Investigation of nonfouling polypeptides of poly(glutamic acid) with lysine side chains synthesized by EDC·HCl/HOBt chemistry

Nonfouling polypeptides with homogenous alternating charges draw peoples’ attentions for their potential capability in biodegradation. Homogenous glutamic acid (E) and lysine (K) polypeptides were proposed and synthesized before. In this work, a new polypeptide formed by poly(glutamic acid) with lysine side chains (poly(E)-K) was synthesized by facile EDC·HCl/HOBt chemistry and investigated. Results show that these polypeptides also have good nonspecific protein resistance determined by enzyme-linked immunosorbent assay. The lowest nonspecific adsorption of the model proteins, anti-IgG and fibrinogen (Fg), on the self-assembling monolayers (SAMs) surface of poly(E)-K was only 3.3 ± 1.8 and 4.4 ± 1.6%, respectively, when protein adsorption on tissue culture polystyrene surface was set as 100%. And, the relative nonspecific protein adsorption increases when the polypeptide molecular weight increases due to the repression of low density polymer brushes. Moreover, almost no obvious cytotoxicity and hemolytic activity in vitro were detected. This work suggests that polypeptides with various formats of homogenous balanced charges could achieve excellent nonspecific protein resistance, which might be the intrinsic reason for the coexistence of high concentration serum proteins in blood.


Introduction
Polypeptide-based nonfouling materials could be biocompatible alternatives of polyvinyl-based ones due to their potential in natural structure-originated properties, such as biodegradability, compatibility and nontoxicity of themselves, and their degradable products. Recently, polypeptides, with homogenous alternating glutamic acid (E) and lysine (K) (poly(EK)), were synthesized and showed high nonfouling property, which proved that the zwitterionic structure mimic at nanometer scale could be an efficient way to obtain resistance to nonspecific protein adsorption. [1][2][3] Moreover, a facile method was successfully developed to solve the uniformity problem of nonfouling peptides caused by copolymerization of EK dimers in large quantities for potential biomedical applications. [2] On the other hand, the methacrylate-based materials containing amino acids were proved to be effective antifouling materials. Liu et al. developed poly(serine methacrylate), [4] poly(lysine methacrylamide) and poly(ornithine methacrylamide) for fouling resistance. [5] Ishii also developed copolymers of N-methacryloyl-L-histidine and n-butyl methacrylate to reduce surface biofouling and found the 50% content of the histidine monomer one as the most excellent candidate. [6] Shiraishi synthesized several poly (methyl methacrylate) copolymer-based microspheres, among which poly(O-methacryloyl -L-serine-co-methyl methacrylate) (poly(SerMA-co-MMA)) showed the most effective suppression of protein adsorption. [7] More interestingly, a series of lysine-containing polymers developed by Chen H group [6][7][8][9] showed both fouling resistance and fibrinolytic activity originated from the synergetic effect of the ε-amino group and the carboxyl group, while additional 2-hydroxyethyl methacylate (HEMA) or poly(ethylene glycol) (PEG) methacylate was introduced to control the density of lysine in these copolymers. Klok also developed an oligo(ethylene glycol) (OEG)-containing surface with lysine brushes by ring-opening polymerization (ROP) of lysine N-carboxyanhydride (NCA). [10] Here, we proposed a new structure, poly(glutamic acid) with lysine side chains to investigate the fouling resistance. This polypeptide has structure similar to lysine-based polymethacrylamide developed by Chen H group [6][7][8][9], which could show extra fibrinolytic activity besides the fouling resistance observed in poly(EK). The detailed synthesis route is shown in Scheme 1. DL-lipoic acid was applied to adjust molecular weight (MW) by attaching it to the N-terminal of the polypeptides. The lipoic acidcapped poly(E)-K can form SAMs on gold surface through two thiol groups. The polymer was verified by nuclear magnetic resonance ( 1 HNMR) and gel permeation chromatography (GPC). Chemical and physical properties of the SAMs were systematically studied by attenuated total reflection Fourier transform infrared (ATR-FTIR) spectroscopy, X-ray photoelectron spectroscopy (XPS), and Ellipsometer (ELL). The relationships between protein resistance, SAM thickness, and polypeptide MW were also discussed. The biocompatibility including in vitro cytotoxicity and hemolytic activity of polypeptides was also investigated. All results indicated that these new polypeptides with lysine side chains are good biocompatible nonfouling materials through reducing nonspecific interactions with protein molecules and cell membranes.

Synthesis route
The synthesis route is summarized in Scheme 1. Scheme 1. Synthesis route of the poly(E)-K polypeptide.

Synthesis of δ-(α-benzyl-N ε -benzyloxycarbonyl-L-lysinyl)-L-glutamic acid (5)
To a 30 mL TFA/CH 2 Cl 2 (V/V = 2:1) solution, 4 (8.0 g, 12.20 mmol) was added. The mixture was stirred for 6 h at 0°C, followed by removal of the solvent by a rotary evaporator under reduced pressure to get an oil residue. The residue was dissolved in ethyl acetate and TEA was added to make a neutral solution. The organic solvent was removed by rotary evaporator and the residue was suspended in water under agitation for several times and the mixture was filtrated to get the compound (5)

Synthesis of poly(N δ -L-lysinyl-L-glumatic acid) (7)
Deprotection of Z moiety was achieved by a mixed solution of 33wt.% HBr/HOAc and TFA(V/V = 1:1). [11] The product was precipitated by ethyl ether, neutralized to pH 7.0 by saturated NaHCO 3 solution, dialyzed against deionized (DI) water, and lyophilized to obtain a white powder (7). Full cleavage was confirmed by total disappearance of peaks at δ 7.22-7.39 and 4.94-5.13 in 1 HNMR spectroscopy. From the (E)-K dimer, the average yield for the polypeptide is about 40%. 1

GPC assay
The MW and polydispersity index (PDI) of polypeptides were determined by a Waters Ultrahydrogel TM 120 column and a Waters Ultrahydrogel TM linear column. The mobile phase was phosphate-buffered saline (PBS) (NaCl 150 mM, pH 7.4) with a flow rate of 0.5 mL/min at 40°C. Poly(ethylene oxide) with different MWs were used as standards.
2.4. ATR-FTIR for film structure and ELL for film thickness Gold-coated chips were rinsed with ethanol and DI water, dried by filtered air, and placed in a UV cleaner for 20 min. Then, chips were incubated with a PBS solution (pH 7.4) of 3 mg/mL polypeptide for 24 h. The chips were then rinsed several times by PBS, followed by drying with filtered air to make clean chips for measurement. ATR-FTIR was used to verify the formation of the SAMs on chips by detecting the characteristic groups of the polypeptides. Thickness of the film was measured by a Spectroscopic Ellipsometer (J. A. Woollam M-2000D). For each sample, five separate spots were measured and bare gold-coated chips were used as references.

XPS measurement [12]
The pretreatment for the chips are the same as that for FTIR and ELL. XPS was conducted on an X-Probe Spectrometer (VG ESCALAB MARK II) equipped with a monochromatic Mg Kα X-ray source (hν = 1253.6 eV), a hemi-spherical analyzer, and a multichannel detector. For each SAM, three separate spots were examined. The bare gold-coated chips were used as references.

Protein adsorption assay
Nonspecific protein adsorption of the polypeptide SAMs on gold surfaces and tissue culture polystyrene (TCPS) control were determined by anti-IgG/HRP and Fg adsorption. The samples were first incubated with anti-IgG/HRP (1 µg/mL) or Fg (10 µg/mL) for 20 min, and rinsed with PBS to remove all free proteins. For Fg adsorption, the samples were further incubated with anti-Fg/HRP (10 µg/mL) for another 10 min. After that, all tested samples and TCPS controls were placed in a 24-well plate and incubated with 1 mL O-phenylenediamine (OPD) (1 mg/mL) in citrate phosphate buffer (0.1 M, pH 5.0) containing 0.03% hydrogen peroxide. Enzyme activity was quenched by adding an equal volume of 2 N H 2 SO 4 after 6-8 min incubation. The tangerine color was measured by a UV spectrophotometer at 492 nm.

MTT assay
A typical procedure for the MTT assay is described as follows: MTT was dissolved in PBS (pH 7.4) to obtain a 5 µg/mL MTT working solution. HUVECs were seeded in a 96-well tissue culture plate at a density of 10,000 cells per well and cultured in a RPMI 1640 medium with 10% FBS. After one day incubation, the medium was removed and the cells were washed by RPMI 1640 medium twice. Then, polypeptides were added at various concentrations of 0.01, 0.05, 0.1, 0.5, 1, and 5 mg/mL, followed by additional 1 day incubation. Then, for each concentration, MTT solutions were loaded for three wells. The MTT solution was removed after 4 h incubation, then 100 µL DMSO was added to each well followed by 10 min shaking to get a complete dissolving of the MTT reduction product in cells. The cell viability was determined by measuring the absorbance of each sample at 492 nm on a Microplate Reader.
2.8. Hemolytic assay [13][14][15] Red blood cells (RBCs) were collected by centrifugation of whole blood in sterile PBS at 1500 rpm for 10 min. The RBCs were further washed three times by sterile PBS. After the supernatant was removed following the last wash, the cells were resuspended in PBS to get a 2% w/v RBC suspension. The tested polypeptide samples were also prepared in sterile PBS. One hundred and fifty microliters of the sample solution and 150 μL of the 2% w/v RBCs solution were added to a centrifuge tube to make a sample with final concentration of 5 mg/mL and incubated for 4 h at 37°C. Then, the mixture was centrifuged and the supernatant was transferred to a 96-well plate. The relative adsorption of supernatants was measured on a Microplate Reader at 575 nm. Complete hemolysis was attained using water as the positive control, and PBS was used as the negative control. The hemolytic activity is defined as follows: Hemolytic activity% = [(sample absorbance − negative control)/(positive control − negative control)] × 100%.  (5). And, the 1 HNMR spectroscopy ( Figure S2) also indicates that the dimer (5) was successfully synthesized for polymerization in the next step. The dimer (5) was then condensed by EDC·HCl/HOBt chemistry to get a Z-protected polypeptide (6), and DL-lipoic acid was used as the end group to adjust the MW of polypeptide. The final product (7) was achieved by total cleavage of the protection moiety of Z, which can be confirmed by the disappearance of peaks at δ 7.27-7.39 and 4.94-5.13 in 1 HNMR spectroscopy. Three poly(E)-K polypeptides at the MW of 3.5 kDa (PDI = 1.08), 5.6 kDa (PDI = 1.44), and 13.7 kDa (PDI = 1.83) were obtained (Table 1) and their 1 HNMR spectroscopy is shown in Figure S3. The difference between the theoretical MW determined by the feeding molar ratio of dimer (5) to lipoic acid and the deviation of MW of the synthesized poly(E)-K polypeptides from theoretical values might be caused by side reactions of minor impurities in dimer (5). The very low PDI of 3.5 kDa poly(E)-K polypeptide is possibly caused by its low polymerization degree and the dialysis during preparation in which some of short poly(E)-K peptides could be removed to narrow MW distribution.

SAM properties characterized by ATR-FTIR, XPS, and ELL
Chemical and physical properties of the polypeptide SAMs were systematically characterized by ATR-FTIR, XPS, and ELL. In Figure 1, the main characteristic peaks of the secondary amide, band I (σ C=O , 1600-1750 cm −1 ), band II (δ N-H , 1500-1560 cm −1 ) can be found, indicating the existence of polypeptide SAMs on the gold surfaces. The detailed spectra of the band I (carbonyl region) clearly show two peaks for amide in lower wavenumber region and carboxyl groups in lysine side chains in higher wavenumber region, respectively. Survey and detailed scans of XPS spectra were obtained to provide further evidences for the existence of the SAMs on gold surface and to investigate relative element contents of the polypeptide SAMs. By comparing the sample and control spectra in Figure 2, both the newly emerging peak of N1s, serving as the characteristic element peak of polypeptides, and the obvious intensity increase of C1s and O1s indicate the polypeptide SAM formation. In order to quantify the relative element contents of the SAMs, detailed scans of carbon, nitrogen, and oxygen were acquired. The detected carbon and nitrogen contents are 63.6 ± 2.4 and 10.2 ± 1.3% (Table 2), respectively, which are slightly lower than their theoretical values of 70.8 and 12.5%. However, oxygen content (26.7 ± 1.6%) increases significantly from its theoretical value (16.7%) with a large relative deviation of 59.9 ± 9.6%, which had also been observed by Chen and Chung et al. [16,17]. According to their results, [16,18] the large amount of the hydration water of the zwitterions of the polypeptides could increase the content of oxygen and lower the contents of carbon and nitrogen, found in XPS measurements comparing the theoretical value. Film thicknesses are determined by ELL. In Figure 3, when the MWs of tested samples are 3.5, 5.6 and 13.7 kDa, the corresponding thicknesses of the SAMs are 2.8 ± 0.1, 3.4 ± 0.2, and 3.8 ± 0.6 nm, respectively. As the MW increases, the slope factor slightly decreases, indicating that the thickness of the SAMs does not linearly increase with the MWs. The possible reason for this phenomenon is that longer polymers have more flexibility and are in a 'collapsed' state, thus leading to a slower thickness increase.

Nonspecific protein adsorption assay
As a fundamental property of the nonfouling materials, the nonspecific protein adsorption of the polypeptide SAMs was evaluated by enzyme-linked immunosorbent assay  Notes: The compositions of the substrate are not considered here and the relative elemental content value is calculated as follows: element% = specific element fitting area/(C + N + O) fitting area × 100%.
using HRP-conjugated anti-IgG and Fg as model proteins. The relative adsorption of Fg was determined by a modified sandwich method. As shown in Figure 4, when MWs increase from 3.5 to 13.7 kDa, the relative nonspecific protein adsorption for anti-IgG and Fg increases from 3.3 ± 1.8, 4.4 ± 1.6 to 11.6 ± 2.8, 15.9 ± 5.6%, respectively. And for all samples tested, Fg has a higher relative adsorption than anti-IgG. It Notes: When the MW increases from 3.5 to 13.7 kDa, the relative nonspecific protein adsorption for anti-IgG and Fg increases from 3.3 ± 1.8, 4.4 ± 1.6 to 11.6 ± 2.8, 15.9 ± 5.6%, respectively. For all samples tested, the polypeptide SAMs show good nonspecific protein resistance properties. The results are means ± standard deviation (SD) (n = 3). is believed that this phenomenon is mainly caused by the higher tendency of Fg in denaturation due to the flexible structure of Fg. [19] The reason for the increased relative nonspecific protein adsorption could be ascribed to the cavity increase when the polypeptide MWs increase. According to Currie [20] and Halperin's results [21], when the particle size is smaller than the space among polymer brushes, the particles can diffuse into the brushes and repress and reorganize polymer brushes, which lead to protein contact with the surface of the substrate and adsorb on it. As for our case, when the MWs increase, cavities among polypeptide chains may become larger due to the low surface packing density of the polymer judged from slow increase of polymer film thickness, which leads to more protein adsorption on the SAM surface, as Currie [20] and Halperin [21] observed. Another possible reason might be the synergetic effect of multiple weak interactions between protein molecule and longer polypeptide chain. [22] Long chain polymer is more flexible and has more contact sites available for protein adsorption, whereas short chain polymer may form a more compact layer with fewer contact sites; thus, lead to lower protein adsorption. It should be noted that all of our samples with various MWs show relatively high efficiency to reduce both anti-IgG and Fg adsorption, even if their PDI is not in a very narrow range, indicating that this kind of polypeptides are highly fouling resistant and the uniformity of MW is not very critical. Comparing the resistant property of the polypeptide SAMs with the OEG SAMs reported by Whitesides et al. [23,24] the protein adsorption on the 3.5KDa polypeptide SAMs, the lowest one among three tested samples, is slightly higher than on the OEG SAMs. However, it is a reasonable results since the complex structure of poly(glutamic acid) with lysine side chains hinders the high density SAMs formation. Last, based on our preliminary results (data not shown), these polypeptides show selective plasminogen-binding capability, which is a key factor for fibrinolytic activity of lysine-based fouling-resistant materials. [6][7][8][9] All suggest that these polypeptides might be ideal candidates for bloodcontact coatings. Notes: From the histogram, we can find that when the polypeptide concentrations increase from 0.01 to 5 mg/mL, the cell viabilities vary from 120.9 ± 5.0% to 85.6 ± 3.2%, indicating no cytotoxicity in vitro for the polypeptides. The results are means ± standard deviation (SD) (n = 3).

MTT and hemolytic activity assay
MTT and hemolytic activity assay were conducted by following the procedures described in 2.7 and 2.8, respectively. When the polypeptide concentration increases from 0.01 to 5 mg/mL, the cell viability gradually decreases from 120.9 ± 5.0 to 85.6 ± 3.2%, indicating no obvious cytotoxicity in vitro for the polypeptides ( Figure 5). It is believed that the decreased cell viability is mainly caused by the high concentration of the hydrophobic lipoic acid terminal group of the polypeptide, which might increase the interaction with the cell membrane of HUVECs and interfer their growth. However, this interference is not very obvious even when the concentration of the polypeptide reaches to 5 mg/mL. Furthermore, the hemolytic activity assay result ( Figure 6) indicates low interactions between the polypeptides and cell membrane at 5 mg/mL polypeptides. It was found that the light absorbance of the supernatants was very low after the removal of RBCs incubated with 3.5, 5.6, and 13.7 kDa polypeptide. For three tested samples, the adsorption was even lower than the negative control (PBS solution). Such phenomenon might come from the protection to RBCs by polypeptides. The highest hemolytic activity was observed for the 3.5 kDa polypeptide, of which hemolytic activity was just −14.2 ± 13.3%. Such low hemolytic activity agrees with the cytotoxicity results and also suggests that the low cytotoxicity should be attributed to low interactions of the polypeptides with protein and cell membrane. In short, all the polypeptides exhibit very good biocompatibility and are excellent candidates for fouling-resistant materials.

Conclusions
In this work, a kind of fouling-resistant polypeptides was synthesized by facile EDC·HCl/HOBt chemistry. Successful synthesis of the polypeptides was verified by 1 HNMR and GPC. The chemical and physical properties of the polypeptide SAMs were systematically evaluated by ATR-FTIR, XPS, and ELL. The existence of the SAMs on the gold surface, the relative element contents of the SAMs, and the thickness of the SAMs were determined. Furthermore, results from nonspecific protein adsorption resistance assay show that this kind of material has good nonspecific protein resistance. In MTT and hemolytic assay, even if the feeding concentration is up to 5 mg/mL, no obvious cytotoxicity or hemolytic activity for the polypeptides in vitro was detected. The design of a homogenous charge distribution and the facile EDC·HCl/ HOBt synthesis chemistry provides a clear guide to explore other fouling-resistant homologies. As a candidate of biofunctional materials, these fouling-resistant polypeptides will be investigated in our successive works.