Lanthanide-Containing Polycations for Monitoring Polyplex Dynamics

Mar 10, 2014 - Bryson , J. M.; Fichter , K. M.; Chu , W.-J.; Lee , J.-H.; Li , J.; Madsen , L. A.; McLendon , P. M.; Reineke , T. M. Proc. Natl. Acad...
0 downloads 0 Views 658KB Size
Download PDF
Article pubs.acs.org/Biomac

Lanthanide-Containing Polycations for Monitoring Polyplex Dynamics via Lanthanide Resonance Energy Transfer Sneha S. Kelkar,†,‡ Lian Xue,§ S. Richard Turner,† and Theresa M. Reineke*,§ †

Department of Chemistry and Macromolecules and Interfaces Institute, Virginia Tech, Blacksburg, Virginia 24060, United States Wake Forest Institute for Regenerative Medicine and Virginia Tech-Wake Forest School of Biomedical Engineering and Sciences, Winston-Salem, North Carolina 27101, United States § Department of Chemistry, University of Minnesota, Minneapolis, Minnesota 55455, United States ‡

S Supporting Information *

ABSTRACT: Theranostic nanomaterials have emerged in the past decade that combine therapeutic delivery and diagnostic imaging into one package. Such materials offer the opportunity to aid diagnosis, track therapeutic biodistribution, and monitor drug release. We have developed a series of nucleic acid delivery polymers containing oligoethylene amines that are able to be protonated at physiological pH (for binding/compacting pDNA) and a lanthanide-chelating domain, which imparts diagnostic functionality. Diamine monomers (containing between 3 and 6 Boc-protected ethyleneamines) were prepared via a multistep procedure involving the selective protection and deprotection of primary and secondary amines. The polymer structures were then synthesized by step-growth polymerization of the oligoethylene diamines with a bisanhydride of diethylenetriamine pentaacetic acid (DTPA-BA), yielding degrees of polymerization between 18 and 24. Chelation of the polymers with gadolinium and terbium was performed to offer MRI contrast agent and luminescence properties, respectively. All of the polymer chelates were found to house approximately one water coordination site, as calculated by the Horrock’s equation and possess longitudinal relaxivities (r1, on a per Gd basis) at least twice that of Magnevist, a clinical contrast agent. All the structures formed polyplexes with pDNA with highly positive zeta potentials and hydrodynamic diameters around 50−80 nm. Lanthanide resonance energy transfer (LRET) was used to monitor polyplex association and dissociation. Polyplexes were formed using the donor−acceptor pair comprising of terbium-chelated polymer with five ethyleneamines within the repeat unit (6c-Tb) and tetramethyl rhodamine (TMR)-labeled pDNA. Association/dissociation in the presence of heparin and NaCl was monitored. The effect of amine number along the polymer backbone on transfection efficiency and cytotoxicity was also investigated. None of the polymers revealed cytotoxic effects with cultured cells; however, the polymer with six ethyleneamines clearly offered the highest transfection efficiency. This preliminary study offers insight into the development of materials with the ability to monitor polyplex unpackaging over time within the cellular environment.

■

modality.6−8 These “theranostic” materials can potentially eliminate the need for a multistep drug administration process for diagnosis and treatment and may enable real-time monitoring of drug delivery and drug efficacy.9 To monitor delivery, nanomedicines can be loaded with imaging agents such as fluorescent markers, MRI contrast agents, and radionuclides, which allow noninvasive imaging.7,10,11 For example, intracellular trafficking and unpackaging of Cy5-labeled chitosan and QD-labeled pDNA has been monitored using QD-FRET by Ho et al. in HEK293 cells.12 In that study, a strong FRET signal from the acceptor (Cy5) was observed in the perinuclear region at 24 and 48 h (indicative of an intact polyplex structure), whereas, after 72 h, there was a significant decrease in FRET signal due to polyplex

INTRODUCTION Nucleic acid based therapeutics hold promise for treating aggressive diseases such as cancer, diabetes, Alzheimer’s, and many common and orphan genetic disorders. However, lack of efficacious delivery has presented impediments for clinical approval.1,2 For a nucleic acid carrier to be effective, it must be designed to be administered and internalized by pathological cells, trafficked to the proper intracellular sites, and released for therapeutic activity.3 For systemic administration, nanomedicine delivery vehicles are also generally designed to accumulate within tumors via “passive targeting” through the enhanced permeability and retention (EPR) effect if the size is well below 100 nm.4,5 In the past decade, interdisciplinary research in the field of chemistry, biology, nanotechnology, materials science, and medicine has led to the development of novel materials that are capable of combining therapy and imaging into one © 2014 American Chemical Society

Received: December 18, 2013 Revised: March 8, 2014 Published: March 10, 2014 1612

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624

Biomacromolecules

Article

channels) can limit their utility in certain cases. To overcome these limitations, inorganic nanoparticles based on semiconductor quantum dots or luminescent lanthanides are currently being examined.23 For example, Johansson et al.24 have developed an LRET-based assay to study oligonucleotide hybridization/enzymatic degradation. Oligonucleotides were labeled with terbium on the 5′ ends and a dark quencher probe (BHQ) on the 3′ ends, which allowed detection of hybridization or enzymatic degradation. To this end, our group has developed a series of lanthanidecontaining theranostic polymers for nucleic acid delivery and detection.25 In our previous study, polymers were synthesized and studied biologically that contain chelates of gadolinium or europium copolymerized with ethyleneamine units along the backbone; analogs containing three or four ethyleneamines were created and analyzed. Analogs of these polymers that chelate Gd3+ revealed MRI contrast agent properties (millimeter to micrometer resolution) in bulk transfected cell samples. Luminescent Eu3+-chelated polymer analogs promoted visualization of delivery and localization within individual cells via confocal microscopy imaging (micro to nanometer resolution). These previous systems were also examined for cellular delivery and toxicity. The previous models containing 3 or 4 ethyleneamine units offered relatively low cytotoxicity and high cellular internalization properties of the polyplex formulations; however, they were found to offer lower transgene expression profiles, all of which has encouraged our group to pursue further studies with these materials. In the current study, we sought to extend this design motif by systematically varying the ethyleneamine number along the polymer backbone from three to six to understand if these chemical changes alter toxicity and improve the transgene expression profiles from the previous analogs. The diagnostic functionality of these structures has also been systematically varied by altering the metal functionality (i.e., Gd3+ or Tb3+) within the chelate structure of the polymer backbone. Furthermore, we have developed a lanthanide resonance energy transfer (LRET)-based method to monitor polyplex unpackaging in vitro using a Tb3+-chelated polymer (donor fluorophore) and tetramethyl rhodamine (TMR)-labeled pDNA, which serves as the acceptor. The terbium-TMR “LRET pair” has been previously used for examining protein conformations26,27 and protein−protein interactions;28 however, to the best of our knowledge, this is the first study to monitor polyplex unpackaging using LRET. In this report, LRET efficiency (Tb3+-polymer/TMR-pDNA binding) was observed through monitoring the change in donor (Tb3+) intensity upon formation and dissociation/destabilization of polyplexes, via time-delayed fluorescence spectrometry. Studies were performed at different N/P ratios (ratio of the number of amines on the polymer to phosphate groups on the pDNA backbone) with the polymer analog containing five ethylene amines in the repeat unit. In vitro cell culture studies reveal differences in transgene expression of these materials as a function of metal chelate type and number of ethylene-amines, with a significant enhancement of transfection with the analogs containing six ethyleneamines. The results obtained with this preliminary in vitro work are promising and suggest pursuit of further LRET research both in vitro and in vivo.

dissociation and cargo unpackaging. A recent study by Shrestha et al.13 examined supramolecular structures comprising of cationic shell-cross-linked knedel-like nanoparticles (cSCKs) that can complex with therapeutic siRNA and anionic shell-crosslinked rods (SCRs) that can be cross-linked with SCKs and covalently linked to a radionuclide (76Br) for nuclear imaging. Cutrin et al.14 developed protein-based theranostic nanoparticles (Apo-CUR-Gd) to monitor in vivo delivery and protection of liver from thioacetamide-induced hepatitis via delivery of curcumin. In vivo MRI studies indicated enhanced uptake of the nanoparticles in the mouse liver and animals pretreated with Apo-CUR-Gd showed significantly reduced hepatic injury (assessed by levels of alanine transferase) compared to the untreated controls. In addition to in vivo imaging, theranostics can be utilized to aid cellular-based imaging techniques and benefit studies to understand delivery vehicle trafficking and unpackaging. Traditionally, intracellular tracking of nanocomplexes containing polynucleotides has been accomplished by labeling nucleic acids and cellular organelles with suitable fluorescent markers to study their colocalization and trafficking mechanisms.6 This methodology yields information about trafficking pathways and kinetics, but issues exist with the inherently low spatial resolution associated with fluorescent dye imaging. With conventional fluorophores, the polymer and the nucleic acid have to diffuse quite far from each other to be considered dissociated, consequently two fluorophores without any chemical/biological interaction could appear to have colocalized signals due to their spatial proximity possibly providing misleading information about polyplex dynamics.15 Additional techniques to monitor polyplex association− dissociation could allow detection and monitoring of polyplex dynamics at higher resolution. Fluorescence resonance energy transfer (FRET), which relies on (nanometer range) distancedependent nonradiative energy transfer between two chromophores (donor and acceptor), has provided useful insights into drug/gene delivery and the release of the vehicle from the payload.16,17 Two chromophores with a spectral overlap and spatial proximity of 10−100 Å can act as “FRET” pairs.18,19 Several methods that are used to measure the FRET efficiency include the measurement of donor and/or acceptor emission, donor lifetime, and fluorescence anisotropy.20,21 For example, Chen et al.22 have developed a FRET-based technique to monitor drug release from polymeric micelles comprising of monomethoxy poly(ethyleneglycol)-block-poly(D,L-lactic acid) copolymer. The polymeric micelles labeled with FITC were loaded with lipophilic organic dyes, DiI and DiO (FRET pairs). The FRET signal was obtained in an intact micelle and the loss of FRET signal was indicative of the drug release. Their studies also indicated that the interaction between the micelles and the plasma membrane played an important role in the dye/drug release. Redy et al.17 studied prodrug activation in the presence of a model enzyme penicillin-G-amidase using FRET-mediated fluorescence quenching and activation. The self-immolative linker between a prodrug and camptothecin was labeled with a pair of identical fluorophores (fluorescein dyes). The close proximity of these dyes was associated with fluorescence quenching; conversely, the drug release resulted in an increase in the fluorescence signal. While these studies clearly show that FRET studies performed with organic dyes have clearly offered insight into drug release, issues such as photobleaching, short fluorescence lifetimes (prohibits use of time-lapse studies), and spectral cross-talk/ overlap (emission of one fluorophore being detected in both

■

EXPERIMENTAL SECTION

General. All reagents obtained were purchased from Aldrich Chemical Co. (Milwaukee, WI) and used without any further purification unless specified otherwise. Diethylenetriamine pentaacetic acid was

1613

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624

Biomacromolecules

Article

Scheme 1. Synthesis of the Oligoethylene Diamine Monomers

Scheme 2. Synthesis of DTPA-BA Monomer

purchased from Alfa Aesar Chemical Co. (Ward Hill, MA). Pentaethylene hexamine, obtained from Sigma Aldrich (95% pure), was further purified by double distillation. MeOH and dichloromethane were purified with MBRAUN MB SPS solvent purification system. Dry dimethyl sulfoxide (DMSO) was purchased from Alfa Aesar. The 20 mM HEPES buffer was prepared with ultrapure MilliQ water and the solution pH 7 was adjusted using a saturated solution of sodium bicarbonate. The plasmid DNA used in this study, gWiz-Luc, was purchased from Aldeveron (Fargo, ND). JetPEI, a linear polyethyleneimine based transfection agent was purchased from Polyplus Transfection (New York, NY). A TMR labeling kit was purchased from Invitrogen (Carlsbad, CA) and pDNA labeling was performed using the manufacturer’s protocol to achieve a 1:5 labeling ratio between pDNA and the TMR dye. Sodium chloride was obtained from Fisher Scientific and heparin sodium salt was obtained from Alfa Aesar. The liquid chromatography−mass spectra (LC-MS) were obtained with an Agilent (Wilmington, DE, U.S.A.) 1100 series HPLC equipped with a diode array detector, column heater, and Thermo Survey (San Jose, CA, U.S.A.) autosampler. The columns used for the HPLC separations were Agilent extended C18 HPLC columns (50 × 2.1 mm) and the mobile phase was one of the following: water (containing 1% formic acid), methanol (containing 1% formic acid) or acetonitrile (containing 1% formic acid). For mass spectral analysis, the HPLC column effluent was directly pumped into the spray chamber of a Thermo Instrument TSQ triple quadrupole mass spectrometer (Thermo Finnigan, San Jose, CA, U.S.A.) equipped with an ESI source. NMR spectra were obtained on Inova 400 MHz or AVANCE 500 MHz instrument. Thin layer chromatography (TLC) was performed on TLC aluminum sheets (silica gel 60 F254) obtained from Whatman (GE healthcare). The dialysis membranes (1000 and 3500 Da MWCO) were manufactured by Spectrum Laboratories, Inc. (Rancho Dominguez, CA) and dialysis was performed with ultrapure water. T1 relaxivity measurements were performed with a Bruker Minispec (Bruker Optics, Billerica, MA, U.S.A.) instrument: mq 20 NMR and mq 60 NMR series at magnetic field strengths of 0.47 and 1.41 T, respectively. Dynamic light scattering was performed with a Zetasizer Nanoseries, Malvern

Instruments (Malvern, U.K.) equipped with 633 nm wavelength laser. Luminescence measurements were conducted on a Spectra-Max M2 micro plate-reader from Molecular Devices (Sunnyvale, CA) or a Cary Eclipse fluorescence spectrometer by Agilent Technologies and analyzed with Spectra-Max or WinFLR software, respectively. Inductively coupled plasma optical emission spectroscopy (ICP-OES) was performed (to determine lanthanide content of the polymers) at University of Illinois at Urbana−Champaign using PerkinElmer 2000DV ICP-OES system. The synthesis of monomers and monomer precursors (1−4 reported in the Supporting Information) and the polymers were performed using procedures our lab has developed and previously published with some modifications,25,29,30 which are described briefly below (Schemes 1−3). Polymer Synthesis. Poly([N2,N3,N4-tris(tert-butoxycarbonyl) tetraethylenepentamine] Amidodiethylenetriaminetriaacetic Acid) (5a). Monomer 1a (1.074 g, 2.2 mmol) was added to a 25 mL dry flask containing 7.0 mL of dry dimethyl sulfoxide (DMSO) and dissolved at 80 °C. The mixture was then cooled to room temperature after dissolution. Monomer 4 (0.78 g, 2.2 mmol), dissolved in 3.5 mL of dry DMSO, was added via cannula to the solution of 1a and the mixture was allowed to polymerize at room temperature under dry atmosphere by stirring with magnetic stir bar for 18 h. The reaction mixture was then deposited into a dialysis bag (MWCO 1000 Da) and dialyzed against methanol for 24 h to remove unreacted monomers and low molecular weight oligomers. The dialysis with MeOH allows solvent exchange from DMSO to low boiling MeOH, which can then be evaporated more easily using a rotary evaporator. The product obtained was dried under vacuum overnight to yield a yellowish-white solid polymer. Yield: 1.25 g (69%). 1 H NMR (DMSO-d6): δ (ppm) = 1.39 (s, 27H), 2.83 (bm, 4H), 2.94 (bm, 4H), 3.25 (bm, 16H), 3.27 (s, 6H), 3.40−3.50 (bm, 4H), 8.0 (s, 1H). Poly([N2,N3,N4,N5-tetra(tert-butoxycarbonyl) Pentaethylenehexamine] Amidodiethylenetriaminetriaacetic Acid) (5b). Polymer 5b was synthesized using the procedure described for 5a, where 1b (0.46 g, 0.72 mmol) was used instead of 1a and the rest of the reagents were added in similar molar proportions. The final polymer was obtained in the form of a yellowish-white solid. 1614

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624

Biomacromolecules

Article

Scheme 3. Polymerization, Polymer Deprotection, and Lanthanide Chelation of the Polymers

Characterization of 6b. Yield: 0.21 g (89%). 1H NMR (D2O): δ (ppm) = 3.02−3.10 (bm, 20H), 3.23 (s, 6H), 3.32 (t, 4H), 3.40 (s, 4H), 3.51 (t, 4H), 3.84 (s, 2H). Characterization of 6c. Yield: 0.08 g (83%). 1H NMR (D2O): δ (ppm) = 2.98−3.04 (bm, 24H), 3.20 (s, 4H), 3.26 (t, 4H), 3.37 (bm, 6H), 3.46 (bm, 4H), 3.77 (s, 2H). Characterization of 6d. Yield: 0.106 g (60%). 1H NMR (D2O): δ (ppm) = 2.89−3.10 (bm, 28H), 3.21 (s, 4H), 3.29 (t, 4H), 3.38 (bm, 6H), 3.55 (bm, 4H), 3.80 (s, 2H). General Procedure for Chelation of Polymers with Gd or Tb (6aTb-6dTb and 6aGd-6dGd). Chelation of the polymers were carried out in 1:1 molar ratio between the metal (Gd3+ or Tb3+) and each polymer. Each polymer (0.082 mmol 6a−d) was dissolved in 1.7 mL of ultrapure water. Next, 0.082 mmol of GdCl3 or TbCl3 was dissolved in 0.8 mL of water and was then added to the polymer solution in three separate aliquots. The pH was adjusted to 6−7 using saturated NaHCO3 after each aliquot addition. The reaction mixture was then stirred at room temperature for 3 h. Next, the solution was deposited into a dialysis bag (MWCO 3500 Da) and dialyzed against ultrapure water for 24 h to remove the excess of LnCl3. After dialysis, the solution was lyophilized to yield a white, fluffy, chelated polymer. The percentage of lanthanide loading on each polymer was determined using ICP-OES analysis (Supporting Information). Yields: 6aGd, 87 mg (yield = 70%); 6aTb, 137.5 mg (yield = 67%); 6bGd, 19.4 mg (yield = 73%); 6bTb, 25.0 mg (yield = 75%); 6cGd, 37.1 mg (yield = 72%); 6cTb, 26.0 mg (yield = 62%); 6dGd, 52.0 mg (yield = 77%); and 6dTb, 43.7 mg (yield = 74%); the conversions for these polymerization were about 90% for all samples. Polymer and Polyplex Characterization. Size Exclusion Chromatography (SEC). Polymers 6aLn-6dLn (Ln = Gd and Tb) were characterized using SEC to determine the weight- average molecular weights and polydispersity indices (Table 1). It should be noted

Yield: 0.45 g (63.9%). 1H NMR (DMSO-d6): δ (ppm) = 1.37 (s, 36H), 2.43 (bm, 4H), 2.96 (bm, 4H), 3.23 (bm, 20H), 3.36 (s, 6H), 3.41−4.50 (bm, 4H), 8.09 (bm, 1H). Poly([N2,N3,N4,N5,N6-penta(tert-butoxycarbonyl) Hexaethylneheptamine] Amidodiethylenetriaminetriaacetic Acid) (5c). Polymer 5c was synthesized using the procedure described for the 5a, where 3c (0.18 g, 0.23 mmol) was used instead of 1a, and the rest of the reagents were added in similar molar proportions. The polymer was obtained in the form of a yellowish-white solid. Yield: 0.18 g (68%). 1H NMR (DMSO-d6): δ (ppm) = 1.36 (s, 45H), 2.82 (bm, 4H), 2.92 (bm, 4H), 3.20 (bm, 24H), 3.35 (s, 6H), 3.41 (s, 4H), 8.07 (bm, 1H). Poly([N2,N3,N4,N5,N6,N7-hexa(tert-butoxycarbonyl) Heptaethyleneoctamine] Amidodiethylenetriaminetriaacetic Acid) (5d). Polymer 5d was synthesized using the procedure described for the 5a, where 3d (0.34 g, 0.37 mmol) was used instead of 1a and the rest of the reagents were added in similar molar proportions. The polymer was obtained in the form of a yellowish-white solid. Yield: 0.34 g (72%). 1H NMR (DMSO-d6): δ (ppm) = 1.36 (s, 54H), 2.79 (bm, 4H), 2.90 (bm, 4H), 3.19 (bm, 28H), 3.31 (s, 6H), 3.37 (s, 4H), 8.09 (bm, 1H). General Procedure for Deprotection of Boc Groups from the Polymers (6a−d). The deprotection of the Boc groups on the polymer backbone was performed using a standard trifluoroacetic acid (TFA) deprotection method.31 Polymer (0.4 mmol of 5a, 5b, 5c, or 5d) was dissolved in 4.0 mL of DCM and cooled to 0 °C. To this solution, 4.0 mL of TFA was added, and the reaction mixture was allowed to warm to room temperature. The reaction mixture was then allowed to stir overnight at room temperature to allow deprotection of all the Boc groups. Complete deprotection was confirmed with NMR. The solvents were evaporated and the resultant polymer was redissolved in ultrapure water and neutralized using saturated NaHCO3. Next, the polymer solution was deposited into a dialysis bag (MWCO 3500 Da) and dialyzed against ultrapure water for 24 h. After dialysis, the solution was lyophilized to yield a white, fluffy, solid polymer. Characterization of 6a. Yield: 0.41 g (52%). 1H NMR (D2O): δ (ppm) = 3.02 (t, 4H), 3.12 (t, 4H), 3.20−3.30 (bm, 12H), 3.36 (t, 4H), 3.43 (s, 4H), 3.61 (bm, 4H), 3.86 (s, 2H).

Table 1. SEC Characterization of 6aLn-6dLn (Ln = Gd and Tb) Polymers

a

polymer

Mw (kDa)

Mw/Mn

na

6aGd 6aTb 6bGd 6bTd 6cGd 6cTb 6dGd 6dTb

17.6 18.9 16.0 18.3 14.2 14.2 15.0 15.1

1.3 1.3 1.2 1.2 1.2 1.2 1.1 1.2

25 26 22 24 18 18 19 18

Degree of polymerization.

that the unchelated polymers (6a−d) interacted (adsorbed) with the columns due to the anionic charge and, thus, were not able to be analyzed with SEC. SEC on the lanthanide-chelated polymers was performed in an aqueous mobile phase containing 0.45 M sodium acetate in water with 20 v% acetonitrile at a pH = 7 (adjusted using acetic acid), and at a flow rate 0.6 mL/min. Polymer dissolved in the mobile phase was separated using GMPWxL and G2500PWxL (Tosoh Bioscience) columns. This GPC system was equipped with a Waters 2489 UV/vis detector (l 1/4 274 nm), a Wyatt Optilab Rex refractometer, and a DAWN HELEOS-II multiangle laser light scattering (MALLS) detector. The data was recorded and analyzed using the ASTRA (version 5.3) software. Gel Electrophoresis Shift Assay. The pDNA binding ability of each polymer was studied using gel electrophoresis shift assays (Figure 2). Agarose gel (0.6 w/v %) containing ethidium bromide (0.06 w/v %) was prepared in Tris-acetate EDTA (TAE) buffer (40 mM Tris-acetate, 1 mM EDTA, pH 8). Polyplexes were formulated at N/P ratios (N = number of amines from the polymer backbone and P = number of phosphate units from DNA) of 0, 2.5, 5, 10, 20, 30, 40, and 50 by mixing equal volumes (10 μL each) of polymer and pDNA (0.1 μg/μL, pcmv lacZ). Polymer solutions were prepared using nuclease-free water 1615

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624

Biomacromolecules

Article

Figure 1. Schematic of monitoring polyplex formation and dissociation with lanthanide resonance energy transfer between terbium-chelated polycations and TMR-labeled pDNA.

Figure 2. Gel electrophoresis shift assay was performed on 6a−d (Gd and Tb) polymers to study the relative pDNA binding efficiencies. The binding ratio can be determined by the absence of a pDNA migration toward the positive electrode. The formulation ratios were varied from 0 to ∞ where 0 and ∞ indicate pDNA and polymer only controls, respectively. The binding ratio decreased from N/P 40 to N/P 5 with an increase in amine number from three to six along the polymer backbone. (Gibco, Invitrogen, Carlsbad, CA). After formulation, the polyplexes were incubated at room temperature for 30 min before addition of 2 μL of gel loading buffer (Blue juice, Invitrogen, Carlsbad, CA). Next, 10 μL of each N/P formulation was loaded into a well and an electrophoretic field (65 V, 60 min) was applied to study the inhibition of DNA migration toward the anode in presence of polymer. The gel was visualized using UV light and a digital camera. Dynamic Light Scattering. Hydrodynamic diameter and zeta potential of the polymer−pDNA complexes (polyplexes) were determined using a dynamic light scattering technique (Figure 3). First, the polyplexes were formulated at N/P ratios of 5, 10, 20, and 40 in nuclease-free water by mixing equal volumes (150 μL) of pDNA (pCMV lacZ, 6732 bp, 0.02 μg/mL) and Gd3+ polymer. Polyplexes were incubated at room temperature for 30 min to allow polymer−pDNA binding and polyplex formulation. Next, the polyplexes were diluted with 600 μL of water before size and zeta potential measurements. All of the measurements were performed in triplicate and error bars represent standard deviation of the three measurements. Determination of Effect on Longitudinal Relaxation Rate (R1) of Water. The effect on T1 relaxation times of water in the presence of 6a−d polymers chelated with Gd3+ and their polyplexes was studied with a Bruker Minispec (20 mq, 0.47 T and 60 mq, 1.41 T series) using an inversion recovery pulse sequence (180°−dt−90°−acquire; Figure 4). Solutions of 6a−d for the Gd3+ polymers and Magnevist (a clinical contrast reagent) were prepared (1.0 mM) in ultrapure water (based on repeat unit molecular weight and on a per “Gd3+” basis). For the polyplex samples, equal volumes of polymer and pDNA were mixed in order to achieve 1 mM concentration on a per Gd3+ basis to standardize all concentrations for accurate comparison. All the measurements were performed in triplicate and error bars represent the standard deviation of the three measurements.

Figure 3. Hydrodynamic diameter was determined using dynamic light scattering for polyplexes formulated with pDNA and polymers 6a−dGd at the N/P ratios indicated and the zeta potential of the polyplex formulations was also measured. The bars (left y-axis) denote the hydrodynamic diameters of the polyplexes formulated at N/P ratios of 5, 10, 20, and 40. Each bar 6a (red), 6b (green), 6c (purple), and 6d (blue) represents an average of three hydrodynamic diameter measurements. The colored lines (right y-axis) represent the zeta potential values for polyplexes formulated with pDNA and polymers 6a−d Gd. Each point on the line 6a (asterisks), 6b (orange spheres), 6c (blue solid line), and 6d (pink rectangles) represents an average of three measurements with standard deviation indicated by an error bar.

Determination of Number of Water Coordination Sites (q) for Tb Polymers. The number of water coordination sites (q) per chelate was determined by measuring the luminescence lifetimes of the 6a−d polymers chelated with Tb3+ (6aTb-6dTb) in D2O and H2O (Table 2) using a Cary Eclipse fluorescence spectrometer (Agilent Technologies, 1616

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624

Biomacromolecules

Article

Figure 4. Water proton relaxation rate constants (R1) in the presence of 6a−dGd polymers and polyplexes formulated at an N/P of 40 was measured at 60 MHz (a) and 20 MHz (b) and compared to Magnevist. Each sample was prepared at 1 mM Gd concentration and each bar represents an average of three measurements and standard deviation represented by an error bar.

⎡⎛ ⎤ 1 1 ⎞⎟ qTb = 5⎢⎜⎜ − − 0.06x ⎥ ⎟ ⎢⎣⎝ τH O ⎥⎦ τD2O ⎠ 2

Table 2. Luminescence Lifetimes of 6a−dTb in H2O and D2O to Calculate q sample

τD2O,ms

τH2O,ms

q

6aTb 6bTb 6cTb 6dTb

2.895 ± 0.035 2.796 ± 0.035 2.695 ± 0.023 2.752 ± 0.034

1.614 ± 0.022 1.603 ± 0.022 1.569 ± 0.015 1.553 ± 0.028

0.77 ± 0.041 0.73 ± 0.041 0.73 ± 0.027 0.80 ± 0.044

(1)

x = number of N−H oscillators from amide groups coordinated to the Tb3+ through the carbonyl. LRET Studies. General Procedure for Preparation of Polyplexes with Tb-Chelated Polymer and TMR-Labeled DNA. Polyplexes were formed at N/P ratios of 20 and 40 by mixing equal volumes (50 μL) of two aqueous solutions of Tb polymer and TMR-labeled pDNA (0.1 μg/μL). The resulting mixture was incubated at room temperature for 30 min to allow polyplex formation (Figure 1). Control samples were as follows: polymer only (50 μL of polymer + 50 μL of water), pDNA only (50 μL of pDNA + 50 μL of water), and unlabeled polyplexes (polyplexes formed with Tb polymer and unlabeled pDNA). Monitoring Change in Terbium Emission Intensity upon Polyplex Formation. A total of 30 min after polyplex formation, 90 μL of each polyplex solution was pipetted into each well of a 96-well plate and analyzed for luminescence intensity of Tb using a microplate reader. Luminescence intensity in each well was measured with a 90−100 μs delay time using end point (excitation 350 ± 5 nm, emission 545 ± 5 nm) and full-spectral scan (400−700 nm) type measurements (Figure 5). Time delayed (90−100 μs) measurements were performed

U.S.A.; formerly Varian). The data was processed with Agilent Cary WinFLR software. Each sample (polymers 6a−d chelated with Tb3+) was prepared at 0.5 mM concentration (800 μL) using D2O and H2O and deposited into a small-volume quartz cuvette. Lifetime measurements were performed with the excitation wavelength of 350 nm (slit 5 nm), emission wavelength of 550 nm (slit 10 nm), a delay time of 0.1 ms, a gate time of 0.3 ms, and a total decay time of 15 ms. All the measurements were performed in triplicate and standard deviation was determined. The revised Horrocks32,33 equation (eq 1) was used to calculate the number of water coordination sites (q) per chelate unit. The number of water coordination sites on each Tb3+ atom chelated by the polymers was found to be ∼1 for all the polymers.

Figure 5. Observing LRET upon polyplex formation with Tb3+-labeled polymer 6c and TMR-labeled pDNA. (a) An overlay of the luminescence spectra (ex = 350 nm) of the lanthanide polymer 6cTb only (6cTb, N ratio = 40, no pDNA, green solid line), polyplexes formulated with polymer 6cTb complexed to unlabeled pDNA at an N/P ratio of 40 (6cTb + pDNA N/P 40, pink dotted line with triangles), and polyplexes formulated with 6cTb polymer complexed with TMR-labeled pDNA at an N/P ratio of 40 (6cTb + TMR pDNA N/P 40, blue dotted line with circles). (b) Plot of the change in luminescence intensity (at 545 nm) of the 6cTb polymer only (no pDNA, blue bars), polyplexes formulated with 6cTb polymer complexed with unlabeled pDNA (green bars), and polyplexes formulated with 6cTb polymer complexed with TMR-labeled pDNA (orange bars) as monitored at different N/P ratios. 1617

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624

Biomacromolecules

Article

to avoid signal interference from direct excitation of TMR and shortlived background fluorescence. Gel Electrophoresis Shift Assay to Monitor Polyplex Destabilization. Polyplexes were formulated by mixing equal volumes (10 μL) of polymer solution and (unlabeled) pDNA (0.1 μg/mL) to achieve an N/P ratio of 40 for the polyplexes; the polyplex solutions were subsequently incubated at room temperature for 30 min to allow binding. Next, 10 μL of NaCl (1.0 M in 20 mM HEPES buffer) or heparin sulfate solution (100−500 μg/mL) was added to each polyplex solution and incubated further for 15 min. Control samples were prepared by mixing 10 μL of water with samples formulated at N/P 40 (polyplexes) and N/P 0 (pDNA only). Gel loading buffer (1 μL, Blue juice, Invitrogen, Carlsbad, CA) was added to each solution before depositing 10 μL of each solution into the well. Dissociation of pDNA from polyplexes was studied via gel electrophoresis shift assay (Figure 6).

performed using Zetasizer Nanoseries from Malvern Instruments (Malvern, U.K.) equipped with 633 nm wavelength laser at 25 °C. Monitoring Change in Terbium Emission upon NaCl and Heparin Addition. The electrostatic interactions and thus binding between polymer and pDNA can be disrupted by the addition of salt (1 M NaCl) or a competitive polyanion such as heparin. The concentration of heparin at which the dissociation takes place was determined to be 500 μg/mL via a gel electrophoresis assay. For both the destabilization experiments, polyplexes were formulated at N/P ratios of 40 and 20. Polyplexes at different N/P ratios along with the control samples were formulated as described above. After 30 min, 90 μL of each of the solution was pipetted into a 96-well plate and Tb emission was monitored with a time delayed (90−100 μs) end point measurement (excitation 350 ± 5 nm, emission 545 ± 5 nm). Next, 50 μL of 1 M NaCl in HEPES buffer (20 mM) or heparin solution (500 μg/mL concentration) was added to each polyplex and control samples within each of the wells. All the samples were incubated further for 15 min and analyzed for a change in Tb luminescence intensity (Figures 7 and 8). Cell Culture Studies. General Procedure for pDNA Transfection of Human Glioblastoma (U87) Cells. Prior to transfection, U87 cells were seeded at 5 × 104 cells/well in 24-well plates, and were incubated for 24 h to yield a confluency of above 80%. Polyplexes (N/P 40) were formulated by mixing equal volumes (150 μL) of each polymer solution (polymers 6a−d chelated with Gd3+ and Tb3+) with pDNA (gWiz-Luc, 0.02 μg/μL) containing the firefly luciferase reporter gene and incubated at room temperature for 1 h. Jet-PEI polyplexes (positive control) were formulated at N/P 5 following the manufacturer’s protocol. Along with the above-mentioned samples, untransfected cells and cells transfected with pDNA only were used as negative controls. Next, each polyplex solution (6aLn-6dLn polymers, pDNA and Jet-PEI) was diluted to 900 μL with serum-free media (Opti-MEM, pH 7.2) and each well of cells was transfected with 300 μL of this polyplex solution containing 1 μg of gWiz-Luc pDNA. At 4 h after transfection, 800 μL of the supplemented DMEM was added to each well. At 24 h after transfection, the media was replaced with fresh supplemented DMEM (1 mL). A total of 48 h after initial transfection, the media was removed and the cells were washed with 500 μL of PBS and treated with cell culture lysis buffer (Promega, Madison, WI). The amount of protein in the cell lysates (as mg of protein) was determined against a standard curve of bovine serum albumin (98%, Sigma, St. Louis, MO), using a Bio-Rad DC protein assay kit (Hercules, CA). Cell lysates were analyzed for Luciferase activity with Promega’s Luciferase assay reagent (Madison, WI). For each sample, luminescence was measured over 10 s in duplicate with a luminometer

Figure 6. Gel electrophoresis shift assay was performed on 6cTb polyplexes formulated with pDNA at an N/P ratio of 40 after addition of different concentrations of heparin (indicated by numbers from 0 to 500 μg/mL) and NaCl (at 1 M concentration). The binding of pDNA to the polymer is indicated by the inhibition of migration of the orange-colored band toward positive electrode. Agarose gel (0.6 w/v %) containing ethidium bromide (0.06 w/v %) was prepared in Tris-acetate EDTA (TAE) buffer (40 mM Tris-acetate, 1 mM EDTA, pH 8) and migration of DNA toward the anode was observed under applied electrophoretic field (65 mV, 60 min). Particle Size Measurement upon Addition of NaCl and Heparin. Change in particle size upon addition of NaCl and heparin was monitored via DLS. Polyplexes were formulated by mixing equal volumes (200 μL) of polymer (6cTb) and (unlabeled) pDNA at N/P 40. The polyplex size was measured before and after addition of 200 μL of 1 M NaCl solution (in 20 mM HEPES buffer) or heparin solution (500 μg/mL in water; Figure S2). Dynamic light scattering was

Figure 7. Overlay of the change in luminescence intensity (emission at 545 nm) of 6cTb polymer only (ex = 350 nm; denoted as 6cTb, N = 40, and 6cTb, N = 20, blue and the purple solid curves, respectively) at concentrations equivalent to “N” ratios of 40 and 20, 6cTb polymer complexed with unlabeled pDNA at N/P ratios of 40 and 20 (denoted as 6cTb + pDNA N/P = 40 and 6cTb + pDNA N/P = 20 in the red and pink dotted curves, respectively), and 6cTb polymer complexed with TMR-labeled pDNA at N/P ratios of 40 and 20 (denoted as 6cTb + TMR pDNA N/P = 40 and 6cTb + TMR pDNA N/P = 20 in the green and orange dashed curves, respectively) before (a) and after (b) addition of 500 μg/mL heparin solution. 1618

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624

Biomacromolecules

Article

Figure 8. Overlay of the change in luminescence intensity (emission at 545 nm) of 6cTb polymer only (ex = 350 nm; denoted as 6cTb, N = 40, and 6cTb, N = 20, blue and the purple solid curves, respectively) at concentrations equivalent to “N” ratios of 40 and 20, 6cTb polymer complexed with unlabeled pDNA at N/P ratios of 40 and 20 (denoted as 6cTb + pDNA N/P = 40 and 6cTb + pDNA N/P = 20 in the red and pink dotted curves, respectively), and 6cTb polymer complexed with TMR-labeled pDNA at N/P ratios of 40 and 20 (denoted as 6cTb + TMR pDNA N/P = 40 and 6cTb + TMR pDNA N/P = 20 in the green and orange dashed curves, respectively) before (a) and after (b) addition of a 1 M NaCl solution.

Figure 9. Luciferase expression (a) and MTT assay (b) were performed 48 and 24 h after transfection, respectively, in the U87 cell line to screen transfection efficiencies and cytotoxicity of the different polyplex types formed with pDNA and the polymer analogs (6a−d, chelating either Gd3+ or Tb3+) at N/P ratio 40, compared with the control Glycofect (N/P ratio 20), naked pDNA, and untransfected cells. (a) The left y-axis (bar graph) represents RLU per mg of protein expressed, whereas the right y-axis (line graph) indicates total protein content used to normalize the expression data and is indicative of cell metabolic level; (b) MTT assay data, the bar graph represents the U87 cell survival for each cell sample exposed to the indicated solutions of polyplexes and controls. (GENios Pro, TECAN US, Research Triangle Park, NC), and the average was utilized. The gene delivery efficiency of each sample was characterized by firefly luciferase expression in U87 cells and denoted as

relative light units (RLU)/mg protein (Figure 9). The cell viability profiles were characterized by the amount of protein in the cell lysates. The protein level of untransfected cells was used to normalize the data 1619

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624

Biomacromolecules

Article

obtained for the protein levels of the cells transfected with naked pDNA and polyplexes formed with different polymers (6aLn-6dLn and Jet-PEI). MTT Assay to Study Cellular Toxicity. MTT reagent (3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) was used to estimate the cytotoxicity of the formulations. MTT can be reduced to purple formazan in living cells under the catalysis of mitochondrial reductase. For this assay, U87 glioblastoma cells were seeded at 5 × 104 cells/well in 24-well plates 24 h prior to transfection. Plasmid DNA transfection was carried out as mentioned above and 1 mL of DMEM was added to each well 4 h after transfection. A total of 24 h posttransfection, the media was aspirated and the cells were washed with PBS (500 μL/well). Serum-containing DMEM (1 mL) with 0.5 mg/mL of MTT was added to each well and cells were incubated for another hour (Figure 9). The media was then replaced with 600 μL of DMSO for 15 min at room temperature. A 200 μL aliquot of the media was transferred to a well of a 96-well plate for analysis by colorimeter with a wavelength at 570 nm. Samples of nontransfected cells were used for normalization.

number from 5 to 6 along the polymer backbone. It was also observed that the polymers with the higher amine numbers (5 and 6, 6c,d) were able to completely inhibit pDNA migration at N/P = 5 (compared to analogues with 3 and 4 amines, which bound at N/P 30 or higher), as indicated by the absence of streaking or a secondary migration band (Figure 2). Particle Size and Zeta Potential Determination. DLS was performed to determine the polyplex size and colloidal stability of the polyplexes in water and to compare if polyplex size differs between the polyplexes formulated with the different polymer structures. The hydrodynamic diameter of these polyplexes was found to be between 80 and 125 nm at an N/P of 5; however, at N/Ps of 10−40, the diameter significantly decreased to 50−80 nm, signifying tighter binding and compaction at higher N/P ratio (Figure 3). The zeta potential values for the polyplexes were all highly positive (10−50 mV) and for each complex type the values increased with N/P ratio. Relaxation Rate Constant Measurements. The effect of both the polymers and polyplexes on T1 relaxation of water protons was examined at two magnetic fields: 0.47 and 1.41 T using a Minispec instrument (Figure 4). The relaxation rate constants (R1) of water protons in the presence of 6a−dGd polymers and the subsequent polyplexes formed with each polymer and pDNA were determined to be between 7.7 and 9.3 s−1 (calculated on a per Gd basis). A specific trend was not observed corresponding to the number of amines along the polymer backbone, which was to be expected due to the concentration of Gd3+ being the same for all samples. However, these relaxivities were at least two times higher than the positive control, Magnevist, a clinical contrast agent, which was examined at same Gd3+ concentration for comparison. The increase in R1 can be attributed to the increase in molecular size/weight (polymerized Gd3+ chelates) and rotational correlation times of the larger polymers in comparison to that of Magnevist, which is a small monomeric contrast agent. The increase in relaxation rates of polymer/polyplexes compared to that of Magnevist (on per Gd3+ basis) can aid in imaging contrast to understand polymer interactions within living systems. Determination of Number of Water Coordination Sites (q) per Chelate in the Repeat Unit. Luminescence lifetime (τ) measurements of polymers 6a−dTb in D2O and H2O were performed to calculate the average number of water molecules (“q”) present in the first coordination sphere of the Tb3+-chelate (Table 2). The rate of luminescence decay (k) was higher in H2O compared to D2O due to quenching of luminescence through O−H bond vibrational overtones.34,35 The N−H bond from the amide group also contributes to this nonradiative quenching of luminescence and is accounted for in the eq 1 through the term 0.06χ to obtain a more accurate value of “q”. The “q” values for 6a−d polymers were found to be 0.73, 0.77, 0.73, and 0.80, respectively, which implies the presence of about 1 water molecule per repeat unit containing a Tb3+ in the chelate structure (we assume that the Gd3+ chelates have similar water coordination numbers). This information is interesting as it points to the fact that these structures have a similar water coordination number to Magnevist, yet, these polymers, on a per Gd basis, have two times the relaxivity profiles. This observation supports the use of these polymer structures to aid understanding of the nucleic acid delivery process by taking advantage of the imaging capability of the lanthanide metals chelated within the polymer structures. LRET Studies. Change in Luminescence Intensity upon Polyplex Formation. Polyplexes were formulated with the

■

RESULTS Synthesis of Monomers and Polymers. The synthesis of the monomers (described in Supporting Information) with the highest purity is crucial for the step-growth polymerization reaction utilized herein. Diamine monomers were separated from impurities by column flash chromatography or recrystallization. The purity of these oligoethylene amine monomers was determined using LC-MS analysis and that of DTPA-BA monomer was determined using NMR integration. A series of polymers with different amine number (3−6, 6a−d) were synthesized via step-growth polymerization of the oligoethyleneamine monomers with DTPA-BA (Schemes 1−3). The monomers were reacted in stoichiometric ratios in an effort to achieve a high degree of polymerization (n). All the polymerizations were performed in dry DMSO (to prevent hydrolysis of the anhydride ring) for 18 h at room temperature to obtain polymers with similar n values between 18 and 26 (Table 1). The completion of every step during polymerization was monitored via NMR. For example, after the first step of polymerization, N−H protons from the newly formed amide bond were observed around 8.5 ppm. Deprotection of Boc groups from the secondary amines along the polymer backbone was confirmed by the absence of t-butyl resonances around 1.4 ppm. After each step, the polymers were purified from unreacted monomers and low molecular weight oligomers via extensive dialysis in ultrapure water. Characterization of Polymers and Polyplexes. SEC Characterization of Polymers. SEC analysis was performed on the library of polymers (6a−d chelated to Gd3+ and Tb3+) to determine the weight average molecular weight (Mw; Table 1). The Mw was found to range from 14.3 to 18.9 kDa (n = 18−26). The polydispersity index (ratio of Mw/Mn) was found to be artificially low, between 1.1 and 1.3. This significantly low PDI can be attributed to the low conversion and extensive dialysis performed after each step before achieving the final polymer structures. Gel Electrophoresis Shift Assay. A gel electrophoresis assay was performed to determine a suitable ratio for formulation of pDNA and each of the polymers for polyplex formulation. Previous results have shown that the N/P ratios of polyplex formation can be rather high for this polymer series (and highly dependent on ethyleneamine number).25 The N/P ratio of binding and polyplex formation was found to decrease from a high N/P ratio of between 40 and 50 (for polymers, 6a,b with lower ethylene amine units 3−4 in the repeat) to a lower N/P ratio between 5 and 10 with the increase in ethyleneamine 1620

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624

Biomacromolecules

Article

Tb3+-chelated polymers and TMR-labeled pDNA to aid understanding of the polymer-pDNA binding process through LRET. The luminescence intensity of terbium chelated-polymer (6cTb) decreased significantly upon formation of polyplexes with TMR-pDNA (6cTb+TMR pDNA) due to nonradiative energy transfer between the donor (Tb)-acceptor (TMR; Figure 5). Lanthanide luminescence (at 490, 545, 590, and 620 nm) was quenched (43%) upon polyplex formation with TMR-pDNA compared to control samples of polyplexes formulated with unlabeled pDNA (pDNA) at the same N/P ratio of 40 (Figure 5a), indicating that pDNA by itself is not an effective acceptor in this energy transfer process. This phenomenon was studied at N/P ratios of 50, 20, 10, and 5 and the extent of quenching was found to increase with the increase in N/P ratio. Polyplex Destabilization with Heparin and NaCl: Gel Electrophoresis and DLS Examination. Polyplexes formulated with 6cTb and pDNA at N/P 40 were subjected to gel electrophoresis after treatment with different concentrations of heparin (100−500 μg/mL) and 1 M NaCl. The heparin began to disrupt the polymer-pDNA binding at a concentration of about 450 μg/mL; however, 1 M NaCl alone was not able to completely release pDNA from the polyplexes (Figure 6). This indicates that the polyplex formation is stable and a multivalent anion (such as heparin) is needed to fully dissociate the polyplexes. These data are interesting as they indicate that the polyplexes are likely stable at physiological salt conditions (150 mM) from dissociation and that a strong interaction with biological polyanions is needed to facilitate pDNA release. DLS data (Figure S2, Supporting Information) shows that the addition of NaCl to the polyplexes (at N/P 40) results in a significant increase in size (from 60 to 1000 nm). This can be attributed to the disruption of electrostatic interaction between the polymer and pDNA (and neutralization of electrostatic repulsion of the complexes) due to the salt, leading to swelling and aggregation of polyplexes. However, in the case of heparin, no significant variation in size was observed (Figure S2, Supporting Information). This was further investigated by comparing the size of the polymer−pDNA and polymer− heparin complexes (Figure S2, Supporting Information) and both types of complexes possess similar particle sizes (60−80 nm). The addition of heparin to the polyplex solution results in formation of a complex composition consisting of polymer−DNA, polymer− heparin, and polyplex−heparin interactions that cannot be distinguished easily by DLS measurement. The results from gel electrophoresis assays (Figure 6, indicated by migration of unbound pDNA toward the anode) provide convincing evidence of disruption of polymer−pDNA binding by this large macromolecule; this experiment mimics the competitive environment that polyplexes face when they encounter highly negative cell surface glycosaminoglycans such as heparin sulfate. Change in Terbium Emission upon Polyplex Destabilization. According to LRET theory,18,36,37 when a donor (Tb3+) and an acceptor (TMR) are separated by more than 100 Å, the “LRET” effect is reversed (i.e., no change in donor emission intensity in presence of acceptor). Herein, we have developed lanthanide-containing polyplex systems to monitor and understand assembly and disassembly of the polyplexes in presence of salt and charged macromolecules (simulating the cellular environment). This phenomenon was tested by monitoring the change (or increase) in the donor (Tb) emission upon polyplex dissociation/destabilization. NaCl destabilizes the electrostatic interaction between the polymer and pDNA,38 which leads to

swelling or aggregation of nanoparticles (Figure S2, Supporting Information). However, at this concentration according to the gel shift assay, pDNA is not released from the polyplex structure (Figure 6). Heparin, a multivalent polyanion, can compete with the pDNA to form nanoparticles with the polymer39 and, thus, can disrupt the polymer−pDNA binding (Figure 6). Both of these events cause a change in the distance between the donor (Tb-polymer) and acceptor (TMR-pDNA) and can be monitored via LRET. LRET was followed via monitoring Tb3+ luminescence intensity upon the addition of heparin or NaCl to the polymer and polyplex solutions (Figures 7 and 8). When polyplexes are disassembled or the binding is significantly “loosened”, a decrease in LRET will result, which causes an increase in luminescence intensity for the Tb3+ polymer. Polyplexes were formulated at N/P ratios of 20 and 40 with TMR-pDNA and Tb3+ emission was measured at 545 nm before and after addition of heparin and NaCl. Controls were also monitored at concentrations corresponding to N/P ratios of 20 and 40 and consisted of 6cTb polymer only (“N ratios” of 20 and 40) and polyplexes formed with 6cTb polymer and unlabeled pDNA at N/P ratios of 20 and 40 (denoted as 6cTb + pDNA, N/P 40, 20). As shown in Figures 7 and 8, the controls of 6cTb polymer only and the polyplexes formed with unlabeled-pDNA yielded the highest luminescence intensity (due to a lack of TMR, the LRET acceptor). However, for polyplexes formed at an N/P ratio of 20 and 40, luminescence was significantly reduced. It should be noted that the luminescence intensity at N/P of 40 is higher due to the higher polymer concentration (and, thus, also chelated metal) in solution. Upon addition of heparin, the LRET intensity was restored to the original value as in the absence of TMR, indicating complete disassembly of polyplexes (Figure 7) due to competitive binding of polyanionic heparin with the 6cTb polymer. This exact experiment was then repeated by adding NaCl to the polyplex solutions and controls (Figure 8). While a minimal change in emission intensity of the controls was observed, LRET was also disrupted for the 6cTb/TMR-pDNA polyplexes (Tb emission recovered to about 80% of the polymer only and polyplexes with unlabeled-pDNA) signifying that monovalent ions also loosen the polyplexes and can play a role in partial disruption of polyplex formation (Figure 8). In this study, LRET has been used to study association and dissociation of polyplexes in vitro by monitoring change in emission of the donor.18,21,40 We have shown that the change in distance between the polymer (donor) and TMR-pDNA (acceptor) can be detected by monitoring the change in emission intensities of the polymer. In our studies, we observed a significant decrease in emission intensity of the Tb-chelated polymer (∼43%) when complexed with TMR-pDNA at different N/P ratios. This effect was completely reversed (resulting in a full recovery of Tb emission intensity) upon addition of heparin, denoting pDNA release from the polyplexes. About 80% recovery of emission intensity was observed with the addition of 1 M NaCl, which signified destabilization of the polyplexes by either weakening the electrostatic interaction between the polymer and pDNA (possibly through polyplex swelling/ aggregation) or partial pDNA release. Cell Culture Studies. Luciferase Assay in U87 Cells to Monitor Transfection Efficiency. U87 cells were transfected with polyplexes formulated with pDNA containing the Luciferase reporter gene (gWiz-Luc, 0.02 μg/mL) and the series of polymers (containing 3−6 amines/repeat units, 6a−d) chelated with Gd3+ or Tb3+. The transfection efficiency and the 1621

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624

Biomacromolecules

Article

They used tricolor FRET consisting of QD and cyanine dye (BOBO-3 or Cy3) labeled pDNA and Cy5-labeled polymer. Turning “ON” and “OFF” of the fluorescence (FRET) signal was used to monitor intracellular release and degradation of pDNA. Lanthanides, however, offer many advantages over organic dyes for intracellular imaging such as narrow emission peaks, lack of photobleaching, large Stoke’s shifts and fluorescence lifetimes, and tunable emission wavelengths that are different than the autofluorescent background characteristic of native biological molecules.23 LRET has been previously utilized to study protein−protein interactions in high-throughput screening of antibacterial drugs.45 In a study conducted by Bergendahl et al., the binding between europium labeled sigma factor (σ-70) and Cy-5 labeled RNA polymerase was studied by monitoring a change in LRET signal from Cy-5 (acceptor) emission.45 Even though the underlying mechanism for LRET is similar to FRET, LRET provides several advantages over FRET such as larger measurable distances (>100 Å), higher signal-to-noise ratios and long fluorescence lifetimes.18,36 In the current study, the structures were fully characterized for molecular weight and then examined for their ability to bind and compact pDNA into polyplexes. The effect of amine number on pDNA binding ratio or N/P ratio was determined using a gel electrophoresis assay. This assay revealed that the N/P binding ratio decreased drastically from 40 to 5 with the increase in ethyleneamine number from three (6a polymer chelates) to six (6d polymer chelates) within the polymer repeat unit. This result indicated that binding affinity and polyplex stability is highly enhanced with the increase in amines within the repeat unit. In addition, it was found that in some cases, the analogs chelating Gd3+ bound to pDNA at a lower N/P ratio, signifying a slightly higher relative binding affinity. The polyplex sizes were determined via DLS and were found to be between 50 and 80 nm, however, a significant trend was not found correlating particle size to amine number. Within each polymer class, there was a slight decrease in size with the increase in N/P ratio from 5 to 40 (due to more compact binding of pDNA with additional polymer). Zeta potential measurements indicated that at the same N/P ratio of 40, the polyplexes all offered a highly positive potential from +25 to +50 mV, which generally did increase with amine number and signified overcharging of the polyplexes. We have shown in a previous collaborative study by Wang et al. that polyplexes formulated with lanthanide-based polymers can be significantly overcharged; NMR studies on related polymer− pDNA complexes have shown that at an N/P ratio of 40, only about 10% of the polymer is free in solution (thus, polyplexes have an effective N/P = 30).46 The ability of these polyplex systems to be imaged via MRI was investigated through T1 relaxivity measurements of aqueous polyplex and polymer solutions created from the systems chelated with Gd3+ at two magnetic field strengths of 0.47 and 1.41 T using a Minispec instrument. The relaxivity rate constants (R1 values) for the 6a−dGd polymers and their complexes with pDNA were measured by formulating the solutions at an N/P ratio of 40. The R1 values, when compared to that of Magnevist (a clinical contrast agent), were two times higher on per Gd basis (same Gd3+ concentration) at the same magnetic field strengths, indicating that these systems could be used as contrast agents for MR imaging. To aid in understanding the average chelate structure of the metals within the DTPA units, the number of water molecules directly coordinated to the polymer-bound metal atoms were measured via luminescence lifetime of polymer solutions in the presence of H2O and D2O and calculated via the

viability of cells exposed to the different polyplex formulations (at an N/P ratio of 40) were compared to the cells transfected with Glycofect (N/P 20, positive control) and the negative controls of naked pDNA and untransfected cells. The luminescence intensity (measured in relative light units, RLU) from the protein expressed by the pDNA encoding firefly Luciferase was used to study transfection and the amount of protein in cell lysates was used to determine viability relative to untransfected cells. The viability of cells transfected with all the polymer chelates and Glycofect was about 100% (Figure 9a). Transfection efficiency of Glycofect was the highest among the samples tested. For all of the polyplex types containing the metal chelates there was no obvious trend; however, the transfection efficiency was significantly higher for the polyplexes formed with models containing six ethyleneamine units (6d series chelating either Tb3+ or Gd3+) compared to the polymer with 3−5 ethyleneamines (6a−c). Interestingly, polyplexes formed with the polymer analogs 6aGd and 6bGd showed anomalous behavior, where the transfection efficiency was higher than the same polymer chelated with Tb3+ (6aTb and 6bTb). Currently, we do not fully understand this effect; however, we have noticed in the gel data (Figure 2) that the polymer analogs chelating Gd3+ appear to bind pDNA at a lower N/P ratio than the Tb3+ analogs. The differences in the transfection data could be, in part, due to differences in binding affinity, however, this effect is not noticed with 6cGd polyplexes. MTT Assay to Study Cellular Toxicity in U87 Cells. MTT assays were carried out to investigate toxicity profiles of the polyplex formulations (6a−dGd and Tb) and the data was compared to the controls of naked pDNA and untranfected cells. Cell viability of all the samples was determined via absorbance (measurement of purple formazan at 595 nm) in each well. All of the lanthanide containing polyplexes (6a−dGd and Tb) showed (Figure 9b) little to no toxicity (viability ∼ 100%) in U87 cells at N/P ratio of 40. These results are promising to pursue delivery studies of these polymer models to further understand and image polyplex dynamics in vitro and potentially in vivo.

■

DISCUSSION In this study, we have designed and developed a series of theranostic polymers that house ethyleneamine units (3−6) and a metal chelate functionality (Gd3+ or Tb3+) within the repeat units. The oligoethyleneamine domains facilitate binding and compaction of pDNA into polyplexes and metal chelate domains can be utilized for polyplex imaging via MRI and fluorescence microscopy techniques. These polymers were synthesized via step-growth polymerization of oligoethylenediamine monomers and DTPA-BA that yielded a degree of polymerization between 18 and 26. With these structures, we sought to explore the chelation of Tb3+ within these architectures and explore the use of LRET to understand polyplex dissociation dynamics in the presence of salt and a polyanionic competitor. We believe these materials offer a unique platform that contains copolymerized fluorescent metals that do not photobleach and offers advantages over conventional techniques of conjugating organic dyes to polymers for intracellular trafficking studies. It should be noted that FRET techniques have been utilized for understanding polyplex dynamics. There are several examples41−44 of monitoring nucleic acid delivery and unpackaging of polymeric and lipid-based carriers via FRET. For example, Leong and coworkers40 have studied polyplex dynamics (polyplex formation, DNA release, and degradation) of chitosan and a polyethylene glycol-polyphosphoramidate block copolymer (PEG-b-PPA). 1622

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624

Biomacromolecules

Article

exploration to monitor polyplex dissociation using LRET and we believe this study opens up avenues to investigate vehicle unpackaging in cellulo experiments in the future. Current efforts are directed toward understanding the trend in transfection of U87 cells, exploring different cell types for the nucleic acid delivery, evaluating LRET in other nucleic acid types (i.e., RNA), and exploring in cellulo LRET experiments.

Horrock’s equation. The water coordination values were found to be on average slightly less than 1.0, indicating about 1 water molecule per chelate is directly bound to the metal center. Because the polymer chelate structure (DTPA) is similar to that of Magnevist, we expected the water coordination number to be similar, indicating that the metal chelate likely has a similar binding stability. Collectively, these data indicate that the increase in R1 is primarily due to the larger polymer and polyplex structures. These results are promising and promote the further use and study of these unique systems via MR imaging in the cell and tissue environment. A primary goal of this study was to develop a lanthanide resonance energy transfer or LRET-based tool to monitor polyplex association and dissociation through monitoring the formation and dissociation of polyplexes (created with the Tb3+labeled polymers and TMR-labeled pDNA). We found that upon formation of polyplexes with TMR-labeled pDNA and 6cTb, a significant decrease in emission of Tb3+ luminescence (43%) was noticed when the emission was monitored at 545 nm at different N/P ratios of polyplex formulation: 50, 40, 20, 10, and 5. However, the emission intensity of 6cTb remained unchanged in polyplexes formulated with unlabeled pDNA, indicating that the pDNA itself is not a LRET acceptor. The LRET effect was reversed upon the addition of a NaCl or heparin solution, albeit, to different degrees. These ions destabilize polyplexes by weakening the electrostatic interaction between the cationic polymer and polyanionic pDNA, which causes an increase in distance between the donor (6cTb) and acceptor (TMR-pDNA) and can be monitored. During this study, we observed a complete recovery of Tb3+ emission upon addition of heparin (indicating complete polyplex disruption) and about an 80% recovery with the addition of NaCl (indicating destabilization) to the polyplex solution. Thus, we have shown that both polyplex formation and destabilization/dissociation can be monitored by observing changes in the emission of Tb3+ (donor) chelated within the polymer structure. These conditions were meant to mimic the environment encountered in the cell. Polyplexes are exposed to high local concentrations of charged macromolecules when being endocytosed, indicating that competitive binding of polycations to polyanions within the cell can destabilize polyplexes and cause nucleic acid release within the cell. MTT assays of cultured U87 cells exposed to each of the polyplex solutions (formed with pDNA and polymers 6a−dTb and 6a−dGd) at 24 h after transfection indicated no significant cytotoxicity (∼100% cell survival) with all analogs. Screens of the cellular transfection efficiency determined using a standard Luciferase expression assay at 48 h after transfection in U87 cells revealed that transfection efficiency increased significantly with the structures containing 6 ethyleneamine units (6d polymers). However, anomalies with these data were found where 6aGd and 6bGd structures also showed some transfection, which could be related, in part, to the gel shift data that the Gd3+ analogs may have a slightly stronger binding affinity to pDNA, yet, this effect is not currently understood.

■

ASSOCIATED CONTENT

S Supporting Information *

Synthesis of monomers, SEC of polymers, DLS data for the LRET studies, and ICP-OES data. 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.

■

ACKNOWLEDGMENTS This project was funded by National Science Foundation (DMR 1105895). The authors would like to thank the following people for use of instrumentation in their laboratories Prof. Valerie Pierre, Prof. Webster Santos, Prof. Richey Davis, Prof. Tim Long, and Prof. Tijana Grove. The authors would also like to thank Drs. Nilesh Ingle and Giovanna Grandinetti for labeling of the pDNA. S.K. would like to thank Dr. Karina Kizjakina, Dr. Antons Sizovs, and Dr. Josh Bryson for helpful discussion with the synthesis. Authors would also like to thank Techulon Inc. for generous donation of Glycofect.

■

REFERENCES

(1) Sizovs, A.; McLendon, P. M.; Srinivasachari, S.; Reineke, T. M. Top. Curr. Chem. 2010, 296, 131−190. (2) Mintzer, M. A.; Simanek, E. E. Chem. Rev. 2009, 109, 259−302. (3) Davis, M. E.; Chen, Z.; Shin, D. M. Nat. Rev. Drug Discovery 2008, 7, 771−782. (4) Iyer, A. K.; Khaled, G.; Fang, J.; Maeda, H. Drug Discovery Today 2006, 11, 812−818. (5) Maeda, H.; Greish, K.; Fang, J. Adv. Polym. Sci. 2006, 193, 103− 121. (6) Kelkar, S. S.; Reineke, T. M. Bioconjugate Chem. 2011, 22, 1879− 1903. (7) Lammers, T.; Aime, S.; Hennink, W. E.; Storm, G.; Kiessling, F. Acc. Chem. Res. 2011, 44, 1029−1038. (8) Perry, J. L.; Herlihy, K. P.; Napier, M. E.; DeSimone, J. M. Acc. Chem. Res. 2011, 44, 990−998. (9) Kunjachan, S.; Jayapaul, J.; Mertens, M. E.; Storm, G.; Kiessling, F.; Lammers, T. Curr. Pharm. Biotechnol. 2012, 13, 609−622. (10) Krasia-Christoforou, T.; Georgiou, T. K. J. Mater. Chem. B 2013, 1, 3002−3025. (11) Wang, Z. N.; Gang, C.; Xiaoyuan. Pharm. Res. 2013, 1−19. (12) Ho, Y.-P.; Chen, H. H.; Leong, K. W.; Wang, T.-H. J. Controlled Release 2006, 116, 83−89. (13) Shrestha, R.; Elsabahy, M.; Luehmann, H.; Samarajeewa, S.; Florez-Malaver, S.; Lee, N. S.; Welch, M. J.; Liu, Y.; Wooley, K. L. J. Am. Chem. Soc. 2012, 134, 17362−17365. (14) Cutrin, J. C.; Crich, S. G.; Burghelea, D.; Dastru, W.; Aime, S. Mol. Pharm. 2013, 10, 2079−2085. (15) Chen, H. H.; Ho, Y.-P.; Jiang, X.; Mao, H.-Q.; Wang, T.-H.; Leong, K. W. Mol. Ther. 2008, 16, 324−332. (16) Cheng, S.-H.; Chen, N.-T.; Wu, C.-Y.; Chung, C.-Y.; Hwu, Y.; Mou, C.-Y.; Yang, C.-S.; Lo, L.-W. J. Chin. Chem. Soc. 2011, 58, 798− 804.

■

CONCLUSION In conclusion, lanthanide resonance energy transfer is a promising tool to aid understanding of therapeutic packaging, delivery, and release in vitro and in vivo. Lanthanide labeling techniques offer many inherent benefits over conventional organic dyes such as longer luminescence lifetimes, elimination of background fluorescence with time-gated measurements, and high signal-to-noise ratios. The current study was our first 1623

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624

Biomacromolecules

Article

(17) Redy, O.; Shabat, D. J. Controlled Release 2012, 164, 276−282. (18) Selvin, P. R. Nat. Struct. Biol. 2000, 7, 730−734. (19) Periasamy, A. J. Biomed. Opt. 2001, 6, 287−291. (20) Kokko, T. Lanthanide Chelates as Donors in Fluorescence Resonance Energy Transfer: Exciting Prospects for Bioaffinity Assay Detection. Ph.D. Dissertation, University of Turku, 2009. (21) Chen, K.-J.; Chiu, Y.-L.; Chen, Y.-M.; Ho, Y.-C.; Sung, H.-W. Biomaterials 2011, 32, 2586−2592. (22) Chen, H.; Kim, S.; Li, L.; Wang, S.; Park, K.; Cheng, J.-X. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 6596−6601. (23) Xiao, M.; Selvin, P. R. J. Am. Chem. Soc. 2001, 123, 7067−7073. (24) Johansson, M. K.; Cook, R. M.; Xu, J. D.; Raymond, K. N. J. Am. Chem. Soc. 2004, 126, 16451−16455. (25) Bryson, J. M.; Fichter, K. M.; Chu, W.-J.; Lee, J.-H.; Li, J.; Madsen, L. A.; McLendon, P. M.; Reineke, T. M. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 16913−16918. (26) Getz, E. B.; Cooke, R.; Selvin, P. R. Biophys. J. 1998, 74, 2451− 2458. (27) Heyduk, T. Methods 2001, 25, 44−53. (28) Sculimbrene, B. R.; Imperiali, B. J. Am. Chem. Soc. 2006, 128, 7346−7352. (29) Srinivasachari, S.; Liu, Y.; Zhang, G.; Prevette, L.; Reineke, T. M. J. Am. Chem. Soc. 2006, 128, 8176−8184. (30) Kizjakina, K.; Bryson, J. M.; Grandinetti, G.; Reineke, T. M. Biomaterials 2012, 33, 1851−1862. (31) Carpino, L. A. Acc. Chem. Res. 1973, 6, 191−198. (32) Supkowski, R. M.; Horrocks, W. D., Jr. Inorg. Chim. Acta 2002, 340, 44−48. (33) Thibon, A.; Pierre, V. C. Anal. Bioanal. Chem. 2009, 394, 107− 120. (34) Hemmila, I.; Laitala, V. J. Fluoresc. 2005, 15, 529−542. (35) Horrocks, W. D., Jr.; Sudnick, D. R. J. Am. Chem. Soc. 1979, 101, 334−340. (36) Selvin, P. R.; Hearst, J. E. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 10024−10028. (37) Selvin, P. R.; Rana, T. M.; Hearst, J. E. J. Am. Chem. Soc. 1994, 116, 6029−6030. (38) Liu, Y.; Reineke, T. M. Bioconjugate Chem. 2007, 18, 19−30. (39) Liu, Y.; Reineke, T. M. J. Am. Chem. Soc. 2005, 127, 3004−15. (40) Chen, H. H.; Ho, Y.-P.; Jiang, X.; Mao, H.-Q.; Wang, T.-H.; Leong, K. W. Nano Today 2009, 4, 125−134. (41) Chen, A. A.; Derfus, A. M.; Khetani, S. R.; Bhatia, S. N. Nucleic Acids Res. 2005, 33, e190. (42) Tan, W. B.; Jiang, S.; Zhang, Y. Biomaterials 2007, 28, 1565−1571. (43) Bagalkot, V.; Zhang, L.; Levy-Nissenbaum, E.; Jon, S.; Kantoff, P. W.; Langer, R.; Farokhzad, O. C. Nano Lett. 2007, 7, 3065−3070. (44) Cheng, S.-H.; Chen, N.-T.; Wu, C.-Y.; Chung, C.-Y.; Hwu, Y.; Mou, C.-Y.; Yang, C.-S.; Lo, L.-W. J. Chin. Chem. Soc. 2011, 58, 798− 804. (45) Bergendahl, V.; Heyduk, T.; Burgess, R. R. Appl. Environ. Microbiol. 2003, 69, 1492−1498. (46) Wang, X.; Kelkar, S. S.; Hudson, A. G.; Moore, R. B.; Reineke, T. M.; Madsen, L. A. ACS Macro Lett. 2013, 2, 1038−1041.

1624

dx.doi.org/10.1021/bm401870z | Biomacromolecules 2014, 15, 1612−1624