an opaque light-yellow viscous substance, which was mannose/stearoyl chloride functionalized polyethyleneimine (SA-Man-PEI).

2.2.4. Preparation of composite particles

A total of 0.5 g of OVA was added to 100 mL deionized water to prepare the 5 mg/mL OVA solution, and a total of 0.1 g of functionalized-PEI was added to 100 mL deionized water to prepare the 1 mg/mL functionalized-PEI solution (pH adjusted to 7.4). Subsequently, the same volume of functionalized PEI solution was slowly dropped into the OVA solution under magnetic stirring. After 10 min, functionalized-PEI/OVA composite particles were obtained, which included Man-PEI, SA-PEI, and SA-Man-PEI.

2.3. Characterization

2.3.1. Structure of functionalized-PEI

The structure of the synthetic polymer was collected by a Fourier transform infrared spectrometer (FT-IR, IRL280301, USA), where the scanning wavenumber was in the range of 4000 to 650 cm⁻¹. To further verify the structure, the synthetic polymer was characterized by a nuclear magnetic resonance hydrogen spectrometer (¹H-NMR, Avance III HD 400 MHz, Germany), and 0.005 g of synthetic polymer was dissolved in 0.6 mL of deuterium water at room temperature.

2.3.2. Determination of particle size, polydispersity index

The particle sizes (average particle size and distribution, polydispersity index) of the functionalized-PEI/OVA composite particles were measured on a PSS Z3000 (Particle Sizing Systems LLC, USA). The samples were prepared by individually diluting 10 μL of functionalized-PEI/OVA composite particles in 990 μL of distilled water.

2.3.3. Microstructure and particle size distribution of composite particles

The microscopic morphologies of the PEI/OVA, Man-PEI_b/OVA, SA-PEI_a/OVA, and SA-Man-PEI/OVA composites were observed by a transmission electron microscope (TEM, JEM-2100 Plus, Japan Electronics). The composite particles were dropped into the copper grid, and 1% phosphotungstic acid solution dropwise was used for staining, followed by drying naturally for 4 h. Next, the microscopic was observed by an electron microscope with an accelerating voltage of 200 kV.

2.3.4. Adsorption of the functionalized-PEI on OVA

The adsorption of functionalized-PEI on OVA was detected by a Multiskan Go instrument (Thermo Scientific, USA) at 37 °C, with the test repeated three times for each sample.

The working solution was prepared according to the manufacturer’s instructions for the BCA kit (PC0020). The 200 μL working solution was added to each well of a 96-well plate, and then a standard curve solution was prepared with OVA solution concentrations of 0, 0.5, 1, 1.5, 2, 3, and 5 mg/mL. Subsequently, 10 μL of each OVA solution concentration was added to a 96-well plate. The functionalized-PEI/OVA composite particles were then evenly mixed with a mass ratio of 0.04, and the composite particles were centrifuged in a high-speed centrifuge at a speed of 16000 rpm for 20 min, where the 10 μL supernatant was then added to a 96-well plate. After 20 min at 37 °C, the absorbance at 562 nm was measured by a Multiskan Go instrument. The standard curves of the OVA solution and the adsorption of functionalized-PEI on OVA were calculated.

2.3.5. Stability of composite particles

The particle size stabilities (average particle size and distribution, and polydispersity index) of the Man-PEI_b/OVA, SA-PEI_a/OVA, SA-Man-PEI/OVA composite particles were measured on a PSS Z3000 instrument (Particle Sizing Systems LLC, USA) to characterize the storage effect of the particles. The samples were placed at 4 °C and 37 °C, awaiting testing. The samples were prepared by individually diluting 10 μL of functionalized-PEI/OVA composite particles in 990 μL of distilled water, and particle size testing was conducted at 12, 24, 48, 72, and 168 h.

2.3.6. Proton buffering capacity of functionalized-PEI

The proton buffering capacity of the carrier material was determined by acid-base titration. First, 30 mg of PEI and functionalized-PEI were accurately weighed and dissolved in 30 mL of 0.15 M NaCl solution, and the initial pH of the material solution was adjusted to 10.0 with 0.1 M NaOH or 0.1 M HCl solutions. The solution to be tested was titrated with 100 μL of 0.1 M HCl solution each time. The pH of the solution was measured until the solution pH for measurement dropped to 3.0, and 0.15 M NaCl solution was used as the blank control.

2.3.7. Cytotoxicity assay

The cytotoxicity of Man-PEI_b, SA-PEI_a, SA-Man-PEI, and PEI was evaluated by the CCK-8 method. The macrophage cell concentration was adjusted to 1 × 10⁶ cells/mL, and the macrophage suspension was added to a 96-cell plate at 100 μL/well for 6 h to allow the macrophages to adhere to the wall. After adhering to the wall, the supernatant was discarded and fresh RPMI 1640 was added to the complete culture medium. Man-PEI_b, SA-PEI_a, SA-Man-PEI, and PEI were added to the cell plates at 100 μL/well, where the drug concentrations were 1 and 2 μg/mL. The process was repeated using five wells for each drug, following 18 h of cultivation, utilizing control and blank wells. In addition, the absorbance at 450 nm was measured with a Multiskan Go instrument (Thermo Scientific, USA), and then the cell viability was calculated according to

$$\text{Cell viability}=\left(\frac{A_{\mathrm{s}}-A_{\mathrm{b}}}{A_{\mathrm{c}}-A_{\mathrm{b}}}\right)\times 100\%,$$

where A_s is the absorbance of the experimental well (including cell, medium, CCK-8), A_c is the absorbance of the control well (including cells, culture medium, CCK-8 solution, no drugs), and A_b is the absorbance of the blank well (including culture medium and CCK-8 solution, no cells, and drugs).

2.3.8. Endocytosis of macrophages

The macrophage cell concentration was adjusted to 1 × 10⁵ cells/mL, and this suspension was added to the cell culture plate at a rate of 1 mL/well for 6 h to make macrophages adhere to the wall. After adherence, the supernatant was discarded and fresh RPMI 1640 complete medium was added. The composite particles formed by Man-PEI_b, SA-PEI_a, SA-Man-PEI, PEI, and FITC-OVA were added to the cell culture plate, and a blank control group (including culture medium, cells, and no drugs) was established. After culturing in a CO₂ incubator for 17 h, the excess liquid was absorbed and discarded, and 1 mL of PBS was added to each well and washed twice. We then added 500 μL of 4% paraformaldehyde fixing solution into each well, which was fixed for 20 min, washed with PBS three times for 3 min each time, and then 150 μL of DAPI staining solution was added. The solution was allowed to remain in the dark for 5 min and rinsed with PBS. Subsequently, 100 μL of antifading mounting medium was added to each well, covered with a NEST cover glass, and the endocytosis of macrophages was observed with an inverted fluorescence microscope. The drug concentration was 1 μg/mL per well, and the concentration of FITC-OVA was 25 μg/mL.

3. Results and discussion

3.1. Structure of functionalized-PEI

Polyethyleneimine was functionalized by mannose and stearyl chloride to prepare Man-PEI, SA-PEI, and SA-Man-PEI. The acyl chloride group in stearyl chloride reacted with the primary and secondary amine groups of PEI, forming amide and imide groups, and the aldehyde group in mannose reacted with the primary amine group of PEI to form a Schiff base structure. The structure diagram of functionalized-PEI is shown in Figure 1, with other diagrams for the synthesis of Man-PEI, SA-PEI, and SA-Man-PEI shown in Figures S1–S3.

Figure 1.

Structure diagram of functionalized-PEI.

FT-IR was used to characterize the molecular structure of PEI and functionalized-PEI, and the infrared spectra and infrared characteristic absorption peaks are shown in Figure S4 and Table S1.

According to Figure S4 and Table S1, the anti-stretching vibration absorption peaks of -NH₂ in the PEI, Man-PEI, SA-PEI, and SA-Man-PEI samples were 3358, 3350, 3351, and 3353 cm⁻¹, respectively, with symmetric stretching characteristic absorption peaks of 3281, 3286, 3269, and 3282 cm⁻¹, respectively. The frequency interval of symmetric and antisymmetric stretching vibration was located at about 80 cm⁻¹. The intensity of the characteristic -NH₂ peak of functionalized-PEI decreased, indicating that a portion of the primary amine participated in the reaction. In addition, the out-of-plane bending vibration peak of -NH- in SA-PEI was located at 750 cm⁻¹, and the peak shape became sharp after the reaction, indicating that the primary amine groups of PEI participated in the reaction and generated more secondary amine groups.

The stretching vibration peak of carbohydrate C-OH was found at 1112 cm⁻¹, and the nucleophilic addition reaction between the aldehyde group of mannose and the primary amine group of PEI resulted in a Schiff base reaction with a C = N structure. The characteristic absorption peak of C = N appeared at 1656 cm⁻¹ for Man-PEI, indicating that mannose successfully modified the molecular structure of PEI.

The characteristic absorption peak of C = O appeared at 1647 cm⁻¹. Theoretically, the stretching vibration frequency (amido I) of secondary carbonyl amide C = O was between 1670 cm⁻¹ and 1650 cm⁻¹, and the stretching vibration frequency (amido I) of tertiary carbonyl amide C = O was between 1675 cm⁻¹ and 1645 cm⁻¹. Combined with the above analysis, the alkyl hydrophobic chain segment successfully modified the molecular structure of PEI.

For SA-Man-PEI, the obvious change occurred at 1633 cm⁻¹, because C = N was generated in the structure of the polymer modified by mannose, which coincided with the characteristic absorption peak of C = O in the amide bond modified by stearyl chloride on the polymer.

To further verify the molecular structure, PEI and functionalized-PEI were tested by a nuclear magnetic resonance hydrogen (¹H-NMR) spectrometer, as shown in Figure S5 and Table S2.

Figure S5 and Table S2 show that the proton chemical shift on -HC = N of the Man-PEI appeared at 8.00 ppm, which did not exist in PEI, indicating that the nucleophilic addition reaction between mannose and PEI produced a C = N bond. In addition, 3.99–3.60 ppm was associated with the proton chemical shift of -OH, -CH₂, and -CH in the mannose structure, and 3.27–2.79 ppm was associated with the proton chemical shift of -CH₂ in CH₂NH and -CH₂ and -CH in PEI in the mannose modified polythene imide structure. The above results indicated that mannose was successfully grafted onto the PEI molecular structure.

The acyl chloride group in the stearyl chloride structure reacted with the primary or secondary amine groups of polyethylenimine to form amide or imide groups, and the end of the stearyl chloride structure contained -CH₃. Therefore, the ¹H-NMR spectra of stearoyl chlorine modified polyethylenimine belonged to -CH₃ at the end of the alkyl hydrophobic chain segment at 0.85 ppm, and the proton chemical shift at 7.99 ppm was the formation of secondary amide between the acyl chloride and primary amine groups.

C = N (δ_{-HC=N = 0}.85 ppm) appeared in SA-Man-PEI, which was obtained after nucleophilic addition reaction with mannose. In addition, the chemical shift of -OH proton in the mannose structure (δ_{-OH, -CH2, -CH}=3.9 9 ∼ 3.60 ppm) and a proton shift with CH₃ at the end of water transport connection structure (δ = 0.85 ppm) were also observed. The above description successfully prepared SA-Man-PEI.

According to the FT-IR and ¹H-NMR results, mannose and stearyl chloride were successfully modified into the molecular structure of PEI.

3.2. Aggregation behavior of functionalized polyethyleneimine and OVA

Man-PEI, SA-PEI, and SA-Man-PEI were obtained by modifying PEI structure with mannose and acyl chloride, respectively. Research on these functionalized polyethyleneimine polymers could serve as a nucleic acid vaccine carrier. PEI consisted of a cationic polymer in which the -NH- group carried a high density of positive charges, allowing it to bind to the negatively charged nucleic acids through electrostatic interactions, mediating effective cell transfection and nucleic acid delivery. Functionalized PEI, such as SA-PEI and SA-Man-PEI, contains a hydrophobic structure. When these polymers bind to negatively charged nucleic acid antigens, the hydrophobic intermolecular forces and electrostatic adsorption of cations synergistically promote their tight binding. This study used OVA as the material model to study the composite formation, mechanization, and morphology of the functionalized polyethyleneimine and OVA.

The particle size and poly-dispersion index (P.I.) of the PEI/OVA, Man-PEI/OVA, SA-PEI/OVA, and SA-Man-PEI/OVA composite particles were measured by a dynamic light scattering particle size-meter with a mass ratio of 0.01 to 0.07 (Figure 2). Figure 2(A)(a–c) and Figure S6 show the appearance of polymers and protein composite particles.

Figure 2.

Appearance and DLS characterization of the different monomer functionalized-PEI and OVA composite particles: (A) Appearance of the polymer/OVA composite particles; (B) Mean particle diameter of composite particles with different mass ratios; (C) P.I. of composite particles with different mass ratios; (D) Mean particle diameter and P.I. of PEI/OVA and single-functionalized-PEI/OVA.

Figure 2B (a,b) and 2C(a, b) show the effects of functionalized PEI with different molar ratios and OVA mass ratios on the mean particle diameter and P.I. When the mass ratio was less than 0.03, the particle size and P.I. of the composite particle gradually decreased and tended to be stable. When the mass ratio reached 0.05 or higher, the particle size and P.I. increased or even showed unstable micron-sized particles. Therefore, when the mass ratio of polymer/OVA was 0.04, it demonstrated the best composite mass ratio of the system. Under this condition, the average particle size uniformity of the composite particles was better, the dispersion index was small, and the solution state was relatively stable.

Figure 2 Da shows the particle size and P.I. of the Man-PEI and OVA composite particles functionalized by different mannose contents at a mass ratio of 0.04. The average particle size of the Man-PEI_b/OVA composite particles was about 168 nm, with a P.I. of less than 0.1 at the abovementioned mass ratio. The molar ratio of mannose to the primary amino group in PEI was 1:10, which could be used in follow-up studies.

Figure 2(Db) shows the particle size and P.I. of the SA-PEI and OVA composite particles functionalized by different stearyl chloride contents at a mass ratio of 0.04. At this mass ratio, the average particle size of the SA-PEI_a/OVA composite particles was about 158 nm, with a P.I. of about 0.1, and the dispersion was relatively uniform. This was due to the presence of hydrophobic alkyl chain segments in the SA-PEI structure, which could undergo hydrophobic association with the hydrophobic microregion of OVA, causing the hydrophobic regions to be close to each other. As a result, the resulting composite particles were bound more tightly. The molar ratio of stearyl chloride to the primary amino group in PEI was 1:100, which could be used for follow-up studies.

According to the above analysis, we chose a mannose and acyl chloride content of 10:1:100 to the primary amino group in PEI as the dosage for double functionalized PEI. Figure 2B(c) and 2C(c) show the particle size and P.I. values of the SA-Man-PEI/OVA composite particles at different mass ratios. When the mass ratio was 0.04, the particle size was the smallest, at about 145 nm, P.I. < 0.4, and the system was uniform and stable.

In summary, whether single functionalized or double functionalized PEI, the composite particle size after combining with OVA was less than 200 nm, which was conducive to the action of cellular uptake (Bachmann & Jennings, 2010).

According to the above analysis, a mannose and acyl chloride content of 10:1:100 to primary amino groups in PEI as the dosage was chosen for double functionalized-PEI, and a functionalized-PEI and OVA mass ratio of 0.04 was chosen for the following test.

The morphology and particle size distributions of the PEI/OVA, Man-PEI/OVA, SA-PEI/OVA, and SA-Man-PEI/OVA composite particles are shown in Figure 3. TEM images showed that the structure of functionalized-PEI and OVA composite particles was more complete and smooth than that of PEI/OVA, as well as that the inclusion of hydrophobic structures makes the particles aggregate and thereby form a more regular shape. The modification of PEI structure with mannose and acyl chloride improved the polymer’s ability to bind more effectively with OVA.

Figure 3.

TEM and particle size distribution of PEI/OVA and functionalized-PEI composite particles: (a) TEM image; (b) particle size distribution.

3.3. Adsorption of SA-Man-PEI on OVA

The BCA protein concentration detection method was used to detect the content of free protein. To prevent the influence of composite particles on detection, the composite particles were removed by the high-speed centrifugal method.

Figure 4(a) presents the standard curve of the OVA solution. The composite particles were centrifuged, PEI/OVA and functionalized-PEI/OVA were removed, the supernatant containing free OVA was retained, and the OD values were detected by an enzyme-labeled apparatus. According to the OVA standard curve, the amount of uncomplexed OVA in the supernatant was calculated, and the adsorption amount of OVA by PEI and functionalized-PEI was obtained. Figure 4(b) shows the adsorption rates of PEI and functionalized-PEI on OVA, indicating that PEI had the highest adsorption rate on OVA because PEI had the highest positive charge density and could adsorb the most negatively charged proteins. Thus, the content of free proteins in the supernatant was the lowest. The positive charge density of the SA-Man-PEI structure was lower than the single functionalized polyethylenimine, thus, the adsorption of OVA was lower than the single functionalized polyethylenimine, however, the adsorption amount was still greater than 60%.

Figure 4.

Adsorption properties of PEI and functionalized-PEI for OVA: (a) standard curve of OVA solution; (b) adsorption rate of PEI/OVA and functionalized-PEI/OVA.

3.4. Stability of functionalized-PEI/OVA composite particles

To test the storage stability of the PEI/OVA, Man-PEI/OVA, SA-PEI/OVA, and SA-Man-PEI/OVA composite particles, the composite particle systems were placed at different ambient temperatures and the particle size changes in the aqueous solutions during the period were investigated by a dynamic light scattering particle sizer, as shown in Figure 5.

Figure 5.

Particle size of PEI and functionalized-PEI composite particles at different ambient temperatures: (a) 4 °C; (b) 37 °C.

As shown in Figure 5(a), at an ambient temperature of 4 °C, the particle size of the composite particles gradually increased with an increase in placement time. Compared to the PEI/OVA composite particles, the particle sizes of Man-PEI/OVA, SA-PEI/OVA, and SA-Man-PEI/OVA showed little change, and the growth trend was relatively slow. As shown in Figure 5(b), the particle size of the composite particles presented a change trend, rapidly increasing at first and then demonstrating slow growth at 37 °C, which was relatively consistent with the change trend at 4 °C. Overall, the composite particles demonstrated better storage stability at 37 °C. In addition, the storage stability of the composite particles formed by functionalized PEI was significantly better than unfunctionalized-PEI.

3.5. Proton buffering capacity

PEI carried a strong positive charge, and even after binding with OVA, DNA, and RNA, the composite particles mediated cell uptake through electrostatic interactions with negatively charged glycoproteins on the cell surface (Dobrovolskaia & McNeil, 2007). After uptake by the cells, complex particles formed endosomes and were transported to the lysosomes or fused with the lysosomes. In the acidic environment of the lysosomes, proteases decomposed the antigen into small amino acid peptides, which were finally presented to the major histocompatibility complex (MHC) molecules and then recognized by the corresponding CD4 T cells. The unique ‘proton sponge’ structure of branched PEI facilitated lysosomal escape (Beyth et al., 2008), allowing the phagocytosed antigens to escape from the lysosome and enter the cytoplasm, thereby improving the efficiency of antigen cross-presentation. Therefore, the unique proton sponge effect of PEI was one of the important reasons for its attention as a carrier for nucleic acid vaccines.

Acid-base titration was used to simulate the change in environmental pH after the composite particles were taken up by cells into lysosomes. The initial pH was adjusted to 10.0, and the volume of HCl was consumed when the pH dropped to 3.0, where the more HCl consumed, the stronger the proton buffering capacity of the cationic polymer adjuvant.

As shown in Figure 6, the volume of hydrochloric acid consumed in descending order followed PEI > SA-PEI > Man-PEI > SA-Man-PEI, which was significantly larger than the blank NaCl ­sample. When the molar ratio of mannose to stearoyl chloride to primary amine was 10:1:100, the proton buffering capacity of the prepared SA-PEI, Man-PEI, and SA-Man-PEI samples decreased as the consumption of primary amine increased in the reaction. However, the proton buffering capacity of functionalized-PEI was close to that of PEI, primarily because the primary amine group was not completely replaced and the degree of substitution was not high. When the system was used for composite particles, it still maintained a good proton buffering effect.

Figure 6.

Proton buffering capacity of cationic polymer adjuvants.

3.6. In vitro

As shown in Figure S7, the cell viability of the cells decreased with increasing drug concentration, because the inhibitory effect of high concentration drugs on macrophages was more obvious. However, even at a high concentration of 2 μg/mL, the cell viability of functionalized SA-Man-PEI was 19.2% higher than that of PEI. The toxicity of single or double functionalized PEI to macrophages was significantly reduced, and the cell viability was significantly higher than unfunctionalized-PEI, because the toxicity of PEI was primarily caused by its higher positive due to charge density after the modification of PEI on the positive charge density was reduced and the toxicity was weakened. According to the macrophage viability results, the cell viability of SA-Man-PEI was higher than SA-PEI or Man-PEI at a polymer drug concentration of 2ug/mL, while the cell viability of Man-PEI was higher than SA-PEI, and the cell viability was correlated with the reduction of amino groups. This further showed that the toxicity was related to the positive charge density.

Macrophages were used for in vitro cell culture experiments. SA-Man-PEI, Man-PEI, and SA-PEI were mixed with FITC-OVA and compounded to prepare nanoparticles at a mass ratio of 0.04, with contrast groups set up using PEI/FITC-OVA and FITC-OVA, and macrophages were used as the blank group. After the macrophages were stained with DAPI, the nuclei appeared blue under an inverted fluorescence microscope, and the endocytosis of FITC-OVA bound to the polymer was observed as bright green.

As shown in Figure 7, no visible green fluorescence was observed in the blank and FITC-OVA samples and the green fluorescence intensity in the PEI was extremely low, with very few green fluorescent areas, indicating that the macrophages in PEI exhibited endocytosis on the composite particles of PEI/FITC-OVA, hut had a poor ability to take up composite particles. Compared to PEI, the green fluorescent areas Man-PEI and SA-PEI significantly increased, indicating that the modification of PEI by mannose and hydrophobic monomers could enhance the endocytosis of the macrophages to the composite particles. This was possibly related to the following reasons. The modification effect of mannose on PEI was beneficial for increasing the targeting effect of the polymer composite particles, with mannose receptors on the surfaces of the macrophages, which was more conducive to promoting the targeted uptake of mannose-containing composite particles by the macrophages. The cell surface was composed of phospholipid bimolecules, and the modification effect of hydrophobic monomers on PEI was more conducive to the entry of composite particles into the cell membrane, promoting the uptake action of macrophages.

Figure 7.

Inverted fluorescence microscope observations of endocytosis of composite particles by macrophages.

Figure 7 also shows that the green fluorescence area in the SA-Man-PEI was noticeably the largest, and green fluorescence overlapped with most of the blue fluorescence, indicating that the nanoparticles of SA-Man-PEI/FITC-OVA were obviously absorbed by the macrophages and their uptake was significantly enhanced. Therefore, in this study, the dual modification of PEI by mannose and stearyl chloride significantly increased the cell targeting and macrophage uptake. This modification enhanced the uptake function of macrophages and theoretically upregulated MHC class II molecule expression, allowing activated macrophages to act as antigen-presenting cells. After engulfing the invader, macrophages could display the protein fragments of the invader to helper T cells via MHC class II molecules. Therefore, SA-Man-PEI, as a carrier for nucleic acid vaccines, was beneficial in promoting a rapid immune response and better protecting the body.

4. Conclusions

In this study, Man-PEI, SA-PEI, and SA-Man-PEI with a certain proton buffering ability were prepared through single and double modification of PEI with mannose and stearyl chloride. These three types of composite particles, each with an average particle size of less than 200 nm, were prepared using OVA as the model antigen through electrostatic adsorption and hydrophobic interactions. SA-Man-PEI, which had a mannose-targeting effect and hydrophobic structure promoting cell uptake, had a small particle size and uniform distribution. Meanwhile, the storage stability of SA-Man-PEI at 37 °C and 4 °C was better than that of single-functionalized PEI. This study verified that the adsorption effect of functionalized-PEI on OVA exceeded 60%, while still maintaining the corresponding proton buffering effect. In the in vitro, macrophage uptake and toxicity experiments, functionalized-PEI significantly reduced cytotoxicity and promoted macrophage uptake. Therefore, functionalized-PEI, especially SA-Man-PEI, could not only enhance the storage stability of the composite particles, but also theoretically promote the body’s specific immune response effectively, providing an alternative carrier for nucleic acid vaccines.

Funding Statement

This work was financially supported by the Basic Research Expenses System Pilot Project of the Hebei Academy of Sciences (2022PF03-2, 2024PF01), the High-level Talents Cultivation and Funding Project of Hebei Academy of Sciences (2023G05), the Natural Science Foundation of Hebei Province (H2021208006), and the Hebei Natural Science Foundation (H2021302001).

Authors’ contributions

Lu Bai: Conceived and designed the study, Performed all the experiments, Collected and analyzed the·data, Methodology, Validation, Writing-original draft, Writing-review & editing. Xiaoqi Chen: Validation, Synthesized materials, Characterized molecular structure and particle size. Chengyu Li: Synthesized composite particles, Performed adsorption experiments, Investigation. Haijun Zhou: Performed experiments·on·SEM and·proton buffering capabilities, Collected the·data. Yantao Li: Analyzed the·data, Validation, Writing-review & editing. Jijun Xiao: Analyzed the·data, Validation. Fen Zhang: Funding resources, Validation, Writing-review & editing. Hua Cheng: Validation, Assisted in cytotoxicity and endocytosis experiments. Mengmeng Zhou: Conceived and designed the study, Performed all experiments, Validation, Writing-review & editing. All authors have read and agreed to the published version of the manuscript.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data availability statement

The data presented in this study are available on request from the corresponding author.