- Email: [email protected]
Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec)
FEBS 35822
No. of Pages 6, Model 5G
20 February 2013 FEBS Letters xxx (2013) xxx–xxx 1
journal homepage: www.FEBSLetters.org 5 6
Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec)
3 4 7 8 9 10 11 12 13 14 15 16 1 3 8 1 19 20 21 22 23 24 25 26 27 28 29 30
Q1
Livia Regina Manzine a, Vitor Hugo Balasco Serrão a, Luis Maurício Trambaioli da Rocha e Lima b, Marcos Michel de Souza a, Jefferson Bettini c, Rodrigo Villares Portugal c, Marin van Heel d,e, Otavio Henrique Thiemann a,⇑ a
Department of Physics and Informatics, Institute of Physics of Sao Carlos, University of Sao Paulo – USP, Sao Carlos, SP 13560-970, Brazil School of Pharmacy, Federal University of Rio de Janeiro – UFRJ, Rio de Janeiro, RJ 21941-617, Brazil Brazilian Nanotechnology National Laboratory – LNNano, Centro Nacional de Pesquisa em Energia e Materiais – CNPEM, Campinas, SP 13083-970, Brazil d Imperial College London, Molecular Biosciences, London SW7 2AZ, UK e Leiden Institute of Chemistry, Gorlaeus Laboratories, Leiden University, Leiden 9502, The Netherlands b c
a r t i c l e
i n f o
Article history: Received 20 November 2012 Revised 9 January 2013 Accepted 6 February 2013 Available online xxxx Edited by Michael Ibba Keywords: Selenocysteine synthase tRNAsec Stoichiometry
a b s t r a c t In bacteria selenocysteyl–tRNAsec (SelC) is synthesized by selenocysteine synthase (SelA). Here we show by fluorescence anisotropy binding assays and electron microscopical symmetry analysis that the SelA–tRNAsec binding stoichiometry is of one tRNAsec molecule per SelA monomer (1:1) rather than the 1:2 value proposed previously. Negative stain transmission electron microscopy revealed a D5 pointgroup symmetry for the SelA–tRNAsec assembly both with and without tRNAsec bound. Furthermore, SelA can associate forming a supramolecular complex of stacked decamer rings, which does not occur in the presence of tRNAsec. We discuss the structure–function relationships of these assemblies and their regulatory role in bacterial selenocysteyl–tRNAsec synthesis. Ó 2013 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies.
32 33 34 35 36 37 38 39 40
41 42 43
1. Introduction
44
Selenocysteine (Sec – U) is incorporated into selenoproteins and is present in the three different domains of life: Bacteria, Archaea and Eukarya [1]. Sec is encoded by a UGA codon associated to an mRNA-specific structure named SElenoCysteine Insertion Sequence (SECIS), and is dependent of a complex biosynthesis and incorporation pathway [2,3] which involves a specific tRNA (tRNAsec or SelC). In prokaryotes Seryl–tRNAsec is converted to Selenocysteyl–tRNAsec by the homodecameric enzyme Selenocysteine Synthase (SelA, EC 2.9.1.1) [3], a pyridoxal 50 -phosphate (PLP) dependent protein [4,5]. The existence of a SelA homolog in Archaea was demonstrated but no homolog is observed in the Eukarya domain. No apparent biological function was found for the putative SelA from the Archaea Methanococcus jannaschii, which exhibit 30% amino acid sequence identity with Escherichia coli SelA [6]. Moreover the putative Archaea SelA shows a dimeric – rather than the bacterial decameric structure.
45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
⇑ Corresponding author. Fax: +55 1633739881. E-mail address: [email protected] (O.H. Thiemann).
The PLP cofactor is covalently linked to E. coli SelA through the lysine residue (K295) [4] and participates in the Ser-Sec conversion reaction in two steps. The first step is the formation of a Schiff’s base between the a-amino group of the serine residue with the formyl group of PLP, resulting in the dehydration of the residue and the formation of the intermediate Aminoacrylyl-tRNAsec [3]. Selenium is then transferred to the Aminoacrylyl-tRNAsec intermediate from selenophosphate (a product of the selenophosphate synthetase enzyme, SelD) resulting in the formation of selenocysteyl– tRNAsec [7–9]. Early gel permeation experiments indicated that one tRNAsec molecule binds to a SelA dimer, resulting in five tRNAsec molecules bound to the SelA decamer [3]. Those samples were then investigated by electron microscopy using negative stain techniques and the stoichiometric composition was determined by scanning transmission electron microscopic mass determination indicating a structure for the SelA–tRNAsec complex with one tRNAsec bound per SelA dimer [10]. Cryo-EM experiments [11] of Moorella thermoacetica SelA confirmed the fivefold symmetry and the dimensions of the SelA protein. In the present communication an optimized purification protocol for the E. coli SelA allowed the characterization of its oligomeric state in solution by fluorescence anisotropy and electron
0014-5793/$36.00 Ó 2013 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies. http://dx.doi.org/10.1016/j.febslet.2013.02.014
Please cite this article in press as: Manzine, L.R., et al. Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec). FEBS Lett. (2013), http://dx.doi.org/10.1016/j.febslet.2013.02.014
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83
FEBS 35822
No. of Pages 6, Model 5G
20 February 2013 2
L.R. Manzine et al. / FEBS Letters xxx (2013) xxx–xxx
94
microscopy symmetry-analysis techniques. Our results show clearly that the binary SelA–tRNAsec complex exhibits a molecular ratio of 1:1 rather than a previously proposed 2:1 ratio. Our new data is of fundamental importance since it corrects the established stoichiometry of five tRNAsec molecules per SelA homodecamer which, in turn, lead to an incorrect assessment of the complex’s structure and to an incorrect interpretation of the molecular binding selectivity of the SelA–tRNAsec binary complex. Future functional and structural investigations of the SelA–tRNAsec complex would make no sense without taking the correct number of tRNAs into account.
95
2. Materials and methods
96
2.1. Optimization of the production of E. coli SelA protein
97
107
To overproduce SelA protein, Escherichia coli WL81460 (kDE3) cells expressing the recombinant selA gene were induced aerobically in LB medium, as described [12]. In the first desalted fraction, NycodenzÒ reagent solution was added to a final concentration of 20% and the mixture was concentrated using Amicon 50 000 MW concentrator. In the next step, the sample was submitted to a Superdex 200 HL column (1.6 60 cm size) (GE) equilibrated with potassium phosphate buffer (20 mM, pH 7.5) without PLP and the protein fractions were concentrated using Amicon 100 000 MW concentrator. Throughout this manuscript, SelA protein concentration is expressed as the total amount of monomeric subunits.
108
2.2. In vitro transcription of tRNAs
109
114
Each gene encoding E. coli selC and tRNAser and Trypanosoma brucei selC, were initially amplified by PCR [SI1] using primers with 20 bases overlap [SI2]. The PCR products purified from 2% agarose gel using Perfect Gel Cleanup kit (Eppendorf) were used in in vitro transcription reaction templates (MEGAscript kit, Ambion, Austin, TX, USA).
115
2.3. RNA synthesis and fluorescein labeling
116
For titration assays, tRNAsec from E. coli and T. brucei, tRNAser from E. coli and a 79 base single-stranded desoxiribonucleotide from part of a M. jannaschii gene were 50 end labeled with fluorescein maleimide, using the kit 50 EndTagTM Nucleic Acid Labeling System (Vector Laboratories, Burlingame, CA, USA). After labeling, the tRNAs or single-stranded DNA (ssDNA) were heated to 80 °C and slowly cooled to 35 °C with the addition of MgCl2 (20 mM). SelA fluorescein labeling (16 h at 10 °C) was performed with 3 lM SelA and 25 mM fluorescein isothiocyanate – FITC (Invitrogen) followed by a HiTrap (GE) desalting chromatography.
84 85 86 87 88 89 90 91 92 93
98 99 100 101 102 103 104 105 106
110 111 112 113
117 118 119 120 121 122 123 124 125 126 127 128 129 130
SelA for steady-state anisotropy measurements with excitation set to 480 nm and emission recorded through a 515 nm cut-off filter. Anisotropy values and total fluorescence were calculated by the ISS program. In all cases maximal dilution was less than 20%. Fluorescence anisotropy binding assays of tRNAs or ssDNA to SelA were performed as described previously [13]. Briefly, aliquots of concentrated SelA (400 lM) were sequentially added to 10 nM of fluorescein-labeled tRNAs or ssDNA. SelA–tRNAsec stoichiometry binding assays were performed using 40 lM unlabeled and 10 nM fluorescein-labeled tRNAsec and aliquots of SelA protein were added as described above. For inverse stoichiometry titration assay, 10 lM unlabeled and 30 nM fluorescein-labeled SelA was titrated with tRNAsec. The mixtures were homogenized and equilibrated for 3 min at 25 °C prior to anisotropy measurements. The fitting of the titration curves was performed as described in Ref. [14].
140
2.5. Transmission electron microscopy
156
For SelA–tRNAsec complex visualization by negative stain, SelA samples (0.5 mg/ml) in potassium phosphate buffer (20 mM, pH
157
2.4. Fluorescence anisotropy binding assay of SelA–SelA and SelA– nucleic acids interactions The fluorescence measurements were performed in an ISS-PC spectrofluorimeter (ISS, Champaign, IL, USA). Protein labeling efficiency was calculated as shown in Eq. (1):
131
FITC=SELA ðmol=molÞ ¼ ðAbs
FITC494nm
=EFITC494nm Þ=ðAbs
133 134 135 136 137 138 139
Ptn280nm
=EPtn280nm Þ ð1Þ
1
1
1
1
using 65 000 M cm and 35 785 M cm as the extinction coefficient (E) of protein-bound FITC at SelA absorbance (Abs) at 494 nm and at 280 nm, respectively [13]. Fluorescence anisotropy assay of SelA–SelA monomers binding was performed by measurements in ‘‘L’’ geometry, in 20 mM potassium phosphate buffer, at 25 °C. Aliquots of an unlabeled SelA were added to 30 nM fluorescein-labeled
Fig. 1. Fluorescence anisotropy titration of SelA protein. (a) Data for SelA isotherms followed by fluorescence anisotropy of fluorescein-labeled SelA. Inset: log scale. Two samples were purified independently (black and white circles). Each curve is an average of three independent experiments. (b) Negative Stain EM image of the SelA sample (0.5 mg/ml concentration). The white box shows a supramolecular stack-structure and the white circle a side-by-side assembly of SelA homodecamers.
Please cite this article in press as: Manzine, L.R., et al. Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec). FEBS Lett. (2013), http://dx.doi.org/10.1016/j.febslet.2013.02.014
141 142 143 144 145 146 147 148 149 150 151 152 153 154 155
158
FEBS 35822
No. of Pages 6, Model 5G
20 February 2013 L.R. Manzine et al. / FEBS Letters xxx (2013) xxx–xxx 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178
7.5) were incubated with tRNAsec in the same buffer for 30 min at 25 °C at a molecular ratio of 10 SelA monomers for 12 tRNAsec molecules. Holey carbon-coated grids were glow discharged for 25 s at 15 mA using an easiGlow system (PELCO). A 3 ll sample was deposited onto the grid for 30 s, followed by two washing steps (HEPES 10 mM, pH 7.5), staining with 3 ll of 2% uranyl acetate (30 s), blotting and air-drying. Images were recorded at 3 lm defocus at 60 000 magnification using a Jeol JEM-2100 operating at 200 kV. For eigenimage analysis, micrographs were recorded on Kodak SO-163 film and digitalized using an ArtixScan F1 (Microtek) scanner. SelA ‘‘stack’’ images were taken using a MultiScan 794 MSC (Gatan). Image analysis was performed using the IMAGIC 4D software package [15]. A total of 7582 particles were picked using a semi-automatic approach, filtered, normalized, centered and subjected to multivariate statistical analysis (MSA) [16,17] to obtain the eigenimages of the dataset and classified to obtain the different characteristic molecular views. The dataset was split in two groups: ‘‘top’’ views (7388 particles) and ‘‘side’’ views (194 particles). ‘‘Top’’ views were rotationally aligned to the second eigenimage of the full dataset and subjected to MSA. The results
3
were classified in 700 classes using the three first eigenimages of ‘‘top’’ views dataset, generating class averages that show a distribution between unbound SelA and tRNAsec bound states.
179
3. Results
182
3.1. SelA supramolecular assembly
183
Solution binding isotherms using fluorescence anisotropy spectroscopy were performed to investigate the SelA assembly process in the absence of tRNAsec. Fluorescein-labeled SelA (30 nM) was titrated with increasing amount of non-labeled SelA and the fluorescence anisotropy was monitored. Representative data of binding isotherms are shown in Fig. 1a. At low SelA concentration, the SelA fluorescein anisotropy is of about 40 mA. Increasing protein concentration results in accompanying increase in fluorescein anisotropy from the conjugated SelA. No clear convergence for a saturating plateau at SelA concentrations up to 10 lM was observed. Further increase in protein concentration was not possible due to technical limitations. Negative stain images of SelA samples
184
Fig. 2. Binding of SelA to tRNAsec and tRNAser from E. coli, tRNAsec from T. brucei and a non-specific DNA. (a) Binding isotherms for tRNAsec (d) and tRNAser from E. coli (s), tRNAsec from T. brucei (.) and non-specific DNA (D) in 20 mM of potassium phosphate buffer, pH 7.5. Inset: log scale. (b) tRNAsec from E. coli binding isotherms with 20 mM of potassium phosphate buffer, pH 7.5 (d), 100 mM of NaCl and 20 mM of MgCl2 (s), 300 mM of NaCl and 50 mM of MgCl2 (.). Inset: log scale. (c) tRNAser from E. coli binding isotherms in same conditions as for tRNAsec. Inset: log scale.
Please cite this article in press as: Manzine, L.R., et al. Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec). FEBS Lett. (2013), http://dx.doi.org/10.1016/j.febslet.2013.02.014
180 181
185 186 187 188 189 190 191 192 193 194 195
FEBS 35822
No. of Pages 6, Model 5G
20 February 2013 4
L.R. Manzine et al. / FEBS Letters xxx (2013) xxx–xxx
202
(10 lM monomer concentration) reveal a polydisperse system with the putative decamers, often associated into supramolecular stacks or side-by-side assemblies (Fig. 1b). This observation supports the fluorescence anisotropy isotherms interpretation of fluorescein-labeled SelA, as a combined process of formation of decamers followed by their association into larger supramolecular assemblies of decamers.
203
3.2. Specificity of SelA–tRNAs interaction
204
223
In order to investigate the mechanisms of SelA binding to tRNAs, we performed fluorescence anisotropy binding assayswith SelA and fluorescein-labeled tRNAs. E. coli tRNAsec and tRNAser, T. brucei (a Kynetoplastidae eukaryote) tRNAsec (Fig. S2) and a 79 nucleotide long single-stranded DNA (ssDNA, as control) at 10 nM concentration were titrated with increasing amounts of unlabeled SelA in separate experiments. The fluorescence anisotropy increased progressively as a function of SelA concentration, for all tRNAs tested (Fig. 2a). Both E. coli tRNAsec and tRNAser show no apparent saturation in the binding curves up to 18 lM (Fig. 2b and c, respectively) and the SelA–tRNA binding was not affected by the solution ionic strength (Na+ and Mg2+). The binding isotherms of SelA to tRNAsec from both E. coli and T. brucei also displayed non-saturating binding curves up to 18 lM of SelA. In contrast to what was observed for SelA–tRNAsec, the binding curve for SelA–tRNAser (Fig. 2a) shows a clear saturation plateau and a Kd of about 2.5 lM. The homodecamer complex binding to the ssDNA control showed a reduced affinity up to 20 lM. These data indicate the selective nature of the SelA binding to tRNAsec.
224
3.3. Stoichiometry of SelA–tRNAsec interaction
225
The binary complex SelA–tRNAsec binding stoichiometry was determined by fluorescence anisotropy binding assays using fluorescein-labeled and unlabeled tRNAsec at a higher concentration than the apparent Kd determined from Fig. 2a. A progressive increase in tRNAsec fluorescence anisotropy was observed as a function of the addition of SelA (Fig. 3a). At a SelA–tRNAsec ratio of 1:1 we observed a change in the tRNAsec dependant anisotropy as a function of SelA concentration as shown by the inflection in Fig. 3a, indicated by a dashed line. The previously proposed ratio of 2:1 is indicated in the same figure by a dotted line. This pattern is compatible with a change in binding mechanism, most likely from a specific to a non-specific binding mode, indicating a specific interaction of 1 SelA monomer to 1 tRNAsec molecule. The reverse stoichiometric binding assay, using unlabeled and fluorescein-labeled SelA, showed high initial anisotropy values that are consistent with the existence of both the decamers and supramolecular oligomers, as observed in Fig. 1. A steep decrease in SelA anisotropy is observed as a function of increasing concentration of tRNAsec (Fig. 3b), reaching a minimum value at 10 lM of tRNAsec and a plateau in the fluorescein anisotropy. Upon addition of the tRNAsec, we believe that the SelA decamers detach from the supramolecular-stack structures, due to a higher affinity to tRNAsec, forming the binary complex. This assay shows a sharp inflection at a SelA–tRNAsec ratio of 1:1, indicating the binding stoichiometry of 10 tRNA to a decamer of SelA. This inflection is indicated as a dashed line in Fig. 3b.
196 197 198 199 200 201
205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222
226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255
3.4. Electron Microscopy (EM) characterization of the SelA–tRNAsec assembly The binary complex stoichiometry (prepared with saturating concentrations of tRNAsec) was investigated by Electron Microscopy (EM) Symmetry Analysis [16]. The first ten eigenimages of
the centered-particles dataset are shown in Fig. 4a. The first eigenimage (Fig. 4a, 1) is equivalent to the sum of all particles; the second and third eigenimages (Fig. 4a, 2–3) together show that the main symmetry component of the dataset is fivefold (these eigenimages relate to each other as a sine and a cosine function along circles). At the same time, these eigenimages are not just fivefold symmetric but also exhibit mirror symmetry (5 m), indicative of a projection image along the fivefold axis of a D5 pointgroup symmetry structure, rather than that of a C5 structure. The sixth eigenimage (Fig. 4a, 6), a rotationally symmetric white ring, may be related to the size variation of the molecule when the tRNAs are present. The seventh and eighth eigenimages (Fig. 4a, 7–8) are associated with the side view images of the complex, which views are relatively rare within the dataset. Fig. 4b shows the first five eigenimages of the top-views dataset after rotationally aligning the images with respect to the second
Fig. 3. Stoichiometric solution binding assay of SelA to tRNAsec. (a) Fluorescence anisotropy binding assay. Absolute values are shown for the interaction of unlabeled (40 lM) and fluorescein maleimide-labeled (10 nM) tRNAsec with the addition of SelA protein in potassium phosphate buffer, pH 7.5. ., s and d are binding curves of independent experiments. SelA–tRNAsec ratio of 1:1 and 2:1 are indicated by dashed and dotted lines respectively. Inset: normalized values. (b) Reverse stoichiometric binding assay. Anisotropy absolute values are shown for the interaction of unlabeled (10 lM) and fluorescein-labeled (30 nM) SelA monomer in potassium phosphate buffer (20 mM, pH 7.5) with the addition of tRNAsec. s and d are binding curves of independent experiments. SelA–tRNAsec ratio of 1:1 is indicated by a dashed line. Inset: normalized values.
Please cite this article in press as: Manzine, L.R., et al. Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec). FEBS Lett. (2013), http://dx.doi.org/10.1016/j.febslet.2013.02.014
256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271
FEBS 35822
No. of Pages 6, Model 5G
20 February 2013 L.R. Manzine et al. / FEBS Letters xxx (2013) xxx–xxx
5
Fig. 4. Electron microscopical analysis of negatively stained complexes. (A) First ten eigenimages obtained from the full 7582 particles dataset. Eigenimages A2 and A3 show the underlying D5 symmetry of this dataset. (B) First 5 eigenimages of MSA of the 7388 ‘‘top’’ views of the dataset after rotational alignment to the second eigenimage of the dataset – A2. (C) Class averages of the unbound state of SelA. (D) Class averages of the SelA–tRNAsec binary complex.
288
eigenimage (Fig. 4a, 2) [17]. The first eigenimage (Fig. 4b, 1) represents the sum of all rotationally aligned images in the dataset and exhibits a fivefold symmetric structure [10] plus mirror symmetry, indicating the presence of an overall D5 pointgroup symmetry. Eigenimages 2 and 3 of Fig. 4b also show fivefold symmetry, and mirror symmetry, also indicating D5 group symmetry for the molecule. Eigenimage 3 (Fig. 4b) shows 10 strong intensities localized at the periphery of the decamers. This eigenimage discriminates between structures with and without the 10 tRNAsec molecules bound. The dataset is a mixture of a population of SelA–tRNAsec complexes and of SelA homodecamers. Different class-averages of these two subpopulations were obtained after classification of the top views using the three first eigenvectors shown in Fig. 4b– d shows five different class-averages for the SelA homodecamer and the binary complex respectively. Extra mass in the perimeter of the SelA–tRNAsec complex can be observed compared to the SelA homodecamer.
289
4. Discussion
290
The selC gene coding for tRNAsec (SelC) was identified as one of four genes whose products are required for the formation of selenoproteins in E. coli [18]. During selenocysteine synthesis and incorporation of the selenocysteine pathway, tRNAsec is aminoacylated with L-serine by Seryl–tRNAsec synthetase [2] and the Seryl– tRNAsec serves as substrate for selenocysteine synthase (SelA), a pyridoxal 50 -phosphate dependent enzyme that catalyzes the conversion of Seryl–tRNAsec to Selenocysteil–tRNAsec [19]. The tRNAsec molecule shows structural differences compared to other tRNAs [20] in regions previously considered invariant. These
272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287
291 292 293 294 295 296 297 298 299
differences are the presence of 8 bp in the acceptor arm, a purine residue at position 8, a pair purine–pyrimidine at positions 11/ 24, pyrimidines at positions 14 and 15, and the UGA-decoding UCA anticodon [21]. In our study, we analyzed the formation of the decameric structure of SelA and its interaction with tRNAsec using the E. coli tRNAser and T. brucei tRNAsec as control RNAs. SelA expression in E. coli WL81460 (kDE3) strain resulted in a tRNAsec-free SelA protein, that would otherwise incur in the interference in binding experiments by having endogenous tRNAsec bound to SelA [12]. This approach resulted in a significant improvement in SelA yield, purity and stability, maintaining the biophysical and biochemical characteristics of this macromolecule. SelA fluorescence anisotropy assays performed to verify the oligomerization process did not show a plateau region in the titration curve. Contrary to expectation, the fluorescence anisotropy values tend to increase without reaching a plateau. Transmission electron microscopy images of SelA showed that the decamer organizes in supramolecular structures as stacks or side by side assemblies, which explains the fluorescence anisotropy results. Since these supramolecular structures are present at low protein concentrations, the determination of SelA decamer dissociation constants by fluorescence anisotropy technique is not possible. Consequently, we inferred that the SelA decamer’s dissociation constant would be in the picomolar range, which are not detectable by fluorescence anisotropy under the conditions employed. Fluorescence anisotropy analysis of SelA–tRNA interaction has presented interesting results. First, the variation of the buffer ionic strength did not affect the interaction of SelA with either E. coli tRNAsec or tRNAser indicating that the protein–tRNA interaction is specific and not due to ionic interaction alone. In vivo studies from
Please cite this article in press as: Manzine, L.R., et al. Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec). FEBS Lett. (2013), http://dx.doi.org/10.1016/j.febslet.2013.02.014
300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329
FEBS 35822
No. of Pages 6, Model 5G
20 February 2013 6
L.R. Manzine et al. / FEBS Letters xxx (2013) xxx–xxx
361
Baron and coworkers, 1990 [22] demonstrated that the tRNAsec acceptor arm length of 8 bp is important for UGA decoding and that a reduction of the variable arm length caused a decrease in Sec incorporation. According to our studies, the maintenance of base pairing of the acceptor arm proved to be essential for the proper interaction of tRNA with SelA and shortening the length of the variable arm (tRNAsec T. brucei) actually promoted a decrease in the binding affinity in relation to the specific ligand tRNAsec. The profile obtained for SelA–tRNAser interaction suggest a different mechanism for this interaction based on the tRNAsec binding affinity with SelA. From these results we propose that tRNAser interacts exclusively with the free SelA decamers resulting in the early stabilization of the fluorescence anisotropy values. Our electron microscopical top-views of the binary complex show additional material in its peripheral region, compared to the SelA homodecamer. Similar results had been obtained by Engelhardt et al. [10]. Our eigenimage analysis also yielded 10 strong intensities localized at the periphery of the decamers. That extra material is thus the likely location of tRNAs in the binary complex. Since symmetry analysis yielded a D5 pointgroup symmetry, both for the SelA homodecamer, and for the binary complex, the binary complex must also have a stoichiometric ratio of 1:1 of SelA monomers and tRNAsec. Our combined data of EM symmetry analysis and direct and reverse fluorescence anisotropy titration showed a stoichiometric ratio of one SelA decamer to ten tRNAsec molecules. The results also revealed the presence of supramolecular stack-structures in the samples. A correct structural and functional interpretation of the SelA–tRNAsec complex is impossible without knowing its stoichiometric composition. For example, crystallization attempts aiming at achieving a 2:1 stoichiometry, may be hampered due to sample heterogeneity.
362
Acknowledgements
363
369
This work was supported in part by the research Grants 98/ 14138-2 and 06/53904-0 from the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), CAPES, from the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq Grant: 400796/2012-0), from the EU/NOE (NOE – PE0748), from the Dutch ministry of economic affairs (Cyttron Project: BIBCR_PX0948/ Cyttron II FES-0908) and from the BBSRC (Grant: BB/G015236/1).
370
Appendix A. Supplementary data
371 373
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.febslet.2013. 02.014.
374
References
330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360
364 365 366 367 368
372
375 376 377
[1] Böck, A. and Stadtman, T.C. (1988) Selenocysteine, a highly specific component of certain enzymes, is incorporated by a UGA-directed co-translational mechanism. BioFactors 1, 245–250.
[2] Leinfelder, W., Zehelein, E., Mandrand-Berthelot, M.A. and Böck, A. (1988) Gene for a novel tRNA species that accepts L-serine and cotranslationally inserts selenocysteine. Nature 331, 723–725. [3] Forchhammer, K. and Böck, A. (1991) Selenocysteine synthase from Escherichia coli analysis of the reaction sequence. J. Biol. Chem. 266, 6324–6328. [4] Tormay, P., Wilting, R., Lottspeich, F., Mehta, P.K., Christen, P. and Böck, A. (1998) Bacterial selenocysteine synthase structural and functional properties. Eur. J. Biochem. FEBS 254, 655–661. [5] Forchhammer, K., Leinfelder, W., Boesmiller, K., Veprek, B. and Böck, A. (1991) Selenocysteine synthase from Escherichia coli. Nucleotide sequence of the gene (selA) and purification of the protein. J. Biol. Chem. 266, 6318–6323. [6] Kaiser, J.T., Rother, G.M., Engelhardt, H., Rodnina, M.V. and Wahl, M.C. (2005) Structural and functional investigation of a putative archaeal selenocysteine synthase. Biochemistry 44, 13315–13327. [7] Glass, R.S., Singh, W.P., Jung, W., Veres, Z., Scholz, T.D. and Stadtman, T.C. (1993) Monoselenophosphate: synthesis, characterization, and identity with the prokaryotic biological selenium donor, compound SePX. Biochemistry 32, 12555–12559. [8] Mihara, H., Kurihara, T., Watanabe, T., Yoshimura, T. and Esaki, N. (2000) CDNA cloning, purification, and characterization of mouse liver selenocysteine lyase. Candidate for selenium delivery protein in selenoprotein synthesis. J. Biol. Chem. 275, 6195–6200. [9] Allmang, C., Wurth, L. and Krol, A. (2009) The selenium to selenoprotein pathway in eukaryotes: more molecular partners than anticipated. Biochim. Biophys. Acta 1790, 1415–1423. [10] Engelhardt, H., Forchhammer, K., Muller, S., Goldie, K.N. and Böck, A. (1992) Structure of selenocysteine synthase from Escherichia coli and location of tRNA in the Seryl–tRNA(sec)-enzyme complex. Mol. Microbiol. 6, 3461–3467. [11] Fischer, N., Paleskava, A., Gromadski, K.B., Konevega, A.L., Wahl, M.C., Stark, H. and Rodnina, M.V. (2007) Towards understanding selenocysteine incorporation into bacterial proteins. Biol. Chem. 388, 1061–1067. [12] Manzine, L.R., Cassago, A., da Silva, M.T.A. and Thiemann, O.H. (2013) An efficient protocol for the production of tRNA-free recombinant selenocysteine synthase (SELA) from Escherichia coli and its biophysical characterization. Protein Expr. Purif. 88, 80–84. [13] Matozo, H.C., Santos, M.A.M., Neto, M.O., Bleicher, L., Lima, L.M.T.R., Iuliano, R., Fusco, A. and Polikarpov, I. (2007) Low-resolution structure and fluorescence anisotropy analysis of protein tyrosine phosphatase g catalytic domain. Biophys. J . 92, 4424–4432. [14] Lima, L.M.T.R. and Silva, J.L. (2004) Positive contribution of hydration on DNA binding by E2c protein from papillomavirus. J. Biol. Chem. 279, 47968–47974. [15] van Heel, M., Portugal, R.V., Rohou, A., Linnemayr, C., Bebeacua, C., Schmidt, R., Grant, T.R. and Schatz, M. (2012) Four-Dimensional Cryo Electron Microscopy at Quasi Atomic Resolution: IMAGIC 4D. International Tables for Crystallography Volume F: Crystallography of Biological Macromolecules (second ed.), pp. 624–628. [16] Dube, P., Tavares, P., Lurz, R. and van Heel, M. (1993) Bacteriophage SPP1 portal protein: a DNA pump with 13-fold symmetry. EMBO J. 12, 1303–1309. [17] van Heel, M., Portugal, R.V. and Schatz, M. (2009) Multivariate Statistical Analysis in Single Particle (Cryo) Electron Microscopy. in: Verkley, A., Orlova, E, (Eds.), An electronic text book: Electron microscopy in Life Science, (3D-EM Network of Excellence). [18] Leinfelder, W., Forchhammer, K., Zinoni, F., Sawers, G., Mandrand-Berthelot, M.A. and Böck, A. (1988) Escherichia coli genes whose products are involved in selenium metabolism. J. Bacteriol. 170, 540–546. [19] Leinfelder, W., Forchhammer, K., Veprek, B., Zehelein, E. and Böck, A. (1990) In vitro synthesis of selenocysteinyl–tRNA(UCA) from Seryl–tRNA(UCA): involvement and characterization of the selD gene product. Proc. Natl. Acad. Sci. USA 87, 543–547. [20] Sprinzl, M., Hartmann, T., Weber, J., Blank, J. and Zeidler, R. (1989) Compilation of tRNA sequences and sequences of tRNA genes. Nucleic Acids Res. 17, r1– 172. [21] Heider, J., Leinfelder, W. and Böck, A. (1989) Occurrence and functional compatibility within enterobacteriaceae of a tRNA species which inserts selenocysteine into protein. Nucleic Acids Res. 17, 2529–2540. [22] Baron, C., Heider, J. and Böck, A. (1990) Mutagenesis of selC, the gene for the selenocysteine-inserting tRNA-species in E. coli: effects on in vivo function. Nucleic Acids Res. 18, 6761–6766.
Please cite this article in press as: Manzine, L.R., et al. Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec). FEBS Lett. (2013), http://dx.doi.org/10.1016/j.febslet.2013.02.014
378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446
No. of Pages 6, Model 5G
20 February 2013 FEBS Letters xxx (2013) xxx–xxx 1
journal homepage: www.FEBSLetters.org 5 6
Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec)
3 4 7 8 9 10 11 12 13 14 15 16 1 3 8 1 19 20 21 22 23 24 25 26 27 28 29 30
Q1
Livia Regina Manzine a, Vitor Hugo Balasco Serrão a, Luis Maurício Trambaioli da Rocha e Lima b, Marcos Michel de Souza a, Jefferson Bettini c, Rodrigo Villares Portugal c, Marin van Heel d,e, Otavio Henrique Thiemann a,⇑ a
Department of Physics and Informatics, Institute of Physics of Sao Carlos, University of Sao Paulo – USP, Sao Carlos, SP 13560-970, Brazil School of Pharmacy, Federal University of Rio de Janeiro – UFRJ, Rio de Janeiro, RJ 21941-617, Brazil Brazilian Nanotechnology National Laboratory – LNNano, Centro Nacional de Pesquisa em Energia e Materiais – CNPEM, Campinas, SP 13083-970, Brazil d Imperial College London, Molecular Biosciences, London SW7 2AZ, UK e Leiden Institute of Chemistry, Gorlaeus Laboratories, Leiden University, Leiden 9502, The Netherlands b c
a r t i c l e
i n f o
Article history: Received 20 November 2012 Revised 9 January 2013 Accepted 6 February 2013 Available online xxxx Edited by Michael Ibba Keywords: Selenocysteine synthase tRNAsec Stoichiometry
a b s t r a c t In bacteria selenocysteyl–tRNAsec (SelC) is synthesized by selenocysteine synthase (SelA). Here we show by fluorescence anisotropy binding assays and electron microscopical symmetry analysis that the SelA–tRNAsec binding stoichiometry is of one tRNAsec molecule per SelA monomer (1:1) rather than the 1:2 value proposed previously. Negative stain transmission electron microscopy revealed a D5 pointgroup symmetry for the SelA–tRNAsec assembly both with and without tRNAsec bound. Furthermore, SelA can associate forming a supramolecular complex of stacked decamer rings, which does not occur in the presence of tRNAsec. We discuss the structure–function relationships of these assemblies and their regulatory role in bacterial selenocysteyl–tRNAsec synthesis. Ó 2013 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies.
32 33 34 35 36 37 38 39 40
41 42 43
1. Introduction
44
Selenocysteine (Sec – U) is incorporated into selenoproteins and is present in the three different domains of life: Bacteria, Archaea and Eukarya [1]. Sec is encoded by a UGA codon associated to an mRNA-specific structure named SElenoCysteine Insertion Sequence (SECIS), and is dependent of a complex biosynthesis and incorporation pathway [2,3] which involves a specific tRNA (tRNAsec or SelC). In prokaryotes Seryl–tRNAsec is converted to Selenocysteyl–tRNAsec by the homodecameric enzyme Selenocysteine Synthase (SelA, EC 2.9.1.1) [3], a pyridoxal 50 -phosphate (PLP) dependent protein [4,5]. The existence of a SelA homolog in Archaea was demonstrated but no homolog is observed in the Eukarya domain. No apparent biological function was found for the putative SelA from the Archaea Methanococcus jannaschii, which exhibit 30% amino acid sequence identity with Escherichia coli SelA [6]. Moreover the putative Archaea SelA shows a dimeric – rather than the bacterial decameric structure.
45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
⇑ Corresponding author. Fax: +55 1633739881. E-mail address: [email protected] (O.H. Thiemann).
The PLP cofactor is covalently linked to E. coli SelA through the lysine residue (K295) [4] and participates in the Ser-Sec conversion reaction in two steps. The first step is the formation of a Schiff’s base between the a-amino group of the serine residue with the formyl group of PLP, resulting in the dehydration of the residue and the formation of the intermediate Aminoacrylyl-tRNAsec [3]. Selenium is then transferred to the Aminoacrylyl-tRNAsec intermediate from selenophosphate (a product of the selenophosphate synthetase enzyme, SelD) resulting in the formation of selenocysteyl– tRNAsec [7–9]. Early gel permeation experiments indicated that one tRNAsec molecule binds to a SelA dimer, resulting in five tRNAsec molecules bound to the SelA decamer [3]. Those samples were then investigated by electron microscopy using negative stain techniques and the stoichiometric composition was determined by scanning transmission electron microscopic mass determination indicating a structure for the SelA–tRNAsec complex with one tRNAsec bound per SelA dimer [10]. Cryo-EM experiments [11] of Moorella thermoacetica SelA confirmed the fivefold symmetry and the dimensions of the SelA protein. In the present communication an optimized purification protocol for the E. coli SelA allowed the characterization of its oligomeric state in solution by fluorescence anisotropy and electron
0014-5793/$36.00 Ó 2013 Published by Elsevier B.V. on behalf of the Federation of European Biochemical Societies. http://dx.doi.org/10.1016/j.febslet.2013.02.014
Please cite this article in press as: Manzine, L.R., et al. Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec). FEBS Lett. (2013), http://dx.doi.org/10.1016/j.febslet.2013.02.014
61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83
FEBS 35822
No. of Pages 6, Model 5G
20 February 2013 2
L.R. Manzine et al. / FEBS Letters xxx (2013) xxx–xxx
94
microscopy symmetry-analysis techniques. Our results show clearly that the binary SelA–tRNAsec complex exhibits a molecular ratio of 1:1 rather than a previously proposed 2:1 ratio. Our new data is of fundamental importance since it corrects the established stoichiometry of five tRNAsec molecules per SelA homodecamer which, in turn, lead to an incorrect assessment of the complex’s structure and to an incorrect interpretation of the molecular binding selectivity of the SelA–tRNAsec binary complex. Future functional and structural investigations of the SelA–tRNAsec complex would make no sense without taking the correct number of tRNAs into account.
95
2. Materials and methods
96
2.1. Optimization of the production of E. coli SelA protein
97
107
To overproduce SelA protein, Escherichia coli WL81460 (kDE3) cells expressing the recombinant selA gene were induced aerobically in LB medium, as described [12]. In the first desalted fraction, NycodenzÒ reagent solution was added to a final concentration of 20% and the mixture was concentrated using Amicon 50 000 MW concentrator. In the next step, the sample was submitted to a Superdex 200 HL column (1.6 60 cm size) (GE) equilibrated with potassium phosphate buffer (20 mM, pH 7.5) without PLP and the protein fractions were concentrated using Amicon 100 000 MW concentrator. Throughout this manuscript, SelA protein concentration is expressed as the total amount of monomeric subunits.
108
2.2. In vitro transcription of tRNAs
109
114
Each gene encoding E. coli selC and tRNAser and Trypanosoma brucei selC, were initially amplified by PCR [SI1] using primers with 20 bases overlap [SI2]. The PCR products purified from 2% agarose gel using Perfect Gel Cleanup kit (Eppendorf) were used in in vitro transcription reaction templates (MEGAscript kit, Ambion, Austin, TX, USA).
115
2.3. RNA synthesis and fluorescein labeling
116
For titration assays, tRNAsec from E. coli and T. brucei, tRNAser from E. coli and a 79 base single-stranded desoxiribonucleotide from part of a M. jannaschii gene were 50 end labeled with fluorescein maleimide, using the kit 50 EndTagTM Nucleic Acid Labeling System (Vector Laboratories, Burlingame, CA, USA). After labeling, the tRNAs or single-stranded DNA (ssDNA) were heated to 80 °C and slowly cooled to 35 °C with the addition of MgCl2 (20 mM). SelA fluorescein labeling (16 h at 10 °C) was performed with 3 lM SelA and 25 mM fluorescein isothiocyanate – FITC (Invitrogen) followed by a HiTrap (GE) desalting chromatography.
84 85 86 87 88 89 90 91 92 93
98 99 100 101 102 103 104 105 106
110 111 112 113
117 118 119 120 121 122 123 124 125 126 127 128 129 130
SelA for steady-state anisotropy measurements with excitation set to 480 nm and emission recorded through a 515 nm cut-off filter. Anisotropy values and total fluorescence were calculated by the ISS program. In all cases maximal dilution was less than 20%. Fluorescence anisotropy binding assays of tRNAs or ssDNA to SelA were performed as described previously [13]. Briefly, aliquots of concentrated SelA (400 lM) were sequentially added to 10 nM of fluorescein-labeled tRNAs or ssDNA. SelA–tRNAsec stoichiometry binding assays were performed using 40 lM unlabeled and 10 nM fluorescein-labeled tRNAsec and aliquots of SelA protein were added as described above. For inverse stoichiometry titration assay, 10 lM unlabeled and 30 nM fluorescein-labeled SelA was titrated with tRNAsec. The mixtures were homogenized and equilibrated for 3 min at 25 °C prior to anisotropy measurements. The fitting of the titration curves was performed as described in Ref. [14].
140
2.5. Transmission electron microscopy
156
For SelA–tRNAsec complex visualization by negative stain, SelA samples (0.5 mg/ml) in potassium phosphate buffer (20 mM, pH
157
2.4. Fluorescence anisotropy binding assay of SelA–SelA and SelA– nucleic acids interactions The fluorescence measurements were performed in an ISS-PC spectrofluorimeter (ISS, Champaign, IL, USA). Protein labeling efficiency was calculated as shown in Eq. (1):
131
FITC=SELA ðmol=molÞ ¼ ðAbs
FITC494nm
=EFITC494nm Þ=ðAbs
133 134 135 136 137 138 139
Ptn280nm
=EPtn280nm Þ ð1Þ
1
1
1
1
using 65 000 M cm and 35 785 M cm as the extinction coefficient (E) of protein-bound FITC at SelA absorbance (Abs) at 494 nm and at 280 nm, respectively [13]. Fluorescence anisotropy assay of SelA–SelA monomers binding was performed by measurements in ‘‘L’’ geometry, in 20 mM potassium phosphate buffer, at 25 °C. Aliquots of an unlabeled SelA were added to 30 nM fluorescein-labeled
Fig. 1. Fluorescence anisotropy titration of SelA protein. (a) Data for SelA isotherms followed by fluorescence anisotropy of fluorescein-labeled SelA. Inset: log scale. Two samples were purified independently (black and white circles). Each curve is an average of three independent experiments. (b) Negative Stain EM image of the SelA sample (0.5 mg/ml concentration). The white box shows a supramolecular stack-structure and the white circle a side-by-side assembly of SelA homodecamers.
Please cite this article in press as: Manzine, L.R., et al. Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec). FEBS Lett. (2013), http://dx.doi.org/10.1016/j.febslet.2013.02.014
141 142 143 144 145 146 147 148 149 150 151 152 153 154 155
158
FEBS 35822
No. of Pages 6, Model 5G
20 February 2013 L.R. Manzine et al. / FEBS Letters xxx (2013) xxx–xxx 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178
7.5) were incubated with tRNAsec in the same buffer for 30 min at 25 °C at a molecular ratio of 10 SelA monomers for 12 tRNAsec molecules. Holey carbon-coated grids were glow discharged for 25 s at 15 mA using an easiGlow system (PELCO). A 3 ll sample was deposited onto the grid for 30 s, followed by two washing steps (HEPES 10 mM, pH 7.5), staining with 3 ll of 2% uranyl acetate (30 s), blotting and air-drying. Images were recorded at 3 lm defocus at 60 000 magnification using a Jeol JEM-2100 operating at 200 kV. For eigenimage analysis, micrographs were recorded on Kodak SO-163 film and digitalized using an ArtixScan F1 (Microtek) scanner. SelA ‘‘stack’’ images were taken using a MultiScan 794 MSC (Gatan). Image analysis was performed using the IMAGIC 4D software package [15]. A total of 7582 particles were picked using a semi-automatic approach, filtered, normalized, centered and subjected to multivariate statistical analysis (MSA) [16,17] to obtain the eigenimages of the dataset and classified to obtain the different characteristic molecular views. The dataset was split in two groups: ‘‘top’’ views (7388 particles) and ‘‘side’’ views (194 particles). ‘‘Top’’ views were rotationally aligned to the second eigenimage of the full dataset and subjected to MSA. The results
3
were classified in 700 classes using the three first eigenimages of ‘‘top’’ views dataset, generating class averages that show a distribution between unbound SelA and tRNAsec bound states.
179
3. Results
182
3.1. SelA supramolecular assembly
183
Solution binding isotherms using fluorescence anisotropy spectroscopy were performed to investigate the SelA assembly process in the absence of tRNAsec. Fluorescein-labeled SelA (30 nM) was titrated with increasing amount of non-labeled SelA and the fluorescence anisotropy was monitored. Representative data of binding isotherms are shown in Fig. 1a. At low SelA concentration, the SelA fluorescein anisotropy is of about 40 mA. Increasing protein concentration results in accompanying increase in fluorescein anisotropy from the conjugated SelA. No clear convergence for a saturating plateau at SelA concentrations up to 10 lM was observed. Further increase in protein concentration was not possible due to technical limitations. Negative stain images of SelA samples
184
Fig. 2. Binding of SelA to tRNAsec and tRNAser from E. coli, tRNAsec from T. brucei and a non-specific DNA. (a) Binding isotherms for tRNAsec (d) and tRNAser from E. coli (s), tRNAsec from T. brucei (.) and non-specific DNA (D) in 20 mM of potassium phosphate buffer, pH 7.5. Inset: log scale. (b) tRNAsec from E. coli binding isotherms with 20 mM of potassium phosphate buffer, pH 7.5 (d), 100 mM of NaCl and 20 mM of MgCl2 (s), 300 mM of NaCl and 50 mM of MgCl2 (.). Inset: log scale. (c) tRNAser from E. coli binding isotherms in same conditions as for tRNAsec. Inset: log scale.
Please cite this article in press as: Manzine, L.R., et al. Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec). FEBS Lett. (2013), http://dx.doi.org/10.1016/j.febslet.2013.02.014
180 181
185 186 187 188 189 190 191 192 193 194 195
FEBS 35822
No. of Pages 6, Model 5G
20 February 2013 4
L.R. Manzine et al. / FEBS Letters xxx (2013) xxx–xxx
202
(10 lM monomer concentration) reveal a polydisperse system with the putative decamers, often associated into supramolecular stacks or side-by-side assemblies (Fig. 1b). This observation supports the fluorescence anisotropy isotherms interpretation of fluorescein-labeled SelA, as a combined process of formation of decamers followed by their association into larger supramolecular assemblies of decamers.
203
3.2. Specificity of SelA–tRNAs interaction
204
223
In order to investigate the mechanisms of SelA binding to tRNAs, we performed fluorescence anisotropy binding assayswith SelA and fluorescein-labeled tRNAs. E. coli tRNAsec and tRNAser, T. brucei (a Kynetoplastidae eukaryote) tRNAsec (Fig. S2) and a 79 nucleotide long single-stranded DNA (ssDNA, as control) at 10 nM concentration were titrated with increasing amounts of unlabeled SelA in separate experiments. The fluorescence anisotropy increased progressively as a function of SelA concentration, for all tRNAs tested (Fig. 2a). Both E. coli tRNAsec and tRNAser show no apparent saturation in the binding curves up to 18 lM (Fig. 2b and c, respectively) and the SelA–tRNA binding was not affected by the solution ionic strength (Na+ and Mg2+). The binding isotherms of SelA to tRNAsec from both E. coli and T. brucei also displayed non-saturating binding curves up to 18 lM of SelA. In contrast to what was observed for SelA–tRNAsec, the binding curve for SelA–tRNAser (Fig. 2a) shows a clear saturation plateau and a Kd of about 2.5 lM. The homodecamer complex binding to the ssDNA control showed a reduced affinity up to 20 lM. These data indicate the selective nature of the SelA binding to tRNAsec.
224
3.3. Stoichiometry of SelA–tRNAsec interaction
225
The binary complex SelA–tRNAsec binding stoichiometry was determined by fluorescence anisotropy binding assays using fluorescein-labeled and unlabeled tRNAsec at a higher concentration than the apparent Kd determined from Fig. 2a. A progressive increase in tRNAsec fluorescence anisotropy was observed as a function of the addition of SelA (Fig. 3a). At a SelA–tRNAsec ratio of 1:1 we observed a change in the tRNAsec dependant anisotropy as a function of SelA concentration as shown by the inflection in Fig. 3a, indicated by a dashed line. The previously proposed ratio of 2:1 is indicated in the same figure by a dotted line. This pattern is compatible with a change in binding mechanism, most likely from a specific to a non-specific binding mode, indicating a specific interaction of 1 SelA monomer to 1 tRNAsec molecule. The reverse stoichiometric binding assay, using unlabeled and fluorescein-labeled SelA, showed high initial anisotropy values that are consistent with the existence of both the decamers and supramolecular oligomers, as observed in Fig. 1. A steep decrease in SelA anisotropy is observed as a function of increasing concentration of tRNAsec (Fig. 3b), reaching a minimum value at 10 lM of tRNAsec and a plateau in the fluorescein anisotropy. Upon addition of the tRNAsec, we believe that the SelA decamers detach from the supramolecular-stack structures, due to a higher affinity to tRNAsec, forming the binary complex. This assay shows a sharp inflection at a SelA–tRNAsec ratio of 1:1, indicating the binding stoichiometry of 10 tRNA to a decamer of SelA. This inflection is indicated as a dashed line in Fig. 3b.
196 197 198 199 200 201
205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222
226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255
3.4. Electron Microscopy (EM) characterization of the SelA–tRNAsec assembly The binary complex stoichiometry (prepared with saturating concentrations of tRNAsec) was investigated by Electron Microscopy (EM) Symmetry Analysis [16]. The first ten eigenimages of
the centered-particles dataset are shown in Fig. 4a. The first eigenimage (Fig. 4a, 1) is equivalent to the sum of all particles; the second and third eigenimages (Fig. 4a, 2–3) together show that the main symmetry component of the dataset is fivefold (these eigenimages relate to each other as a sine and a cosine function along circles). At the same time, these eigenimages are not just fivefold symmetric but also exhibit mirror symmetry (5 m), indicative of a projection image along the fivefold axis of a D5 pointgroup symmetry structure, rather than that of a C5 structure. The sixth eigenimage (Fig. 4a, 6), a rotationally symmetric white ring, may be related to the size variation of the molecule when the tRNAs are present. The seventh and eighth eigenimages (Fig. 4a, 7–8) are associated with the side view images of the complex, which views are relatively rare within the dataset. Fig. 4b shows the first five eigenimages of the top-views dataset after rotationally aligning the images with respect to the second
Fig. 3. Stoichiometric solution binding assay of SelA to tRNAsec. (a) Fluorescence anisotropy binding assay. Absolute values are shown for the interaction of unlabeled (40 lM) and fluorescein maleimide-labeled (10 nM) tRNAsec with the addition of SelA protein in potassium phosphate buffer, pH 7.5. ., s and d are binding curves of independent experiments. SelA–tRNAsec ratio of 1:1 and 2:1 are indicated by dashed and dotted lines respectively. Inset: normalized values. (b) Reverse stoichiometric binding assay. Anisotropy absolute values are shown for the interaction of unlabeled (10 lM) and fluorescein-labeled (30 nM) SelA monomer in potassium phosphate buffer (20 mM, pH 7.5) with the addition of tRNAsec. s and d are binding curves of independent experiments. SelA–tRNAsec ratio of 1:1 is indicated by a dashed line. Inset: normalized values.
Please cite this article in press as: Manzine, L.R., et al. Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec). FEBS Lett. (2013), http://dx.doi.org/10.1016/j.febslet.2013.02.014
256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271
FEBS 35822
No. of Pages 6, Model 5G
20 February 2013 L.R. Manzine et al. / FEBS Letters xxx (2013) xxx–xxx
5
Fig. 4. Electron microscopical analysis of negatively stained complexes. (A) First ten eigenimages obtained from the full 7582 particles dataset. Eigenimages A2 and A3 show the underlying D5 symmetry of this dataset. (B) First 5 eigenimages of MSA of the 7388 ‘‘top’’ views of the dataset after rotational alignment to the second eigenimage of the dataset – A2. (C) Class averages of the unbound state of SelA. (D) Class averages of the SelA–tRNAsec binary complex.
288
eigenimage (Fig. 4a, 2) [17]. The first eigenimage (Fig. 4b, 1) represents the sum of all rotationally aligned images in the dataset and exhibits a fivefold symmetric structure [10] plus mirror symmetry, indicating the presence of an overall D5 pointgroup symmetry. Eigenimages 2 and 3 of Fig. 4b also show fivefold symmetry, and mirror symmetry, also indicating D5 group symmetry for the molecule. Eigenimage 3 (Fig. 4b) shows 10 strong intensities localized at the periphery of the decamers. This eigenimage discriminates between structures with and without the 10 tRNAsec molecules bound. The dataset is a mixture of a population of SelA–tRNAsec complexes and of SelA homodecamers. Different class-averages of these two subpopulations were obtained after classification of the top views using the three first eigenvectors shown in Fig. 4b– d shows five different class-averages for the SelA homodecamer and the binary complex respectively. Extra mass in the perimeter of the SelA–tRNAsec complex can be observed compared to the SelA homodecamer.
289
4. Discussion
290
The selC gene coding for tRNAsec (SelC) was identified as one of four genes whose products are required for the formation of selenoproteins in E. coli [18]. During selenocysteine synthesis and incorporation of the selenocysteine pathway, tRNAsec is aminoacylated with L-serine by Seryl–tRNAsec synthetase [2] and the Seryl– tRNAsec serves as substrate for selenocysteine synthase (SelA), a pyridoxal 50 -phosphate dependent enzyme that catalyzes the conversion of Seryl–tRNAsec to Selenocysteil–tRNAsec [19]. The tRNAsec molecule shows structural differences compared to other tRNAs [20] in regions previously considered invariant. These
272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287
291 292 293 294 295 296 297 298 299
differences are the presence of 8 bp in the acceptor arm, a purine residue at position 8, a pair purine–pyrimidine at positions 11/ 24, pyrimidines at positions 14 and 15, and the UGA-decoding UCA anticodon [21]. In our study, we analyzed the formation of the decameric structure of SelA and its interaction with tRNAsec using the E. coli tRNAser and T. brucei tRNAsec as control RNAs. SelA expression in E. coli WL81460 (kDE3) strain resulted in a tRNAsec-free SelA protein, that would otherwise incur in the interference in binding experiments by having endogenous tRNAsec bound to SelA [12]. This approach resulted in a significant improvement in SelA yield, purity and stability, maintaining the biophysical and biochemical characteristics of this macromolecule. SelA fluorescence anisotropy assays performed to verify the oligomerization process did not show a plateau region in the titration curve. Contrary to expectation, the fluorescence anisotropy values tend to increase without reaching a plateau. Transmission electron microscopy images of SelA showed that the decamer organizes in supramolecular structures as stacks or side by side assemblies, which explains the fluorescence anisotropy results. Since these supramolecular structures are present at low protein concentrations, the determination of SelA decamer dissociation constants by fluorescence anisotropy technique is not possible. Consequently, we inferred that the SelA decamer’s dissociation constant would be in the picomolar range, which are not detectable by fluorescence anisotropy under the conditions employed. Fluorescence anisotropy analysis of SelA–tRNA interaction has presented interesting results. First, the variation of the buffer ionic strength did not affect the interaction of SelA with either E. coli tRNAsec or tRNAser indicating that the protein–tRNA interaction is specific and not due to ionic interaction alone. In vivo studies from
Please cite this article in press as: Manzine, L.R., et al. Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec). FEBS Lett. (2013), http://dx.doi.org/10.1016/j.febslet.2013.02.014
300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329
FEBS 35822
No. of Pages 6, Model 5G
20 February 2013 6
L.R. Manzine et al. / FEBS Letters xxx (2013) xxx–xxx
361
Baron and coworkers, 1990 [22] demonstrated that the tRNAsec acceptor arm length of 8 bp is important for UGA decoding and that a reduction of the variable arm length caused a decrease in Sec incorporation. According to our studies, the maintenance of base pairing of the acceptor arm proved to be essential for the proper interaction of tRNA with SelA and shortening the length of the variable arm (tRNAsec T. brucei) actually promoted a decrease in the binding affinity in relation to the specific ligand tRNAsec. The profile obtained for SelA–tRNAser interaction suggest a different mechanism for this interaction based on the tRNAsec binding affinity with SelA. From these results we propose that tRNAser interacts exclusively with the free SelA decamers resulting in the early stabilization of the fluorescence anisotropy values. Our electron microscopical top-views of the binary complex show additional material in its peripheral region, compared to the SelA homodecamer. Similar results had been obtained by Engelhardt et al. [10]. Our eigenimage analysis also yielded 10 strong intensities localized at the periphery of the decamers. That extra material is thus the likely location of tRNAs in the binary complex. Since symmetry analysis yielded a D5 pointgroup symmetry, both for the SelA homodecamer, and for the binary complex, the binary complex must also have a stoichiometric ratio of 1:1 of SelA monomers and tRNAsec. Our combined data of EM symmetry analysis and direct and reverse fluorescence anisotropy titration showed a stoichiometric ratio of one SelA decamer to ten tRNAsec molecules. The results also revealed the presence of supramolecular stack-structures in the samples. A correct structural and functional interpretation of the SelA–tRNAsec complex is impossible without knowing its stoichiometric composition. For example, crystallization attempts aiming at achieving a 2:1 stoichiometry, may be hampered due to sample heterogeneity.
362
Acknowledgements
363
369
This work was supported in part by the research Grants 98/ 14138-2 and 06/53904-0 from the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP), CAPES, from the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq Grant: 400796/2012-0), from the EU/NOE (NOE – PE0748), from the Dutch ministry of economic affairs (Cyttron Project: BIBCR_PX0948/ Cyttron II FES-0908) and from the BBSRC (Grant: BB/G015236/1).
370
Appendix A. Supplementary data
371 373
Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.febslet.2013. 02.014.
374
References
330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360
364 365 366 367 368
372
375 376 377
[1] Böck, A. and Stadtman, T.C. (1988) Selenocysteine, a highly specific component of certain enzymes, is incorporated by a UGA-directed co-translational mechanism. BioFactors 1, 245–250.
[2] Leinfelder, W., Zehelein, E., Mandrand-Berthelot, M.A. and Böck, A. (1988) Gene for a novel tRNA species that accepts L-serine and cotranslationally inserts selenocysteine. Nature 331, 723–725. [3] Forchhammer, K. and Böck, A. (1991) Selenocysteine synthase from Escherichia coli analysis of the reaction sequence. J. Biol. Chem. 266, 6324–6328. [4] Tormay, P., Wilting, R., Lottspeich, F., Mehta, P.K., Christen, P. and Böck, A. (1998) Bacterial selenocysteine synthase structural and functional properties. Eur. J. Biochem. FEBS 254, 655–661. [5] Forchhammer, K., Leinfelder, W., Boesmiller, K., Veprek, B. and Böck, A. (1991) Selenocysteine synthase from Escherichia coli. Nucleotide sequence of the gene (selA) and purification of the protein. J. Biol. Chem. 266, 6318–6323. [6] Kaiser, J.T., Rother, G.M., Engelhardt, H., Rodnina, M.V. and Wahl, M.C. (2005) Structural and functional investigation of a putative archaeal selenocysteine synthase. Biochemistry 44, 13315–13327. [7] Glass, R.S., Singh, W.P., Jung, W., Veres, Z., Scholz, T.D. and Stadtman, T.C. (1993) Monoselenophosphate: synthesis, characterization, and identity with the prokaryotic biological selenium donor, compound SePX. Biochemistry 32, 12555–12559. [8] Mihara, H., Kurihara, T., Watanabe, T., Yoshimura, T. and Esaki, N. (2000) CDNA cloning, purification, and characterization of mouse liver selenocysteine lyase. Candidate for selenium delivery protein in selenoprotein synthesis. J. Biol. Chem. 275, 6195–6200. [9] Allmang, C., Wurth, L. and Krol, A. (2009) The selenium to selenoprotein pathway in eukaryotes: more molecular partners than anticipated. Biochim. Biophys. Acta 1790, 1415–1423. [10] Engelhardt, H., Forchhammer, K., Muller, S., Goldie, K.N. and Böck, A. (1992) Structure of selenocysteine synthase from Escherichia coli and location of tRNA in the Seryl–tRNA(sec)-enzyme complex. Mol. Microbiol. 6, 3461–3467. [11] Fischer, N., Paleskava, A., Gromadski, K.B., Konevega, A.L., Wahl, M.C., Stark, H. and Rodnina, M.V. (2007) Towards understanding selenocysteine incorporation into bacterial proteins. Biol. Chem. 388, 1061–1067. [12] Manzine, L.R., Cassago, A., da Silva, M.T.A. and Thiemann, O.H. (2013) An efficient protocol for the production of tRNA-free recombinant selenocysteine synthase (SELA) from Escherichia coli and its biophysical characterization. Protein Expr. Purif. 88, 80–84. [13] Matozo, H.C., Santos, M.A.M., Neto, M.O., Bleicher, L., Lima, L.M.T.R., Iuliano, R., Fusco, A. and Polikarpov, I. (2007) Low-resolution structure and fluorescence anisotropy analysis of protein tyrosine phosphatase g catalytic domain. Biophys. J . 92, 4424–4432. [14] Lima, L.M.T.R. and Silva, J.L. (2004) Positive contribution of hydration on DNA binding by E2c protein from papillomavirus. J. Biol. Chem. 279, 47968–47974. [15] van Heel, M., Portugal, R.V., Rohou, A., Linnemayr, C., Bebeacua, C., Schmidt, R., Grant, T.R. and Schatz, M. (2012) Four-Dimensional Cryo Electron Microscopy at Quasi Atomic Resolution: IMAGIC 4D. International Tables for Crystallography Volume F: Crystallography of Biological Macromolecules (second ed.), pp. 624–628. [16] Dube, P., Tavares, P., Lurz, R. and van Heel, M. (1993) Bacteriophage SPP1 portal protein: a DNA pump with 13-fold symmetry. EMBO J. 12, 1303–1309. [17] van Heel, M., Portugal, R.V. and Schatz, M. (2009) Multivariate Statistical Analysis in Single Particle (Cryo) Electron Microscopy. in: Verkley, A., Orlova, E, (Eds.), An electronic text book: Electron microscopy in Life Science, (3D-EM Network of Excellence). [18] Leinfelder, W., Forchhammer, K., Zinoni, F., Sawers, G., Mandrand-Berthelot, M.A. and Böck, A. (1988) Escherichia coli genes whose products are involved in selenium metabolism. J. Bacteriol. 170, 540–546. [19] Leinfelder, W., Forchhammer, K., Veprek, B., Zehelein, E. and Böck, A. (1990) In vitro synthesis of selenocysteinyl–tRNA(UCA) from Seryl–tRNA(UCA): involvement and characterization of the selD gene product. Proc. Natl. Acad. Sci. USA 87, 543–547. [20] Sprinzl, M., Hartmann, T., Weber, J., Blank, J. and Zeidler, R. (1989) Compilation of tRNA sequences and sequences of tRNA genes. Nucleic Acids Res. 17, r1– 172. [21] Heider, J., Leinfelder, W. and Böck, A. (1989) Occurrence and functional compatibility within enterobacteriaceae of a tRNA species which inserts selenocysteine into protein. Nucleic Acids Res. 17, 2529–2540. [22] Baron, C., Heider, J. and Böck, A. (1990) Mutagenesis of selC, the gene for the selenocysteine-inserting tRNA-species in E. coli: effects on in vivo function. Nucleic Acids Res. 18, 6761–6766.
Please cite this article in press as: Manzine, L.R., et al. Assembly stoichiometry of bacterial selenocysteine synthase and SelC (tRNAsec). FEBS Lett. (2013), http://dx.doi.org/10.1016/j.febslet.2013.02.014
378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446