Convenient Approach to Polypeptide Copolymers Derived from Native

May 4, 2012 - Programming Supramolecular Biohybrids as Precision Therapeutics ... Programmable Biopolymers for Advancing Biomedical Applications of ...
0 downloads 4 Views 3MB Size
Download PDF
Article pubs.acs.org/Biomac

Convenient Approach to Polypeptide Copolymers Derived from Native Proteins Yuzhou Wu,†,§ Goutam Pramanik,†,§ Klaus Eisele,† and Tanja Weil*,†,‡ †

Institute for Organic Chemistry III/Macromolecular Chemistry, Ulm University, Albert-Einstein-Allee 11, 89073 Ulm, Germany Max Planck Institute for Polymer Research, Ackermannweg 10, 55128 Mainz, Germany

‡

S Supporting Information *

ABSTRACT: A convenient approach for the synthesis of narrowly dispersed polypeptide copolymers of defined compositions is presented. The controlled denaturation of the proteins serum albumin and lysozyme followed by an in situ stabilization with polyethylene(oxide) chains yields polypeptide side chain copolymers of precisely defined backbone lengths as well as the presence of secondary structure elements. Supramolecular architectures are formed in solution because of the presence of hydrophobic and hydrophilic amino acids along the polypeptide main chain. Polypeptide copolymers reported herein reveal excellent solubility and stability in aqueous media and no significant cytotoxicity at relevant concentrations, and they can be degraded via proteolysis, which is very attractive for biomedical applications. This “semi-synthetic chemistry” approach is based on a novel and convenient concept for producing synthetic polypeptides from native protein resources, which complements traditional polypeptide synthesis and expression approaches and offers great opportunities for the preparation of diverse polypeptides with unique architectures.

■

INTRODUCTION Peptide materials are of emerging interest because of their biocompatibility, biodegradability, and the potential to achieve unique macromolecular architectures with tailored properties.1 They have been applied successfully for a number of applications such as self-assembly into complex nanostructures2 or to improve the pharmacokinetic properties of protein therapeutics.3 To date, there is an increasing interest in the preparation of peptide materials for biomedical applications, for example, for tissue engineering4 or as scaffolds for the delivery of drug molecules.5 Biomedical applications often impose that the biological and physical properties of the polypeptides are precisely controlled,6 which is usually considered to be rather challenging to achieve. Ring-opening polymerization of α-amino acid N-carboxyanhydrides represents one of the most commonly used synthetic schemes to prepare polypeptides with repeating amino acid sequences.7 However, monodisperse polypeptides composed of different kinds of amino acids or particular peptide sequences with defined structures and distinct molecular weights are often challenging to achieve via polymerization reactions.8 As complementary strategies, the expression of recombinant polypeptides9 and the solid-phase peptide synthesis10 yields polypeptides with more diverse © 2012 American Chemical Society

sequences and defined molecular weights. Still, sophisticated expression or synthesis procedures, low yields, and high costs represent common limitations. Therefore, it is highly desirable to develop more efficient approaches for the preparation of defined polypeptides with tunable properties. Proteins represent natural polypeptides of precisely defined sequences, molecular weights, structures, and 3D architectures, and they are ubiquitous available in nature.11 It has been proposed that the chemical modification of the side chains of proteins should facilitate monodisperse polymers with sequences of natural diversity and functionalities.8 However, because of the tightly folded polypeptide backbone of native proteins, functional groups are mostly hidden inside the protein structure, thus impeding efficient chemical reactions at the protein surface. Denaturing the 3D structure of proteins usually causes precipitation and destabilization of the polypeptide backbone,12 thus preventing the use of proteins as polypeptide sources. Herein, we present a comprehensive and highly versatile approach to achieve polypeptide-based copolymers with distinct molecular weights, chain lengths, ordered secondary Received: March 16, 2012 Revised: May 1, 2012 Published: May 4, 2012 1890

dx.doi.org/10.1021/bm300418r | Biomacromolecules 2012, 13, 1890−1898

Biomacromolecules

Article

room temperature for 15 min. Tris(2-carboxy-ethyl)phosphine hydrochloride (TCEP) (8.6 mg, 30 μmol) was then added as a solid, and the solution was stirred for an additional 30 min under an argon atmosphere. Subsequently, PEO-5000-MI (154 mg, 31 μmol) was allowed to react with the protein solution under an argon atmosphere for 3 h. After the attachment of the PEO chains, N-(2aminoethyl)maleimide trifluoroacetate salt (8 mg, 31.5 μmol) was introduced and stirred for another 3 h to react with the remaining thiol groups. Thereafter, the reaction mixture was prepurified five times by ultrafiltration using Vivaspin 20 (MWCO 30k) centrifugal concentrator with 20 mM Tris-HCl buffer (pH 7.4, containing 150 mM NaCl, 2 mM EDTA) and then further purified by size-exclusion FPLC (AKTÄ Purifier, Sephacyl S-100 HR gel filtration column, 20 mM Tris-HCl pH 7.4 buffer with 150 mM NaCl). The first protein peak was collected and concentrated by Vivaspin 20 (MWCO 30k) centrifugal concentrator. Subsequently, the product mixture was passed through a Bio-Gel P30 desalting column and lyophilized to afford 24 mg dHSA-PEO(5000)27 (2) (daHSA-PEO(5000)27 (5) or dcBSA-PEO(5000)27 (8)) as white solid (overall yield ∼40%). Preparation of dLY-PEO(5000)8 (10). Five mg LY was denatured using 5 mL of guanidine-tris buffer (8 M guanidine, 0.2 M NaCl, 1 mM EDTA, and 0.1 M Tris/HCl) and stirred at room temperature for 4 h. Then, the disulfide bond of LY was reduced in 1 mL of 1 M DLdithiothreitol (DTT) by stirring at room temperature for 24 h. After reduction, the pH of the reaction mixture was brought below 4 using HCl. Thereafter, DTT was removed from the reaction mixture by a Sephadex G25 column using 0.1 M acetic acid. Subsequently, the denatured protein was lyophilized to obtain denatured lysozyme (9). PEO(5000)-MI (41 mg, 8.2 μmol) was dissolved in degassed ureaphosphate buffer (10 mL, 10 mM phosphate buffer, pH 7.4, 5 M urea, and 2 mM EDTA); then, (9) (5 mg, 0.34 mmol) was added and stirred under an argon atmosphere overnight. The reaction mixture was prepurified by ultrafiltration five times with pure water using Vivaspin 20 (MWCO 10 kDa) centrifugal concentrator and then further purified through a Bio Gel P-30 desalting column and lyophilized to afford dLY-PEO(5000)8 (10) as white solid (overall yield 35%). MALDI-TOF MS: 54 kDa (M+). Gel Permeation Chromatography. Gel permeation chromatography (GPC) was recorded on Waters 515 HPLC system using PL Aquagel−OH 30/40/50 and PL Guard Column (Agilent) and Waters 244 refractive index detector. One mg of sample was injected. Pure water was used as mobile phase with a flow rate of 1 mL/min. Dynamic Light Scattering. Protein copolymers (2), (5), (8), and (10) were prepared at 0.1 mg/mL in aqueous solution and measured at 25 °C using a Malvern Zetasizer ZEN3600 (Malvern) The analysis was accomplished with a laser wavelength of 633 nm and a scattering angle of 173°. Solutions were filtered through a 0.2 μm pore size Supor Membrane (Life Science) filter prior to data acquisition. Autocorrelation functions were analyzed by applying the cumulants method and COTIN routine to estimate the hydrodynamic diameter and polydispersity index (PDI). The hydrodynamic diameter distribution was discussed based on both intensity distribution and number distribution. Characterization of the Polymer Surface Charge. The net surface charges of all samples were characterized by agarose gel electrophoresis using 0.5% agarose gel in pH 7.4 TAE (Tris-AcetateEDTA) buffer. Positively charged species move toward the negative electrode, whereas the negatively charged molecules move toward the positive electrode. The zeta potential of all samples was also acquired in 1 mM KCl solution using a Malvern Zetasizer ZEN3600 (Malvern). Proteolysis of Polypeptides - Stability Study. One mg/mL of each sample was prepared in PBS buffer at pH 7.4 and incubated in 0.1% trypsin at 37 °C. To monitor the digestion process on SDSPAGE, we took 7 μL of the sample out of the reaction mixture at desired time points, and it was immediately treated by phenylmethanesulfonyl fluoride (PMSF) (1 μL, 10 mM), DTT (1 μL, 1 M) and 3 μL of NuPAGE LDS sample buffer (Invitrogen) and boiled at 95 °C for 5 min. The samples were then directly analyzed by SDSPAGE.

structural elements, and high numbers of functional groups by applying strong denaturing conditions. To overcome the precipitation problem, the backbone was stabilized by grafting poly(ethylene oxide) (PEO) chains. Such copolymers are well soluble in water, and various reactive groups are distributed along the polypeptide backbone that are accessible for further modifications. We have previously demonstrated the preparation of a 200 kDa molecular weight polypeptide copolymer with negative net charge derived from a human serum albumin (HSA) precursor,13 which is attractive for the stabilization of nanoparticles such as quantum dots (QDs)13 and the delivery of drug molecules.14 In this Article, we discuss a general scheme for the synthesis and functionalization of polypeptide copolymers derived from different proteins. These highly versatile macromolecules achieved herein could be considered as a novel class of materials featuring biocompatibility, biodegradability, and multiple reactive functional groups along the backbone as well as the presence of ordered structural elements. Such biopolymers are highly attractive for a broad range of applications such as nanopatterning, drug delivery, as well as tissue engineering. The “semi-synthetic” chemistry approach” reported herein could be considered to be a valuable alternative to conventional polypeptide synthesis and expression providing fast and efficient access to numerous polypeptides of natural diversity.

■

MATERIALS AND METHODS

Materials. All chemical reagents were obtained from commercial suppliers and used without further purification unless otherwise noted. Albumins from human serum (HSA), lysozyme from hen egg white (LY), and bovine serum albumin (BSA) were obtained from SigmaAldrich. cBSA-147 (cBSA) was synthesized according to the our previous report.15 Bio-Gel P30 from Bio-Rad was used for desalting column, and Vivaspin centrifugal concentrators were purchased from GE healthcare. Dulbecco’s modified Eagle’s medium (D-MEM) (1×) liquid (high glucose), fetal bovine serum (FBS) standard quality (EU approved) from PAA Laboratories, and MEM non-essential amino acids solution 10 mM (100×) from Invitrogen were used for cell culture. CellTiter-Glo Luminescent Cell Viability Assay was obtained from Promega. Instrumentation. AKTÄ Purifier FPLC and Sephacyl S-100 HR gel filtration column were used for protein copolymer purification. Precast NuPAGE TA 3%-8% Gel and NuPAGE Bis-Tris 4%-12% Gel were purchased from Invitrogen, and gel electrophoresis was performed in Invitrogen Novek Mini-Cell. Agarose gel electrophoresis was performed using Bio-Rad Mini-Sub Cell GT horizontal electrophoresis system. Circular dichroism (CD) spectra were measured on a JASCO J-810 spectropolarimeter. The MALDI-TOF mass spectrum was obtained on a Bruker Reflex III MALDI-TOF spectrometer. Dynamic light scattering (DLS) and zeta potential measurements were performed on a Malvern Zetasizer ZEN3600 (Malvern, Ltd., Malvern, U.K.). Preparation of Anionic HSA (aHSA). 50 mg (∼0.75 μmol) of HSA and 26.4 mg (0.2 mmol) glutaric acid were allowed to dissolve completely in 7.5 mL of degassed phosphate buffer (50 mM, pH8.4), followed by adding 1-ethyl-3-(3-dimethylaminopropyl)carbodiimide (EDC) (91 mg, 0.47 mol) and stirring for 2.5 h at room temperature. Reaction mixture was then directly added to a Vivaspin 20 (MWCO 30k) centrifugal concentrator and washed five times with deionized distilled water. After washing, aHSA can be lyophilized to obtain 48 mg white fluffy solid (yield: 90%). MALDI-TOF MS, 71.4 kDa (M+). Preparation of dHSA-PEO(5000)27 (2), daHSA-PEO(5000)27 (5), and dcBSA-PEO(5000)27 (8). HSA (20 mg, 0.3 μmol) (use 0.3 μmol aHSA to prepare daHSA-PEO(5000)27 or use 0.3 μmol cBSA147 to prepare dcBSA-PEO(5000)27) were dissolved in degassed ureaphosphate buffer (10 mL, 20 mM phosphate buffer, pH 7.4, 5 M urea, and 2 mM ethylenediamine tetraacetic acid (EDTA)) and stirred at 1891

dx.doi.org/10.1021/bm300418r | Biomacromolecules 2012, 13, 1890−1898

Biomacromolecules

Article

Scheme 1. Preparation of HSA-, aHSA-, cBSA-, and LY-Derived Polypeptide Copolymers

Circular Dichroism. All copolymer materials were dissolved in pure water in ∼2 mg/mL concentration, respectively. Tris buffer (0.75 mL, 50 mM, pH7) was added to 0.1 mL of this solution separately. The sample was measured in a 1 mm cuvette with a volume of 700 μL. The CD signal was recorded from 240 to 190 nm. The bandwidth was set to 1 nm and a response at 1 s. Standard sensitivity was used with the data pitch 0.1 and 100 nm/min scanning speed. Temperature was kept constant at 20 °C. The data were recorded using five times data accumulation. The secondary structure elements were estimated using the CDSSTR program in CDPro software package (available at website: http://lamar.colostate.edu/∼sreeram/CDPro/main.html). Cytotoxicity Assay. A549 cells, a carcinomic human alveolar basal epithelial cell line, was obtained from DSMZ, German Collection of Microorganisms and Cell Cultures, Braunschweig. The cells were cultured in DMEM medium with high glucose supplemented with 10% FBS, 100 U/mL penicillin, 0.1 mg/mL streptomycin, and 0.1 mM nonessential amino acids at 37 °C in a humidified 5% CO2 incubator. For cytotoxicity tests, 8000 A549 cells per well were plated onto a white 96-well plate and incubated overnight for attachment. The wells were 70−80% confluent on the day of the experiment. The cells were then treated with 0.1−50 μM of (2), (5), (8), and (10) and incubated for another 24 h. Blank cells without treatment were used as positive control. After treatment, Cell-Titer Glo (Promega) cell viability assay kit was used to quantify the viability of the cell culture in each well, according to the manufacturer’s manual. Confocal Laser Scanning Microscope (CLSM). A549 cells were plated in a Coverglass Lab-Tek 8-well chamber (Nunc, Denmark) at a density of 30 000 cells per well in 300 μL of DMEM containing 10% FCS, 1% penicillin/streptomycin, and 1% MEM. The cells were incubated overnight at 37 °C in 5% CO2 to allow adhesion. The next day, 4 μg of rhodamine labeled (2) and (8) (see preparation in the Supporting Information) were added to cells, respectively, and

incubated for 24 h. Before imaging, cells were washed three times with PBS buffer. Cell nuclei were stained with 1 μL of DAPI solution (5 mg/mL) for 15 min, and the cell membrane was further stained with 0.5 μL of CellMask Deep Red plasma membrane stain solution for 5 min (0.5 mg/mL). Imaging was then performed using an LSM 710 CLSM system (Zeiss, Germany) coupled to an XL-LSM 710 S incubator and equipped with a 63× oil immersion objective. The DAPI nuclei stain and CellMask Deep Red plasma membrane stain was excited with a Diode405−30 laser and a HeNe633 laser, respectively, and the emission was collected using 410−582 nm and 647−759 nm filters, respectively. The fluorescence of rhodamine was recorded separately using a 531−648 nm filter and a 514 nm Argon laser for excitation. The acquired images were processed with Zen software developed by Carl Zeiss.

■

RESULTS AND DISCUSSION Preparation of Polypeptide Copolymers with Different Backbone Lengths. Polypeptide copolymers with different lengths of the main chains were achieved from three native precursor proteins by applying a “denaturationpegylation” strategy.16 The high-molecular-weight proteins human (HSA, 65 kDa) and bovine (BSA, 67 kDa) serum albumin and the smaller protein lysozyme from hen egg white (LY, 24 kDa) were selected as representative examples. The synthesis of protein-derived polypeptide copolymers was accomplished in the presence of high urea or guanidinium concentrations to destabilize the tertiary protein structure by diminishing secondary interactions and a reducing agent, such as tris(2-carboxyethyl)phosphine (TCEP) or DTT, to reduce all disulfide bridges, as well as in situ reaction of the reduced

1892

dx.doi.org/10.1021/bm300418r | Biomacromolecules 2012, 13, 1890−1898

Biomacromolecules

Article

Figure 1. Characterization of albumin- and LY-derived polypeptide copolymers. (a) SDS-PAGE of HSA, aHSA, cBSA, LY and their corresponding polypeptide copolymers (2), (5), (8) and (10). (b) MALDI-TOF MS spectra of native LY (calcd MW = 14.3 kDa) and dLY-PEO(5000)8 (10, calcd. MW = 54.3 kDa); matrix: sinapinic acid. (c) MALDI-TOF MS spectrum of aHSA compared with native HSA. (d) MALDI-TOF MS spectrum of (2). (e) Gel permeation chromatogram of dHSA-PEG(5000)27 (2) (compared with a commercial, monodisperse polyamidoamine “PAMAM” generation 4.0 dendrimer). The first signal corresponds to dHSA-PEG(5000)27 (2), and the second, lower signal refers to the commercial PAMAM 4.0 dendrimer. dHSA-PEG(5000)27 (2) has a similar elution profile as the commercial PAMAM dendrimer (which is considered to be monodisperse) supporting the narrow dispersity of (2).

thiol groups with O-(2-maleimidoethyl)-O′-methylpolyethylene(oxide)-5000 (PEO(5000)-MI) (Scheme 1). Via this procedure, the HSA-based copolymer (2)16 was achieved as previously reported carrying about 27 PEO chains, indicating that 27 out of 35 cysteine residues have reacted with PEO(5000)-MI. Because the presence of remaining free sulfhydryl groups has a negative impact on the long-term stability of such oligopeptides, N-(2-aminoethyl)maleimide was added after the pegylation step to increase further the shelf life of the copolymer. Quantitative capping of all free thiol groups was demonstrated by applying the 4,4-dipyridyl disulfide test (DPS test, Table S1, Supporting Information).13 After size exclusion chromatography, dHSA-PEO(5000) 27 (2) was isolated in moderate yields. The molecular weight of (2) was assessed by MALDI-TOF MS spectrum, SDS-PAGE (sodium dodecyl sulfate polyacrylamide gel electrophoresis), and GPC (Figure 1a,d,e), suggesting that a high-molecular-weight biopolymer of ∼200 kDa with a narrow molecular weight distribution was achieved.16 The polypeptide copolymer based on the smaller protein LY was synthesized by a similar strategy. Unlike HSA, LY required more drastic reaction conditions, for example, 8 M guanidinium hydrochloride, excess of the reduction reagent DTT, as well as prolonged reaction times (>24 h). Subsequent pegylation of all eight thiol groups of LY with PEO(5000)-MI and purification with size exclusion chromotography yielded dLY-PEO(5000)8 (10) with a

significantly shorter length of the main chain and a molecular weight of ∼54 kDa according to MALDI-TOF MS and SDSPAGE (Figure 1a,b), indicating that all eight thiol groups of the LY backbone have reacted with PEO(5000)-MI, thus resulting in the formation of a narrowly dispersed copolymer. Preparation of Polypeptide Copolymers from Priori Modified Proteins. Protein-derived copolymers bear a large number of reactive groups suitable for further chemical modifications. In this way, the introduction of functional groups could be accomplished by a priori modification of the native globular protein precursor before protein denaturation. For instance, charges play an important role in cell biology, and chemical modification of the serum albumin precursor allows the preparation of polycationic or polyanionic polypeptides that are attractive for cellular uptake, cell trafficking, targeting of subcellular structures,17 and complex formation with, for example, DNA.15 The synthesis of the polycationic albumin derivative cBSA-147 (cBSA, cationized bovine serum albumin with 147 primary amino groups on average) has been previously reported, which facilitates fast cell uptake and efficient DNA delivery15,18 as well as cell19 and vesicle immobilization20 to surfaces. cBSA (6) was achieved from native BSA after reacting accessible carboxylic acid residues with ethylenediamine,15 yielding a modified albumin derivative with high numbers of positive charges at physiological pH. Following a similar approach, the polyanionic serum albumin 1893

dx.doi.org/10.1021/bm300418r | Biomacromolecules 2012, 13, 1890−1898

Biomacromolecules

Article

Figure 2. (a) Amino acid sequence of lysozyme. Blue color denotes hydrophilic amino acid residues, purple color stands for lipophilic amino acids, and yellow color indicates cystein residues. In the case of LY, all eight cysteine residues carry a PEO chain. (b) Amino acid sequence of HSA. Colors have been allocated as under panel a, and red arrows indicate the presence of two neighboring cysteine residues. Because of steric hindrance, the PEO chain is most likely attached to only one of these cysteine residues. (c) Folding of the copolymers into micelles and formation of larger aggregates.

Figure 3. Secondary structure analysis of (a) (2), (b) (5), (c) (8), and (d) (10) compared with their corresponding protein precursors. The percentage of the secondary structure elements is calculated from circular dichroism (CD) spectra using CDSSTR program.

derivative aHSA (3) was prepared by reacting ∼40 accessible lysine residues with glutaric acid to introduce additional carboxylic acid groups. The globular precusor proteins can be

characterized conveniently via MALDI-TOF (Figure 1c), whereas it is very challenging to detect the polypeptide copolymers, most likely due to their high molecular weight as 1894

dx.doi.org/10.1021/bm300418r | Biomacromolecules 2012, 13, 1890−1898

Biomacromolecules

Article

Figure 4. Characterization of charge properties of polypeptide copolymers. (a) Agarose gel electrophoresis at pH 7.4 compares charge properties of HSA, aHSA, cBSA, LY, and their corresponding polypeptide copolymers (2), (5), (8), and (10). (b) Zeta potential of all polypeptide copolymers comparing with their precursor proteins.

example, helical and β sheet structures, due to noncovalent interactions among the amino acid residues. The presence of regular structural elements in polymers is highly attractive because secondary interactions could yield additional features such as self-assembly into ordered suprastructures such as amyloid fiber formation21 due to the presence of β-sheet structures. According to CD studies, the predominant secondary structure elements of the precursor proteins are partially conserved for all protein-derived polypeptide copolymers reported herein (Figure 3). For instance, HSA reveals a high content of α-helix (∼60%) and only 20% of β-sheet elements (Figure 3), and the corresponding polypeptide copolymer (2) also contains mainly α-helix (∼43%) structures. A similar trend is observed for (5), (8), and (10). However, in all cases, the polypeptide copolymers reveal an increase in random structures elements, most likely due to the denaturation and pegylation steps. Alternating Hydrophilic and Lipophilic Amino Acids Induce the Formation of Nanoscopic Micelles. According to the amino acid sequences of the native precursor proteins, all polypeptide copolymers contain hydrophobic and hydrophilic amino acids as well as PEO chains distributed along their main chain. The polarity of the side chains in proteins is essential for protein folding to adopt a defined and compact 3D structure in aqueous solution. Therefore, it is also very likely that polypeptide copolymers reveal micelle formation in aqueous solution, as known from other polypeptides.22 The precursor proteins HSA, aHSA, cBSA, and LY reveal hydrodynamic diameters (dH) of about 4.7, 9.9, 7.2, and 3.1 nm, respectively, according to the number-average diameter obtained by DLS. The extended structure of aHSA and cBSA compared with the native proteins is most likely due to electrostatic repulsion as well as a larger solvent shell, as previously discussed.15 In contrast, the size distribution by light scattering (LS) intensity yields slightly larger dH as well as a second population with dH between 80 and 150 nm. It has been previously described that serum albumins as well as LY form supramolecular structures in solution,23 which could be detected by DLS. These results again reflect the extreme sensitivity of DLS intensity toward large aggregates. The size distribution profiles of the polypeptide copolymers obtained by LS intensity indicate

well as fragmentation during the measurement.16 Therefore, in contrast with the functionalization of the polypeptide backbone, a priori modification of the native precursor proteins offers the great advantage that the number of functional groups could be assessed from the MALDI-TOF spectra (Figure 1c, representative MALDI-TOF spectra of cBSA have been previously published15). Chemically modified globular proteins such as aHSA (3) or cBSA (6) reveal well-defined signals in the MALDI-TOF spectra, which are of comparable resolution as those of native proteins. Preparation of denatured polypeptide copolymers from both modified albumin precursor proteins could be achieved via the similar procedure as discussed above for dHSA-PEO(5000)27 (2) (Scheme 1). The resulting polypeptide copolymers dcBSA-PEO(5000) 27 (8) and daHSA-PEO(5000)27 (5) exhibit similar molecular weight as dHSA-PEO(5000)27 (2) according to SDS-PAGE (Figure 1a). Polypeptide Copolymers Possess Highly Defined Structure and Maintain Secondary Elements of Their Precursor Proteins. Four polypeptide copolymers (2), (5), (8), and (10) derived from macromolecular but structurally distinct precursor proteins were prepared, revealing excellent water solubility (>20 mg/mL) and long shelf lives (>6 months) at 4 °C. In the case of dLY-PEO(5000)8 (10), the location of the PEO-chains of (10) is precisely defined due to the exactly known location of all eight cysteine residues along the main chain (Figure 2a). In the case of the albumin-derived polypeptides, about 27 out of 35 thiol groups have reacted with PEO(5000)-MI, suggesting a larger polydispersity of albumin-derived copolymers (2), (5), and (8) compared with (10). However, the attachment of about 27 substituents along the main chain is also very reproducible for other substituents than PEO,14 which might be explained in view of the polypeptide sequence because eight cysteine residues are located directly next to each other. Therefore, it is very unlikely that a second PEO chain is attached after the attachment of the first chain. In this way, even though the PEO substituents might not be located at the same cysteine residue, the number and distribution of the PEO chains along the backbone could still be considered to be well-defined. The primary amino acid sequence of the protein backbone contributes to the formation of the secondary structure, for 1895

dx.doi.org/10.1021/bm300418r | Biomacromolecules 2012, 13, 1890−1898

Biomacromolecules

Article

Table 1. Summary of Molecular Weight (M.W.), Zeta Potential, Size Distribution in Aqueous Solution, Complete Proteolysis Digestion Time, and Cell Uptake Efficiency of Various Polypeptide Copolymers and Their Precursor Proteins sample name HSA dHSA-PEO(5000)27 (2) cBSA (6) dcBSA-PEO(5000)27 (8) aHSA (3) daHSA-PEO(5000)27 (5) LY dLY-PEO(5000)8 (10)

zeta potential (mV)

diameter (H2O, nm)a

PDIb

b

66

−3.1 ± 1.3

4.7 ± 0.5

0.34

200b

−0.7 ± 1.7

13.1 ± 3.5

0.46

71b

22.7 ± 3.0

7.2 ± 0.5

0.27

205e

21.6 ± 1.5

15.4 ± 3.9

0.41

71b

−10.2 ± 3.7

9.9 ± 1.7

0.26

205e 14b

−22.3 ± 3.4 4.9 ± 0.6

119.8 ± 20.6 3.1 ± 0.9

0.27 0.35

54b

−8.4 ± 2.8

114.7 ± 11.5

0.28

M.W. (kDa)

diameter (H2O, nm)c 6.8 121.1 39.5 186.9 12.9 148.7 23.0 151.0 12.0 119 162.6 4.3 78.4 170.4

± ± ± ± ± ± ± ± ± ± ± ± ± ±

digestion time (min)d

0.1 9.1 4.3 15.2 1.2 36.3 3.7 28.3 0.9 37.8 18.3 0.8 7.4 3.4

cell uptake

/

−

90

−

/

+

30

+

/

/

120 /

/ /

60

/

According to DLS measurements based on number using ∼0.1 mg/mL of each protein in aqueous solution. Molecular weight obtained by MALDI-TOF spectra. cAccording to DLS measurements based on intensity using ∼0.1 mg/mL of each protein in aqueous solution. dDigestion time was estimated with 0.1% trypsin and analyzed by SDS-PAGE. Detailed SDS-PAGE pictures are shown in the Supporting Information (Figure S1). e Molecular weight according to SDS-PAGE and theoretical calculation. a

b

for aHSA (Figure 4b and Table 1). The surface zeta potential of the respective polypeptide copolymers revealed some differences compared with the precursor proteins: dHSAPEO(5000)27 (2) and dcBSA-PEO(5000)27 (8) reveal similar net charges as HSA and cBSA. In contrast, daHSAPEO(5000)27 (5) and dLY-PEO(5000)8 (10) bear high negative net charges. These findings underline that polypeptide polyelectrolytes with tunable charges and charge densities have been achieved. Polypeptides that allow fine-tuning of their surface charges are highly attractive to achieve biopolymers with tunable cell and tissue penetration properties. In addition, the presence of multiple functional groups located along the polypeptide backbone open several opportunities to introduce further functionalities to achieve sophisticated biohybrid materials such as polymer-drug micelles.13,14 Polypeptide Copolymers Are Noncytotoxic and Biodegradable and Cell Uptake Is Adjustable. For drug delivery applications, sufficient stability in the bloodstream, no or low cell toxicity, as well as biodegradability into nontoxic metabolites represent a key concern. Because polypeptide copolymers are derived proteins, they should offer promising biocompatibility, and they are degraded in cell endosome and lysosomes in the presence of high concentrations of proteases. We have investigated the cytoxicity of these copolymers in A549 cells at 0.1 and 1 μM concentrations, and no significant cytoxicity was observed for any copolymer (Figure 5). Moreover, after in vitro treatment with 0.1% trypsin, all copolymers (2), (5), (8), and (10) were digested into small peptide fragments within 2 h only (Table 1 and Figure S1in the Supporting Information). Interestingly, the polyanionic copolymer (5) exhibits a considerably slower digestion rate most likely due to the modification on lysine residues because trypsin is known to cleave peptide chains at the C-terminus of lysine and arginine.24 Notably, without the addition of proteases, all copolymers reveal high stability in the solid state as well as in aqueous solution for several months. Cell uptake of positively charged and negatively charged copolymers has been studied by applying rhodamine-labeled derivatives in A549 cells. It has been previously shown that cBSA is quickly uptaken by clathrin-mediated endocytosis and

that two populations of (2) and (8) coexist in aqueous solution. The smaller species with diameters of about 39.5 nm (2) and 23.0 nm (8) are most likely single-chain macromolecules, whereas the larger polypeptide aggregates exhibit diameters of about 187.0 (2) and 151.0 nm (8). In the case of (5) and (10), only polypeptide aggregates of 162.6 and 170.4 nm were observed, respectively. If the number distribution is considered, then the majority of species of (2) and (8) in solution are present as single chains with average dH of 13.1 and 15.4 nm, indicating that the number of polypeptide aggregates generated is much smaller than the number of nonaggregated, single-chain polypeptides. Differences in the sizes of the different polypeptide species found in solution can be due to a variety of effects ranging from electrostatic repulsion of the negatively charged backbone, leading to coil expansion, different levels of counterion condensation depending on the average spacing between backbone charges, a hydrophobic collapse due to a larger content of hydrophobic amino acids, or a higher degree of more compact secondary structures such as α-helix or βsheet structures.22 This property could be highly attractive to design efficient nanocarriers for medicinal applications. Polypeptide Copolymers Contain Multiple Positive and Negative Charges along the Backbone. The distinct differences of the net surface charges of cBSA and aHSA compared with native HSA were demonstrated by gel electrophoresis at pH 7.4 (Figure 4a): cBSA shifts toward the negative electrode with strong tailing due to the interaction with the agarose gel, whereas aHSA moves toward positive electrode, indicating a negative net charge. Native HSA also bears slightly negative net charge, therefore moving toward positive electrode, however, with significantly slower speed compared with aHSA. dcBSA-PEO(5000)27 (8), daHSAPEO(5000)27 (5), and dHSA-PEO(5000)27 (2) reveal similar characteristics at physiological pH (pH 7.4) as their native precursor proteins, suggesting similar net charges. The surface charges of all polypeptide copolymers were also characterized by zeta potential measurements, and the results are summarized in Figure 4b and Table 1. A positive surface zeta potential of 22.7 and 4.9 was obtained for cBSA and LY, respectively, whereas a negative zeta potential of −10 was found 1896

dx.doi.org/10.1021/bm300418r | Biomacromolecules 2012, 13, 1890−1898

Biomacromolecules

■

Article

CONCLUSIONS

In conclusion, we have presented a general reaction scheme toward various large polypeptides with highly defined structures and molecular weights as well as attractive physical and biological properties. These polypeptides are prepared by controlled denaturation of native proteins and in situ stabilization with several PEO(5000) chains. In this way, polypeptides with different chain lengths (e.g., HSAPEO(5000)27 and LY-PEO(5000)8), varying net charges (e.g., HSA-PEO(5000) 27 , cBSA-PEO(5000) 2 7 , and aHSAPEO(5000)27), as well as the presence of ordered secondary structural elements have been achieved successfully. This reaction scheme could, in principle, be applied for other precursor proteins, but the denaturing conditions required to achieve polypeptide copolymers might vary depending on the stability of the respective native protein. The presence of alternating hydrophobic and hydrophilic “patches” along the polypeptide backbone contributes to the formation of micellar architectures in diluted aqueous solution, which is attractive for drug encapsulation and drug delivery. Diverse functional groups available at precise positions along protein backbone enable further orthogonal modifications. All of these copolymers reveal excellent solubility and stability in aqueous media, no cytotoxicity at relevant concentrations, as well as fast enzymatic degradation, indicating their promising potential as nanotransporter scaffolds for medicinal applications. Because biomedical applications often impose that no immune responses are elicited, polypeptide copolymers derived from human proteins such as (2) or (5) could be particularly attractive for therapeutic applications. In view of these unique features, protein-derived polypeptide copolymers represent attractive and versatile biomaterials for drug delivery, nanopatterning due to self-assembly, as well as tissue engineering. The “semi-chemistry” polypeptide preparation approach reported herein emphasizes a novel concept for producing synthetic polypeptides from native protein resources, which could complement traditional polypeptide synthesis and expression and open access to unique macromolecular architectures.

Figure 5. Cell viability assay showing no significant cytotoxicity of polypeptide copolymer (2), (5), (8), and (10).

released into the cytoplasm;15 in contrast, HSA reveals no significant cellular uptake. The corresponding copolymers display a very similar uptake profile. By CLSM study, cBSA derived cationic polypeptide (8) was significantly uptaken by cells, whereas HSA-derived polyanionic polypeptide (2) was not found inside the cell, most likely due to repulsive forces between the biopolymer and the negatively charged cellular membranes (Figure 6). In this way, cell uptake of the copolymers could be fine-tuned by attaching, for example, positively charged functionalities to the precursor proteins. There are several applications, where cell uptake of the copolymer might not be desirable, for example, if extracellular targets such as membrane receptors or structures of the extracellular matrix should be addressed for bioimaging purposes. In this case, high nonspecific cell uptake might lead to reduced contrast, and undesired intracellular effects might be evoked. In contrast, efficient cell uptake and low proteolytic stability are particularly attractive for drug delivery systems, and here cBSA represents an attractive platform for transporting drug molecules into cells.

Figure 6. CLSM imaging of rhodamine-labeled (2) (B1−B5) and (8) (A1-A5) incubated with A549 cells for 24 h. A1 and B1 represent bright field images; A2 and B2 are cell nucleus stained by DAPI nucleus staining; A3 and B3 are cell membranes stained by CellMask Deep Red plasma membrane stain; A4 and B4 are fluorescence channel for rhodamine emission; and A5 and B5 are overlay of channel 2 to 4. 1897

dx.doi.org/10.1021/bm300418r | Biomacromolecules 2012, 13, 1890−1898

Biomacromolecules

■

Article

Hest, J. C.; Tirrell, D. A. Protein-based materials, toward a new level of structural control. Chem. Commun. (Cambridge, U. K.) 2001, 2001, 1897−904. (10) Brunsveld, L.; Kuhlmann, J.; Alexandrov, K.; Wittinghofer, A.; Goody, R. S.; Waldmann, H. Lipidated ras and rab peptides and proteins--synthesis, structure, and function. Angew. Chem., Int. Ed. 2006, 45, 6622−46. (11) Klok, H.-A. Protein-inspired materials: synthetic concepts and potential applications. Angew. Chem., Int. Ed. 2002, 41, 1509−1513. (12) (a) Calmettes, P.; Durand, D.; Desmadril, M.; Minard, P.; Receveur, V.; Smith, J. C. How random is a highly denatured protein? Biophysical Chemistry 1994, 53, 105−113. (b) Dill, K. A.; Shortle, D. Denatured states of proteins. Annu. Rev. Biochem. 1991, 60, 795−825. (13) Wu, Y.; Chakrabortty, S.; Gropeanu, R. A.; Wilhelmi, J.; Xu, Y.; Er, K. S.; Kuan, S. L.; Koynov, K.; Chan, Y.; Weil, T. pH-Responsive quantum dots via an albumin polymer surface coating. J. Am. Chem. Soc. 2010, 132, 5012−5014. (14) Wu, Y.; Er, K. S.; Ramanathan, A.; Vasudevan, S.; Weil, T., Nano-sized albumin-copolymer micelles for efficient doxorubicin delivery. Biointerphases 2012, 7, DOI: 10.1007/s13758-011-0005-7. (15) Eisele, K.; Gropeanu, R. A.; Zehendner, C. M.; Rouhanipour, A.; Ramanathan, A.; Mihov, G.; Koynov, K.; Kuhlmann, C. R. W.; Vasudevan, S. G.; Luhmann, H. J.; Weil, T. Fine-tuning DNA/albumin polyelectrolyte interactions to produce the efficient transfection agent cBSA-147. Biomaterials 2010, 31, 8789−8801. (16) Li, L. Overview of MS and MALDI MS for Polymer Analysis; John Wiley & Sons, Inc.: Hoboken, NJ, 2009; p 8. (17) (a) Boddohi, S.; Kipper, M. J. Engineering nanoassemblies of polysaccharides. Adv. Mater. 2010, 22, 2998−3016. (b) He, Q.; Cui, Y.; Li, J. Molecular assembly and application of biomimetic microcapsules. Chem. Soc. Rev. 2009, 38, 2292−303. (18) Zöphel, L.; Eisele, K.; Gropeanu, R.; Rouhanipour, A.; Koynov, K.; Lieberwirth, I.; Müllen, K.; Weil, T. Preparation of defined albumin−polymer hybrids for efficient cell transfection. Macromol. Chem. Phys. 2010, 211, 146−153. (19) Ng, J. F.; Jaenicke, S.; Eisele, K.; Dorn, J.; Weil, T. cBSA-147 for the preparation of bacterial biofilms in a microchannel reactor. Biointerphases 2011, 5, FA41−47. (20) Ritz, S.; Eisele, K.; Dorn, J.; Ding, S.; Vollmer, D.; Putz, S.; Weil, T.; Sinner, E. K. Cationized albumin-biocoatings for the immobilization of lipid vesicles. Biointerphases 2010, 5, FA78−87. (21) Dong, H.; Hartgerink, J. D. Role of hydrophobic clusters in the stability of α-helical coiled coils and their conversion to amyloid-like βsheets. Biomacromolecules 2007, 8, 617−623. (22) Siddique, B.; Duhamel, J. Effect of polypeptide sequence on polypeptide self-assembly. Langmuir 2011, 27, 6639−6650. (23) Georgalis, Y.; Umbach, P.; Raptis, J.; Saenger, W. Lysozyme aggregation studied by light scattering. i. influence of concentration and nature of electrolytes. Acta Crystallogr., Sect. D 1997, 53, 691−702. (24) Strader, M. B.; Tabb, D. L.; Hervey, W. J.; Pan, C.; Hurst, G. B. Efficient and specific trypsin digestion of microgram to nanogram quantities of proteins in organic−aqueous solvent systems. Anal. Chem. 2005, 78, 125−134.

ASSOCIATED CONTENT

S Supporting Information *

SDS-PAGE pictures of all copolymers digested by trypsin, DLS diagrams, and preparation of rhodamine-labeled dHSAPEO(5000)27 (2) and dcBSA-PEO(5000)27 (8). This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Author Contributions §

Shared first authorship due to equal contribution.

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge technical support from Ms. Magdalene Zimmermann and financial support from German Research Foundation (DFG) under the SFB 625 grant P3246 029.

■

REFERENCES

(1) (a) Collier, J. H.; Rudra, J. S.; Gasiorowski, J. Z.; Jung, J. P. Multicomponent extracellular matrices based on peptide self-assembly. Chem. Soc. Rev. 2010, 39, 3413−24. (b) de la Rica, R.; Matsui, H. Applications of peptide and protein-based materials in bionanotechnology. Chem. Soc. Rev. 2010, 39, 3499−509. (c) Lowik, D. W.; Leunissen, E. H.; van den Heuvel, M.; Hansen, M. B.; van Hest, J. C. Stimulus responsive peptide based materials. Chem. Soc. Rev. 2010, 39, 3394−412. (2) (a) Cui, H.; Muraoka, T.; Cheetham, A. G.; Stupp, S. I. Selfassembly of giant peptide nanobelts. Nano Lett. 2009, 9, 945−951. (b) Verch, A.; Hahn, H.; Krause, E.; Colfen, H.; Börner, H. G. A modular approach towards functional decoration of peptide-polymer nanotapes. Chem. Commun. (Cambridge, U. K.) 2010, 46, 8938−8940. (3) (a) Apostolovic, B.; Deacon, S. P.; Duncan, R.; Klok, H. A. Hybrid polymer therapeutics incorporating bioresponsive, coiled coil peptide linkers. Biomacromolecules 2010, 11, 1187−95. (b) Deacon, S. P.; Apostolovic, B.; Carbajo, R. J.; Schott, A. K.; Beck, K.; Vicent, M. J.; Pineda-Lucena, A.; Klok, H. A.; Duncan, R. Polymer coiled-coil conjugates: potential for development as a new class of therapeutic “molecular switch”. Biomacromolecules 2011, 12, 19−27. (4) (a) Gil, E. S.; Mandal, B. B.; Park, S. H.; Marchant, J. K.; Omenetto, F. G.; Kaplan, D. L. Helicoidal multi-lamellar features of RGD-functionalized silk biomaterials for corneal tissue engineering. Biomaterials 2010, 31, 8953−63. (b) Sahoo, S.; Toh, S. L.; Goh, J. C. A bFGF-releasing silk/PLGA-based biohybrid scaffold for ligament/ tendon tissue engineering using mesenchymal progenitor cells. Biomaterials 2010, 31, 2990−8. (5) (a) van Dongen, S. F. M.; de Hoog, H.-P. M.; Peters, R. J. R. W.; Nallani, M.; Nolte, R. J. M.; van Hest, J. C. M. Biohybrid polymer capsules. Chem. Rev. 2009, 109, 6212−6274. (b) Wenk, E.; Wandrey, A. J.; Merkle, H. P.; Meinel, L. Silk fibroin spheres as a platform for controlled drug delivery. J. Controlled Release 2008, 132, 26−34. (6) Hunter, A. C.; Moghimi, S. M. Therapeutic synthetic polymers: a game of Russian roulette? Drug Discovery Today 2002, 7, 998−1001. (7) (a) Hadjichristidis, N.; Iatrou, H.; Pitsikalis, M.; Sakellariou, G. Synthesis of well-defined polypeptide-based materials via the ringopening polymerization of α-amino acid N-carboxyanhydrides. Chem. Rev. 2009, 109, 5528−5578. (b) Broyer, R. M.; Grover, G. N.; Maynard, H. D. Emerging synthetic approaches for protein-polymer conjugations. Chem. Commun. (Cambridge, U. K.) 2011, 47, 2212−26. (8) Yang, J.; Gitlin, I.; Krishnamurthy, V. M.; Vazquez, J. A.; Costello, C. E.; Whitesides, G. M. Synthesis of monodisperse polymers from proteins. J. Am. Chem. Soc. 2003, 125, 12392−12393. (9) (a) Rabotyagova, O. S.; Cebe, P.; Kaplan, D. L. Protein-based block copolymers. Biomacromolecules 2011, 12, 269−289. (b) van 1898

dx.doi.org/10.1021/bm300418r | Biomacromolecules 2012, 13, 1890−1898