YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx 1
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi 4 5
Folding and stability studies on C-PE and its natural N-terminal truncant
3 6 7 8 9 11 10 12 1 2 4 6 15 16 17 18 19 20 21 22 23 24 25
Q1
Khalid Anwer a, Asha Parmar b, Safikur Rahman a, Avani Kaushal b, Datta Madamwar b, Asimul Islam a, Md. Imtaiyaz Hassan a,⇑, Faizan Ahmad a a b
Centre for Interdisciplinary Research in Basic Sciences, Jamia Millia Islamia (Central University), Jamia Nagar, New Delhi 110 025, India BRD School of Biosciences, Sardar Patel University, Vallabh Vidyanagar, Gujarat 388 120, India
a r t i c l e
i n f o
Article history: Received 18 November 2013 and in revised form 3 January 2014 Available online xxxx Keywords: Biliproteins C-phycoerythrin Chromophore Protein denaturation Folding and stability
a b s t r a c t The conformational and functional state of biliproteins can be determined by optical properties of the covalently linked chromophores. a-Subunit of most of the phycoerythrin contains 164 residues. Recently determined crystal structure of the naturally truncated form of a-subunit of cyanobacterial phycoerythrin (Tr-aC-PE) lacks 31 N-terminal residues present in its full length form (FL-aC-PE). This provides an opportunity to investigate the structure–function relationship between these two natural forms. We measured guanidinium chloride (GdmCl)-induced denaturation curves of FL-aC-PE and Tr-aC-PE proteins, followed by observing changes in absorbance at 565 nm, fluorescence at 350 and 573 nm, and circular dichroism at 222 nm. The denaturation curve of each protein was analyzed for DGD , the value of Gibbs free energy change on denaturation (DGD) in the absence of GdmCl. The main conclusions of the this study are: (i) GdmCl-induced denaturation (native state M denatured state) of FL-aC-PE and Tr-aC-PE is reversible and follows a two-state mechanism, (ii) FL-aC-PE is 1.4 kcal mol1 more stable than Tr-aC-PE, (iii) truncation of 31-residue long fragment that contains two a-helices, does not alter the 3-D structure of the remaining protein polypeptide chain, protein–chromophore interaction, and (iv) amino acid sequence of Tr-aC-PE determines the functional structure of the phycoerythrin. Ó 2014 Published by Elsevier Inc.
27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
43 44
Introduction
45
Cyanobacteria are the most primordial among oxygenic photosynthetic organisms, and their antenna system consists of characteristic phycobiliproteins (PBPs)1 which are categorized as phycoerythrin (PE), phycocyanin (PC) and allophycocyanin (APC) that form a supramolecular assembly called phycobilisomes (PBS) [1–3]. PBS is a light harvesting antennae of photosystem II in cyanobacteria, red algae and glaucophytes. These protein complexes are anchored to thylakoid membranes which are made of stacks of PBPs and their associated linker polypeptides [4]. Each PBS consists of a core made of APC from which several outwardly oriented rods made of stacked disks of PC and (if present) PE or phycoerythrocyanin [3,5]. The energy transfer in the organism proceeds successively in the direction PE ? PC ? APC ? chlorophyll a, with an overall efficiency of almost 100% [1,2,6]. The lights absorbed by PBP are in the regions of visible spectrum where chlorophyll a has lower efficiency. The spectral properties of PBPs are mainly governed by their
46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
⇑ Corresponding author. E-mail address: [email protected] (M.I. Hassan). Abbreviations used: GdmCl, guanidinium chloride; CD, circular dichroism techniques; FL-aC-PE, full length aC-PE; Tr-aC-PE, truncated aC-PE; a6b6, hexamer; a2b2, dimers; (a3b3)2, dimers of trimers; PE, phycoerythrocyanin; PBPs, phycobiliproteins; PE, phycoerythrin; PC, phycocyanin; APC, allophycocyanin; PBS, phycobilisomes; PEBs, phycoerythrobilins; PyMOL, Molecular Graphics System. 1
prosthetic groups which are linear tetrapyrroles known as phycobilin including phycocyanobilin, phycoerythrobilin, phycourobilin and phycoviolobilin. Interestingly, spectral properties of phycobilins are highly influenced by the protein environment [1,2,7]. PE is one of the unique and special accessory light harvesting pigment and exists in the form of hexamer (a6b6), dimers (a2b2) or dimers of trimers (a3b3)2 in light harvesting system of bluegreen algae, cryptophytes and red algae [1]. PE of blue-green algae (C-PE) contains 6a and 6b subunits arranged in ring-like assemblies where the 3 monomers are present in the threefold symmetry [8,9]. Each a-subunit of PE consists of 164 amino acid residues while each b-subunit is 171-residue long [1,9]. The structure of C-PE resembles that of the globin protein which contains two open-chain linear tetrapyrrole chromophores known as phycoerythrobilins (PEBs) covalently attached to each a-subunit at Cys82 and Cys139. While b-subunit of C-PE contains three chromophores where two are covalently attached to the Cys84 and Cys155, and third chromophore is doubly linked to Cys50 and Cys61 [1,8,9]. The interactions between PEB with protein atoms and protein–chromophore environments are critically responsible for the photon absorption properties of C-PE. Recently, we identified and determined the crystal structure of a-subunit of C-PE from Phormidium tenue, which is devoid of 31 residues from its amino terminal side [9] as compared to the full length aC-PE containing complete 164 residues [8]. It is interesting
0003-9861/$ - see front matter Ó 2014 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.abb.2014.01.005
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82
83 84 85
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 2 86 87 88 89
90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
to note that truncation of 31 N-terminal residues occurs naturally when the organism is grown or kept in prolonged starved condition. However, the truncated aC-PE maintains its light absorbing capabilities [8,9]. The full length aC-PE (FL-aC-PE) and truncated aC-PE (Tr-aC-PE) can be used as a model system to study protein folding problems in order to establish the role of 31 N-terminal residues. We, therefore, studied folding/unfolding of the full length and truncated aC-PEs using absorption, fluorescence and circular dichroism (CD) techniques. Apart from determination of role of 31 N-terminal residues in protein folding and stability, our findings may be helpful to understand the mechanism of survival of this cyanobacterium and other marine algae under stress conditions. As under these conditions including starvation the organism maintains its pigment proteins and absorption properties for food preparation and survival. Furthermore, our study may also provide an insight to grow higher photosynthetic plants under this and similar types of stress environmental conditions. This report on structure, folding and stability of FL-aC-PE and naturally truncated Tr-aC-PE will be helpful for better understanding of mechanism of energy transfer in photosynthetic cyanobacteria, even under stress conditions like starvation and salinity. Furthermore, we established the role of 31 N-terminal residues in the folding and stability of C-PE.
109
Materials and methods
110
Materials
111
Guanidinium chloride (GdmCl) of ultra pure grade was purchased from MP Biomedicals, LLC (Illkirch, France). Sodium cacodylate trihydrate and Sephadex G-150 matrix were obtained from Sigma–Aldrich (St Louis, USA) and GE Healthcare UK Limited, respectively. Other chemicals were purchased from local suppliers. All chemicals used were of molecular biology research grade.
112 113 114 115 116
117
Isolation and purification
118
Protein was isolated and purified from P. tenue sp. A27DM collected from rocky shores of Bet Dwarka, Okha and estuarine mouth of river Daman Ganga (India). The enrichment, provision of growth conditions, extraction and purification of C-PE were carried out using our well optimized protocol described elsewhere [10–12]. Briefly, the cyanobacterium was harvested in artificial sea water nutrients, and the cell mass was washed with 1.0 M Tris–HCl buffer (pH 8.1) containing 3.0 mM sodium azide. For the purification of FL-aC-PE and Tr-aC-PE, cultures were harvested when they were young (20 days) and old (90–200 days), respectively. The procedures for getting both types of proteins have already been published [9–11]. Washed cell mass was resuspended in the same buffer, and repeated freeze–thaw cycles (30 °C and 4 °C) were applied for C-PE release. This was followed by two-step (20% and 70%) ammonium sulfate precipitation. The precipitate was dissolved in 10 mM Tris buffer and subjected to gel (G-150) permeation chromatography column pre-equilibrated with 10 mM Tris buffer. The purity of the purified protein was checked by the native-PAGE and SDS–PAGE using silver staining [11,13,14]. The entire purification procedure was carried out in dark at 4 °C. All buffer solutions required above during isolation and purification were prepared in mili-Q water supplemented with 0.01% sodium azide. Stock solutions of both proteins were stored in the Tris– HCl buffer (pH 8.1) containing 0.01% sodium azide at 4 °C. To remove buffer constituents and sodium azide, stock solutions were dialyzed against several changes with 0.1 M KCl (pH 7.0). The concentrations of the dialyzed and filtered stock solutions
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
of FL-aC-PE and Tr-aC-PE were determined using a value of 265,000 M1 cm1 molar absorption coefficient at 565 nm [15].
145
Sample preparation for denaturation and renaturation experiments
147
For the study of GdmCl-induced denaturation, 8 M stock solution of the denaturant was prepared in 50 mM sodium cacodylate buffer (pH 7.0) containing 0.1 M KCl. Following the procedure reported earlier [16], protein solutions for denaturation and renaturation experiments were prepared. Briefly, for each denaturation experiment known amounts of the protein stock solution, buffer and concentrated denaturant solution were mixed and incubated for a time long enough to ensure the complete denaturation. This was cross-checked by taking absorbance. A similar procedure was employed in preparing the solution for renaturation experiments with the exception that the protein was first denatured by adding denaturant solution and then dilated with the buffer. The GdmCl-induced denaturation was studied at pH 7.0. We have always checked the pH of the protein solutions thus prepared.
148
Measurements of absorption spectra
162
Absorption spectra of FL-aC-PE and Tr-aC-PE were measured in Jasco UV/visible spectrophotometer (Jasco V-660, Model B 028661152) equipped with peltier-type temperature controller (ETCS-761). All measurements were carried out in the wavelength range 700–350 nm using 1 cm path length cuvettes. The protein concentration used was in the range of 0.1–0.3 mg ml1. All spectral measurements were done at least three times at 25 ± 0.1 °C.
163
Measurements of fluorescence spectra
170
Fluorescence measurements were carried out in Jasco spectrofluorimeter (Model FP-6200) using 5 mm quartz cuvette. All experiments were carried out at 25 ± 0.1 °C which is maintained by an external thermostated water circulator. FL-aC-PE and Tr-aC-PE have intrinsic (Trp77) and extrinsic (PEB) fluorophores. To measure Trp77 fluorescence the protein was excited at 287 nm and emission spectra was recorded in the range 300–400 nm. To measure PEB’s fluorescence, protein solution was excited at 479 nm, and emission spectra were recorded in the range 500–600 nm [17]. The protein concentration used was 0.1–0.2 mg ml1.
171
CD measurements
181
The far-UV and Soret CD measurements of FL-aC-PE and Tr-aC-PE were carried out in Jasco spectropolarimeter (J-715) equipped with peltier type of temperature controller (PTC348WI). CD spectra of FL-aC-PE and Tr-aC-PE were collected with a response time of 1 s and a scan speed of 100 nm min1. Each spectrum was an average of at least five scans. All measurements were carried out at 25 ± 0.1 °C. The far-UV spectra were collected in the wavelength range 250–200 nm using 0.1 cm path length cuvette. The protein concentration used was 0.2–0.4 mg ml1. Soret CD spectra measurements of FL-aC-PE and Tr-aC-PE were carried out in the wavelength range 600–400 nm using protein concentration of 0.3–0.5 mg ml1 in a cuvette of a path length of 1.0 cm. The raw data at a given wavelength, k (nm) were converted into concentration-independent parameter [h]k (deg cm2 dmol1), the mean residue ellipticity at wavelength k using the relation:
182
½hk ¼ M 0 hk =10lc
ð1Þ
where hk is the observed ellipticity in millidegrees at wavelength k, M0 is the mean residue weight of the protein, l is the path length of the cell in centimeters, and c is the protein concentration in mg ml1.
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
146
149 150 151 152 153 154 155 156 157 158 159 160 161
164 165 166 167 168 169
172 173 174 175 176 177 178 179 180
183 184 185 186 187 188 189 190 191 192 193 194 195 196
197 199 200 201 202 203
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx 204
Analysis of data
205
Sigmoidal GdmCl-induced denaturation curve (plot of a physical property (y) versus [GdmCl], the molar concentration of the denaturant) were analyzed for stability parameters (DGD , the value of DGD (Gibbs free energy change) in the absence of the GdmCl; m, the slope (=oDGD/o[GdmCl]); and Cm, the midpoint of the denaturation curve, i.e., [GdmCl] at which DGD = 0. This analysis involves a least-squares method to fit the entire denaturation curve, obtained at constant pH and temperature T (K), according to the relation:
206 207 208 209 210 211 212
213
y ¼ yN þ yD Exp½ðDGD m½GdmClÞ=RT=1 þ Exp½ðDGD 215 216 217 218 219 220 221 222 223 224 225
226
m½GdmClÞ=RT
ð2Þ
where yN and yD are the properties of the native (N) and denatured (D) protein molecules under the same experimental condition in which y has been measured, R is gas constant, and T is temperature (K). It should be noted that Eq. (2) assumes that a plot of DGD versus [GdmCl] is linear, and the dependencies of yN and yD on [GdmCl] are also linear (i.e., yN = aN+bN [GdmCl], yD = aD+bD [GdmCl], where aN, bN, aD and bD are [GdmCl]-independent parameters, and subscripts N and D signify that parameters are for the native and denatured states, respectively). The denatured fraction of the protein, fD at a given [GdmCl] was determined using the relation:
fD ¼ ðy yN Þ=ðyD yN Þ 228
¼ y ðaN þ bN ½GdmClÞ=ðaD aN Þ þ ðbD bN Þ½GdmCl
ð3Þ
229
In silico structural analysis
230
241
The crystal structure of Tr-aC-PE from P. tenue is already (PDB code 3MWN). However, the crystal structure of FL-aC-PE is still not determined. Hence, we modeled the structure of FL-aC-PE using modeler [18]. The energy minimized coordinate [19] was validated using online Structural Analysis and Verification Server (http://nihserver.mbi.ucla.edu/SAVES/), showing most of the residues in allowed region of Ramachandran plot (Fig. S1). A superimposed ribbon diagrams of full length and truncated aC-PE were drawn using PyMOL (The PyMOL Molecular Graphics System, Version 1.2r3pre, Schrödinger, LLC) [20]. Contacts offered by various residues were measured by using the contact program from the CCP4 suite [21].
242
Results
243
Polyacrylamide gel electrophoresis
244
247
The purified FL-aC-PE and Tr-aC-PE showed single band on both the native PAGE and SDS–PAGE (results not shown) [10]. This observation led us to conclude that both full length and truncated proteins exist as monomer.
248
Spectral characterization
249
Insets in Fig. 1 show visible absorbance spectra of the native FL-aC-PE and Tr-aC-PE at pH 7.0 (50 mM sodium cacodylate containing 0.1 M KCl) and 25 ± 0.1 °C. It has been argued that this absorption of both proteins is due to the interaction between the folded polypeptide backbone and the two thioether bonded PEBs [10,11]. This argument was based on the fact that the apoprotein does not absorb in the visible region, and PEB has a very low absorbance [1,2,22]. These findings suggest that the folding/unfolding reactions of FL-aC-PE and Tr-aC-PE can be followed by observing changes in the visible absorption at 565 nm. To see the effect of GdmCl on the structural stability of FL-aC-PE and Tr-aC-PE, we
231 232 233 234 235 236 237 238 239 240
245 246
250 251 252 253 254 255 256 257 258 259
3
measured the absorption spectra of both proteins in the presence of different concentrations of GdmCl in the range 700–350 nm. Results of these measurements are shown in Fig. S2 (also see insets in Fig. 1). Fig. 1A and B respectively show plots of De565 (difference in molar absorption coefficient at 565 nm in the presence of GdmCl and that in the absence of GdmCl) of FL-aC-PE and Tr-aC-PE as a function of [GdmCl]. These plots were analyzed for DGD m and Cm values for each protein according to Eq. (2). Results of this analysis are shown in Table 1. Both the FL- and Tr-aC-PE proteins have a lone tryptophan at position 77 (numbering is according to the full length protein). This intrinsic fluorophore is buried and is at a distance of 17 Å from the second chromophore (PEB2) attached at Cys139 [9]. After excitation at 287 nm, we measured tryptophan emission spectra of both proteins in the native (in 50 mM cacodylate buffer) and denatured (GdmCl solution) states at pH 7.0 and 25 ± 0.1 °C (see insets in Fig. 2). To make corrections for the effect of solvent (GdmCl solution) on the D state of the protein, following procedure was used. Values of fluorescent intensity at different wavelengths were determined from the observed emission spectra of a protein in the presence of [GdmCl] > 2.5 M i.e., the post transition region (Figs. S3 and S4). At a given wavelength, values of fluorescent intensity were plotted against [GdmCl]. To obtain the solvent independent value, an extrapolated value at [GdmCl] = 0 M was determined from this plot. These extrapolated values of fluorescent intensity values at different wavelengths were used to obtain the fluorescence spectrum of the denatured state which does not depend on the denaturant concentration. It has been shown that kmax of tryptophan fluorescence spectrum shifts to longer wavelengths and the intensity at kmax decreases on transferring this fluorophore from non-polar medium to polar medium [23,24]. As can be seen in Fig. 2, when a protein is transferred from the native buffer to the denaturant solution, there is an expected red shift of the emission spectrum from 350 to 360 nm. However, contrary to this expectation, the emission intensity of the proteins is increased on denaturation. This increase can be explained by invoking quenching phenomenon due to the presence of PEB2 near Trp77, which will decrease when the protein unfolds [22]. We exploited this phenomenon to see the effect of GdmCl on emission spectra of FL- and Tr-aC-PEs. These spectra are shown in Fig. S3. Values of emission intensity at 350 nm (F350) of both proteins are read from these spectra and plotted as a function of [GdmCl] in Fig. 2A and B. To obtain stability parameters (DGD , m and Cm) each denaturation curve shown in Fig. 2 was fitted to Eq. (2). Results of this analysis for both proteins are shown in Table 1. PEB has fluorescence property [25–27]. It has been shown that the covalently linked PEBs bonded to the folded protein polypeptide chain via thioether bonds have a strong emission spectrum in the wavelength region 500–600 nm, which vanishes on protein denaturation [26,28]. After exciting FL-aC-PE and Tr-aC-PE at 479 nm, we measured emission spectra of both proteins in the presence of different concentrations of GdmCl at pH 7.0 and 25 ± 0.1 °C. These spectra are shown in Fig. S4. Fig. 3A and B show the change in F573, the fluorescence intensity of FL-aC-PE and Tr-aC-PE at 573 nm as a function of [GdmCl], respectively. For each protein, GdmCl-induced denaturation curves were analyzed for DGD , m and Cm values by fitting the entire denaturation curve according to Eq. (2). The results of this analysis are shown in Table 1. To see the effect of GdmCl on the stability of FL-aC-PE and Tr-aC-PE, we measured the far-UV CD spectra of both proteins in the presence of different concentrations of GdmCl. These measurements are shown in Fig. S5. Values of [h]222 (probe for the secondary structure) in the presence of different concentrations of GdmCl were determined from these CD spectra shown in this figure. Fig. 4A and B show plots of [h]222 versus [GdmCl] for FL-aC-PE and Tr-aC-PE,
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 4
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
1
A
30
Native FL-αC-PE
ελ x 10-4, M-1 cm-1
25
20
0
-3 -1 -1 Δε565 x 10 , M cm
15
10
5
-1
Denatured FL-αC-PE 0 350
400
450
500
550
600
650
700
Wavelength, nm
-2
-3
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M 1
B
30
Native Tr-αC-PE
ελ x 10-4, M-1cm-1
25
-3 -1 -1 Δε565 x 10 , M cm
0
20
15
10
5
-1
Denatured Tr-αC-PE 0 350
400
450
500
550
600
650
700
Wavelength, nm
-2
-3
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M Fig. 1. GdmCl-induced denaturation of FL-aC-PE and Tr-aC-PE at pH 7.0 and 25 °C. (A) Denaturation curve of FL-aC-PE followed by observing changes in De565 as a function of [GdmCl]. (B) Plots of De565 of Tr-aC-PE versus [GdmCl]. Open and filled symbols show the results obtained from denaturation and renaturation experiments. The insets in (A) and (B) show the absorption spectra of FL-aC-PE and Tr- aC-PE of the native and GdmCl-induced denatured states, respectively.
Table 1 Stability parameters of FL-aC-PE and Tr-aC-PE associated with the GdmCl-induced denaturation at pH 7.0 and 25 ± 0.1 °C.a Probes (Fit parameter)
De565 F350 F573 [h]222 a
DGD (kcal mol1 M1)
m (kcal mol1 M1)
Cm (M)
FL-aC-PE
Tr-aC-PE
FL-aC-PE
Tr-aC-PE
FL-aC-PE
Tr-aC-PE
11.45 ± 0.34 11.77 ± 0.38 11.56 ± 0.23 11.82 ± 0.21
10.11 ± 0.40 10.42 ± 0.66 10.16 ± 0.27 10.47 ± 0.56
5.67 ± 0.44 6.13 ± 0.35 5.99 ± 0.56 6.16 ± 0.49
5.77 ± 0.36 6.35 ± 0.22 5.83 ± 0.39 6.41 ± 0.57
2.02 ± 0.43 1.92 ± 0.24 1.93 ± 0.35 2.03 ± 0.27
1.75 ± 0.29 1.68 ± 0.22 1.74 ± 0.34 1.77 ± 0.30
A ± with a value is a mean of errors from the average of three independent measurements.
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
5
Fig. 2. GdmCl-induced denaturation of FL-aC-PE and Tr-aC-PE followed by observing changes in F350 at pH 7.0 and 25 °C. (A) and (B) show denaturation curves of FL-aC-PE and Tr-aC-PE, respectively. Open and filled symbols have same meaning as in Fig. 1. Fluorescence spectra of FL-aC-PE (inset in (A)) and Tr-aC-PE (inset in (B)) in their native and denatured states, respectively. The excitation wavelength was 287 nm. D state of each protein was corrected for the solvent effect.
326 327 328 329 330
331 332
respectively. Each of these plots (denaturation curves) was analyzed for values of DGD , m and Cm by fitting the entire denaturation curve according to Eq. (2). These stability parameters thus obtained are given in Table 1. The insets in Fig. 4 show the far-UV CD spectra of FL-aC-PE and TraC-PE. We have used k2d program (http://dichroweb.cryst.bbk.ac.uk/ html/process.shtml) to estimate the percentage of elements of sec-
ondary structure of those proteins. Analysis of the far-UV CD spectrum of the native FL-aC-PE (see inset in Fig. 4A) for the contents of a-helix and b-sheet yielded values of 75% and 2%, respectively. Analysis of the far-UV CD spectrum of Tr-aC-PE (see inset in Fig. 4B) for the secondary structure gives 60% and 2% for the a-helix and b-structure, respectively, which is in excellent agreement with that determined by the X-ray diffraction studies [9].
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
333 334
335 336 337 338 339
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 6
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
60
A
F573 (a. u.)
Fluorescence (a. u.)
60
Native FL-αC-PE 40
Denatured FL-αC-PE
20
40 0 520
540
560 Wavelength, nm
580
600
20
0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M 60
B Fluorescence (a. u.)
Native Tr-αC-PE
F573 (a. u.)
60
40
Denatured Tr-αC-PE
20
40 0 520
540
560 Wavelength, nm
580
600
20
0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M Fig. 3. Denaturation curves of FL-aC-PE and Tr-aC-PE induced by GdmCl at pH 7.0 and 25 °C. (A) Plot of F573 of FL-aC-PE versus [GdmCl]. (B) Plot of F573 of Tr-aC-PE versus [GdmCl]. Open and filled symbols have the same meaning as in Fig. 1. Insets in (A) and (B) show the fluorescence spectra of the native and denatured FL-aC-PE and Tr-aC-PE, respectively. Spectra of the denatured proteins were corrected for the solvent effect. For both proteins excitation wavelength was 479 nm.
340
Refolding experiments
341
In order to check the reversibility of denaturation of FL-aCPE and Tr-aC-PE, proteins were exposed to concentrated GdmCl solution ([GdmCl] P 2.5 M) for a time long enough to ensure complete denaturation. The denatured protein was diluted with the native buffer to check the reversibility of denaturation at 3.0, 2.5, 2.0 and 0.5 M GdmCl. Fig. 5A–H show representative denaturation (curve 1) and renaturation (curve 2) spectra. It has been observed that for each protein in a given denaturant
342 343 344 345 346 347 348
concentration, spectra obtained from denaturation and renaturation experiments are indistinguishable (also see Figs. 1–4).
349
In silico structure analysis
351
A structural comparison of FL-aC-PE and Tr-aC-PE suggests a difference of 31 residues from N-terminal side, which is comprised of two a-helices. This segment is completely away from the whole compact polypeptide chain (Fig. 6). It is seen in Fig. 6 that there are only 6 hydrogen bonds between residues
352
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
350
353 354 355 356
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 7
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx 5
A
[θ ]222 x 10-3, deg cm2 dmol-1
0
-5
[θ ]λ x 103, deg cm2 dmol-1
10
5
Denatured FL-α C-PE 0
-5
-10
-15
Native FL-αC-PE
-20
-25 200
205
210
215
220
225
230
235
240
245
250
Wavelength, nm -10
-15
-20
-25 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M 5
B
2 -1 [θ ] x 10-3, deg cm dmol 222
0
-5
[θ ]λ x 103, deg cm2 dmol-1
10
5
Denatured Tr-α C-PE
0
-5
-10
-15
Native Tr-α C-PE -20
-25 200
205
210
215
220
225
230
235
240
245
250
Wavelength, nm
-10
-15
-20
-25 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M Fig. 4. GdmCl-induced denaturation curves of FL-aC-PE and Tr-aC-PE at pH 7.0 and 25 °C. For the full length and truncated proteins (A) and (B) show the plots of [h]222 versus [GdmCl], respectively. Open and filled symbols have the same meaning as in Fig. 1. The insets in (A) and (B) show the far-UV CD spectra of the native and denatured states of FL-aC-PE and Tr-aC-PE, respectively.
357 358 359 360 361 362 363 364 365 366
Lys2 and Gly105, Gly29 and Arg33, Gln28 and Gln32, Asn30 and Gln32, Gln28 and Gln32 (through side chain) and Ile31 and Gln32 (see Table 2). However, for the sake of clarity the first three hydrogen bonds are shown in Fig. 6. Furthermore, this N-terminal segment is involved in 25 van der Waals interactions (see Table 2). Since these 31 residues of FL-aC-PE are completely away from the packed polypeptide core, these are expected to be highly exposed to water, if the protein exists in the form of a monomer. The crystal structure of Tr-aC-PE was recently determined [9]. It was observed that this protein belongs to a class of all
a-protein. Likewise, our modeled structure of FL-aC-PE shows the presence of 75% a-helical content. Fig. 6 shows overlapping 3-D structures of Tr-aC-PE and FL-aC-PE. The 31-residue long N-terminal sequence of FL-aC-PE contains 12% a-helix.
367
Discussion
371
The analysis of transition curves (Figs. 1–4) for the stability parameters, DGD , m and Cm (Table 1) according to Eq. (2) is based
372
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
368 369 370
373
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 8
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx 30
A
3 2
25
x 10-4 , M-1 cm-1
15 10
2
20 15 10 5
5
1
1 0
0 350
400
450
500
550
600
650
350
700
400
450
500
550
600
650
700
Wavelength, nm
Wavelength, nm 15
15
C
D 1 Fluorescence (a. u.)
1
Fluorescence (a. u.)
B
3 25
20
x 10-4 , M-1 cm-1
30
10
2 5
10
2 5
3 3 0 300
320
340
360
380
0 300
400
320
340
360
380
400
Wavelength, nm
Wavelength, nm 60
E
3
60
F
40
Fluorescence (a. u.)
Fluorescence (a. u.)
3
2
20
2
40
20
1
0 520
540
1
560
580
0 520
600
540
560
10
H
5
[ ]λ x 10-3 , deg cm2 dmol-1
[ ]λ x 10-3 , deg cm2 dmol-1
600
10
G
5 0 -5
1
-10 -15
2
-20 -25 200
580
Wavelength, nm
Wavelength, nm
210
215
220
225
-5
1
-10
2 -15 -20
3
-25
3 205
0
230
Wavelength, nm
235
240
245
250
200
205
210
215
220
225
230
235
240
245
250
Wavelength, nm
Fig. 5. Renaturation of the GdmCl-denatured FL-aC-PE and Tr-aC-PE at pH 7.0 and 25 °C. Absorption spectra of (A) FL-aC-PE and (B) Tr-aC-PE: (1) denatured protein in 2.5 M GdmCl, (2) renatured protein in 0.5 M GdmCl, and (3) native protein. (C) and (D) show Trp fluorescence spectra, (E) and (F) show PEB fluorescence spectra, and (G) and (H) the far-UV CD spectra of FL-aC-PE and Tr-aC-PE, respectively. Spectra meaning has the same meaning as in (A).
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
9
Gly105
Lys2
Asn30
Gln32
Gly29 Ile31
Gln28
Arg33
Fig. 6. Overlaid structure of Tr-aC-PE (light green) and FL-aC-PE illustrated in cartoon model. The chromophores, PEBs are shown in ball and stick model. 31 N-terminal residues forming hydrogen bonded interactions with other part (32–164) of molecule are shown in ball and stick. Structure is drawn in PyMol using the atomic coordinates of FL-aC-PE and aC-PE from the Phormidium tenue (PDB code: 3MWN). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395
on a few assumptions. A few comments are therefore necessary. First, it was assumed that the GdmCl-induced denaturation of the full length and truncated aC-PE proteins are reversible. To check the reversibility of denaturation, we denatured each protein by adding different concentrations of the denaturant. The GdmCl-denatured protein was diluted with the native buffer. After incubation of the denatured protein in the buffer for a time enough for the completion of the renaturation process, optical spectra (absorption, fluorescence and CD) were recorded. A spectrum of the renatured protein was then compared with that obtained during denaturation experiments. It was observed that the spectral properties of renatured protein in a given [GdmCl] are identical to those obtained during denaturation experiments. Fig. 5 shows the representative data which were obtained on denaturing the protein by 2.5 M GdmCl (curve 1), and renaturing this denatured protein by diluting with the buffer to bring the denaturant concentration to 0.5 M (curve 2). It has been observed that the spectra of the protein in a given [GdmCl] obtained from the denaturation and renaturation experiments were indistinguishable. The spectra of the denatured and renatured proteins are also compared with those of the native proteins (curve 3; Fig. 5). The coincidence of optical properties of the proteins obtained from denaturation and
renaturation experiments (Fig. 5 and also see Figs. 1–4) led us to conclude that the GdmCl-induced denaturation of FL-aC-PE and Tr-aC-PE is reversible. The second assumption was that the GdmCl-induced transition between N and D states is a two-state process. The most authentic criterion for a two-state (native and denatured) behavior of the protein denaturation is to show that the calorimetric and van’t Hoff enthalpy changes on denaturation are identical. We could not succeed to make differential scanning calorimetry measurements on Fl-aC-PE and Tr-aC-PE proteins due to the fact that the heat denatured proteins formed aggregates. However, another test for the two-state behavior of the protein denaturation is to show coincidence of the normalized sigmoidal curves of different physical property induced by GdmCl [29]. Fig. 7A and B show normalized denaturation curves (plots of fD versus [GdmCl]) of the full length and truncated proteins, respectively. It is seen in these figures that for a protein, the normalization of each transition curve of Figs. 1– 4, using Eq. (3) yields a coincidence of curves of different physical properties. As pointed out earlier [29], this is a necessary but not sufficient criterion for a two-state behavior of the protein denaturation. Customarily, the observation of non-coincidence of normalized transition curves of different physical properties constitutes a
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 10
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
Table 2 Hydrogen bonds and van der Waals Interactions that the N-terminal segment (1–31 residues) forms with atoms of the remaining polypeptide chain (32–164) of aR-PE.a Residues involved in non-covalent interactions
Bond length
1MetCa 1MetSd 1MetCe
3.30 3.43 3.47 3.42 3.48 3.27 3.12 3.18 3.48 2.84 3.31 3.18 3.10 3.22 3.08 3.41 2.81 3.42 3.23 3.05 2.86 2.89 2.88 3.00 3.23 3.29 2.62 3.43 2.79 3.15 3.38
2LysN 2LysO
28GlnO
28GlnCd 28GlnNe2 29GlyO 30Asn N 30AsnC 30AsnO c
30AsnC 30AsnOd1 30AsnNd2 31IleN 31IleCa 31IleC 31IleO
a
418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449
109GluOe2 106ProCd 109GluCd 109GluOe1 105GlyCa 105GlyN 103GlyCa 103GlyC 103GlyO 32GlnN 32GlnCc 32GlnNe2 32GlnNe2 32GlnN 33ArgN 32GlnN 34AlaN 34AlaCb 33ArgN 37ArgNg2 37ArgNg2 37ArgNg2 32GlnN 32GlnC 33ArgN 34AlaN 32GlnCa 32GlnO 32GlnC 33ArgN 32GlnN
Residues forming hydrogen bonds are shown in bold letters.
proof of existence of stable intermediate(s) on the folding M unfolding pathways [29]. On the contrary, basing their arguments on theoretical grounds, Dill and Shortle [30] have shown that the non-coincidence of sigmoidal curves of two different physical properties is also observed for a two-state protein denaturation. Recently, Rahaman et al. [31] provided experimental evidence that this non-coincidence is consistent with two-state denaturation behavior. In another approach for testing the two-state behavior of GdmCl-induced denaturation, DGD values of the protein are measured from sigmoidal curves of different structural proteins, and it is shown that this thermodynamic property is independent of techniques (probes) used to monitor the denaturation [32]. Table 1 shows that values of DGD of each protein, measured by four different structural properties are, within experimental errors, identical. This overlap in DGD values suggest that GdmCl-induced denaturation of Fl-aC-PE and Tr-aC-PE is a two-state process. It is known that if absorption spectra of two non-interacting species separately intersect exactly at one or more wavelengths, then each intersecting point (isosbestic point) of absorption no longer depends on the concentration ratios of these species in their mixtures. The presence of isosbestic points on the absorption spectra of the native and denatured proteins alone and in their equilibrium mixtures is therefore taken as an evidence for a two-state denaturation [22,29]. Fig. S2 shows absorption spectra of the FL- and Tr- aC-PEs in the presence of different concentration of GdmCl in which a protein exists either in the N state (pre-transition region) or as an equilibrium mixture of N and D states (transition region) or in the D state (post-transition region). It is seen in this figure that there exist isosbestic points at wavelengths 460 and 594 nm in the case of FL-aC-PE and at wavelengths 440 and 602 nm in case of Tr-aC-PE. These observations
therefore support an assumption that GdmCl-induced denaturation of both proteins follows a two-state mechanism. Estimates of DGD from the denaturation curves depend not only on the mechanism of denaturation but also on the models describing the dependencies of the observed DGD on [GdmCl]. There are three models to estimate DGD from the extrapolation of DGD to 0 [M] GdmCl concentrations [33,34]. One model assumes that DGD value varies linearly with [denaturant]. The other one (binding model) assumes the presence of ‘specific’ binding sites for the chemical denaturant on the native and denatured protein molecules. In this model dependence of DGD on the [denaturant] is non-linear. The third model known as transfer-free energy model, describes the dependence of DGD on the transfer free energy of small model compounds for the protein groups from water to different concentrations of the chemical denaturants. It has been observed that the analysis of the same set of denaturation data using different models give different values of DGD of a protein. However, we used the linear free energy model for the analysis of denaturation curves of both FL-aC-PE and Tr-aC-PE (Figs. 1–4), for it has supports of both theory [35,36] and experiments [37–41]. It is seen in Table 1 that the full length protein is only 1.4 kcal mol1 more stable than the truncated protein, or in other words, the contribution of 31 N-terminal residues to protein stability (DGD ) of FL-aC-PE is equivalent to the stability provided by only one buried hydrogen bond (1.5 ± 1 kcal mol1) [42]. To understand this difference in stability between FL-aC-PE and Tr-aC-PE, we performed in silico study. Since the sequence of FL-aC-PE is not determined so far we have taken a conserved sequence of 31 N-terminal residues from cyanobacteria. We assumed that our model is quite similar to the crystal structure of FL-aC-PE. This assumption is supported by following observations. (i) We aligned the sequences of all a-PEs of cyanobacteria and red algae available in the UniProt database using multiple sequence alignment protcol on ClustalW2 [43]. Out of 31 N-terminal residues, only 5 positions are not fully conserved; 5A/I/A, 8V/A, 9I/V, 10A/S/G/T and 21 T/A/Q/S (see Fig. 8). This comparison indicates that there is 97% sequence homology as far as 31 N-terminal residues are concerned. (ii) A sequence alignment of Tr-aC-PE with that of 32–164 residues shows 94% homology (Fig. 8). (iii) An overlaying of 3-D structure of Tr-aCPE on that of FL-aC-PE shows almost identical structural homology (Fig. 6). (iv) Analysis of the far-UV CD spectra for the elements of secondary structures in FL-aC-PE and Tr-aC-PE (Fig. 4A and B), and comparison of these elements of secondary structure with those of Tr-aC-PE obtained from the X-ray diffraction studies (Fig. 6) led to the conclusion that 31 N-terminal residues of the full length intact protein contain 12% a-helix. This observation is identical to that observed for the same segment in the FL-aC-PE (see Fig. 6). Table 1 shows m-values of FL-aC-PE and Tr-aC-PE. The average values of this parameter (m) obtained from measurements of GdmCl-induced denaturation using different optical properties are 5.99 kcal mol1 M1 for the full length protein and 5.65 kcal mol1 M1 for the truncated protein. Myer et al. [44] have developed a method to predict m-value of a protein from its known change in the accessible surface area (DASA) on denaturation. Using their equation that relates m-value with DASA (i.e., m = 953 + 0.23 DASA), we estimated m-values of 5.65 and 5.71 kcal mol1 M1 for FL-aC-PE and Tr-aC-PE, respectively. This predicted value of m for each protein is in excellent agreement with the average value determined experimentally (also see Table 1). It should be noted that for the determination of total DASA of the protein, we have used DSSP method [45] for the estimation of ASA of the folded protein and the procedure of Creamer et al. [46] for estimation of the lower bound value of ASA of the unfolded protein. The a-subunit of each PBP contains 164 residues and consists of 9-helices packed in globular shape to form main core of the
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 11
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
1.2
A
1.0
0.8
fD
0.6
0.4
0.2
F350 F573
0.0
Δε565 [θ]222
-0.2 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M 1.2
B
1.0
fD
0.8
0.6
0.4
F350 0.2
F573 Δε565 [θ]222
0.0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M Fig. 7. Normalized GdmCl-induced denaturation curves of (A) FL-aC-PE and (B) Tr-aC-PE at pH 7.0 and 25 °C. Results shown in Figs. 1–4 were used to construct plots of fD versus [GdmCl]. 516 517 518 519 520 521 522 523 524 525 526 527
protein molecule [9]. The N-terminal segment (1–31 residues) contains only 2-helices X and Y which are antiparallel to each other and perturbed away from the main core (32–164 residues). Structural analysis suggests that residues of the N-terminal segment form a total of 30 hydrogen bonds and many van der Waals interactions among themselves (see Table 3). In addition to these non-covalent interactions this segment forms six inter-segment hydrogen bonds: Lys2 and Gly105, Gly29 and Arg33, Gln28 and Gln32, Asn30 and Gln32, Gln28 and Gln32 (through side chain) and Ile32 and Gln32) and 25 van der Waals interactions (see Table 2). It seems that among these six hydrogen bonds, only one hydrogen bond, 2LysOA105GlyN may have a significant role hold-
ing the N-terminal segment to the main core, for other hydrogen bonds are formed between neighboring atoms. Both types of non-covalent interactions (hydrogen bonding and van der Waals interactions) are known to provide stability to folded proteins [47–49]. It is then expected that the full length protein (FL-aC-PE) should be more stable than truncated protein (Tr-aCPE) which lacks 31-residues long N-terminal segment of the full length protein. This is indeed true; FL-aC-PE is 1.4 kcal mol1 more stable than the Tr-aC-PE (see Table 1). A buried hydrogen bond in the protein core (dielectric constant 4–6) provides a stability of about 2 kcal mol1 [42]. Thus an observed stability difference of 1.4 kcal mol1 between the full length
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
528 529 530 531 532 533 534 535 536 537 538 539
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 12
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
Fig. 8. Multiple sequence alignment of a-subunit of PBPs. Names of sequences is given according to the sequence identifier on the UniProt (www.uniprot.org). Non-conserved residues from N-terminal side (1–31) are shaded in yellow. Secondary structures of PBPs are illustrated on the top of sequence. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx Table 3 Residues of the N-terminal segment (1–31 residues) involved in the formation of hydrogen bonds.
540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556
Serial number
Residues involved in the hydrogen bond formation
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
6ThrN 7ThrN 7ThrOc1 8ThrN 8ThrOc1 8ThrOc1 9IleN 10SerN 11AlaN 11AlaN 12AlaN 12AlaN 13GluN 14AlaN 15AlaN 15AlaO 16GlyN 16GlyN 17ArgN 17ArgN 20SerN 23GluN 23GluN 24LeuN 24LeaN 25GlnN 25GlnO 26SerN 29GlyN 30AsnN
3SerOc1 3SerO 3SerO 4ValO 4ValO 5IleO 5IleO 6ThrO 8ThrO 9IleO 8ThrO 9IleO 9IleO 10SerO 12AlaO 13GluO 12AlaO 13GluO 12AlaO 15AlaO 23AspOd2 20SerO 21ThrO 20SerO 21hrO 21ThrO 21ThrOc1 22SerO 26SerO 27ValO
and truncated proteins appears to be too small, for 31-residue long N-terminal is involved in six hydrogen bonds formation [20]. A possible reason for this small difference in stability could be due to full or partial exposure of amino acid residues involved in non-covalent interactions, to the highly polar solvent water (dielectric constant 80). The reason for saying this is that both hydrogen bonds and van der Waals interactions are electrostatic in nature and inversely proportional to the dielectric constant of the medium. For instance, hydrogen bonds in water do not have a net stabilizing effect on protein structure due to the fact that in order to form a DAH A (where A and D are N and O atoms) bond one needs to disrupt another set of very favorable hydrogen bonds, those the protein groups (D-A and A) can make with water. Using software RVP-Net (version 1), we estimated exposure of all residues involved in non-covalent interactions. Interestingly, it has been observed that almost all groups involved in such interactions are fully or partially exposed.
557
Conclusions
558
(a) The 31-residue long N-terminal segment of the full length phycoerythrin (i) does not contribute significantly to the overall stability of the protein, for the full length protein is only 1.4 kcal mol1 more stable than the truncated protein, (ii) has 12% a-helix, and (iii) has no effect on the mechanism of folding of the protein fragment (32–164 residues), for both proteins undergo a two-state transition between N and D states. (b) The non-covalent interactions due to 31 N-terminal residues in the FL-aC-PE monomeric protein provide structural rigidity, not significant stability. (c) Deletion of the N-terminal segment does not perturb the protein–PEBs interaction, thus both FL-aC-PE and Tr-aC-PE have same absorbance capability.
559 560 561 562 563 564 565 566 567 568 569
13
Acknowledgements
570
M.I.H. and F.A. are thankful to the Department of Science and Technology (DST) for financial support (Project No. SR/SO/BB0124/2010A). K.A. is thankful to University Grant Commission (UGC) for providing Maulana Azad National Fellowship (MANF).
571
Appendix A. Supplementary data
575
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.abb. 2014.01.005.
576
References
578
[1] [2] [3] [4] [5] [6] [7] [8] [9]
[10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [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]
R. MacColl, J. Struct. Biol. 124 (1998) 311–334. R. MacColl, Biochim. Biophys. Acta 1657 (2004) 73–81. B.H. Gray, C.A. Lipschultz, E. Gantt, J. Bacteriol. 116 (1973) 471–478. J. Cornejo, S.I. Beale, M.J. Terry, J.C. Lagarias, J. Biol. Chem. 267 (1992) 14790– 14798. A.N. Glazer, Rev. Microbiol. 36 (1982) 173–198. M. Yokono, S. Akimoto, K. Koyama, T. Tsuchiya, M. Mimuro, Biochim. Biophys. Acta 1777 (2008) 55–65. R. MacColl, Polar Biol. 22 (1999) 384–388. W.R. Chang, T. Jiang, Z.L. Wan, J.P. Zhang, Z.X. Yang, D.C. Liang, J. Mol. Biol. 262 (1996) 721–731. B.R. Soni, M.I. Hasan, A. Parmar, A.S. Ethayathulla, R.P. Kumar, N.K. Singh, M. Sinha, P. Kaur, S. Yadav, S. Sharma, D. Madamwar, T.P. Singh, J. Struct. Biol. 171 (2010) 247–255. A. Parmar, Process Biochem. 46 (2011) 1793–1799. A. Parmar, N.K. Singh, A. Kaushal, S. Sonawala, D. Madamwar, Bioresour. Technol. 102 (2011) 1795–1802. G.N. Shah, V. Madamwar, Acta Bot. Mal. 26 (2001) 188–193. A. Parmar, J. Phycol. 46 (2010) 285–289. N.K. Singh, A. Parmar, D. Madamwar, Bioresour. Technol. 100 (2009) 1663– 1669. C.N. Pace, Protein Sci. 4 (1995). F. Ahmad, C.C. Bigelow, J. Biol. Chem. 257 (1982) 12935–12938. C.A. Royer, Chem. Rev. 106 (2006) 1769–1784. N. Eswar, D. Eramian, B. Webb, M.Y. Shen, A. Sali, Methods Mol. Biol. 426 (2008) 145–159. K. Joo, J. Lee, J.H. Seo, K. Lee, B.G. Kim, Proteins 75 (2009) 1010–1023. W.L. DeLano, Delano Scient, San Carlos, 2004, pp. 1–62. P. Emsley, K. Cowtan, Acta Crystallogr. D: Biol. Crystallogr. 60 (2004) 2126– 2132. K. Anwer, R. Sonani, D. Madamwar, P. Singh, F. Khan, K. Bisetty, F. Ahmad, M.I. Hassan, J. Biomol. Struct. Dyn. 31 (2013) 1–13. A. White, Biochemistry 71 (1958) 217–220. J.T. Vivian, P.R. Callis, Biophys. J. 80 (2001) 2093–2109. C.S. French, V.K. Young, J. Gen. Physiol. 35 (1952) 873–890. L.N. Liu, H.N. Su, S.G. Yan, S.M. Shao, B.B. Xie, X.L. Chen, X.Y. Zhang, B.C. Zhou, Y.Z. Zhang, Biochim. Biophys. Acta 1787 (2009) 939–946. T. Dammeyer, N. Frankenberg-Dinkel, J. Biol. Chem. 281 (2006) 27081–27089. M. Kupka, H. Scheer, Biochim. Biophys. Acta 1777 (2008) 94–103. N.C. Pace, C. Tanford, Biochemistry 7 (1968) 198–208. K.A. Dill, D. Shortle, Annu. Rev. Biochem. 60 (1991) 795–825. H. Rahaman, M.K. Alam Khan, M.I. Hassan, A. Islam, A.A. Moosavi-Movahedi, F. Ahmad, J. Chem. Thermodyn. 58 (2013) 351–358. D.W. Bolen, Methods 34 (2004) 312–322. C. Tanford, K.C. Aune, Biochemistry 9 (1970) 206–211. K.C. Aune, C. Tanford, Biochemistry 8 (1969) 4579–4585. J.A. Schellman, Biopolymers 17 (1978) 1305–1322. J.A. Schellman, Biophys. Chem. 96 (2002) 91–101. F. Ahmad, C.C. Bigelow, Can. J. Biochem. 56 (1978) 1003–1005. C.N. Pace, CRC Crit. Rev. Biochem. 3 (1975) 1–43. D.W. Bolen, M.M. Santoro, Biochemistry 27 (1988) 8069–8074. R. Gupta, F. Ahmad, Biochemistry 38 (1999) 2471–2479. R. Gupta, S. Yadav, F. Ahmad, Biochemistry 35 (1996) (1930) 11925–11930. C.N. Pace, G.R. Grimsley, S.T. Thomas, G.I. Makhatadze, Protein Sci. 8 (1999) 1500–1504. J.D. Thompson, T.J. Gibson, D.G. Higgins, Curr. Protoc. Bioinformatics 56 (2002) 59–63. J.K. Myers, C.N. Pace, J.M. Scholtz, Protein Sci. 4 (1995) 2138–2148. W. Kabsch, C. Sander, Biopolymers 22 (1983) 2577–2637. T.P. Creamer, R. Srinivasan, G.D. Rose, Biochemistry 36 (1997) 2832–2835. D. Freilder, Phys. Biochem.Ed. 2 (1982) 86–176. K.A. Dill, Biochemistry 29 (1990) 7133–7155. W. Kauzmann, Adv. Protein Chem. 14 (1959) 1–63.
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
572 573 574
577
Q2
579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642
No. of Pages 13, Model 5G
16 January 2014 Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx 1
Contents lists available at ScienceDirect
Archives of Biochemistry and Biophysics journal homepage: www.elsevier.com/locate/yabbi 4 5
Folding and stability studies on C-PE and its natural N-terminal truncant
3 6 7 8 9 11 10 12 1 2 4 6 15 16 17 18 19 20 21 22 23 24 25
Q1
Khalid Anwer a, Asha Parmar b, Safikur Rahman a, Avani Kaushal b, Datta Madamwar b, Asimul Islam a, Md. Imtaiyaz Hassan a,⇑, Faizan Ahmad a a b
Centre for Interdisciplinary Research in Basic Sciences, Jamia Millia Islamia (Central University), Jamia Nagar, New Delhi 110 025, India BRD School of Biosciences, Sardar Patel University, Vallabh Vidyanagar, Gujarat 388 120, India
a r t i c l e
i n f o
Article history: Received 18 November 2013 and in revised form 3 January 2014 Available online xxxx Keywords: Biliproteins C-phycoerythrin Chromophore Protein denaturation Folding and stability
a b s t r a c t The conformational and functional state of biliproteins can be determined by optical properties of the covalently linked chromophores. a-Subunit of most of the phycoerythrin contains 164 residues. Recently determined crystal structure of the naturally truncated form of a-subunit of cyanobacterial phycoerythrin (Tr-aC-PE) lacks 31 N-terminal residues present in its full length form (FL-aC-PE). This provides an opportunity to investigate the structure–function relationship between these two natural forms. We measured guanidinium chloride (GdmCl)-induced denaturation curves of FL-aC-PE and Tr-aC-PE proteins, followed by observing changes in absorbance at 565 nm, fluorescence at 350 and 573 nm, and circular dichroism at 222 nm. The denaturation curve of each protein was analyzed for DGD , the value of Gibbs free energy change on denaturation (DGD) in the absence of GdmCl. The main conclusions of the this study are: (i) GdmCl-induced denaturation (native state M denatured state) of FL-aC-PE and Tr-aC-PE is reversible and follows a two-state mechanism, (ii) FL-aC-PE is 1.4 kcal mol1 more stable than Tr-aC-PE, (iii) truncation of 31-residue long fragment that contains two a-helices, does not alter the 3-D structure of the remaining protein polypeptide chain, protein–chromophore interaction, and (iv) amino acid sequence of Tr-aC-PE determines the functional structure of the phycoerythrin. Ó 2014 Published by Elsevier Inc.
27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42
43 44
Introduction
45
Cyanobacteria are the most primordial among oxygenic photosynthetic organisms, and their antenna system consists of characteristic phycobiliproteins (PBPs)1 which are categorized as phycoerythrin (PE), phycocyanin (PC) and allophycocyanin (APC) that form a supramolecular assembly called phycobilisomes (PBS) [1–3]. PBS is a light harvesting antennae of photosystem II in cyanobacteria, red algae and glaucophytes. These protein complexes are anchored to thylakoid membranes which are made of stacks of PBPs and their associated linker polypeptides [4]. Each PBS consists of a core made of APC from which several outwardly oriented rods made of stacked disks of PC and (if present) PE or phycoerythrocyanin [3,5]. The energy transfer in the organism proceeds successively in the direction PE ? PC ? APC ? chlorophyll a, with an overall efficiency of almost 100% [1,2,6]. The lights absorbed by PBP are in the regions of visible spectrum where chlorophyll a has lower efficiency. The spectral properties of PBPs are mainly governed by their
46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
⇑ Corresponding author. E-mail address: [email protected] (M.I. Hassan). Abbreviations used: GdmCl, guanidinium chloride; CD, circular dichroism techniques; FL-aC-PE, full length aC-PE; Tr-aC-PE, truncated aC-PE; a6b6, hexamer; a2b2, dimers; (a3b3)2, dimers of trimers; PE, phycoerythrocyanin; PBPs, phycobiliproteins; PE, phycoerythrin; PC, phycocyanin; APC, allophycocyanin; PBS, phycobilisomes; PEBs, phycoerythrobilins; PyMOL, Molecular Graphics System. 1
prosthetic groups which are linear tetrapyrroles known as phycobilin including phycocyanobilin, phycoerythrobilin, phycourobilin and phycoviolobilin. Interestingly, spectral properties of phycobilins are highly influenced by the protein environment [1,2,7]. PE is one of the unique and special accessory light harvesting pigment and exists in the form of hexamer (a6b6), dimers (a2b2) or dimers of trimers (a3b3)2 in light harvesting system of bluegreen algae, cryptophytes and red algae [1]. PE of blue-green algae (C-PE) contains 6a and 6b subunits arranged in ring-like assemblies where the 3 monomers are present in the threefold symmetry [8,9]. Each a-subunit of PE consists of 164 amino acid residues while each b-subunit is 171-residue long [1,9]. The structure of C-PE resembles that of the globin protein which contains two open-chain linear tetrapyrrole chromophores known as phycoerythrobilins (PEBs) covalently attached to each a-subunit at Cys82 and Cys139. While b-subunit of C-PE contains three chromophores where two are covalently attached to the Cys84 and Cys155, and third chromophore is doubly linked to Cys50 and Cys61 [1,8,9]. The interactions between PEB with protein atoms and protein–chromophore environments are critically responsible for the photon absorption properties of C-PE. Recently, we identified and determined the crystal structure of a-subunit of C-PE from Phormidium tenue, which is devoid of 31 residues from its amino terminal side [9] as compared to the full length aC-PE containing complete 164 residues [8]. It is interesting
0003-9861/$ - see front matter Ó 2014 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.abb.2014.01.005
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82
83 84 85
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 2 86 87 88 89
90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
to note that truncation of 31 N-terminal residues occurs naturally when the organism is grown or kept in prolonged starved condition. However, the truncated aC-PE maintains its light absorbing capabilities [8,9]. The full length aC-PE (FL-aC-PE) and truncated aC-PE (Tr-aC-PE) can be used as a model system to study protein folding problems in order to establish the role of 31 N-terminal residues. We, therefore, studied folding/unfolding of the full length and truncated aC-PEs using absorption, fluorescence and circular dichroism (CD) techniques. Apart from determination of role of 31 N-terminal residues in protein folding and stability, our findings may be helpful to understand the mechanism of survival of this cyanobacterium and other marine algae under stress conditions. As under these conditions including starvation the organism maintains its pigment proteins and absorption properties for food preparation and survival. Furthermore, our study may also provide an insight to grow higher photosynthetic plants under this and similar types of stress environmental conditions. This report on structure, folding and stability of FL-aC-PE and naturally truncated Tr-aC-PE will be helpful for better understanding of mechanism of energy transfer in photosynthetic cyanobacteria, even under stress conditions like starvation and salinity. Furthermore, we established the role of 31 N-terminal residues in the folding and stability of C-PE.
109
Materials and methods
110
Materials
111
Guanidinium chloride (GdmCl) of ultra pure grade was purchased from MP Biomedicals, LLC (Illkirch, France). Sodium cacodylate trihydrate and Sephadex G-150 matrix were obtained from Sigma–Aldrich (St Louis, USA) and GE Healthcare UK Limited, respectively. Other chemicals were purchased from local suppliers. All chemicals used were of molecular biology research grade.
112 113 114 115 116
117
Isolation and purification
118
Protein was isolated and purified from P. tenue sp. A27DM collected from rocky shores of Bet Dwarka, Okha and estuarine mouth of river Daman Ganga (India). The enrichment, provision of growth conditions, extraction and purification of C-PE were carried out using our well optimized protocol described elsewhere [10–12]. Briefly, the cyanobacterium was harvested in artificial sea water nutrients, and the cell mass was washed with 1.0 M Tris–HCl buffer (pH 8.1) containing 3.0 mM sodium azide. For the purification of FL-aC-PE and Tr-aC-PE, cultures were harvested when they were young (20 days) and old (90–200 days), respectively. The procedures for getting both types of proteins have already been published [9–11]. Washed cell mass was resuspended in the same buffer, and repeated freeze–thaw cycles (30 °C and 4 °C) were applied for C-PE release. This was followed by two-step (20% and 70%) ammonium sulfate precipitation. The precipitate was dissolved in 10 mM Tris buffer and subjected to gel (G-150) permeation chromatography column pre-equilibrated with 10 mM Tris buffer. The purity of the purified protein was checked by the native-PAGE and SDS–PAGE using silver staining [11,13,14]. The entire purification procedure was carried out in dark at 4 °C. All buffer solutions required above during isolation and purification were prepared in mili-Q water supplemented with 0.01% sodium azide. Stock solutions of both proteins were stored in the Tris– HCl buffer (pH 8.1) containing 0.01% sodium azide at 4 °C. To remove buffer constituents and sodium azide, stock solutions were dialyzed against several changes with 0.1 M KCl (pH 7.0). The concentrations of the dialyzed and filtered stock solutions
119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144
of FL-aC-PE and Tr-aC-PE were determined using a value of 265,000 M1 cm1 molar absorption coefficient at 565 nm [15].
145
Sample preparation for denaturation and renaturation experiments
147
For the study of GdmCl-induced denaturation, 8 M stock solution of the denaturant was prepared in 50 mM sodium cacodylate buffer (pH 7.0) containing 0.1 M KCl. Following the procedure reported earlier [16], protein solutions for denaturation and renaturation experiments were prepared. Briefly, for each denaturation experiment known amounts of the protein stock solution, buffer and concentrated denaturant solution were mixed and incubated for a time long enough to ensure the complete denaturation. This was cross-checked by taking absorbance. A similar procedure was employed in preparing the solution for renaturation experiments with the exception that the protein was first denatured by adding denaturant solution and then dilated with the buffer. The GdmCl-induced denaturation was studied at pH 7.0. We have always checked the pH of the protein solutions thus prepared.
148
Measurements of absorption spectra
162
Absorption spectra of FL-aC-PE and Tr-aC-PE were measured in Jasco UV/visible spectrophotometer (Jasco V-660, Model B 028661152) equipped with peltier-type temperature controller (ETCS-761). All measurements were carried out in the wavelength range 700–350 nm using 1 cm path length cuvettes. The protein concentration used was in the range of 0.1–0.3 mg ml1. All spectral measurements were done at least three times at 25 ± 0.1 °C.
163
Measurements of fluorescence spectra
170
Fluorescence measurements were carried out in Jasco spectrofluorimeter (Model FP-6200) using 5 mm quartz cuvette. All experiments were carried out at 25 ± 0.1 °C which is maintained by an external thermostated water circulator. FL-aC-PE and Tr-aC-PE have intrinsic (Trp77) and extrinsic (PEB) fluorophores. To measure Trp77 fluorescence the protein was excited at 287 nm and emission spectra was recorded in the range 300–400 nm. To measure PEB’s fluorescence, protein solution was excited at 479 nm, and emission spectra were recorded in the range 500–600 nm [17]. The protein concentration used was 0.1–0.2 mg ml1.
171
CD measurements
181
The far-UV and Soret CD measurements of FL-aC-PE and Tr-aC-PE were carried out in Jasco spectropolarimeter (J-715) equipped with peltier type of temperature controller (PTC348WI). CD spectra of FL-aC-PE and Tr-aC-PE were collected with a response time of 1 s and a scan speed of 100 nm min1. Each spectrum was an average of at least five scans. All measurements were carried out at 25 ± 0.1 °C. The far-UV spectra were collected in the wavelength range 250–200 nm using 0.1 cm path length cuvette. The protein concentration used was 0.2–0.4 mg ml1. Soret CD spectra measurements of FL-aC-PE and Tr-aC-PE were carried out in the wavelength range 600–400 nm using protein concentration of 0.3–0.5 mg ml1 in a cuvette of a path length of 1.0 cm. The raw data at a given wavelength, k (nm) were converted into concentration-independent parameter [h]k (deg cm2 dmol1), the mean residue ellipticity at wavelength k using the relation:
182
½hk ¼ M 0 hk =10lc
ð1Þ
where hk is the observed ellipticity in millidegrees at wavelength k, M0 is the mean residue weight of the protein, l is the path length of the cell in centimeters, and c is the protein concentration in mg ml1.
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
146
149 150 151 152 153 154 155 156 157 158 159 160 161
164 165 166 167 168 169
172 173 174 175 176 177 178 179 180
183 184 185 186 187 188 189 190 191 192 193 194 195 196
197 199 200 201 202 203
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx 204
Analysis of data
205
Sigmoidal GdmCl-induced denaturation curve (plot of a physical property (y) versus [GdmCl], the molar concentration of the denaturant) were analyzed for stability parameters (DGD , the value of DGD (Gibbs free energy change) in the absence of the GdmCl; m, the slope (=oDGD/o[GdmCl]); and Cm, the midpoint of the denaturation curve, i.e., [GdmCl] at which DGD = 0. This analysis involves a least-squares method to fit the entire denaturation curve, obtained at constant pH and temperature T (K), according to the relation:
206 207 208 209 210 211 212
213
y ¼ yN þ yD Exp½ðDGD m½GdmClÞ=RT=1 þ Exp½ðDGD 215 216 217 218 219 220 221 222 223 224 225
226
m½GdmClÞ=RT
ð2Þ
where yN and yD are the properties of the native (N) and denatured (D) protein molecules under the same experimental condition in which y has been measured, R is gas constant, and T is temperature (K). It should be noted that Eq. (2) assumes that a plot of DGD versus [GdmCl] is linear, and the dependencies of yN and yD on [GdmCl] are also linear (i.e., yN = aN+bN [GdmCl], yD = aD+bD [GdmCl], where aN, bN, aD and bD are [GdmCl]-independent parameters, and subscripts N and D signify that parameters are for the native and denatured states, respectively). The denatured fraction of the protein, fD at a given [GdmCl] was determined using the relation:
fD ¼ ðy yN Þ=ðyD yN Þ 228
¼ y ðaN þ bN ½GdmClÞ=ðaD aN Þ þ ðbD bN Þ½GdmCl
ð3Þ
229
In silico structural analysis
230
241
The crystal structure of Tr-aC-PE from P. tenue is already (PDB code 3MWN). However, the crystal structure of FL-aC-PE is still not determined. Hence, we modeled the structure of FL-aC-PE using modeler [18]. The energy minimized coordinate [19] was validated using online Structural Analysis and Verification Server (http://nihserver.mbi.ucla.edu/SAVES/), showing most of the residues in allowed region of Ramachandran plot (Fig. S1). A superimposed ribbon diagrams of full length and truncated aC-PE were drawn using PyMOL (The PyMOL Molecular Graphics System, Version 1.2r3pre, Schrödinger, LLC) [20]. Contacts offered by various residues were measured by using the contact program from the CCP4 suite [21].
242
Results
243
Polyacrylamide gel electrophoresis
244
247
The purified FL-aC-PE and Tr-aC-PE showed single band on both the native PAGE and SDS–PAGE (results not shown) [10]. This observation led us to conclude that both full length and truncated proteins exist as monomer.
248
Spectral characterization
249
Insets in Fig. 1 show visible absorbance spectra of the native FL-aC-PE and Tr-aC-PE at pH 7.0 (50 mM sodium cacodylate containing 0.1 M KCl) and 25 ± 0.1 °C. It has been argued that this absorption of both proteins is due to the interaction between the folded polypeptide backbone and the two thioether bonded PEBs [10,11]. This argument was based on the fact that the apoprotein does not absorb in the visible region, and PEB has a very low absorbance [1,2,22]. These findings suggest that the folding/unfolding reactions of FL-aC-PE and Tr-aC-PE can be followed by observing changes in the visible absorption at 565 nm. To see the effect of GdmCl on the structural stability of FL-aC-PE and Tr-aC-PE, we
231 232 233 234 235 236 237 238 239 240
245 246
250 251 252 253 254 255 256 257 258 259
3
measured the absorption spectra of both proteins in the presence of different concentrations of GdmCl in the range 700–350 nm. Results of these measurements are shown in Fig. S2 (also see insets in Fig. 1). Fig. 1A and B respectively show plots of De565 (difference in molar absorption coefficient at 565 nm in the presence of GdmCl and that in the absence of GdmCl) of FL-aC-PE and Tr-aC-PE as a function of [GdmCl]. These plots were analyzed for DGD m and Cm values for each protein according to Eq. (2). Results of this analysis are shown in Table 1. Both the FL- and Tr-aC-PE proteins have a lone tryptophan at position 77 (numbering is according to the full length protein). This intrinsic fluorophore is buried and is at a distance of 17 Å from the second chromophore (PEB2) attached at Cys139 [9]. After excitation at 287 nm, we measured tryptophan emission spectra of both proteins in the native (in 50 mM cacodylate buffer) and denatured (GdmCl solution) states at pH 7.0 and 25 ± 0.1 °C (see insets in Fig. 2). To make corrections for the effect of solvent (GdmCl solution) on the D state of the protein, following procedure was used. Values of fluorescent intensity at different wavelengths were determined from the observed emission spectra of a protein in the presence of [GdmCl] > 2.5 M i.e., the post transition region (Figs. S3 and S4). At a given wavelength, values of fluorescent intensity were plotted against [GdmCl]. To obtain the solvent independent value, an extrapolated value at [GdmCl] = 0 M was determined from this plot. These extrapolated values of fluorescent intensity values at different wavelengths were used to obtain the fluorescence spectrum of the denatured state which does not depend on the denaturant concentration. It has been shown that kmax of tryptophan fluorescence spectrum shifts to longer wavelengths and the intensity at kmax decreases on transferring this fluorophore from non-polar medium to polar medium [23,24]. As can be seen in Fig. 2, when a protein is transferred from the native buffer to the denaturant solution, there is an expected red shift of the emission spectrum from 350 to 360 nm. However, contrary to this expectation, the emission intensity of the proteins is increased on denaturation. This increase can be explained by invoking quenching phenomenon due to the presence of PEB2 near Trp77, which will decrease when the protein unfolds [22]. We exploited this phenomenon to see the effect of GdmCl on emission spectra of FL- and Tr-aC-PEs. These spectra are shown in Fig. S3. Values of emission intensity at 350 nm (F350) of both proteins are read from these spectra and plotted as a function of [GdmCl] in Fig. 2A and B. To obtain stability parameters (DGD , m and Cm) each denaturation curve shown in Fig. 2 was fitted to Eq. (2). Results of this analysis for both proteins are shown in Table 1. PEB has fluorescence property [25–27]. It has been shown that the covalently linked PEBs bonded to the folded protein polypeptide chain via thioether bonds have a strong emission spectrum in the wavelength region 500–600 nm, which vanishes on protein denaturation [26,28]. After exciting FL-aC-PE and Tr-aC-PE at 479 nm, we measured emission spectra of both proteins in the presence of different concentrations of GdmCl at pH 7.0 and 25 ± 0.1 °C. These spectra are shown in Fig. S4. Fig. 3A and B show the change in F573, the fluorescence intensity of FL-aC-PE and Tr-aC-PE at 573 nm as a function of [GdmCl], respectively. For each protein, GdmCl-induced denaturation curves were analyzed for DGD , m and Cm values by fitting the entire denaturation curve according to Eq. (2). The results of this analysis are shown in Table 1. To see the effect of GdmCl on the stability of FL-aC-PE and Tr-aC-PE, we measured the far-UV CD spectra of both proteins in the presence of different concentrations of GdmCl. These measurements are shown in Fig. S5. Values of [h]222 (probe for the secondary structure) in the presence of different concentrations of GdmCl were determined from these CD spectra shown in this figure. Fig. 4A and B show plots of [h]222 versus [GdmCl] for FL-aC-PE and Tr-aC-PE,
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 4
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
1
A
30
Native FL-αC-PE
ελ x 10-4, M-1 cm-1
25
20
0
-3 -1 -1 Δε565 x 10 , M cm
15
10
5
-1
Denatured FL-αC-PE 0 350
400
450
500
550
600
650
700
Wavelength, nm
-2
-3
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M 1
B
30
Native Tr-αC-PE
ελ x 10-4, M-1cm-1
25
-3 -1 -1 Δε565 x 10 , M cm
0
20
15
10
5
-1
Denatured Tr-αC-PE 0 350
400
450
500
550
600
650
700
Wavelength, nm
-2
-3
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M Fig. 1. GdmCl-induced denaturation of FL-aC-PE and Tr-aC-PE at pH 7.0 and 25 °C. (A) Denaturation curve of FL-aC-PE followed by observing changes in De565 as a function of [GdmCl]. (B) Plots of De565 of Tr-aC-PE versus [GdmCl]. Open and filled symbols show the results obtained from denaturation and renaturation experiments. The insets in (A) and (B) show the absorption spectra of FL-aC-PE and Tr- aC-PE of the native and GdmCl-induced denatured states, respectively.
Table 1 Stability parameters of FL-aC-PE and Tr-aC-PE associated with the GdmCl-induced denaturation at pH 7.0 and 25 ± 0.1 °C.a Probes (Fit parameter)
De565 F350 F573 [h]222 a
DGD (kcal mol1 M1)
m (kcal mol1 M1)
Cm (M)
FL-aC-PE
Tr-aC-PE
FL-aC-PE
Tr-aC-PE
FL-aC-PE
Tr-aC-PE
11.45 ± 0.34 11.77 ± 0.38 11.56 ± 0.23 11.82 ± 0.21
10.11 ± 0.40 10.42 ± 0.66 10.16 ± 0.27 10.47 ± 0.56
5.67 ± 0.44 6.13 ± 0.35 5.99 ± 0.56 6.16 ± 0.49
5.77 ± 0.36 6.35 ± 0.22 5.83 ± 0.39 6.41 ± 0.57
2.02 ± 0.43 1.92 ± 0.24 1.93 ± 0.35 2.03 ± 0.27
1.75 ± 0.29 1.68 ± 0.22 1.74 ± 0.34 1.77 ± 0.30
A ± with a value is a mean of errors from the average of three independent measurements.
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
5
Fig. 2. GdmCl-induced denaturation of FL-aC-PE and Tr-aC-PE followed by observing changes in F350 at pH 7.0 and 25 °C. (A) and (B) show denaturation curves of FL-aC-PE and Tr-aC-PE, respectively. Open and filled symbols have same meaning as in Fig. 1. Fluorescence spectra of FL-aC-PE (inset in (A)) and Tr-aC-PE (inset in (B)) in their native and denatured states, respectively. The excitation wavelength was 287 nm. D state of each protein was corrected for the solvent effect.
326 327 328 329 330
331 332
respectively. Each of these plots (denaturation curves) was analyzed for values of DGD , m and Cm by fitting the entire denaturation curve according to Eq. (2). These stability parameters thus obtained are given in Table 1. The insets in Fig. 4 show the far-UV CD spectra of FL-aC-PE and TraC-PE. We have used k2d program (http://dichroweb.cryst.bbk.ac.uk/ html/process.shtml) to estimate the percentage of elements of sec-
ondary structure of those proteins. Analysis of the far-UV CD spectrum of the native FL-aC-PE (see inset in Fig. 4A) for the contents of a-helix and b-sheet yielded values of 75% and 2%, respectively. Analysis of the far-UV CD spectrum of Tr-aC-PE (see inset in Fig. 4B) for the secondary structure gives 60% and 2% for the a-helix and b-structure, respectively, which is in excellent agreement with that determined by the X-ray diffraction studies [9].
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
333 334
335 336 337 338 339
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 6
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
60
A
F573 (a. u.)
Fluorescence (a. u.)
60
Native FL-αC-PE 40
Denatured FL-αC-PE
20
40 0 520
540
560 Wavelength, nm
580
600
20
0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M 60
B Fluorescence (a. u.)
Native Tr-αC-PE
F573 (a. u.)
60
40
Denatured Tr-αC-PE
20
40 0 520
540
560 Wavelength, nm
580
600
20
0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M Fig. 3. Denaturation curves of FL-aC-PE and Tr-aC-PE induced by GdmCl at pH 7.0 and 25 °C. (A) Plot of F573 of FL-aC-PE versus [GdmCl]. (B) Plot of F573 of Tr-aC-PE versus [GdmCl]. Open and filled symbols have the same meaning as in Fig. 1. Insets in (A) and (B) show the fluorescence spectra of the native and denatured FL-aC-PE and Tr-aC-PE, respectively. Spectra of the denatured proteins were corrected for the solvent effect. For both proteins excitation wavelength was 479 nm.
340
Refolding experiments
341
In order to check the reversibility of denaturation of FL-aCPE and Tr-aC-PE, proteins were exposed to concentrated GdmCl solution ([GdmCl] P 2.5 M) for a time long enough to ensure complete denaturation. The denatured protein was diluted with the native buffer to check the reversibility of denaturation at 3.0, 2.5, 2.0 and 0.5 M GdmCl. Fig. 5A–H show representative denaturation (curve 1) and renaturation (curve 2) spectra. It has been observed that for each protein in a given denaturant
342 343 344 345 346 347 348
concentration, spectra obtained from denaturation and renaturation experiments are indistinguishable (also see Figs. 1–4).
349
In silico structure analysis
351
A structural comparison of FL-aC-PE and Tr-aC-PE suggests a difference of 31 residues from N-terminal side, which is comprised of two a-helices. This segment is completely away from the whole compact polypeptide chain (Fig. 6). It is seen in Fig. 6 that there are only 6 hydrogen bonds between residues
352
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
350
353 354 355 356
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 7
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx 5
A
[θ ]222 x 10-3, deg cm2 dmol-1
0
-5
[θ ]λ x 103, deg cm2 dmol-1
10
5
Denatured FL-α C-PE 0
-5
-10
-15
Native FL-αC-PE
-20
-25 200
205
210
215
220
225
230
235
240
245
250
Wavelength, nm -10
-15
-20
-25 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M 5
B
2 -1 [θ ] x 10-3, deg cm dmol 222
0
-5
[θ ]λ x 103, deg cm2 dmol-1
10
5
Denatured Tr-α C-PE
0
-5
-10
-15
Native Tr-α C-PE -20
-25 200
205
210
215
220
225
230
235
240
245
250
Wavelength, nm
-10
-15
-20
-25 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M Fig. 4. GdmCl-induced denaturation curves of FL-aC-PE and Tr-aC-PE at pH 7.0 and 25 °C. For the full length and truncated proteins (A) and (B) show the plots of [h]222 versus [GdmCl], respectively. Open and filled symbols have the same meaning as in Fig. 1. The insets in (A) and (B) show the far-UV CD spectra of the native and denatured states of FL-aC-PE and Tr-aC-PE, respectively.
357 358 359 360 361 362 363 364 365 366
Lys2 and Gly105, Gly29 and Arg33, Gln28 and Gln32, Asn30 and Gln32, Gln28 and Gln32 (through side chain) and Ile31 and Gln32 (see Table 2). However, for the sake of clarity the first three hydrogen bonds are shown in Fig. 6. Furthermore, this N-terminal segment is involved in 25 van der Waals interactions (see Table 2). Since these 31 residues of FL-aC-PE are completely away from the packed polypeptide core, these are expected to be highly exposed to water, if the protein exists in the form of a monomer. The crystal structure of Tr-aC-PE was recently determined [9]. It was observed that this protein belongs to a class of all
a-protein. Likewise, our modeled structure of FL-aC-PE shows the presence of 75% a-helical content. Fig. 6 shows overlapping 3-D structures of Tr-aC-PE and FL-aC-PE. The 31-residue long N-terminal sequence of FL-aC-PE contains 12% a-helix.
367
Discussion
371
The analysis of transition curves (Figs. 1–4) for the stability parameters, DGD , m and Cm (Table 1) according to Eq. (2) is based
372
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
368 369 370
373
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 8
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx 30
A
3 2
25
x 10-4 , M-1 cm-1
15 10
2
20 15 10 5
5
1
1 0
0 350
400
450
500
550
600
650
350
700
400
450
500
550
600
650
700
Wavelength, nm
Wavelength, nm 15
15
C
D 1 Fluorescence (a. u.)
1
Fluorescence (a. u.)
B
3 25
20
x 10-4 , M-1 cm-1
30
10
2 5
10
2 5
3 3 0 300
320
340
360
380
0 300
400
320
340
360
380
400
Wavelength, nm
Wavelength, nm 60
E
3
60
F
40
Fluorescence (a. u.)
Fluorescence (a. u.)
3
2
20
2
40
20
1
0 520
540
1
560
580
0 520
600
540
560
10
H
5
[ ]λ x 10-3 , deg cm2 dmol-1
[ ]λ x 10-3 , deg cm2 dmol-1
600
10
G
5 0 -5
1
-10 -15
2
-20 -25 200
580
Wavelength, nm
Wavelength, nm
210
215
220
225
-5
1
-10
2 -15 -20
3
-25
3 205
0
230
Wavelength, nm
235
240
245
250
200
205
210
215
220
225
230
235
240
245
250
Wavelength, nm
Fig. 5. Renaturation of the GdmCl-denatured FL-aC-PE and Tr-aC-PE at pH 7.0 and 25 °C. Absorption spectra of (A) FL-aC-PE and (B) Tr-aC-PE: (1) denatured protein in 2.5 M GdmCl, (2) renatured protein in 0.5 M GdmCl, and (3) native protein. (C) and (D) show Trp fluorescence spectra, (E) and (F) show PEB fluorescence spectra, and (G) and (H) the far-UV CD spectra of FL-aC-PE and Tr-aC-PE, respectively. Spectra meaning has the same meaning as in (A).
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
9
Gly105
Lys2
Asn30
Gln32
Gly29 Ile31
Gln28
Arg33
Fig. 6. Overlaid structure of Tr-aC-PE (light green) and FL-aC-PE illustrated in cartoon model. The chromophores, PEBs are shown in ball and stick model. 31 N-terminal residues forming hydrogen bonded interactions with other part (32–164) of molecule are shown in ball and stick. Structure is drawn in PyMol using the atomic coordinates of FL-aC-PE and aC-PE from the Phormidium tenue (PDB code: 3MWN). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395
on a few assumptions. A few comments are therefore necessary. First, it was assumed that the GdmCl-induced denaturation of the full length and truncated aC-PE proteins are reversible. To check the reversibility of denaturation, we denatured each protein by adding different concentrations of the denaturant. The GdmCl-denatured protein was diluted with the native buffer. After incubation of the denatured protein in the buffer for a time enough for the completion of the renaturation process, optical spectra (absorption, fluorescence and CD) were recorded. A spectrum of the renatured protein was then compared with that obtained during denaturation experiments. It was observed that the spectral properties of renatured protein in a given [GdmCl] are identical to those obtained during denaturation experiments. Fig. 5 shows the representative data which were obtained on denaturing the protein by 2.5 M GdmCl (curve 1), and renaturing this denatured protein by diluting with the buffer to bring the denaturant concentration to 0.5 M (curve 2). It has been observed that the spectra of the protein in a given [GdmCl] obtained from the denaturation and renaturation experiments were indistinguishable. The spectra of the denatured and renatured proteins are also compared with those of the native proteins (curve 3; Fig. 5). The coincidence of optical properties of the proteins obtained from denaturation and
renaturation experiments (Fig. 5 and also see Figs. 1–4) led us to conclude that the GdmCl-induced denaturation of FL-aC-PE and Tr-aC-PE is reversible. The second assumption was that the GdmCl-induced transition between N and D states is a two-state process. The most authentic criterion for a two-state (native and denatured) behavior of the protein denaturation is to show that the calorimetric and van’t Hoff enthalpy changes on denaturation are identical. We could not succeed to make differential scanning calorimetry measurements on Fl-aC-PE and Tr-aC-PE proteins due to the fact that the heat denatured proteins formed aggregates. However, another test for the two-state behavior of the protein denaturation is to show coincidence of the normalized sigmoidal curves of different physical property induced by GdmCl [29]. Fig. 7A and B show normalized denaturation curves (plots of fD versus [GdmCl]) of the full length and truncated proteins, respectively. It is seen in these figures that for a protein, the normalization of each transition curve of Figs. 1– 4, using Eq. (3) yields a coincidence of curves of different physical properties. As pointed out earlier [29], this is a necessary but not sufficient criterion for a two-state behavior of the protein denaturation. Customarily, the observation of non-coincidence of normalized transition curves of different physical properties constitutes a
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 10
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
Table 2 Hydrogen bonds and van der Waals Interactions that the N-terminal segment (1–31 residues) forms with atoms of the remaining polypeptide chain (32–164) of aR-PE.a Residues involved in non-covalent interactions
Bond length
1MetCa 1MetSd 1MetCe
3.30 3.43 3.47 3.42 3.48 3.27 3.12 3.18 3.48 2.84 3.31 3.18 3.10 3.22 3.08 3.41 2.81 3.42 3.23 3.05 2.86 2.89 2.88 3.00 3.23 3.29 2.62 3.43 2.79 3.15 3.38
2LysN 2LysO
28GlnO
28GlnCd 28GlnNe2 29GlyO 30Asn N 30AsnC 30AsnO c
30AsnC 30AsnOd1 30AsnNd2 31IleN 31IleCa 31IleC 31IleO
a
418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449
109GluOe2 106ProCd 109GluCd 109GluOe1 105GlyCa 105GlyN 103GlyCa 103GlyC 103GlyO 32GlnN 32GlnCc 32GlnNe2 32GlnNe2 32GlnN 33ArgN 32GlnN 34AlaN 34AlaCb 33ArgN 37ArgNg2 37ArgNg2 37ArgNg2 32GlnN 32GlnC 33ArgN 34AlaN 32GlnCa 32GlnO 32GlnC 33ArgN 32GlnN
Residues forming hydrogen bonds are shown in bold letters.
proof of existence of stable intermediate(s) on the folding M unfolding pathways [29]. On the contrary, basing their arguments on theoretical grounds, Dill and Shortle [30] have shown that the non-coincidence of sigmoidal curves of two different physical properties is also observed for a two-state protein denaturation. Recently, Rahaman et al. [31] provided experimental evidence that this non-coincidence is consistent with two-state denaturation behavior. In another approach for testing the two-state behavior of GdmCl-induced denaturation, DGD values of the protein are measured from sigmoidal curves of different structural proteins, and it is shown that this thermodynamic property is independent of techniques (probes) used to monitor the denaturation [32]. Table 1 shows that values of DGD of each protein, measured by four different structural properties are, within experimental errors, identical. This overlap in DGD values suggest that GdmCl-induced denaturation of Fl-aC-PE and Tr-aC-PE is a two-state process. It is known that if absorption spectra of two non-interacting species separately intersect exactly at one or more wavelengths, then each intersecting point (isosbestic point) of absorption no longer depends on the concentration ratios of these species in their mixtures. The presence of isosbestic points on the absorption spectra of the native and denatured proteins alone and in their equilibrium mixtures is therefore taken as an evidence for a two-state denaturation [22,29]. Fig. S2 shows absorption spectra of the FL- and Tr- aC-PEs in the presence of different concentration of GdmCl in which a protein exists either in the N state (pre-transition region) or as an equilibrium mixture of N and D states (transition region) or in the D state (post-transition region). It is seen in this figure that there exist isosbestic points at wavelengths 460 and 594 nm in the case of FL-aC-PE and at wavelengths 440 and 602 nm in case of Tr-aC-PE. These observations
therefore support an assumption that GdmCl-induced denaturation of both proteins follows a two-state mechanism. Estimates of DGD from the denaturation curves depend not only on the mechanism of denaturation but also on the models describing the dependencies of the observed DGD on [GdmCl]. There are three models to estimate DGD from the extrapolation of DGD to 0 [M] GdmCl concentrations [33,34]. One model assumes that DGD value varies linearly with [denaturant]. The other one (binding model) assumes the presence of ‘specific’ binding sites for the chemical denaturant on the native and denatured protein molecules. In this model dependence of DGD on the [denaturant] is non-linear. The third model known as transfer-free energy model, describes the dependence of DGD on the transfer free energy of small model compounds for the protein groups from water to different concentrations of the chemical denaturants. It has been observed that the analysis of the same set of denaturation data using different models give different values of DGD of a protein. However, we used the linear free energy model for the analysis of denaturation curves of both FL-aC-PE and Tr-aC-PE (Figs. 1–4), for it has supports of both theory [35,36] and experiments [37–41]. It is seen in Table 1 that the full length protein is only 1.4 kcal mol1 more stable than the truncated protein, or in other words, the contribution of 31 N-terminal residues to protein stability (DGD ) of FL-aC-PE is equivalent to the stability provided by only one buried hydrogen bond (1.5 ± 1 kcal mol1) [42]. To understand this difference in stability between FL-aC-PE and Tr-aC-PE, we performed in silico study. Since the sequence of FL-aC-PE is not determined so far we have taken a conserved sequence of 31 N-terminal residues from cyanobacteria. We assumed that our model is quite similar to the crystal structure of FL-aC-PE. This assumption is supported by following observations. (i) We aligned the sequences of all a-PEs of cyanobacteria and red algae available in the UniProt database using multiple sequence alignment protcol on ClustalW2 [43]. Out of 31 N-terminal residues, only 5 positions are not fully conserved; 5A/I/A, 8V/A, 9I/V, 10A/S/G/T and 21 T/A/Q/S (see Fig. 8). This comparison indicates that there is 97% sequence homology as far as 31 N-terminal residues are concerned. (ii) A sequence alignment of Tr-aC-PE with that of 32–164 residues shows 94% homology (Fig. 8). (iii) An overlaying of 3-D structure of Tr-aCPE on that of FL-aC-PE shows almost identical structural homology (Fig. 6). (iv) Analysis of the far-UV CD spectra for the elements of secondary structures in FL-aC-PE and Tr-aC-PE (Fig. 4A and B), and comparison of these elements of secondary structure with those of Tr-aC-PE obtained from the X-ray diffraction studies (Fig. 6) led to the conclusion that 31 N-terminal residues of the full length intact protein contain 12% a-helix. This observation is identical to that observed for the same segment in the FL-aC-PE (see Fig. 6). Table 1 shows m-values of FL-aC-PE and Tr-aC-PE. The average values of this parameter (m) obtained from measurements of GdmCl-induced denaturation using different optical properties are 5.99 kcal mol1 M1 for the full length protein and 5.65 kcal mol1 M1 for the truncated protein. Myer et al. [44] have developed a method to predict m-value of a protein from its known change in the accessible surface area (DASA) on denaturation. Using their equation that relates m-value with DASA (i.e., m = 953 + 0.23 DASA), we estimated m-values of 5.65 and 5.71 kcal mol1 M1 for FL-aC-PE and Tr-aC-PE, respectively. This predicted value of m for each protein is in excellent agreement with the average value determined experimentally (also see Table 1). It should be noted that for the determination of total DASA of the protein, we have used DSSP method [45] for the estimation of ASA of the folded protein and the procedure of Creamer et al. [46] for estimation of the lower bound value of ASA of the unfolded protein. The a-subunit of each PBP contains 164 residues and consists of 9-helices packed in globular shape to form main core of the
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 11
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
1.2
A
1.0
0.8
fD
0.6
0.4
0.2
F350 F573
0.0
Δε565 [θ]222
-0.2 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M 1.2
B
1.0
fD
0.8
0.6
0.4
F350 0.2
F573 Δε565 [θ]222
0.0 0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
[GdmCl], M Fig. 7. Normalized GdmCl-induced denaturation curves of (A) FL-aC-PE and (B) Tr-aC-PE at pH 7.0 and 25 °C. Results shown in Figs. 1–4 were used to construct plots of fD versus [GdmCl]. 516 517 518 519 520 521 522 523 524 525 526 527
protein molecule [9]. The N-terminal segment (1–31 residues) contains only 2-helices X and Y which are antiparallel to each other and perturbed away from the main core (32–164 residues). Structural analysis suggests that residues of the N-terminal segment form a total of 30 hydrogen bonds and many van der Waals interactions among themselves (see Table 3). In addition to these non-covalent interactions this segment forms six inter-segment hydrogen bonds: Lys2 and Gly105, Gly29 and Arg33, Gln28 and Gln32, Asn30 and Gln32, Gln28 and Gln32 (through side chain) and Ile32 and Gln32) and 25 van der Waals interactions (see Table 2). It seems that among these six hydrogen bonds, only one hydrogen bond, 2LysOA105GlyN may have a significant role hold-
ing the N-terminal segment to the main core, for other hydrogen bonds are formed between neighboring atoms. Both types of non-covalent interactions (hydrogen bonding and van der Waals interactions) are known to provide stability to folded proteins [47–49]. It is then expected that the full length protein (FL-aC-PE) should be more stable than truncated protein (Tr-aCPE) which lacks 31-residues long N-terminal segment of the full length protein. This is indeed true; FL-aC-PE is 1.4 kcal mol1 more stable than the Tr-aC-PE (see Table 1). A buried hydrogen bond in the protein core (dielectric constant 4–6) provides a stability of about 2 kcal mol1 [42]. Thus an observed stability difference of 1.4 kcal mol1 between the full length
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
528 529 530 531 532 533 534 535 536 537 538 539
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 12
K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx
Fig. 8. Multiple sequence alignment of a-subunit of PBPs. Names of sequences is given according to the sequence identifier on the UniProt (www.uniprot.org). Non-conserved residues from N-terminal side (1–31) are shaded in yellow. Secondary structures of PBPs are illustrated on the top of sequence. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
YABBI 6602
No. of Pages 13, Model 5G
16 January 2014 K. Anwer et al. / Archives of Biochemistry and Biophysics xxx (2014) xxx–xxx Table 3 Residues of the N-terminal segment (1–31 residues) involved in the formation of hydrogen bonds.
540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556
Serial number
Residues involved in the hydrogen bond formation
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
6ThrN 7ThrN 7ThrOc1 8ThrN 8ThrOc1 8ThrOc1 9IleN 10SerN 11AlaN 11AlaN 12AlaN 12AlaN 13GluN 14AlaN 15AlaN 15AlaO 16GlyN 16GlyN 17ArgN 17ArgN 20SerN 23GluN 23GluN 24LeuN 24LeaN 25GlnN 25GlnO 26SerN 29GlyN 30AsnN
3SerOc1 3SerO 3SerO 4ValO 4ValO 5IleO 5IleO 6ThrO 8ThrO 9IleO 8ThrO 9IleO 9IleO 10SerO 12AlaO 13GluO 12AlaO 13GluO 12AlaO 15AlaO 23AspOd2 20SerO 21ThrO 20SerO 21hrO 21ThrO 21ThrOc1 22SerO 26SerO 27ValO
and truncated proteins appears to be too small, for 31-residue long N-terminal is involved in six hydrogen bonds formation [20]. A possible reason for this small difference in stability could be due to full or partial exposure of amino acid residues involved in non-covalent interactions, to the highly polar solvent water (dielectric constant 80). The reason for saying this is that both hydrogen bonds and van der Waals interactions are electrostatic in nature and inversely proportional to the dielectric constant of the medium. For instance, hydrogen bonds in water do not have a net stabilizing effect on protein structure due to the fact that in order to form a DAH A (where A and D are N and O atoms) bond one needs to disrupt another set of very favorable hydrogen bonds, those the protein groups (D-A and A) can make with water. Using software RVP-Net (version 1), we estimated exposure of all residues involved in non-covalent interactions. Interestingly, it has been observed that almost all groups involved in such interactions are fully or partially exposed.
557
Conclusions
558
(a) The 31-residue long N-terminal segment of the full length phycoerythrin (i) does not contribute significantly to the overall stability of the protein, for the full length protein is only 1.4 kcal mol1 more stable than the truncated protein, (ii) has 12% a-helix, and (iii) has no effect on the mechanism of folding of the protein fragment (32–164 residues), for both proteins undergo a two-state transition between N and D states. (b) The non-covalent interactions due to 31 N-terminal residues in the FL-aC-PE monomeric protein provide structural rigidity, not significant stability. (c) Deletion of the N-terminal segment does not perturb the protein–PEBs interaction, thus both FL-aC-PE and Tr-aC-PE have same absorbance capability.
559 560 561 562 563 564 565 566 567 568 569
13
Acknowledgements
570
M.I.H. and F.A. are thankful to the Department of Science and Technology (DST) for financial support (Project No. SR/SO/BB0124/2010A). K.A. is thankful to University Grant Commission (UGC) for providing Maulana Azad National Fellowship (MANF).
571
Appendix A. Supplementary data
575
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.abb. 2014.01.005.
576
References
578
[1] [2] [3] [4] [5] [6] [7] [8] [9]
[10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [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]
R. MacColl, J. Struct. Biol. 124 (1998) 311–334. R. MacColl, Biochim. Biophys. Acta 1657 (2004) 73–81. B.H. Gray, C.A. Lipschultz, E. Gantt, J. Bacteriol. 116 (1973) 471–478. J. Cornejo, S.I. Beale, M.J. Terry, J.C. Lagarias, J. Biol. Chem. 267 (1992) 14790– 14798. A.N. Glazer, Rev. Microbiol. 36 (1982) 173–198. M. Yokono, S. Akimoto, K. Koyama, T. Tsuchiya, M. Mimuro, Biochim. Biophys. Acta 1777 (2008) 55–65. R. MacColl, Polar Biol. 22 (1999) 384–388. W.R. Chang, T. Jiang, Z.L. Wan, J.P. Zhang, Z.X. Yang, D.C. Liang, J. Mol. Biol. 262 (1996) 721–731. B.R. Soni, M.I. Hasan, A. Parmar, A.S. Ethayathulla, R.P. Kumar, N.K. Singh, M. Sinha, P. Kaur, S. Yadav, S. Sharma, D. Madamwar, T.P. Singh, J. Struct. Biol. 171 (2010) 247–255. A. Parmar, Process Biochem. 46 (2011) 1793–1799. A. Parmar, N.K. Singh, A. Kaushal, S. Sonawala, D. Madamwar, Bioresour. Technol. 102 (2011) 1795–1802. G.N. Shah, V. Madamwar, Acta Bot. Mal. 26 (2001) 188–193. A. Parmar, J. Phycol. 46 (2010) 285–289. N.K. Singh, A. Parmar, D. Madamwar, Bioresour. Technol. 100 (2009) 1663– 1669. C.N. Pace, Protein Sci. 4 (1995). F. Ahmad, C.C. Bigelow, J. Biol. Chem. 257 (1982) 12935–12938. C.A. Royer, Chem. Rev. 106 (2006) 1769–1784. N. Eswar, D. Eramian, B. Webb, M.Y. Shen, A. Sali, Methods Mol. Biol. 426 (2008) 145–159. K. Joo, J. Lee, J.H. Seo, K. Lee, B.G. Kim, Proteins 75 (2009) 1010–1023. W.L. DeLano, Delano Scient, San Carlos, 2004, pp. 1–62. P. Emsley, K. Cowtan, Acta Crystallogr. D: Biol. Crystallogr. 60 (2004) 2126– 2132. K. Anwer, R. Sonani, D. Madamwar, P. Singh, F. Khan, K. Bisetty, F. Ahmad, M.I. Hassan, J. Biomol. Struct. Dyn. 31 (2013) 1–13. A. White, Biochemistry 71 (1958) 217–220. J.T. Vivian, P.R. Callis, Biophys. J. 80 (2001) 2093–2109. C.S. French, V.K. Young, J. Gen. Physiol. 35 (1952) 873–890. L.N. Liu, H.N. Su, S.G. Yan, S.M. Shao, B.B. Xie, X.L. Chen, X.Y. Zhang, B.C. Zhou, Y.Z. Zhang, Biochim. Biophys. Acta 1787 (2009) 939–946. T. Dammeyer, N. Frankenberg-Dinkel, J. Biol. Chem. 281 (2006) 27081–27089. M. Kupka, H. Scheer, Biochim. Biophys. Acta 1777 (2008) 94–103. N.C. Pace, C. Tanford, Biochemistry 7 (1968) 198–208. K.A. Dill, D. Shortle, Annu. Rev. Biochem. 60 (1991) 795–825. H. Rahaman, M.K. Alam Khan, M.I. Hassan, A. Islam, A.A. Moosavi-Movahedi, F. Ahmad, J. Chem. Thermodyn. 58 (2013) 351–358. D.W. Bolen, Methods 34 (2004) 312–322. C. Tanford, K.C. Aune, Biochemistry 9 (1970) 206–211. K.C. Aune, C. Tanford, Biochemistry 8 (1969) 4579–4585. J.A. Schellman, Biopolymers 17 (1978) 1305–1322. J.A. Schellman, Biophys. Chem. 96 (2002) 91–101. F. Ahmad, C.C. Bigelow, Can. J. Biochem. 56 (1978) 1003–1005. C.N. Pace, CRC Crit. Rev. Biochem. 3 (1975) 1–43. D.W. Bolen, M.M. Santoro, Biochemistry 27 (1988) 8069–8074. R. Gupta, F. Ahmad, Biochemistry 38 (1999) 2471–2479. R. Gupta, S. Yadav, F. Ahmad, Biochemistry 35 (1996) (1930) 11925–11930. C.N. Pace, G.R. Grimsley, S.T. Thomas, G.I. Makhatadze, Protein Sci. 8 (1999) 1500–1504. J.D. Thompson, T.J. Gibson, D.G. Higgins, Curr. Protoc. Bioinformatics 56 (2002) 59–63. J.K. Myers, C.N. Pace, J.M. Scholtz, Protein Sci. 4 (1995) 2138–2148. W. Kabsch, C. Sander, Biopolymers 22 (1983) 2577–2637. T.P. Creamer, R. Srinivasan, G.D. Rose, Biochemistry 36 (1997) 2832–2835. D. Freilder, Phys. Biochem.Ed. 2 (1982) 86–176. K.A. Dill, Biochemistry 29 (1990) 7133–7155. W. Kauzmann, Adv. Protein Chem. 14 (1959) 1–63.
Please cite this article in press as: K. Anwer et al., Arch. Biochem. Biophys. (2014), http://dx.doi.org/10.1016/j.abb.2014.01.005
572 573 574
577
Q2
579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642
