A pH Responsive Ternary Gene Carrier based on Branched-Poly(ethylenimine) and Poly(2-ethyl-2-oxazoline)-block- Poly(methacrylic acid)

This study demonstrates new ternary polyion gene carriers, based on the P H -responsive diblock copolymer poly(2-ethyl-oxazoline)- block -poly(methacrylic acid) (PEOz- b -PMAA), and the branched-poly (ethylenimine) (B-PEI). This study complexes the plasmid DNA with B-PEI and further with PEOz- b -PMAA to obtain ternary polyplexes (DNA/B-PEI/PEOz- b -PMAA). PMAA were partially dissociated under neutral P H with a negative charge, to attach to the positive charge surface of the B-PEI/DNA polyplex. The ternary polyplexes will also desorb and return to the original pre-poly complex to help gene release after cell uptake due to PMMA become neutral charge under an acid environment in endosome. The ternary polyplexes showed suitable mean particle size, low cytotoxicity, and acceptable transfection at P H 7.4 because of shielding of B-PEI by PEOz- b -PMAA. TEM morphology showed that the stable core-shell structure of ternary polyplexes at P H 7.4 collapsed and released plasmid at P H 5. Observations of cell uptake of the B-PEI/DNA polyplex and ternary polyplexes by CLSM revealed that ternary polyplexes started to accumulate after 3 h incubation and accumulated significantly after 6 h. In conclusion, the ternary polyplex improves cytotoxicity of the single B-PEI/DNA polyplex, and presents a P H -responsive behavior to enhance gene escape from the polyplex. The ternary polyplex constitutes a useful approach for gene carrier design.


Introduction
Non-viral vectors, such as cationic polymers, peptides, and liposomes, have attracted interest because of their advantages compared to viral systems. For example, research suggests that non-viral vectors are safer, with more flexible structures and chemical properties [1][2][3]. Among the various types of polymer gene carriers, branched polyethylenimine (B-PEI) is the most effective vector for delivering genes because the BPEI/DNA polyplexes exhibit great transgene expression in vitro and in vivo [4][5]. The high transfection efficiency of B-PEI is because of its high buffering capacity, which disrupts the endosomal membrane, and releases the polyplex into the cytoplasm [6]. The intrinsic high cytotoxicity of B-PEI limits its application in vivo. Excessive positive charge on the cationic polymer carriers' surface is toxic to cells and tissues [7], and adversely influences biodistribution of the complexes in vivo [8]. Poly (2-ethyl-2-oxazoline) (PEOz) is a water-soluble polyelectrolyte, with low toxicity and high hydrophilicity similar to poly (ethylene glycol) (PEG) [9] and it is approved by the FDA as a biocompatibility polymer. The charge density in the linear polyethylenimine (LPEI) in our past research strongly affected cell viability, and the PEOz-coupled LPEI reduced cationic charge density and cytotoxicity [10].
Studies in recent years have examined the multilayer gene carriers made by "layer by layer self-assembly" [11][12]. Trubetskoy used the poly-L-lysine (PLL) and succinylated PLL (SPLL) as polycations and polyanions to form the DNA/PLL/SPLL complex with a charge stoichiometry close to 1:1:1. The zeta potential of the previously mentioned complexes became negative, indicating effective surface recharging [13]. Koyama reported a new ternary gene carrier structure by preparing the carriers from DNA polycation as the core structure (binary polyplex), and covering them with a negative charge of PEG-carbocylic acid. This ternary polyplex showed a steady property instead of the binary polyplex appearing as an aggregated state in BSA [14]. As the cell uptake carried the polyplex into the endosome, the carbocylic acids protonated and lost hydrophilicy. The structure of the hydrophilic acids fused with the endosome membrane [15], to enhance escape from the endosome.

Synthesis of poly (2-ethyl-2-oxazoline)-poly (methacrylic acid), (PEOz-b-PMAA)
The diblock copolymer PEOz-b-PMAA was synthesized by free radical polymerization with the macroinitiator (PEOz) 2-ABCPA. The experiment dissolved (PEOz) 2-ABCPA in dichloromethane and heated it until 80°C, before adding MAA and stirring for 18 h. The product was purified by dialysis for 2 d using cutoff dialysis tubing with a molecular weight of 6000, and dried in a vacuum, obtaining the PEOz-b-PMAA product. The composition was analyzed by 1 H NMR.

Preparation of B-PEI/DNA polyplex and PEOz-PMAA/B-PEI/DNA ternary polyplex
Polyplex preparation included diluting the required amount of polymer and 1 μg of luciferase encoded plasmid pUHC-13-3 in 100 μl of distilled water and gently mixing. The polyplex formulation was incubated at room temperature for 30 m before using. After preparing the B-PEI/DNA polyplex, the ternary polyplex was created by mixing the B-PEI/DNA solution and the PEOz-b-PMAA solution, and incubated for 30 m before using.

Measurement of particle size and zeta potential
To measure the particle size and zeta potential, various formulations of polyplexes were prepared. Dynamic light scattering was used to decide the particle size of polyplex particles in distilled water at 25°C (Zetasizer 3000HS, Malvern Instruments, Worcestershire, U.K). Laser doppler micro-electrophoresis in the capillary electrophoresis cell of the Zetasizer 3000HS at 25°C determined the zeta potential of polyplexes prepared in distilled water. The mean values for these measurements were calculated from the data obtained in ten runs.

Cell culture and MTT assay
Human cervix carcinoma HeLa cells (1 x 10 4 cells/well) were cultured onto a plate with ninety-six wells in DMEM media supplemented with 10% FBS in a humid atmosphere of 5% CO2 at 37°C for 24 h. Then the growth medium was replaced with 200 μl medium that contained the desired amount of polymers and allowed to incubate for 72 h. The cytotoxicity of each sample was determined by measuring the cell viability using a tetrazolium dye (MTT) assay. 100 μl of the medium containing 10 μl of MTT PBS solution (5 mg/ml) was added. After incubation for 2 h, the formazan crystals were dissolved in 100 μl of DMSO/ EtOH solution. The absorbance of each well was measured using a microplate reader (Stat Fax 2100, Awareness, Palm City, FL, USA) at a test wavelength of 570 nm and a reference wavelength of 630 nm.

In vitro transfection and total protein assay
To evaluate the transfection potential of polyplexes, human cervix carcinoma HeLa cells (2 x 10 5 cells/well) were incubated in DMEM media supplemented with 10% FBS and plated in six-well plates in a humid atmosphere of 5% CO 2 at 37°C for 24 h. The growth medium (2 mL) contained DNA or polyplexes that contained 5 μg of plasmid incubated at 37°C. After incubation for 48 h, the cells were washed with cold PBS and then lysed using 400 μL of the 1X cell lysis buffer. The cell lysate was then transferred into Eppendorf tubes and centrifuged for 15 s at 12,000 rpm. Then luciferase activity was measured in relative light units (RLU), using a ninety-six-well plate luminometer (Wallac 1420 Multilabel Counter, Perkin-Elmer, USA) and a luciferase assay kit (Promega, Tokyo, Japan). The total amount of protein was measured at 595 nm using a protein assay.

Morphology of the polyplex
The polyplex morphology prepared as described above was observed by transmission electron microscopy (TEM, Hitachi, H-7500, Tokyo, Japan). To prepare the TEM sample, 10 μl of polyplex solution was placed onto a copper grid coated with carbon, tapped with a filter paper to remove water, and air-dried for 5 m. Then, the grid was negatively stained with 2 wt% uranyl acetate solution for 30 s, tapped with a filter paper to remove water, and vacuum-dried for 24 h.

Cellular uptake of polyplexes
The amino end group of B-PEI was used to label the Cy5.5 dye to observe the cellular uptake behavior of polyplexes by a confocal laserscanning microscope (CLSM). B-PEI was mixed with Cy5.5 NHS ester in DMSO at room temperature for 24 h. The product was purified by dialysis for 2 d in a dark bottle. Polyplex accumulated in HeLa cells was localized using LSM5 PASCAL CLSM (Carl Zeiss, USA). The HeLa cells were seeded on cover-slides for 24 h and were then treated with free DNA or polyplexes. After specific time, the cells were washed twice with PBS, and then LysoTracker was added to the culture medium without FBS and incubated for 20 m. Then the samples were stored at 4°C refrigerator for 16 h, removed from the formaldehyde solution, and we added DAPI water solution (10μg/ml) for 5 m.

Measurement of particle size and zeta potential
Laser light scattering estimated the particle size and zeta potential of the polyplexes. The polyplexes prepared in an aqueous solution of physiological salt (150 mM NaCl) or in distilled water measured the polyplex size to explain whether the ionic strength of the media influences polyplex stability.
Prepolycomplex B-PEI/DNA: To form prepolycomplex B-PEI/ DNA, plasmid DNA was complexed with B-PEI. Figure 2 shows the particle size and zeta potential of the B-PEI/DNA polyplex in distilled water. The B-PEI/DNA polyplex formed large particles (particles size were greater than 300 nm) as the polymer/DNA weight ratio (μg/μg) dropped below 1. When the B-PEI/DNA weight ratio was above 5, the mean diameter of particle sizes was around 150 nm, and the tendency kept constant. The PDI of the B-PEI/DNA polyplex grew after the weight ratio went above 5. The zeta potential of the polyplex expressed a similar tendency. Initially, the zeta potential indicated a negative charge as with a B-PEI/DNA weight ratio below 1. With increasing weight ratio of B-PEI/DNA, the zeta potential became positive, and the  value maintained an almost constant value of 30 mV with a weight ratio above 5. The previously mentioned results indicated that the B-PEI/ DNA polyplex formed a most stable conformation with a weight ratio around 5. The following experiments adopted the B-PEI/DNA weight ratio of 5, and selected the B-PEI/DNA weight ratio of 1、3 as contrast forms. The code "B-PEI mX" represents m for the B-PEI/DNA weight ratio (μg/μg). For example, the code B-PEI 5X means the B-PEI/DNA weight ratio equals 5, as used below.
Ternary polyplexes: This study prepared five compositions of PEOz-b-PMAA (E6M2, E6M4, E10M2, E10M4, and E15M2) to form ternary polyplexes by covering PEOz-b-PMAA polymers on the B-PEI/ DNA core structure. The ternary polyplexes with a longer hydrophilic segment PEOz from PEOz-b-PMAA (E10M2, E10M4, and E15M2) showed higher particle size (above 200 nm) and zeta potential (around 30 mV). Increased positive charges and large size may be due to the higher hydrophilic force from the longer PEOz segment, exposing the positively charged PEI parts to the outer polyplex. The section below discusses the two kinds of PEOz-b-PMAA with a short PEOz segment (that is, E6M2 and E6M4). Figure 3 shows the results of particle size and zeta potential for the inner core, B-PEI 5X, and the outer shell, PEOz-b-PMAA (E6M2 and E6M4), with various weight ratios based on DNA in distilled water. Comparing it with the pre-poly complex B-PEI/ DNA, the particle size increased slightly with a mean diameter from 150 nm to about 200 nm, but zeta potential decreased from 30 mV to about 10 mV. The results indicated that as the ternary polyplex weight ratio of PEOz-b-PMAA increased, the particle size and zeta potential decreased. This change may be due to the increasing neutralization effect between negative charges from PMMA and positive charges from B-PEI. Figure 4 and Figure 5 show similar results for B-PEI 3X and B-PEI 1X, but zeta potential decreased to about -20 mV. It may be due to the ternary polyplexes having more negative charges from PMMA at the same weight ratio conditions. The B-PEI parts are well-set inside the polyplex attracted by DNA, and the neutral PEOz parts are expected to come outside to provide less charge to the polyplex surface. The hydrophilic and neutral segment, PEOz, creates an optimal particle size and zeta potential to the polyplex for cell entry. The incoming negative charges from PMAA cover the positive charges of B-PEI/DNA. The greater the weight ratio of B-PEI/ DNA is, the more PEOz-b-PMAA is needed to neutralize the B-PEI/ DNA positive charge. In other words, the smaller the weight ratios of Polyplexes in salt-containing media: Figure 6 shows the results of ternary polyplexes prepared in an aqueous solution of 150 mM NaCl to discuss the stability of ternary polyplexes. The ternary polyplexes did not aggregate in salt-containing media. The results indicated that the neutral BPEI and LPEI polyplexes were prone to aggregate, because of the lack of electrostatic repulsion. The DNA/B-PEI/PEOz-b-PMAA ternary polyplexes were stable because they possessed a core-shell structure with a hydrophobic B-PEI/DNA core and a hydrophilic PEOz shell that prevented the polyplexes from aggregating.

In vitro transfection and total protein assay
This study evaluated in vitro transfection efficiencies of naked DNA, B-PEI/DNA polyplexes, and DNA/B-PEI/PEOz-b-PMAA (E6M2, E6M4) ternary polyplexes under various P/D ratios on the HeLa cell. The luciferase assay for the luciferase encoding plasmid determines transfection efficiency. This work used total protein production, an indirect measure of induced toxicity, to normalize the RLU. This work compared the transfections of ternary polyplexes with B-PEI/DNA polyplexes. The ternary polyplex prepared from B-PEI 3X showed a decreasing tendency with increasing PEOz-b-PMAA. However, the ternary polyplex formed from B-PEI 5X showed better transition efficiency than the ternary polyplex formed from B-PEI 3X. The ternary polyplex formed from B-PEI 5X did not decline, but remained at the same level. The above phenomena match the results of zeta potential, the surplus negative charges from PMMA of ternary polyplex prepared from B-PEI 3X diminish the cell uptake outcome, but not in 5X.   and E6M4) at various concentrations after 72 h incubation on HeLa cultures. This study treated the wells containing only media without polycations as positive controls and set them at a cell viability of 100%, calculating relative cell viability as [Abs] sample /[Abs] control x 100. The B-PEI exhibited stronger cytotoxicity with an IC 50 = 2.5 μg/ml compared to PEOz-b-PMAA with IC 50 = about 150 μg/ml for E6M2, and that of E6M4 was higher than 500 μg/ml. The cell viability of PEOzb-PMAA declined as the PMMA percentage increased. It may be due to E6M4 has more negative charge from PMMA segment than E6M2 which make E6M4 hard to be absorbed by cell. Figure 10 shows polyplex cytotoxicities with B-PEI and PEOz-b-PMAA at various concentrations after 72 h incubation on HeLa cultures. To obtain the difference in biocompatibility between B-PEI/DNA and the PEOz-b-PMAA ternary polyplex, this study selected the ternary polyplex with a zeta potential of about 0~10mV for the HeLa cell viability test and compared it with the B-PEI/DNA polyplex. The cell culture incubated with DNA was the control. The results indicated cell viability: B-PEI 1X > B-PEI 3X > B-PEI 5X. After we covered it with PEOz-b-PMAA, the cell viability of these ternary polyplexes maintained high viability independent of the content composition of the core structure. Decreased toxicity resulted from shielding the positively charged polymer in the core by the PEOz hydrophilic shell structure. The previously mentioned results indicated a slight positive charge of the PEOz does not damage the cells as in B-PEI.

Polyplex morphology by TEM observation
TEM observed the polyplex morphology. The study incubated the polyplexes in the phosphate buffer (P H =7) and succinate buffer (P H =5) solutions at room temperature for 1 h to observe the P H -sensitive behavior. The TEM images in Figure 11(a) show that free DNA has a random coil structure. Figure 11(b), (c) and (d) show the B-PEI/DNA polyplex does not have a core-shell, but a global structure. In this case, when the B-PEI wraps the DNA to form a compact particle, the DNA cannot react with uranyl acid (UA) easily. The images of B-PEI/DNA demonstrate a white contrast in the particle center. Unlike B-PEI, E6M2/DNA and E6M4/DNA polyplexes form an obvious core-shell micelle structure as shown in Figures 12(a)~(d). The inner core, P(EOz/EI) complex with DNA, is negatively stained by UA and exhibits a dark image, where the nuclei acid from the DNA bond with UO 2 (Ac) + due to UA dissociation [16]. The outer shell, the PEOz hydrophilic segment, provides less contrast. The hydrophilic and neutral charge segment PEOz of the shell may protect polyplexes from immune system destruction.    Cy5.5, LysoTracker, and DAPI after incubating B-PEI/DNA polyplexes and the ternary polyplex with HeLa cells. The current study used the LysoTracker and DAPI molecules as indicators in the acidic compartment and nucleus [17]. The CLSM traced the cell uptake process. The fluorescence signals of DAPI revealed a blue color, the signals of Lysotracker revealed a green color, and Cy5.5 was red. For B-PEI/DNA, the Cy5.5 signals intensified around the HeLa cell periphery. The B-PEI remained in the cell nucleus after 6 h. The cell uptake progress for the ternary polyplex was slower. Observations showed significant amounts of Cy5.5 signals at 6 h, and the ternary polyplex entered the cell nucleus at 12 h. The B-PBI/DNA polyplex showed a more obvious signal after 6 h, compared with the ternary polyplex because of the higher positive charge density on B-PBI for enhancing cell uptake.

Conclusion
This study synthesized the P H -responsive diblock copolymer PEOzb-PMAA by polymerizing the MMA monomer with the macroinitiator (PEOz) 2 -ABCPA. The study successfully prepared the ternary polyplex based on PEOz-b-PMAA and B-PEI, and evaluated the physicochemical and biological characteristics. The results showed the ternary polyplex efficiently covered up the positive charge of the B-PEI/DNA precomplex and offered higher cell viability than the B-PEI/DNA polyplex. The ternary polyplex formed a clear core-shell structure by complexing with DNA under TEM observation. These polyplexes show P H -sensitivity to destroy core-shell structures and gene release in acidic buffers at a P H of 5. The non-viral gene carrier based on polyion PEOz-b-PMAA creates an optimal particle size and zeta potential for the polyplex, and  The P H response of the polyplex was carried out at P H 5. The random coil structure from DNA interconnected with the broken particles and the core-shell structures broke under acid conditions. The ruptured ternary polyplex formed a gel-like state and stuck together. The P H responsive PEOz-b-PMAA produced by hydrogen bonding in an acidic environment may produce this state. The particle size showed the same results ( Figure 14). Different from the B-PEI/DNA polyplex, the particle size of the ternary polyplex showed a great rise in the acidic environment. In short, the P H -sensitivity from PEOz-b-PMAA of the ternary polyplex may help DNA release from the polyplexes in the endosome after cell uptake.