Grafting of Amino Functional Monomer onto Initiator-Modified

Leonardo Foti , Andre Sionek , Elis Moura Stori , Paula Poli Soares , Miriam Marzall Pereira , Marco Aurélio Krieger , Cesar Liberato Petzhold , Wido...
0 downloads 0 Views 477KB Size
Langmuir 2005, 21, 2209-2217

2209

Grafting of Amino Functional Monomer onto Initiator-Modified Polystyrene Particles Anna Musyanovych*,† and Hans-Ju¨rgen P. Adler‡ Organic Chemistry III, Macromolecular Chemistry, University of Ulm, Albert-Einstein-Allee 11, 89069 Ulm, Germany, and Institute of Macromolecular Chemistry and Textile Chemistry, Dresden University of Technology, Mommsenstr. 4, 01062 Dresden, Germany Received August 14, 2004. In Final Form: December 17, 2004 Polystyrene nanoparticles with grafted chains of an amino functionalized polymer were prepared by a two-step polymerization process. In the first step, the polystyrene seed particles were synthesized by the conventional batch emulsion polymerization using terpolymer HAS (hydroperoxide monomer, acrylic acid, and styrene) as a surface-active initiator. The surface of the obtained particles contains carboxyl groups, which are responsible for the latex stability, and residual undecomposed hydroperoxide groups. Therefore, in the second step, an amino functional monomer was grafted onto the hydroperoxide modified polystyrene particles by a “grafting from” approach. X-ray photoelectron spectroscopy, NMR, and scanning electron microscopy were used to examine the surface of the amino functionalized particles. The amount of incorporated amino groups onto the particles was determined by fluorescenometric titration. In general, the number of amino groups on the particle surface increased with the increase of the functional monomer content in the reaction mixture. The incorporation of the functional monomer was also confirmed by electrophoretic measurements. Final particles possess amphoteric character due to the presence of amino and carboxyl groups on the surface. Adsorption of human immunoglobulins G onto the amino functionalized particles was studied as a function of pH and ionic strength. The covalent binding of human IgG was performed using the glutaraldehyde preactivation method. The immunoreactivity of the latex-IgG complex was examined by the latex agglutination test.

Introduction Polymer dispersions with well-defined particle morphology and colloidal characteristics are of great interest in the biomedical field.1,2 Such particles have been extensively used in the separation and purification of biomaterials,3 drug delivery systems,4-6 clinical assays as carriers of antigens or antibodies,7,8 and so forth. Among all particles the polystyrene ones are most widely used as a result of their low cost and availability. However, the surface of the polystyrene is quite hydrophobic and usually leads to a nonspecific adsorption of biomolecules. To favor the covalent binding of proteins onto the particles surface, they should contain certain functional groups. A number of papers have described the synthesis of the polystyrene latex particles bearing various functional surface groups, for example, carboxyl,9-11 hydroxyl,12 marcapto,13 ac* To whom correspondence should be addressed. E-mail: [email protected]. Tel.: +49 (0)731 50-22896. Fax: +49 (0)731 50-22883. † University of Ulm. ‡ Dresden University of Technology. (1) Arshady, R. Microspheres, Microcapsules & Liposomes, 1st ed.; Citus Books: London, 1999; Vol. 2. (2) Piskin, E.; Tuncel, A.; Denizli, A.; Ayhan, H. J. Biomater. Sci., Polym. Ed. 1994, 5, 451-471. (3) (a) Yoon, J.-Y.; Park, H.-Y.; Kim, J.-H.; Kim, W.-S. J. Colloid Interface Sci. 1996, 177, 613-620. (b) Yoon, J.-Y.; Kim, J.-H.; Kim, W.-S. Colloids Surf., A 1999, 153, 413-419. (c) Yoon, J. Y.; Kim, J. H.; Kim, W. S. Colloids Surf., B 1998, 10, 356-377. (4) Luck, M.; Paulke, B.-R.; Schroder, W.; Blunk, T.; Muller, R. H. J. Biomed. Mater. Res. 1998, 39, 478-485. (5) Kurisawa, M.; Terano, M.; Yui, N. Macromol. Rapid Commun. 1995, 16, 663-666. (6) Yang, S. C.; Ge, H. X.; Hu, Y.; Jiang, X. Q.; Yang, C. Z. Colloid Polym. Sci. 2000, 278, 285-292. (7) Chern, C.-S.; Lee, C.-K.; Chang, C.-J. Colloid Polym. Sci. 2003, 281, 1092-1098. (8) Radomska-Galant, I.; Basinska, T. Biomacromolecules 2003, 4, 1848-1855.

etal,14,15 thymine,16 chlormethyl,14,17,18 amine,19-23 ester,24 and so forth. There are several main methods available for the preparation of the functionalized particles: (i) emulsion or dispersion copolymerization in the presence of functional comonomer(s), (ii) multistep emulsion polymerization in which the polystyrene “core” particle is synthesized in the first stage and the functional monomer is added in the second stage of the polymerization, forming the functionalized polymer “shell” on the “core” particle, and (iii) chemical transformation of the presented func(9) Lee, C.-F.; Young, T.-H.; Huang, Y.-H.; Chiu, W.-Y. Polymer 2000, 41, 8565-8571. (10) Tuncel, A.; Tuncel, M.; Ergun, B.; Alagos, C.; Bahar, T. Colloids Surf., A 2002, 197, 79-94. (11) Reb, P.; Margarit-Puri, K.; Klapper, M.; Mu¨llen, K. Macromolecules 2000, 33, 7718-7723. (12) Tamai, H.; Hasegawa, M.; Suzawa, T. J. Appl. Polym. Sci. 1989, 38, 403-412. Nilsson, K. G. I. J. Immun. Methods 1989, 122, 273-277. (13) Nilsson, K. G. I. J. Immunol. Methods 1989, 122, 273-277. (14) Izquierdo, M. P. S.; Martı´n-Molina, A.; Ramos, J.; Rus, A.; Borque, L.; Forcada, J.; Galisteo-Gonza´lez, F. J. Immunol. Methods 2004, 287, 159-167. (15) Santos, R. M.; Forcada, J. J. Polym. Sci., Part A 1997, 35, 16051610. (16) Dahman, Y.; Puskas, J. E.; Margaritis, A.; Merali, Z.; Cunningham, M. Macromolecules 2003, 36, 2198-2205. (17) Park, J.-G.; Kim, J.-W.; Suh, K.-D. Colloids Surf., A 2001, 191, 193-199. (18) Sarobe, J.; Molina-Bolı´var, J.; Forcada, J.; Galisteo, F.; HidalgoA Ä lvarez, R. Macromolecules 1998, 31, 4282-4287. (19) Cousin, P.; Smith, P. J. Appl. Polym. Sci. 1994, 54, 1631-1641. (20) Ganachaud, F.; Sauzedde, F.; Elaı¨ssari, A.; Pichot, C. J. Appl. Polym. Sci. 1997, 65, 2315-2330. (21) Ganachaud, F.; Mouterde, G.; Elaı¨ssari, A.; Pichot, C. Polym. Adv. Technol. 1995, 6, 480-488. (22) Miraballes-Martı´nez, I.; Forcada, J. J. Polym. Sci., Part A 2000, 38, 4230-4237. (23) Miraballes-Martı´nez, I.; Martı´n-Molina, A.; Galisteo-Gonza´lez, F.; Forcada, J. J. Polym. Sci., Part A 2001, 39, 2929-2936. (24) Nagai, K.; Ohashi, T.; Kaneko, R.; Taniguchi, T. Colloids Surf., A 1999, 153, 133-136.

10.1021/la047960+ CCC: $30.25 © 2005 American Chemical Society Published on Web 02/08/2005

2210

Langmuir, Vol. 21, No. 6, 2005

Musyanovych and Adler

Scheme 1. Schematic Representation of the Synthesis of Amino Functionalized Latex Particles

tional groups into another which enable the covalent binding of proteins. When dealing with the polymer particles for bioapplication it is also important that the particles remain stable in a wide range of pH values and electrolyte concentrations.25 Therefore, amphoteric latex particles possessing both steric and electrostatic stabilization could be of great interest. In our previous paper,26 we described the synthesis of polystyrene latex particles prepared by an emulsion polymerization in the presence of a reactive surfactant. As a reactive surfactant (i.e. inisurf), we used the copolymer of hydroperoxide monomer (5-hydroperoxy-5methyl-1-hexene-3-yne), acrylic acid, and styrene. The hydroperoxide groups worked as an initiator of the polymerization, and the carboxyl groups worked in a salt form that stabilized the particles. The obtained particles are stabilized electrosterically, and the stabilizer molecules are covalently attached to the surface. Employing the iodometric analysis, it was found that after polymerization the residual amount of hydroperoxide groups is present on the particle surface. In this regard, they can generate free radicals and initiate the polymerization of a second monomer. The goal of the current work was to prepare stable amino functionalized latex particles. The synthetic route is based on an amino monomer grafting onto the polystyrene seed particle via the “grafting from” approach (Scheme 1). The advantages of such amino particles are that they have the functional groups for a covalent binding of proteins; they possess both charges, originating from the amino monomer and stabilizer molecules, and the stabilizer molecules cannot be released from the surface. These particles are potentially useful for bioapplications, where the amount of impurities (i.e., desorbed emulsifier molecules) should be as small as possible. In this regard, the Human immunoglobulin G (abbreviated hereafter to IgG) was covalently bonded onto the surface of the aminated particles. The immunoresponse of the obtained latexIgG complexes was examined by the latex agglutination test (LAT), which is frequently used in the diagnosis of infections. (25) Peula, J.; Santos, R.; Forcada, J.; Hidalgo-Alvarez, R.; de las Nieves, F. J. Langmuir 1998, 14, 6377-6384. (26) Musyanovych, A.; Adler, H.-J. P. Langmuir 2003, 19, 96199624.

Experimental Section Chemicals. All chemicals were of analytical grade quality and were used without further purification. The amino functional monomers, 2-aminoethyl methacrylate hydrochloride (AEMH) and N-(3-aminopropyl)methacrylamide hydrochloride (APMH), were supplied by Polysciences, Inc. Fluorescamine, glutaraldehyde solution (50% in water), and Tween 20 [polyoxyethylen(20)-sorbitanmonopalmitat] were supplied by Fluka. HumanIgG and sodium borohydride (NaBH4) were purchased from Sigma and Merck, respectively. The goat anti-human IgG was kindly provided by Dr. B. Neef (Medical Faculty, Institute of Immunology, TUD, Germany). Distilled water was used as a dispersion medium throughout the work. Synthesis of Polystyrene Seed Latex Particles. The polystyrene seed latex particles were synthesized by emulsion polymerization using the terpolymer of the hydroperoxide monomer (5-hydroperoxy-5-methyl-1-hexene-3-yne), acrylic acid, and styrene as an inisurf (abbreviated HAS). HAS is a random terpolymer with a weight-average molecular weight of 46 750 g/mol. The amount of hydroperoxide polymer, poly(acrylic acid), and polystyrene in the terpolymer composition corresponds to 35, 45.1, and 19.9 mol %, respectively. After neutralization of the carboxyl groups of the acrylic acid, HAS becomes water-soluble and shows surface-active properties, decreasing the surface tension on the air-water interface until 41.0 mN/m. The decomposition kinetics of HAS has been studied in the aqueous medium. The homolytical decomposition rate constant (kd) at 85 °C was estimated to be 0.154 × 10-3 s-1. The calculated decomposition half-life time (t1/2) is equal to 75 min. The detailed synthesis procedure, the kinetic study of polymerization, and the characterization of obtained particles were reported in ref 26. Briefly, distilled water (30 g) and an aqueous solution of the terpolymer (6 mL, of 10% solids) were introduced into a reactor. Styrene (4 g) was then added. The mixture was constantly stirred for 3 h at 500 rpm, and the reaction temperature was maintained at 85 °C. The particles produced were monodisperse (Dv/Dn ) 1.017) with a size of 100.2 nm. The number of particles was found to be Np ) 1.017 × 1017 L-1. The amount of the residual undecomposed hydroperoxide groups that are present on the surface after emulsion polymerization is equal to 5.86 × 104 groups/particle (1.86 groups/nm2), which was found by the iodometric titration. “Grafting from” Polymerization. The chemical structures of the amino functionalized monomers that were used for the “graft polymerization” reaction are given in Figure 1. The recipes and reaction conditions for the preparation of the amino functionalized latex particles are presented in Table 1. The chosen amount of AEMH or APMH was dissolved in distilled water, and the pH was adjusted until 8.6 by the addition of 1 M sodium hydroxide aqueous solution. The seed polystyrene latex particles with the initiator (hydroperoxide) groups on the surface and

Grafting of Amino Functional Monomer

Langmuir, Vol. 21, No. 6, 2005 2211

Figure 1. Chemical structures of the amino functionalized monomers. Table 1. Recipe Used To Prepare Amino Functionalized Particles AEMH, g APMH, g distilled water, g seed latex, g

0.0059-0.0177 0.0064-0.0189 33.0 1.0

AEMH or APMH solutions were introduced into the reactor and deaerated by argon gas bubbling at room temperature for 1 h. The stirrer speed was maintained at 350 rpm. Then, the temperature was increased until 75 °C. The reaction was allowed to proceed for a period of 24 h. Characterization of the Amino Functionalized Latex Particles. Before any characterization study, all latexes were cleaned by repetitive centrifugation and redispersion using deionized Milli-Q water. Particle Size and Electrophoretic Mobility. The hydrodynamic particle size measurements were carried out by the photon correlation spectroscopy using a Malvern Zetasizer 3000 (Malvern Instruments, Ltd., U.K.) device at 20 °C. The ionic strength was kept constant by diluting the samples in a 0.001 M potassium chloride solution. A total of 10 measurements were made for each sample, and the mean size and polydispersity values were calculated. The electrophoretic mobility was measured as a function of pH using the same equipment as above. The pH of the latex suspensions was controlled by adding 1 M KOH or HCl aqueous solutions. The samples have been prepared by the dilution of latex (100-200 µL) into 10 mL of the appropriate solution. Three measurements were recorded for each latex sample at each pH. The zeta potential was calculated from the electrophoretic mobility using the Smoluchowsky equation:27

µe )

r0ζ η

where µe is the electrophoretic mobility, η the liquid viscosity, and r and 0 are the permittivity of a vacuum and the relative permittivity of the medium, respectively. Scanning Electron Microscopy (SEM). SEM images were obtained using a DSM 982 Gemini (ZEISS) instrument. A drop of the latex particles diluted in the distilled water was placed on a clean aluminum surface. The samples were dried at room temperature and then in a vacuum. Before measurements, the samples were coated with a thin layer of gold/palladium. X-ray Photoelectron Spectroscopy (XPS). XPS spectra were obtained using an Axis Ultra photoelectron spectrometer (Kratos Analytical, Manchester, U.K.), which is equipped with a monochromatic aluminum KR X-ray source of 300 W at 20 mA. The pressure in the main vacuum chamber during the analysis was typically 10-8 mbar. The kinetic energy of photoelectrons was determined using a hemispherical analyzer with a constant pass energy of 160 eV for wide-scan spectra and 20 eV for highresolution spectra [C(1s), O(1s), and N(1s) regions]. The samples for XPS measurements were prepared as follows. The washed latex particles were dried at room temperature and were manually grinded. The powder of the dry latex particles was placed on a sample holder. After 3 h of storage in the ultrahigh vacuum preparation chamber of the spectrometer, the sample was transferred into the main chamber and then measured. (27) Hunter, R. J. In Fundations of Colloid Science; Hunter, R. J., Ed.; Oxford University Press: Oxford, 1986; Vol. 1, p 557.

Nuclear Magnetic Resonance (NMR). Samples in a dry form were placed into the NMR tube and mixed with dimethylsulfoxided6 (DMSO-d6). The 1H NMR spectra of the dissolved part were recorded with a Bruker DRX 500 spectrometer. Analysis of Amino Groups Content. The concentration of surface amino groups was determined by fluorescence titration according to the method published by Ganachaud et al.20 Briefly, this method is based on the reaction of fluorescamine with primary amino groups, yielding a highly fluorescent product. The measurements were carried out in a MicroMax microwell plate reader, FluoroMax-3 spectrofluorometer (Jobin Yvon, Inc.). The amount of amino groups was measured either in the latex serum (obtained after centrifugation of latex particles) or directly on the latex particles. The calibration curve for serum titration was plotted from given concentrations of the AEMH and APMH solutions in a borate buffer (0.1 M, pH 9.5), ranging from 0.01 to 0.1 g/L. The amount of a 100 µL monomer solution or sample dilution was added to 2.9 mL of borate buffer. Then, 1 mL of fluorescamine solution in acetone (0.3 g/L) was added, and 100 µL of mixture was pipetted into a well of the microtiter plate. The excitation wavelength was 410 nm, and the emission was measured at a wavelength of 470 nm. In the case of latex titration, the calibration curve was obtained by mixing the polystyrene seed latex particles (redispersed in a borate buffer) and a known concentration of the amino monomer. The measurements were performed under the same conditions as described above. The overall content of the surface amino groups on the latex particles was calculated as follows:

[NH2] )

C MFMSC

where C is the functional monomer amount directly on the particle surface, MFM is the molecular weight of the functional monomer, and SC is the experimental solid content of the latex. Physical Adsorption of Human IgG. APMH functionalized latex particles (1 mL of 1.2% solids) were mixed with 5 mg of human IgG dissolved in 3 mL of sodium phosphate buffer (SPB), pH 7.2. The mixture was stirred at 37 °C for 20 h. After the incubation time, the reaction mixture was centrifuged at 15 °C and the unbound IgG remaining in the supernatant was determined by UV spectroscopy (maximum at 750 nm) according to the Lowry et al. method.28 Glutaraldehyde Activation of Amino Latex Particles. The covalent binding of human IgG via the glutaraldehyde activation was performed on the amino functionalized latex particles (pSt-AP1) using the method described in ref 29 with some modifications corresponding to the particles. The aminated latex (2 mL of 1.5% solids) in the SPB, pH 7.2, was placed into a glass to which 0.25 mL of a 0.25 mg/mL aqueous glutaraldehyde solution was added. The latex suspension was then mixed at 37 °C for 12 h. Afterward, the unbound glutaraldehyde was removed from the particle surface by multiple centrifugation/redispersion in SPB. The imine double bonds were reduced by the addition of the sodium borohydride (NaBH4). A total of 35 mg of NaBH4 was dissolved in 1 mL of distilled water and mixed with 3 mL of the latex suspension (solid content 1 wt %) in SPB. The reduction step was carried out at room temperature for 24 h (28) Lowry, O.; Rosenbrough, N.; Farr, A.; Randall, R. J. Biol. Chem. 1951, 193, 265-275. (29) Harlow, E.; Lane, D. P. Antibodies: A Laboratory Manual; Cold Spring Harbor Laboratory: New York, 1988; Chapter 13.

2212

Langmuir, Vol. 21, No. 6, 2005

under gentle stirring. Then, the latex was cleaned by repetitive centrifugation/redispersion in SPB. Covalent Immobilization of Human IgG. Glutaraldehyde treated microspheres (1 mL of 1.2% solids) were mixed with 5 mg of IgG dissolved in 3 mL of SPB, pH 7.2. The mixture was stirred at 37 °C for 20 h. After the incubation time, the reaction mixture was centrifuged at 15 °C and the unbound IgG remaining in the supernatant was determined in the same manner as described above. Physically adsorbed IgG was removed from the surface by mixing microspheres with a 0.1% (w/v) of Tween 20. The mixture was incubated at room temperature for 24 h with a slow shaking. Then, the latex was centrifuged and the amount of desorbed protein was determined in the same manner as above. Reactivity of the Latex-IgG Conjugate. The LAT was employed to determine the immunological response of the latexIgG complex. The test was performed with anti-human IgG. An aliquot of the goat anti-human IgG (20 µL) was placed onto a clean glass slide and mixed with 90 µL of SPB (pH 7.2). The latex-IgG complex (40 µL of 1% solids) was added to the serum dilution and the slide was incubated at 37 °C for 5 min. The results of the agglutination reaction were checked visually.

Results and Discussion After emulsion polymerization the obtained polystyrene seed particles are stable due to the electrosteric repulsion, which is provided by the inisurf polymeric chains that are covalently bonded onto the particle surface and by inisurf carboxyl groups. The concentration of the surface carboxyl groups is 2.65 × 104 groups/particle (0.84 groups/nm2). From the literature, the close-packed parking area occupied by one carboxyl group is around 2 nm.30 Therefore, we can conclude that the surface of the seed particles is not completely covered with the negatively charged carboxyl groups which is an advantage for performing the graft polymerization of the positively charged amino monomer. “Grafting from” Polymerization. Two different amino monomers, namely, AEMH and APMH, were used in this work to prepare particles with amino functionality. In the beginning the acid dissociation constants (pKa) for AEMH and APMH were determined. The values were estimated from the experimental curves obtained after potentiometric titration with the NaOH aqueous solution: pKaAEMH ) 8.5 (in comparison pKaAEMH ) 8.7)31 and pKaAPMH ) 8.3. A series of AEMH- and APMH-grafted polystyrene latexes were prepared with various contents of amino monomers. The amount of AEMH/APMH was systematically varied from 0.017 to 0.056 wt %. It is important to note that we wanted to achieve the grafting of a positively charged amino monomer onto a negatively charged polystyrene seed latex particle. Therefore, it was necessary to maintain the pH of the reaction mixture in the range of 8.2-8.6 during the whole duration of polymerization. Under these pH values, about 50% of amino monomer already exists in a nonsalt form (see pKa). The free Clions that are released from the amino monomer couple with Na+ ions. As a result, the ionic strength of the reaction medium increases. By increasing the ionic strength, the double electric layer on the particle surface is compressed. This leads to the decrease of the particles’ negative charge, therefore, reducing the possibility of the salt bond formation between carboxyl groups on the particles and amino groups from the monomer molecule. The highest concentration of the functional monomer that could be used in the polymerization reaction was (30) Bangs, L. Uniform Latex Particles; Seradyn, Particle Technology Div.: Indianapolis, 1987; p 19. (31) Sauzedde, F.; Ganachaud, F.; Elaı¨ssari, A.; Pichot, C. J. Appl. Polym. Sci. 1997, 65, 2331-2342.

Musyanovych and Adler Table 2. Characteristics of the Latex Particles AEMH sample pSt-seed pSt-AE1 pSt-AE2 pSt-AE3 pSt-AE4 pSt-AE5 pSt-AP1 pSt-AP2 pSt-AP3 pSt-AP4 pSt-AP5

g/L

mmol/L

0.536 0.357 0.250 0.214 0.179

3.214 2.143 1.429 1.286 1.071

APMH g/L

0.571 0.382 0.268 0.229 0.193

mmol/L

Dn

3.214 2.143 1.429 1.286 1.071

100.2 103.1 102.2 102.0 101.7 101.4 102.9 102.3 101.4 101.5 101.3

δ, nm Dv/Dn ζ, mV 2.9 2.0 1.8 1.5 1.4 2.7 2.1 1.4 1.3 1.1

1.017 1.065 1.041 1.037 1.028 1.025 1.050 1.055 1.043 1.031 1.027

-58.1 -17.4 -25.7 -27.5 -30.1 -35.4 -18.1 -27.0 -29.3 -34.8 -40.9

found to be 0.536 g/L (0.177 g/glatex) and 0.571 g/L (0.189 g/glatex) for AEMH and APMH, respectively. With increased concentrations, a coagulation of latex was observed. The coagulation effect can be explained by two facts. First, the high hydrophilic character of the amino functionalized polyelectrolytes which are formed during the graft polymerization in the continuous phase might course the bridging flocculation of the latex particles. Second, the high loading concentration of the amino monomer leads to the formation of the polyamine functionalized shell on the particle surface. At that moment, the overall charge of the particle changes from the negative (carboxyl groups) to positive (ammonium groups). Because this process is relatively slow and not homogeneous within the system, the repulsive electrostatic force between particles becomes weak and coagulation takes place. The average particle size Dn, particle size distribution Dv/Dn, thickness of the polymer layer δ obtained after graft polymerization, and zeta potential ζ (measured at pH 6.8) are reported in Table 2 for both monomers. The results indicate that after graft polymerization the aminated particles have a higher degree of size distribution than the polystyrene seed latex particles. It was found that the final size of the particles slightly increased upon the increase of the amino monomer content used in the polymerization. The thickness of the amino polymer layer was calculated as a difference between the final size and the size of the seed polystyrene latex. For both monomers, the obtained values were in the range between 1 and 3 nm, which is close to the sensitivity of the technique. In fact, with such a thickness of second polymer, the morphology of final particles should be “core-shell”. In our case, we were not observing this effect, because as it can be seen from the zeta potential measurements, the final particles are still charged negatively due to the presence of surface carboxyl groups from the inisurfs’ molecules. SEM was employed to examine the surface morphology of the amino functionalized particles. With a low initial concentration of AEMH and APMH, the surface of the final particles was still relatively smooth. When the amount of AEMH and APMH was increased to 0.536 g/L and 0.571 g/L, respectively, the surface of the particles became noticeably rougher (Figure 2). The surface roughness is more pronounced in the case of AEMH functionalized latex particles. We speculate that this is due to a higher reactivity of the methacrylate group compared to the methacrylamide. Comparing the structures of both monomers, it can be noticed that the carbonyl in the ester group (AEMH) is more electronegative compared to the carbonyl in the amide group (APMH). Therefore, the double bond in the case of AEMH will be more electronpoor and more reactive to the radical attack. XPS and 1H NMR. The presence of the amino functionality on the surface layer of the latex particles was

Grafting of Amino Functional Monomer

Langmuir, Vol. 21, No. 6, 2005 2213

Figure 2. SEM images of the amino functionalized particles: (a) pSt-AE1 (CAEMH ) 0.536 g/L) and (b) pSt-AP1 (CAPMH ) 0.571 g/L).

Figure 4. C(1s) core-line spectra of (a) non-functionalized polystyrene latex particles and (b) functionalized latex particles after graft polymerization of the aminated polymer (latex pStAP1).

Figure 3. XPS wide-scan spectra recorded for (a) nonfunctionalized polystyrene seed latex particles (pSt-seed) and (b) functionalized latex particles after graft polymerization of the aminated polymer (latex pSt-AP1).

shown using the XPS analysis and 1H NMR. In Figure 3 the XPS wide-scan spectra of the pure polystyrene seed latex (pSt-seed) and for one selected amino functionalized latex (pSt-AP1) are presented. For the pure polystyrene seed latex (see Figure 3a), the main contributions to the spectrum are the C(1s) peak at 284.7 eV and O(1s) peak at 533 eV, but there is also a low-intensity signal at 1071.0 eV corresponding to the Na(1s) core level. Figure 3b shows the XPS wide scan of the APMH functionalized latex particles prepared with 0.571 g/L APMH. In comparison with the polystyrene latex, the presence of the N(1s) peak at 400.0 eV is observed. This confirms the presence of amino functionality at the particle surface. Moreover, the presence of the Na(1s) peak at 1071.1 eV indicates the presence of carboxyl groups in a salt form on the particle surface, which is originated from the inisurf molecule. This information provides further evidence about the amphoteric nature of the final latex particles. The presence of amino functionality can be also confirmed by a closer inspection of the C(1s) core level spectra shown in Figure 4.

For the polystyrene seed particles, the C(1s) peak envelope shows the presence of three carbon species. They can be attributed to CdC, CsC, and C-H (284.7-285.0 eV); CsO (286.0 eV); and CdO (287.0 eV). For the APMH functionalized particles, the presence of another carbon is detected at 285.7 eV attributed to CsN structures. The presence of shake-up signals at a binding energy higher than 291.0 eV in both C(1s) spectra is indicative of the polystyrene aromatic rings and is associated with a lowenergy π-π* electron transfer.32 The presence of the amino polymer on the particle surface was also confirmed by 1H NMR spectroscopy. The method has been established for analyzing the surface of copolymerized material and was recently described.33 The principle is based on dispersing the dried functionalized latex particles in DMSO-d6. Because polystyrene-based cross-linked particles are poorly soluble in DMSO, the protons from the seed are not detected. Therefore, only protons from the amino functionalized macromolecules located in the surface layer of the particle can be determined due to their motional freedom. 1H NMR spectra for the AEMH monomer, polystyrene seed (pSt-seed), and AEMH functionalized (pSt-AE1) latexes are presented in Figure 5. In comparison with the spectrum of polystyrene latex particles before graft polymerization, the presence of characteristic resonance peaks belonging to the AEMH functional polymer are clearly observed in the spectrum of amino functionalized latex. Due to the addition reaction (32) Briggs, D.; Seah, M. P. Practical Surface Chemistry; Wiley: Chichester, 1990; Vol. 1. (33) Charreyre, M.-T.; Razafindrakoto, V.; Veron, L.; Delair, T.; Pichot, C. Macromol. Chem. Phys. 1994, 195, 2153-2167.

2214

Langmuir, Vol. 21, No. 6, 2005

Musyanovych and Adler

Figure 5. 1H NMR spectra in DMSO-d6 of (a) AEMH functionalized particles (pSt-AE1), (b) seed latex particles before functionalization (pSt-seed), and (c) AEMH monomer.

to the monomer double bond, the signal of the methyl group shifted from 1.9 ppm to the region 0.7-1.3 ppm. The signal is close to the signal of CH3 groups from the inisurf molecule (hydroperoxide polymer). Two weak signals located at 5.8 ppm represent hydrogen protons near the polymerizable bond of the amino monomer, which is grafted onto the particle surface. In conclusion, it appears that both analytical methods (XPS and 1H NMR) permit the presence of the amino functionality onto the particle surface to be shown. However, it was not possible to interpret the obtained data in a quantitative way due to the sensitivity limit of these techniques and a low concentration of the amino polymer on the latex particles. Fluorescenometric Titration. The fluorescenometric titration was performed to determine quantitatively the content of primary amino groups that are actually presented on the surface. Different latexes were measured according to the procedure described in the experimental section. This method allows determining the available surface amino groups from direct measurements either on the particle surface or on the residual monomer that present in the latex serum after centrifugation. In the last case, the concentration was determined from the difference between the initial amount of functionalized monomer and the amount obtained from fluorescenometric titration. Table 3 summarizes the results of amino group densities obtained from both analyses and the grafting yields of the reactions. The surface amino groups’ density was calculated from the particle size and the value of amino groups’ amount obtained from the direct titration of latex particles. It is notable that the amount of surface amino groups increases with increasing of the functional monomer concentration.

Table 3. Results of Fluorescence Titration of the Amino Functionalized Particles surface -NH2 density sample

[-NH2],a µmol/glatex

[-NH2],b µmol/glatex

groups/ nm2

×103 groups/ particle

grafting yield, %

pSt-AE1 pSt-AE2 pSt-AE3 pSt-AE4 pSt-AE5 pSt-AP1 pSt-AP2 pSt-AP3 pSt-AP4 pSt-AP5

16.43 13.48 11.08 9.07 5.89 15.88 11.85 8.30 5.95 3.92

17.22 14.37 11.91 8.75 5.37 16.52 12.41 8.01 5.73 4.30

0.15 0.18 0.10 0.08 0.05 0.14 0.11 0.07 0.05 0.04

4.78 3.92 3.22 2.64 1.71 4.62 3.45 2.42 1.73 1.14

14.7 18.8 22.3 20.3 15.5 14.2 15.9 16.7 13.3 10.5

a,b The amount of amino monomer calculated from the direct titration of latex particles (a) and calculated from the values obtained from the serum titration (b).

The yields for the grafted amino polymer varied between 10 and 22 wt %. These values were calculated from the ratio between the amount of amino groups obtained from the latex titration and the total amount of monomer introduced into the reactor. The aqueous continuous phase contains about 60-75 wt % of the amino polymers and 10-15% of the amino monomer. The values were calculated from the 1H NMR spectra performed in the deuterium oxide by the comparison of the signal integrals of monomer double bond (5.8 and 6.3 ppm) and NH3+ groups (8.5 ppm). Although the grafting efficiency is relatively low, it is in agreement with a high hydrophilic character31 and small initial concentration of the functionalized monomers. Moreover, both amino-containing monomers (in the hydrochloride form) are acting as transfer agents.34,35 This fact also explains the formation of the low molecular weight

Grafting of Amino Functional Monomer

Figure 6. Zeta potential versus pH of the latex particles before and after amino functionalization. The ionic strength was kept constant at 0.001 mol/L KCl.

water-soluble polyelectrolytes and short chain length copolymers grafted onto the surface of the seed latex particle. In the case of the reactions carried out with the APMH functional monomer, the amount of available amino groups and the percentage of functionalization are lower compared to the values obtained in the AEMH based reactions. The amount of amino groups on the particle surface was a factor of ≈10 less than that for hydroperoxide groups before graft polymerization. The reason for that is probably due to the location of the hydroperoxide groups. For example, it is difficult for the amino monomer to reach the hydroperoxide groups that are very near to the particle surface or if the hydroperoxide groups are surrounded by the negatively charged carboxyl groups, when the electrostatic attraction is predominant. In the end of the graft polymerization, the residual hydroperoxide groups were completely decomposed, which was proven by differential scanning calorimetry measurements. Electrokinetic Measurements. Electrophoresis is a convenient method to characterize the surface properties of polymer particles. The colloidal stability of the amino functionalized particles was analyzed in terms of the ζ potential of diluted samples. As it was mentioned in the introduction, no additional stabilizer was used during the graft polymerization. The presence of different charged groups on the surface, provided by the incorporated inisurf molecules and grafted amino polymer, determines the colloidal stability of the particles. The carboxylic groups originated from the inisurf molecules are closer to the particle surface compared to the ammonium groups, which are attached to the hydroperoxide groups through the polymer spacer. In addition, the charge distribution within the “hairy” layer of the amino functionalized particle is not absolutely homogeneous, taking into account the random structure of the inisurf molecule and the location of the grafting center (negative-poor charge region). The electrophoretic mobility of aminated latexes was measured as a function of pH in a 0.001 mol/L KCl solution at 20 °C and then converted into the ζ potential using Smoluchowski equation. As it was expected, the colloidal stability of the amphoteric latex strongly depends on the pH, because of the presence on the surface ionic carboxyl and amino groups that have the opposite charges. The evolution of the ζ potential as a function of pH is plotted in Figure 6 for AEMH/APMH functionalized latex particles (34) Duracher, D.; Sauzzede, F.; Elaı¨ssari, A.; Perrin, A.; Pichot, C. Colloid Polym. Sci. 1998, 276, 219-231. (35) Rossi, S.; Lorenzo-Ferreira, C.; Battistoni, J.; Elaı¨ssari, A.; Pichot, C.; Delair, T. Colloid Polym. Sci. 2004, 282, 215-222.

Langmuir, Vol. 21, No. 6, 2005 2215

Figure 7. Zeta potential of APMH functionalized latex particles as a function of pH. The ionic strength was kept constant at 0.001 mol/L KCl.

containing the highest amount of grafted polymer and polystyrene seed latex particles (pSt-seed). For both types of amino functionalized latexes, the value of the ζ potential exhibits a positive sign in the region of pH 1.5-4.0, therefore, proving the presence of the labile cationic charges on the particle surface. In the acidic region of pH, the predominant amount of the amino groups is converted to ammonium cations, whereas the carboxyl groups are partially dissociated. Both AEMH and APMH functionalized latexes exhibit the isoelectric point (IEP), which is located arround pH 4. This point also corresponds to the point of zero charge, where the amount of positive charge (NH3+) is equal to the number of negative charges (COO-). According to Homola and James,36 when the ratio of acid to amine is 5.5:1, the IEP is located near pH 4. That is in a good agreement with our data, obtained from the amount of carboxyl and amino groups, that is, [COO-]/ [NH3+] ) 0.84/0.15 ) 5.6:1. As the pH increased above 4, the surface of the particles has a negative charge. In this region the amino groups are still positively charged, but the overall surface charge density is negative due to the presence of excess anionic surface charges that are originated from the surface carboxyl groups. In contrast, the surface charge of the polystyrene seed latex particles shows only a small change with pH increase, and it is due to the presence of acid (carboxyl) surface groups in particular. In the absence of functionalized monomers, the seed latex bearing only carboxyl groups on the surface exhibits no IEP in the range of pH 1.8-11. The amount of amino groups on the surface has also the influence on the ζ potential, because as the amino monomer content increases, more amino groups are present on the surface of the final particles. As an example, Figure 7 shows the influence of the initial APMH concentration on the surface potential. It can be seen that the ζ-pH behavior shows a series of similar shaped curves along the pH axis. The IEP is located in the range of 2.34.0 depending on the ratio of carboxyl to amine groups. A slight shift of the IEP toward higher pH was observed with an increase in the concentration of the amino groups on the particle surface. Next, we studied the influence of the ionic strength on the ζ potential at different pH values. Figure 8 presents the ζ potential versus pH dependencies for the pSt-AP1 (IEP ) 3.9) measured in a various KCl solutions, that is, 0.001, 0.01, 0.1, and 0.5 mol/L. One can see that the ζ (36) Homola, A.; James, R. J. Colloid Interface Sci. 1976, 59, 123133.

2216

Langmuir, Vol. 21, No. 6, 2005

Musyanovych and Adler

Figure 8. Zeta potential of APMH functionalized latex particles as a function of pH at different ionic strengths.

potential of the particles decreases with an increase in the ionic strength. However, all curves have the location of IEP in the same region of pH 4. This trend confirms that the concentration of the surface charge groups is not changing and these groups are strongly anchored to the surface.36 Although the ζ potential decreases with the increase in ionic strength, the latex particles were completely stable. According to the Derjaguin, Landau, Verwey, and Overbeek theory, the addition of electrolytes should cause the particle aggregation due to the suppression of the electric double layer. This disagreement with a theory can be attributed to the hydration of the adsorbed cations on the hydrophilic particle surface at the high values of pH. The existence of such hydration forces was observed by several other authors.37-40 Adsorption and Covalent Binding of the Proteins. In the present part of our research we studied the latexprotein interactions. It is known that the first contact between the protein and the surface is based on the physical nature. Already after some minutes of the reaction the physically adsorbed protein achieves saturation. Physical or passive adsorption is mediated by (i) hydrophobic interactions between amino acid residues of the protein molecules and the polystyrene surface of the microspheres, (ii) hydrogen bond formation between carboxyl and amino groups, and (iii) electrostatic interactions between the particle and the protein molecule. The physical adsorption of human IgG was studied as a function of pH and ionic strength of the reaction medium. The obtained results are presented in Figure 9. The highest amount of physically adsorbed IgG onto the latex particles pSt-AP1 is obtained at pH 5.8 for both ionic strengths (0.02 and 0.1 M). This is because at pH 5.8 the overall charge of the protein is positive whereas the latex particle has a negative charge. The adsorption takes place at a low pH mainly due to the electrostatic interactions between the antibody and the latex particles. At pH 7.2, which is near the IEP of the human IgG, the influence of the electrostatic interactions is reduced, and the adsorbed amount of antibody decreases. A further increase in the pH toward the basic values causes the two reactants (particle and antibody molecule) to exhibit the same charge. As a result the adsorbed amount of IgG decreases (37) Israelachvili, J. N. Intermolecular and Surface Forces; Academic Press: New York, 1992. (38) Pashley, R. M.; Israelachvili, J. N. J. Colloid Interface Sci. 1984, 97, 446-455. (39) Valle-Delgado, J. J.; Molina-Bolı´var, J. A.; Galisteo-Gonza´lez, F.; Ga´lvez-Ruiz, M. J. Colloid Polym. Sci. 2003, 281, 708-715. (40) Molina-Bolı´var, J. A.; Galisteo-Gonza´lez, F.; Hidalgo-A Ä lvarez, R. J. Colloid Interface Sci. 1998, 206, 518-526.

Figure 9. Effect of the pH and ionic strength on the physical adsorption of human IgG. Table 4. Amounts of IgG Adsorbed and Covalently Bonded onto Different Latex Particles at Various pH and Temperatures mg/m2

pSt-AP1-Gl

pSt-AP1

pH 5.8

total adsorption, latex covalently bonded, mg/m2latex

3.42 1.81

3.22 0.11

pH 7.2

total adsorption, mg/m2latex covalently bonded, mg/m2latex

3.15 1.76

3.07 0.10

pH 8.2

total adsorption, mg/m2latex covalently bonded, mg/m2latex

2.71 1.69

2.64 0.13

down to 2.71 mg/m2 of latex. When the protein has the same charge sign as the particle, the adsorption is mainly caused by the hydrophobic interactions. The increase of the ionic strength from 0.02 up to 0.1 M leads to the decrease of the charge, and the repulsion force between the particle surface and IgG molecule becomes weaker, thus, allowing more IgG to adsorb onto the surface at pH 8.2 compared to pH 5.8. The covalent binding between the proteins and amino particles was achieved through a preactivation step with glutaraldehyde. The covalent binding takes place via the formation of the chemical bonds between the surface aldehyde groups of the latex particles and the amino groups of lysine (one of the amino acids which is present in the protein molecule). To avoid the formation of agglomerates during the activation step, the reaction was carried out with a diluted latex sample and in the presence of an excess amount of the glutaraldehyde. The glutaraldehyde groups react with the amino groups, leading to the Schiff base linkage formation. This bond is quite unstable and can be easily hydrolyzed, giving free amine and aldehyde groups.15 To overcome this drawback, the double bonds were reduced to a single bond by adding a reducing agent, that is, sodium borohydride (NaBH4). A major problem when working with active groups that are capable of coupling the protein covalently is the elucidation of the amount of chemically or physically attached protein. Therefore, it is necessary to carry out a desorption process to remove the adsorbed proteins and to ensure that the proteins are covalently bonded. This was achieved by the treatment of the latex-protein complex with a nonionic emulsifier (Tween 20).41 The emulsifier will force desorption of the physically attached protein. (41) Rapoza, R. J.; Horbett, T. A. J. Colloid Interface Sci. 1990, 136, 480-493.

Grafting of Amino Functional Monomer

Langmuir, Vol. 21, No. 6, 2005 2217

to the latex particle are recognized by corresponded antigens and, therefore, confirms a good performance of the immobilization process. The same LAT was performed after 7 months of storage the latex-IgG at +4 °C. The results were similar to those obtained with the freshly prepared latex-IgG complex, confirming the high stability and the absence of the IgG desorption from the particle surface. Figure 10. Photographs of (a) the latex-IgG complex reacted with goat anti-human IgG serum and (b) the same complex mixed with a normal human serum.

The covalent binding of IgG (IEP ) 7.0) onto the latex particles was performed at 37 °C and in various buffer solutions as a function of pH. The ionic strength was kept constant in all runs at 0.02 M. For comparison, the experiments on the binding of antibodies were performed with the nonactivated latex particles (pSt-AP1). The obtained total amount of IgG (physically and chemically anchored) and the amount that is bonded covalently onto the particle surface are summarized in Table 4. From the presented results, it can be concluded that the amount of physically adsorbed antibody increases with the decrease of pH value. In contrast, the covalent binding of IgG onto the latex particle does not highly depend on the pH. The obtained results prove that more than 60% of the total immobilized antibodies are chemically attached to the surface of the particles. Immunoreactivity. The activities of IgGs that are covalently attached to the surface amino groups were examined in a LAT. The principle of LAT is that the latex particles coated with a given antibody are mixed with the analyzing sample (blood, saliva, etc.). If the corresponding antigen is present in the sample, the latex particles agglutinate (clump together). Aminated particles (pSt-AP1-Gl) with the highest amount of covalently attached human IgG (at pH 5.8) were used as a solid support for determination of the goat anti-human IgG. As it can be seen in Figure 10a, the latexIgG complex exhibited a specific agglutination with an anti-human IgG, whereas no agglutination occurred when the particles were mixed with a normal human serum (Figure 10b). This indicates that the antibodies attached

Conclusions A series of polyAEMH/APMH-grafted polystyrene latexes were successfully synthesized by a two-step polymerization process. Different concentrations of both amino monomers were examined. The content of amino groups on the particle surface could be controlled by the initial amount of amino monomer used in the polymerization. The highest amount of the monomer that can be grafted onto the particle surface and retain the system stability corresponds to 0.536 and 0.571 g/L for AEMH and APMH, respectively. Characterization of the final latexes by XPS, NMR, and electrophoretic measurements confirmed the presence of the amino functionality on the surface. The measurements of the zeta potential of the latexes showed that the amino functionalized particles exhibit the amphoteric character. They display an increase in colloidal stability at low pH attributed to the grafted amino polymer that stabilizes the particle through ammonium groups, and at high pH they become negatively charged due to the ionized carboxyl groups. It has been shown that the adsorption of the proteins (i.e., human IgG) onto the amino functionalized particles could be controlled by the pH and the ionic strength of the reaction medium. The high degree of the covalently bonded protein molecules onto the surface and the fact that the particles are stabilized by the covalently attached surfactant molecules make these microspheres attractive in biomedical applications. Acknowledgment. The authors are grateful to Prof. S. Voronov, Dr. O. Budishevska, and Dr. V. Samaryk for helpful discussions. We also thank Dr. F. Simon, Mrs. A. Rudolf, and Mrs. E. Kern for providing assistance in XPS, NMR, and SEM measurements. This work was financially supported by the Deutsche Forschungsgemeinschaft, Project No. A1 - SFB 287 “Reactive Polymers”. LA047960+