Article pubs.acs.org/Biomac
Oxidation as a Facile Strategy To Reduce the Surface Charge and Toxicity of Polyethyleneimine Gene Carriers Wei Yang Seow, Kun Liang, Motoichi Kurisawa, and Charlotte A. E. Hauser* Institute of Bioengineering and Nanotechnology, 31 Biopolis Way, Singapore 138669 S Supporting Information *
ABSTRACT: Polyethyleneimine (PEI) is widely regarded as one of the most efficient non-viral transfection agents commercially available. However, a key concern is its pronounced cytotoxicity, ascribed mainly to its high amine content and cationic charge density. Significant past efforts to mitigate its toxicity usually involved lengthy synthetic procedures. We now propose a simple strategy using hydrogen peroxide (H2O2) to oxidize the amine groups. PEI/DNA complexes were first formed before some amine groups were removed with H2O2. This reduced surface charge while the remaining cationic charges still allowed for efficient transfection. The DNA was not damaged and remained bound after oxidation. Furthermore, H2O2 was quantitatively removed with sodium pyruvate prior to cell culture. Oxidized complexes caused no cytotoxicity even at high polymer concentrations. Compared to non-oxidized complexes used at subtoxic doses, oxidized complexes mediated significantly more GFP expression. A key strength of this approach is its simplicity as it involves only simple mixing of solutions. This strategy promises to further realize the potential of using PEI for the delivery of nucleic acids or other cargos.
1. INTRODUCTION Polyethyleneimine (PEI) as a cationic vehicle for nucleic acids delivery was first described in 1995.1 Today, PEI is still widely regarded as one of the most successful off-the-shelf non-viral transfection agent.2−4 PEI is cheaply available in a range of molecular weights and can be either linear or branched. It is, however, the branched form (usually 25 kDa) that is particularly successful2 and frequently serves as a yardstick for other novel carriers under development.5−7 The chemical structure of branched PEI is such that nitrogen makes up close to a third of its mass as a mixture of primary, secondary, and tertiary amines. This accounts for the highly cationic nature of PEI and its superior ability to bind DNA for the penetration of cellular membranes. Additionally, due to the spread in pKa values and the close proximity of amines (such that protonation in one amine may likely suppress the protonation of another neighboring amine due to the unfavorable energetic of juxtaposing like charges8), PEI is seldom fully protonated.9 This means that PEI maintains a stockpile of amines that can still undergo protonation during, for instance, the acidification of endosomes into lysosomes. As a result, PEI possesses a large buffering capacity that, studies suggested, allows it to exploit the proton-sponge hypothesis10 in order to gain cytosolic access for efficient transfection.1,5 Nonetheless, a complete understanding of the sequence of events of a PEI/DNA complex remains elusive,3 and issues such as how PEI gains nuclear entry or when it releases its DNA cargo are still controversial.11 PEI has been extensively modified and explored for various applications.12 For instance, mannose13 or transferrin14 was conjugated onto 25 kDa or 800 kDa PEI to improve its target © XXXX American Chemical Society
specificity for gene delivery, and 22 kDa JetPEI was used for the systemic delivery of DNA into neonatal mice and found to transfect several organs effectively.15 Also, 1.8 kDa PEI was either grafted onto polyorganophosphazene to form hydrogels for the delivery of human growth hormones16 or used with polycarbonates as a biodegradable gene carrier.7 A common concern in many studies, however, was the significant cytotoxicity of PEI, which increased with molecular weight and transfection efficiency.7,17,18 While important for DNA binding, buffering capacity, and membrane translocation, the high amine content and charge density of PEI have also been suggested to cause membrane disruption,18,19 leading to toxicity.7,17 Indeed, cationic charges function like a doubleedged sword, and its usage is a long-standing dilemma of researchers in the field of gene delivery. In efforts to reduce the toxicity of PEI carriers, disulfide bonds were introduced into the backbone of PEI to make it biodegradable in the reductive cytosolic environment.18 In another study, varying percentages of amine groups on PEI were removed by acetylation, which expectedly reduced surface charge, buffering, and DNA binding ability.20 Interestingly, toxicity decreased with increasing degree of acetylation, and the overall effect was still an increase in transfection efficiency due to the easy release of DNA from the weaker binding. A similar effect was observed when PEI was functionalized with esters, which upon hydrolysis contributed negative charges to make PEI less cationic, thereby facilitating the release of its DNA cargo.21 Poly(ethylene glycol) (PEG) Received: April 2, 2013 Revised: May 28, 2013
A
dx.doi.org/10.1021/bm4004628 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Scheme 1. Oxidation Strategy To Reduce the Surface Charge and Toxicity of PEI/DNA Complexesa
a PEI and DNA are typically mixed to form complexes with an overall positive charge. While important for DNA binding, buffering, and cellular internalization, the cationic amine groups have also been implicated in the cytotoxicity of PEI. Here, H2O2 was used to oxidize some of the amine groups to reduce surface charge while the remainder was left behind for transfection efficiency. Residual H2O2 was then quantitatively removed with sodium pyruvate before the complexes were added to the cells. Oxidized complexes caused no cytotoxicity up to N/P 80 and mediated a peak GFP expression in close to 90% of cells.
nitrogen on PEI to phosphorus on the DNA backbone) by adding appropriate volumes of PEI to DNA. The solution was mixed gently by pipetting and allowed to stand at room temperature for 30 min. Upon complex formation, an equivolume amount of 0.5% H2O2 was added such that its final concentration was 0.25%. Oxidation occurred at room temperature for the desired length of time. Water was used in place of H2O2 where oxidation was not intended. Prior to cell culture, excess H2O2 was removed by the addition of sodium pyruvate (Invitrogen, Singapore) to the complex solution for 30 min at room temperature. The amount of H2O2 remaining was quantified with the Pierce Quantitative Peroxide Assay Kit (Thermo Scientific, Singapore) according to the manufacturer’s recommendations. 2.3. Hydrodynamic Size and ζ Potential Measurements. A Malvern Zetasizer Nano ZS (Worcestershire, U.K.) was used to measure the hydrodynamic size and ζ potential of PEI/DNA complexes (N/P 15) following various treatments. Results were reported as average ± SD of at least triplicates. 2.4. Agarose Gel Electrophoresis. The integrity or presence of DNA was investigated on a 1% agarose gel (Invitrogen) preloaded with the SYBR Safe dye (Invitrogen). Images were then captured with a Versadoc 4000 MP instrument (Bio-Rad Laboratories, Singapore). 2.5. GFP Expression Assay. Human embryonic kidney (HEK293) cells were purchased from ATCC (USA) and maintained in Dulbecco’s modified Eagle’s medium (DMEM, Invitrogen) completed with 10% (v/v) fetal bovine serum (Invitrogen) and 1% (v/v) penicillin/streptomycin (Invitrogen) at 37 °C with 5% CO2. One day prior to transfection, 1 mL of cell suspension containing 160,000 cells was added into 12-well plates for attachment. On the day of transfection, 200 μL of PEI/DNA complexes at various N/P ratios with or without oxidation and/or sodium pyruvate treatment were prepared as above and then dripped into each well. Complexes without oxidation had the remaining volume made up with an equivalent amount of water and sodium pyruvate. The amount of DNA introduced into each well was fixed at 3 μg. Two days post transfection, cells were harvested and suspended in PBS for flow cytometry analysis (LSR II, BD Bioscience, CA, USA). Then, 10,000 cells were gated, and the percentage of cells expressing GFP was reported as mean ± SD of at least triplicates. Cells were defined to be positive for GFP if they fell within the gating region preset to include 80% cell viability), oxidized PEI mediated significantly more green fluorescent protein (GFP) expression in HEK293 cells while allowing for 100% cell viability. To our knowledge, such a simple oxidative method to chemically modify PEI has not been reported previously. This promises to be an effective and convenient strategy to further realize the potential of using PEI as an efficient vehicle for nucleic acids delivery.
2. EXPERIMENTAL SECTION 2.1. Nuclear Magnetic Resonance (NMR). NMR signals were recorded with a Bruker 400 MHz Ultrashield Plus (Bruker BioSpin, Singapore), and chemical shifts were expressed as parts per million (ppm). For 1H NMR, branched 25 kDa PEI (Sigma, Singapore) was dissolved in a mixture of H2O/D2O (50/50 v/v) (Cambridge Isotope Laboratories, Inc., MA, USA) to a final concentration of 2 mg/mL in the presence or absence of 0.25% H2O2 (Merck, Singapore). Solutions were stored at room temperature until measurement. For 13C NMR (100 MHz, 10000 scans), PEI was dissolved to a final concentration of ∼60 mg/mL in a mixture of H2O/D2O (80/20 v/v) containing varying amounts of H2O2. 2.2. Formation and Oxidation of PEI/DNA Complexes. Plasmid (pEGFP-C1, 4.7 kbp) encoding for GFP was obtained from Clontech Laboratories, Inc. (CA, USA). PEI/DNA complexes were first formed at various N/P ratios (defined here as the molar ratio of B
dx.doi.org/10.1021/bm4004628 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 1. 1H and 13C NMR analysis of PEI. (a) For 1H NMR, PEI was incubated in 0.25% H2O2. Initially, PEI produced two main peaks corresponding to the methylene protons. Following further incubation, the original peaks reduced in intensity. At the same time, additional peaks appeared downfield, presumably as the amine groups became oxidized. Protons could not be assigned due to insufficient peak resolution. (b) 13C NMR of non-oxidized PEI with all eight major peaks being assigned. (c) 13C NMR spectra of PEI after incubation in varying concentrations of H2O2 for 30 h. The dose-dependent reduction of all eight major peaks coincided with the appearance of additional peaks (presumably due to the amines being oxidized), suggesting that all three classes of amines present in PEI are susceptible to oxidation by H2O2. fluorescence was measured using a Tecan Infinite Microplate Reader (Ex/Em, 545/590 nm). Viability was expressed as a percentage of untreated control cells and data was reported as mean ± SD of sextuplicates. 2.7. Statistical Analysis. ANOVA testing (OriginLab Corporation, MA, USA) of sample means was performed with p < 0.05 (denoted by *) being accepted to be statistically significant.
conditions, the peaks were insufficiently resolved for assignment, and we were not able to determine which class(es) of amines was affected by H2O2. To gain further insights into the oxidation process, we turned to 13C NMR, which produced better peak resolution. On the basis of available literature,25,26 the eight major peaks in the 13C spectrum of non-oxidized PEI can be assigned (Figure 1b). For instance, peaks 1−3, 4−6, and 7 and 8 belong to carbons immediately beside tertiary, secondary, and primary amines, respectively. It is tempting to speculate that the tertiary amines, being most amendable to oxidation,27,28 are the predominant species being oxidized, in which case, zwitterionic amine-oxides are likely to be the main oxidation product.27,28 Recently, betaine-modified PEI with a partial zwitterionic character was reported to reduce vector toxicity due to enhanced surface properties and reduced protein adsorption,23 suggesting that it may be advantageous to have zwitterionic species within the polymer backbone. Surprisingly, the 13C spectra of PEI following incubation in increasing concentration of H2O2 showed a gradual reduction of all of the eight original peaks in a dose-dependent fashion (Figure 1c). Additionally, there was the appearance of additional peaks, presumably due to the oxidized amines, which became more prominent with increasing concentration of H2O2. This suggests that all three classes of amines present in PEI are susceptible to oxidation by H2O2. More studies are currently being conducted in our laboratory to identify the main oxidized product. We further point out that a high polymer concentration was used in this series of experiments to obtain meaningful 13C NMR spectra. Therefore, oxidation conditions discussed here may not be directly indicative of those that will be used to prepare the PEI/ DNA complexes for subsequent transfection studies (where PEI is present in much lower amounts). Finally, incubating PEI in the absence of H2O2 for 2 days produced no change in the
3. RESULTS AND DISCUSSION The high positive charge density of PEI is indeed a doubleedged sword as it is frequently cited to be a main cause of both its transfection efficiency and toxicity.18,19 In this study, we attempted to reduce the toxicity of PEI/DNA complexes by partially removing some cationic charges while leaving the remaining behind for transfection efficiency (Scheme 1). We also aimed to avoid lengthy synthetic and purification procedures or the need for harmful catalysts to chemically modify PEI. We therefore used oxidation to remove the amine groups on PEI as it involved just simple mixing of commercially available solutions. To evaluate the susceptibility of PEI to oxidation by H2O2, we dissolved PEI in an aqueous solution of 0.25% H2O2 before studying its chemical structure using 1H NMR at various time intervals. From Figure 1a, PEI at the start of experiment produced two main peaks between 2.5 and 2.6 ppm that are assignable to the methylene protons. Upon oxidation with H2O2 for 5 h, additional peaks appeared between 2.7 and 3.3 ppm. Continued oxidation after 1 or 2 days produced new peaks downfield of 6.4 ppm with a concomitant decrease in the original peaks at 2.5−2.6 ppm. This suggests that the amines were most likely consumed, and the new peaks presumably belonged to the methylene protons now situated adjacent to oxidized amines. The coexistence of primary, secondary, and tertiary amines in PEI has complicated the interpretation of the 1 H NMR spectra. Moreover, under current 1H NMR C
dx.doi.org/10.1021/bm4004628 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 2. PEI/DNA complexes were formed and then incubated in the presence or absence of 0.25% H2O2 for various durations at room temperature. The (a) ζ potential and (b) diameter of the complexes were then measured using dynamic light scattering. In the absence of H2O2, the complexes were highly cationic (∼+40 mV). They were also stable and could maintain their size (∼100−130 nm) over 3 days. Upon the introduction of H2O2, ζ potential became less positive with time, supporting the hypothesis that some amine groups became consumed. The oxidized complexes also tended to aggregate more due to the reduction in interparticle repulsion forces, and size gradually increased with time, most sharply after 1 day. 1
H NMR spectrum (Figure S1 in Supporting Information). This indicates that air oxidation was insignificant and that H2O2 was essential for the chemical transformation. Next, the effects of oxidation on the surface ζ potential and size of PEI complexes were investigated by dynamic light scattering. First, PEI was mixed with plasmids encoding for GFP to form complexes. H2O2 at a final concentration of 0.25% or water was then added to the complexes. The current concentration of 0.25% H2O2 was chosen after preliminary experiments with a lower concentration led to less consistent results (data not shown). As seen in Figure 2a, in the absence of H2O2, PEI complexes were highly cationic as expected (∼+40 mV). The high surface charge also meant that the particles could resist aggregation and remained stable in solution. This was evident by the consistent surface charge and hydrodynamic diameter (100−130 nm) (Figure 2b) of the complexes throughout the 3 days of measurement. Upon incubation with H2O2, however, ζ potential became less positive, and the sharpest fall occurred between 6 and 24 h post oxidation. This was consistent with earlier NMR data indicating that the amines were being consumed and thus could no longer contribute positive charges. As a result, the size of the complexes increased as they became less positive and less able to repel neighboring complexes. On the basis of these observations, 24 h was selected to be an optimal oxidation time for the complexes prior to transfection since ζ potential was significantly reduced (by ∼40% compared to non-oxidized samples) just before the onset of particle aggregation. We had two concerns about the oxidation procedure: first, if the DNA could be damaged by H2O2 and second, if the reduction in surface charge could cause complex dissociation. To answer the first question, we exposed naked unprotected DNA to 0.25% H2O2 for 24 h at 37 °C before assessing their integrity on an agarose gel. From Figure 3, DNA exposed to either heat alone or heat and H2O2 produced bands that were identical to the control DNA. This suggests that oxidative DNA damage is insignificant at the current concentration of H2O2 used. Moreover, it has to be pointed out that naked DNA was used for this experiment, whereas in actual application, oxidation would occur only after the formation of complexes where the DNA would already be protected by PEI from any insults by the external environment. To address the second concern, PEI complexes were first formed and then similarly exposed to 0.25% H2O2 for 24 h. As can be seen in Figure 3, no bands were seen before and after exposure to H2O2. This indicates that complex dissociation was not evident and that the
Figure 3. Agarose gel electrophoresis to evaluate the integrity and presence of DNA. DNA was subjected to various treatments as stated above. Compared to the control DNA, incubation of naked DNA in 0.25% H2O2 at 37 °C for 24 h did not compromise its integrity. Furthermore, the oxidation process did not cause the DNA to dissociate from its PEI carrier within the time frame of this experiment.
DNA remained tightly bound to PEI in the face of reduced surface charge. These observations also argue for the current order of oxidation practiced, i.e., post- rather than precomplex formation as the DNA enjoys extra protection and is not affected by the change in binding strength of PEI. A potential barrier to bioapplication of these novel oxidized PEI complexes is that even a small amount of residual H2O2 can be toxic to cells. We surveyed several methods to remove the residual H2O2 before cell culture and were unsuccessful initially. For instance, we tried an enzymatic approach by adding catalase to the complex solution. The enzyme quantitatively removed H2O2 (breaking it down into H2O and O2) (Supplementary Figure S2a) but caused massive particle aggregation (Supplementary Figure S2b) and a reversal of surface charge (Supplementary Figure S2c). This is likely because catalase itself is a negatively charged macromolecule. We next tried a redox approach by adding molar equivalents of reducing agents such dithiothreitol (DTT) and tris(2-carboxyethyl)phosphine (TCEP) to remove H2O2. The addition of DTT increased complex size (Supplementary Figure S3a) but did not significantly affect surface charge (Supplementary Figure S3b). Unfortunately, cell viability suffered significantly after exposure to DTT-treated complexes (data not shown). TCEP, on the other hand, was too acidic, turned the growth medium yellow immediately upon contact, and was not compatible with D
dx.doi.org/10.1021/bm4004628 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
Figure 4. Removal of residual H2O2 from the complex solution using sodium pyruvate as a free-radical scavenger. (a) Sodium pyruvate was added to 0.25% H2O2 for various durations before the amount of remaining H2O2 was quantified. Sodium pyruvate was used in a slight molar excess to remove >98% of H2O2 after 30 min. After the sodium pyruvate treatment, there was (b) a slight increase in the size but (c) no significant effect on the ζ potential of the oxidized complexes.
Figure 5. GFP expression and cytotoxicity assay using PEI complexes with or without oxidation. For both species of PEI, sodium pyruvate was added to account for its effect on transfection efficiency and cytotoxicity. (a) Flow cytometry was used to quantify the percentage of HEK293 cells expressing GFP following transfection with various PEI formulations. (*** p < 0.001) (b) The AlamarBlue assay was used to quantify the percentage of metabolically active cells after exposure to various PEI formulations. 100% of cells remained viable (up to N/P 80), and a peak transfection efficiency of close to 90% was achieved at N/P 80 after transfection with oxidized complexes. In contrast, significant toxicity was observed with nonoxidized complexes past N/P 10. (c−f) Representative images of GFP-positive cells following different treatments. A representative scale bar for 200 μm is provided in panel c.
S4a). Doing so, >95% of H2O2 was removed after 2 h, and the size of complexes was only slightly increased (Supplementary Figure S4b). Cell viability, however, suffered following this treatment, which can be due to either the residual H2O2 or FeCl3. Finally, we ruled out dialysis to remove H2O2 as the method is time-consuming. Also, due to the effects of volume gain and diffusive polymer lost (despite choosing a membrane with a small cutoff pore size), there is expected to be uncertainty and batch-to-batch variation in the concentration
cells (data not shown). Arguably, the redox approach may work after further fine-tuning of the molar ratio of reducing agent to H2O2. However, this approach has a limited working window and may not be robust enough to accommodate fluctuations in PEI or H2O2 concentration in future experiments. Yet another method we tried was a catalytic approach that involved decomposing H2O2 at 60 °C in the presence of FeCl3. Heat alone was insufficient to decompose H2O2 and required the addition of a catalytic amount of FeCl3 (Supplementary Figure E
dx.doi.org/10.1021/bm4004628 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
Article
The lack of toxicity of the oxidized complexes made it possible to extend the therapeutic range of PEI to N/P 100, although transfection efficiency was optimal at N/P 80. In another study, betaine-modified PEI with a partial zwitterionic character also demonstrated reduced toxicity, allowing its application at a high N/P ratio.23 On the other hand, the lack of transfection efficiency of non-oxidized complexes below N/P 10 meant that there was a lower limit at which such complexes could be used. Therefore, compared to just simply reducing N/ P ratio, the oxidation strategy appears to be a superior method to reduce toxicity. Admittedly, compared to non-oxidized complexes, more polymer is required to achieve high transfection efficiency for the oxidized complexes. However, we stress that the oxidized species caused no toxicity below N/ P 80. Moreover, for in vitro, ex vivo, or even nonchronic in vivo applications where accumulation in the body is not a major concern, this approach appears to be useful. We currently cannot provide a full mechanistic description of the mode of action of oxidized PEI complexes, which is the subject of our future investigation. Nonetheless, this simple oxidation strategy promises to extend the suitability of using PEI as a non-viral gene carrier.
of DNA particles after dialysis. Furthermore, large volumes of sterile water/buffer are required for the dialysis process, which adds to cost and a more complicated experimental setup to maintain sterility. Eventually, we tried the free-radical scavenger approach by using sodium pyruvate, which reacts with H2O2 to form sodium acetate, CO2, and H2O.24 The advantage here is that sodium pyruvate itself is frequently used as a growth medium supplement to provide an alternative energy source for the cells. It is thus extremely compatible with cells and can be added in molar excess to ensure complete removal of H2O2. Indeed, a slight molar excess of sodium pyruvate was needed to remove >98% of H2O2 after only 30 min of incubation at room temperature (Figure 4a). Following such a treatment, particle size increased modestly (Figure 4b), which could be due to the presence of charge-screening counterions. On the other hand, surface charge was not affected significantly (Figure 4c). We nonetheless note that it was difficult to obtain good ζ potential data as, presumably, the addition of sodium pyruvate altered the ionic strength of the sample to which the measurement was sensitive to. The oxidized PEI/DNA complexes were next evaluated for their ability to mediate GFP expression and the extent of toxicity induced in HEK293 cells. PEI complexes were either oxidized as above or unmodified. In both cases, equivalent volume of sodium pyruvate was added to the complex solution for 30 min before dripping into the culture medium. This was to exclude the effects of sodium pyruvate on transfection efficiency and toxicity. As seen from Figure 5a, regular nonoxidized PEI complexes could efficiently transfect cells, and 100% of cells expressed GFP past N/P 20. However, this transfection efficiency came at the expense of toxicity. For instance, only ∼83% and ∼71% of cells remained metabolically active at N/P 10 and 20, respectively (Figure 5b). Toxicity also increased steeply at N/P 30 where 90% of cells maintained metabolic activity at N/P 100 when oxidized complexes were used. Therefore, comparing the peak transfection efficiencies of the two species of PEI complexes when used at a concentration where >80% of cells maintained metabolic activity (i.e, N/P 10 and 80 for non-oxidized and oxidized complexes respectively), the oxidized PEI complexes were far more efficient (∼47% vs ∼88%) in transfecting cells. The gulf in transfection efficiency would increase even further if the requirement on cell viability was more stringent. For instance, if >90% cell viability was required, non-oxidized PEI had to be used at N/P 5, in which case only ∼8% of cells were transfected. Images corresponding to the respective transfection conditions are presented in Figure 5c−f and Supplementary Figure S5 to provide a visual representation of the distribution of GFP-positive cells. As expected, oxidized complexes without sodium pyruvate treatment were highly toxic, mainly due to the residual H2O2, and ∼95% of cells lost metabolic activity (Figure 5b). The sodium pyruvate treatment is therefore an essential step in the overall oxidation strategy. Finally, we note that full serum conditions were maintained throughout all cell culture experiments.
4. CONCLUSIONS A simple and effective strategy of modifying PEI to reduce its cytotoxicity has been reported. PEI complexes were first formed, before H2O2 was introduced to oxidize a portion of the amine groups. The remaining amine groups were left behind for transfection efficiency. This resulted in complexes with reduced surface charge. The DNA was not damaged, and the complex did not dissociate after oxidation. Furthermore, H2O2 could be quantitatively removed with sodium pyruvate as a free-radical scavenger prior to cell culture. Oxidized complexes caused no toxicity in cells up to N/P 80 (defined here as the molar ratio of nitrogen on PEI to phosphorus on the DNA backbone). Compared to regular non-oxidized complexes used at a concentration where >80% of cells maintained metabolic activity, oxidized complexes mediated significantly more GFP expression while allowing for 100% cell viability. A key strength of this approach is its simplicity as it involves only simple mixing of solutions. Lengthy and complicated synthetic or purification procedures are therefore avoided. The use of potentially harmful catalysts is also not needed. This strategy promises to extend the suitability of using PEI for the delivery of nucleic acids or other cargos to a wide range of applications.
■
ASSOCIATED CONTENT
S Supporting Information *
Figures including time course NMR analysis of PEI dissolved in a mixture of H2O/D2O in the absence of H2O2; enzymatic, redox, and catalytic approaches to remove H2O2; and representative images of cells expressing GFP after being transfected with various formulations of PEI/DNA complexes. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Notes
The authors declare no competing financial interest. F
dx.doi.org/10.1021/bm4004628 | Biomacromolecules XXXX, XXX, XXX−XXX
Biomacromolecules
■
Article
ACKNOWLEDGMENTS We thank Kiat-Hwa Chan and Mangesh Joshi (Institute of Bioengineering and Nanotechnology, IBN) for assistance in obtaining and interpreting the NMR spectra. This work was supported by IBN, Biomedical Research Council, Agency for Science, Technology and Research, Singapore.
■
REFERENCES
(1) Boussif, O.; Lezoualc’h, F.; Zanta, M. A.; Mergny, M. D.; Scherman, D.; Demeneix, B.; Behr, J.-P. Proc. Natl. Acad. Sci. U.S.A. 1995, 92, 7297−7301. (2) Putnam, D. Nat. Med. 2006, 5, 439−451. (3) Won, Y.-Y.; Sharma, R.; Konieczny, S. F. J. Controlled Release 2009, 139, 88−93. (4) Hoffman, A. S. J. Controlled Release 2008, 132, 153−163. (5) Wong, S. Y.; Sood, N.; Putnam, D. Mol. Ther. 2009, 17, 480−490. (6) Baumhover, N. J.; Anderson, K.; Fernandez, C. A.; Rice, K. G. Bioconjugate Chem. 2010, 21, 74−83. (7) He, F.; Wang, C.-F.; Jiang, T.; Han, B.; Zhuo, R.-X. Biomacromolecules 2010, 11, 3028−3035. (8) Suh, J.; Paik, H.-J.; Hwang, B. K. Bioconjugate Chem. 1994, 22, 318−327. (9) Ziebarth, J. D.; Wang, Y. Biomacromolecules 2010, 11, 29−38. (10) Seow, W. Y.; George, A. J. T. In Non-viral Gene Therapy; Yuan, X., Ed.; InTect Open Access Publisher: Rijeka/Vienna, 2011; pp 615− 644 (Online open access article). (11) Lam, A. P.; Dean, D. A. Gene Ther. 2010, 17, 439−447. (12) Jager, M.; Schubert, S.; Ochrimenko, S.; Fischer, D.; Schubert, U. S. Chem. Soc. Rev. 2012, 41, 4755−4767. (13) Diebold, S. S.; Kursa, M.; Wagner, E.; Cotten, M.; Zenke, M. J. Biol. Chem. 1999, 274, 19087−19094. (14) Kircheis, R.; Kichler, A.; Wallner, G.; Kursa, M.; Ogris, M.; Felzmann, T.; Buchberger, M.; Wagner, E. Gene Ther. 1997, 4, 409− 418. (15) Wong, S. P.; Argyros, O.; Howe, S. J.; Harbottle, R. P. J. Controlled Release 2011, 150, 298−306. (16) Seo, B.-B.; Park, M.-R.; Chun, C.; Lee, J.-Y.; Song, S.-C. Biomaterials 2011, 32, 8271−8280. (17) Moore, N. M.; Sheppard, C. L.; Sakiyama-Elbert, S. E. Acta Biomater. 2009, 5, 854−864. (18) Lee, Y.; Mo, H.; Koo, H.; Park, J.-Y.; Cho, M. Y.; Jin, G.-w.; Park, J.-S. Bioconjugate Chem. 2007, 18, 13−18. (19) Mahon, K. P.; Love, K. T.; Whitehead, K. A.; Qin, J.; Akinc, A.; Leshchiner, E.; Leshchiner, I.; Langer, R.; Anderson, D. G. Bioconjugate Chem. 2010, 21, 1448−1454. (20) Gabrielson, N. P.; Pack, D. W. Biomacromolecules 2006, 7, 2427−2435. (21) Liu, X.; Yang, J. W.; Lynn, D. M. Biomacromolecules 2008, 9, 2063−2071. (22) Zhong, Z.; Feijen, J.; Lok, M. C.; Hennink, W. E.; Christensen, L. V.; Yockman, J. W.; Kim, Y.-H.; Kim, S. W. Biomacromolecules 2005, 6, 3440−3448. (23) Jun, S.; Zeng, F.; Jian, H.; Wu, S. Biomacromolecules 2013, 14, 728−736. (24) Jagtap, J. C.; Chandele, A.; Chopde, B. A.; Shastry, P. J. Chem. Neuroanat. 2003, 26, 109−118. (25) von-Harpe, A.; Petersen, H.; Li, Y.; Kissel, T. J. Controlled Release 2000, 69, 309−322. (26) Idris, S. A.; Mkhatresh, O. A.; Heatley, F. Polym. Int. 2006, 55, 1040−1048. (27) Sharma, V. B.; Jain, S. L.; Sain, B. Tetrahedron Lett. 2003, 44, 3235−3237. (28) Colladon, M.; Scarso, A.; Strukul, G. Green Chem. 2008, 10, 793−798.
G
dx.doi.org/10.1021/bm4004628 | Biomacromolecules XXXX, XXX, XXX−XXX
