Direct Correlation Between Zeta Potential and Cellular Uptake of Poly(methacrylic acid) Post-Modified with Guanidinium Functionalities

1 Introduction

The delivery of drugs using nanoparticles and polymers is expected to be able to address issues such as low drug stability, side effects, and drug resistance issues.1 The cellular uptake of the drug carrier by cells is often identified as a key parameter to the efficient action of drugs. A high cellular uptake is often in directly correlates with the toxicity of the drug.2 It is therefore not surprising that copious reports are available that investigate the influence of size, shape, and surface chemistry of the drug carrier on drug delivery.3 The cellular uptake of nanoparticles and polymers is directed by a two-step process. The initial surface attachment to the cellular membrane is followed by the internalization step, usually by endocytosis. It is therefore plausible that the surface chemistry of the polymer nanoparticles will play a pivotal role in cell uptake as it is the interface to the biological world. The composition of the cell surface is dominated by negatively charged species such as heparan sulfate proteoglycans, but also other negatively charged compounds such as phosphatidylserine and anionic phospholipids are present.4 It is therefore not surprising that cell-penetrating peptides (CPPs)5 that are rich in the positively charged arginine are frequently used to enhance the entry into cells. Decoration of nanoparticles with Tat peptide6 or oligoarginine7 was frequently used to enhance the delivery of nanoparticles, but a similar effect can be observed when nanoparticles are coated with cationic polymers such as chitosan[1] and polyethylene imine.8

Inspired by CPPs with their abundance of arginine, researchers have mimicked these biological molecules using guanidinium-rich polymers.9 Attachment of guanidinium functionalities to polymers, which coincides with the introduction of positive charges along the polymer, can significantly enhance the cellular uptake.9 The arrangement and distribution of cationic groups along the polymer can have detrimental effects on the properties with arginine residues bundled in a block-like structure showing superior properties to evenly distributed charges along the chain.10 Moreover, a direct relationship between the zeta potential, which is one measure for the surface charge, and the cellular uptake has been reported.11

Despite the frequent use of cationic polymers to enhance cellular uptake, concerns in regards to their cytotoxicity remain. The high toxicity of these polymers even facilitates their use as antimicrobial agents.12 As a result, the design of drug delivery carriers based on cationic polymers needs to be carefully planned. Recently, a zwitterionic polymer based on arginine displayed a high cellular uptake but no cytotoxicity.13 The balance of negative and positive charges led the formation of an almost neutral polymer. However, the polymer still displayed enhanced cellular uptake while the polymer was non-toxic even at high concentrations. Although the mechanism of the interaction of this polymer with cells is unknown, the observation may suggest that polymers carrying positive and negative charges may display overall superior properties as drug carriers.

In this paper, a systematic study was put in place that seeks to answer the question if there is a direct correlation between the zeta potential of guanidinium-functionalized polymers and the cellular uptake. Therefore, poly(methacrylic acid) will be post-functionalized with agmatine to generate polymers with various ratios of cationic and anionic functional groups (Scheme 1). Agmatine is a guanidine derivative formed by decarboxylation of arginine, which was previously attracted attention as an endogenous ligand for the I₁ receptor, which is located on the plasma membrane.14 The structural feature of agmatine is essential for enhanced cellular uptake due to its guanidine group, which can interact with the phosphate or other anionic moieties on the cell surface. Therefore, several reports have described the preparation of chitosan–agmatine bioconjugates, dextran–agmatine bioconjugates, and polyamidoamine bearing agmatine for efficient drug delivery system.15

In contrast to earlier work,16 the two charges are not present as zwitterionic structure, but they are randomly distributed along the polymer chain. The aim of this study will be to identify the effect of the fraction of agmatine on the toxicity toward A2780 ovarian cancer cells and to evaluate the relationship between the net surface charge of the polymer and the cellular uptake.

2 Experimental Section

2.4 Cytotoxicity Tests

The cytotoxicity tests were performed by a standard sulforhodamine B colorimetric proliferation assay (SRB assay). The ovarian cancer cell lines A2780 were seeded in 96-well cell culture plates at a density of 4000 cells per well for 24 h. The polymers were sterilized by filtration through 0.45 μm sterile membrane in bio-safety cabinet and then the growth medium was replaced with 100 μL fresh 2 × concentrated medium. Subsequently, 100 μL of serially diluted (2 × dilution) polymers were added to the plates, followed by incubation at 37 °C for 48 h.

After that, the culture medium was removed and the living cells were fixed in 10% (w/v) TCA (trichloroacetic acid) for 30 min at 4 °C. Next, the plates were washed five times with water. After removal of water, the TCA-fixed cells were stained using 100 μL of 0.4% (w/v) SRB dissolved in 1% acetic acid and incubated for 20 min at room temperature in dark, followed by washing five times with 1% acetic acid. After the plates were air dried for 2 h, the dye (SRB) was solubilized with 200 μL of 10 × 10⁻³ m Tris buffer. The absorbance was determined at 490 nm using a Bio-Rad BenchMarkmicroplate reader. Finally, the data were analyzed and plotted using GraphPad Prism 6.0 (non-treated cells were used as controls). The cell viability was calculated using optical density (OD):

2.8 Characterization

2.8.1 Nuclear Magnetic Resonance Spectroscopy

¹H NMR spectra were recorded by using Bruker Avance III 400 MHz NMR equipped with ultrashielded magnet and BBFO probe. Samples were analyzed in deuterated solvent (D₂O). The theoretical molecular weight (M_n,theo) was calculated as follow

(1)

2.8.2 Size Exclusion Chromatography

Size exclusion chromatography analyses were performed in Agilent modular system, comprising of an auto-injector, a PL aquagel-OH 8 μm bead size guard column (50 × 7.5 mm), two PL aquagel-OH MIXED-M 8 μm bead size columns (300 × 7.5 mm), a differential refractive index detector and a UV detector (λ = 305 nm). 0.02% w/v sodium azide solution in Milli-Q water was used as eluent at a flow rate of 0.8 mL min⁻¹ at 30 °C. The apparent molecular weights and polydispersities were determined with a calibration based on linear polyethylene glycol/polyethylene oxide (PEG/PEO) standards with narrow molecular weight distributions. The molecular weights of standards are ranging from 106 to 1.5 × 10⁶ g mol⁻¹.

2.8.3 Fluorescence Spectrophotometer

Cellular uptake studies at different time intervals were carried out using Cary Eclipse fluorescence spectrophotometer. The excitation and emission wavelengths were 495 nm and 519 nm.

2.8.4 Dynamic Light Scattering

Zeta potential measurements were performed on a Malvern Zetasizer Nano ZS (He−Ne laser 633 nm, max 5 mW) with a light scattering angle of 173° and then the data were recorded by Malvern Zetasizer Software version 6.20. Samples in the concentration of 1 mg mL⁻¹ were used for analysis without filtration. The measurements were run for three times at 25 °C.

3 Results and Discussion

MAA was polymerized in the presence of CPADB as a RAFT agent and AIBN as the initiator. The ¹H NMR spectrum and SEC analysis of PMAA, shown in Figure 2 and Figure 3, respectively, confirmed that PMAA was successfully synthesized with a good control over the dispersity. After 3 h of polymerization, 60% of MAA was consumed and a monomodal distribution was obtained resulting in PMAA of 60 repeating units and a M_n = 111 300 g mol⁻¹ (SEC) and Ð = 1.19, according to aqueous SEC using PEG/PEO standards.

The aim of this work was the design of polymers with negative and positive charged functionality at various ratios to identify the relationship between the overall charge and the cellular uptake. Post-functionalization with agmatine was identified as a suitable means to introduce both functionalities as this can ensure even distribution of both charges along the chain. Initial attempts included post-functionalization of activated PMAA such as poly(N-succinimidyl methacrylate). However, the disparate solubilities of the lipophilic polymer and the hydrophilic agmatine only led to negligible yields due to the absence of a common solvent. Therefore, PMAA was directly functionalized with various amounts of agmatine in water (Table 1). This experiment was carried out without using any coupling agents such as EDC/NHS since the modification of COOH group with NHS ester, which is the typical intermediate in this reaction, may lead to the insolubility of products in water. After a reaction time of 24 h, the polymers were purified by dialysis against water. The degree of agmatine conjugation was determined by ¹H NMR analysis in D₂O (Figure 2). The methylene functionalities at around 2.98 and 3.2 ppm, respectively, are indicative of the successful reaction with agmatine.18 The degree of side chain substitution was determined from the peak of methyl group in polymer backbone (δ ≈ 0.6–1.4 ppm) to the side chain group, and is referred to as a percentage value. At low agmatine concentration, the agmatine conjugation reached 100%. However, an excess of amine could not improve the maximum agmatine content of around 60 mol% indicating that steric hindrance may play a role. In addition, the introduction of more cationic charges may lead to repulsion along the polymer chain. Repulsion may lead to unfavorable chain stretching while a conjugation efficiency of around 50% leads to a balance of negative and positive charges. Molecular weight analysis using SEC revealed a monomodal distributions and low dispersity (Table 1 and Figure 3). The molecular weight is significantly higher than the expected value as a result of the analysis in water and the used PEG/PEO standards. Moreover, according to the previous study, it was reported that SEC was found to be the least accurate than other common laboratory methods such as ¹H NMR and MALDI-TOF in the determination of the molecular weight of polymers.19 Nevertheless, the SEC analysis shows that side reactions such as crosslinking reactions are absent as evident by the narrow-molecular-weight distribution.

The polymers were analyzed in water and 0.1 m NaCl using DLS. Hydrodynamic diameters below 5 nm confirm the existence of single polymer chains indicating that aggregate formation despite the presence of negative and positive charges is absent. Analysis of the zeta potential revealed a direct relationship between the amount of guanidine functionalities and the surface charge. It is interesting to note that in Milli-Q water the polymer Agm.38 has a slight negative charge although more guanidinium functionalities are present.

3.1 Cellular Uptake and In Vitro Cytotoxicity of Polymers

Human ovarian cancer cells (A2780) were chosen for the biological evaluation. Initially, the cytotoxicity of the agmatine-conjugated copolymers listed in Table 1 was tested using sulforhodamine B assay (SRB assay). Polymers obtained via post-functionalization modification were shown to be nontoxic. The negative charged carboxylate functionalities are therefore counterbalancing the typically toxic cationic charges. This is in agreement with earlier reports where zwitterionic polymers prepared using an arginine-based monomer showed no toxicity.13 In contrast to this report where negative and positive charges are distributed along the polymer backbone, zwitterionic polymers carry both charges in adjacent positions. According to Figure S1 (Supporting Information), it is illustrated that all polymers were found to be nontoxic to cells up to concentrations of 200 μg mL⁻¹.

Following the in vitro cytotoxicity, the uptake of the post-functionalized polymers was monitored in order to evaluate the effect of arginine groups and the increasing positive charge. The polymers were therefore incubated with A2780 cells for 1 h and the uptake was monitored using laser scanning confocal microscopy and flow cytometry (Figures 3-5). Both techniques reveal that the fluorescence-labeled polymers were taken up (Figures 3-5). Interestingly, the increase in polymer uptake was in almost linear fashion with the amount of conjugated agmatine. The cationic-charged polymers possess a higher uptake because the cell membrane is anionic as a result of the anionic head group of phospholipids and the presence of carbohydrates such as sialic acid.20

The results obtained from flow cytometry were subsequently normalized by taking the individual fluorescence intensity (Table 1) into account. As illustrated in Figure 6, PMAA exhibits poor uptake, but still higher than that measured by confocal microscopy. Although fluorescent microscopy can be considered more reliable than flow cytometry,21 the differences may not be as significant taking the error bars into account. Another possibility is that flow cytometry only measures the overall fluorescence of each cell, which may lead to errors from the polymer that is absorbed on the surface of the cell, but not internalized by the cells. In contrast, confocal microscopy will be able to distinguish the fluorescence caused by absorbed polymer and polymer in the cytosol.21

Moreover, the remaining solution that was used to culture the cells was analyzed and the decline in fluorescence intensity was recorded as a measure of the absorbed or internalized polymer. The internalization analysis over a period of time indicated around of 40% of the original agmatine conjugated polymers was taken up in a period of 45 h, while there was little uptake for PMAA (Figure 1). These results show little internalization of PMAA, which agrees with the confocal microscopy analyses. Interesting is the behavior of the agmatine-functionalized polymers over a period of 45 h. Although flow cytometry and laser scanning confocal microscopy suggest a slightly higher uptake of Agm.38 after 1 h, the differences between any of the modified PMAA polymers seem negligible when monitored over a long period of time. In overall, the cell uptake analyses demonstrated the efficient uptake of homopolymers with arginine functionality, and some internalization of PMAA, which might be as a result of the interaction of anionic polymer with cationic lipid domains in the cell membrane.20

4 Conclusions

We successfully synthesized polymers bearing both COOH and guanidinium functionalities. The polymers were incubated with A2780 human ovarian cancer cells at 37 °C in order to investigate in vitro cytotoxicity and cell uptake. The cytotoxicity analyses proved that the guanidinium-functional polymers did not introduce any toxicity, most likely due to the presence of negatively charged carboxylate functionalities. The polymers conjugated with agmatine were rapidly taken up by the cells. The increasing amount of guanidinium functional groups in the polymer chain led to a less negative zeta potential, which led to a highly efficient uptake. In contrast, PMAA itself was internalized to a negligible amount only. Interestingly, although higher agmatine functionalization led to a slightly higher uptake after 1 h, monitoring the cell uptake over a longer period of time suggests that there is only a small advantage of having higher guanidinium content. However, it can be concluded that the efficient cellular uptake and virtually nontoxic nature of agmatine-conjugated polymer enable this polymer to become an efficient carrier for drugs.