Synthesis and Characterization of Recombinant Abductin-Based

Oct 22, 2013 - Stefan Roberts , Michael Dzuricky , Ashutosh Chilkoti ... Nathan P. Cowieson , Christopher M. Elvin , Anita J. Hill , Namita R. Choudhu...
0 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/Biomac

Synthesis and Characterization of Recombinant Abductin-Based Proteins Renay S.-C. Su,† Julie N. Renner,† and Julie C. Liu*,†,‡ †

School of Chemical Engineering, Purdue University, West Lafayette, Indiana 47907-2100, United States Weldon School of Biomedical Engineering, Purdue University, West Lafayette, Indiana 47907-2032, United States

‡

S Supporting Information *

ABSTRACT: Recombinant proteins are promising tools for tissue engineering and drug delivery applications. Protein-based biomaterials have several advantages over natural and synthetic polymers, including precise control over amino acid composition and molecular weight, modular swapping of functional domains, and tunable mechanical and physical properties. In this work, we describe recombinant proteins based on abductin, an elastomeric protein that is found in the inner hinge of bivalves and functions as a coil spring to keep shells open. We illustrate, for the first time, the design, cloning, expression, and purification of a recombinant protein based on consensus abductin sequences derived from Argopecten irradians. The molecular weight of the protein was confirmed by mass spectrometry, and the protein was 94% pure. Circular dichroism studies showed that the dominant structures of abductin-based proteins were polyproline II helix structures in aqueous solution and type II β-turns in trifluoroethanol. Dynamic light scattering studies illustrated that the abductin-based proteins exhibit reversible upper critical solution temperature behavior and irreversible aggregation behavior at high temperatures. A LIVE/DEAD assay revealed that human umbilical vein endothelial cells had a viability of 98 ± 4% after being cultured for two days on the abductin-based protein. Initial cell spreading on the abductin-based protein was similar to that on bovine serum albumin. These studies thus demonstrate the potential of abductin-based proteins in tissue engineering and drug delivery applications due to the cytocompatibility and its response to temperature.

■

INTRODUCTION

tunable Young’s modulus that matched the mechanical properties of human vocal fold tissue.2 In the field of drug delivery, sequence design can be used to modulate the molecular architecture and the environmentally induced responsiveness of protein-based materials.3 In this paper, we focus on developing and characterizing a recombinant protein consisting of an abductin-based mechanical domain. Previous studies have identified repetitive sequences that contribute to the superior mechanical properties of natural elastomeric proteins, such as elastin,4−6 resilin,7−9 spider silk,10−12 titin,13−15 and abductin.16 These repetitive sequences have been widely utilized as mechanical domains to confer the desired mechanical properties to recombinant

Recombinant protein-based biomaterials have been widely used in fields such as tissue engineering and regenerative medicine.1 They have advantages over natural and synthetic materials, including precise control over molecular weight and amino acid sequence. Furthermore, the exquisite control over sequence allows one to specifically tune the physical and mechanical properties for the desired application. In addition, the modularity of recombinant proteins allows incorporation of different domains, such as mechanical domains or bioactive domains. Mechanical domains confer structural integrity to proteins, whereas bioactive domains provide cellular cues, such as cell adhesion domains or sequences derived from growth factors. All of these properties make recombinant proteins promising candidates for applications in tissue engineering and drug delivery systems. For example, recombinant proteins based on a repetitive resilin sequence were designed to have a © 2013 American Chemical Society

Received: August 4, 2013 Revised: October 7, 2013 Published: October 22, 2013 4301

dx.doi.org/10.1021/bm401162g | Biomacromolecules 2013, 14, 4301−4308

Biomacromolecules

Article

proteins.1,17 Recent studies demonstrate the ease with which mechanical properties of protein-based hydrogels can be tailored by varying factors such as protein concentration and the ratio between cross-linking sites and cross-linker.18−24 Thus, protein-based biomaterials with a wide range of mechanical properties can be achieved for various applications in tissue engineering and drug delivery. Among natural elastomeric proteins, abductin has not been as well studied. Natural abductin was first identified in the abductor ligament located in the internal hinge of bivalves such as scallops and clams.25 It has rubber-like properties, serves as a compression spring to keep the shells open while the adductor muscle relaxes,26 and plays an important role in the swimming mechanism of scallops.27−29 Natural abductin has a tensile modulus of 1.25 MPa,25 which is higher than elastin (0.3−0.6 MPa)30 but on the same order of magnitude as resilin (0.6−2 MPa).31,32 On the other hand, natural abductin has a compressive modulus of 4 MPa, which is higher than resilin (0.6−0.7 MPa).9,25,32 The superior mechanical properties of natural abductin offer the potential of designing protein-based biomaterials that can be utilized in a broader number of applications. Consensus amino acid sequences derived from Argopecten irradians are rich in glycine (57.3%) and methionine (14.3%).16 Sequences such as MGGG, FGGMG, FGGMGGG, GGFGGMGGG, and FGGMGGGNAG are repeated throughout the entire abductin gene.33 The secondary structures of synthetic peptides based on repetitive sequences from natural abductin were characterized.33,34 There are two different structures that dominate, depending on the solvent. Most synthetic abductin-based peptides adopt polyproline II (PPII) structures, which are left-handed helices, in aqueous solution, whereas they have type II β-turns in trifluoroethanol (TFE), which is a more hydrophobic solvent.33 Coexistence of PPII and type II β-turns and temperature-induced multiconformational transitions were observed with longer synthetic abductinlike peptides such as (FGGMGGGNAG)4 in hexafluoroisopropanol (HFIP).34 To our knowledge, this work is the first study to describe the synthesis and expression of recombinant abductin-based proteins. We designed the sequence based on the repetitive decapeptide found in abductin derived from A. irradians. Abductin-based DNA repeats were seamlessly multimerized using recursive cloning. We characterized the secondary structures and temperature-induced aggregation behavior of the abductin-based protein. We also assessed its cytocompatibility and initial interactions with human umbilical vein endothelial cells (HUVECs).

■

England BioLabs, Ipswich, MA) were chosen as specific restriction sites. Standard cloning techniques were performed.36 Protein Expression. Recombinant plasmids containing the final DNA construct were transformed into E. coli strain BL21-CodonPlus(DE3)-RIPL. Cells were grown overnight at 37 °C in 2xYT medium containing kanamycin (25 μg/mL) and chloramphenicol (34 μg/mL). Overnight cultures were diluted at a ratio of 1:250 and used to inoculate a fermentor (BioFlo, BL110, 14 L capacity, New Brunswick, Enfield, CT) containing 10 L of Terrific Broth (TB). Cell growth was monitored by optical density (O.D.) at 600 nm. When the O.D. at 600 nm reached 4−5, a final concentration of 2.5 mM isopropyl β-D-1thiogalactopyranoside (EMD Chemicals) was added to induce protein expression. The culture temperature was lowered to 20 °C after induction. Cells were further cultured for 4−5 h. Cells were harvested by centrifugation at 8000g for 23 min at 4 °C, and the cell pellet was stored at −80 °C before purification. Protein Purification. The cell pellet was resuspended in lysis buffer (8 M urea, 100 mM NaH2PO4, 100 mM Tris-Cl (J.T. Baker, Phillipsburg, NJ), pH 8.0) with addition of at least 0.5 mM phenylmethanesulfonylfluoride (PMSF). The cell suspension was subjected to at least 3 freeze−thaw cycles and sonicated with a Misonix XL-2000 (Qsonica, Newtown, CT) for 1 min and then cooled on ice for 1 min for a total of ≥2 h. To separate cell debris from the supernatant, the cell lysate was centrifuged at 10 000g for 20 min at 4 °C. To remove undesired bacterial proteins, 15 wt % ammonium sulfate was added to the supernatant. Additional ammonium sulfate was added to a final concentration of 30 wt % to precipitate the abductinbased protein. The protein pellet was resuspended in Milli-Q water (Millipore, Billerica, MA) at a concentration of up to 250 mg/mL (based on the wet weight of the pellet) and incubated at 80 °C for 15 min. Undesired bacterial proteins aggregated at this high temperature and were separated from the supernatant through centrifugation. The supernatant containing the abductin-based protein was further purified on an Ni-NTA agarose-packed column (Qiagen, Germantown, MD) under denaturing conditions. The supernatant was applied to the column twice. Sequential washes were carried out with the lysis buffer containing 50 mM or 75 mM imidazole (EMD Chemicals) (pH 6.3). The abductin-based protein was eluted with the lysis buffer containing 250 mM imidazole (pH 4.5). Dialysis (MWCO 8000, Spectrum Laboratories, Rancho Dominguez, CA) against reverse osmosis water was carried out at 4 °C to remove residue salt. The proteins were freeze-dried. Protein expression and purification were confirmed by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and Western blot using standard techniques.37 The abductin-based protein was detected using an anti-T7 tag antibody conjugated to horseradish peroxide (EMD Chemicals) and a chemiluminescent agent (Pierce ECL substrate, Thermo Scientific, Rockford, IL) or a colorimetric substrate (3, 3′,5,5′-tetramethylbenzidine, Kirkegaard & Perry Laboratories, Gaithersburg, MD). Coomassie Brilliant Blue R-250 was used to stain SDS-PAGE gels. The purity of the abductin-based protein was determined through densitometry analysis using ImageJ (NIH, Bethesda, MD).38 The molecular weight and the amino acid composition of purified protein samples were verified by matrix assisted laser desorption ionization−time-of-flight (MALDI-TOF) mass spectroscopy and amino acid analysis, respectively. Both measurements were performed by the Purdue Proteomic Facility in the Bindley Bioscience Center. Circular Dichroism (CD). CD spectroscopy was used to determine the secondary structure of the abductin-based protein in solution (0.3 mg/mL of protein in Milli-Q water and 0.2 mg/mL of protein in TFE). Scans were performed on a Jasco-815 circular dichroism spectrometer using a 1 mm path length, a data pitch of 1 nm, a bandwidth of 2.00 nm, and a scanning speed of 200 nm/min. The resulting spectrum was an average of five scans. The baseline was an average of five scans of either water or TFE. Dynamic Light Scattering (DLS). Thermal aggregation behavior of the abductin-based protein was assessed using a Zetasizer Nano S dynamic light scattering instrument (Malvern Instruments, UK). The

MATERIALS AND METHODS

Chemicals and Bacteria Strains. All chemicals were purchased from Sigma-Aldrich (St. Louis, MO) unless mentioned otherwise. Escherichia coli strains BL21(DE3) and BL21-CodonPlus-(DE3)-RIPL were generous gifts from Dr. Jo Davisson (Purdue University), BL21(DE3)pLysS was a gift from Dr. Chongli Yuan (Purdue University), and Rosetta2(DE3)pLysS was purchased from EMD Chemicals (Gibbstown, NJ). DNA Cloning. The DNA sequence was designed and optimized using an online optimizer for E. coli (http://genomes.urv.cat/ OPTIMIZER/). We modified the cloning scheme developed by Renner et al.35 Abductin-based DNA inserts were first ligated into pAL (a modified pUC19 vector with an extra linker). To seamlessly obtain multiple abductin-like sequences, enzymes SgrAI and BspEI (New 4302

dx.doi.org/10.1021/bm401162g | Biomacromolecules 2013, 14, 4301−4308

Biomacromolecules

Article

Figure 1. A recombinant abductin-based protein (AB12) was designed and cloned successfully. (A) Full amino acid sequence of AB12. AB12 includes a mechanical domain with 12 abductin-based repeats, a T7 tag for antibody detection, a heptahistidine tag (His tag) for purification purposes, and an enterokinase cleavage site that allows removal of all tags. Both the N-terminal and C-terminal sequences flanking the abductin-like domain contain restrictions sites that can potentially be used to incorporate other functional domains into AB12. (B) Recursive cloning scheme used to seamlessly multimerize abductin repeats (in gray). The restriction sites SgrAI and BspEI are compatible and ablate after ligation. (C) The number of abductin repeats can be tuned precisely. Plasmids containing 3, 6, and 12 abductin repeats were digested with EcoRI and BamHI, and the products were separated by agarose gel electrophoresis. The expected insert sizes of 157, 247, and 427 base pairs were observed. on TCP were used as a positive control. HUVEC viability after 2 d was quantified using a LIVE/DEAD Viability/Cytotoxicity Kit (Molecular Probes L-3224, Life Technologies, Carlsbad, CA). Cells were rinsed with excess PBS, incubated at 37 °C with a solution of 0.5 μM calceinAM and 1.5 μM ethidium homodimer-1 in PBS for 45 min, and rinsed again before imaging. PBS supplemented with 0.49 mM MgCl2 and 0.90 mM CaCl2 was used to prevent cell detachment. Images were taken using a Nikon Ti-E microscope with a 4x objective lens. Fluorescein isothiocyanate (FITC) and tetramethyl rhodamine isothiocyanate (TRITC) standard filter cubes were used to detect calcein AM and ethidium homodimer-1 signals. Each group had three replicates, and each replicate contained at least 2000 cells. Cells incubated with 95% ethanol at 37 °C for 30 min served as a negative control. The numbers of live and dead cells were determined using NIS-Elements software (Nikon Instruments, Melville, NY). HUVEC Spreading. HUVECs were cultured on TCP (positive control), adsorbed protein surfaces, or BSA-blocked surfaces (negative control) in a 24-well plate at a density of 2500 cells/cm2. Phase contrast images were taken using a 10x objective lens at 15, 60, 95, and 180 min after cells were seeded. Cell area was determined by outlining the boundary of each cell using ImageJ software.38 Each treatment had three replicates, and each replicate consisted of at least 100 cells. Statistical Analysis. All data are presented as the mean ± standard deviation. Comparisons of cell viability between the positive control and cells grown on adsorbed abductin were analyzed using a twotailed, unpaired t test (Excel). Cell areas on different surfaces at a single time point were analyzed using a one-way analysis of variance (ANOVA). Tukey’s post hoc analysis was performed to organize treatments into statistically different (p < 0.05) subgroups using JMP 8 (SAS, Cary, NC). For all statistical tests, a threshold of α = 0.05 was chosen.

protein solution (10 mg/mL in Milli-Q water) was cooled from room temperature to 2 °C in the precooled DLS chamber. The temperature was increased or decreased in intervals of 1 °C. At each temperature, three replicates, which were composed of ten 1-min measurements, were taken. Data were reported as an average of three replicate measurements ± standard deviation. Oscillatory Rheology. Rheological measurements of the abductinbased protein were performed on a stress-controlled AR2000 rheometer (TA Instruments, New Castle, DE) using a 15-mm diameter plate-on-plate geometry. Protein solutions (10 mg/mL in Milli-Q) were made at room temperature and loaded onto the precooled bottom plate (2 °C). The top plate was then lowered to a gap distance of 825 μm. Low-viscous mineral oil was applied to the edge of the protein solution to prevent evaporation. After the sample had equilibrated at 2 °C for 1 min, dynamic oscillatory time and frequency sweeps were performed. Oscillatory time sweeps were performed at an angular frequency of 1 rad/s and at 1% strain. Frequency sweeps were conducted from 0.1−55 rad/s at 1% strain. Data are reported as an average of three replicates ± standard deviation. Human Umbilical Vein Endothelial Cell Culture. Human umbilical vein endothelial cells (HUVECs, Lonza, Walkersville, MD) were cultured at 37 °C and 5% CO2 in endothelial cell growth medium-2 (EGM-2, Lonza, Walkersville, MD). Cells were subcultured after reaching 70−80% confluence. Passage 4 cells were used in all experiments. Adsorbed Protein Surface. Abductin-based protein solutions (1 mg/mL in phosphate buffered saline (PBS)) were sterile-filtered and adsorbed onto tissue culture polystyrene (TCP, BD Falcon, Durham, NC) surfaces at 4 °C overnight. Adsorbed protein surfaces were rinsed with excess PBS twice and blocked with bovine serum albumin (BSA, fraction V, EMD Chemicals, 2 mg/mL in PBS, heat-inactivated at 85 °C for 10 min, sterile-filtered) for 30 min at room temperature. Adsorbed protein surfaces were rinsed with excess PBS before cell culture. The amount of abductin-based protein adsorbed on the surface was quantified using a bicinchoninic acid (BCA) assay (Thermo Scientific). BSA was used to develop a linear standard curve for absorbance readings of 0.002−0.42. The absorbance reading of adsorbed AB12 was 0.07, which corresponded to a density of 203 pmol/cm2. HUVEC Viability. HUVECs were cultured on adsorbed protein surfaces in 24-well plates at a density of 2500 cells/cm2. Cells cultured

■

RESULTS AND DISCUSSION Protein Sequence Design. A modular scheme for cloning recombinant protein-based biomaterials was previously designed by our lab.35,39 The N-terminus of our recombinant proteins contains a T7-tag for antibody detection and a heptahistidine tag (His tag) for purification purposes. The proteins also possess a mechanical domain with repetitive amino acid sequences that contribute to the mechanical and physical properties. In this work, an abductin-based protein, 4303

dx.doi.org/10.1021/bm401162g | Biomacromolecules 2013, 14, 4301−4308

Biomacromolecules

Article

Figure 2. Successful purification of AB12. (A) SDS-PAGE and (B) Western blotting showed successful purification of AB12 (1 mg/mL in Milli-Q water). The purity of AB12 was calculated to be 94% using densitometry. (C) The molecular weight of AB12 was confirmed by MALDI-TOF mass spectroscopy. AB12 has an expected molecular weight of 14.639 kDa, and the spectrum shows a peak at 14.658 kDa, which is within 0.06% of the expected value. Peaks at 29.312 kDa and 7.322 kDa represent dimers and doubly charged molecules, respectively.

chromatography was therefore performed to further purify AB12. SDS-PAGE and Western blotting were used to confirm successful purification of AB12 (Supporting Information, Figure S2, Figure 2A,B). The final purity of AB12 was 94% as assessed by densitometry analysis. The average yield of purified AB12 was 1.4 mg/L, which is relatively low compared to the lowest yields reported for other recombinant elastomeric proteins based on elastin (25−30 mg/L)48 and resilin (14 mg/ L).49 We hypothesize that the low yield of AB12 might result from its high glycine content (∼60%). Other glycine-rich proteins, such as proteins based on spider silk sequences, have low yields in E. coli systems.50 A recent study to improve the expression level of spider silk proteins engineered an E. coli system with elevated levels of tRNAs that recognize glycine codons and increased the glycine pool by overexpressing serine hydroxymethyltransferase, an enzyme that converts serine to glycine.51 This strategy resulted in enhanced expression levels of spider silk proteins, and application of this method could potentially improve the yield of AB12. The molecular weight of AB12 (expected molecular weight of 14.639 kDa) was confirmed by MALDI-TOF mass spectroscopy, which detected a molecular weight within 0.06% of the expected molecular weight (Figure 2C). The composition of AB12 was verified by amino acid analysis (Table 1).

AB12, was designed to have 12 repeats of the decapeptide FGGMGGGNAG, which is derived from the natural abductin sequence from A. irradians.16 The full amino acid sequence of AB12 is shown in Figure 1A. Within the abductin-based domain, one out of every three asparagine (N) residues was replaced with a lysine (K) residue to serve as potential crosslinking sites. The phenylalanine (F) residues also serve as a potential cross-linking sites; through genetic engineering, E. coli can replace phenylalanine with the noncanonical amino acid, para-azidophenylalanine,40,41 which is sensitive to ultraviolet light and can be utilized to photocross-link protein-based materials.42,43 Protein Cloning, Expression, and Purification. Molecular cloning techniques can be used to precisely control the number of desired repeats in recombinant proteins.44 Selection of restriction sites allowed seamless multimerization of the abductin-based repeats (Figure 1B). Vectors containing 3, 6, and 12 abductin-based repeats were obtained (Figure 1C). Protein expression was assessed in four different E. coli strains, including BL21(DE3), BL21(DE3)pLysS, Rosetta2(DE3)pLysS, and BL21-CodonPlus-(DE3)-RIPL, and the BL21CodonPlus-(DE3)-RIPL strain had the highest expression of AB12 (data not shown). Western blotting confirmed successful expression (Supporting Information, Figure S1). Purification of AB12 was achieved by using a combination of salting-out, heating, and affinity chromatography. The saltingout and heating methods were developed based on a facile purification method of resilin-based proteins.35,39,45 Heat stability is characteristic of some elastomeric proteins such as resilin9 and spider silk.46 Therefore, heating has been used as part of the purification process of resilin-based proteins;35,45,47 upon heating, other proteins aggregate, whereas resilin-based proteins remain soluble in the supernatant. The thermal stability of AB12 was assessed by heating aqueous solutions of AB12 to high temperatures. AB12 stayed in solution and did not aggregate after heating for 5 min at temperatures ranging from 50−80 °C (data not shown). Thus, after proteins were salted out, solutions containing the abductin-based proteins were heated to further purify AB12. After salting-out and heating, there were still multiple bands that appeared on SDS-PAGE gels (Supporting Information, Figure S2A). Thus, additional purification was needed. Affinity

Table 1. Amino Acid Analysis Results of AB12 Confirm Successful Manufacture of AB12

ASX SER GLX GLY HIS ARG THR ALA MET LYS LEU PHE 4304

observed mol %

calculated mol %

8.59 1.47 2.19 45.82 6.33 2.31 1.46 8.84 10.66 3.34 1.82 7.17

8.59 1.23 1.84 47.85 5.52 1.84 1.23 8.59 11.04 3.07 1.84 7.36

dx.doi.org/10.1021/bm401162g | Biomacromolecules 2013, 14, 4301−4308

Biomacromolecules

Article

Figure 3. Effect of temperature and solvents on the secondary structure of AB12. (A) CD spectra of AB12 (0.3 mg/mL in Milli-Q water) at different temperatures. AB12 displays PPII helical structures and unordered structures in aqueous solution. (B) CD spectra of AB12 in different solvents at 25 °C. The structure of AB12 is predominantly PPII helices in water and β-turns in TFE.

Secondary Structure Analysis of AB12. The secondary structure of AB12 was analyzed using CD. Previous studies have demonstrated that CD deconvolution programs, which utilize libraries based on globular proteins, give inaccurate results for peptides and nonglobular proteins such as elastin peptides, abductin peptides, and resilin.34,52,53 Therefore, the secondary structure of AB12 was qualitatively analyzed by comparing the CD spectra to other peptides with known secondary structures.34 The CD spectra of aqueous solutions of AB12 (Figure 3A) show a strong negative peak at 200 nm and a tendency toward positive values at ∼218 nm, which are characteristics of PPII helices.54 An isodichroic point at ∼208 nm suggests an equilibrium existence between the PPII structure and other conformations.54,55 In addition, because the peak at 218 nm never exceeds zero, the spectra suggest the coexistence of unordered structures and PPII helices.34 A small negative band can be observed at ∼225 nm, which likely results from the aromatic residue, phenylalanine, in the sequence.33,56 The effect of temperature on secondary structure was studied. With increasing temperature, the magnitude of both peaks at 200 and 218 nm decreased (Figure 3A), which is typical for PPII helix conformations.52,57 In addition, the change in structure as a result of temperature was fully reversible and did not display any hysteresis (Supporting Information, Figure S3). The PPII conformation, which is widely present in elastomeric proteins such as elastin53,57,58 and titin,59 is believed to play an important role in determining the elasticity of these proteins.60 The effect of different solvents on secondary structure was also investigated. Specifically, we used a more hydrophobic solvent, TFE, which is known to induce more folded conformations, such as α-helices and β-turns, as well as destabilize the PPII conformation.53 The CD spectrum of AB12 in TFE (Figure 3B) shows a strong positive peak at 195 nm and a small negative band at ∼225 nm, which indicates a dominant conformation of type II β-turns together with a small contribution of unordered structures.33,61 The dominant PPII helix structure of AB12 in aqueous solution was also observed with other short synthetic abductinlike peptides, such as GGMGGG, GMGGG, FGGMGGG, and FGGMGGGNAG.33 GMGGG and GGMGGG also exhibit type II β-turns in TFE.33,34 The coexistence of PPII and unordered structures of AB12 in water is similar to what was seen with longer synthetic abductin-like peptides (AGGMGGGNAGAGGMGGGMAGAGGMG).34

In contrast, CD results showed that the longest synthetic abductin-like peptide previously reported ((FGGMGGGNAG)4) showed a mixture of unordered structures, PPII helices, and type II β-turns in HFIP at room temperature.34 With increasing temperature, the peptide underwent a structural change from PPII helices to type II βturns. Interestingly, we did not observe any similar thermal transitions between multiple conformations with AB12; however, we did not examine AB12 in HFIP. Temperature-Induced Aggregation Behavior of AB12. A solution of AB12 (10 mg/mL in Milli-Q water) was visually observed to turn from transparent to opaque when cooled from room temperature to lower temperatures (incubated on ice). DLS was used to further investigate the temperature responsiveness of AB12. An abrupt decrease in the hydrodynamic diameter (Dh) of AB12 was observed when the protein solution was heated from 2 to 5 °C (Figure 4). This

Figure 4. Temperature-induced responsiveness of AB12. AB12 (10 mg/mL in Milli-Q water) exhibited a sudden decrease in hydrodynamic diameter (Dh) at 5 °C, which indicated upper critical solution temperature behavior. Dh increased moderately from 35 °C and abruptly at 57 °C, which was characterized as a thermal aggregation temperature.

phenomenon is indicative of upper critical solution temperature (UCST) behavior. The change in Dh at low temperatures was reversible and displayed some hysteresis (Supporting Information, Figure S4A). A moderate increase in Dh was observed from 35 °C, and a sharper increase in Dh occurred starting at 57 °C (aggregation temperature). Compared to the reversible UCST behavior, the transition that occurred at the aggregation temperature was irreversible (Supporting Information, Figure S4B). 4305

dx.doi.org/10.1021/bm401162g | Biomacromolecules 2013, 14, 4301−4308

Biomacromolecules

Article

has a higher storage modulus (AB12 has a G′ of ∼100 Pa, whereas the resilin-based protein has a G′ of ∼6 Pa).64 Thus, using abductin-based sequences instead of resilin-based sequences as a fusion partner could potentially result in faster purification and easier separation between the coacervate and supernatant. It should be noted that high temperatures were used during the purification of AB12; at 80 °C, other undesired bacterial proteins precipitated, whereas AB12 remained in solution. In contrast, DLS showed that AB12 surprisingly formed irreversible aggregates at high temperatures. Although these two observations appear to be inconsistent, resilin-based proteins demonstrate similar phenomena. Resilin-based proteins are heat stable and are routinely purified using heating methods;45 however, they also form aggregates at high temperatures under certain conditions.39,63 This apparent inconsistency can be explained by the differences in protein or salt concentration between the different procedures.64,65 In particular, the concentration of AB12 was relatively low during the purification process (∼1 mg/mL) compared to the highly concentrated samples for the DLS tests (10 mg/mL). Increasing the concentration of resilin-based proteins decreases the aggregation temperature,64 and we are likely witnessing a similar phenomenon with AB12. In addition, it is likely that the protein solution that is heated during the purification process contains residual ammonium sulfate salt, which may affect the heat stability behavior of AB12. Cell Response to AB12 Surfaces. HUVECs were cultured on adsorbed AB12 surfaces (203 pmol protein/cm2) for 2 d. Cells had a viability of 98.4 ± 4% when cultured on AB12 surfaces and 97.4 ± 5% on positive control surfaces (TCP) (Figure 6A). There were no significant differences in cell viability when cells were cultured on TCP or AB12 surfaces. HUVECs were seeded on TCP (positive control), adsorbed AB12 surfaces, and BSA-coated surfaces (negative control), and the cell area was measured at 15, 60, 95, and 180 min. Initial cell areas were similar on AB12 and BSA surfaces (Figure 6B). After 3 h, there were no significant differences between the cell mean area when seeded on TCP, AB12 surfaces, or BSA surfaces. Example images of cell spreading at different time points are shown in the Supporting Information, Figure S5. Our results showed that HUVECs spread on AB12 surfaces slowly. This trend might result from the fact that there are no cell-binding motifs in AB12. However, cell interactions with

To further characterize the phase transition of AB12 at low temperatures, the shear modulus of a solution of AB12 (10 mg/ mL in Milli-Q water) was measured at 2 °C. The protein solution exhibited stable storage (G′) and loss (G″) moduli with G′ > G″ during the time sweep (data not shown). It remained viscoelastic over a frequency range of 0.1−55 rad/s (Figure 5), and the G′ value was ∼100 Pa. This modulus is

Figure 5. AB12 showed gel-like rheological properties (G′ > G″) below the phase transition temperature. Frequency sweeps from 0.1− 55 rad/s were performed, and the shear modulus of AB12 was measured at 2 °C, which was below the upper critical solution temperature. AB12 showed viscoelastic properties over the whole frequency range tested. Black symbols represent G′ (storage modulus) and gray symbols represent G″ (loss modulus).

similar to the shear modulus of the coacervate formed from an elastin-like peptide (ELP) at its lower critical solution temperature (LCST) (80 Pa).62 The temperature responsiveness of AB12 is similar to that observed for resilin-like polypeptides.63 Resilin-like proteins exhibit a dual-phase transition behavior with a UCST and a LCST. The UCST behavior of resilin-like proteins is similar to that of AB12. However, the aggregation behavior of AB12 at higher temperatures is different from the reversible behavior of resilin-like proteins at the LCST. Previous work with resilinbased proteins demonstrated the benefits of coacervation at low temperatures;49 the thermal responsiveness of AB12 could potentially be utilized as a cheap and facile purification method for abductin-based proteins. Furthermore, abductin-based sequences could be used as a fusion partner with other proteins to facilitate purification. Although the UCST of resilinbased proteins and AB12 are similar (both proteins have a UCST of ∼4−6 °C63,64), AB12 coacervates more quickly (AB12 coacervates in 1 min, whereas it takes ∼11 min for the resilin-based protein to coacervate64), and the AB12 coacervate

Figure 6. Cells cultured on AB12 were viable but did not initially spread. (A) HUVECs cultured for 2 d had similar viability on AB12 and TCP. Live cells are stained green, and dead cells are stained red. Scale bar represents 250 μm. (B) HUVECs were seeded on different surfaces, and the cell spread area was monitored over time. HUVECs did not spread on AB12 or BSA surfaces initially. After 3 h, no differences in mean cell area were observed between TCP, AB12, and BSA surfaces. ANOVA and Tukey’s honestly significant difference post hoc tests were performed at each time point. Groups in brackets are not significantly different from each other (p > 0.05). 4306

dx.doi.org/10.1021/bm401162g | Biomacromolecules 2013, 14, 4301−4308

Biomacromolecules

Article

Science Foundation (Award No. 0927100-EEC and a graduate fellowship to J.N.R), and the 3M Nontenured Faculty Award. We appreciate the generous gifts of the pET28aRW plasmid from Dr. David Tirrell (California Institute of Technology), the BL21(DE3)pLysS expression host from Dr. Chongli Yuan (Purdue University), and the BL21-CodonPlus(DE3)-RIPL and the BL21(DE3) expression hosts from Dr. Jo Davisson (Purdue University).We thank Dr. Dorota Inerowicz (Purdue Proteomics Facility) for conducting the MALDI-TOF mass spectroscopy and amino acid analyses. We thank Dr. Elizabeth Topp (Purdue University) for access to the CD spectroscopic instrument, Dr. Jue Chen (Purdue University) for access to the DLS instrument, and Dr. Sherry Voytik-Harbin (Purdue University) for access to the rheometer.

materials, especially cell spreading, are important for tissue engineering.66 Initial cell interactions can regulate subsequent cell behavior such as proliferation and migration. To promote specific cell adhesion and spreading, bioactive cues such as adhesion motifs can be incorporated in biomaterials. For example, two cell-binding domains derived from fibronectin (CS5 and RGD) were incorporated into recombinant elastinlike polypeptides (ELPs).67,68 When HUVECs were cultured on ELPs containing the authentic cell-binding domains, cells exhibited improved spreading and attachment compared to sequence-scrambled negative control proteins. Thus, in future studies to target specific cell receptors and enhance initial cell spreading, a cell-binding domain could easily be added to AB12 through the modular cloning scheme. Given that AB12 surfaces are cytocompatible, this modification could expand the versatility of recombinant AB12 in tissue engineering applications.

■

■

CONCLUSIONS To our knowledge, ours is the first study to successfully clone, express, and purify proteins based on consensus abductin sequences derived from A. irradians. In aqueous solution, the abductin-based protein adopted predominantly a PPII conformation, which is an important structural feature found in many elastomeric proteins. The abductin-based protein was also responsive to changes in temperature. It possessed reversible UCST behavior and formed a gel-like structure. At high temperatures, it displayed irreversible aggregation behavior. The thermal responsiveness is a useful property for engineering drug delivery systems because the encapsulation and release of drugs can easily be controlled via temperature change. The abductin-based protein was cytocompatible, and cells spread slowly when first seeded on the abductin-based protein. Future tissue engineering studies will focus on controlling cell interactions with the abductin-based protein through inclusion of a bioactive domain, such as a cell-adhesion motif. In addition, future studies will elucidate the relationship between protein sequence and transition temperatures for application in drug delivery.

■

ASSOCIATED CONTENT

S Supporting Information *

Supporting Information includes the Western blot of AB12 samples during protein expression (Figure S1), the SDS-PAGE images of samples during purification (Figure S2), the secondary structure of AB12 at 37 °C before and after heating and cooling (Figure S3), the reversible UCST behavior and the irreversible thermal aggregation behavior of AB12 (Figure S4), and sample images of HUVECs spreading on AB12 surfaces and controls (Figure S5). This material is available free of charge via the Internet at http://pubs.acs.org.

■

REFERENCES

(1) Maskarinec, S. A.; Tirrell, D. A. Curr. Opin. Biotechnol. 2005, 16, 422−426. (2) Li, L.; Teller, S.; Clifton, R. J.; Jia, X.; Kiick, K. L. Biomacromolecules 2011, 12, 2302−2310. (3) Chilkoti, A.; Dreher, M. R.; Meyer, D. E. Adv. Drug Delivery Rev. 2002, 54, 1093−1111. (4) Foster, J. A.; Bruenger, E.; Gray, W. R.; Sandberg, L. B. J. Biol. Chem. 1973, 248, 2876−2879. (5) Urry, D. J. Protein Chem. 1984, 3, 403−436. (6) Urry, D. J. Protein Chem. 1988, 7, 1−34. (7) Andersen, S. O. Biochim. Biophys. Acta 1963, 69, 249−262. (8) Ardell, D. H.; Andersen, S. O. Insect Biochem. Mol. Biol. 2001, 31, 965−970. (9) Weis-Fogh, T. J. Exp. Biol. 1960, 37, 889−907. (10) Hayashi, C. Y.; Lewis, R. V. J. Mol. Biol. 1998, 275, 773−784. (11) Hayashi, C. Y.; Lewis, R. V. Science 2000, 287, 1477−1479. (12) Hayashi, C. Y.; Shipley, N. H.; Lewis, R. V. Int. J. Biol. Macromol. 1999, 24, 271−275. (13) Greaser, M. Proteins 2001, 43, 145−149. (14) Labeit, S.; Kolmerer, B. Science 1995, 270, 293−296. (15) Linke, W. A.; Ivemeyer, M.; Mundel, P.; Stockmeier, M. R.; Kolmerer, B. Proc. Natl. Acad. Sci. U. S. A. 1998, 95, 8052−8057. (16) Cao, Q.; Wang, Y.; Bayley, H. Curr. Biol. 1997, 7, 677−678. (17) Li, L.; Charati, M. B.; Kiick, K. L. Polym. Chem. 2010, 1, 1160− 1170. (18) Annabi, N.; Mithieux, S. M.; Zorlutuna, P.; Camci-Unal, G.; Weiss, A. S.; Khademhosseini, A. Biomaterials 2013, 34, 5496−5505. (19) Fang, J.; Li, H. Langmuir 2012, 28, 8260−8265. (20) Hopkins, A. M.; De Laporte, L.; Tortelli, F.; Spedden, E.; Staii, C.; Atherton, T. J.; Hubbell, J. A.; Kaplan, D. L. Adv. Funct. Mater. 2013, 23, 5140−5149. (21) Lampe, K. J.; Antaris, A. L.; Heilshorn, S. C. Acta Biomater. 2013, 9, 5590−5599. (22) Li, L.; Tong, Z.; Jia, X.; Kiick, K. L. Soft Matter 2013, 9, 665− 673. (23) Nowatzki, P. J.; Tirrell, D. A. Biomaterials 2004, 25, 1261−1267. (24) Xu, D.; Asai, D.; Chilkoti, A.; Craig, S. L. Biomacromolecules 2012, 13, 2315−2321. (25) Kelly, R. E.; Rice, R. V. Science 1967, 155, 208−210. (26) Alexander, R. J. Exp. Biol. 1966, 44, 119−130. (27) Cheng, J.; Davison, I.; Demont, M. J. Exp. Biol. 1996, 199, 1931−1946. (28) Cheng, J. Y.; DeMont, M. E. Can. J. Zool. 1996, 74, 1734−1748. (29) Vogel, S. Nature 1997, 385, 21−22. (30) Fung, Y. C. In Biomechanics: Mechanical Properties of Living Tissues, 2nd ed.; Springer-Verlag: New York, 1993; p 321−391. (31) Gosline, J. M. In Mechanical Properties of Biological Materials; Vincent, J. F. V., Currey, J. D., Eds.; Society for Experimental Biology: London, 1980; p 331−357. (32) Weis-Fogh, T. J. Mol. Biol. 1961, 3, 648−667.

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]; Phone: 765-494-1935; Fax: 765494-0805. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Purdue School of Chemical Engineering and the College of Engineering, the National 4307

dx.doi.org/10.1021/bm401162g | Biomacromolecules 2013, 14, 4301−4308

Biomacromolecules

Article

(66) Bačaḱ ová, L.; Filová, E.; Rypácě k, F.; Švorčík, V.; V., S. Physiol. Res. 2004, 53, S35−S45. (67) Heilshorn, S. C.; DiZio, K. A.; Welsh, E. R.; Tirrell, D. A. Biomaterials 2003, 24, 4245−4252. (68) Liu, J. C.; Heilshorn, S. C.; Tirrell, D. A. Biomacromolecules 2004, 5, 497−504.

(33) Bochicchio, B.; Pepe, A.; Tamburro, A. M. Chirality 2005, 17, 364−372. (34) Bochicchio, B.; Jimenez-Oronoz, F.; Pepe, A.; Blanco, M.; Sandberg, L. B.; Tamburro, A. M. Macromol. Biosci. 2005, 5, 502−511. (35) Renner, J. N.; Kim, Y.; Cherry, K. M.; Liu, J. C. Protein Expression Purif. 2012, 82, 90−96. (36) Current Protocols in Molecular Biology; Ausubel, F. M., Brent, R., Kingston, R. E., Moore, D. D., Seidman, J. G., Smith, J. A., Struhl, K., Eds.; John Wiley & Sons: New York, 2003. (37) Current Protocols in Cell Biology; Bonifacino, J. S., Dasso, M., Harford, J. B., Lippincott-Schwartz, J., Yamada, K. M., Eds.; John Wiley & Sons: New York, 2002. (38) Abramoff, M. D.; Magalhaes, P. J.; Ram, S. J. Biophotonics Int. 2004, 11, 36−42. (39) Renner, J. N.; Cherry, K. M.; Su, R. S.-C.; Liu, J. C. Biomacromolecules 2012, 13, 3678−3685. (40) Chin, J. W.; Santoro, S. W.; Martin, A. B.; King, D. S.; Wang, L.; Schultz, P. G. J. Am. Chem. Soc. 2002, 124, 9026−9027. (41) Kirshenbaum, K.; Carrico, I. S.; Tirrell, D. A. ChemBioChem 2002, 3, 235−237. (42) Carrico, I. S.; Maskarinec, S. A.; Heilshorn, S. C.; Mock, M. L.; Liu, J. C.; Nowatzki, P. J.; Franck, C.; Ravichandran, G.; Tirrell, D. A. J. Am. Chem. Soc. 2007, 129, 4874−4875. (43) Nowatzki, P. J.; Franck, C.; Maskarinec, S. A.; Ravichandran, G.; Tirrell, D. A. Macromolecules 2008, 41, 1839−1845. (44) Mi, L. Biomacromolecules 2006, 7, 2099−2107. (45) Lyons, R. E.; Lesieur, E.; Kim, M.; Wong, D. C. C.; Huson, M. G.; Nairn, K. M.; Brownlee, A. G.; Pearson, R. D.; Elvin, C. M. Protein Eng., Des. Sel. 2007, 20, 25−32. (46) Scheller, J.; Gührs, K.-H.; Grosse, F.; Conrad, U. Nat. Biotechnol. 2001, 19, 573−577. (47) Qin, G.; Lapidot, S.; Numata, K.; Hu, X.; Meirovitch, S.; Dekel, M.; Podoler, I.; Shoseyov, O.; Kaplan, D. L. Biomacromolecules 2009, 10, 3227−3234. (48) Trabbic-Carlson, K.; Liu, L.; Kim, B.; Chilkoti, A. Protein Sci. 2004, 13, 3274−3284. (49) Lyons, R. E.; Elvin, C. M.; Taylor, K.; Lekieffre, N.; Ramshaw, J. A. M. Biotechnol. Bioeng. 2012, 109, 2947−2954. (50) Winkler, S.; Kaplan, D. L. Rev. Mol. Biotechnol. 2000, 74, 85−93. (51) Xia, X.-X.; Qian, Z.-G.; Ki, C. S.; Park, Y. H.; Kaplan, D. L.; Lee, S. Y. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 14059−14063. (52) Lyons, R. E.; Nairn, K. M.; Huson, M. G.; Kim, M.; Dumsday, G.; Elvin, C. M. Biomacromolecules 2009, 10, 3009−3014. (53) Tamburro, A. M.; Bochicchio, B.; Pepe, A. Biochemistry 2003, 42, 13347−13362. (54) Drake, A. F.; Siligardi, G.; Gibbons, W. A. Biophys. Chem. 1988, 31, 143−146. (55) Bienkiewicz, E. A.; Moon Woody, A. Y.; Woody, R. W. J. Mol. Biol. 2000, 297, 119−133. (56) Woody, R. W. In Methods in Protein Structure and Stability Analysis; Uversky, V., Permyakov, E., Eds.; Nova Science: Hauppauge, NY, 2007; p 291−344. (57) Bochicchio, B.; Tamburro, A. M. Chirality 2002, 14, 782−792. (58) Tamburro, A. M.; Bochicchio, B.; Pepe, A. Pathol. Biol. 2005, 53, 383−389. (59) Ma, K.; Kan, L.-S.; Wang, K. Biochemistry 2001, 40, 3427−3438. (60) Bochicchio, B.; Pepe, A.; Tamburro, A. M. Chirality 2008, 20, 985−994. (61) Perczel, A.; Hollosi, M.; Sandor, P.; Fasman, G. D. Int. J. Pept. Protein Res. 1993, 41, 223−236. (62) Betre, H.; Setton, L. A.; Meyer, D. E.; Chilkoti, A. Biomacromolecules 2002, 3, 910−916. (63) Dutta, N. K.; Truong, M. Y.; Mayavan, S.; Choudhury, N. R.; Elvin, C. M.; Kim, M.; Knott, R.; Nairn, K. M.; Hill, A. J. Angew. Chem., Int. Ed. 2011, 50, 4428−4431. (64) Renner, J. N. Modular Protein Matrices for Cartilage Repair. Ph.D. Dissertation, Purdue University, West Lafayette, IN, 2012. (65) Kumar, S.; Tsai, C.-J.; Nussinov, R. Protein Eng. 2000, 13, 179− 191. 4308

dx.doi.org/10.1021/bm401162g | Biomacromolecules 2013, 14, 4301−4308