Application of an HIV gp41-Derived Peptide for Enhanced Intracellular

920

Bioconjugate Chem. 2008, 19, 920–927

Application of an HIV gp41-Derived Peptide for Enhanced Intracellular Trafficking of Synthetic Gene and siRNA Delivery Vehicles Ester J. Kwon, Jamie M. Bergen, and Suzie H. Pun* Department of Bioengineering, University of Washington, 1705 NE Pacific Street, Seattle, Washington 98195. Received December 5, 2007; Revised Manuscript Received January 25, 2008

Endosomal release is an efficiency-limiting step for many nonviral gene delivery vehicles. In this work, nonviral gene delivery vehicles were modified with a membrane-lytic peptide taken from the endodomain of HIV gp41. Peptide was covalently linked to polyethylenimine (PEI) and the peptide-modified polymer was complexed with DNA. The resulting nanoparticles were shown to have similar physicochemical properties as complexes formed with unmodified PEI. The gp41-derived peptide demonstrated significant lytic activity both as free peptide and when conjugated to PEI. Significant increases in transgene expression were achieved in HeLa cells when compared to unmodified polyplexes at low polymer to DNA ratios. Additionally, peptide-modified polyplexes mediated significantly enhanced siRNA delivery compared to unmodified polyplexes. Despite increases in transgene expression and siRNA knockdown, there was no increase in internalization or binding of modified carriers as determined by flow cytometry. The hypothesis that the gp41-derived peptide increases the endosomal escape of vehicles is supported by confocal microscopy imaging of DNA distributions in transfected cells. This work demonstrates the use of a lytic peptide for improved trafficking of nonviral gene delivery vehicles.

INTRODUCTION The development of safe and efficient gene delivery vehicles is necessary for clinical gene therapy. Synthetic vehicles offer potential advantages over viral-based vectors including safety and low immunogenicity, but transfection efficiencies from many of these systems remain suboptimal (1, 2). An effective nucleic acid delivery vehicle must mediate either cytoplasmic or nuclear delivery, depending on its therapeutic agent. Upon internalization, the majority of carriers accumulate within endocytic vesicles (3). Efficient release from these vehicles before lysosomal degradation has been shown to be a major barrier for many of these systems. Thus, it would be advantageous to develop methods to improve the endosomal release of new and existing vectors for future applications of nonviral gene delivery. Polyethylenimine (PEI) has been used widely as a nonviral vector and is among the most efficient of cationic polymer delivery vehicles (4). The efficacies of PEI and other protonatable cationic polymers as gene delivery agents has been attributed to their buffering properties, which have been hypothesized to mediate endosomal escape via the proton sponge mechanism (5, 6). In brief, pH-sensitive polymers buffer endosomes during the natural acidification process, resulting in counterion and water accumulation in the vesicles, eventually leading to osmolysis. However, the buffering capacity of PEI is correlated to the amount of PEI that is present (6, 7). Purification of free PEI from polyplexes after formulation results in polyplexes with only a slight charge excess of PEI. These polyplexes are much less efficient at mediating gene transfer than unpurified polyplexes (8, 9). This has implications for in ViVo delivery where unassociated polymer can be separated from DNA before endocytosis occurs. The development of agents that have more potent endosomal escape properties is therefore desirable. * To whom correspondence should be addressed. Phone: (206) 685-3488. Fax: (206) 616-3928. E-mail: [email protected].

In an effort to develop more potent mediators of endosomal escape, research has looked to nature for inspiration. Peptides taken from the membrane-disrupting domains of proteins, such as melittin (10-12), Tat (13-15), and hemagglutinin (16, 17), have been incorporated into nonviral gene delivery vectors, as reviewed elsewhere (18, 19). In this paper, we report the application of a peptide from the endodomain of HIV gp41 envelope glycoprotein found to have lytic properties from a peptide screen recently reported in the literature (20). This sequence of HIV gp41, corresponding to residues 783-806 of gp160, has been shown to have high membrane association and the potential to form multimers (21). Peptides from this region of gp41 have been shown to adopt amphipathic R-helical structuresandhavethepotentialtoformporesinmembranes(22,23). The goal of this work was to evaluate the lytic peptide from the endodomain of HIV gp41, HGP, as a mediator of enhanced intracellular trafficking of nonviral gene delivery vehicles. Our results demonstrate that modification of PEI with HGP increases transgene expression and siRNA knockdown, likely by increasing the efficiency of endosomal release.

METHODS Synthesis of HGP-Modified PEI. C-terminal cysteineterminated HGP peptide (LLGRRGWEVLKYWWNLLQYWSQELC) was synthesized and HPLC purified by GenScript Corporation (Piscataway, NJ). Branched PEI with molecular weight 25 000 (Sigma, St. Louis, MO) was modified with succinimidyl 4-[p-maleimidophenyl]butyrate (SMPB) purchased from Pierce (Rockford, IL) at a molar excess of 5 according to manufacturer’s protocol. SMPB-modified PEI was purified using a PD-10 column (GE Healthcare, Piscataway, NJ) and lyophilized. For peptide conjugation, SMPB-modified PEI was dissolved in DMF and reacted with 5 equiv HGP peptide in the presence of triethylamine (TEA) for 72 h. Unreacted peptide and TEA were removed by dialysis against a MWCO 10 000 membrane (Pierce) with water over 3 days. PEI concentration was determined by monitoring cuprammonium complexes formed by PEI and copper (II) acetate at 630 nm as described

10.1021/bc700448h CCC: $40.75  2008 American Chemical Society Published on Web 04/01/2008

HIV-Derived Peptide for Gene and siRNA Delivery

Bioconjugate Chem., Vol. 19, No. 4, 2008 921

previously (24). HGP conjugation efficiency was determined by quantifying absorbance at 280 nm using UV/vis spectrophotometry. Polyplex Formulation and Physicochemical Characterization. Polyplexes, complexes of polycations with nucleic acids, were formulated by adding equal volumes of polymer to nucleic acid at the desired charge ratio. The charge ratio is calculated on the basis of molar amount of polymer nitrogen to nucleic acid phosphate (N/P) ratio. Polyplexes were allowed to incubate for 10 min at room temperature to allow for complexation. Complexation of DNA was monitored by agarose gel electrophoresis. Polyplexes were loaded on a 0.8% agarose gel containing 0.3 µg/mL ethidium bromide. Hydrodynamic size was measured in triplicate using a ZetaPALS zeta potential and particle size analyzer (Brookhaven Instruments Corp., Holtsville, NY) as described previously (25). Dye Release Assay. Liposomes composed of L-R-phosphotidylcholine and cholesterol (Avanti Polar Lipids, Alabaster, AL) encapsulating 50 mM sulforhodamine B (Invitrogen, Carlsbad, CA) were prepared by extrusion through 100 nm polycarbonate membranes (Whatman, Florham Park, NJ) using the Avanti Mini-Extruder. Unencapsulated sulforhodamine B was removed by dialysis into 10 mM Tris pH 7.4, 20 mM NaCl, 0.1 mM EDTA. Liposome lysis was monitored by dequenching of sulforhodamine B fluorescence upon release from liposomes. Fluorescence intensity was monitored using a Tecan Safire2 microplate reader (Tecan Systems, Inc., San Jose, CA) by excitation at 565 nm and reading emission at 586 nm. Dye release was calculated as a percentage according to the equation % dye release )

Fm - F0 × 100 F100 - F0

(1)

where Fm is the fluorescence value after 15 min of incubation with peptide or polyplex at 37 °C, F0 is the fluorescence value of the initial liposome suspension, and F100 is the fluorescence value after incubation with a 0.05% Triton X-100 solution. Cell Culture. HeLa cells purchased from ATCC (Manassas, VA) were cultured in complete growth medium (minimum essential medium with 10% fetal bovine serum and antibiotics). Cells were passaged every 2-3 days. Plasmid Delivery to HeLa Cells, Evaluation of Transfection Efficiencies, and Cytotoxicity Assay. Transfection experiments were performed in triplicate. HeLa cells suspended in complete growth medium were seeded at 30 000 cells/well in 24-well plates and allowed to attach overnight. Polyplexes were formulated as described above using 1 µg of gWiz-Luciferase plasmid DNA (Aldevron, Fargo, ND) and then diluted in OptiMEM medium (Invitrogen). Cells were rinsed once with phosphate buffered saline, pH 7.4 (PBS), and incubated with polyplexes for 4 h at 37 °C in a 5% CO2 atmosphere. Cells were rinsed with PBS and medium was replaced with fresh complete growth medium. Cells were returned to a 37 °C, 5% CO2 atmosphere and luciferase expression was quantified after 48 h using a luciferase assay kit (Promega Corp., Madison, WI). Cells were washed with PBS, lysed with 200 µL of reagent lysis buffer (Promega Corp.), and frozen at -20 °C. 100 µL of luciferase substrate was added to 20 µL of lysate and luminescence was measured using a TECAN Safire2 microplate reader. Luminescence was integrated for 1 s and recorded in relative light units (RLU). Luciferase activity is reported in luminescence normalized by protein content (RLU/ mg), as measured by a BCA Protein Assay Kit (Pierce). Cytotoxity was measured by incubating polymer or polyplexes with HeLa cells as described above and adding the reagent 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-

sulfophenyl)-2H-tetrazolium (MTS) (Promega Corp.) after 24 h. Cells were incubated in a 37 °C, 5% CO2 atmosphere for 4 h and absorbance measurements of lysate at 490 nm were recorded. siRNA Delivery to HeLa Cells and Expression Analysis by Quantitative PCR and GAPDH Activity Assay. siRNA targeting endogenous glyceraldehyde-3-phosphate dehydrogenase (siGAPDH) was purchased from Ambion (Silencer GAPDH siRNA; Austin, TX), and control siRNA targeting green fluorescent protein (siGFP) was synthesized by Dharmacon (Lafayette, CO) with the following sequence: 5′-GACGUAAACGGCCACAAGUUC-3′ (sense) and 5′-ACUUGUGGCCGUUUACGUCGC-3′ (antisense). HeLa cells were seeded on a 24well plate as described for the plasmid transfection studies. For transfection, polyplexes were formulated by condensing either siGAPDH or siGFP with either PEI or PEI-HGP at an N/P ratio of 10 (60 pmol siRNA/well, triplicate samples). Polyplexes were diluted in 400 µL OptiMEM and incubated with cells for 5 h, after which the polyplex solution was removed and replaced with complete growth medium. 48 h after transfection, total RNA was isolated using an RNeasy Mini Kit (Qiagen, Valencia, CA). 500 ng of RNA from each sample was reverse-transcribed using Omniscript RT (Qiagen) and random hexamers as primers (Operon, Huntsville, AB). Quantitative PCR was performed using a model 7300 Real Time PCR system (Applied Biosystems, Foster City, CA) following universal thermal cycling parameters. GAPDH expression levels were determined in 20 µL reactions using TaqMan Universal PCR Master Mix (Applied Biosystems) and a TaqMan gene expression assay for human GAPDH, and were normalized by expression levels for β-actin (TaqMan gene expression assay for human ACTB). Relative GAPDH expression levels for each sample were determined on the basis of a comparison with untreated control samples, and were calculated by the 2-∆∆CT method (26). To measure GAPDH activity, an identical transfection experiment was performed, and cell lysate was collected 48 h after transfection. GAPDH activity was measured in the lysate using the fluorescence-based KDalert GAPDH Assay Kit (Ambion). For each formulation, the degree of reduction in GAPDH activity mediated by siGAPDH was assessed by comparing GAPDH activity levels in siGAPDH- versus siGFP-treated samples. Internalization and Binding Studies. Uptake and binding of PEI-HGP polyplexes relative to PEI polyplexes was investigated by flow cytometry. Nitro-2,1,3-benzoxadiazol-4-yl (NBD)labeled oligonucleotides were prepared as described previously (27). HeLa cells were seeded at 300 000 cells/well in 6-well plates and allowed to attach overnight. One hour before transfection, cells were washed with PBS, supplemented with Gibco OptiMEM medium, and then equilibrated to either 37 or 4 °C. Polyplexes were formulated at an N/P ratio of 3 as described above using 4 µg of NBD-labeled oligo. Polyplex solution was added to cells and allowed to incubate for 3 h at either 37 or 4 °C. Cells were then washed twice with cold PBS, trypsinized, and resuspended in complete growth medium. Cellassociated fluorescence was quantified by flow cytometry using a BD FACScan (BD Biosciences, Franklin Lakes, NJ). Live cell populations were identified using forward and side-scattering profiles, and polyplex uptake was assessed by NBD fluorescence using FlowJo flow cytometry analysis software. Cell association was reported as mean fluorescence in relative fluorescence units (RFU). Confocal Imaging. Confocal imaging was performed to determine the intracellular distribution of polyplexes. gWizLuciferase plasmid DNA was labeled with TOTO-3 (Invitrogen, Eugene, OR) at 1 dye per 100 base pairs and dialyzed into 10 mM Tris pH 7.4, 1 mM EDTA overnight at 4 °C. Glass coverslips were precoated with poly(L-lysine) in a 24-well plate

922 Bioconjugate Chem., Vol. 19, No. 4, 2008

Kwon et al.

Figure 1. Gel retardation assay to monitor DNA complexation in PEI-HGP and PEI polyplexes at various N/P ratios. M: New England Biolabs 1kb ladder.

and HeLa cells were seeded at 75 000 cells/well in complete medium. 3 µg of TOTO-3 labeled DNA was complexed with PEI or PEI-HGP at N/P ratio 3 as described above and added to cells with 2 mg/mL Alexa Fluor 488-labeled dextran (MW 10 000, Invitrogen) for 1 or 3 h in a 37 °C, 5% CO2 environment. Cells were then washed with PBS and Cell Scrub Buffer (Gene Therapy Systems, San Diego, CA), and then fixed for 7 min in 4% paraformaldehyde. Coverglasses were mounted onto a coverslip with Fluoromount-G (Southern Biotech, Birmingham, AL) and imaged with a Zeiss LSM 510 confocal microscope. Chloroquine-Mediated Transfection Studies. HeLa cells were seeded at 120 000 cells/well in 6-well plates and allowed to attach overnight. Polyplexes were formulated at an N/P ratio of 3 as described above using 4 µg of pEGFP-C1 (Clontech Laboratories, Inc., Mountain View, CA) plasmid DNA that codes for the enhanced green fluorescent protein (EGFP) reporter. Polyplex was diluted in OptiMEM and added to cells. A 5 mM chloroquine stock solution was diluted to a final concentration of 0.1 mM and cells were allowed to incubate for 4 h in a 37 °C, 5% CO2 atmosphere. Cells were rinsed with PBS and incubated for an additional 48 h in complete growth medium. Cells were imaged by brightfield and fluorescence microscopy. Transgene expression was quantified by EGFP fluorescence using flow cytometry as described above. EGFP expression is reported as the percent of EGFP positive cells. Tail-Vein Injection. Female C57/Bl6 mice were purchased from Jackson Laboratory (Bar Harbor, ME). Polyplexes were formulated with 50 µg gWiz Luciferase plasmid DNA in 250 µL of 20 mM HEPES pH 7.4, 5% glucose. Polyplexes were delivered via tail-vein injections in 12-16 week old mice and lung tissue was excised 24 h after injection. Lungs were collected in reagent lysis buffer (Promega) supplemented with protease inhibitors (Roche) and underwent three freeze-thaw cycles. Tissue was mechanically homogenized using an IKA T8 Disperser (Wilmington, NC) and cell debris was spun down at 14 000 g for 15 min at 4 °C. Luciferase activity and protein content were measured as described above.

RESULTS Synthesis of PEI-HGP and Polyplex Characterization. Moreno et al. recently screened 15 amino acid sequences from the HIV surface glycoprotein, gp41, in order to identify regions in the protein that interact with membranes. The sequence of HGP was chosen on the basis of its ability to mediate membrane rupture and is from the C-terminal tail region of HIV gp41 designated as the lentivirus peptide sequences (LLP-2/3) (20).

Figure 2. Hydrodynamic size of PEI-HGP (gray bars) and PEI (white bars) polyplexes were measured at N/P ratios 2, 3, and 4 using dynamic light scattering. Triplicate formulations were characterized at each charge ratio. Results are reported as mean diameter ( SD.

PEI-HGP was synthesized by conjugating a cysteine-terminated HGP to PEI with the heterobifunctional cross-linker SMPB. The concentration of peptide was determined by absorbance measurements at 280 nm and labeling efficiency was found to be 2-3 HGP per PEI on average. Polyplexes with plasmid DNA were formed by addition of an equal volume of polymer to DNA followed by rapid mixing and incubation at room temperature for 10 min. Complex formation was examined by agarose gel electrophoresis (Figure 1). Uncomplexed DNA migrates into the gel, whereas complexed DNA is occluded and retained in the loading well. Uncomplexed gWiz-Luciferase plasmid is shown in lanes marked as N/P ratio 0. For both PEI and PEI-HGP, DNA is fully complexed at N/P ratio of 2. Therefore, HGP conjugation does not interfere with complex formation. This is expected, as the percentage of modification is low. Hydrodynamic sizes of PEI and PEI-HGP polyplexes were measured using dynamic light scattering (DLS) in triplicate samples (Figure 2). For N/P ratios 2, 3, and 4, sizes of PEIHGP polyplexes were 153 ( 14.7 nm, 144.8 ( 10.6 nm, and 134.4 ( 7.1 nm, respectively. Sizes of PEI polyplexes were 139.7 ( 7.7 nm, 106.7 ( 4.9 nm, and 107.0 ( 6.7 nm. PEIHGP polyplexes are slightly larger in size. Lytic Activity of HGP. The lytic activity of free HGP peptide and PEI-HGP polyplexes was determined by monitoring the release of sulforhodamine B dye from liposomes after incubation for 15 min at various peptide concentrations (28). Figure 3A shows that HGP displayed concentration-dependent lytic activity when incubated with liposomes. At a concentration of 0.63 µM,

HIV-Derived Peptide for Gene and siRNA Delivery

Bioconjugate Chem., Vol. 19, No. 4, 2008 923

Figure 3. (A) Lytic activity of HGP was tested as free peptide using a fluorescence-based, liposomal leakage assay. (B) Lytic activities of HGP peptide (black bars), PEI polyplexes (gray bars), and PEI-HGP polyplexes (white bars) were compared. 100% dye release corresponded to fluorescence levels of liposomes treated with Triton X-100 surfactant for complete lysis. Results are reported as mean % dye release ( SD for triplicate samples.

HGP was able to mediate approximately 50% dye release. The lytic activities of PEI and PEI-HGP polyplexes formulated at various N/P ratios were also evaluated (Figure 3B). The lytic activity of the equivalent amount of HGP incorporated in the PEI-HGP polyplexes was also determined for comparison. The concentrations of HGP peptide present in polyplex formulations of N/P ratios of 2, 3, and 4 were 0.13 µM, 0.19 µM, and 0.26 µM, respectively. Free peptides at these concentrations resulted in dye release percentages of 1.0%, 2.2%, and 3.0%. The dye release percentages for N/P ratios 2, 3, and 4 were 0.5%, 4.4%, and 50.3% for PEI and 6.5%, 55.3%, and 76.6% for PEI-HGP. Evaluation of PEI-HGP for Plasmid DNA Delivery. Transfection efficiency of PEI and PEI-HGP polyplexes in HeLa cells as a function of N/P ratio was determined using the luciferase reporter gene system. PEI-HGP polyplexes showed increased expression profiles over unmodified polyplexes at N/P ratios of 2 and 3 (Figure 4A). At an N/P ratio of 2, transgene expression from PEI-HGP polyplexes (8.9 × 106 ( 3.8 × 106 RLU/mg) was 38-fold higher than PEI polyplexes (2.3 × 105 ( 4.0 × 104 RLU/mg) with a significance value of p < 0.05. At an N/P ratio of 3, transgene expression from PEI-HGP polyplexes (1.7 × 109 ( 2.3 × 108 RLU/mg) was 30-fold higher than PEI polyplexes (5.5 × 107 ( 3.7 × 107 RLU/mg) with a significance value of p < 0.001. At an N/P ratio of 4, transgene expression between PEI-HGP polyplexes (4.5 × 108 ( 4.3 × 108 RLU/mg) was not statistically different than PEI polyplexes (1.7 × 108 ( 2.0 × 108 RLU/mg). Increases in transfection efficiency at N/P ratios of 2 and 3 were confirmed in several separate experiments. Cellular viability after transfection with PEI and PEI-HGP polyplexes was determined using an MTS assay and compared to untreated cells (Figure 4B). No toxicity was observed with these materials at N/P ratios 2, 3, and 4. At charge ratios greater than an N/P ratio of 15, the PEI-HGP polyplexes were more toxic than PEI polyplexes (data not shown). Evaluation of PEI-HGP as an siRNA Carrier. The ability of HGP to enhance PEI-mediated siRNA delivery was evaluated

Figure 4. (A) Luciferase activity of PEI-HGP (gray bars) and PEI polyplexes (white bars) were compared in HeLa cells at N/P ratios 2, 3, and 4. (B) Cell viability of HeLa cells after transfection with PEIHGP (gray bars) and PEI (white bars) polyplexes at N/P ratios 2, 3, and 4 as measured by MTS. Results are reported as mean RLU/mg ( SD for triplicate samples. (Student’s t-test, *p < 0.05, **p < 0.001).

in HeLa cells using siRNA targeting the endogenous GAPDH gene. While both PEI and PEI-HGP incorporating siGAPDH mediated significant GAPDH knockdown compared to identical polymer formulations incorporating siGFP, PEI-HGP was responsible for a significant enhancement in specific GAPDH knockdown activity compared to unmodified PEI (53.4 ( 2.2% reduction in GAPDH expression for PEI vs 82.3 ( 1.6% reduction for PEI-HGP, p < 0.001) (Figure 5). In a similar experiment, cell lysates were collected and analyzed for GAPDH protein activity. Compared to the corresponding siGFP control, PEI/siGAPDH reduced GAPDH activity by 24.9 ( 9.5%, while PEI-HGP/siGAPDH reduced GAPDH activity by 57.0 ( 8.4%. Binding and Internalization of PEI-HGP Polyplexes. Cellular binding and uptake of NBD-labeled PEI and PEI-HGP polyplexes was investigated using flow cytometry (Figure 6). Fluorescently labeled polyplexes were incubated with HeLa cells at 4 °C (for binding) and 37 °C (for binding and uptake) for three hours before analysis. At 4 °C, association of PEI polyplexes was greater than PEI-HGP polyplexes (p < 0.005). At 37 °C, there was no significant difference in total associated fluorescence between PEI and PEI-HGP polyplexes. Confocal Imaging. In order to determine the intracellular distribution of carriers, PEI and PEI-HGP polyplexes containing TOTO-3-labeled plasmid DNA were incubated with cells for either 1 or 3 h, fixed, and imaged using confocal microscopy. Endosomes were labeled with Alexa488-dextran, a fluid-phase uptake marker. At 1 h, both PEI and PEI-hgp polyplexes exhibited punctate staining (data not shown). When PEI polyplexes were delivered for 3 h, DNA exhibited punctate intracellular staining that was colocalized with dextran (Figure

924 Bioconjugate Chem., Vol. 19, No. 4, 2008

Kwon et al.

0.01), and PEI-HGP transfection was increased 3.8-fold in the presence of chloroquine (76.8% ( 3.7%, p < 0.001). Results from flow cytometry were verified by fluorescence microscopy (Figure 8B).

DISCUSSION

Figure 5. Quantification of GAPDH mRNA levels in HeLa cells following treatment with PEI or PEI-HGP polyplexes incorporating siRNA. Cells were transfected with polyplexes incorporating either siRNA against GAPDH (siGAPDH, gray bars) or siRNA against GFP (siGFP, white bars) as a negative control (N/P ) 10). Total RNA was isolated 48 h after transfection, and relative GAPDH expression levels were measured by quantitative PCR. Values are normalized to an untreated control sample and are reported as the mean relative GAPDH expression ( SD for triplicate samples. (Student’s t test, *p < 0.001).

Figure 6. Binding and internalization of fluorescently labeled PEI and PEI-HGP polyplexes in HeLa cells. Cells were treated with polyplexes formulated at an N/P ratio 3 for 3 h at 4 °C (binding studies) and 37 °C (internalization studies) and analyzed by flow cytometry. Results are reported as mean fluorescence ( SD for triplicate samples. (Student’s t-test, *p < 0.005).

7A). There were few examples of dextran staining that were not colocalized with DNA. When PEI-HGP polyplexes were delivered for 3 h, both punctate and diffuse DNA staining were observed (Figure 7B). Incidences of punctate DNA were colocalized with dextran, although there were examples of endocytic vesicles that were not colocalized with DNA. A substantial fraction (∼30%) of cells had diffuse DNA fluorescence throughout the cells (white arrows). Chloroquine-Mediated Transfection. In order to investigate the effects of HGP as a mediator of endosomal escape, PEI and PEI-HGP polyplex transfections were performed in the presence of chloroquine. Chloroquine is a small molecule buffering agent used to mediate endosomal escape of gene delivery vehicles (29-31). Polyplex transfections to HeLa cells were conducted in the presence of 100 µM chloroquine using an enhanced green fluorescent protein (EGFP) reporter gene. The percent of EGFP positive cells was measured by flow cytometry (Figure 8A). The percentage of EGFP positive cells after PEI-HGP transfection (20.1% ( 1.8%) was increased 16.1fold when compared to PEI transfection (1.3% ( 0.3%, p < 0.001). Chloroquine increased the percentage of EGFP positive cells for both PEI and PEI-HGP; PEI transfection was increased 11.7-fold in the presence of chloroquine (14.6% ( 4.0%, p