Article pubs.acs.org/ac
Fast Characterization of Polyelectrolyte Complexes by Inline Coupling of Capillary Electrophoresis to Taylor Dispersion Analysis Laurent Leclercq† and Hervé Cottet*,‡ †
Institut des Biomolécules Max Mousseron, UMR CNRS 5247, Université de Montpellier 1, Université de Montpellier 2, 15 Avenue Charles Flahault 34060 Montpellier, France ‡ Institut des Biomolécules Max Mousseron, UMR CNRS 5247,Université de Montpellier 1, Université de Montpellier 2, Place Eugène Bataillon, 34095 Montpellier Cedex 5, France ABSTRACT: The inline coupling of capillary electrophoresis (CE) to Taylor dispersion analysis was used for the characterization of polyelectrolyte complexes. The charge stoichiometry and the hydrodynamic radii of the two polyelectrolyte constituents were determined in a fully automated single run using standard commercial CE apparatus with a single detection point. The proposed methodology utilizes unusual high ionic strength (1.3 M) background electrolyte to obtain the dissociation and electrophoretic separation of polyelectrolyte constituents. Such highly saline conditions in combination with neutrally coated capillary were found to avoid any polyelectrolyte interactions onto the capillary surface. This innovative methodology should greatly contribute to simplify and accelerate the characterization of polyelectrolyte complexes or polyplexes.
P
characterization. However, studying selectivity phenomena10−15 requires the analysis of the PEC constituents. Ion exchange chromatography (IEC) can be used to separate the PEC components and to determine the charge stoichiometry.16 Nevertheless, off-line coupling to size exclusion chromatography (SEC) using two different columns was required to characterize the polyelectrolyte molar masses after IEC. Such an approach is time-consuming, nonautomated, with dialysis/ freeze-drying steps, and requires at least a few milligrams of each component. Taylor dispersion analysis (TDA) is an absolute method for the determination of hydrodynamic radius (Rh) based on the dispersion of a sample plug in a laminar flow.17 A regain of interest in TDA has been recently observed18−22 due to the simplicity and the rapidity of the approach, the low consumption of analyte (a few nanoliters), the absence of calibration, the possibility to determine molecular dimensions down to angströms, and the determination of the weightaverage Rh in the case of polydisperse samples. In this work, we report a new methodology for PEC characterization using the inline coupling of CE to TDA, where the charge stoichiometry and the hydrodynamic size of the two polyelectrolytes are determined in a single run.
olyanions (PA) and polycations (PC) interact electrostatically to form polyelectrolyte complexes (PEC).1 PEC play a crucial role in nature, where many charged systems like proteins, polysaccharides, and cells are in contact and interact. Let us exemplify with the condensation of DNA on histones2 and PEC formation between glycosaminoglycans (GAG) and collagen that fix the properties of connective tissues.3 DNA and genes condensing on polycationic vectors for the purpose of gene transfection into cells are other examples of PEC biological applications.4 Drug transport using polyelectrolyte carriers and gene transfection imply the injection of PA and PC into blood where interactions may occur with proteins and cells.5−7 Dynamics of PEC formation depends on various factors including pH, ionic strength, molar mass, charge density, concentration, +/− molar ratio, and even mixing order.8,9 For instance, when poly(L-lysine) was progressively added to poly(acrylic acid) or poly(L-lysine citramide), longer PA molecules were found in the first precipitated fractions, while the shorter ones remained in solution.10 In contrast, when PA was added to PC, fractionation was weak.11 Similarly, molecular recognition occurred between poly(ethylene glycol)-b-poly(aspartic acid) and poly(ethylene glycol)-b-poly(L-lysine). Complexes of PA and PC blocks with the same lengths were formed selectively from mixtures of copolymers with disparate block lengths.12 Last but not least, the inhibitory effect of GAG on gene transfer was explained by the replacement of DNA by GAG in the polyplex, resulting in the uptake of GAG into cells instead of DNA.13 So far, little attention has been paid to the chemical composition of PEC, probably because of the complexity of the © 2012 American Chemical Society
Received: December 3, 2011 Accepted: January 4, 2012 Published: January 20, 2012 1740
dx.doi.org/10.1021/ac203208k | Anal. Chem. 2012, 84, 1740−1743
Analytical Chemistry
■
Article
EXPERIMENTAL SECTION Chemicals and Materials. Hydroxypropyl cellulose (HPC, Mw = 100 000 g/mol) and poly(L-lysine) hydrobromide (Mw = 26 300 g/mol and I = 1.9) were purchased from Sigma-Aldrich, Saint-Quentin Fallavier, France. The pKa value of the amine groups of poly(L-lysine) is in the range of 10.2−10.5.23 Poly(Llysine citramide) sodium salt (Mw = 39 000 g/mol and I = 2.0) was kindly supplied by Dr. Mahfoud Boustta (Max Mousseron Institute for Biomolecules, Montpellier, France). The pKa value of the acid groups of poly(L-lysine citramide) is in the range of 4.3−4.6.24 Ultrapure water was obtained using a Milli-RO system from Millipore (Molsheim, France). Analytical grade NaCl was purchased from Merck, Fontenay-sous-Bois, France. Formation of PEC Fraction. A 10 mg PA/0.3 mL phosphate buffer saline (PBS, pH = 7.4, 0.154 ionic strength) solution was added stepwise with NPA/NPC = 0.2 steps to a vigorously stirred 12.1 mg PC/0.9 mL PBS solution. Similarly, a 12.1 mg PC/0.3 mL PBS solution was added stepwise with NPC/NPA = 0.2 steps to a vigorously stirred 10 mg PA/0.9 mL PBS solution. Titrations were achieved in plastic vials at room temperature. The precipitate formed after the first addition (fraction 1) was collected by centrifugation and washed with fresh PBS and finally rinsed with deionized water prior to drying under vacuum. The same procedure was applied to the other additions and yielded successive complex fractions up to the absence of precipitation, i.e., when no turbidity was detected (Figure 1). The residual supernatant (fraction S) was
equipped with a UV detector and an anionic CM-CL6B Sepharose gel column (60 cm length × 1 cm diameter). The mobile phase was 1 M NaCl buffered by 0.13 M NaH2PO4 at pH 7.4. Three poly(L-lysine citramide) of known molar masses determined by static laser light scattering, namely, 56 000 g/ mol, 34 000 g/mol, and 26 000 g/mol, were used for calibration. The molar mass of PLL molecules engaged in the various PEC fractions was determined by SEC using a DEAE Sepharose cationic gel column (40 cm length × 1 cm diameter). The mobile phase was 1.6 M NaCl adjusted to pH 4.0. Four poly(L-lysine) of known molar masses determined by static laser light scattering, namely, 73 500 g/mol, 51 000 g/ mol, 26 300 g/mol, and 12 700 g/mol, were used for calibration. All SEC experiments were operated at 214 nm, and the elution rate was 0.30 mL/min. The injected volume was 0.5 mL. Capillary Electrophoresis Coupled to Taylor Dispersion Analysis. CE-TDA experiments were performed on a PACE MDQ Beckman Coulter (Fullerton, CA) apparatus. Capillaries were prepared from bare silica tubing purchased from Composite Metal Services (Worcester, United Kingdom). Capillary dimensions were 60.7 cm (50.35 cm to the detector) × 50 μm i.d. Capillaries were coated with hydroxypropylcellulose according to a previously described procedure.25 The temperature of the capillary cartridge was set at 25 °C. Solutes were monitored by UV absorbance at 214 nm. Before sample injection, the capillary was filled with a mixture of sodium phosphate buffer (42.5 mM) and NaCl (1.28 M) at pH 6.9.
■
RESULTS AND DISCUSSION Poly(L-lysine citramide) and poly(L-lysine) were selected as examples of biosourced PA and PC, respectively, to generate the PEC. Additions, made according to Figure 1, were characterized by the NPA/NPC or NPC/NPA ratios, where NPA and NPC are the mole number of negative and positive charges introduced in the medium, respectively. Charge neutralization of the polyelectrolyte was progressively obtained by the addition of the oppositely charged polymer, by steps of 20% (in mol) of charge neutralization. At each step of the neutralization, the PEC was centrifuged, washed, dried (see the Experimental Section), and further analyzed by the inline coupling of CE to TDA and by the offline coupling of IEC to SEC. The new methodology for the characterization of the PEC fraction by CE-TDA is schematically described in Figure 2 with an example of an electropherogram/taylorgram obtained for one PEC fraction. Step 1 consisted of the electrophoretic separation of PA and PC constituents assisted by hydrodynamic flow to allow the detection of PC that migrates against the hydrodynamic flow. It should be noted that the PEC sample was injected from the outlet end (50 mbar, 6 s) and that the ionic strength of the electrolyte was sufficiently high (1.3 M) to allow the PEC dissociation. This injection was followed by an injection of buffer (25 mbar, 3 s) to avoid any loss in PC constituent. The applied voltage was 4.5 kV with a copressure of −10 mbars (step 1, Figure 2). It is worth noting that the current intensity (∼190 μA) was stable. The 4.5 kV voltage was sufficient to get the PA/PC separation while maintaining the current intensity below 200 μA. The 10 mbar copressure was set by a trial and error process with the idea to use the lowest pressure as possible to detect the PC while achieving good PA/ PC separation. Steps 2 and 3 correspond to TDA steps of the separated PA and PC zones. In step 2, the voltage was turned
Figure 1. Preparation of PEC fractions by successive additions (1/5 charge molar increment) using a titration protocol.
then dried under vacuum. Centrifugation was carried out using a Sigma 112 centrifuge at 20 000 rpm for 5 min. Five precipitated fractions (fraction 1, 1.2 mg; fraction 2, 2.4 mg; fraction 3, 2.5 mg; fraction 4, 2.5 mg; and fraction 5, 2.2 mg) were collected when PC was added to PA. In contrast, only four precipitated fractions (fraction 1, 2.8 mg; fraction 2, 3.9 mg; fraction 3, 2.9 mg; and fraction 4, 0.6 mg) were collected when PA was added to PC, the fourth precipitated fraction amounting to less than 6% from the total. Ion Exchange Chromatography. IEC experiments were performed using an anionic CM-CL6B Sepharose gel column 70 cm length × 1 cm diameter (Pharmacia, France) to separate the PA and PC components of the PEC as previously described.16 Typically, the solid PEC was dissociated in 2 M NaCl and then injected in the column. With a 1 M NaCl mobile phase, the PC was retained by the gel and the PA was eluted. The PC was then eluted using 2 M NaCl solution. Detection was made at 214 nm. The flow rate was set at 0.5 mL/min. The injection volume was 1 mL. Size Exclusion Chromatography. SEC experiments were performed using a Pharmacia Biotech LCC-501 chromatograph 1741
dx.doi.org/10.1021/ac203208k | Anal. Chem. 2012, 84, 1740−1743
Analytical Chemistry
Article
Figure 2. Typical CE-TDA electropherogram of a PEC fraction obtained with the three steps methodology schematically depicted on the top: () voltage and ( - ) pressure.
off and a −50 mbar mobilization pressure was applied. Note that the mobilization pressure was reversed in step 3 to allow the double detection of the constituents for differential measurement on a single detection point. The Rh,i value was obtained using eq 1:
R h, i =
4kBT (σ2, i 2 − σ1, i 2) πηrc 2(t 2, i − t1, i)
(1)
where kB is the Boltzmann constant; T is the absolute temperature; η is the viscosity, rc2 is the radius of the capillary, σ1,i2 and σ2,i2 are the peak variances of compound i at the first (step 2) and second (step 3) detection, respectively. In eq 1, the experimentally observed second detection time t2,i,app was corrected to take into account the pressure ramps and the operating delays set by the equipment according to
t 2, i = t 2, i ,app − 3
(2)
Figure 3. Charge stoichiometry within PEC fractions as a function of charge titration.
where the times are expressed in seconds. Note that eq 1 is valid when the following equations are fulfilled26 (this was verified in this work):
Di(t 2, i − t1, i) ≥ 1.4 rc 2
(3)
urc ≥ 69 Di
(4)
demonstrates that both CE-TDA and IEC methods gave similar PC/PA molar ratios. It also demonstrates that when PC was added to PA, the PC/PA molar ratio was close to 1. In contrast, when PA was added to PC, complexes were enriched in PC with a decreasing PC/PA molar ratio. The mixing orderdependent behavior can be explained by considering the schematic model of stoichiometric poly(L-lysine citramide)/ poly(L-lysine) complex proposed from consideration of free energies.27,28 In this model, PA/PC complex particles present a core−shell structure where the inner part contains an excess of hydrophobic PC monomer units and with salt counterions for electroneutrality. This leaves the exterior part enriched in hydrophilic PA macromolecules. Accordingly, when PA is
where Di is the diffusion coefficient of solute i and u is the linear velocity of the mobile phase. Di is related to Rh,i via the classical Stokes−Einstein equation. Charge stoichiometry of PEC was obtained by integration of the PA and PC peaks and by using calibration curves. Figure 3 1742
dx.doi.org/10.1021/ac203208k | Anal. Chem. 2012, 84, 1740−1743
Analytical Chemistry
■ ■
added stepwise to PC, the formed complex particles have a PArich outer layer that complex PC macromolecules in excess in solution, thus creating an unbalanced situation in favor of PC when considering the global composition of the particles. In contrast, when PC was added to PA, the PA-rich surface prevents the PA macromolecules in excess in solution from interacting and leaves the global composition stoichiometric, an interpretation in agreement with other experimental data.10,11 Figure 4 displays the PA (bottom) and PC (top) hydrodynamic radii within the PEC vs the corresponding
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected]. Fax: (+33) 4 67 63 10 46.
ACKNOWLEDGMENTS H.C. thanks the Institut Universitaire de France support and the Région Languedoc-Roussillon for the “Chercheur d’Avenir” fellowship. L.L. thanks Dr. Mahfoud Boustta for furnishing the PA.
■
REFERENCES
(1) Fuoss, R. M.; Sadek, H. Science 1949, 110, 552−573. (2) Hsiang, M. W.; Cole, R. D. Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 4852−4856. (3) Vaughan-Thomas, A.; Young, R. D.; Phillips, A. C.; Duance, V. C. J. Biol. Chem. 2001, 276, 5303−5309. (4) Lynn, D. M.; Langer, R. J. Am. Chem. Soc. 2000, 122, 10761− 10768. (5) Katchalsky, A.; Danon, D.; Nevo, A.; De Vries, A. Biochim. Biophys. Acta 1959, 33, 120−138. (6) Kabanov, V. A.; Mustafaev, M. I.; Blokhina, V. D.; Agafieva, V. S. Molek. Biol. 1980, 14, 47−57. (7) Bohidar, H.; Dubin, P. L.; Majhi, P. R.; Tribet, C.; Jaeger, W. Biomacromolecules 2005, 6, 1573−1585. (8) Bekturov, E. A.; Bimendina, L. A. Adv. Polym. Sci. 1981, 41, 99− 147. (9) Tsuchida, E.; Abe, K. Adv. Polym. Sci. 1982, 45, 1−119. (10) Leclercq, L; Boustta, M.; Vert, M. J. Bioact. Comp. Polym. 2011, 26, 3−19. (11) Leclercq, L; Boustta, M.; Vert, M. J. Bioact. Comp. Polym. 2011, 26, 301−316. (12) Harada, A.; Kataoka, K. Science 1999, 283, 65−67. (13) Ruponen, M.; Rönkkö, S.; Honkakoski, P.; Pelkonen, J.; Tammi, M.; Urtti, A. J. Biol. Chem. 2001, 276, 33875−33880. (14) Bronich, T. K.; Nguyen, H. K.; Eisenberg, A.; Kabanov, A. V. J. Am. Chem. Soc. 2000, 122, 8339−8343. (15) Zelikin, A. N.; Trukhanova, E. S.; Putnam, D.; Izumrudov, V. A.; Litmanovich, A. A. J. Am. Chem. Soc. 2003, 125, 13693−13699. (16) Boustta, M.; Leclercq, L.; Vert, M. J. Bioact. Comp. Polym. 2004, 19, 155−171. (17) Taylor, G. Proc. R. Soc. London A 1953, 219, 186−203. (18) Bello, M. S.; Rezzonico, R.; Righetti, P. G. Science 1994, 266, 773−776. (19) Sharma, U.; Gleason, N. J.; Carbeck, J. D. Anal. Chem. 2005, 77, 806−813. (20) Jensen, H; Østergaard, J. J. Am. Chem. Soc. 2010, 132, 4070− 4071. (21) Le Saux, T.; Cottet, H. Anal. Chem. 2008, 80, 1829−1832. (22) Østergaard, J.; Jensen, H. Anal. Chem. 2009, 81, 8644−8648. (23) Frey, B. L.; Corn, R. M. Anal. Chem. 1996, 68, 3187−3193. (24) Boustta, M.; Huguet, J.; Vert, M. Makromol. Chem. Macromol. Symp. 1991, 47, 345−355. (25) Shen, Y.; Smith, R. D. J. Microcolumn Sep. 2000, 12, 135−141. (26) d’Orlyé, F.; Varenne, A.; Gareil, P. J. Chromatogr., A 2008, 1204, 226−232. (27) Vasilevskaya, V. V.; Leclercq, L.; Boustta, M.; Vert, M. Macromolecules 2007, 40, 5934−5940. (28) Krotova, M. K.; Vasilevskaya, V. V.; Leclercq, L.; Boustta, M.; Vert, M.; Khokhlov, A. R. Macromolecules 2009, 42, 7495−7503.
Figure 4. Hydrodynamic radii of the PA (squares, bottom) and PC (triangles, top) within PEC fractions obtained by CE-TDA vs number average molar mass by IEC-SEC. The numbers correspond to the order of precipitation, and S is the supernatant.
number average molar mass Mn determined by SEC, in double logarithmic scale. The experimental results demonstrate that strong molar mass segregation occurred on the titrated polyelectrolyte, in good agreement with previous results.10,11 Very good correlations between Mn and Rh were obtained. The slope is proportional to (1 + a)/3 with a being the Mark− Houwink exponent. For both PA and PC, the slope was close to 0.5, which means the polyelectrolytes were close to theta solvent in the CE-TDA electrolyte at 25 °C.
■
Article
CONCLUSIONS
This work clearly states the possibility to get fast characterization of PEC constituents in a single run using inline CETDA coupling performed in commercial CE apparatus using a single detection point. Within less than 20 min, the charge stoichiometry of PEC and the hydrodynamic radii of the polyelectrolytes can be determined under highly saline conditions (1.3 M) that preserve the PEC dissociation and use a neutrally coated capillary to avoid any adsorption onto the capillary wall. Alternative techniques, such as offline coupling of IEC to SEC require much more time (several hours) and much higher injected quantities and could not be fully automated (dialysis and freeze-drying steps between chromatographic steps). This new methodology was much faster and straightforward. It should greatly contribute to get better and faster characterization of polyelectrolyte complexes for better insight into the physicochemical understanding of PEC. 1743
dx.doi.org/10.1021/ac203208k | Anal. Chem. 2012, 84, 1740−1743
