Cholesterol-conjugated polyethylenimine 25K for safe gene delivery into human carcinoma cells

Low cytotoxicity is a major requirement of a safe gene delivery vehicle. In this study, cholesterol-conjugated polyethylenimine 25K (PEI25K-CL) was used to evaluate the buffer capacity, the complexing ability with plasmid DNA, pEGFP, colloidal properties, pEGFP expression and transfection efficiency, and cytotoxicity in HeLa cervical carcinoma cells. The buffer capacity of PEI25K-CL was close to that of PEI25K. The PEI25K-CL/pEGFP and PEI25K/pEGFP complexes were completely retained at the polymer/DNA ratios of above 0.5:1. The particle size of PEI25K-CL/pEGFP complexes was 220 ± 7 to 402 ± 34 nm, depending on the polymer/DNA ratio. The pEGFP expression and transfection efficiency of the PEI25K-CL/pEGFP complexes were higher than those of pEGFP and lower than those of the PEI25K/pEGFP and Lipofectamine TM 2000/pEGFP complexes. The cytotoxicity of the PEI25K-CL/pEGFP complexes was significantly lower than that of the PEI25K/pEGFP complexes. This study concluded that PEI25K-CL could be an efficient and safe gene delivery vehicle.


INTRODUCTION
The viral vector has been an effective vehicle and practical in clinics currently (Duan, 2018;Ong et al., 2019). Among nonviral vectors, polyethylenimine 25K (PEI25K) has been reported as an effective gene delivery carrier among cationic polymers (Jiang et al., 2017;Ryu et al., 2018). Modification of PEI has been reported to improve the transfection efficiency of the gene into the cells (Lisha Mali et al., 2021;Wu et al., 2020).
Plasmid DNA is a gene-based strategy. It has potential to be an alternative therapy for many diseases including cancer (Shimada, 2018), inherited retinal disease (Zhang et al., 2021), and cardiovascular disease (Gorabi et al., 2018).
Safe gene delivery is a vital issue that serves the vehicle to be effective and practical for clinical application and treatment (Clanchy et al., 2008;Ma et al., 2019). The objective of this current study was to investigate PEI25K-CL on the improved transfection efficiency of pEGFP and low cytotoxicity in HeLa cervical carcinoma cells.

Materials
Polyethylenimine (branched, 25 KDa) was purchased from Sigma-Aldrich (St. Louis, MO). PEI 25K-CL (1:1 mole ratio) was provided by Dr. Boon-ek Yingyongnarongkul, Department of Chemistry and Center of Excellence for Innovation in Chemistry, Faculty of Science, Ramkhamhaeng University, Bangkok, Thailand. The XLarge DNA Ladder was purchased from GeneDireX (Las Vegas city, NV). pEGFP encoding enhanced green fluorescent protein was purchased from Clontech (Palo Alto, CA). HeLa human cervical carcinoma cells were obtained from the American Type Culture Collection (ATCC, Rockville, MD).

Plasmid DNA preparation
pEGFP was extracted and purified from DH5α Escherichia coli containing pEGFP using the Geneaid Plasmid Maxi Kit (Taipei City, Taiwan). The plasmid DNA was extracted according to the manufacturer's instructions. The purified plasmid was dissolved in tris-ethylenediamine tetraacetic acid (EDTA) buffer, pH 8.0. DNA concentration was measured at 260 nm using a cuvette photometer (BioPhotometer, Eppendorf AG, Hamburg, Germany).

Preparation of PEI25K-CL/pEGFP complexes
PEI25K-CL (0.1 mg/ml) in sterile water and pEGFP (1 mg/ml) in tris-EDTA buffer, pH 8.0, were diluted in ultrapure water or in Minimum Essential Medium (MEM). The PEI25K-CL and pEGFP complexes based on weight ratios of 0.125:1 to 2.5:1 were formed by adding a pEGFP solution to a PEI25K-CL solution, pipetted up-down for 15 times, and allowing spontaneous complex assembling for 15-20 minutes at room temperature (Weecharangsan et al., 2021).

Measurement of buffer capacity of PEI 25K-CL
PEI 25K-CL (1:1 molar ratio) was dissolved in ultrapure water at a concentration of 1 mg/ml and then diluted with 10 ml of a 0.9% sodium chloride solution to a concentration of 0.1 mg/ml. The PEI 25K-CL solution was then titrated with a 0.1 N hydrochloric acid solution of 0.5-8.0 µmol. Simultaneously, the pH of the PEI 25K-CL solution was monitored using a pH meter (Bench 700, Oakton Instrument, IL). The buffer capacity of PEI 25K-CL was calculated by dividing the amount of hydrochloric acid by pH change at a high resistance of pH change (Sinko, 2006).

Particle size, polydispersity index, and zeta potential
The particle size, polydispersity index, and zeta potential of the PEI25K-CL/pEGFP complexes in ultrapure water were determined using the Zetasizer Nano Series (Malvern Instruments Ltd., Worcestershire, UK) at the temperature of 25°C.

In vitro transfection and pEGFP expression
HeLa cells were plated in a 24-well plate at a density of 2 × 10 4 cells/well with MEM + 10% fetal bovine serum (FBS) for 24 hours. 1 µg/well of pEGFP was gently mixed with 0.5-2.5 µg/well of PEI25K-CL and incubated for 15 minutes at room temperature. The PEI25K-CL/pEGFP complexes were subjected to HeLa cell transfection for 4 hours, and the transfected cells were rinsed with phosphate buffered saline (PBS) pH 7.4 and incubated with a complete growth medium at 37°C for 24 hours. The Lipofectamine TM 2000/pEGFP complexes at liposome/DNA ratio of 0.92/1 (v/w) were used as control. The cells were observed under a fluorescence microscope (Eclipse TS100, Nikon, Tokyo, Japan). The number of GFP-expressed cells was counted. The transfection efficiency was defined as the ratio of GFP expressed cell number to the plate area (Paecharoenchai et al., 2012).

Cytotoxicity of PEI25K-CL
HeLa cells were seeded at a density of 8 × 10 3 cells/ well in a 96-well plate in MEM containing 10% FBS and grown for 24 hours. PEI25K-CL and PEI25K at the concentration of 0 to 20 µg/ml were added to each well and incubated for 4 hours. Cells treated with a growth medium were used as control. After 4 hours, the cells were then rinsed with PBS pH 7.4 and incubated with a growth medium for 24 hours. The 3-(4,5-dimethylthiazol-2yl)-2,5-diphenyl tetrazolium bromide assay was used to detect the cytotoxicity of PEI25K-CL (Weecharangsan et al., 2014).

Statistical analysis
The data were collected as mean ± SD. The significant difference in the data was analyzed using Student's t-test and analysis of variance as p < 0.05.

Buffer capacity of PEI25K-CL
Buffer capacity of PEI25K-CL and PEI25K was calculated at the high region resistance of pH change 7.12 ± 0.05 to 7.76 ± 0.04 and 7.01 ± 0.07 to 7.77 ± 0.12, respectively, at the amount of HCl of 1.5 × 10 −3 µmol (Fig. 1). The buffer capacity of PEI25K-CL and PEI25K was 0.0023 ± 0.0002 and 0.002 ± 0.0002, respectively. The buffer capacity of PEI25K-CL was not different from that of PEI25K. However, the amount of HCl needed for neutralization of PEI25K was higher than that of PEI25K-CL for the pH range of 7.0-7.7. This could be due to the amount of free NH 2 group of PEI25K being higher than that of PEI25K-CL.

Electrophoresis of PEI25K-CL/pEGFP complexes
Gel electrophoresis was used to examine the PEI25K-CL/pEGFP complexes and PEI25K/pEGFP complexes (Fig. 2). The electrophoretic mobility of pEGFP of the PEI25K-CL/pEGFP and PEI25K/pEGFP complexes was completely retained at the polymer/DNA ratios of above 0.5:1. At the polymer/DNA ratio of 0.5:1, the PEI25K-CL/pEGFP complexes exhibited higher DNA band intensity than that of the PEI25K/pEGFP complexes. This suggested that PEI25K-CL had a lower ability to form a complex with pEGFP than that of PEI25K. Figure 3 shows the particle size, polydispersity index, and zeta potential of the PEI25K-CL/pEGFP and PEI25K/pEGFP complexes performed in ultrapure water. The particle size, polydispersity index, and zeta potential of the PEI25K-CL/pEGFP and PEI25K/pEGFP complexes were 220 ± 7 to 402 ± 34 and 271 ± 12 to 462 ± 8 nm, 0.39 ± 0.05 to 0.51 ± 0.16 and 0.47 ± 0.07 to 0.6 ± 0.07, and 32.9 ± 1.4 to 43.7 ± 1.7 and 37.2 ± 4.6 to 43.5 ± 1.8 mV, respectively.

Colloidal properties of PEI25K-CL/pEGFP complexes
The particle size, polydispersity index, and zeta potential of the PEI25K-CL/pEGFP and PEI25K/pEGFP complexes were dependent on the polymer/DNA ratio. At the polymer/DNA ratios of 0.5:1 and 2.5:1, the particle size of the PEI25K-CL/pEGFP complexes was significantly larger than that of the PEI25K/ pEGFP complexes, while at the polymer/DNA ratios of 0.5:1 and 2.5:1 the particle size of the PEI25K-CL/pEGFP complexes was significantly smaller than that of the PEI25K/pEGFP complexes. At the polymer/DNA ratio of 0.5:1, the polydispersity index and zeta potential of the PEI25K-CL/pEGFP complexes were significantly lower than those of the PEI25K/pEGFP complexes while not significantly different at other polymer/DNA ratios.

pEGFP expression of PEI25K-CL/pEGFP and transfection efficiency
pEGFP expression of the PEI25K-CL/pEGFP and PEI25K/pEGFP complexes in HeLa cells analyzed by fluorescent microscopy is shown in Figure 4. PEI25K-CL/pEGFP exhibited lower pEGFP expression than that of PEI25K-CL/pEGFP. Transfection efficiency of pEGFP into HeLa cells of the PEI25K-CL/pEGFP complexes was 8 ± 4 to 679 ± 295 cells/cm 2 depending on the polymer/DNA (Fig. 5A). The transfection efficiency of pEGFP into HeLa cells of the PEI25K-CL/pEGFP complexes was significantly higher than in cells treated with pEGFP (Fig. 5A) while significantly lower than in cells treated with the PEI25K/ pEGFP complexes and Lipofectamine TM 2000/pEGFP complexes (Fig. 5B).

Cytotoxicity of PEI25K-CL
The cytotoxicity of PEI25K-CL and PEI25K in HeLa cells is shown in Figure 6. The cell viability of PEI25K-CL and PEI25K was 23.6% ± 0.8% to 99.5% ± 1.5% and 20.0% ± 0.9% to 97.2% ± 3.4%, respectively. The cell viability of PEI25K-CL and PEI25K was not different at the polymer concentration of 0.1 µg/ ml, while that of PEI25K-CL was significantly lower than that of PEI25K at the polymer concentration of all above 0.1 µg/ml.

DISCUSSION
Our study demonstrated that PEI25K-CL had the ability to deliver plasmid DNA and yielded an efficient transfection efficiency and low cytotoxicity. The low cytotoxicity and  efficient transfection efficiency of PEI25K-CL could be due to the amine groups of PEI25K partially substituted by cholesterol (Weecharangsan et al., 2021).
PEI25K-CL had a buffer capacity similar to that of PEI25K and a lower neutralizing ability of HCl at the pH range of 7.12 ± 0.05 to 7.76 ± 0.04. Wang et al. (2002) showed that PEI1.8K-CL and PEI10K-CL had a high buffer capacity in the pH range of 5-7. Weecharangsan et al. (2021) showed that the buffer capacity of PEI25K-CA was not different from that of PEI25K.
PEI25K-CL had the ability to condense plasmid DNA, pEGFP. Jiang et al. (2010) exhibited that polyethyleniminecholesterol cationic lipopolymer could condense plasmid DNA. Wang et al. (2002) showed that PEI1.8K-CL and PEI10K-CL could condense DNA to the nanosized and positively charged complexes.   The colloidal property often dictates the cellular entry mechanism of carrier/DNA complexes. The PEI25K-CL/pEGFP complexes yielded nanosized and positively charged particles. Li et al. (2018) demonstrated that polyethylenimine-cholesterol/ miR was positively charged and nanosized.
PEI25K-CL improved pEGFP expression and transfection efficiency and low cytotoxicity in HeLa oral carcinoma cells. Jiang et al. (2010) exhibited that polyethylenimine-cholesterol lipopolymer/DNA had a good transfection efficiency in A549 and MCF-7 cells. Wang et al. (2002) showed that the PEI1.8K-CL/ pCMS-EGFP and PEI10K-CL/pCMS-EGFP complexes had high GFP expression in Jurkat cells. Remant et al. (2020) demonstrated that cholesterol-grafted low-molecular-weight PEI enabled siRNA delivery into K562 myeloid leukemia cells depending on the degree of substitution of PEI backbone. Sun et al. (2014) reported that PEI-cholesterol liposomes improved cellular uptake and endosomal escape of cisplatin, as did Wu et al. (2020) in hydrophobization of PEI with cholesterol of siRNA.
PEI25K-CL had low cytotoxicity in HeLa oral carcinoma cells. Gusachenko et al. (2009) showed that lipophilic conjugated-PEI25K had low cytotoxicity at particular modification of amino groups of the polymer. Jiang et al. (2010) exhibited that the polyethylenimine-cholesterol cationic lipopolymer/ DNA complexes were low cytotoxic in A549 and MCF-7 cells. Wu et al. (2020) exhibited that hydrophobization of PEI with cholesterol was low cytotoxic compared with PEI. Wang et al. (2002) showed that PEI1.8K-CL and PEI10K-CL had slight cytotoxicity in Jurkat cells. Sun et al. (2014) exhibited that PEIcholesterol liposomes did not show noticeable systemic toxicity in H22 hepatoma-bearing mice. Li et al. (2018) demonstrated that polyethylenimine-cholesterol was nontoxic in CAL-27 cells at the concentration of 0-20 µg/ml. Lin et al. (2017) demonstrated that PEI25K induced autophagy in mouse fibroblast cells. Anwer et al. (2013) showed that PEG-cholesterol lipopolymer/plasmid IL-12 did not exhibit worsened side effects in the chemotherapy treatment of a phase I trial.

CONCLUSION
This study concluded that PEI25K-CL had potential to be an efficient and safe gene delivery carrier.

AUTHORS' CONTRIBUTIONS
WW and BY contributed to the concept and design. WW contributed to the data acquisition and interpretation and the writing and final approval of the manuscript. PO contributed to the research supervision.

CONFLICTS OF INTEREST
The authors have no conflicts of interest to declare.

FUNDING
This study was supported by the Faculty of Pharmacy, Srinakharinwirot University (034/2556).

ETHICAL APPROVALS
This study does not involve experiments on animals or human subjects.

DATA AVAILABILITY
All data generated and analyzed are included within this research article.

PUBLISHER'S NOTE
This journal remains neutral with regard to jurisdictional claims in published institutional affiliation.