Communication pubs.acs.org/Biomac
Hydrogel Biopolymer Created from the Self-Assembly of a Designed Protein Containing a Four-Helix Bundle Forming Motif Andrew F. Mehl,*,† Stefan P. Feer,‡ and John S. Cusimano‡ †
Department of Chemistry and ‡The Program in Biochemistry, Knox College, Galesburg, Illinois 61401, United States S Supporting Information *
ABSTRACT: A protein hydrogel system based on the assembly of a four-helix bundle motif was proposed and synthesized by genetic engineering methods. This new polypeptide, named GBH1, consists of identical amphipathic helices of 22 residues in length oriented in opposite fashion to one another at each end of a polypeptide with a total length of 227 amino acids. The middle portion of the polypeptide (residues 79−147) is an unstructured random coil. The region between the amphipathic and unstructured segment is an α-helical stretch (23− 78 and 148−204) not possessing a sequence compatible with a coiledcoil conformation, but rather possesses regions that have overwinding of the helix. The thermal unfolding of GBH1 shows more than one inflection point (Tm1 = 30.5 °C, Tm2 = 64.6 °C), indicative of a partially unfolded intermediate and, thus, multiple interactions in the folded state. A qualitative assessment of hydrogel formation with varying pH showed that acidic conditions did not support the gel state, indirectly indicating that the proposed four-helix bundle is the major cross-linking structure and not a leucine zipper motif. Scanning electron microscopy reveals a network of interacting protein molecules forming a spongelike matrix with numerous pores that would be occupied with water molecules.
■
(see Figure 1), these α-helices (two from each monomer) interact to form a four-helix bundle.13 Residues 33−88 also form an α-helix in the dimeric structure, but do not, however, form a coiled-coil structure but instead are in the same plane, creating the long “tail” portion of GrpE. This is due to the fact that the heptad repeat of hydrophobic residues causes overwinding in some sections due to “stutters” in the sequence.13,14 For the deletion mutant GrpE1−112, it was postulated that the tetramer was forming via a four-helix bundle with one dimeric “tail” being formed with two of the GrpE1− 112 monomers and then another dimer “tail” would align in the opposite direction of the first and interact to form a four-helix bundle at the COOH-terminal ends of each dimer.12 Here we report the construction of a designed protein that uses the amino acid sequence in the four-helix bundle forming region and the “tail” portion of the GrpE1−112 protein at both the NH2-terminal and COOH-terminal ends. The new protein has the 1−112 sequence inverted at the NH2-terminal, resulting in a new polypeptide sequence. Having an amphipathic α-helix at each end would allow for the potential formation of a four helix bundle at each end of the protein and, thus, could lead to the creation of a biopolymer. Purification and characterization of this designed protein, named GBH1 (GrpE1−112 based hydrogel), showed that a biopolymer is in fact forming and
INTRODUCTION Interest in hydrogel materials has increased dramatically over the past few years,1 in particular, those biopolymer hydrogels that can self-assemble from genetically engineered peptides and polypeptides.2,3 Protein-based hydrogels are used in biotechnology as support vehicles for drug delivery and in tissue engineering because of their cell adhesion properties. Furthermore, the self-assembly process is of value to investigate because many naturally occurring biological phenomenon rely on self-assembly of proteins. The self-assembly of protein-based hydrogels involves a variety of interactions depending on the protein or peptide sequence; some make use of electrostatic residues,4 β-sheets,2,5,6 β-hairpins,7,8 and α-helices9,10 to build larger structures. Considering those that use the α-helix, the most common arrangement of the building scaffold is the coiled-coil motif.11,10 The four-helix bundle motif is a very common protein structural element and is often the site of interaction for oligomeric proteins via the creation of a hydrophobic core. Prior research probing the structural requirements for dimerization via a four-helix bundle of the GrpE protein from E. coli led to the creation of a deletion mutant (GrpE1−112, see Figure 1) that was found to form a tetrameric species.12 This mutant contains the first 112 of 197 residues at the NH2terminal end of the full length protein. Residues 88−112 contain an amphipathic α-helix that is followed by a loop and then another amphipathic α-helix sequence (residues 116− 138) in the full length protein. In the dimeric GrpE structure © 2012 American Chemical Society
Received: February 16, 2012 Revised: April 17, 2012 Published: April 18, 2012 1244
dx.doi.org/10.1021/bm300250r | Biomacromolecules 2012, 13, 1244−1249
Biomacromolecules
Communication
Figure 1. (a) Ribbon diagram of the GrpE dimer (amino acids 34−197). The structure is based on the X-ray crystallographic data collected for a deletion mutant of GrpE that is missing the first 33 amino acids at the NH2-terminal end and is in a complex with the ATPase domain of DnaK.13 The three main regions depicted in the figure are the NH2-terminal α-helical “tail” portion composed of residues 34−87 from each monomer, the four-helix bundle composed of residues 88−137 from each monomer and the COOH-terminal β-sheet domain. The figure was produced using MOLMOL.18. (b) Schematic representation for the GrpE1−112 deletion mutant protein based on the known sequence and structure and a schematic representation for the proposed GBH1 polypeptide. The three main secondary structural regions of each are given as indicated. The full sequence for GBH1 is available online as Supporting Information.
Table 1. Primer DNA Sequences Employed for the Creation GBH1 Protein primer name
primer sequence
F112−1 RAva1 F1Ava1 R112
5′-CCACCGTCGACAAGAAGGAGATATATTAATGGCTAAAGATGCTGT-3′ 5′-CCACGGATGAAATTCCTAGGCTCGAGCATACTACTTTTTTCCTG-3′ 5′-CCACCTAGGAATTCCTAGGATGCTCGAGATGAGTAGTAAAGAACA-3′ 5′-CCACGGCGCCCTGCAGGTATACCTATTAAGCTTTATCAGCCCACTTC-3′
plasmid containing this new “gene” was then used as template DNA in a PCR reaction containing a forward primer (F112−1, see Table 1) having a Sal1 restriction enzyme site, a consensus ribosome binding site, a start codon, and the beginning coding sequence (underlined). A reverse primer (RAva1, see Table 1) containing an Ava1 restriction enzyme site and the ending coding sequence (underlined). PCR amplification was performed in a reaction mixture (100 μL) containing 25 ng of the template plasmed DNA (mentioned above), 100 pmol of each primer, 50 μM of each of the four dNTPs, 10 mM Tris-HCl (pH 8.3), 50 mM KCl, 1.5 mM MgCl2, 0.001% (w/v) gelatin, and 2.5 units of Amplitaq DNA polymerase. To generate the second piece of DNA coding for the GrpE1−112 polypeptide sequence, a forward primer (F1Ava1, see Table 1) containing an Ava1 restriction enzyme site and the coding sequence (underlined) starting residue 1 was utilized. A reverse primer (R112, see Table 1) contained a PstI restriction enzyme site, two tandem translation stop codons, and the complement grpE ending coding sequence (underlined). PCR amplification was carrying
has properties of known hydrogels. We believe that this new biopolymer represents the first protein-based hydrogel material that is formed from an α-helix building block motif using simple hydrophilic and hydrophobic type interactions similar to those found in a four-helix bundle and not coiled-coil interactions between α-helices.
■
EXPERIMENTAL METHODS
Design Strategy and Molecular Cloning of GBH1. Briefly, we used a PCR-based approach to first amplify both segments of DNA containing the desired coding region for the polypeptide sequence needed; one that coded for GrpE112−1 and one that coded for GrpE1−112. These two pieces were ligated together and then inserted into the pRLM156 overexpression vector.15 More specifically, to create the brand new polypeptide sequence of GrpE112−1, the corresponding DNA sequence was synthesized by IDT (Coralville, IA); the 1245
dx.doi.org/10.1021/bm300250r | Biomacromolecules 2012, 13, 1244−1249
Biomacromolecules
Communication
temperatures (Tm) for each transition were obtained from the fitting results. Hydrogel Preparation. Hydrogels were either formed by directly concentrating the solution in a Speed-Vac concentrator until the gel consistency was achieved or by dissolving lyophilized protein in buffer or water. To probe what approximate weight percent the hydrogel would form, protein aliquots with known concentrations and volumes were lyophilized and then rehydrated with 100 μL of distilled water. Mixing with a pipet tip and low speed centrifugation was utilized to facilitate the dissolution process in a 1.5 mL centrifuge tube. The extent of hydrogel formation was determined by observing if the material would adhere in the bottom of the tube after inversion and repeated mechanical movements of the tube in a downward motion. To investigate the ability of hydrogel formation with varying pH, an appropriate buffer at the desired pH was used during the step of the purification that involved dialysis of the ammonium sulfate precipitate and then concentrated as described above. Scanning Electron Microscopy. Purified GBH1 in the hydrogel state (greater than 7% w/w) was smeared on a glass coverslip and allowed to air-dry for 72 h. Small sections of the coverslip that contained the hydrogel material were sent to the microscopy suite at the Beckman Institute for Advanced Science and Technology at the University of Illinois at Urbana−Champaign, where the samples were coated with ∼6 nm of gold−palladium using a Denton (Moorestown NJ) Desk-2 turbo sputter coater. Images were collected using a Philips/FEI (Hillsboro OR) XL30 environmental scanning electron microscope with a field-emission electron gun (ESEM-FEG) in HiVac mode at 5 kV with a spot size of 3 (2.1 nm).
out in the same manner as above using plasmid DNA that contained the full length grpE gene (pRLM159)16 as the template DNA. After gel purification (1.7% agarose) of both of the PCR-generated pieces of DNA, each was digested to completion with AvaI and gel purified, then ligated together. After gel purification (1.7% agarose) of the ligated product, this DNA fragment was digested to completion with SalI and PstI and then ligated to pRLM 156 which had been similarly digested. DNA from the ligation reaction was transformed into RLM 56916 cells, and ampicillin-resistant clones were screened for the ability to overproduce a polypeptide with the appropriate size when grown at 42 °C. Plasmid preparations (Qiagen Midi Kits, Valencia, CA, U.S.A.) were carried out on a number of clones that showed significant overexpression, and these were sequenced to verify the correct amino acid sequence (Protein and Nucleic Acid Chemical Laboratory, Washington University School of Medicine, St. Louis, MO, U.S.A.). Note that due to the cloning strategy there are two additional amino acid residues in the middle between the number 1 (methionine) positions, namely, leucine and glutamic acid. The resulting new plasmid was named pAFM41 and the strain was named AFM66e. The sequence has been deposited with GenBank (accession No. JQ927343). Expression and Purification of GBH1. E. coli cells carrying pAFM41 were grown aerobically at 30 °C in 1 L of LB broth to an optical density of 1.0 at 595 nm. Cells were then transferred to 42 °C and grown an additional 3.5 h. Cell density typically was 1.4 at 595 nm after this induction period. Cells were aliquoted into four centrifuge tubes and spun at 7000 rpm in a Beckman SLA-1500 rotor (∼5000g). To begin cell lysis, 5.0 mL of lysis buffer (50 mM HEPES/KOH, pH 7.60, 2 mM DTT, 2 mM MgCl2, 1.0 M NaCl, and 2 M guanidine HCl) was added to each centrifuge tube and then the cells were resuspended. Cells from two tubes will combined together in a smaller centrifuge tube to which lysozyme was added (1 mg/mL). Cells were frozen in liquid nitrogen and then allowed to thaw slowly; this was repeated 4 times with a long incubation period of 40 min after the third thawing. All subsequent steps were carried out at 0−4 °C. Cell debris was removed by centrifugation for 60 min at 18500 rpm in a Beckman SS34 rotor (∼30000g). The supernatant was dialyzed for 16 h against 2 L of Buffer T (25 mM Hepes/KOH, pH 7.60, 0.5 mM EDTA, 1.0 mM DTT, and 5% (v/v) glycerol) to remove the denaturant. The dialyzed solution was centrifuged for 40 min at 12500 rpm in a Beckman SS34 rotor (∼16000g) to remove the solid material formed during the dialysis. To the supernatant, ammonium sulfate was slowly added to 55% saturation (0.351 g of ammonium sulfate/ml of supernatant) and then stirred for an additional 20 min. The precipitate that formed was removed by a 40 min centrifugation at 12500 rpm in a Beckman SS34 rotor (∼16000g). The pelleted precipitate was transferred to dialysis tubing and dialyzed for 16 h against either Buffer T (without glycerol) or 25 mM Potassium Phosphate buffer at pH 7.20. If certain experiments required additional purification, size exclusion chromatography was carried out; this involved a Superdex 200 column (FPLC system) equilibrated in the above phosphate buffer. Elution of the protein was obtained using a flow rate of 0.4 mL/ min. Circular Dichroism Spectroscopy. Circular dichroism spectra were recorded using an OLIS (Online Instruments Inc., Bogart, GA U.S.A.) RSM 1000 spectrometer configured with a CD module for dual beam CD. Spectra were obtained used the following conditions: T, 20 °C; path length, 0.1 cm; protein concentrations, 0.5−0.7 mg/ mL; scanned from 260 to 190 nm (2 nm step size) with a rate dependent on the amount of signal produced at each wavelength. Temperature was controlled using a Peltier-type temperature control system (OLIS and Quantum Northwest, Spokane, WA, U.S.A.). Thermal unfolding studies were carried out using the same conditions described above. A full scan was taken at each temperature from 6 to 94 °C (every 2 °C) with an equilibration time of 2 min at each temperature. The entire data set was fitted to a three species model using the GlobalWorks (OLIS, Bogart, GA, U.S.A.) software. The software allows a singular value decomposition (SVD) factor analysis on the entire data set using all the points from each scan at each temperature (a three-dimensional data matrix). The transition
■
RESULTS AND DISCUSSION Figure 1 shows a schematic representation of the new GrpE1− 112 based hydrogel (GBH1) protein and the various regions predicted in the sequence based on the known structure of GrpE1−112. GBH1 consists of identical amphipathic helices of 22 residues in length oriented in opposite fashion to one another at each end of a polypeptide with a total length of 227 amino acids. The middle portion of the polypeptide (residues 79−147) is an unstructured random coil. The region between the amphipathic and unstructured segment is an α-helical stretch (23−78 and 148−204) not possessing a sequence compatible with a coiled-coil conformation, but rather possesses overwinding of the helix. Figure 2 shows the proposed hydrogel formation from the GBH1 protein molecules. The biopolymer structure would be fairly amorphous with the protein scaffold created by the formation of the four-helix bundle motif. Once the clone for GBH1 had been constructed, we discovered that, during the initial purification process, when overexpressed, the newly designed protein remained with the cell debris pellet after cell lysis and centrifugation, perhaps forming inclusion bodies. To alleviate this problem, cell lysis was carried out in a low concentration of denaturant (2 M guanidine HCl) and then after centrifugation the supernatant was dialyzed against buffer without any denaturant. Lane 1 of Figure 3 shows the SDS-PAGE of whole-cell lysate (after dialysis to remove the denaturant) of E. coli expressing the GBH1 protein. As with the deletion mutant GrpE1−112,12 the GBH1 protein migrates to a much higher monomeric position than one would expect based on its predicted molar mass of ∼25 kDa. Two additional purification steps of ammonium sulfate fractionation and size exclusion chromatography produced a sample of >95% purity. Additionally, the first notice of a hydrogel-like material was observed after the dialysis step to remove the ammonium sulfate; Figure 4 shows an image of this material. Studying the hydrogel formation process 1246
dx.doi.org/10.1021/bm300250r | Biomacromolecules 2012, 13, 1244−1249
Biomacromolecules
Communication
Figure 4. Digital image of the GBH1 hydrogel biopolymer shown on the tip of a small spatula.
Figure 2. Schematic representation of hydrogel formation through self-assembly via noncovalent α-helical interactions to form four-helix bundles.
Figure 3. 13% SDS-polyacrylamide gel showing the various steps in the purification of the GrpE112−1:1−112 protein. Approximately 5 μg of total protein was loaded in each lane. Lane 1, cell lysate containing 2 M guanidine HCl; Lane 2, material after the first dialysis to remove the 2 M guanidine HCl; Lane 3, material after the second dialysis to remove ammounium sulfate; Lane 4, following a gel filtration step. Molecular weight markers are (from top to bottom) phosphorylase b (94000), albumin (66000), ovalbumin (45000), carbonic anhydrase (30000), trypsin inhibitor (20100), and α-lactalbumin (14400).
Figure 5. Far-UV CD spectra of the GBH1 protein at a concentration below that needed for hydrogel formation (∼3.5 mg/mL): (○) represents the protein solution initially at 10 °C; (□) represents the protein solution at 94 °C; (●) represents the protein solution upon returning to 10 °C.
at room temperature revealed that a concentration of 7% by weight or greater was necessary to achieve the gel consistency. Additionally, the material maintains the gel-like consistency when kept at 4 °C (determined over a period of 1 year). Initial characterization utilized CD spectroscopy to access αhelical content and, thus, proper folding of the polypeptide. Figure 5 shows the results of the CD analysis. The characteristic minimum in ellipiticity is observed at 222 and 208 nm, indicating significant α-helical content. To check for the capability of the designed protein to refold after becoming denatured, a temperature-induced unfolding experiment was carried out (see Figure 6) and then the temperature was returned to allow for refolding. The results indicate that the majority of protein molecules are refolding and not aggregating, and the process of thermal denaturation was reversible. The thermal unfolding shows two points of inflection, indicative of an unfolded intermediate (see Figure 6). Additionally, when the
thermal-induced unfolding data was analyzed using a three species model, a first transition Tm of 30.5 °C was determined, and for the second transition, a Tm of 64.6 °C was determined. To study the sensitivity of pH to the ability of GBH1 to form the hydrogel material, we qualitatively assessed gel formation after exhaustive dialysis in the appropriate buffer solution (see Table 2). Under acidic conditions (pH 5.0 and 6.0) no hydrogel formation was observed, while at neutral and basic conditions the hydrogel formed (Figure 4). This study supports the notion that the major cross-linking structure is not coiledcoil α-helices, but rather via four-helix bundle formation because acidic conditions would help stabilize coiled-coil aggregates with protonation of the Glu side chains.9 Finally, to verify that a polymer is forming and is responsible for the hydrogel properties, SEM microscopy was utilized. Figure 7 shows a network of interacting protein molecules forming a spongelike matrix with numerous pores that would 1247
dx.doi.org/10.1021/bm300250r | Biomacromolecules 2012, 13, 1244−1249
Biomacromolecules
Communication
Incorporating an amphipathic α-helical region of 22 residues that facilitates the formation of a four-helix bundle at each end of a polypeptide along with a highly charged region of residues in the middle resulted in the self-assembly of a protein-based hydrogel material. Future work will investigate what changes in the structural features will produce a more robust hydrogel; for example, could replacement of certain aliphatic nonpolar residues with aromatic ones within the α-helical segment of the four-helix bundle forming region create a better hydrophobic core due to an aromatic cluster? Additionally, could new bifunctional hydrogel materials be created where specific enzyme activities are fused with the hydrogel forming polypeptide?
■
ASSOCIATED CONTENT
S Supporting Information *
The full amino acid and corresponding nucleotide sequence for GBH1. This material is available free of charge via the Internet at http://pubs.acs.org.
Figure 6. Thermal unfolding of the GBH1 protein followed by CD spectroscopy at 222 nm. Conditions are described in the Experimental Methods.
■
Table 2. Qualitative Analysis of GBH1 Hydrogel Formation as a Function of pH buffer component (25 mM)
pH
ability to form a hydrogel
acetic acid MES HEPES HEPES TRIS glycine
5.0 6.0 7.0 8.0 9.0 10.0
none none yes yes yes minimal
AUTHOR INFORMATION
Corresponding Author
*E-mail:
[email protected].
a
Notes
The authors declare no competing financial interest.
■
ACKNOWLEDGMENTS We are grateful for the support of Knox College; specifically, the support from the Paul K. Richter and Evalyn E. Cook Richter Memorial Trusts and the Andrew Mellon Foundation.
■
a
Determined by visual inspection of whether the material would stay adhered on the tip of a spatula after inversion.
REFERENCES
(1) Van Vlierberghe, S.; Dubruel, P.; Schacht, E. Biomacromolecules 2011, 5, 1387−1408. (2) Rughani, R. V.; Salick, D. A.; Lamm, M. S.; Yucel, T.; Pochan, D. J.; Schneider, J. P. Biomacromolecules 2009, 5, 1295−1304. (3) King, W. J.; Toepke, M. W.; Murphy, W. L. Chem. Commun. 2011, 47, 526−528. (4) King, W. J.; Pytel, N. J.; Ng, K.; Murphy, W. L. Macromol. Biosci. 2010, 6, 580−584. (5) Fang, J.; Zhang, X.; Cai, Y.; Wei, Y. Biomacromolecules 2011, 5, 1578−1584. (6) Kyle, S.; Aggeli, A.; Ingham, E.; McPherson, M. J. Biomaterials 2010, 36, 9395−9405. (7) Nagarkar, R. P.; Hule, R. A.; Pochan, D. J.; Schneider, J. P. J. Am. Chem. Soc. 2008, 13, 4466−4474. (8) Rajagopal, K.; Lamm, M. S.; Haines-Butterick, L. A.; Pochan, D. J.; Schneider, J. P. Biomacromolecules 2009, 9, 2619−2625. (9) Petka, W. A.; Harden, J. L.; McGrath, K. P.; Wirtz, D.; Tirrell, D. A. Science 1998, 5375, 389−392. (10) Lu, H. D.; Wheeldon, I. R.; Banta, S. Protein Eng. Des. Sel. 2010, 7, 559−566. (11) Yang, J.; Xu, C.; Wang, C.; Kopecek, J. Biomacromolecules 2006, 4, 1187−1195. (12) Mehl, A. F.; Heskett, L. D.; Jain, S. S.; Demeler, B. Protein Sci. 2003, 6, 1205−1215. (13) Harrison, C. J.; Hayer-Hartl, M.; Di Liberto, M.; Hartl, F.; Kuriyan, J. Science 1997, 5311, 431−435. (14) Lupas, A. Trends Biochem. Sci. 1996, 10, 375−382. (15) Karzai, A. W.; McMacken, R. J. Biol. Chem. 1996, 19, 11236− 11246. (16) Mehl, A. F.; Heskett, L. D.; Neal, K. M. Biochem. Biophys. Res. Commun. 2001, 2, 562−569. (17) Xu, C.; Breedveld, V.; Kopecek, J. Biomacromolecules 2005, 3, 1739−1749.
Figure 7. SEM image of the GBH1 hydrogel.
be occupied with water molecules. The pores created are typically much smaller than those observed with other proteinbased hydrogels that utilize a coiled-coil interaction as the basis for polymer formation.11,17
■
CONCLUSION When the known properties of a previously created deletion mutant protein of the GrpE heat shock protein from E. coli were used, a new hydrogel biopolymer (GBH1) was designed. 1248
dx.doi.org/10.1021/bm300250r | Biomacromolecules 2012, 13, 1244−1249
Biomacromolecules
Communication
(18) Koradi, R.; Billeter, M.; Wuthrich, K. J. Mol. Graphics 1996, 14, 51−55.
■
NOTE ADDED AFTER ASAP PUBLICATION This Communication posted ASAP on April 24, 2012. The toc graphic and abstract graphic have been revised. The correct version posted on April 27, 2012.
1249
dx.doi.org/10.1021/bm300250r | Biomacromolecules 2012, 13, 1244−1249
