Fast Characterization of Polyplexes by Taylor Dispersion Analysis

Article pubs.acs.org/Macromolecules

Fast Characterization of Polyplexes by Taylor Dispersion Analysis Laurent Leclercq,† Sören Reinhard,‡ Joseph Chamieh,† Markus Döblinger,§ Ernst Wagner,‡ and Hervé Cottet*,† †

Institut des Biomolécules Max Mousseron (IBMM, UMR 5247 CNRS, Ecole Nationale Supérieure de Chimie de Montpellier), Université de Montpellier, Place Eugène Bataillon, CC 1706, 34095 Montpellier, Cedex 5, France ‡ Pharmaceutical Biotechnology, Department of Pharmacy, and Center for NanoScience, and §Department of Chemistry, Ludwig-Maximilians-Universität, Butenandtstrasse 5-13, D-81377 Munich, Germany S Supporting Information *

ABSTRACT: In a single procedure, Taylor dispersion analysis (TDA) was used for the size characterization of polyplexes and the quantification of free polycation contained in excess within the polyplex sample. TDA analysis was carried out in frontal mode for a better sensitivity of detection. The proof of concept was established using a model polyplex generated from the mixture of linear polylysine (DP 20) and DNA from salmon testes at nitrogen to phosphate (N/P) ratio of 12. Polyplex hydrodynamic radius was compared to the values obtained by dynamic light scattering measurements. TDA was found to give access to the weight-average hydrodynamic radius, while DLS basically gives an intensityaverage (harmonic z-average) value. The method was next applied to the study of various polyplexes issued from polylysines of various DP (50, 100) and different topologies (dendrigraft polylysines of generation 2 and 3). This new methodology should greatly contribute to the physicochemical characterization of polyplexes used for gene transfection.

1. INTRODUCTION Polycations bind to polyanionic DNA to form complexes, named polyplexes, having varying shapes (e.g., rods, toroids, spheres) and sizes (on the order of 20−200 nm).1,2 Generally, these polyplexes are prepared with a large excess of polycation molecules that have a crucial role in the DNA transfection process.2−6 Polyplexes having a global net positive charge can interact with negatively charged proteoglycans on cell membrane and undergo nonspecific or caveolae-mediated endocytosis which internalizes the polyplexes into endosomes.7,8 Successful transfection requires polyplexes of appropriate size, surface charge, and physical and chemical stability.7,8 Physicochemical characterization of polyplexes is very important in understanding polymer−DNA binding affinity, complex stability (under physiological conditions), size, shape, and colloidal behavior in the various media used for in vitro delivery (or in blood for in vivo applications) because these parameters can significantly affect efficiency and side effects. Usually, polyplexes are characterized by gel electrophoresis shift assays, which reveal the required nitrogen to phosphate (N/P) ratio of complex neutralization. Dynamic light scattering (DLS) combined with electrophoretic mobility measurement via the Doppler effect (also called zetametry) is typically completed to assess the general nanoparticle hydrodynamic radius and surface charge (or zeta potential).9 DLS and zetametry techniques are primary methods for characterizing nanoparticle size and charge and are also used to monitor polyplex colloidal stability upon the addition of salt and serum media. For © XXXX American Chemical Society

additional information on the morphologies of polyplexes (rodlike or toroidal), direct visualization techniques, such as transmission electron microscopy (TEM), scanning electron microscopy (SEM) and atomic force microscopy (AFM), are also used to observe general polyplex size and shape.10−14 It is well-known that the periphery of polyplexes in solution should be positively charged in order to achieve reasonable transfection efficiency.15,16 Previous studies showed that nearly all DNA chains were complexed by polyethylenimine (PEI) to form polyplexes (∼100 nm size) when the polycation/DNA ratio equaled or exceeded N/P = 3, irrespective of the chain length of PEI and ionic strength (water or PBS) used.15−20 Optimum, high in vitro gene transfection efficiency was only achieved with N/P around 10 or higher. Therefore, it was concluded that at high N/P ratio there are two kinds of PEI chains in solution: PEI chains bound to DNA and free PEI chains in solution. The free PEI chains are essential to promote gene transfection. These findings were further confirmed with other polycations like poly(L-lysine),7 poly(dimethylaminoethyl methacrylate),7 and chitosan.17 However, for significantly toxic polycations such as PEI, the previous studies also showed that the unbound polycationic molecules, existing as free individual chains in solution, not only improve the transfection activity but also trigger enhanced cytotoxicity. It is therefore very important to quantify the free polycation in polyplexes. Received: August 17, 2015 Revised: September 22, 2015

A

DOI: 10.1021/acs.macromol.5b01824 Macromolecules XXXX, XXX, XXX−XXX

Article

Macromolecules

given distance from the injection end of the capillary. For a monodisperse sample, the time recording detection signal is given by

Taylor dispersion analysis (TDA) is an absolute method for the determination of diffusion coefficient (and thus hydrodynamic radius, Rh) based on the dispersion of a sample plug in a laminar flow.21 A regain of interest in TDA has been recently observed22−28 due to the simplicity and the rapidity of the approach, the low consumption of analyte (a few nanoliters), the absence of calibration, and the possibility to determine molecular dimensions down to angstroms. In addition, in the case of polydisperse sample, TDA leads to the weight-average Rh which is not biased toward the larger aggregates or nanoparticles contained in the sample as observed for the intensity-average Rh measured in DLS.29−31 In a previous work, the online coupling of capillary electrophoresis to TDA was used for the characterization of polyelectrolyte complexes after dissociation at high ionic stength.27 The charge stoichiometry and the hydrodynamic radii of the two polyelectrolyte constituents were determined in a fully automated single run. In the present work, we report the use of frontal mode TDA for polyplex characterization, where both the weight-average Rh of polyplex and the amount of free polycation in excess are determined in a single run.

S( t ) =

⎡ −(t − t )2 ⎤ S0 0 ⎥ exp⎢ 2 σ 2π 2σ ⎣ ⎦

(1)

where t0 is average elution time, σ is the temporal variance of the elution profile, and S0 is a constant that depends on the response factor and injected quantity of the solute. In the conditions of application of TDA that are remembered later on in this section, the diffusion coefficient D of the solute, and the corresponding hydrodynamic radius Rh, are related to σ2 according to 2

D=

Rh =

R c 2t0 24σ 2

(2)

kBT 4σ 2kBT = 6πηD πηR c 2t0

(3)

where kB is the Boltzmann constant, T is the temperature (in K), Rc is the capillary diameter, and η is the viscosity of the eluent. It means that a monodisperse sample, having one Rh and one corresponding D value, leads to a Gaussian peak with a temporal variance which is directly related to D (and thus to Rh). The higher Rh is, the lower D is, the broader the Gaussian peak is. To increase the sensitivity of detection, TDA can be implemented in frontal mode (see Figure 1B). In that case, the sample is continuously, and hydrodynamically, injected in the capillary, leading to the detection of a front instead of a peak. The temporal signal front is given by

2. THEORETICAL SECTION Taylor dispersion is based on the analysis of band broadening of a short initial solute plug under a laminar Poiseuille flow in an open tube (see Figure 1A). Because of the parabolic velocity

S(t ) 1 1 ⎛ t − t0 ⎞ ⎟ = + erf⎜ S0 2 2 ⎝σ 2 ⎠

(4)

where the first derivative of eq 4 leads to eq 1. For bimodal mixture composed of two different populations having different sizes, eq 4 becomes S(t ) =

⎛ t − t 0 ⎞ S2 ⎛ t − t 0 ⎞ S1 + S2 S + 1 erf⎜ erf⎜ ⎟+ ⎟ 2 2 2 ⎝ σ2 2 ⎠ ⎝ σ1 2 ⎠

(5)

Equations 1, 2, 4, and 5 are valid when two conditions are fulfilled.32 First, t0 should be much longer than the characteristic diffusion time of the solute in the cross section of the capillary, i.e., t0 ≥ 1.25Rc2/D for a relative error ε on the determination of D lower than 3%. Second, the axial diffusion should be negligible compared to convection (i.e., the Peclet number Pe = Rcu/D should be superior to 40 for ε lower than 3%32−34 (with u being the average linear mobile phase velocity)).

Figure 1. Schematic representation of the Taylor dispersion analysis of polyplexes using plug injection (A) or frontal mode (B). Free polycations lead to low dispersion band (A)/front (B), while polyplexes of larger size lead to broader zone/front.

3. METHODS AND MATERIALS 3.1. Raw Materials. TRIS was purchased from Carlo Erba (Paris, France). Hydroxypropyl cellulose (HPC, Mw 1.0 × 105 g/mol), poly(diallyldimethylammonium chloride) (PDADMAC, Mw 5.0 × 105 g/mol), and deoxyribonucleic acid sodium salt from salmon testes (DNA, Mw 1.3 × 106 Da, ∼ 2000 base pairs) were purchased from Sigma-Aldrich, Saint-Quentin Fallavier, France. Linear poly(L-lysine) hydrochloride of various molar masses (PLKC20 (degree of polymerization DP 20, lot no. KC020-101), Mw 3300 g/mol and polydispersity index (PDI) 1.05; PLKC50 (lot no. KC050-102), Mw 8200 g/mol and PDI 1.04, and PLKC100 (lot no. KC100-105), Mw

profile, the analyte initial band is dispersed according to the combination of a convection/diffusion process, also named Taylor−Aris dispersion.21 The analytes are redistributed along the tube cross section owing to the molecular diffusion. When the characteristic diffusion time is lower than the average detection time, the Taylor−Aris dispersion leads to a Gaussian peak for a monodisperse sample. The elution profile is generally obtained by online UV detection through the capillary tube at a B


Article

Macromolecules

Figure 2. Taylorgrams obtained for DNA in frontal mode (A) and its first derivative (B). Experimental conditions: HPC-coated capillary 60 cm total length (51.5 cm to the UV detector) × 50 μm i.d. Eluent: 10 mM TRIS, HCl pH 7.4. Mobilization pressure: 50 mbar. UV detection: 200 and 260 nm. Temperature: 25 °C. DNA concentration: 0.2 g/L in the eluent. Erf fitting according to eq 4 are displayed in red; experimental data are in black.

Figure 3. Taylorgrams obtained for PLKC20 in frontal mode (A) and its first derivative (B). Experimental conditions as in Figure 2. PLKC20 concentration: 5 g/L in the eluent. 16 000 g/mol and PDI 1.13) were from Alamanda Polymers, Huntsville, AL. Dendrigraft poly(L-lysine) of various generations (DGL-G2 (lot no. DC1210-02B), Mw 8600 g/mol and PDI 1.38 and DGL-G3 (lot no. DC1201-03), Mw 22 000 g/mol and PDI 1.46) were supplied by COLCOM, Montpellier, France. Polybead polystyrene microspheres of 25 and 100 nm theoretical hydrodynamic radius (2.5% solid (w/v) aqueous suspensions) were purchased from Polysciences, Eppelheim, Germany. Deionized water was further purified with a Milli-Q system from Millipore (Molsheim, France). 3.2. Formation of Polyplexes. Polyplexes were prepared directly before TDA experiments. A 0.2 g/L DNA solution and a 5 g/L polycation solution were prepared in 10 mM TRIS, HCl at pH 7.4. A 24 μL aliquot of PLKC (or 32 μL aliquot of DGL) solution was first added to 76 μL (or 68 μL) of TRIS buffer solution. The resulting mixture was then added to a 100 μL DNA solution and shortly stirred at room temperature. Therefore, final concentrations were 0.1 g/L for DNA and 0.6 g/L (0.8 g/L) for the PLKC (DGL) polycation. The N/ P ratio was 12 for all the systems. 3.3. Taylor Dispersion Analysis Experiments (TDA). Capillaries were prepared from bare silica tubing purchased from Composite Metal Services (Worcester, United Kingdom). All TDA experiments were carried out at 25 °C using frontal mode (i.e., by continuous injection of the sample; see Figure 1B). Each sample is prepared in the background electrolyte (10 mM TRIS, HCl at pH 7.4). Before sample analysis, the capillary was previously filled with 10 mM TRIS HCl buffer. All TDA experiments were realized in duplicates. In the case of PLKC20/DNA polyplex studied for the proof of concept, TDA experiments were performed on a CE Agilent (Waldbronn, Germany) apparatus using 50 μm i.d. × 60 cm (× 51.5 cm to the detector) capillaries. Capillaries were coated with

hydroxypropyl cellulose (HPC) according to a previously described procedure.35 Solutes were monitored by UV absorbance at both 200 and 260 nm. TDA experiments were carried out using 50 mbar mobilization pressure. In the case of all other polyplexes studied in this work, TDA experiments were performed on a P/ACE MDQ system (Beckman, USA) using 50 μm i.d. × 40 cm (× 29.75 cm to the detector). Capillaries were coated with poly(diallyldimethylammonium chloride) (PDADMAC) instead of HPC since the protocol was faster and because no free DNA was present in the polyplex samples. For PDADMAC coating, the capillaries were first conditioned with 1 M NaOH for 20 min and water for 5 min, then flushed with a 0.2% PDADMAC aqueous solution for 10 min, and finally flushed with 10 mM TRIS, HCl buffer at pH 7.4 for 10 min. Solutes were monitored by UV absorbance at 214 nm. TDA experiments were carried out using 30 mbar mobilization pressure. As our CE Beckman apparatus is not equipped with a multiwavelength detector, CE Agilent apparatus was selected for the experiments on PLKC20/DNA polyplex to be able to detect solutes both at 200 nm (where PLKC and DNA absorb) and 260 nm (where only free or bound DNA absorb). Determination of Rh led to the same results at the two wavelengths. All the other polyplexes were then characterized at 200 nm only on the Beckman instrument. 3.4. Dynamic Light Scattering (DLS). The polyplexes were prepared in a total volume of 200 μL as described above and measured in a folded capillary cell (DTS1070) using a Zetasizer Nano ZS with backscatter detection (Malvern Instruments, Worcestershire, UK). The equilibration time was 0 min, the temperature was 25 °C, and an automatic attenuator was used. The refractive index of the solvent water was 1.330, and the viscosity was 0.8872 cP. For polyplex analysis C


Article

Macromolecules of the particles the refractive index of protein (1.45) was used. Each sample was measured three times. 3.5. Transmission Electron Microscopy (TEM). The polyplexes were prepared in a total volume of 20 μL as described in section 3.4. 80 μL of 10 mM TRIS was added to dilute the samples by 1:5. A carbon-coated 400 mesh copper grid (Plano GmbH, Germany) was activated with hydrogen/oxygen gas for 30 s using the Solarus advanced plasma cleaning system (Gatan GmbH, Germany). A 5 μL drop of the polyplex sample was placed onto the grid and incubated for 10 min. Samples were rinsed cautiously with two drops of doubledistilled water to remove buffer salts and then stained with 5 μL of 1% phosphotungstic acid aqueous solution (Science Services, Germany) for 3 min. All samples were subsequently air-dried prior to imaging at an accelerating voltage of 80 kV using a FEI Titan 80-300 operated at 80 kV.

prepared in the same background electrolyte (eluent), with initial concentrations in DNA and in polycation that are similar in the individual samples and in the mixture (polyplex sample). TDA experiments were carried out in frontal mode (instead of plug injection) to improve the sensitivity of detection by avoiding the effect of dilution. A HPC-coated capillary was used to avoid any interaction between the polycation (PLKC20), or the cationic polyplex, and the capillary wall. Figures 2A and 3A were fitted with eq 3, where t0 is the elution time at the inflection point of the erf function, S0 is the absorbance at the top of the plateau, and σ2 is the temporal variance of the erf function. To limit the appearance of noisy baseline, it is preferable to directly fit the taylorgram in frontal mode instead of fitting the first derivative of the initial frontal taylorgram. As expected, (free) DNA leads to a much more disperse TDA profile than PLKC due to the large difference in size (or molar mass) between these two samples: the broader the TDA signal, the larger the solute size. Indeed, Taylor dispersion increases with lower diffusion coefficient (or larger size). Free DNA hydrodynamic radius was ca. 110 nm versus 1.1 nm for PLKC20. Figures 2B and 3B display the first derivative of the “erf” taylorgrams presented in Figures 2A and 3A, respectively. Figures 3B and 4B actually represent the Gaussian curves that would have been obtained in “plug” TDA, if the sensitivity would have been sufficient to ensure correct detection. It is worth noting that both DNA and PLKC20 do absorb at 200 nm, while only DNA could be detected at 260 nm (with a lower molar extinction coefficient than at 200 nm). DNA size obtained at 200 and 260 nm was found similar (Rh of 111 nm vs 106 nm, respectively). Elution times are not different between PLKC20 and DNA samples, meaning that the variation in viscosity between the electrolyte and the samples could be neglected due to the high dilution of the samples.The TDA analysis of the DNA/PLKC20 polyplex sample was carried out under the same conditions (see Figure 4A). Contrary to the case of the individual PLKC20 and DNA constituents, the sum of two erf functions should be used (see eq 4) for the fitting of the taylorgram of the polyplex mixture due to the bimodal size distribution. The polyplex sample contains PLKC20/DNA complex and a certain amount of free PLKC20, contained in excess in the sample. The bimodal characteristic of the mixture can be clearly visualized in Figure 4B (first derivative of the frontal taylorgram), which appears as the sum of two Gaussian peaks at 200 nm. The broader contribution is due to the polyplex itself, while the thinner peak corresponds to the free PLKC (this situation corresponds to the schematic representation shown in Figure 1). It is worth noting that the polyplex absorbs at both 200 and 260 nm due to the DNA contribution to the signal absorbance, while the free PLKC20 is only detected at 200 nm. The average hydrodynamic radius of the polyplex was found to be 39 nm (determined at both 200 and 260 nm), while the PLKC contribution observed at 200 nm leads to a Rh of 1.2 nm, in good agreement with the value obtained for the individual PLKC20 sample, as previously observed.28 The Gaussian trace at 260 nm in Figure 4B, which is specific to the DNA detection, proves that all DNA chains are entrapped in the polyplex since free DNA would lead to much broader peak. This result is in good agreement with the high N/P = 12 ratio. It is interesting to note that DNA was condensed (compacted) within the polyplex since the hydrodynamic size of DNA was smaller in the polyplex than in free solution. Using a calibration curve for PLKC20 based on the front heights (see inset in Figure 4A),

4. RESULTS AND DISCUSSION 4.1. Proof of Concept on PLKC20/DNA Polyplex. The objective of the present work is to demonstrate that it is possible to determine by TDA, in a single run, the hydrodynamic radius of polyplexes and the quantity of free polycation contained in the polyplex samples. Figures 2A, 3A, and 4A respectively display the frontal taylorgrams obtained for the DNA, the polycationic vector poly(L-lysine) hydrochloride (PLKC20, DP 20), and the polyplex sample, keeping the same elution time window from 5 to 20 min. The samples were

Figure 4. Taylorgrams obtained for the PLKC20/DNA polyplex sample (N/P = 12) in frontal mode (A) and its first derivative (B). Experimental conditions as in Figure 2. After mixture, the final PLL concentration is 0.6 g/L and DNA concentration is 0.1 g/L. Inset in (A) displays the external calibration curve obtained for the PLKC20, based on front heights of PLKC20 solutions. Inset in (B) is a TEM picture of the polyplexes (free PLKC cannot be visualized). D


Article

Macromolecules the quantity of free (unbound) PLKC20 was ∼82% of the initial concentration (18% are complexed with DNA). Table 1

Equation 7 is established for a mass−concentration sensitive detector such as UV detector in the case of a polymer absorbing via the repeating unit. DNA, polylysines, and their polyplexes correspond to that case, and therefore eq 7 does apply. Of course, in the case of polydisperse samples, eqs 6 and 7 are intrinsically different with Rh,TDA ≤ Rh,DLS. The two values only converge for monodisperse samples. Interestingly, the relative difference between the two values is a direct estimation of sample size polydispersity. TEM experiments were also performed (see inset of Figure 4B). TEM experiments revealed the presence of a majority of polyplexes with Rh close to TDA values, together with some larger polyplexes, but in much lower number, that confirms a certain polydispersity of the sample. Even the presence of a small proportion of larger aggregates can explain the observed difference between the intensity-average value obtained by DLS and the TDA value. 4.2. Application to Polyplexes Obtained with Different Polylysines (PLKC and DGL). The present study was next generalized to other polyplexes using other polylysines of different molar masses (PLKC50, PLKC100) and different topologies (dendrigraft polylysines, DGL-G2 and DGL-G3), keeping constant the N/P ratio to 12. DGL-G2 is a branched comblike polylysine having a DP of 48. DGL-G3 is an hyperbranched polylysine, obtained from G2, and having a DP of 123. For more detailed on the structures of DGL, the reader can refer to the original paper describing their synthesis and characterization.36 The taylorgrams of the polyplexes are presented in the Supporting Information (see Figures SI-1 to SI-4). For all samples, no free DNA could be detected. The numerical values corresponding to the free proportion of polycation and the polyplex Rh derived from DLS and frontal TDA are gathered in Table 2. It appears that the free polycation mass proportion is similar for PLKC20 and PLKC50 (about 80%) and is slightly lower for PLKC100 (about 75%) and even lower for DGL-G2 (68%) or DGL-G3 (65%). Regarding the size of the polyplexes, the weight-average Rh is about 35−40 nm for PLKC and slightly lower, about 25−30 nm for DGL G2-G3. Intensity-average Rh from DLS are typically about 20 nm higher than the weight-average Rh due to the presence of some aggregates in the samples. Based on the ratio of the Rh obtained by DLS and TDA, size polydispersity of the samples seems quite similar for PLKC and DGL.

Table 1. Average Hydrodynamic Radii (Rh) of PLKC20, DNA, and DNA/PLKC20 Polyplex at 200 and/or 260 nma polymer

PLKC20

detection wavelength (nm) fitting function Rh (nm) τ Pe

PLKC20/DNA polyplex

DNA

200

200

260

200

260

1 erf 1.1 301 71

1 erf 106 2.4 8021

1 erf 110 2.3 8370

2 erf 39/1.2 4.3/301 4899/71

1 erf 38 4.4 4787

In each case, numerical values of the dimensional τ and Pe parameters are given to demonstrate that the conditions of validity of TDA were fulfilled, i.e., Pe = Rcu/D ≥ 40 and τ = t0D/Rc2 ≥ 1.25. The precision of average size determination by TDA is ∼5%. a

gathers all the numerical values determined in this work together with the two characteristic adimensional parameters (τ and Pe), the values of which verified the conditions of validity of TDA (see the end of Theoretical Section). These results were compared to DLS (see Table 2 for numerical values). The intensity-average Rh obtained by DLS Table 2. Polyplex Average Hydrodynamic Radii (Rh) Obtained by DLS and by TDA for the Different Systems Based on PLKC20, PLKC50, PLKC100, and DGL (G2− G3)a DLS

TDA

Rhb

[nm] in intensity

polyplex PLKC20/DNA PLKC50/DNA PLKC100/DNA DGL-G2/DNA DGL-G3/DNA

66.4 55.4 52.4 46.5 53.5

± ± ± ± ±

1.7 1.8 1.1 1.2 1.6

Rhc [nm] 41.7 33.0 39.2 26.3 30.7

± ± ± ± ±

3.8 1.0 2.5 2.4 3.8

% of free polycationd 83 81 75 68 65

± ± ± ± ±

1 1 1 2 2

Average values ± one standard deviation (n = 3 (DLS) or n = 2 (TDA)). bHarmonic z-average Rh (see eq 6) given by cumulants analysis (intensity averaged particle diameter). cWeight-average Rh (see eq 7) obtained by curve fitting of the frontal taylorgram at 214 nm using eq 5. The reported Rh corresponds to the polyplex population. dMass proportion relative to the introduced polycation concentration. a

5. CONCLUSIONS (using the cumulant analysis of the autocorrelation function) is an harmonic z-average value given by the following equation:29 R h,DLS =

This work demonstrates that frontal TDA can give access, in a single run and in a few minutes, to two major key characteristics of polyplexes, namely, the weight-average polyplex hydrodynamic radius and the amount of free polycation in solution, at equilibrium. This methodology can be applied even for cationic polymers with polydisperse molar mass distributions, as far as the polyplexes have different size compared to the free polycation. Owing to the advantages relative to the TDA methodology (straightforward and simple method, low sample volume, low sensitivity to dust, weight-average Rh which is not biased toward larger size aggregates, sizing down to angstroms, absolute determination without calibration), we believe that the proposed methodology could become a method of choice for fast polyplex characterization and physicochemical studies in gene delivery.

2 ∑i NM i i

∑i

2 NM i i R h, i

(6)

with Ni being the number of moles of macromolecules of molar mass Mi in the sample. As expected, this value obtained for the PLKC20/DNA sample (66.5 nm) was higher than the result obtained by TDA (39.2 nm). In TDA, a weight-average hydrodynamic radius is determined for a mass−concentration sensitive detector:29

R h,TDA =

∑i NM i iR h, i ∑i NM i i

(7) E


Article

Macromolecules

■

(24) Jensen, H.; Østergaard, J. J. Am. Chem. Soc. 2010, 132, 4070− 4071. (25) Le Saux, T.; Cottet, H. Anal. Chem. 2008, 80, 1829−1832. (26) Østergaard, J.; Jensen, H. Anal. Chem. 2009, 81, 8644−8648. (27) Leclercq, L.; Cottet, H. Anal. Chem. 2012, 84, 1740−1743. (28) Jin, X. Y.; Leclercq, L.; Sisavath, N.; Cottet, H. Macromolecules 2014, 47, 5320−5327. (29) Cottet, H.; Biron, J. P.; Martin, M. Anal. Chem. 2007, 79, 9066− 9073. (30) Cottet, H.; Martin, M.; Papillaud, A.; Souaïd, E.; Collet, H.; Commeyras, A. Biomacromolecules 2007, 8, 3235−3243. (31) Hawe, A.; Hulse, W. L.; Jiskoot, W.; Forbes, R. T. Pharm. Res. 2011, 28, 2302−2310. (32) Cottet, H.; Biron, J. P.; Martin, M. Analyst 2014, 139, 3552− 3562. (33) d’Orlyé, F.; Varenne, A.; Gareil, P. J. Chromatogr. A 2008, 1204, 226−232. (34) Taylor, G. Proc. R. Soc. London, Ser. A 1954, 225, 473−477. (35) Shen, Y.; Smith, R. D. J. Microcolumn Sep. 2000, 12, 135−141. (36) Collet, H.; Souaïd, E.; Cottet, H.; Deratani, A.; Boiteau, L.; Dessalces, G.; Rossi, J. C.; Commeyras, A.; Pascal, R. Chem. - Eur. J. 2010, 16, 2309−2316.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.macromol.5b01824. Figures SI-1 to SI-4 (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Tel +33 4 6714 3427, Fax +33 4 6763 1046; e-mail herve. [email protected] (H.C.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS H.C. gratefully acknowledges the support from the Institut Universitaire de France and from the Region LanguedocRoussillon for the fellowship “Chercheurs d’Avenir”. E.W. appreciates financial support by the DFG Cluster of Excellence Nanosystems Initiative Munich (NIM).

■

REFERENCES

(1) Tang, M. X.; Szoka, F. C. Gene Ther. 1997, 4, 823−832. (2) Lächelt, U.; Wagner, E. Chem. Rev. 2015, DOI: 10.1021/ cr5006793. (3) Reineke, T. M.; Davis, M. E. Bioconjugate Chem. 2003, 14, 255− 261. (4) Anderson, D. G.; Lynn, D. M.; Langer, R. Angew. Chem., Int. Ed. 2003, 42, 3153−3158. (5) Anderson, D. G.; Akinc, A.; Hossain, N.; Langer, R. Mol. Ther. 2005, 11, 426−434. (6) Yue, Y. N.; Wu, C. Biomater. Sci. 2013, 1, 152−170. (7) Lechardeur, D.; Verkman, A. S.; Lukacs, G. L. Adv. Drug Delivery Rev. 2005, 57, 755−767. (8) Pack, D. W.; Hoffman, A. S.; Pun, S.; Stayton, P. S. Nat. Rev. Drug Discovery 2005, 4, 581−593. (9) Uskoković, V. J. Dispersion Sci. Technol. 2012, 33, 1762−1786. (10) Ulasov, A. V.; Khramtsov, Y. V.; Trusov, G. A.; Rosenkranz, A. A.; Sverdlov, E. D.; Sobolev, A. S. Mol. Ther. 2011, 19, 103−112. (11) Troiber, C.; Kasper, J. C.; Milani, S.; Scheible, M.; Martin, I.; Schaubhut, F.; Küchler, S.; Rädler, J.; Simmel, F. C.; Friess, W.; Wagner, E. Eur. J. Pharm. Biopharm. 2013, 84, 255−264. (12) Lai, E.; van Zanten, J. H. Biophys. J. 2001, 80, 864−867. (13) Lin, W. J.; Hsu, W. Y. Carbohydr. Polym. 2015, 120, 7−14. (14) Agirre, M.; Zarate, J.; Ojeda, E.; Puras, G.; Desbrieres, J.; Pedraz, J. L. Polymers 2014, 6, 1727−1755. (15) Yue, Y. A.; Jin, F.; Deng, R.; Cai, J. G.; Chen, Y. C.; Lin, M. C. M.; Kung, H. F.; Wu, C. J. Controlled Release 2011, 155, 67−76. (16) Yue, Y. A.; Jin, F.; Deng, R.; Cai, J. G.; Dai, Z. J.; Lin, M. C. M.; Kung, H. F.; Mattebjerg, M. A.; Andresen, T. L.; Wu, C. J. Controlled Release 2011, 152, 143−151. (17) Thibault, M.; Astolfi, M.; Tran-Khanh, N.; Lavertu, M.; Darras, V.; Merzouki, A.; Buschmann, M. D. Biomaterials 2011, 32, 4639− 4646. (18) Clamme, J. P.; Azoulay, J.; Mely, Y. Biophys. J. 2003, 84, 1960− 1968. (19) Clamme, J. P.; Krishnamoorthy, G.; Mely, Y. Biochim. Biophys. Acta, Biomembr. 2003, 1617, 52−61. (20) Boeckle, S.; Von Gersdorff, K.; Van der Piepen, S.; Culmsee, C.; Wagner, E.; Ogris, M. J. Gene Med. 2004, 6, 1102−1111. (21) Taylor, G. Proc. R. Soc. London, Ser. A 1953, 219, 186−203. (22) Bello, M. S.; Rezzonico, R.; Righetti, P. G. Science 1994, 266, 773−776. (23) Sharma, U.; Gleason, N. J.; Carbeck, J. D. Anal. Chem. 2005, 77, 806−813. F


Fast Characterization of Polyplexes by Taylor Dispersion Analysis

Recommend Documents