Time-Resolved Fluorescence Spectroscopy Reveals Fine Structure

Article pubs.acs.org/JPCB

Time-Resolved Fluorescence Spectroscopy Reveals Fine Structure and Dynamics of Poly(L‑lysine) and Polyethylenimine Based DNA Polyplexes Ekaterina S. Lisitsyna,*,†,‡ Tiia-Maaria Ketola,‡,⊥ Emmanuelle Morin-Picardat,†,§,# Huamin Liang,‡,○ Martina Hanzlíková,† Arto Urtti,†,§ Marjo Yliperttula,†,∥ and Elina Vuorimaa-Laukkanen‡ †

Division of Pharmaceutical Biosciences, Centre for Drug Research, Faculty of Pharmacy, University of Helsinki, P.O. Box 56, FI-00014 Helsinki, Finland ‡ Department of Chemistry and Bioengineering, Tampere University of Technology, P.O. Box 541, FI-33101 Tampere, Finland § School of Pharmacy, Faculty of Health Sciences, University of Eastern Finland, P.O. Box 1627, FI-70211 Kuopio, Finland ∥ Department of Pharmaceutical and Pharmacological Sciences, University of Padova, via F. Marzolo, 5, 35131 Padova, Italy S Supporting Information *

ABSTRACT: Structural dynamics of the polyethylenimine− DNA and poly(L-lysine)−DNA complexes (polyplexes) was studied by steady-state and time-resolved fluorescence spectroscopy using the fluorescence resonance energy transfer (FRET) technique. During the formation of the DNA polyplexes, the negative phosphate groups (P) of DNA are bound by the positive amine groups (N) of the polymer. At N/P ratio 2, nearly all of the DNA’s P groups are bound by the polymer N groups: these complexes form the core of the polyplexes. The excess polymer, added to this system to increase the N/P ratio to the values giving efficient gene delivery, forms a positively charged shell around the core polyplex. We investigated whether the exchange between the core and shell regions of PEI and PLL polyplexes takes place. Our results demonstrated a clear difference between the two studied polymers. Shell PEI can replace PEIs previously attached to DNA in the polyplex core, while PLL cannot. Such a dynamic structure of PEI polyplexes compared to a more static one found for PLL polyplexes partially explains the observed difference in the DNA transfection efficiency of these polyplexes. Moreover, the time-resolved fluorescence spectroscopy revealed additional details on the structure of PLL polyplexes: in between the core and shell, there is an intermediate layer where both core and shell PLLs or their parts overlap.

■

different steps, e.g., DNA escape from the endosomes,5 buffering of pH within the endosomes,6 endosomal swelling and lysis resulting from the proton sponge effect,6 and interactions of the polyplexes with the glycosaminoglycans on the cell surface.7,8 Therefore, structural studies of PEI-DNA complexes are expected to yield new information useful for rational design of gene therapy vectors. PEI as well as PLL can condense DNA into DNA−polymer polyplexes stabilized by positive charges.1 The complete DNA condensation occurs at approximately N/P ratio 2 when the polyplex core is formed. At this point, nearly all DNA phosphates are bound by polymer amines, leading to charge neutralization, and the resulting core polyplexes have a strong tendency to form large aggregates.9,10 Further addition of

INTRODUCTION

Polyethylenimine (PEI) was shown to be an efficient vehicle for DNA delivery into the cells by means of forming complexes with DNA (polyplexes).1 PEI polyplexes provide efficient gene transfer at high ratios of PEI amino groups to DNA phosphates (N/P ratios), whereas at lower N/P ratios PEI shows low efficacy of DNA delivery.2 Compared to many other polycations, such as poly-L-lysine (PLL), PEI shows high levels of DNA transfection,2 and the higher efficacy of gene transfer is attributed to the buffering capacity of PEI within the endosomes.1 Indeed, PLL contains only primary amine groups that are fully charged under physiological conditions, whereas branched PEI has primary, secondary, and tertiary amine groups about only a half of which are protonated.3 Thus, the rest of PEI amines can be further protonated during the acidification in endosomes and facilitate endosomal release by the proton sponge effect.3,4 It has been previously shown that the structural properties of PEI affect the delivery process at © XXXX American Chemical Society

Received: August 22, 2017 Revised: November 7, 2017 Published: November 8, 2017 A

DOI: 10.1021/acs.jpcb.7b08394 J. Phys. Chem. B XXXX, XXX, XXX−XXX

Article

The Journal of Physical Chemistry B

the backbone amines are part of the peptide bonds. Thus, only the primary amines bind DNA and only they were taken into account when calculating the N/P ratios (the molar ratio of potentially ionizable polymer nitrogen to DNA phosphate). For PEI, the nitrogens of all, primary, secondary, and tertiary, amines were taken into account in the calculation of N/P ratios. DNA, PLL, and PEI Labeling. PEI, PLL, and DNA were covalently labeled with fluorescent probes in order to visualize the formation of the polyplexes at different N/P ratios by fluorescence changes. DNA was labeled with FITC (F) using a Label IT Tracker Intracellular Nucleic Acid Localization Kit (Mirus Bio Corporation, USA) following the manufacturer’s protocol. The labeling density was 1.3 mol % of bases for FDNA. Both PEI and PLL were covalently labeled with Cy3 (C) in the form of Cy3 NHS ester (GE Healthcare) with labeling densities of 0.2 and 0.6 mol % amines for C-PLL and C-PEI, respectively. Polyplex Preparation. All solutions were prepared in a buffer containing 50 mM MES, 50 mM HEPES, and 75 mM NaCl (adjusted to pH 7.4 using 5 M NaOH). The final nucleotide concentration of DNA was 30 μM. The polyplexes were prepared stepwise: independently on the final N/P ratio between the cationic polymer and DNA, the initial solution with N/P = 0.6 was prepared by vigorous mixing of equal volumes of DNA and cationic polymer solutions. The fluorescence spectrum of the initial solution was monitored to follow the complexation. After the measurement, the next N/P ratio was reached by adding an appropriate amount of polymer solution. The measured N/P values ranged from 0.6 to 6. The samples are denoted as “DNA:core polymer:shell polymer” with N/P ratios as subscripts. For example, the sample with FITC-labeled DNA, unlabeled PLL until N/P = 2, and Cy3-labeled PLL until the total N/P = 4 is described as FDNA:PLL2:C-PLL2. Fluorescence Measurements. The design of the fluorescence measurements is shown in Scheme 1, and Table 1 shows the materials used in those experiments. In the experimental route B, we utilized FRET with FITClabeled DNA as the donor and Cy3-labeled polymers as acceptors to reveal the core−shell dynamics of the studied nanoparticles. Steady-state fluorescence and excitation spectra

polymer to the neutral core results in the formation of a shell of extra PEI around the core and polyplex stabilization. Aggregation of these stable, positively charged polyplexes is significantly reduced.11,12 The core−shell structure of DNA− polycation complexes was previously described by Kabanov et al.13,14 Later, it was demonstrated that interaction of DNA and PEI is a two-step process that is in accordance with the core− shell structure of PEI polyplexes.12 Apparently, the core−shell structure plays a key role in the gene delivery process, since many polyplexes deliver DNA only at high N/P ratios. The structure of DNA−PEI nanoparticles has been studied by fluorescent methods using ethidium bromide as an external chromophore for visualization.2,15 However, the methodology has a limitation due to the total displacement of ethidium bromide from the complexes by polymer at a N/P ratio of about 2.12,15−17 Thus, there remains a need for clarification of the polyplex structure and processes occurring at N/P ratios over 2 where the highest gene transfer activity is demonstrated. Here, we present a work that discloses the DNA−polymer interactions at high polymer concentrations by means of tracking the behavior of covalently labeled DNA and polymer using the fluorescence resonance energy transfer (FRET) technique. FRET is a powerful spectroscopic tool to study spatiotemporal dynamics of molecular interactions.18,19 It is a distancedependent process by which energy is transferred nonradiatively from the donor fluorophore in an excited state to the acceptor fluorophore by intermolecular long-range dipole− dipole interaction.20−22 There are several requirements for efficient energy transfer occurring: (1) the fluorescence spectrum of the donor and the absorption spectrum of the acceptor must have a good overlap; (2) donor and acceptor molecules must be in close proximity (typically 1−10 nm); (3) the fluorescence quantum yield of the donor as well as the absorption coefficient of the acceptor must be sufficiently high. The fluorophore pair for the present study was selected according to these requirements to visualize the interactions between target DNA plasmid and polymer molecules (PEI, PLL) by steady-state and time-resolved fluorescence spectroscopy.12,15,16,23 The hypothesis that motivated the present study is the possibility of migration or exchange of PEI molecules in the polyplex composition. We suppose that the PEI movements between the core and shell of the polyplex could explain the excess polymer effect of facilitating the gene delivery into the cells.

Scheme 1. Core Dynamics Studiesa

■

MATERIALS AND METHODS Materials. The plasmid DNA (pCLuc4) encoding luciferase as a reporter gene with cytomegalovirus (CMV) viral promoter was grown in E. coli and was isolated and purified using a Qiagen Plasmid Giga kit (Qiagen, Hilden, Germany) according to the manufacturer’s protocol. The identity and purity of the plasmid were confirmed by agarose gel electrophoresis using restriction enzymes and by measuring the absorbance at 260 and 280 nm (A260/A280 > 1.8), respectively. Branched polyethylenimine, PEI (Sigma-Aldrich, USA), with an average molecular weight of 25 kDa and end primary, secondary, and tertiary amines in a ratio of 1:1.2:0.76 was purchased from Sigma-Aldrich and used as a 1 mg/mL aqueous solution adjusted to pH 7. Poly(L-lysine), PLL, of an average molecular weight of 200 kDa and containing only primary amines as end groups (Sigma-Aldrich, USA) was used in this study. For PLL,

a

In parts A and B, the polyplex core was prepared from FITC-DNA and unlabeled polymer and the shell either from (A) unlabeled polymer or (B) Cy3-labeled polymer. In part C, the polyplex core was prepared from Cy3-polymer and either FITC-DNA or DNA and the shell from the same Cy3-polymer.

B


Article


empirical parameter showing the heterogeneity of the system. The parameter β lies in the range 0 < β ≤ 1, where β = 1 leads back to purely exponential dependence without any distribution of the states. Thus, the smaller β values correspond to a broader distribution and higher heterogeneity of the system and vice versa.24,25 The amplitudes (αi(λ)) were corrected depending on the sensitivity of the detector at different wavelengths. Decayassociated spectra (DAS) are represented by the amplitudes αi(λ) plotted against wavelength. In the case of a mixture of several noninteracting emitting dyes, the DAS provide a means of separating the individual spectra of the species on the base of fluorescence lifetime. Energy transfer efficiency was estimated as

Table 1. Components of the Polyplexes: DNA Labeled with FITC (F-DNA), Unlabeled DNA (DNA), Unlabeled Polymers (PLL and PEI), and Polymers Labeled with Cy3 (C-PLL and C-PEI)a PLL

PEI

scheme

DNA

core, N/P = 2

shell, N/P = 6

core, N/P = 2

shell, N/P = 6

A C C B

F-DNA DNA F-DNA F-DNA

PLL C-PLL C-PLL PLL

PLL C-PLL C-PLL C-PLL

PEI C-PEI C-PEI PEI

PEI C-PEI C-PEI C-PEI

a

The characters A−C refer to Scheme 1.

ΦET ∼ I 0 /I ∼ τ0/τ

were recorded with a Fluorolog Yobin Yvon-SPEX fluorometer. The excitation wavelength was 483 nm, and the spectra were automatically corrected using a correction function provided by the manufacturer. The time-resolved fluorescence was measured using a timecorrelated single photon counting (TCSPC) system (PicoQuant GmBH) consisting of a PicoHarp 300 controller and a PDL 800-B driver. The samples were excited with the pulsed diode laser head LDH-P-C-485 at 483 nm at a time resolution of 130 ps. The signals were detected with a microchannel plate photomultiplier tube (Hamamatsu R2809U). The influence of the scattered excitation light was reduced with a cutoff filter (transmission >490 nm) in front of the monitoring monochromator. Fluorescence decays were collected with a constant accumulation time in the 500−630 nm wavelength range with steps of 10 nm. The instrumental response function (IRF) was measured separately, and the decays were simultaneously deconvoluted and fitted by applying the iterative least-squares method either to the sum of exponents (eq 1) or to the sum of exponential and Kohlrausch stretched exponential functions (eq 2) I (t , λ ) =

∑ ai(λ)e−t/τ

where I and I0 are fluorescence intensities at donor (FITC) wavelength 520 nm, τ and τ0 are decay times of the donor (FITC) in the presence and absence of the acceptor (Cy3), respectively.26 For F-DNA:C-PEI and F-DNA:PEI:C-PEI systems, the donor (FITC) lifetime was clearly resolved by the exponential component when fitting with eq 2 and subsequently used for determination of energy transfer efficiencies. According to the data analysis with eq 1, F-DNA and C-PLL fluorescence lifetimes are too close to each other to be well separated by the fitting program (Table 2). Thus, for FDNA:C-PLL and F-DNA:PLL:C-PLL systems, the mean amplitude weighted lifetime ⟨τ⟩ was obtained from twoexponential fitting of single decay curves at monitoring wavelength 520 nm for each N/P ratio as follows ⟨τ ⟩ =

(4)

■

(1)

I(t , λ) = ai(λ)e−(t / τ1) + aj(λ)e−(t / τ2)

∑ aiτi ∑ ai

and used for the calculation of energy transfer efficiencies. Since Cy3 dye does not fluoresce at 520 nm, it was supposed that both calculated lifetimes belong to FITC.

i

i

(3)

RESULTS Before discussing the results obtained from the FRET studies, it is important to know the behavior of the labeled species in solution and in polyplexes in the absence of the other FRET partner.

β

(2)

where τi is the global lifetime, αi(λ) is the local amplitude (preexponential factor) at a specific wavelength, and β is the

Table 2. Maximum Wavelengths of the Absorption, Excitation, and Fluorescence Spectra and the Fluorescence Lifetimes for the Labeled Species in the Absence and Presence of Unlabeled Polymers or DNA sample

λmax,ABS (nm)

λmax,EX (nm)

λmax,FL (nm)

fitting modela

τ (ns)

F-DNA F-DNA:PLL, N/P = 6

495 499

517 522

exp (eq 1) exp (eq 1)

F-DNA:PEI, N/P = 6

498

527

exp (eq 1)

560, 525 I525/I560 = 0.52 559, 524 I524/I559 = 0.52 544, 512 I515/I545 = 0.70 546, 514 A514/A546 = 0.81

572, 620

exp (eq 1)

3.81 ± 0.03 3.50 ± 0.04 (87%) 0.109 (13%)b 2.25 ± 0.07 (44%) 0.33 ± 0.07 (56%) 2.67 ± 0.03

572, 619

exp (eq 1)

2.67 ± 0.02

565

exp + st exp (eq 2)

578

exp + st exp (eq 2)

0.28 ± 0.03 (7%) 0.033 (93%)b 0.54 ± 0.03 (13%) 0.103 (87%)b

C-PLL DNA:C-PLL, N/P = 6 C-PEI DNA:C-PEI, N/P = 6

560, 525 A525/A560 562, 526 A526/A562 545, 510 A510/A545 549, 514 A510/A545

= 0.55 = 0.57 = 0.97 = 0.92

β

0.519 ± 0.007 1

a

exp: mono- or biexponential model (eq 1). exp + st exp: mixed model containing exponential and stretched exponential parts (eq 2). bIt is not possible to determine the accuracy for lifetimes ≤0.1 ns, the time resolution of the system. C


Article


Figure 1. (a) Fluorescence and excitation spectra for F-DNA and F-DNA:PLL at N/P = 6 and F-DNA:PEI at N/P = 4. (b) Fluorescence decay curves for F-DNA and F-DNA:PLL (N/P = 6) and F-DNA:PEI (N/P = 6) monitored at 520 nm. In the fluorescence spectra and decay curves, the excitation wavelength was 483 nm and the excitation spectra were monitored at 535 nm. In all samples, the F-DNA concentration was constant corresponding to the 30 μM total DNA nucleotide concentration.

Figure 2. Normalized fluorescence and excitation spectra of (a) C-PLL and (b) C-PEI at the concentration corresponding to N/P ratio 6. Insets: Absorption spectra of the same (a) C-PLL and (b) C-PEI solutions. (c) Fluorescence decay curves for C-PLL and C-PEI at the concentration corresponding to N/P ratio 6 monitored at 570 and 560 nm, respectively. In the fluorescence spectra and decay curves, the excitation wavelength was 483 nm and the excitation spectra were monitored at 610 nm.

DNA Labeled with FITC (F-DNA). The fluorescence and excitation spectra for F-DNA in the presence and absence of unlabeled PLL and PEI (Scheme 1, route A) are shown in Figure 1a. In the presence of PLL and PEI, both spectra shift to red (Table 2) and the fluorescence efficiency decreases considerably up to N/P = 1.5 for PLL and N/P = 3 for PEI (Figure S1a). In the presence of both PLL and PEI, FITC fluorescence decays became two-exponential at N/P = 1.5 (Figure 1b). For PLL, the lifetime of the longer living component decreased from 3.81 ns (no PLL) to 3.50 ns at N/P ≥ 1.5. In the case of PEI, the lifetime of the longer living component

decreased from 3.81 to 2.30 ns at N/P ≥ 3. The lifetimes of the shorter living components remained nearly constant at 0.1 ns for PLL and 0.33 ns for PEI (Figure S1b). The full width at half of IRF maximum is 130 ps for the exploited TCSPC system. Thus, 0.1 ns should be considered as the upper limit for all of the lifetimes ≤0.1 ns obtained for the shorter component of all the studied systems. During the formation of DNA polyplexes, the positive amine groups (N) of the cationic polymers bind the negative phosphate groups (P) of DNA. According to previous studies, the P of DNA are completely bound by the N of the polymer at N/P ratios of 1.5 and 2.5 for PLL and PEI, respectively.2,3 The fluorescence quenching is due to the D


Article


Figure 3. Normalized fluorescence spectra for (a) C-PLL in the presence of unlabeled DNA at N/P = 0.6, 2, and 6 and (b) C-PEI in the presence of unlabeled DNA at N/P = 2 and 6. Normalized fluorescence decay curves for (c) C-PLL in the presence of unlabeled DNA at N/P = 0.6 and 6 and (d) C-PEI in the presence of unlabeled DNA at N/P = 2 and 6 monitored at 560 nm. The excitation wavelength for both fluorescence spectra and decay kinetics was 483 nm.

quenched fluorescence. The fluorescence decay curves for CPLL are one-exponential with a lifetime of 2.65 ns and the maximum at 572 nm (Figures 2c and S2b). C-PEI fluorescence decay curves are not monoexponential (Figure 2c) and were fitted using eq 2. The exponential component gave a lifetime of 0.30 ns and the stretched component a lifetime of 0.03 ns and parameter β ∼ 0.5 (Table 2). The maximum of the stretched component was at 566 nm, but no clear maximum was observed for the weak longer-living component (Figure S2b). The fluorescence of C-PEI is quenched by the Cy3 aggregates, leading to a distribution of quenched Cy3 states with short lifetimes. Both Cy3 aggregates and quenched Cy3 states can contribute to the stretched component, and the 0.03 ns lifetime represents the average lifetime of the distribution. Since in the present study Cy3 acts as a FRET acceptor, its aggregation should not hamper the further measurements and analysis. In the presence of unlabeled DNA, the fluorescence properties of Cy3 labeled polymers depended on the N/P ratio. At low N/P ratios (N/P < 2), the polyplex core was formed and the maxima of steady-state fluorescence and excitation spectra were red-shifted 5 nm compared to the spectra without DNA (Figures 3 and S3). For C-PLL, upon increasing N/P ratio, the spectra shifted back to blue and at N/P = 2 the spectra reached the same position as in the absence of DNA (Figure 3a, blue line). Thus, with increasing N/P ratio, increasingly fewer Cy3 moieties have direct interaction with DNA. At N/P ratios over 2, i.e., during polyplex shell formation, the spectral maximum stops shifting, since the shell C-PLL has no direct contact with DNA (Figure 3a, red line). The same change was observed in the DAS and

changes in the microenvironment of the FITC dye during the polycation complexation of DNA. Once the core is complete, further addition of polymer does not have any effect on the fluorescence of F-DNA. PEI caused a more dramatic lifetime decrease of F-DNA compared to PLL (Table 2 and Figure 1b). The amine density of PEI (only two carbon atoms between amines) is higher than that in the case of PLL (12 carbons between amines), and thus, its effect on the FITC fluorescence lifetime is stronger than that of PLL.4 PLL and PEI Labeled with Cy3 (C-PLL and C-PEI). Fluorescence and excitation spectra of C-PEI were blue-shifted (7 nm) compared to those of C-PLL, although they had similar shapes (Figure 2). However, the absorption spectra of C-PLL and C-PEI differed from each other significantly (Figure 2, insets). The spectrum of C-PLL corresponded to the spectrum of monomeric Cy3 with an absorption maximum at 560 nm and a vibrational shoulder at 525 nm. However, the spectrum of CPEI was blue-shifted with two maxima at 545 and 510 nm of nearly equal absorbance. Also, the C-PEI absorption was about 4 times higher than the absorption of C-PLL, indicating that the density of Cy3 labels in C-PEI is much higher than that in C-PLL, high enough to lead to formation of Cy3 H-dimers or even higher order aggregates.27 The ratio of the two absorption peaks decreased from 0.97 in the absorption spectrum to 0.70 in the excitation spectra of C-PEI, whereas it stayed nearly constant at about 0.5 for C-PLL (Table 2). At the same time, C-PEI fluorescence efficiency was only 7% of that for C-PLL (Figure S2a). This difference additionally referred to the formation of Cy3 assemblies in C-PEI with substantively E


Article


Figure 4. Normalized fluorescence spectra for (a) the F-DNA:C-PLL system at N/P 0.6 and 6 and (b) the F-DNA:C-PEI system at N/P = 0.6 and 4. Normalized fluorescence decay curves for (c) F-DNA:C-PLL and (d) F-DNA:C-PEI at N/P = 0, 0.6, and 4 (black, blue, and red curves, respectively) monitored at 520 nm. The excitation wavelength for fluorescence spectra and decays measurements was 483 nm.

of steady-state and time-resolved fluorescence spectroscopy in order to reveal the dynamics of the complexes. Namely, polyplexes containing core and shell made of Cy3-labeled polymers at N/P ratios 0 to 6 (Scheme 1, route C) and those prepared with unlabeled polymers in core (until N/P = 2) followed by further additions of Cy3-labeled polymers up to N/ P = 6 (Scheme 1, route B) were used:

the fluorescence lifetime decreased from 2.94 to 2.67 ns (Figures 3c and S4a,c,e). However, the addition of DNA to CPEI caused only a red shift up to N/P = 6 both in corresponding fluorescence spectra (Figure 3b, blue and red lines) and DAS (Figure S4b,f), while the maximum of the excitation spectrum shifted to the blue region and broadened (Figure S3b). These different changes in fluorescence and excitation spectra of DNA:C-PEI compared to those for the DNA:C-PLL system could appear due to the contribution of the fluorescence of Cy3 H-dimers or higher order aggregates.27 DNA:C-PEI decay curves were fitted with eq 2, as was previously shown for C-PEI. The lifetime decrease was observed for both components (1.8 to 0.5 ns and 0.14 to 0.016 ns) of C-PEI based polyplexes (Figure S4b,d). The more pronounced lifetime drop with the increasing N/P ratio in the case of C-PEI polyplexes with respect to those of C-PLL (Figure 3c,d) speaks also of the Cy3 self-quenching processes in the system due to the higher labeling density of PEI. It is consistent with the steady-state fluorescence data. Although the system is more complex, the parameter β increased with the addition of DNA to the C-PEI from 0.5 to 1 (Table 2), indicating a lower heterogeneity of the system. This result can be explained by the domination of one of the quenched Cy3 states over the others, leading to a smaller heterogeneity in the presence of DNA. Thus, the fluorescence properties of C-PLL and C-PEI depend on whether it is interacting with DNA or with the polyplex core. Dynamics of Polyplexes. The exchange of polymers in the polyplex core and shell was studied using energy transfer between F-DNA as the donor and C-PLL or C-PEI as the acceptor. Two different polyplex designs were studied by means

+C ‐ P

(5)

F‐DNA ⎯⎯⎯⎯⎯→ F‐DNA:C‐P +P

+C ‐ P

F‐DNA ⎯→ ⎯ F‐DNA:P2 ⎯⎯⎯⎯⎯→ F‐DNA:P2:C‐P1 − 4

(6)

Polyplexes with Core and Shell Made of Cy3 Labeled Polymers (Scheme 1, Route C). For both C-PLL and C-PEI polyplexes (N/P = 0.6), the presence of F-DNA was seen as the fluorescence band at 520 nm (Figure 4a,b). The presence of CPLL was observed as the fluorescence bands at 578 and 625 nm as well as the presence of C-PEI as the band at 568 nm. At N/P = 6, most of the F-DNA fluorescence was quenched and the fluorescence of C-PLL and C-PEI dominated the spectra (Figure 4a,b). Corresponding excitation spectra are shown in Figure S5a,b. The quenching was stronger than in the absence of Cy3 labels, because energy transfer from F-DNA to C-PLL took place (Figure S5c,d). Thus, the FRET technique is feasible to acquire the information about the polyplex formation. In the presence of Cy3 labeled polymers, the quenching was seen at N/P ≤ 1, whereas, in the presence of unlabeled ones, the quenching continued until N/P = 1.5 and 3.0 for PLL and PEI, respectively. Again, the excitation and fluorescence bands of CPLL shifted 5 nm toward the blue region at increasing N/P ratios (Figures 4a and S5a). However, fluorescence bands of FDNA:C-PEI shifted to red wavelengths similar to C-PEI F


Article


Figure 5. Normalized fluorescence and excitation spectra for F-DNA:P2 (black lines), F-DNA:P2:C-P1 (red lines), and F-DNA:P2:C-P4 (blue lines), where P is either PLL (a, b) or PEI (c, d). The excitation spectra were monitored at 610 nm, and the excitation wavelength was 483 nm.

Figure 6. Normalized fluorescence decay curves for F-DNA:C-P4 (black curves) and F-DNA:P2:C-P4 (red curves) where P is either (a) PLL or (b) PEI monitored at 520 nm. The excitation wavelength was 483 nm.

the component corresponds to F-DNA (donor) fluorescence and its lifetime decreased (3.77 to 3.13 ns) with the addition of C-PEI (acceptor) to the system. The short-living component (≤0.1 ns) is observed at 550−600 nm indicating that it corresponds to the fluorescence of the quenched Cy3 states distribution due to Cy3 aggregation (Table S1, Figure S6). The parameter β increased from 0.39 to 0.54 with the increase of CPEI concentration, showing the lower heterogeneity of the system while it became more complex. Domination of one of the quenched Cy3 states over the others explains this seeming decrease in the heterogeneity of the system. Polyplexes with Unlabeled Polymers in Core and Cy3 Labeled Polymers in Shell (Scheme 1, Route B). A clear difference between PLL and PEI polyplexes was observed when comparinng their excitation and emission spectra with each other at corresponding N/P ratios (Figure 5).

polyplexes with unlabeled DNA (Figure 4b), while the excitation maximum was blue-shifted as well as for F-DNA:CPLL (Figure S5a,b). Addition of Cy3 labeled polymers to F-DNA also affected fluorescence decay kinetics (Figure 4c,d). For F-DNA:C-PLL, the decay curves became two-exponential upon C-PLL (acceptor) addition at all N/P ratios (Figure 4c, blue and red curves, Table S1). The curves monitored at donor fluorescence wavelength 520 nm were used for the analysis in order to exclude the acceptor lifetime contribution. Thus, all of the lifetime components obtained correspond to F-DNA (the donor) fluorescence. For F-DNA:C-PEI, the decay curves were fitted globally at all wavelengths using the stretch exponential function in order to take into account the distribution of different Cy3 states as mentioned above. The long-living component has maximum amplitude at 520 nm, indicating that G


Article


Figure 7. Fluorescence intensity ratio versus N/P ratio plots for F-DNA:P0.6−6 (black squares), F-DNA:C-P0.6−6 (red circles), and F-DNA:P2:C-P1−4 (blue diamonds), where P is either (a) PLL or (b) PEI. The FITC fluorescence intensity at 520 nm was used to calculate the intensity ratio I0/I. Fluorescence lifetime (for donor F-DNA) ratio versus N/P ratio plots for F-DNA:C-P0.6−6 (blues squares) and F-DNA:P2:C-P1−4 (red squares), where P is either (c) PLL or (d) PEI.

though it was added after unlabeled PEI, i.e., confirming the exchange between polymers in the core and shell parts of the polyplex structure. Comparative data of steady-state and time-resolved fluorescence of all the systems for both polymers is summarized in Figure 7. The fluorescence intensity ratio plotted versus N/P ratio was used to compare the F-DNA:P0.6−6, F-DNA:C-P0.6−6, and F-DNA:P2:C-P1−4 systems in terms of donor (F-DNA) fluorescence quenching (Figure 7a,b). Both unlabeled polymers negligibly affected the donor fluorescence for F-DNA:P0.6−6 (Figure 7a,b, black squares). However, the addition of the acceptor-labeled PLL as well as PEI resulted in a significant quenching of F-DNA fluorescence in F-DNA:C-P0.6−6 systems (Figure 7a,b, red circles). This indicates the presence of FRET between the F-DNA and CPLL or C-PEI in the polyplex core. The difference between the two studied polymers was observed for their mixed polyplexes consisting of unlabeled core and labeled shell, F-DNA:P2:CP1−4. For F-DNA:PLL2:C-PLL1−4, the fluorescence intensity changes were similar to those obtained for F-DNA:PLL0.6−4, suggesting that unlabeled PLL in the polyplex core interferes with direct interaction of C-PLL of the polyplex shell with DNA and no FRET takes place (Figure 7a, blue diamonds and black squares). For the F-DNA:PEI2:C-PEI1−4 system, the curve was close to that for F-DNA:PEI at small N/P ratios ≤2. However, upon further addition of C-PEI (N/P ratio >2), the more pronounced quenching of F-DNA fluorescence was observed and it reached the same maximum values as in the FDNA:C-PEI0.6−6 system at higher N/P ratios up to 6 (Figure 7b, blue diamonds). This clearly demonstrates that C-PEI can

For the F-DNA:PLL2:C-PLL1−4 system, the maximum wavelength of Cy3 remained unchanged at different N/P ratios (Figure 5a,b, red and blue lines). Therefore, one can conclude that C-PLL does not interact directly with DNA and there is no PLL exchange between the polyplex core and shell. Meanwhile, in the case of the PEI system, a direct interaction between C-PEI and F-DNA was observed as a spectral shift at Cy3 maximum wavelength (Figure 5c,d, red and blue lines). This affirms the more dynamic structure of PEI polyplexes wherein the exchange takes place between the core and shell polymers. Of note is that the shift in the fluorescence spectra of PEI polyplexes is less than that in the corresponding excitation spectra and the spectrum of F-DNA:PEI2:C-PEI4 is broadened compared to that of F-DNA:PEI2:C-PEI1. This can be explained by a superposition of two effects, namely, blue spectral shift due to the direct interaction of C-PEI and F-DNA and red shift and spectral broadening mediated by Cy3 aggregation events similar to the DNA:C-PEI system at high N/P ratios. Normalized decay kinetics of the two systems, F-DNA:C-P4 and F-DNA:P2:C-P4, are presented for both polymers in Figure 6. A clear difference between the two polymers is seen. Direct C-PLL addition to F-DNA in the polyplex core results in the shorter living excited state of the donor (⟨τ⟩ = 1.45 ns) than the addition of C-PLL after unlabeled polymer (⟨τ⟩ = 1.97 ns) (Figure 6a). In the case of PEI, the decay kinetics are similar regardless of the presence of unlabeled polymer in the core of the polyplexes and characterized by a donor lifetime of about 3 ns (Figure 6b, Table S1). This leads to the same conclusion that C-PEI is capable of interacting directly with DNA even H


Article


Figure 8. Illustration of PLL and PEI polyplex structures based on steady-state and time-resolved fluorescence data.

according to steady-state fluorescence data. Cy3-Labeled PLL in the shell did not directly interact with the DNA in the nanoparticle core at all N/P ratios tested. Time-resolved fluorescence data further confirmed the more dynamic structure of PEI polyplexes and ability of shell PEIs to replace PEI in the polyplex core. For PLL polyplexes, the analysis of fluorescence lifetime changes allowed for the better understanding of the static polyplex structure. Although the exchange between shell and core PLL polymers was not observed, there was no clear interface between core and shell polymer layers, as was expected according to steady-state fluorescence spectroscopy data alone. The existence of the intermediate layer comprising both core and shell polymers was confirmed by time-resolved fluorescence spectroscopy. The overall conclusion deduced from both steady-state and time-resolved fluorescence data is presented schematically in Figure 8. Thus, the time-resolved fluorescence technique seemed to be able to reveal the fine structure of PLL polyplexes. This shows elevated power and sensitivity of the lifetime measurements in combination with the FRET technique over those related to fluorescence intensities. The current results are consistent with the molecular dynamic simulation study reported by Sun et al.28 stating that extra PEI can replace PEI previously bound to DNA. A clear difference in the behavior of PEI and PLL formed polyplexes is also in agreement with recently published results of Pigeon et al.29 where the possibility of plasmid DNA exchange between linear PEI polyplexes was stated but none was observed for polyplexes made of PLL. The reported difference in the binding constants for polyplex core formation between the two studied polymers is not big enough to explain a dramatic distinction in DNA transfection efficacy for PLL and PEI polyplexes.2,12 The polymer structure effect on the polyplex formation in terms of type, content, and distribution of the amine groups can elucidate the above conclusions. PLL has only primary amines at lysine side chains that are available for DNA binding, while branched PEI contains primary, secondary, and tertiary amines, thus having a higher density of amine groups than PLL. Thus, PLL−DNA binding is almost exclusively mediated by primary amines that are fully charged under physiological conditions, whereas only a part of all amines is protonated in PEI.3 The primary amine groups of PLL seem to bind DNA more tightly, providing the static structure of the polyplex. In turn, the combination of the primary, secondary, and tertiary amine groups of branched PEI allows for a more loose structure of the polyplex, thereby facilitating polymer dynamics between core and shell parts of the complex.

interact directly with F-DNA regardless of the presence of unlabeled PEI molecules in the polyplex core; i.e., C-PEI of the polyplex shell can replace PEI previously bound to DNA in the core part, allowing FRET between C-PEI and F-DNA. The summary of the time-resolved fluorescence study of the systems is presented as plots of fluorescence lifetime ratio versus N/P ratio for PEI and PLL polyplexes. The plots for the F-DNA:C-P0.6−6 and F-DNA:P2:C-P1−4 systems are shown in Figure 7c,d. Virtually the same final fluorescence lifetime changes were observed for the F-DNA:C-PEI0.6−4 and FDNA:PEI2:C-PEI1−4 systems independently of the presence of unlabeled polymer in the polyplex core (Figure 7d, red and blue squares). This additionally confirms the possibility of direct interaction between C-PEI from polyplex shell and FDNA, i.e., core−shell polymer exchange. For F-DNA:PLL2:CPLL1−4, some donor fluorescence lifetime change was detected (Figure 7c, red squares), but it was less pronounced than that obtained for F-DNA:PLL0.6−4 (Figure 7c, blue squares), supporting that C-PLL in the polyplex shell did not have direct interaction with DNA in the core. Therefore, some of the C-PLL molecules in the shell part of the polyplex can still be located close enough to the F-DNA in the polyplex core, thereby providing less efficient energy transfer at a bigger distance compared to the direct interaction between donor labeled DNA and acceptor labeled polymer. An intermediate layer is formed between unlabeled core and labeled shell parts in the polyplex structure consisting of both labeled and unlabeled PLL molecules or their parts. Thus, studying the systems of interest by means of both steady-state and timeresolved fluorescence spectroscopies allows for observing of the polyplex structural dynamics with enhanced detail.

■

DISCUSSION The core−shell dynamic studies were successfully performed using steady-state and time-resolved fluorescence spectroscopy and the corresponding FRET pair of fluorescent labels. In contrast to the TCSPC method using ETI dye,12 the FRETbased method enabled the detection of changes in the polyplex state after the core formation, i.e., at N/P ratios >2. The results further confirmed that the polyplexes can be described as core− shell structures. In general, for PEI and PLL, a very strongly bound and stable nanoparticle core is formed at N/P ≤ 2, and the core is surrounded by a loose, positively charged shell when an extra amount of polymer is added. For PEI polyplexes formed at N/P ratios >2, it was clearly shown that PEI molecules exchanged between the nanoparticle core and the shell according to the excitation and emission spectra changes. However, different behavior was observed for PLL polyplexes I


Article


additional value of the excited state lifetime measurements in combination with the FRET technique for clarification of the polyplex fine structures. The obtained results may inspire the experts to apply the presented approach for studying other systems based on complex formation. Moreover, the current study is an important step on the way to cell studies of the above polymeric gene carriers by means of fluorescence lifetime microscopy and confocal microscopy techniques paying attention to the selection of suitable fluorescent labels. The approach applied in the current work enables the study of structural dynamics of various spontaneously assembling nanoparticles in both solutions and cell-based systems that will shed light on the functionality of the particles in gene and drug delivery.

Another possible reason for the different polyplex dynamics revealed in the study can be the molecular weights of the polymers used. The selection of the polymers with different molecular weights was mediated by their efficiency in DNA complexing and gene transfer. Indeed, it has been previously reported that PLL of the small weight 20 kDa cannot condense DNA at the same N/P ratios as PLL (200 kDa) and PEI (25 kDa).2 Moreover, PLL of molecular weight 24 kDa shows an even lower transfection efficiency than the one with a higher molecular weight of 224 kDa.30 A comparative study of branched PEI (25 kDa), linear PEI (22 kDa), and short PEI (700 Da) polyplexes for their binding with DNA also clearly showed the role of polymer molecular weight in the complex formation. The above-mentioned reports suggested that the difference in molecular weights between PEI and PLL could also partially explain the findings of the study. We hypothesize that the binding of smaller PEI polymer to DNA is essentially reversible and results in the formation of thermodynamically stable polyplexes. However, larger PLL polymer binds DNA almost irreversibly, leading to DNA release blockage due to the inability of the polymer rearrangement. This in turn leads to the formation of polyplexes, which are kinetically trapped and thermodynamically unstable. Although tightly bound PLL polyplexes are locked in a thermodynamically unstable configuration, they still strive to reach the more preferable stable configuration, providing some release of the content and existence of negligible transfection efficiency. Thus, the structure of PLL polyplexes is more static compared to PEI polyplexes but could not be considered as a strictly static one. In other words, the bulkier PLL chain additionally limits the movements and the exchange of the polymer molecules in the polyplexes compared to the smaller PEIs. However, more studies are needed to confirm or rule out the hypothesis of molecular weight contribution to the dynamic/static nature of the polyplex structure. The dynamics between the core and shell observed in the present study should be an intrinsic feature of polyplexes, and it can partially explain the difference in the gene transfer efficiencies. Thus, the poor transfection performance of PLL is also related to the difference observed in structural dynamics between PLL and PEI. The rigid core−shell structure of PLL polyplexes can considerably hinder the release of DNA from the carriers and lower the gene transcription. On the other hand, the mobility of PEI molecules between the core and shell of PEI polyplexes makes these polyplexes more dynamic, and hence sensitive to dissociation and DNA release at the cellular level. The polyplexes and lipoplexes with secondary and tertiary amines also show increased DNA release from the nanoparticle carrier compared to those containing only primary amines.5,8 Furthermore, the secondary and tertiary amines of PEI are capable of buffering the pH changes in the endosomes,6,31−33 thereby contributing to gene transfer efficiency. This is also related to the PEI exchange between the core and shell, since PEI can accept and release not only DNA molecules but also hydrogen ions.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.jpcb.7b08394. Fluorescence and excitation spectra and fluorescence intensities at different N/P ratios for different nanoparticle systems, fluorescence lifetimes at different N/P ratios, and DAS (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected], ekaterina.lisitsyna@tut.fi. Phone: +358469521118. ORCID

Ekaterina S. Lisitsyna: 0000-0003-4228-1157 Arto Urtti: 0000-0001-6064-3102 Elina Vuorimaa-Laukkanen: 0000-0002-3610-785X Present Addresses ⊥

T.-M.K.: No current affiliation. E.M.-P.: National Institute of Agricultural Research (INRA), Rue de la Géraudière BP 71627, 44 316 Nantes Cedex 3, France. ○ H.L.: Building 2-413, Hefei Institutes of Physical Science, CAS, Shushanhu Road 350, 230031 Hefei, Anhui, P.R. China. #

Author Contributions

All authors have given approval to the final version of the manuscript Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The study was supported by Fellowships programme of the Centre for International Mobility (CIMO Winter School, Finland) and Academy of Finland (Project Nos. 259990, 283721, and 140357).

■

ABBREVIATIONS PEI, polyethylenimine; PLL, poly(L-lysine); FRET, fluorescence resonance energy transfer; N/P, ratios of amino groups of a polymer to the phosphates of DNA; FITC, fluorescein isothiocyanate; Cy3, indocarbocyanine 3; F-DNA, FITClabeled DNA; C-P, Cy3-labeled polymer; MES, 2-(Nmorpholino)ethanesulfonic acid; HEPES, 4-(2-hydroxyethyl)1-piperazineethanesulfonic acid; DAS, decay associated spectrum; IRF, instrumental response function

■

CONCLUSIONS In the present study, we have proved the working hypothesis and presented the evidence that PEI molecules do undergo exchange between the core and shell of the polyplexes. The clarified fact that PEI polyplex is a much more dynamic system than PLL polyplex could explain its favorable DNA transfection via several mechanisms. Furthermore, our data demonstrates an J


Article


■

(19) Joo, C.; Balci, H.; Ishitsuka, Y.; Buranachai, C.; Ha, T. Advances in Single-Molecule Fluorescence Methods for Molecular Biology. Annu. Rev. Biochem. 2008, 77, 51−76. (20) Förster, T. Energiewanderung und Fluoreszenz. Naturwissenschaften 1946, 33, 166−175. (21) Clegg, R. M. Fluorescence Resonance Energy Transfer. Curr. Opin. Biotechnol. 1995, 6, 103−110. (22) Breunig, M.; Lungwitz, U.; Liebl, R.; Goepferich, A. Fluorescence Resonance Energy Transfer: Evaluation of the Intracellular Stability of Polyplexes. Eur. J. Pharm. Biopharm. 2006, 63, 156−165. (23) Vuorimaa-Laukkanen, E.; Lisitsyna, E. S.; Ketola, T.-M.; MorinPickardat, E.; Liang, H.; Hanzlíková, M.; Yliperttula, M. Difference in the Core-Shell Dynamics of Polyethyleneimine and Poly(L-lysine) DNA Polyplexes. Eur. J. Pharm. Sci. 2017, 103, 122−127. (24) Berberan-Santos, M. N.; Bodunov, E. N.; Valeur, B. Mathematical Functions for the Analysis of Luminescence Decays with Underlying Distributions 1. Kohlrausch Decay Function (Stretched Exponential). Chem. Phys. 2005, 315, 171−182. (25) Lee, J.; Lee, J.; Lee, M.; Lee, K. J. B.; Ko, D. S. Scanning Confocal Fluorescence Microscopy of Single DNA-EtBr Complexes Dispersed in Polymer. Chem. Phys. Lett. 2004, 394, 49−53. (26) Valeur, B. Molecular Fluorescence: Principles and Applications; Wiley-VCH Verlag GmbH: New York, 2002. (27) McPhee, J. T.; Scott, E.; Levinger, N. E.; Van Orden, A. Cy3 in AOT Reverse Micelles I. Dimer Formation Revealed through SteadyState and Time-Resolved Spectroscopy. J. Phys. Chem. B 2011, 115, 9576−9584. (28) Sun, C.; Tang, T.; Uludağ, H. Molecular Dynamics Simulations of PEI Mediated DNA Aggregation. Biomacromolecules 2011, 12, 3698−3707. (29) Pigeon, L.; Gonçalves, C.; Pichon, C.; Midoux, P. Evidence for Plasmid DNA Exchange after Polyplex Mixing. Soft Matter 2016, 12, 7012−7019. (30) Wolfert, M. A.; Dash, Ph. R.; Nazarova, O.; Oupicky, D.; Seymour, L. W.; Smart, S.; Strohalm, J.; Ulbrich, K. Polyelectrolyte Vectors for Gene Delivery: Influence of Cationic Polymer on Biophysical Properties of Complexes Formed with DNA. Bioconjugate Chem. 1999, 10, 993−1004. (31) Kichler, A.; Leborgne, C.; Coeytaux, E.; Danos, O. Polyethylenimine-Mediated Gene Delivery: a Mechanistic Study. J. Gene Med. 2001, 3, 135−144. (32) Benjaminsen, R. V.; Mattebjerg, M. A.; Henriksen, J. R.; Moghimi, S. M.; Andresen, T. L. The Possible ‘Proton Sponge’ Effect of Polyethylenimine (PEI) Does Not Include Change in Lysosomal pH. Mol. Ther. 2013, 21, 149−157. (33) Liang, W.; Lam, J. K. W. Endosomal Escape Pathways for NonViral Nucleic Acid Delivery Systems. In Molecular Regulation of. Endocytosis; Ceresa, B., Ed.; InTech: Rijeka, Croatia, 2012; Chapter 17, pp 429−456.

REFERENCES

(1) Boussif, O.; Lezoualc’h, F.; Zanta, M. A.; Mergny, M. D.; Scherman, D.; Demeneix, B.; Behr, J.-P. A Versatile Vector for Gene and Oligonucleotide Transfer into Cells in Culture and In Vivo: Polyethylenimine. Proc. Natl. Acad. Sci. U. S. A. 1995, 92, 7297−301. (2) Ruponen, M.; Ylä-Herttuala, S.; Urtti, A. Interactions of Polymeric and Liposomal Gene Delivery Systems with Extracellular Glycosaminoglycans: Physicochemical and Transfection Studies. Biochim. Biophys. Acta, Biomembr. 1999, 1415, 331−341. (3) Tang, M. X.; Szoka, F. C. The Influence of Polymer Structure on the Interactions of Cationic Polymers with DNA and Morphology of the Resulting Complexes. Gene Ther. 1997, 4, 823−832. (4) Hyvönen, Z.; Plotniece, A.; Reine, I.; Chekavichus, B.; Duburs, G.; Urtti, A. Novel Cationic Amphiphilic 1,4-Dihydropyridine Derivatives for DNA Delivery. Biochim. Biophys. Acta, Biomembr. 2000, 1509, 451−466. (5) Ruponen, M.; Rönkkö, S.; Honkakoski, P.; Pelkonen, J.; Tammi, M.; Urtti, A. Extracellular Glycosaminoglycans Modify Cellular Trafficking of Lipoplexes and Polyplexes. J. Biol. Chem. 2001, 276, 33875−33880. (6) Sonawane, N. D.; Szoka, F. C.; Verkman, A. S. Chloride Accumulation and Swelling in Endosomes Enhances DNA Transfer by Polyamine-DNA Polyplexes. J. Biol. Chem. 2003, 278, 44826−31. (7) Ruponen, M.; Honkakoski, P.; Tammi, M.; Urtti, A. Cell-Surface Glycosaminoglycans Inhibit Cation-Mediated Gene Transfer. J. Gene Med. 2004, 6, 405−414. (8) Hanzlíková, M.; Ruponen, M.; Galli, E.; Raasmaja, A.; Aseyev, V.; Tenhu, H.; Urtti, A.; Yliperttula, M. Mechanisms of PolyethylenimineMediated DNA Delivery: Free Carrier Helps to Overcome the Barrier of Cell-Surface Glycosaminoglycans. J. Gene Med. 2011, 13, 402−409. (9) van de Wetering, P.; Cherng, J.-Y.; Talsma, H.; Crommelin, D. J.; Hennink, W. 2-(Dimethylamino)ethyl Methacrylate Based (Co)polymers as Gene Transfer Agents. J. Controlled Release 1998, 53, 145−153. (10) Wagner, E. Strategies to Improve DNA Polyplexes for In Vivo Gene Transfer: Will ‘Artificial Viruses’ Be the Answer? Pharm. Res. 2004, 21, 8−14. (11) Ikonen, M.; Murtomäki, L.; Kontturi, K. Controlled Complexation of Plasmid DNA with Cationic Polymers: Effect of Surfactant on the Complexation and Stability of the Complexes. Colloids Surf., B 2008, 66, 77−83. (12) Ketola, T.-M.; Hanzlíková, M.; Leppänen, L.; Raviña, M.; Bishop, C. J.; Green, J. J.; Urtti, A.; Lemmetyinen, H.; Yliperttula, M.; Vuorimaa-Laukkanen, E. Independent Versus Cooperative Binding in Polyethylenimine−DNA and Poly(L-lysine)−DNA Polyplexes. J. Phys. Chem. B 2013, 117, 10405−10413. (13) Kabanov, A. V.; Szoka, F. C.; Seymour, L. W. In Self-Assembling Complexes for Gene Delivery: From Laboratory to Clinical Trial; Kabanov, A. V., Felgner, P. L., Seymour, L. W., Eds; John Wiley: Chichester, U.K., 1998; pp 197−218. (14) Vinogradov, S. V.; Bronich, T. K.; Kabanov, A. V. Self-Assembly of Polyamine−Poly(ethylene glycol) Copolymers with Phosphorothioate Oligonucleotides. Bioconjugate Chem. 1998, 9, 805−812. (15) Ketola, T.-M.; Hanzlíková, M.; Urtti, A.; Lemmetyinen, H.; Yliperttula, M.; Vuorimaa, E. Role of Polyplex Intermediate Species on Gene Transfer Efficiency: Polyethylenimine−DNA Complexes and Time-Resolved Fluorescence Spectroscopy. J. Phys. Chem. B 2011, 115, 1895−1902. (16) Vuorimaa, E.; Urtti, A.; Seppänen, R.; Lemmetyinen, H.; Yliperttula, M. Time-Resolved Fluorescence Spectroscopy Reveals Functional Differences of Cationic Polymer−DNA Complexes. J. Am. Chem. Soc. 2008, 130, 11695−11700. (17) Banerjee, P.; Reichardt, W.; Weissleder, R.; Bogdanov, A., Jr. Novel Hyper-branched Dendron for Gene Transfer In Vitro and In Vivo. Bioconjugate Chem. 2004, 15, 960−968. (18) Grecco, H. E.; Verveer, P. J. FRET in Cell Biology: Still Shining in the Age of Super-Resolution? ChemPhysChem 2011, 12, 484−490. K


Time-Resolved Fluorescence Spectroscopy Reveals Fine Structure

Recommend Documents