A Low-Molecular-Weight Polyethylenimine/pDNA-VEGF Polyplex System Constructed in a One-Pot Manner for Hindlimb Ischemia Therapy

Peripheral arterial disease (PAD) is often characterized by continued reduction in blood flow supply to limbs. Advanced therapeutic strategies like gene therapy could potentially be applied to limb ischemia therapy. However, developing a gene delivery system with low toxicity and high efficiency remains a great challenge. In this study, a one-pot construction was used to integrate vector synthesis and polyplex fabrication simultaneously in a simple and robust manner. We fabricated an interpenetrating gene delivery network through the physical interaction between low-molecular-weight polyethylenimine (PEI 1.8 kDa) and plasmid DNA (pDNA) and the chemical bonding between PEI and glutaraldehyde (GA), which was named the glutaraldehydelinked-branched PEI (GPEI) polyplex. The final GPEI polyplex system was pH-responsive and biodegradable due to the imine linkage and it could successfully deliver desired vascular endothelial growth factor (VEGF) pDNA. Compared with PEI (25 kDa)/pDNA polyplexes, GPEI polyplexes showed lower cytotoxicity and higher transfection efficiency both in vitro and in vivo. In addition, we demonstrated that GPEI polyplexes could efficiently promote the formation of new capillaries in vivo, which may provide a practicable strategy for clinical hindlimb ischemia therapy in the future.


Introduction
Characterized by continued reduction in blood flow to one or more limbs, peripheral arterial disease (PAD) is mainly caused by atherosclerosis, and it can cause severe ischemic limb disease [1]. The prevalence of PAD has been gradually increasing in recent years due to the changes in dietary structure, aging of the population, and the growing frequency of vascular surgery. For example, the average risk of developing arteriosclerosis for diabetic patients is 19 times higher than that of healthy individuals, and these diabetic patients are more likely to suffer from PAD [2,3]. Besides, thromboangiitis obliterans and other chronic diseases can induce severe PAD [4,5]. Clinically, current therapeutic approaches to PAD include antiplatelet therapy (aspirin and clopidogrel) [6], anticoagulants (cilostazol) [7], statin therapy [8], and vascular dilation drugs (Anplag) [9] as well as cell-based therapy [10]. Even with such multiple choices, existing clinical strategies are still fraught with some challenges like low therapy efficacy and poor prognosis. Gene therapy may provide an advanced therapeutic strategy for PAD, especially chronic lower extremity ischemia, via the delivery of nucleic acid-based drugs (DNA or

Preparation of GPEI Polyplexes
According to the formula showing in Table 1, GPEI polyplexes were prepared at various PEI 1.8 kDa/pDNA weight/weight (w/w) ratios. First, pDNA solution (20 ng/μL) was mixed with PEI 1.8 kDa solution (2 mg/mL) and H2O. These mixtures were incubated at room temperature for 30 min (first condensation). Then, GA solution (0.125%) was added and these mixtures were heated in a water bath at 65 °C with oscillation for 30 min (second condensation). Finally, the prepared GPEI polyplex solution was stored at 4 °C for further use.

Particle Size, Zeta Potential and Morphology Measurements
The particle size and zeta potential of GPEI polyplexes at various w/w ratios were measured by Particle Size Analyzer (Brookhaven Particle Size Analyzer 90 Plus, Brookhaven Instruments Corporation, Holtsville, NY, USA), and PEI 25 kDa polyplexes (w/w ratio = 2) were used as the positive control. The morphology of GPEI polyplexes at w/w ratio of 5 was examined by transmission electron microscopic (TEM) (Tecnai G2 spirit Biotwin, Thermo Fisher Scientific, Waltham, MA, USA).

In Vitro Cytotoxicity
In vitro cytotoxicity of GPEI polyplexes was measured by using Cell Counting Kit-8 (CCK-8) (DOJINDO LABORATORISE, Shanghai, China) reagent. First, HUVECs were seeded into 96-well plates (1 × 10 4 cells/well) and cultured in cell incubator for about 24 h. Next, the medium was replaced by 50 µL new culture medium (without FBS), and 10 µL GPEI polyplex solutions at various w/w ratios (from 1 to 20) were added into wells for 4-h and 24-h incubation. Each ratio was repeated 6 times. Then, each well was replaced by 10 µL of CCK-8 reagent and 50 µL of new culture medium. About 2 h later, the absorbance was measured by multifunctional microplate reader (SpectraMax M3 Multi-Mode Microplate Reader, Sunnyvale, CA, USA). PEI 25 kDa polyplexes at the same w/w ratios were used as the control.

Intracellular Uptake
In order to have a direct observation of the intracellular uptake of GPEI polyplexes, the intracellular localization of GPEI polyplexes in HUVECs was measured by a Super-Resolution Multiphoton Confocal Microscope (SMCM) (TCS SP8 STED 3X, Lecia, Germany) [22]. GPEI polyplex solutions (w/w ratio = 5, 10) were prepared as regular methods and the pDNA was labeled with the fluorophore Cy3 (GenePharma, Shanghai, China). HUVECs were seeded into 12-well plates (5 × 10 5 cells/well) with cell slides (WHB Scientific, Shanghai, China) and cultured for about 24 h. Then, the medium was replaced by 1 mL new culture medium (without FBS) and 200 µL GPEI polyplex solutions. After 2 h, cells were washed five times with phosphate buffer saline (PBS) and 2 mL LysoTracker Green (100 nM) (Beyotime, Shanghai, China) were added to mark the lysosome for 50 min at 37 • C. Next, cells were washed five times with PBS and 1 mL trypan blue solution (0.4%) (Yeasen Biotech, Shanghai, China) was added to mark viable cells for 2 min at room temperature. Then, cells were washed five times with PBS and 1 mL 4% paraformaldehyde was added to fix cells for 30 min at room temperature. Finally, cells were washed five times with PBS and 2 mL DAPI (1 µg/mL) (Roche Diagnostics, Mannheim, Germany) was added to mark the cell nucleus for 4 min at 37 • C and investigated by SMCM. Similarly, the 4-h intracellular uptake was conducted by the same method.

In Vitro Cell Transfection
HUVECs were seeded into 24-well plates ((5~10) × 10 4 cells/well) and cultured in cell incubator for about 24 h. Then, the medium was replaced by 500 µL new culture medium (without FBS) and 100 µL GPEI polyplex solutions at various w/w ratios (from 1 to 20) were added into wells for 4 h incubation, and each ratio was repeated for 3 times. PEI 25 kDa polyplexes (w/w ratio = 2) were used as the positive control, naked pDNA as the negative control, and PBS as the blank control. Then, each well was replaced by 1 mL new culture medium. Seventy-two hours later, flow cytometry (FCM) (BD LSRFortessa, Franklin, NJ, USA) was utilized to determine the percentage of cells with green fluorescence.

Hindlimb Ischemia Model Study
All animal experiments were conducted strictly in accordance with the guidelines approved by the Regulations for the Administration of Affair Concerning Laboratory Animals for Shanghai Jiao Tong University, the National Institutes of Health Guide for Care and Use of Laboratory Animals Sixteen healthy Sprague-Dawley rats (male, weight 150-200 g) were raised under standard laboratory conditions in the laboratory animal facility in the school of pharmacy, Shanghai Jiao Tong University. One week later, rats weighing about 250 g were intraperitoneally injected with pentobarbital sodium (50 mg/kg) for anesthesia, and operations were performed according to reported article as well as the instructional video [23]. After the operation, rats were randomly divided into four groups (n = 4). When the skin wound at surgical sites was completely healed, drugs (0.1 mL/100 g) were injected intramuscularly at the model site once every three days. The blank control group was injected with normal saline. The negative control group was injected with VEGF pDNA (100 ng/µL). The positive control group was injected with PEI 25 kDa polyplexes (w/w ratio = 2) containing VEGF pDNA (100 ng/µL). And the GPEI group was injected with GPEI polyplexes (w/w ratio = 5) containing VEGF pDNA (100 ng/µL). In addition, each rat was injected with 5-Bromo-2-deoxyuridine (BrdU) (Sigma-Aldrich, St. Louis, MO, USA) once every three days [24]. After four weeks of treatment, rats were sacrificed for further study.

In Vivo Cytotoxicity
In vivo cytotoxicity of GPEI polyplexes was evaluated by hematoxylin-eosin (HE) staining. When rats were sacrificed, the main organs (including heart, liver, spleen, lungs and kidneys) of rats were immediately harvested and fixed in 4% paraformaldehyde, sectioned, and stained with HE. After staining, the glass slides were observed under optical microscope.

In Vivo Therapeutic Effects
Gastrocnemii at the surgical site of sacrificed rats were collected and fixed in 4% paraformaldehyde and assessed by HE staining. Angiogenesis was assessed by both CD34 immunofluorescence staining [25] and CD31 immunohistochemical staining [26], and the capillary density of gastrocnemii was analyzed semi-quantitatively by software ImageJ. 5-Bromo-2-deoxyUridine (BrdU) antibody was used to label proliferating cells in the tissue. Besides that, part of gastrocnemii was immediately frozen Pharmaceutics 2019, 11, 171 6 of 14 at −80 • C after the sacrifice, which were then used to quantitatively evaluate the expression of VEGF protein in the tissue by Western blotting (WB).

Statistical Analysis
Data of each independence measurement were presented as mean values ± standard deviation (S.D.). Statistical analysis was tested by independent sample t-test, and statistical significance was assigned as *: p < 0.05, **: p < 0.01, or ***: p < 0.001.

Agarose Gel Electrophoresis (AGE)
The ability of GPEI to condense pDNA into nanoparticles was measured by AGE. As shown in Figure 2, when the w/w ratio was over 0.03:1, there was no pDNA electrophoretic band present, which indicated that GPEI had a strong ability to condense nucleic acids efficiently, even at a small w/w ratio. gastrocnemii was immediately frozen at −80 °C after the sacrifice, which were then used to quantitatively evaluate the expression of VEGF protein in the tissue by Western blotting (WB).

Statistical Analysis
Data of each independence measurement were presented as mean values ± standard deviation (S.D.). Statistical analysis was tested by independent sample t-test, and statistical significance was assigned as *: p < 0.05, **: p < 0.01, or ***: p < 0.001.

Characterization of Polyplexes
The ability of GPEI to condense pDNA into nanoparticles was measured by AGE. As shown in Figure 2, when the w/w ratio was over 0.03:1, there was no pDNA electrophoretic band present, which indicated that GPEI had a strong ability to condense nucleic acids efficiently, even at a small w/w ratio.

Particle Size, Zeta Potential, and Morphology Measurements
The particle size and zeta potential of the nanoparticles have a great influence on the delivery efficiency of GPEI polyplexes. As shown in Figure 3A, particle sizes of GPEI polyplexes at various w/w ratios were in the range of 200 to 300 nm, which were significantly lower than that of PEI 1.8 kDa. And the polydispersity index (PDI) of GPEI polyplexes was low ( Figure 3B), confirming that particle sizes of GPEI polyplexes in the system was uniform. We mixed VEGF pDNA, PEI 1.8 kDa, and GA aqueous solution together in a one-pot manner to prepare GPEI polyplexes. The PDI of some groups exceeded 0.2, which may be caused by uneven cross-linking during preparation. Excessive and uneven cross-linking of glutaraldehyde and amine groups may cause multiple polyplexes to link together, contributing to heterogeneity in particle size. Therefore, polyplexes should be oscillated continuously during heating to ensure a narrow range of particle size. The zeta potential of GPEI polyplexes ranged from 30 to 50 mV, which was higher than that of PEI 1.8 kDa ( Figure 3C). In addition, TEM images also directly demonstrated that particle sizes were around 150 nm at the w/w ratio of 5 ( Figure 3D). All above data showed that the GPEI polyplexes formed by twice-condensation of PEI 1.8 kDa with GA can efficiently enhance the condensation capacity of PEI 1.8 kDa, as well as reduce the particle sizes of polyplexes.

In Vitro Cytotoxicity
Safety and biocompatibility are important indicators for evaluating the practicability of applying polycation in biological systems. Thus, in vitro cytotoxicity of GPEI polyplexes was measured by

Particle Size, Zeta Potential, and Morphology Measurements
The particle size and zeta potential of the nanoparticles have a great influence on the delivery efficiency of GPEI polyplexes. As shown in Figure 3A, particle sizes of GPEI polyplexes at various w/w ratios were in the range of 200 to 300 nm, which were significantly lower than that of PEI 1.8 kDa. And the polydispersity index (PDI) of GPEI polyplexes was low ( Figure 3B), confirming that particle sizes of GPEI polyplexes in the system was uniform. We mixed VEGF pDNA, PEI 1.8 kDa, and GA aqueous solution together in a one-pot manner to prepare GPEI polyplexes. The PDI of some groups exceeded 0.2, which may be caused by uneven cross-linking during preparation. Excessive and uneven cross-linking of glutaraldehyde and amine groups may cause multiple polyplexes to link together, contributing to heterogeneity in particle size. Therefore, polyplexes should be oscillated continuously during heating to ensure a narrow range of particle size. The zeta potential of GPEI polyplexes ranged from 30 to 50 mV, which was higher than that of PEI 1.8 kDa ( Figure 3C). In addition, TEM images also directly demonstrated that particle sizes were around 150 nm at the w/w ratio of 5 ( Figure 3D). All above data showed that the GPEI polyplexes formed by twice-condensation of PEI 1.8 kDa with GA can efficiently enhance the condensation capacity of PEI 1.8 kDa, as well as reduce the particle sizes of polyplexes.
we could also find that there were fewer cells alive in 24-h cytotoxicity test compared with that in 4h cytotoxicity test. The cell viability of some groups exceeded 100%, probably because the cytotoxicity of polyplexes with small w/w ratios was low and cells were still proliferating during the basal medium incubation. One reason accounting for the low cytotoxicity of GPEI in this study is that GPEI here is prepared by conjugating the aldehyde groups in GA with amine groups in PEI (1.8 kDa) to generate imine linkage, which is pH-responsive and could degrade under the acidic environment of endosomes.

In Vitro Cytotoxicity
Safety and biocompatibility are important indicators for evaluating the practicability of applying polycation in biological systems. Thus, in vitro cytotoxicity of GPEI polyplexes was measured by CCK-8 reagent in this study. As shown in Figure 4, the cytotoxicity of GPEI polyplexes was significantly lower than that of PEI 25 kDa polyplexes. GPEI showed no obvious cytotoxicity when at low w/w ratios such as 1, 3, 5, and 10. The cytotoxicity of GPEI was higher with increasing w/w ratios, and the cell viability of GPEI was only 50% when the w/w ratio reached 30. From the results we could also find that there were fewer cells alive in 24-h cytotoxicity test compared with that in 4-h cytotoxicity test. The cell viability of some groups exceeded 100%, probably because the cytotoxicity of polyplexes with small w/w ratios was low and cells were still proliferating during the basal medium incubation. One reason accounting for the low cytotoxicity of GPEI in this study is that GPEI here is prepared by conjugating the aldehyde groups in GA with amine groups in PEI (1.8 kDa) to generate imine linkage, which is pH-responsive and could degrade under the acidic environment of endosomes.
Pharmaceutics 2019, 11, x 7 of 14 CCK-8 reagent in this study. As shown in Figure 4, the cytotoxicity of GPEI polyplexes was significantly lower than that of PEI 25 kDa polyplexes. GPEI showed no obvious cytotoxicity when at low w/w ratios such as 1, 3, 5, and 10. The cytotoxicity of GPEI was higher with increasing w/w ratios, and the cell viability of GPEI was only 50% when the w/w ratio reached 30. From the results we could also find that there were fewer cells alive in 24-h cytotoxicity test compared with that in 4h cytotoxicity test. The cell viability of some groups exceeded 100%, probably because the cytotoxicity of polyplexes with small w/w ratios was low and cells were still proliferating during the basal medium incubation. One reason accounting for the low cytotoxicity of GPEI in this study is that GPEI here is prepared by conjugating the aldehyde groups in GA with amine groups in PEI (1.8 kDa) to generate imine linkage, which is pH-responsive and could degrade under the acidic environment of endosomes.

Intracellular Uptake
Intracellular uptake is a critical step in the delivery of pDNA by GPEI. In this study, we performed intracellular localization of GPEI polyplexes to capture the intracellular uptake of GPEI polyplexes in HUVECs. The pDNA used in GPEI polyplexes was labeled with Cy3, which could emit red fluorescence under excitation. Therefore, the red fluorescent signals indicated the existence of pDNA, which is also the representation of the GPEI polyplex position, while the green fluorescent signals in the system suggested the positions of endosomes. As shown in Figure 5, a few GPEI polyplexes were endocytosed into cytoplasm after 2 h and many polyplexes were encapsulated in endosomes after 4 h. These results indicated that GPEI polyplexes could be endocytosed into cells and deliver pDNA successfully. In addition, some red dots appeared in the nuclear area ( Figure 5), indicating that some of GPEI polyplexes could escape from endosomes and readily enter into nucleus for gene expression.

Intracellular Uptake
Intracellular uptake is a critical step in the delivery of pDNA by GPEI. In this study, we performed intracellular localization of GPEI polyplexes to capture the intracellular uptake of GPEI polyplexes in HUVECs. The pDNA used in GPEI polyplexes was labeled with Cy3, which could emit red fluorescence under excitation. Therefore, the red fluorescent signals indicated the existence of pDNA, which is also the representation of the GPEI polyplex position, while the green fluorescent signals in the system suggested the positions of endosomes. As shown in Figure 5, a few GPEI polyplexes were endocytosed into cytoplasm after 2 h and many polyplexes were encapsulated in endosomes after 4 h. These results indicated that GPEI polyplexes could be endocytosed into cells and deliver pDNA successfully. In addition, some red dots appeared in the nuclear area ( Figure 5), indicating that some of GPEI polyplexes could escape from endosomes and readily enter into nucleus for gene expression.

In Vitro Cell Transfection
Transfection efficiency is an important aspect when evaluating the performance of a gene delivery system. As shown in Figure 6, the transfection efficiency of GPEI group become higher with the increasing w/w ratios (from 1 to 20). Compared to that in PEI 25 kDa group (w/w ratio = 2), the transfection efficiency in GPEI group was higher. Significant improvement in transfection efficiency could be seen in GPEI group, when compared to that in naked pDNA group. No fluorescent signal was detected in the naked DNA group, indicating that naked DNA could hardly enter cells and been expressed, therefore, a safe and efficient delivery vector is required.

In Vivo Cytotoxicity
Main organs including heart, liver, spleen, lungs, and kidneys stained with HE were imaged by optical microscope to evaluate long-term toxicity. As shown in Figure 7, the staining results in all groups were comparable with each other, which means there is no significant difference among different groups. This may be explained by two reasons. Firstly, polyplexes in this study were injected intramuscularly rather than intravenously at the model site, and therefore they may not circulate around the body and cause the obvious toxicity in various organs. Secondly, the injected polyplexes are macromolecules that cannot be easily absorbed and distributed after local injection. They may be first degraded into small molecules and then absorbed and distributed, but the metabolized small molecules during this stage have low toxicity and thus cannot cause significant organ toxicity.

In Vitro Cell Transfection
Transfection efficiency is an important aspect when evaluating the performance of a gene delivery system. As shown in Figure 6, the transfection efficiency of GPEI group become higher with the increasing w/w ratios (from 1 to 20). Compared to that in PEI 25 kDa group (w/w ratio = 2), the transfection efficiency in GPEI group was higher. Significant improvement in transfection efficiency could be seen in GPEI group, when compared to that in naked pDNA group. No fluorescent signal was detected in the naked DNA group, indicating that naked DNA could hardly enter cells and been expressed, therefore, a safe and efficient delivery vector is required.

In Vivo Cytotoxicity
Main organs including heart, liver, spleen, lungs, and kidneys stained with HE were imaged by optical microscope to evaluate long-term toxicity. As shown in Figure 7, the staining results in all groups were comparable with each other, which means there is no significant difference among different groups. This may be explained by two reasons. Firstly, polyplexes in this study were injected intramuscularly rather than intravenously at the model site, and therefore they may not circulate around the body and cause the obvious toxicity in various organs. Secondly, the injected polyplexes are macromolecules that cannot be easily absorbed and distributed after local injection. They may be first degraded into small molecules and then absorbed and distributed, but the metabolized small molecules during this stage have low toxicity and thus cannot cause significant organ toxicity.

In Vivo Therapeutic Effects
According to the experimental results in vitro, the w/w ratio of 5 and 10 might be the ideal formulations for research in vivo, considering the cytotoxicity and transfection efficiency. Although the GPEI 5:1 showed a lower transfection efficiency, its cell viability was higher than that of GPEI 10:1. Therefore, we chose GPEI 5:1 for in vivo animal model study so that animals would face fewer safety risks.
In order to evaluate the in vivo therapeutic effect of the GPEI polyplexes, hindlimb ischemia model and corresponding evaluation methods were established. As shown in Figure 8A, there were visible gaps in the muscle fiber of blank group and naked VEGF group. In contrast, PEI 25 kDa 2:1 group and GPEI 5:1 group showed denser muscle fibers and presented no symptoms of muscular atrophy.

In Vivo Therapeutic Effects
According to the experimental results in vitro, the w/w ratio of 5 and 10 might be the ideal formulations for research in vivo, considering the cytotoxicity and transfection efficiency. Although the GPEI 5:1 showed a lower transfection efficiency, its cell viability was higher than that of GPEI 10:1. Therefore, we chose GPEI 5:1 for in vivo animal model study so that animals would face fewer safety risks.
In order to evaluate the in vivo therapeutic effect of the GPEI polyplexes, hindlimb ischemia model and corresponding evaluation methods were established. As shown in Figure 8A, there were visible gaps in the muscle fiber of blank group and naked VEGF group. In contrast, PEI 25 kDa 2:1 group and GPEI 5:1 group showed denser muscle fibers and presented no symptoms of muscular atrophy. BrdU is a thymidine analog that can incorporate into the DNA molecule being replicated during the DNA synthesis phase (S phase) instead of thymine. Therefore, the cell proliferation of gastrocnemii can be analyzed by detecting the fluorescence signal of BrdU. As shown in Figure 8B, the signal density in each group was not significantly different, implying that the proliferation of cells in each group was comparable. CD34 immunofluorescence staining and CD31 immunohistochemical staining were used for morphological observation to evaluate neovascularization in gastrocnemii. In Figure 8C, although some CD34 signals could be detected in the blank group and naked VEGF group, almost no intact vascular morphology was observed in gastrocnemii. In contrast, a greater number of areas of neovascularization were clearly observed in the PEI25 kDa 2:1 group and the GPEI 5:1 group. And results in Figure 8D also showed that the PEI25 kDa 2:1 group and GPEI 5:1 group showed more newborn capillaries. The capillary density of gastrocnemii was then analyzed semi-quantitatively ( Figure 8E,F). The result was consistent with the morphological observation and confirmed our expectation that the gene therapy of VEGF could promote capillary angiogenesis.
The VEGF protein expression in gastrocnemii in vivo was determined by WB and GAPDH served as the loading control. As shown in Figure 8G,H, VEGF protein levels in two polyplex groups were both significantly higher than that in other two groups, indicating that VEGF pDNA in the polyplex groups was successfully delivered, transcribed, translated, and finally released.
All experimental results above demonstrated that GPEI polyplexes could deliver VEGF pDNA into cells efficiently and increase the expression level of VEGF protein in vivo. According to our previous work [19,20,27,28], plasmid DNA encoding anti-VEGF-shRNA or anti-VEGF-siRNA [29] could be delivered by polyplexes to inhibit the expression of VEGF, which could further limit the growth of tumors. Therefore, we could conclude that the increased VEGF expression and capillary BrdU is a thymidine analog that can incorporate into the DNA molecule being replicated during the DNA synthesis phase (S phase) instead of thymine. Therefore, the cell proliferation of gastrocnemii can be analyzed by detecting the fluorescence signal of BrdU. As shown in Figure 8B, the signal density in each group was not significantly different, implying that the proliferation of cells in each group was comparable.
CD34 immunofluorescence staining and CD31 immunohistochemical staining were used for morphological observation to evaluate neovascularization in gastrocnemii. In Figure 8C, although some CD34 signals could be detected in the blank group and naked VEGF group, almost no intact vascular morphology was observed in gastrocnemii. In contrast, a greater number of areas of neovascularization were clearly observed in the PEI25 kDa 2:1 group and the GPEI 5:1 group. And results in Figure 8D also showed that the PEI25 kDa 2:1 group and GPEI 5:1 group showed more newborn capillaries. The capillary density of gastrocnemii was then analyzed semi-quantitatively ( Figure 8E,F). The result was consistent with the morphological observation and confirmed our expectation that the gene therapy of VEGF could promote capillary angiogenesis.
The VEGF protein expression in gastrocnemii in vivo was determined by WB and GAPDH served as the loading control. As shown in Figure 8G,H, VEGF protein levels in two polyplex groups were both significantly higher than that in other two groups, indicating that VEGF pDNA in the polyplex groups was successfully delivered, transcribed, translated, and finally released.
All experimental results above demonstrated that GPEI polyplexes could deliver VEGF pDNA into cells efficiently and increase the expression level of VEGF protein in vivo. According to our previous work [19,20,27,28], plasmid DNA encoding anti-VEGF-shRNA or anti-VEGF-siRNA [29] could be delivered by polyplexes to inhibit the expression of VEGF, which could further limit the growth of tumors. Therefore, we could conclude that the increased VEGF expression and capillary formation demonstrated in this study come from the high translation of delivered VEGF pDNA rather than simple upregulated expression induced by injury. As reported, VEGF protein could promote the formation of new capillaries, thereby improving the blood supply of the ischemic hind-limbs, increasing the nutrient transport of the ischemic hindlimbs, and even alleviating muscle atrophy to some extent. Therefore, we believe the GPEI/pDNA-VEGF polyplex system developed in this study could provide a practicable strategy for future clinical hindlimb ischemia therapy. formation demonstrated in this study come from the high translation of delivered VEGF pDNA rather than simple upregulated expression induced by injury. As reported, VEGF protein could promote the formation of new capillaries, thereby improving the blood supply of the ischemic hindlimbs, increasing the nutrient transport of the ischemic hindlimbs, and even alleviating muscle atrophy to some extent. Therefore, we believe the GPEI/pDNA-VEGF polyplex system developed in this study could provide a practicable strategy for future clinical hindlimb ischemia therapy.

Conclusions
In order to achieve effective treatment of PAD, researchers have focused on gene therapy with angiogenic factors [30]. As the most classic non-viral polycation carrier, PEI has been widely studied in recent years [31]. Many scientists have put a lot of effort to reducing the toxicity of PEI, with methods like modifying PEI with lipids [32], polyethylene glycol (PEG) [33,34], and fluorine [35,36], and crosslinking low-molecular-weight PEI with biodegradable chemical bonds [18][19][20]. However, all these methods complicate the system and limit practical use since intricate synthesis, purification, and the use of organic reagents are often involved. In contrast, the preparation of GPEI polyplexes in our one-pot manner was simple, practicable, and convenient. It avoided long-term synthesis and purification and the use of organic solvents, and saved lots of time. However, the GPEI polyplex presented here still lacks targeting groups, thus it can only be injected locally at the ischemic site, which needs further improvement.
In summary, the novel gene vector GPEI has been developed by twice-condensation in our study. Compared with other existing polycation vectors, advantages of GPEI are as follows: (1) the preparation method of GPEI polyplexes is quite simple, robust, and practicable, allowing researchers to reproduce the delivery system successfully; (2) the reaction was fast and no organic solvents were used in this biomaterial system, avoiding long-term synthesis and time-consuming purification; (3) low-molecular-weight PEI (1.8 kDa) was condensed twice to form the interpenetrating network, enhancing the stability of polyplexes and reducing nucleic acid leakage effectively; (4) the imine linkage conjugated in the polymer could respond to acid environment, promoting the release of nucleic acid intracellularly; and (5) GPEI polyplexes linked with imine bonds are biodegradable, leading to low cytotoxicity.
In our constructed polyplex delivery system, GPEI can condense nucleic acid into nanoparticles effectively with homogeneous particle size and appropriate zeta potential and the nucleic acid can be successfully delivered and released. GPEI polyplexes show low cytotoxicity and high transfection efficiency both in vitro and in vivo. Therefore, we believe the GPEI/pDNA-VEGF polyplex system described here could be a promising candidate for the future hindlimb ischemia therapy.
Author Contributions: W.-E.Y. conceived the initial idea, the conceptualization, and the study design, and participated in the data extraction and analysis and revised the manuscript. X.G., Z.Y., Y.X., X.Z., and Z.F. participated in the study design, searched databases, extracted and assessed studies and helped draft the manuscript. X.G. wrote the manuscript. All authors have read and approved the final manuscript.