YABBI 6973
No. of Pages 7, Model 5G
5 May 2015 Archives of Biochemistry and Biophysics xxx (2015) xxx–xxx 1
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi 4 5 3
IR-spectroscopic characterization of an elongated OmpG mutant
6
Filiz Korkmaz a,⇑, Katharina van Pee b, Özkan Yildiz b
7 8 9 10 1 4 2 2 13 14 15 16 17 18 19 20 21 22 23
a b
Atilim University, Physics Unit, Biophysics Laboratory, 06836 Golbasi, Ankara, Turkey Max Planck Institute for Biophysics, Department of Structural Biology, Max-von-Laue Str. 3, D-60438 Frankfurt am Main, Germany
a r t i c l e
i n f o
Article history: Received 13 January 2015 and in revised form 10 April 2015 Available online xxxx Keywords: Outer membrane protein (OMP) IR spectroscopy Protein engineering Porin b-Barrel
a b s t r a c t OmpG is a nonselective, pH dependent outer membrane protein from Escherichia coli. It consists of 281 residues, forming a 14-stranded b-sheet structure. In this study, OmpG is extended by 38 amino acids to produce a 16-stranded b-barrel (OmpG-16S). The resulting protein is investigated by IR-spectroscopy. The secondary structure, pH-dependent opening/closing mechanism, buffer accessibility and thermal stability of OmpG-16S are compared to OmpG-WT. The results show that OmpG-16S is responsive to pH change as indicated by the Amide I band shift upon a switch from acidic to neutral pH. This spectral shift is consistent with that observed in OmpG-WT, which confirms the existence of structural differences consistent with the presence of the open or closed state. Secondary structure analysis after curve-fitting of Amide I band revealed that the additional residues do not fold into b-sheet; rather they are in the form of turns and unordered structure. In thermal stability experiments, OmpG-16S is found to be as stable as OmpG-WT. Additionally, H/D exchange experiments showed no difference in the exchange rate of OmpG-16S between the acidic and alkaline pH, suggesting that the loop L6 is no longer sufficient to block the pore entrance at acidic pH. Ó 2015 Published by Elsevier Inc.
25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
Introduction OmpG is an outer membrane protein from Escherichia coli forming a 14-stranded b-barrel. It is a nonspecific, monomeric porin for the passage of ions and molecules up to 900 Da [1–3]. OmpG has been extensively studied using many techniques like X-ray crystallography [4], single molecule force spectroscopy [5], atomic force microscopy [3] and FTIR spectroscopy [6,7]. In these studies, the details of its structure in various conditions have been revealed. There is a pile of structural and functional information available for OmpG that are referable to its excellent stability and pH-regulated channel activity. OmpG shows two different structural properties at acidic and neutral pH values [4,7]. The reason behind this difference is due to the electrostatic interaction of the histidines, His231 and His261, located at the pore lumen. At acidic pH, the imidazole ring of these histidines are protonated and repel each other, which causes the bending of loop L6 that connects b-strand S11–S12 into the channel to block it. At neutral pH, both histidines are deprotonated and hydrogen bonded, stabilizing the loop L6 in the extended conformation, leaving the channel open for passive diffusion. This structural difference is reflected in the IR-spectra by the shift of the low frequency component of b-sheet by 1–2 cm1. The ⇑ Corresponding author. E-mail address: fi[email protected] (F. Korkmaz).
role of histidines in opening/closing the channel was shown by mutating both histidines to alanines and cysteines [6]. It was also shown that the opening/closing is driven by the formation/breaking of hydrogen bridges in b-strands S11–S13 as well as by the electrostatic interactions among positive and negatively charged side chains [7]. FTIR spectroscopy further revealed that OmpG is stable up to 80 °C in detergent solubilized form and up to 100 °C in lipid bilayer. The difference in stability is associated with the aromatic belt of the protein that interacts with the interfacial region of the lipids [8]. Due to its robust structure and its ability to attain an open and closed state, OmpG is a potential candidate for biotechnological applications such as targeted drug delivery and biosensor applications [9] (and references therein). With this motivation, the first structural alteration is made to enlarge the pore diameter. The number of b-strands is aimed to increase from 14 to 16, while keeping the structural stability and the pH-dependent channel activity unaltered. Increasing the b-strand number is accomplished by introducing two more b-strands with a similar amino acid composition like the last hairpin of the wild type protein (OmpG-WT), which resulted in the OmpG mutant, so called OmpG-16S. In this study, FTIR spectroscopy is used to investigate the structural and functional properties of OmpG-16S. The results are compared with the results previously obtained for OmpG-WT [6,7]. FTIR spectroscopy provides direct information on bond orders, electrostatic interactions, H-bonding and protonation states of
http://dx.doi.org/10.1016/j.abb.2015.04.010 0003-9861/Ó 2015 Published by Elsevier Inc.
Please cite this article in press as: F. Korkmaz et al., Arch. Biochem. Biophys. (2015), http://dx.doi.org/10.1016/j.abb.2015.04.010
64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89
YABBI 6973
No. of Pages 7, Model 5G
5 May 2015 2
F. Korkmaz et al. / Archives of Biochemistry and Biophysics xxx (2015) xxx–xxx
resolution of 2 cm1, apodized using Happ-Genzel apodization function and Fourier-transformed with a zero-filling factor of 1. A sample shuttle device is used in the sample chamber that enables scanning first the background and then the sample without opening the chamber lid. An empty pair of CaF2 cells is used for background measurement, which is subtracted automatically by the software OMNIC 8.2 (Thermo Fisher Scientific Inc.). For thermal profiling experiments, both OmpG-WT and OmpG-16S are incubated in 0.1 M K-phosphate buffer, pD 5.5/7.5, 0.5% OG for 24 h at +4 °C before the experiment. A heated cell that is in contact with the sample cells controls the temperature. The temperature is increased by 2 °C at each step and the sample is incubated at each temperature set point for 5 min. Spectra are recorded with the instrument parameters described above. In order to follow changes in protein structure induced by increasing temperature, the position of low frequency component of b-sheet signal is followed as the temperature is increased from 20 to 110 °C. The signal is followed from the second derivative of spectra.
148
109
residues in protein research, and thus, is widely used to study the structure and dynamics of polypeptides and their interaction with ligands [10–12]. FTIR studies particularly focus on the Amide I band (1700–1600 cm1) that is mainly originated from C@O stretching vibration of the peptide bond with contributions from CAN stretching, CCN deformation and NAH in-plane bending modes. Amide I band position, width and shape depend on the folding motif of the polypeptide backbone. Therefore, it is sensitive to the secondary structure composition of peptides. Amide II band (1600–1500 cm1) is originated from CAN stretching coupled with NAH bending mode, which is generally used to monitor 1 H/2H exchange rate and kinetics of the polypeptide backbone. In FTIR spectroscopy studies, proteins are studied in their close-to-natural environment and are not labeled in order to follow particular molecular groups. Beside the structural properties of OmpG-16S, we have also studied the stepwise response of the mutant with respect to pH and temperature. 1H/2H exchange experiments are also performed to observe solvent accessibility differences at two pH values and compared to the results of OmpG-WT.
Processing of spectra
166
110
Materials and methods
167
111
Expression and purification of OmpG-16S
112
132
Expression and purification of OmpG-16S protein is done as already described for the wild type OmpG [4]. The sequence of OmpG-WT was extended by 38 residues corresponding to the b-strand S13 and S14 of the original protein. The synthetic gene coding for the extended protein (OmpG-16S) is cloned in the pET26b vector and expressed in E. coli C43-(DE3) cells, which are grown in TB medium. The protein is purified out of inclusion bodies, which are collected by a low-speed centrifugation after disrupting the cells with cell disruptor (Constant Systems). The inclusion bodies are washed in buffer (25 mM Tris–HCl pH 8.0, 1 M urea and 1% Triton-X 100) and dissolved in the same buffer with 8 M urea. The solubilized protein is loaded onto an anion exchange column. Unfolded OmpG-16S is eluted using a NaCl step gradient and refolded by dilution to a final urea concentration of 3 M in 1% (wt/vol) n-octy-b-D-glucopyranoside (OG). Refolding takes place overnight following a second anion exchange chromatography to remove unfolded or partially folded OmpG-16S. The protein purity and refolding is monitored by SDS–PAGE. Afterwards the buffer is exchanged to 25 mM Tris–HCl pH 8.0 containing 0.5% OG and the pure protein (>95%) is concentrated to 10 mg/ml by ultrafiltration (Centricon).
133
FTIR measurements
134
For spectroscopic analysis of the protein, the buffer (25 mM Tris pH 8.0, 0.5% OG) is exchanged against 0.1 M K-phosphate buffer, pH/pD 5.5 or pH/pD 7.5, 0.5% OG by ultrafiltration (Millipore, USA) using 30 kDa cut-off filters. The pD, i.e. the equivalent of pH in D2O (2H2O), is determined by adding 0.4 pH units to the pH meter reading [13]. Two microliter of protein at a concentration of 25 mg/ml is pipetted between two CaF2 cells. Spectra are recorded with a Nicolet 6700 (Thermo Fisher Scientific Inc., USA) FTIR spectrometer equipped with both Deuterated Tri-Glycine Sulfate (DTGS)1 and a liquid N2-cooled Mercury Cadmium Telluride (MCT) detector. MCT detector is used for fast scanning in thermal profiling experiments, otherwise DTGS detector is used throughout the study. The sample compartment is continuously purged with dry air. 128 interferograms are recorded at a
The sample spectra are first buffer corrected when necessary using the software OMNIC 8.2. Spectra are then used for Amide I peak position and second derivative comparisons. For secondary structure analysis, the baseline is corrected for the region 1710– 1580 cm1 by subtracting a straight line, connecting the two points. Side chain contributions are subtracted prior to curve fitting using the parameters reported previously [14]. For the protein in D2O buffer, residues having absorbance contribution in the Amide I region are first grouped as hydrophilic and hydrophobic using the X-ray structure of OmpG-WT. Their spectral contributions are subtracted from the protein spectrum separately since the ones in hydrophilic region are in deuterated form but those in hydrophobic region are in hydrated form. Protonation state of side chains depending on the buffer pH are also taken into account before the subtraction. Side chain subtraction coefficient is scaled with the tyrosine ring CAC mode at 1515 cm1 as the reference. Prior to curve fitting, the number and position of underlying sub-bands are determined from the second derivative of the spectrum. The built-in iterative curve-fitting macro of the software OMNIC is used. The number and position of sub-bands are entered manually to the software and kept constant during the iteration; however, the software optimizes the width and height of bands in order to obtain the best match between the original spectrum and the sum of fitted spectra. Curve-fitting success is determined by comparing the second derivative profile of the raw spectrum and that of the sum of fitted bands while trying to achieve minimal root mean square (RMS) error at the same time. Spectra taken at 20 °C in temperature profiling experiments are used for calculating the H/D exchange rates since they are first equilibrated in D2O buffer for 24 h. Spectra are buffer corrected and side chain subtracted before the analyses. Areas of Amide I and II bands are calculated for each buffer condition (H2O buffer pH 5.5/7.5 and D2O buffer pD 5.5/7.5). Although the change in Amide II band area is the indicator of H/D exchange rate, the ratio of areas (Amide II/Amide I) is used to counteract the absorbance differences among different measurements. This ratio is set to 0% for the samples in H2O buffer. Complete disappearance of Amide II band corresponds to 100% exchange. Results and discussion
205
pH response of OmpG-16S
206
OmpG-16S has 38 additional residues at the carboxyl end compared with OmpG-WT. The purpose of this extension of the
207
90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108
113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131
135 136 137 138 139 140 141 142 143 144 145 146 147
1 Abbreviations used: OMP, outer membrane protein; DTGS, Deuterated Tri-Glycine Sulfate; MCT, Mercury Cadmium Telluride.
Please cite this article in press as: F. Korkmaz et al., Arch. Biochem. Biophys. (2015), http://dx.doi.org/10.1016/j.abb.2015.04.010
149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165
168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204
208
YABBI 6973
No. of Pages 7, Model 5G
5 May 2015 F. Korkmaz et al. / Archives of Biochemistry and Biophysics xxx (2015) xxx–xxx 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233
polypeptide chain is to add two more b-strands to the b-barrel to increase the diameter of the channel. We have investigated the structural properties of this supposedly enlarged porin by IR-spectroscopy and compared the results with previously reported results of OmpG-WT. The b-sheet content of OmpG-WT at acidic pH is 68% and at neutral pH is 71% as shown in the X-ray structure [4] and as also confirmed by IR-spectroscopy [6,7]. A detailed analysis of the Amide I band of OmpG-WT showed that the b-sheet signal is shifted to lower wavenumbers at neutral pH compared to the same signal at acidic pH. This shift in the b-sheet signal position has been previously correlated with the hydrogen-bond strength, so that the protein is more rigid and stable at neutral pH (open pore) than it is at acidic pH (closed pore) [7]. In order to investigate the structure of OmpG-16S, the protein is equilibrated both in H2O and in D2O buffer. Many side chain groups and unordered structure are shifted when the buffer is changed, which enables the user to identify the signals and make assignments accordingly. When the spectra of OmpG-16S mutant at two pH values in H2O and in D2O buffer are analyzed, a similar shift of b-sheet signal is seen from both absorbance spectra and their respective second derivative profiles (Fig. 1). Amide I peak position for OmpG-16S is located approximately at 1630 cm1 in both buffer conditions, which indicates the overwhelming b-sheet in structure similar to OmpG-WT [11,15,16]. However, this position depends on the buffer pH. While the Amide I band is located at
3
1631 cm1 at pH 5.5, it shifts down to 1628 cm1 at pH 7.5 in H2O buffer (Fig. 1A). A similar spectral shift of 1.2 cm1 is also seen in D2O buffer (Fig. 1B). The second derivative profiles of these spectra show that the observed shift is due to the component located at 1628 cm1 (Fig. 1C and D). The shift, as seen from the second derivative spectra, is by 2.1 cm1 in H2O buffer and 1.9 cm1 in D2O buffer. There are also minor differences in the 1640–1680 cm1 region between the two pH/pD conditions of the protein. The Amide I peak position also differs between the spectra taken in H2O and D2O buffers of the same pH values (Fig. 1A and B). While the peak is located at 1631 cm1 in H2O buffer pH 5.5, it is located at 1628 cm1 in D2O buffer. This difference is mainly due to the unordered structure signal that is located at 1660 cm1 in H2O buffer and at 1640 cm1 in D2O buffer [16]. Another factor contributing to the difference is the side chain absorbance of Arg, which is located at 1680 cm1 in H2O buffer and at 1580 cm1 in D2O buffer [17]. Side chain group and unordered structure absorbance shifts observed in two buffer media explain the difference in the corresponding Amide I band profiles. Inspection of the Amide I band and its second derivative profile definitely points to the fact that pH change affects the secondary structure of OmpG-16S in a way similar to OmpG-WT. We also wanted to verify that the spectral changes in response to pH are not special to the two pH values used, i.e. pH 5.5 and 7.5; rather
Fig. 1. IR absorbance spectra of OmpG-16S in 0.1 M K-phosphate +0.5% OG buffer at acidic (–) or neutral ( ) pH/pD in H2O (A) and in D2O buffer (B). Their corresponding second derivative profiles are shown below (C and D). Absorbance spectra are shown for the entire fingerprint region (1750–1300 cm1); however, the second derivative 1 profiles are shown only for the Amide I and II regions (1700–1500 cm ) for better visualization of the difference between the two spectra, particularly the shift of low frequency b-sheet component.
Please cite this article in press as: F. Korkmaz et al., Arch. Biochem. Biophys. (2015), http://dx.doi.org/10.1016/j.abb.2015.04.010
234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258
YABBI 6973
No. of Pages 7, Model 5G
5 May 2015 4
F. Korkmaz et al. / Archives of Biochemistry and Biophysics xxx (2015) xxx–xxx Table 1 Curve-fitting results of Amide I region, listing the positions of sub-bands and their respective band area for pD 5.5 and pD 7.5 for OmpG-WT and OmpG-16S. Suggested band assignments are based on X-ray data of OmpG-WT. Band position (1/ cm)
1625 1628 1637 1640 1647 1656 1665 1673 1683 1686 1693 1696 Fig. 2. The low frequency mode of b-sheet signal position as a function of buffer pD. The peak position is followed from the second derivative of IR absorbance spectra of OmpG-16S incubated in 0.1 M K-phosphate +0.5% OG buffer in D2O, incubated for 24 h at indicated pD values.
279
it is the increase in pH that the protein structure responds. Thus, a pH-titration experiment is performed and the low frequency b-sheet signal position is followed at each pH value (Fig. 2). Protein samples are equilibrated in D2O buffer for 24 h with pD values ranging from 5.0 to 8.0. The b-sheet signal position is observed to shift gradually down to lower wavenumbers starting from pD 5.0 to pD 7.0; however, further increasing pD to 7.5 caused a dramatic downshift of the b-sheet peak position. Between pD 7.5 and 8.0, there is no significant difference in the position. Although data clearly show that OmpG-16S exhibits a gradual structural response to increasing pD, the profile of the spectral shift is somewhat different than the previously reported profile of OmpG-WT [6]. The b-sheet peak position of OmpG-WT has been observed to stay constant below pD 6.0 and above 7.5, which pointed out to well-defined conformations as open (>pD 7.5) and closed state (
No. of Pages 7, Model 5G
5 May 2015 Archives of Biochemistry and Biophysics xxx (2015) xxx–xxx 1
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi 4 5 3
IR-spectroscopic characterization of an elongated OmpG mutant
6
Filiz Korkmaz a,⇑, Katharina van Pee b, Özkan Yildiz b
7 8 9 10 1 4 2 2 13 14 15 16 17 18 19 20 21 22 23
a b
Atilim University, Physics Unit, Biophysics Laboratory, 06836 Golbasi, Ankara, Turkey Max Planck Institute for Biophysics, Department of Structural Biology, Max-von-Laue Str. 3, D-60438 Frankfurt am Main, Germany
a r t i c l e
i n f o
Article history: Received 13 January 2015 and in revised form 10 April 2015 Available online xxxx Keywords: Outer membrane protein (OMP) IR spectroscopy Protein engineering Porin b-Barrel
a b s t r a c t OmpG is a nonselective, pH dependent outer membrane protein from Escherichia coli. It consists of 281 residues, forming a 14-stranded b-sheet structure. In this study, OmpG is extended by 38 amino acids to produce a 16-stranded b-barrel (OmpG-16S). The resulting protein is investigated by IR-spectroscopy. The secondary structure, pH-dependent opening/closing mechanism, buffer accessibility and thermal stability of OmpG-16S are compared to OmpG-WT. The results show that OmpG-16S is responsive to pH change as indicated by the Amide I band shift upon a switch from acidic to neutral pH. This spectral shift is consistent with that observed in OmpG-WT, which confirms the existence of structural differences consistent with the presence of the open or closed state. Secondary structure analysis after curve-fitting of Amide I band revealed that the additional residues do not fold into b-sheet; rather they are in the form of turns and unordered structure. In thermal stability experiments, OmpG-16S is found to be as stable as OmpG-WT. Additionally, H/D exchange experiments showed no difference in the exchange rate of OmpG-16S between the acidic and alkaline pH, suggesting that the loop L6 is no longer sufficient to block the pore entrance at acidic pH. Ó 2015 Published by Elsevier Inc.
25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40
41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63
Introduction OmpG is an outer membrane protein from Escherichia coli forming a 14-stranded b-barrel. It is a nonspecific, monomeric porin for the passage of ions and molecules up to 900 Da [1–3]. OmpG has been extensively studied using many techniques like X-ray crystallography [4], single molecule force spectroscopy [5], atomic force microscopy [3] and FTIR spectroscopy [6,7]. In these studies, the details of its structure in various conditions have been revealed. There is a pile of structural and functional information available for OmpG that are referable to its excellent stability and pH-regulated channel activity. OmpG shows two different structural properties at acidic and neutral pH values [4,7]. The reason behind this difference is due to the electrostatic interaction of the histidines, His231 and His261, located at the pore lumen. At acidic pH, the imidazole ring of these histidines are protonated and repel each other, which causes the bending of loop L6 that connects b-strand S11–S12 into the channel to block it. At neutral pH, both histidines are deprotonated and hydrogen bonded, stabilizing the loop L6 in the extended conformation, leaving the channel open for passive diffusion. This structural difference is reflected in the IR-spectra by the shift of the low frequency component of b-sheet by 1–2 cm1. The ⇑ Corresponding author. E-mail address: fi[email protected] (F. Korkmaz).
role of histidines in opening/closing the channel was shown by mutating both histidines to alanines and cysteines [6]. It was also shown that the opening/closing is driven by the formation/breaking of hydrogen bridges in b-strands S11–S13 as well as by the electrostatic interactions among positive and negatively charged side chains [7]. FTIR spectroscopy further revealed that OmpG is stable up to 80 °C in detergent solubilized form and up to 100 °C in lipid bilayer. The difference in stability is associated with the aromatic belt of the protein that interacts with the interfacial region of the lipids [8]. Due to its robust structure and its ability to attain an open and closed state, OmpG is a potential candidate for biotechnological applications such as targeted drug delivery and biosensor applications [9] (and references therein). With this motivation, the first structural alteration is made to enlarge the pore diameter. The number of b-strands is aimed to increase from 14 to 16, while keeping the structural stability and the pH-dependent channel activity unaltered. Increasing the b-strand number is accomplished by introducing two more b-strands with a similar amino acid composition like the last hairpin of the wild type protein (OmpG-WT), which resulted in the OmpG mutant, so called OmpG-16S. In this study, FTIR spectroscopy is used to investigate the structural and functional properties of OmpG-16S. The results are compared with the results previously obtained for OmpG-WT [6,7]. FTIR spectroscopy provides direct information on bond orders, electrostatic interactions, H-bonding and protonation states of
http://dx.doi.org/10.1016/j.abb.2015.04.010 0003-9861/Ó 2015 Published by Elsevier Inc.
Please cite this article in press as: F. Korkmaz et al., Arch. Biochem. Biophys. (2015), http://dx.doi.org/10.1016/j.abb.2015.04.010
64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89
YABBI 6973
No. of Pages 7, Model 5G
5 May 2015 2
F. Korkmaz et al. / Archives of Biochemistry and Biophysics xxx (2015) xxx–xxx
resolution of 2 cm1, apodized using Happ-Genzel apodization function and Fourier-transformed with a zero-filling factor of 1. A sample shuttle device is used in the sample chamber that enables scanning first the background and then the sample without opening the chamber lid. An empty pair of CaF2 cells is used for background measurement, which is subtracted automatically by the software OMNIC 8.2 (Thermo Fisher Scientific Inc.). For thermal profiling experiments, both OmpG-WT and OmpG-16S are incubated in 0.1 M K-phosphate buffer, pD 5.5/7.5, 0.5% OG for 24 h at +4 °C before the experiment. A heated cell that is in contact with the sample cells controls the temperature. The temperature is increased by 2 °C at each step and the sample is incubated at each temperature set point for 5 min. Spectra are recorded with the instrument parameters described above. In order to follow changes in protein structure induced by increasing temperature, the position of low frequency component of b-sheet signal is followed as the temperature is increased from 20 to 110 °C. The signal is followed from the second derivative of spectra.
148
109
residues in protein research, and thus, is widely used to study the structure and dynamics of polypeptides and their interaction with ligands [10–12]. FTIR studies particularly focus on the Amide I band (1700–1600 cm1) that is mainly originated from C@O stretching vibration of the peptide bond with contributions from CAN stretching, CCN deformation and NAH in-plane bending modes. Amide I band position, width and shape depend on the folding motif of the polypeptide backbone. Therefore, it is sensitive to the secondary structure composition of peptides. Amide II band (1600–1500 cm1) is originated from CAN stretching coupled with NAH bending mode, which is generally used to monitor 1 H/2H exchange rate and kinetics of the polypeptide backbone. In FTIR spectroscopy studies, proteins are studied in their close-to-natural environment and are not labeled in order to follow particular molecular groups. Beside the structural properties of OmpG-16S, we have also studied the stepwise response of the mutant with respect to pH and temperature. 1H/2H exchange experiments are also performed to observe solvent accessibility differences at two pH values and compared to the results of OmpG-WT.
Processing of spectra
166
110
Materials and methods
167
111
Expression and purification of OmpG-16S
112
132
Expression and purification of OmpG-16S protein is done as already described for the wild type OmpG [4]. The sequence of OmpG-WT was extended by 38 residues corresponding to the b-strand S13 and S14 of the original protein. The synthetic gene coding for the extended protein (OmpG-16S) is cloned in the pET26b vector and expressed in E. coli C43-(DE3) cells, which are grown in TB medium. The protein is purified out of inclusion bodies, which are collected by a low-speed centrifugation after disrupting the cells with cell disruptor (Constant Systems). The inclusion bodies are washed in buffer (25 mM Tris–HCl pH 8.0, 1 M urea and 1% Triton-X 100) and dissolved in the same buffer with 8 M urea. The solubilized protein is loaded onto an anion exchange column. Unfolded OmpG-16S is eluted using a NaCl step gradient and refolded by dilution to a final urea concentration of 3 M in 1% (wt/vol) n-octy-b-D-glucopyranoside (OG). Refolding takes place overnight following a second anion exchange chromatography to remove unfolded or partially folded OmpG-16S. The protein purity and refolding is monitored by SDS–PAGE. Afterwards the buffer is exchanged to 25 mM Tris–HCl pH 8.0 containing 0.5% OG and the pure protein (>95%) is concentrated to 10 mg/ml by ultrafiltration (Centricon).
133
FTIR measurements
134
For spectroscopic analysis of the protein, the buffer (25 mM Tris pH 8.0, 0.5% OG) is exchanged against 0.1 M K-phosphate buffer, pH/pD 5.5 or pH/pD 7.5, 0.5% OG by ultrafiltration (Millipore, USA) using 30 kDa cut-off filters. The pD, i.e. the equivalent of pH in D2O (2H2O), is determined by adding 0.4 pH units to the pH meter reading [13]. Two microliter of protein at a concentration of 25 mg/ml is pipetted between two CaF2 cells. Spectra are recorded with a Nicolet 6700 (Thermo Fisher Scientific Inc., USA) FTIR spectrometer equipped with both Deuterated Tri-Glycine Sulfate (DTGS)1 and a liquid N2-cooled Mercury Cadmium Telluride (MCT) detector. MCT detector is used for fast scanning in thermal profiling experiments, otherwise DTGS detector is used throughout the study. The sample compartment is continuously purged with dry air. 128 interferograms are recorded at a
The sample spectra are first buffer corrected when necessary using the software OMNIC 8.2. Spectra are then used for Amide I peak position and second derivative comparisons. For secondary structure analysis, the baseline is corrected for the region 1710– 1580 cm1 by subtracting a straight line, connecting the two points. Side chain contributions are subtracted prior to curve fitting using the parameters reported previously [14]. For the protein in D2O buffer, residues having absorbance contribution in the Amide I region are first grouped as hydrophilic and hydrophobic using the X-ray structure of OmpG-WT. Their spectral contributions are subtracted from the protein spectrum separately since the ones in hydrophilic region are in deuterated form but those in hydrophobic region are in hydrated form. Protonation state of side chains depending on the buffer pH are also taken into account before the subtraction. Side chain subtraction coefficient is scaled with the tyrosine ring CAC mode at 1515 cm1 as the reference. Prior to curve fitting, the number and position of underlying sub-bands are determined from the second derivative of the spectrum. The built-in iterative curve-fitting macro of the software OMNIC is used. The number and position of sub-bands are entered manually to the software and kept constant during the iteration; however, the software optimizes the width and height of bands in order to obtain the best match between the original spectrum and the sum of fitted spectra. Curve-fitting success is determined by comparing the second derivative profile of the raw spectrum and that of the sum of fitted bands while trying to achieve minimal root mean square (RMS) error at the same time. Spectra taken at 20 °C in temperature profiling experiments are used for calculating the H/D exchange rates since they are first equilibrated in D2O buffer for 24 h. Spectra are buffer corrected and side chain subtracted before the analyses. Areas of Amide I and II bands are calculated for each buffer condition (H2O buffer pH 5.5/7.5 and D2O buffer pD 5.5/7.5). Although the change in Amide II band area is the indicator of H/D exchange rate, the ratio of areas (Amide II/Amide I) is used to counteract the absorbance differences among different measurements. This ratio is set to 0% for the samples in H2O buffer. Complete disappearance of Amide II band corresponds to 100% exchange. Results and discussion
205
pH response of OmpG-16S
206
OmpG-16S has 38 additional residues at the carboxyl end compared with OmpG-WT. The purpose of this extension of the
207
90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108
113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131
135 136 137 138 139 140 141 142 143 144 145 146 147
1 Abbreviations used: OMP, outer membrane protein; DTGS, Deuterated Tri-Glycine Sulfate; MCT, Mercury Cadmium Telluride.
Please cite this article in press as: F. Korkmaz et al., Arch. Biochem. Biophys. (2015), http://dx.doi.org/10.1016/j.abb.2015.04.010
149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165
168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204
208
YABBI 6973
No. of Pages 7, Model 5G
5 May 2015 F. Korkmaz et al. / Archives of Biochemistry and Biophysics xxx (2015) xxx–xxx 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233
polypeptide chain is to add two more b-strands to the b-barrel to increase the diameter of the channel. We have investigated the structural properties of this supposedly enlarged porin by IR-spectroscopy and compared the results with previously reported results of OmpG-WT. The b-sheet content of OmpG-WT at acidic pH is 68% and at neutral pH is 71% as shown in the X-ray structure [4] and as also confirmed by IR-spectroscopy [6,7]. A detailed analysis of the Amide I band of OmpG-WT showed that the b-sheet signal is shifted to lower wavenumbers at neutral pH compared to the same signal at acidic pH. This shift in the b-sheet signal position has been previously correlated with the hydrogen-bond strength, so that the protein is more rigid and stable at neutral pH (open pore) than it is at acidic pH (closed pore) [7]. In order to investigate the structure of OmpG-16S, the protein is equilibrated both in H2O and in D2O buffer. Many side chain groups and unordered structure are shifted when the buffer is changed, which enables the user to identify the signals and make assignments accordingly. When the spectra of OmpG-16S mutant at two pH values in H2O and in D2O buffer are analyzed, a similar shift of b-sheet signal is seen from both absorbance spectra and their respective second derivative profiles (Fig. 1). Amide I peak position for OmpG-16S is located approximately at 1630 cm1 in both buffer conditions, which indicates the overwhelming b-sheet in structure similar to OmpG-WT [11,15,16]. However, this position depends on the buffer pH. While the Amide I band is located at
3
1631 cm1 at pH 5.5, it shifts down to 1628 cm1 at pH 7.5 in H2O buffer (Fig. 1A). A similar spectral shift of 1.2 cm1 is also seen in D2O buffer (Fig. 1B). The second derivative profiles of these spectra show that the observed shift is due to the component located at 1628 cm1 (Fig. 1C and D). The shift, as seen from the second derivative spectra, is by 2.1 cm1 in H2O buffer and 1.9 cm1 in D2O buffer. There are also minor differences in the 1640–1680 cm1 region between the two pH/pD conditions of the protein. The Amide I peak position also differs between the spectra taken in H2O and D2O buffers of the same pH values (Fig. 1A and B). While the peak is located at 1631 cm1 in H2O buffer pH 5.5, it is located at 1628 cm1 in D2O buffer. This difference is mainly due to the unordered structure signal that is located at 1660 cm1 in H2O buffer and at 1640 cm1 in D2O buffer [16]. Another factor contributing to the difference is the side chain absorbance of Arg, which is located at 1680 cm1 in H2O buffer and at 1580 cm1 in D2O buffer [17]. Side chain group and unordered structure absorbance shifts observed in two buffer media explain the difference in the corresponding Amide I band profiles. Inspection of the Amide I band and its second derivative profile definitely points to the fact that pH change affects the secondary structure of OmpG-16S in a way similar to OmpG-WT. We also wanted to verify that the spectral changes in response to pH are not special to the two pH values used, i.e. pH 5.5 and 7.5; rather
Fig. 1. IR absorbance spectra of OmpG-16S in 0.1 M K-phosphate +0.5% OG buffer at acidic (–) or neutral ( ) pH/pD in H2O (A) and in D2O buffer (B). Their corresponding second derivative profiles are shown below (C and D). Absorbance spectra are shown for the entire fingerprint region (1750–1300 cm1); however, the second derivative 1 profiles are shown only for the Amide I and II regions (1700–1500 cm ) for better visualization of the difference between the two spectra, particularly the shift of low frequency b-sheet component.
Please cite this article in press as: F. Korkmaz et al., Arch. Biochem. Biophys. (2015), http://dx.doi.org/10.1016/j.abb.2015.04.010
234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258
YABBI 6973
No. of Pages 7, Model 5G
5 May 2015 4
F. Korkmaz et al. / Archives of Biochemistry and Biophysics xxx (2015) xxx–xxx Table 1 Curve-fitting results of Amide I region, listing the positions of sub-bands and their respective band area for pD 5.5 and pD 7.5 for OmpG-WT and OmpG-16S. Suggested band assignments are based on X-ray data of OmpG-WT. Band position (1/ cm)
1625 1628 1637 1640 1647 1656 1665 1673 1683 1686 1693 1696 Fig. 2. The low frequency mode of b-sheet signal position as a function of buffer pD. The peak position is followed from the second derivative of IR absorbance spectra of OmpG-16S incubated in 0.1 M K-phosphate +0.5% OG buffer in D2O, incubated for 24 h at indicated pD values.
279
it is the increase in pH that the protein structure responds. Thus, a pH-titration experiment is performed and the low frequency b-sheet signal position is followed at each pH value (Fig. 2). Protein samples are equilibrated in D2O buffer for 24 h with pD values ranging from 5.0 to 8.0. The b-sheet signal position is observed to shift gradually down to lower wavenumbers starting from pD 5.0 to pD 7.0; however, further increasing pD to 7.5 caused a dramatic downshift of the b-sheet peak position. Between pD 7.5 and 8.0, there is no significant difference in the position. Although data clearly show that OmpG-16S exhibits a gradual structural response to increasing pD, the profile of the spectral shift is somewhat different than the previously reported profile of OmpG-WT [6]. The b-sheet peak position of OmpG-WT has been observed to stay constant below pD 6.0 and above 7.5, which pointed out to well-defined conformations as open (>pD 7.5) and closed state (
