Multiepitope Templates Imprinted Particles for the Simultaneous

May 17, 2016 - To achieve the simultaneous capture of various target proteins, the multiepitope templates imprinted particles were developed by phase ...
0 downloads 0 Views 3MB Size
Download PDF
Letter pubs.acs.org/ac

Multiepitope Templates Imprinted Particles for the Simultaneous Capture of Various Target Proteins Kaiguang Yang,† Senwu Li,†,‡ Jianxi Liu,†,‡ Lukuan Liu,†,‡ Lihua Zhang,*,† and Yukui Zhang† †

Key Lab of Separation Sciences for Analytical Chemistry, National Chromatographic R&A Center, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 457 Zhongshan Road, Dalian, 116023, China ‡ University of Chinese Academy of Sciences, Beijing, 100049, China S Supporting Information *

ABSTRACT: To achieve the simultaneous capture of various target proteins, the multiepitope templates imprinted particles were developed by phase inversion-based poly(ether sulfone) (PES) self-assembly. Herein, with the top three high-abundance proteins in the human plasma, serum albumin, immunoglobulin G, and transferrin, as the target proteins, their N-terminal peptides were synthesized as the epitope templates. After the preorganization of three epitopes and PES in dimethylacetamide, the multiepitope templates imprinted particles were formed in water through self-assembly, by which the simultaneous recognition of three target proteins in human plasma was achieved with high selectivity. Furthermore, the binding kinetics study proved that the adsorption mechanism in this imprinting system toward three epitope templates was the same as that on the single-epitope imprinting polymer. These results demonstrate that our proposed multiepitope templates imprinting strategy might open a new era of artificial antibodies to achieve the recognition of various targets simultaneously.

I

protein, epitopes have been successfully employed as templates to fabricate the recognition sites for several proteins, including albumin,20,23 immunoglobulin G,24 cytochrome c,20 HIV-1related proteins (glycoprotein 41 and anti-HIV-1),25,26 and anthrax protective antigen.27 Also, multiple glycans have been employed as the templates to mimic lectins for the recognition of an intact glycoprotein, in which glycans functioned very similar to epitopes, though they were not termed as epitopes.15 Compared with protein templates, epitope templates are much more robust and stable during the imprinting process. Moreover, epitope templates could be obtained in large scale with low cost, even for the rare proteins.28,29 Therefore, the choice of polymerization formats for epitope imprinting is more flexible. Polymer self-assembly provides novel and facile tools for a tailored, bottom-up approach to design organized materials through noncovalent bonds.30 In the present work, poly(ether sulfone) (PES) self-assembly was achieved through the phase inversion with N,N-dimethylacetamide (DMAc) as the dispersion medium.31 As an organic solvent with strong polarity, DMAc favors the homogeneity of PES and facilitates the solubility of epitopes with different physical and chemical properties. This would promote the interaction between the

n recent years, the simultaneous capture of various proteins has played a crucial role in precise diagnosis and highthroughput analysis.1,2 Up until now, many efforts have been made to accomplish these tasks,3,4 among which immunoassays with different antibody-modified particles have gained popularity.5 However, theoretical and experimental evidence have proved that the nonequilibration of the recognition sites in these physically mixed particles would lead to the loss of the selectivity.6−8 Meanwhile, such multiple component immunoaffinity materials are increasingly controversial due to its high cost, delicate application condition and high nonspecific binding.9,10 Besides, for some important proteins, such as acetylated proteins and hydroxylated proteins, the acquiration of antibodies is of great challenging.11,12 As artificial antibodies, protein-imprinted materials offer many advantages, such as high selectivity, potential applicability to all proteins, long-term storage stability, and resistance to harsh environments.13−19 However, two hurdles impede their practical use in the simultaneous capture of multiple proteins: (1) unavailability of rare target proteins in pure and stable forms for use as templates and (2) scarcity of universal polymerization system for different proteins imprinting. Epitopes are special regions within the structure of a globular protein that can be classified as the antigenic determinant, i.e., the epitope’s structural specificity usually represents the entire protein.20−22 Moreover, they can be easily obtained through artificial synthesis. Currently, as an alternative to the target © XXXX American Chemical Society

Received: March 30, 2016 Accepted: May 13, 2016

A

DOI: 10.1021/acs.analchem.6b01247 Anal. Chem. XXXX, XXX, XXX−XXX

Letter

Analytical Chemistry S(O)2 groups on the free PES chains and the amino/ hydroxyl groups on the epitopes through electronic conjugation and hydrogen bond interaction.32,33 Herein, we attempted to develop a universal method to prepare multiepitope templates imprinted particles through phase inversion-based PES self-assembly to achieve the simultaneous capture of multiple proteins from human plasma. Three proteins, human serum albumin (HSA, Mw = 66 kDa, pI = 4.7−5.5), immunoglobulin G (IgG, Mw = 150 kDa, pI = 8.6), and transferrin (TRF, Mw = 77 kDa, pI = 5.6−6.6) were selected as the target proteins, based on their respective high abundance in plasma and their obviously different physical and chemical properties.34 The N-terminal peptides of these three proteins, MKWVTFISL (E HSA , N-terminal for HSA), ASTKGPSVF (EIgG, N-terminal for IgG), and MRLAVGALL (ETRF, N-terminal for TRF) were synthesized as the epitope templates. As shown in Scheme 1, PES (300 mg), EHSA (42 Scheme 1. Fabrication of Multiepitope Templates Imprinted Particles via PES Self-Assembly and Application in the Simultaneous Capture of Various Target Proteins

Figure 1. SEM photographs of the cross sections of MIPs: entire particle (a), skin layer (b), interpenetrated channel (c), and porous internal structure (d).

separation occurred at the inner surface when a high concentration of solvent was used.31 A finger-like structure is observed under the outer skin (Figure 1b), which demonstrates that the instantaneous phase separation starts from the outer surface.35 The inside channels, with diameters ranging from 1 to 20 μm, interpenetrate each other, which would favor mass transfer inside the particles (Figure 1c). Furthermore, many small pores are observed inside the microspheres, which could facilitate their application as adsorbent materials (Figure 1d). All of these results suggested that the PES transformation from the polymer solution to the solid state occurred quickly in the water, followed by the instantaneous precipitation of PES.31 A similar hierarchical pore structure and intact microsphere morphologies were also found in the NIPs, as shown in Figures S-1a-d and S-2a−d in the Supporting Information. According to the drying loss measurements, the porosities of the MIPs and NIPs were 81.51% and 78.23%, respectively.32,33 Clearly, no significant morphological variations could be observed between the MIPs and NIPs. It was thus concluded that the addition of templates did not affect the phase inversion process in terms of morphology. To evaluate the recognition abilities of multiepitope templates imprinted particles, binding experiments were conducted separately in EHSA, EIgG, and ETRF aqueous solutions with the initial epitope concentration of 0.25 mg/mL at 20 °C. After incubation at different time intervals, the concentration of the residual epitopes was measured via MALDI-TOF MS, with an artificial synthesized peptide (APGDRIYVHPF, 25 μg/mL, Mw = 1271.4) as the internal standard. As shown in Figure 2a1, b-1, c-1, the saturated adsorption amounts of MIPs ([S](imprinted)) to EHSA, EIgG, and ETRF are 12.00, 45.00, and 12.89 mg/g, respectively, while those of NIPs ([S](nonimprinted)) are 5.24, 4.78, and 1.72 mg/g, respectively. When the value of the recognition factor (α), the ratio of [S](imprinted) to [S](nonimprinted) is larger than 1 and the imprinting process could be considered successful. Here, the αHSA, αIgG, and αTRF were calculated as 2.29, 9.41, and 7.54, respectively, demonstrating that multiple epitopes were simultaneously imprinted on the PES matrix. The low amount of NIP

mg), EIgG (10 mg), and ETRF (4 mg) were dispersed in DMAc (1200 mg), with the mass ratio of epitopes the same as that of the target proteins in human plasma.34 The resulting polymer solution was added dropwise to distilled water at room temperature to fabricate the multiepitope templates imprinted particles (MIPs) via phase inversion. Here, water was selected as the nonsolvent because it was miscible with DMAc but immiscible with PES.32 Therefore, phase inversion occurred by the rapid exchange of DMAc and water. Subsequently, these particles were incubated in water to remove the residual DMAc. Finally, the templates were removed by washing with methanol/acetic acid (v:v = 9:1) at 40 °C until no epitopes could be detected at m/z = 1124, 893, and 943 by matrixassisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS). Simultaneously, nonimprinted particles (NIPs) were prepared with a 20 wt % PES solution using the same protocol but without the addition of the templates. Scanning electron micrographs (SEMs) were acquired for the cross sections of the prepared MIPs. As shown in Figure 1, macrovoids are distributed throughout the porous PES microspheres (Figure 1a), indicating that a delayed phase B

DOI: 10.1021/acs.analchem.6b01247 Anal. Chem. XXXX, XXX, XXX−XXX

Letter

Analytical Chemistry

where Qe and Qt (mg g−1) are the final equilibrium adsorption capacity and the actual adsorption capacity at time t (h), respectively; v0 (mg g−1 h−1) is the initial adsorption rate; and k2 (g mg−1 h−1) represents the rate constant of second-order adsorption. As shown in Figure 2a-2, b-2, c-2 and Table S-1, the secondorder kinetic model fits all types of template epitope adsorption data well according to the relatively high correlation factors (r = 0.965, 0.998, and 0.987, for EHSA, EIgG, and ETRF). This indicated that (1) the rate-limiting step occurred by virtue of chemical adsorption performed through electronic conjugation or hydrogen bond interaction suggesting, in turn, that the affinity of MIP played a role between adsorbent and adsorbate,36 and (2) the adsorption capacity was proportional to the number of active recognition sites on the imprinted PES particles. In contrast, the adsorption on NIPs did not fit the second-order kinetic model well because the correlation factors were relatively low (r = 0.961, 0.907, and 0.772, for the EHSA, EIgG, and ETRF, respectively, as shown in the Supporting Information, Figure S-3). This indicated that the adsorption mechanisms of MIPs were different from those of NIPs. Moreover, in single-protein imprinting, it has been reported that the adsorption kinetics also followed the Lagergren model,36 further illustrating that different templates led to their respective recognition sites in this imprinting system. The equilibrium adsorption was carried out with the various epitope concentrations from 50 to 400 μg mL−1. As shown in Figure S4, with the increase of each epitope concentration, the adsorption capacity of MIP was higher than that of NIPs, which indicated that the imprinted sites promoted the high specific affinity toward their respective targets. The selectivity test was carried out with three standard target proteins (HSA, IgG, and TRF) and interfering protein (Cytochrome c). The MIPs show obvious recognition ability toward the target proteins, HSA and TRF (Figure S-5). Plasma, the liquid portion of blood, can be considered as the most important and complicated clinical biosample for monitoring physiological, pathological, and pharmacological states.34 Despite the current multiple component immunoaffinities, such as ProteoPrep20 and the Seppro IgY14 kit, that were supposed to deplete multiple high-abundance proteins, the limited selectivity and highly cost-consuming properties make them very controversial for the practical application in the real sample.9,10 Herein, we applied our prepared multiepitope templates imprinted particles to achieve the capture of the three highest abundance proteins from human plasma with high selectivity. The MIPs and NIPs were separately incubated with 100× diluted native and denatured human plasma each. Then, the proteins on the particles and in the supernatant were collected, followed by a typical shotgun proteome analysis protocol.37 Finally, a label-free quantitative method, spectral counting,38 was used to quantify the captured protein amount in this study. As shown in Figure 3a and Table S-2, with the native human plasma as the sample, the MIPs could recognize the target proteins, HSA (P1), IgG (P2), and TRF (P3), simultaneously with obviously greater binding amounts than the NIPs. For other high abundance proteins, such as KRT1 keratin type II cytoskeletal 1 (P4), macroglobulin (P5), fibrinogen-alfa (P6), and alpha-2-glycoprotein (P7), the binding amounts on the MIPs were similar to those on the NIPs, due to the nonspecific adsorption. For proteins such as alpha-1-antitrypsin (P8), complement C3 (P9), and apolipoprotein A-IV (P10), their

Figure 2. Time course of the binding of MIPs (◊) and NIPs (□) toward EHSA (a-1), EIgG (b-1), and ETRF (c-1); and curves of the second-order rate equation of the Lagergren kinetic model toward the adsorption of EHSA (a-2), EIgG (b-2), and ETRF (c-2) by MIPs. Initial epitope concentrations, 0.25 mg/mL for each template; incubation temperature, 20 °C; concentration detection method, MALDI-TOF MS with APGDRIYVHPF as the internal standard.

adsorption might be attributed to the antifouling property of the PES matrix with abundant amino and hydroxyl groups.33 Since the GRAVY values (grand average of hydropathicity, parameters to calculate the relative hydrophobicity or hydrophilicity proteins or peptides) of EHSA, EIgG, and ETRF were 1.211, 0.067, and 1.800, respectively, the recognition abilities toward different epitopes indicated that such a phase inversion self-assembly method was suitable for the imprinting of multiple templates with different properties. It should be mentioned that quite long saturation time (7.5 h and even longer) was observed in Figure 2. For more effective application as the sorbent, this imprinting strategy should be modified by (1) enlarging the channel pores on the surface to facilitate the mass transfer or (2) the surface imprinting on the nanoparticles. Furthermore, for the multiepitope imprinting, the α-TRF was similar to that of single TRF epitope imprinting,33 indicating that multiple templates did not bind each other to form complexes in this imprinting system and that each recognition site could be formed separately. To further study the adsorption mechanism of the multiple recognition sites, the second-order rate equation of the Lagergren kinetic model was applied to fit the kinetics data,36 expressed as follows: t 1 t 1 t = + = + Qt Qe vo Qe k 2Q e 2 C

DOI: 10.1021/acs.analchem.6b01247 Anal. Chem. XXXX, XXX, XXX−XXX

Letter

Analytical Chemistry

compositions, played an important role in the construction of recognition sites. The multiepitope templates imprinted particles were also applied to deplete the high abundance proteins in the human plasma. Original serum and serum sample treated by MIPs and NIPs were digested by trypsin. Then, the peptides were analyzed by RPLC−MS/MS (detailed proteome analysis protocol shown in the Supporting Information). The identified protein group number in original serum was 297, while this number in the serum treated by MIP increased to 381. The HSA, IgG, and TRF were the top three high abundance proteins in plasma; hence, more low abundance proteins were identified after MIP deplete the three target proteins simultaneously. In conclusion, we proposed a general method to prepare multiepitope templates imprinted particles through phase inversion-based PES self-assembly. The imprinted particles could capture the target proteins simultaneously with high recognition factors, even in such a complex matrix as human plasma. To the best of our knowledge, the present study is the first instance of fabricating multiepitope template imprinted particles for the selective and simultaneous capture of various proteins in real biological samples. The results also demonstrated nondiscrimination toward the epitope templates, maintenance of the epitope conformation, and low nonspecific binding. Such a multiepitope imprinting strategy might promote the wide application of MIPs as artificial antibodies to achieve simultaneous recognition of various target proteins.

Figure 3. Binding amounts (spectrum counts) of 10 highly abundant proteins by MIPs and NIPs in the native (a) and denatured (b) human plasma samples. P1, HSA; P2, TRF; P3, IgG; P4, KRT1 keratin type II cytoskeletal 1; P5, macroglobulin; P6, fibrinogen-alpha; P7, alpha-2glycoprotein; P8, alpha-1-antitrypsin; P9, complement C3; P10, apolipoprotein A-IV.

■

binding amounts on the MIPs were also apparently higher compared to NIPs. Deductively, IgG (P2) could strongly interact with the alpha-1-antitrypsin (P8) and complement C3 (P9) through protein−protein interactions.39 HSA and apolipoprotein A-IV (P10) could be complexed and adsorbed on the materials together as a result of protein corona effect.40 It should be mentioned that nonspecific binding of MIPs toward P8, P9, and P10 could be decreased by increasing the intensity of the elution solution to disrupt tenacious protein− protein interactions. This study is undertaken in our laboratory comprehensively. Such property is beyond the ability of antibody-based immunoassays. All these results demonstrated that MIPs, with multiple recognition sites, could selectively and simultaneously recognize the target proteins from complex plasma matrix. We also evaluated the recognition ability of the imprinted materials in the denatured human plasma. As shown in Figure 3b, the binding amounts of HSA, IgG, and TRF on MIPs were just slightly higher compared to NIPs, and the binding amounts of other proteins were nearly equal to those on the NIPs. The above-mentioned results showed that the selectivity of MIPs for the target proteins was more obvious in the native than that in the denatured human serum proteome sample, which further demonstrated that under the conditions for PES self-assembly, the various artificial epitopes could form their native conformation, respectively, in an aqueous environment, followed by memorization in the PES particles. Therefore, the recognition could be easily achieved in native plasma, in which the conformation of the proteins was maintained. In contrast, in the denatured plasma, the native conformation of proteins was disassembled. Therefore, the epitope-oriented recognition sites could not be well-complemented with the proteins without the initial steric structures. This phenomenon further indicated that the conformations of the epitopes, as well as their chemical

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.analchem.6b01247. Detailed experimental procedures and additional data (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*Phone/fax: +86-411-84379720. E-mail: [email protected]. cn. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We gratefully acknowledge funding from the National Natural Science Foundation of China (Grants 21375128 and 21321064) and the Ministry of Science and Technology of China (Grants 2012CB910601, 2012AA020202, and 2013BAK10B01).

■

REFERENCES

(1) Azzarito, V.; Long, K.; Murphy, N. S.; Wilson, A. J. Nat. Chem. 2013, 5, 161−173. (2) Pfleger, K. D.; Eidne, K. A. Nat. Methods 2006, 3, 165−174. (3) Zong, C.; Wu, J.; Wang, C.; Ju, H.; Yan, F. Anal. Chem. 2012, 84, 2410−2415. (4) Wang, D.; Gan, N.; Zhang, H.; Li, T.; Qiao, L.; Cao, Y.; Su, X.; Jiang, S. Biosens. Bioelectron. 2015, 65, 78−82. (5) Bellei, E.; Bergamini, S.; Monari, E.; Fantoni, L. I.; Cuoghi, A.; Ozben, T.; Tomasi, A. Amino Acids 2011, 40, 145−156. (6) Yoshimatsu, K.; Koide, H.; Hoshino, Y.; Shea, K. J. Nat. Protoc. 2015, 10, 595−604. D

DOI: 10.1021/acs.analchem.6b01247 Anal. Chem. XXXX, XXX, XXX−XXX

Letter

Analytical Chemistry (7) LeJeune, J.; Spivak, D. A. Biosens. Bioelectron. 2009, 25, 604−608. (8) Albonetti, S.; Cavani, F.; Trifiro, F.; Venturoli, P.; Calestani, G.; Granados, M. L.; Fierro, J. L. G. J. Catal. 1996, 160, 52−64. (9) Kullolli, M.; Warren, J.; Arampatzidou, M.; Pitteri, S. J. J. Chromatogr. B: Anal. Technol. Biomed. Life Sci. 2013, 939, 10−16. (10) Millioni, R.; Tolin, S.; Puricelli, L.; Sbrignadello, S.; Fadini, G. P.; Tessari, P.; Arrigoni, G. PLoS One 2011, 6, e19603. (11) McDonough, M. A.; Li, V.; Flashman, E.; Chowdhury, R.; Mohr, C.; Lienard, B. M.; Zondlo, J.; Oldham, N. J.; Clifton, I. J.; Lewis, J.; McNeill, L. A.; Kurzeja, R. J.; Hewitson, K. S.; Yang, E.; Jordan, S.; Syed, R. S.; Schofield, C. J. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 9814−9819. (12) Yang, F.; Lin, S.; Dong, X. Chem. Commun. 2015, 51, 7673− 7676. (13) Takeuchi, T.; Mori, T.; Kuwahara, A.; Ohta, T.; Oshita, A.; Sunayama, H.; Kitayama, Y.; Ooya, T. Angew. Chem., Int. Ed. 2014, 53, 12765−12770. (14) Ma, Y.; Pan, G.; Zhang, Y.; Guo, X.; Zhang, H. Angew. Chem., Int. Ed. 2013, 52, 1511−1514. (15) Bie, Z. J.; Chen, Y.; Ye, J.; Wang, S. S.; Liu, Z. Angew. Chem., Int. Ed. 2015, 54, 10211−10215. (16) Shinde, S.; Bunschoten, A.; Kruijtzer, J. A. W.; Liskamp, R. M. J.; Sellergren, B. Angew. Chem., Int. Ed. 2012, 51, 8326−8329. (17) Bi, X. D.; Liu, Z. Anal. Chem. 2014, 86, 12382−12389. (18) Bi, X. D.; Liu, Z. Anal. Chem. 2014, 86, 959−966. (19) Li, L.; Lu, Y.; Bie, Z. J.; Chen, H. Y.; Liu, Z. Angew. Chem., Int. Ed. 2013, 52, 7451−7454. (20) Nishino, H.; Huang, C. S.; Shea, K. J. Angew. Chem., Int. Ed. 2006, 45, 2392−2396. (21) Zhang, Y.; Deng, C.; Liu, S.; Wu, J.; Chen, Z.; Li, C.; Lu, W. Angew. Chem., Int. Ed. 2015, 54, 5157−5160. (22) Liu, L.; Zhong, T.; Xu, Q.; Chen, Y. Anal. Chem. 2015, 87, 10910−10919. (23) Li, S.; Yang, K.; Liu, J.; Jiang, B.; Zhang, L.; Zhang, Y. Anal. Chem. 2015, 87, 4617−4620. (24) Yang, H. H.; Lu, K. H.; Lin, Y. F.; Tsai, S. H.; Chakraborty, S.; Zhai, W. J.; Tai, D. F. J. Biomed. Mater. Res., Part A 2013, 101, 1935− 1942. (25) Lu, C. H.; Zhang, Y.; Tang, S. F.; Fang, Z. B.; Yang, H. H.; Chen, X.; Chen, G. N. Biosens. Bioelectron. 2012, 31, 439−444. (26) Zhou, J.; Gan, N.; Li, T.; Hu, F.; Li, X.; Wang, L.; Zheng, L. Biosens. Bioelectron. 2014, 54, 199−206. (27) Tai, D. F.; Jhang, M. H.; Chen, G. Y.; Wang, S. C.; Lu, K. H.; Lee, Y. D.; Liu, H. T. Anal. Chem. 2010, 82, 2290−2293. (28) Xu, J. J.; Ambrosini, S.; Tamahkar, E.; Rossi, C.; Haupt, K.; Bui, B. T. S. Biomacromolecules 2016, 17, 345−353. (29) Chen, Y.; Li, D. J.; Bie, Z. J.; He, X. P.; Liu, Z. Anal. Chem. 2016, 88, 1447−1454. (30) Seo, M.; Kim, S.; Oh, J.; Kim, S. J.; Hillmyer, M. A. J. Am. Chem. Soc. 2015, 137, 600−603. (31) Zhao, C.; Xue, J.; Ran, F.; Sun, S. Prog. Mater. Sci. 2013, 58, 76− 150. (32) Yang, K. G.; Liu, Z. B.; Mao, M.; Zhang, X. H.; Zhao, C. S.; Nishi, N. Anal. Chim. Acta 2005, 546, 30−36. (33) Yang, K.; Liu, J.; Li, S.; Li, Q.; Wu, Q.; Zhou, Y.; Zhao, Q.; Deng, N.; Liang, Z.; Zhang, L.; Zhang, Y. Chem. Commun. 2014, 50, 9521−9524. (34) Issaq, H. J.; Xiao, Z.; Veenstra, T. D. Chem. Rev. 2007, 107, 3601−3620. (35) Qin, J. J.; Cao, Y. M.; Oo, M. H. J. Appl. Polym. Sci. 2006, 99, 430−435. (36) Gao, R. X.; Mu, X. R.; Hao, Y.; Zhang, L. L.; Zhang, J. J.; Tang, Y. H. J. Mater. Chem. B 2014, 2, 1733−1741. (37) Zhang, Y.; Fonslow, B. R.; Shan, B.; Baek, M. C.; Yates, J. R., 3rd. Chem. Rev. 2013, 113, 2343−2394. (38) Liu, J.; Deng, Q.; Tao, D.; Yang, K.; Zhang, L.; Liang, Z.; Zhang, Y. Sci. Rep. 2014, 4, 5487. (39) Vivanco, F.; Munoz, E.; Vidarte, L.; Pastor, C. Mol. Immunol. 1999, 36, 843−852.

(40) Winzen, S.; Schoettler, S.; Baier, G.; Rosenauer, C.; Mailaender, V.; Landfester, K.; Mohr, K. Nanoscale 2015, 7, 2992−3001.

E

DOI: 10.1021/acs.analchem.6b01247 Anal. Chem. XXXX, XXX, XXX−XXX