Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule level and at the in cell level

Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule level and at the in cell level

BBAPAP-38959; No. of pages: 12; 4C: 2, 4, 6, 7, 8, 10 Biochimica et Biophysica Acta xxx (2013) xxx–xxx Contents lists available at SciVerse ScienceDi...

2MB Sizes 0 Downloads 60 Views

BBAPAP-38959; No. of pages: 12; 4C: 2, 4, 6, 7, 8, 10 Biochimica et Biophysica Acta xxx (2013) xxx–xxx

Contents lists available at SciVerse ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbapap

4 5 6 7

F

Q1 3

Carlos Marcuello a, 1, Sonia Arilla-Luna b, 1, Milagros Medina b,⁎, Anabel Lostao a, c,⁎⁎

O

2

Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule level and at the in cell level a

Laboratorio de Microscopías Avanzadas (LMA), Instituto de Nanociencia de Aragón (INA), Universidad de Zaragoza, Spain Departamento de Bioquímica y Biología Molecular y Celular, Facultad de Ciencias, and Instituto de Biocomputación y Física de Sistemas Complejos (BIFI)-Joint Unit BIFI-IQFR (CSIC), Universidad de Zaragoza, Spain c Fundación ARAID, Spain b

R O

1

8

i n f o

a b s t r a c t

P

a r t i c l e

Article history: Received 22 September 2012 Received in revised form 11 December 2012 Accepted 21 December 2012 Available online xxxx

D

Biochemical characterization of Corynebacterium ammoniagenes FADS (CaFADS) pointed to certain confusion about the stoichiometry of this bifunctional enzyme involved in the production of FMN and FAD in prokaryotes. Resolution of its crystal structure suggested that it might produce a hexameric ensemble formed by a dimer of trimers. We used atomic force microscopy (AFM) to direct imaging single CaFADS molecules bound to mica surfaces, while preserving their catalytic properties. AFM allowed solving individual CaFADS monomers, for which it was even possible to distinguish their sub-molecular individual N- and C-terminal modules in the elongated enzyme. Differences between monomers and higher stoichiometries were easily imaged, enabling us to detect formation of oligomeric species induced by ligand binding. The presence of ATP:Mg 2+ particularly induced the appearance of the hexameric assembly whose mean molecular volume resembles the crystallographic dimer of trimers. Finally, the AFM results are confirmed in cross-linking solution, and the presence of such oligomeric CaFADS species detected in cell extracts. All these results are consistent with the formation of a dimer of trimers during the enzyme catalytic cycle that might bear biological relevance. © 2012 Published by Elsevier B.V.

T

E

Keywords: FAD synthetase Atomic force microscopy Single-molecule Macromolecular protein ensemble Catalytic cycle

E

C

9 10 11 12 13 14 16 15 17 18 19 20 21 22 23

41

R

40

1. Introduction

43

Riboflavin (RF) can be de novo synthesized by plants, yeast and most prokaryotes, but RF uptake from the environment is essential for human nutrition and animal feeding. All organisms are able to transform RF, first into FMN, and then into FAD, by the sequential action of two activities, an ATP:riboflavin kinase (RFK, EC 2.71.26) and an ATP:FMN adenylyltransferase (FMNAT, EC 2.71.26) [1]. However, whereas eukaryotes use two different enzymes for FMN and FAD production, most prokaryotes depend on a single bifunctional enzyme, FAD

48 49 50

N C O

46 47

Abbreviations: FADS, FAD synthetase; RF, riboflavin; FMN, flavin mononucleotide; FAD, flavin adenine dinucleotide; ATP, adenosine 5′-triphosphate; RFK, ATP:riboflavin kinase; FMNAT, ATP:FMN adenylyltransferase; FADpp, FAD pyrophosphorylase; PIPES, 1,4-Piperazine diethane sulfonic acid; AFM, atomic force microscopy; UV, ultra-violet; SDS, sodium dodecyl-sulfate; BS3, bis[sulfosuccinimidyl]suberate; BSOCOES, bis[2-(succinimidyloxycarbonyloxy)ethyl]sulfone; DTT, dithiothreitol ⁎ Correspondence to: M. Medina, Departamento de Bioquímica y Biología Molecular y Celular. Facultad de Ciencias. Universidad de Zaragoza. Pedro Cerbuna, 12, 50009 Zaragoza, Spain. Tel.: +34 976762476; fax: +34 976762123. ⁎⁎ Correspondence to: A. Lostao, Laboratorio de Microscopías Avanzadas. Instituto de Nanociencia de Aragón. Universidad de Zaragoza. Edificio I+D+i. Campus Río Ebro. Mariano Esquillor s/n. 50018 Zaragoza, Spain. Tel.: +34 876555357; fax: +34 976762776. E-mail addresses: [email protected] (M. Medina), [email protected] (A. Lostao). 1 C. M. and S. A.-L. contributed equally to this work.

U

44 45

R

42

24 25 26 27 28 29 30 31 32 33 34 35 36 37 39 38

synthetase (FADS) [2,3]. In some prokaryotic FADS the FMN activity can also take place reversibly, being the backward activity known as FAD pyrophosphorylase activity (FADpp, EC 2.71.26). Differential molecular characteristics to carry out the same chemistry between the enzymatic components for FMN and FAD production in prokaryotes and eukaryotes envisage selective inhibition of prokaryotic FADS as a feasible treatment for diseases produced by pathogens [4–10]. The first stage in this process is to understand the structure-function relationship of all these enzymes, but particularly of prokaryotic FADS. Using as model the FADS from Corynebacterium ammoniagenes (CaFADS), binding and catalytic parameters for the substrates at the two catalytic sites, and the presence of one ATP and one flavin binding site at each one of the catalytic sites have been determined [9,11,12]. Crystal structures of Thermotoga maritima FADS (TmFADS) and CaFADS showed that the protein folds in two almost independent modules (Fig. 1a) [10,13]. The N-terminal (residues 1–186) forms an α/β dinucleotide binding domain with a typical Rossmann fold, ending in a small subdomain built by a β-hairpin and two short α-helices. It has not homology with mammal FMNATs, but belongs to the nucleotidyltransferase superfamily and is proposed to catalyze the adenylylation of FMN and, therefore, it is known as FMNAT-module [4,10,14]. The C-terminal (187–338) folds in a globular module consisting of a β barrel with six antiparallel strands, a terminal α-helix perpendicular to the barrel, and seven loops connecting them. It shares sequence and structural homology

1570-9639/$ – see front matter © 2012 Published by Elsevier B.V. http://dx.doi.org/10.1016/j.bbapap.2012.12.013

Please cite this article as: C. Marcuello, et al., Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule..., Biochim. Biophys. Acta (2013), http://dx.doi.org/10.1016/j.bbapap.2012.12.013

51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74

2

C. Marcuello et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

3.5 nm

10.5 nm

b

F

2.5 nm

P

6.2 nm

R O

4.0 nm

O

90 o

FMNAT

7.4 nm

RFK

2.9 nm

a

D

2.9 nm

85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107

C

E

R

83 84

R

81 82

O

79 80

C

77 78

with monofunctional RFKs and catalyzes the RFK activity by itself being known as the RFK-module [2,10,12,14–16]. Despite the homology of this RFK module with monofunctional eukaryotic enzymes carrying out this activity a more complex structural reorganization of its secondary structural elements is envisaged during catalysis [9]. The CaFADS crystal structure is also consistent with a quaternary organization formed by a dimer of trimers [10]. In the trimer the protomers are connected in a head-to-tail disposition that might allow ligand channeling between active sites (Fig. 1b). One of the still controversial points in the biochemistry of the RFK and FMNAT activities in eukaryotes is the presence of different isoforms of these proteins, their sub-cellular localization, and, in consequence, the role these factors might have in flavin homeostasis and flavoprotein biogenesis [6,7,17]. In prokaryotes the lack of cellular compartmentalization and of FADS isoforms will prevent a similar type of regulation. However, as described in several nucleotidyl transferases stabilizing hexameric organizations [18–20], the quaternary organization of CaFADS might contribute to regulate flavin homeostasis. In solution, oligomeric CaFADS ensembles have been detected together with the monomeric form [10,12]. Further work must be done to understand the factors inducing its formation, its stoichiometry, its putative formation in vivo and its functional role, as well as whether they correspond to the oligomers observed in the crystal structure. Many events in biological pathways are implemented by protein oligomers or multiprotein assemblies rather than by individual protein molecules. Determining the stoichiometry of these assemblies, as well as the factors involved in their formation-dissociation, are essential aspects for better understanding the molecular mechanisms governing such processes. Atomic force microscopy (AFM) is the only method presently available to obtain images in aqueous media at sub-nm resolution [21]. In the last decade AFM has been increasingly used for the study of biomolecules at the single-molecule level in physiologically relevant buffers [22,23]. It allows studying the

N

76

U

75

T

E

Fig. 1. Crystal structure of CaFADS (PDB code 2X0K). (a) Cartoon representation of the CaFADS protomer with the N-terminal FMNAT (residues 1–186) and C-terminal RFK (187– 338) modules coloured in green and orange, respectively. Ligands are modelled as sticks at their putative binding positions at each catalytic site. (b) Cartoon representation of the topology of the CaFADS dimer of trimers. The three protomers of the upper and lower trimers are coloured in pink and green scales, respectively. Dimensions are included.

topology, adhesion, elasticity, dynamics and other properties of biological samples [24]. However, despite the achieved resolution, few studies on protein morphology or protein association processes, allowing description of the architecture of subunit arrangement, have been reported [25–32]. Here, we show how AFM allows not only observing the topology of single CaFADS molecules, even identifying each domain, but also how it confirms that the presence of some ligands induces the formation of the dimer of trimers. The sample must be immobilized on a surface, but most of used methods preserve or even improve the enzymatic functionality regardless of using electrostatic or covalent protocols [33,34]. However, we have not found any previous study about the activity of enzyme oligomers, so we have performed some experiments to shed light on this question. Additionally, we also show that such oligomeric species can be detected in Escherichia coli cultures overexpressing CaFADS, as well as in those of C. ammoniagenes.

108

2. Materials and methods

124

2.1. Biological material

125

CaFADS was produced by heterologous over-expression of the previously produced pET28-CaFADS plasmid in E. coli, and purified as reported [12,14]. The C-terminal (RFK-module, Δ(1–182)FADS) and N-terminal (FMNAT-module, Δ(188–338)FADS) modules of CaFADS were individually produced using a similar protocol from the corresponding plasmids, pET28a-Δ(1–182)FADS and pET28a-Δ(188–338) FADS, but replacing either the DEAE-cellulose or the PhenylSheparose chromatographies with a HiPrep 26/60 Sephacryl S-200 one (GE Healthcare), respectively. After purification, samples were dialyzed in either 20 mM PIPES, pH 7.0 or potassium phosphate 50 mM, pH 8.0, and stored at −80 °C until used. All FADS samples were quantified through UV-Vis spectrometry; an ε279 nm of 27.8 mM−1.cm−1

126

Please cite this article as: C. Marcuello, et al., Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule..., Biochim. Biophys. Acta (2013), http://dx.doi.org/10.1016/j.bbapap.2012.12.013

109 110 111 112 113 114 115 116 117 118 119 120 121 122 123

127 128 129 130 131 132 133 134 135 136 137

C. Marcuello et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201

C

162 163

E

160 161

R

158 159

R

156 157

N C O

Additional approaches were used to confirm the production of oligomeric ensembles. Herein in vitro is referred to experiments carried out with purified CaFADS samples, while in vivo refers to those experiments in which the oligomerization state was analyzed in cellular crude extracts. For in vitro experiments both CaFADS and CaFADSm were incubated with ligands under the different conditions stated in the corresponding Fig. 7 in the results section, and subsequently the oligomerization state was analyzed by 10% native PAGE (N-PAGE) and 7% denaturing PAGE (SDS-PAGE) electrophoresis, as well as by gel filtration chromatography. N-PAGE and SDS-PAGE were carried out following standard protocols (BioRad electrophoresis guide), and visualized with coomassie staining. Gel filtration chromatography was carried out on a Superdex™ 200 10/ 300 GL column (GE Healthcare) previously equilibrated with potassium phosphate 50 mM, pH 8.0, and calibrated with the Gel Filtration Calibration Kit LMW (GE Healthcare). Chromatograms were fit to a set of Gaussian functions to identify the number of components in the elution profile. Two approaches were also conducted for in vivo detection of oligomeric ensembles of CaFADS. On one side, E. coli over-expressing CaFADS cells and C. ammonigenes cells were re-suspended in potassium phosphate 50 mM, pH 8.0 and broken by ultrasonic treatment at 4 °C using 18 cycles of 30 s in a DRH UP200 DR Hielsher sonicator. Crude extracts were centrifuged at maximum speed for 2 min to remove the pellet and the supernatants were incubated for 2 h at 4 °C with the bis[sulfosuccinimidyl]suberate (BS3) cross-linker (Thermo Scientific). The cross-linker reaction was stopped by adding the quenching solution (TRIS-HCl 500 mM (10 ×), pH 8.0) up to 50 mM. In a different set of experiments fresh cultures of either E. coli over-expressing CaFADS or C. ammoniagenes cells were incubated with bis[2-(succinimidyloxycarbonyloxy)ethyl]sulfone (BSOCOES) (Thermo Scientific), a cross-linker able to cross membranes and to get introduced into the cells, at a final concentration of 2 mM. After 2 h incubation at 4 °C the quenching solution was added, cells were boiled at 100 °C for 5 min, centrifuged at maximum speed for 2 min to remove the pellet, and the supernatant analyzed. Visualization of the association degree of CaFADS in the different in vivo samples was obtained by immunoblot (BioRad protein blotting guide). Samples were first separated by 7% SDS-PAGE and subsequently transferred onto polyvinylidene difluoride membranes. Then, the membranes were probed with the anti-CaFADS serum diluted at 1:3000 (Zeu Immunotec S.L.) and the peroxidase-conjugated anti-rabbit IgG antibody produced in goat (Sigma-Aldrich) diluted 1:10000. The membrane was then colour-developed in phosphocitrate buffer containing dioctyl sulfosuccinate (Sigma-Aldrich), 3,3′,5,5′-tetramethylbenzidine (Sigma-Aldrich) and hydrogen peroxide, following the green precipitate appearance.

F

149 150

U

147 148

O

AFM measurements were performed with a Cervantes Fullmode Scanning Probe Microscope (Nanotec Electrónica S.L.). Images were taken using the Jumping Mode operation (JM) [35], considered as a suitable non-intrusive scanning method for soft samples and especially for biomolecules [36]. JM works by performing a force curve at each point of the sample surface with a feedback time in between, the feedback signal being the loading normal force. For each pixel of the image maximum tip–sample adhesion and height are registered, allowing the quickly production of topography images. V-shaped silicon nitride cantilevers with integrated pyramidal tips and a spring constants of 0.06 N/m (Bruker Probes, SNL lever Probes) were employed. Cleaning of levers and production of images with JM were carried out as described elsewhere [36]. Unless otherwise stated 0.5 μM solutions of the different CaFADS samples were used. When samples required the addition of ligands or reactants, the mixtures were preincubated at 4 °C for 10 min to favor mixing and complex formation. In particular cases this time was reduced to b30 s or extended up to 16 hr. These solutions were then incubated on fresh exfoliated muscovite mica pieces (Electron Microscopy Sciences) for 10 min at room temperature. Subsequently the mica was washed three times with 20 mM PIPES, pH 6.0, to remove weakly joined molecules. The immobilized sample and the cantilever holder were introduced into a liquid cell (previously cleaned with 20% isopropanol and Millipore ultrapure water). AFM measurements were conducted in 20 mM PIPES, pH 6.0 at 20 °C, except for the RFK-module sample, that was imaged at pH 5.0. When reducing conditions were required freshly prepared dithiothreitol (DTT; Sigma-Aldrich) was added at 2–4 mM. In general, 0.8 mM MgCl2 was added in the experiments involving ligands, but controls at 8 mM MgCl2 were also performed. Analyses of the effect produced by the different ligands in the CaFADS oligomerization state were carried out by incubation of CaFADSm samples with different combinations of ligands. Typically concentrations used were 250 μM for ATP, ADP and PPi, and 50 μM for RF, FMN and FAD (Sigma-Aldrich). Samples containing flavins were protected from light to minimize side photo-chemical reactions. Image processing was performed with the WSxM software [37], and percentages for the different aggregation states were estimated as described [29]. At least 10 images of 10 different areas of the 500 × 500 nm sample surface were analyzed, and a minimum of 250 features were examined for each sample. Only clearly identified features were quantified, discarding unclear elements. Each feature was analyzed by zooming with the WSxM software, without losing image information. This function allows observing each oligomer in detail, and when complemented with the height profile makes possible to differentiate each monomer within the oligomer, therefore, allowing to unequivocally distinguishing the type of association. Percentages of each species were determined as the existing number of

145 146

202 203 204 205

2.3. Analytical methods for the detection of in vitro and in vivo oligomeric 206 states 207

R O

154 155

144

P

2.2. Atomic force microscopy imaging

142 143

D

153

140 141

this species with respect to the total number of species in the sample. The error was calculated from the dispersion of results in the analysis of different AFM images corresponding to different areas of the sample.

T

151 152

calculated in PIPES 20 mM, pH 7.0 was used for CaFADS [12], while for Δ(1–182)FADS and Δ(188–338)FADS the theoretical ε280 nm values were used, 14.4 mM−1.cm−1 and 14.0 mM−1.cm−1, respectively. Two different samples of CaFADS were subsequently used: 1) the sample obtained after purification to homogeneity with the standard protocol, herein CaFADS, and 2) the quantitatively isolated monomeric form, herein labeled CaFADSm, obtained by subsequent purification of the above CaFADS sample by gel filtration chromatography with a Superdex™ 200 10/300G-L (GE Healthcare). A rabbit anti-CaFADS antiserum was produced by injection of a CaFADS sample (Zeu Inmunotec S.L., Zaragoza, Spain) and used for Western blot (WB) assays (see below). Cultures of C. ammoniagenes strain DSM 20305 (ATCC 6872) were grown at 30 °C in a medium containing 0.5% glucose, 0.5% NaCl, 1.0% tryptone and 0.5% yeast extract, pH 7.2–7.4.

E

138 139

3

208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254

2.4. Evaluation of the functionality and folding of mica immobilized 255 CaFADS samples 256 Samples of CaFADSm (15 μM) were preincubated, either in the absence or presence of ATP (7.5 mM) and Mg 2+ (10 mM), during 10 min at room temperature, and subsequently immobilized on 2 cm × 2 cm mica pieces as described in Section 2.2. The time dependent ability of the immobilized samples to transform RF into FAD (including both RFK plus FMNAT activities) and of FMN into FAD (only FMNAT activity) was evaluated at 25 °C in a Synergy HT

Please cite this article as: C. Marcuello, et al., Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule..., Biochim. Biophys. Acta (2013), http://dx.doi.org/10.1016/j.bbapap.2012.12.013

257 258 259 260 261 262 263

273 274 275 276 277 278 279 280 281 282 283

υ0 ¼

ΔF Δt ⋅ðK FMN −K FAD Þ

297

3. Results

298

3.1. Single-molecule visualization of CaFADS by AFM

299 300

AFM scanning of CaFADS samples under conditions mimicking the physiological ones (20 mM PIPES, pH 6.0) and without any additional chemical modification or treatment, resulted in a clear set of images (Fig. 2a). Negatively charged CaFADS molecules can be immobilized on the mica surface by use of Mg 2+ in the buffer solution, allowing

301 302 303

C

E R

R

294 295

O

292 293

C

291

N

289 290

U

287 288

T

296

where ΔF is the decrease in the value of the fluorescence expressed in arbitrary units, Δt is the measurement of reaction time expressed in min−1, and KRF =KFMN and KFAD are the RF/FMN and FAD fluorescent rate constants expressed in μM−1. Qualitative conversions of RF into FMN and, of FMN into FAD were also assayed by incubation of the desorbed CaFADS variants with substrates and subsequent resolution of the reaction products by Thin Layer Chromatography on Silica Gel SIL-G-25 plates [14]. To check for folding of desorbed samples circular dichroism spectra were recorded in a Chirascan spectropolarimeter (Applied Photophysics Ltd.) at 16 °C in 5 mM PIPES with a path length of 0.1 cm in the far-UV and in 20 mM PIPES with a path length of 0.4 cm for the near-UV. pH values were adjusted for each one of the assayed samples.

284 285 286

F

271 272

O

270

R O

268 269

working at pH values slightly higher than the isoelectric point theoretically calculated (pI 5.5 for CaFADS). The counter ions act as a bridge between the negatively charged mica and the negatively charged molecules [38]. The produced images show that the protein concentration used was suitable to individually resolve molecules. Monomers and dimers were easy to analyze and differentiate in the same layer since they were on the plane, with maximum heights ~ 7 nm. We did not find trimers nor overlapped dimers. Tetramers and hexamers form two overlapped layers with maximum heights ~ 17 nm. The zoom function made possible to unequivocally attribute the association pattern of each oligomer. No more than two layers were detected in any case. It is important to remark that the measurement in the Z-height (height profiles) presents a high accuracy, less than 1 nm, whereas in the XY size of the feature, the width, is less accurate due to the well documented broadening effect related to the AFM tip dilation [39]. Despite this broadening effect, XY width can be compared in relative terms and it can be observed that dimers and tetramers have twice the width of the monomer features with the difference that tetramers have also twice the Z-height of monomers and dimers. The systematic analysis indicated that 65% of the features detected in the absence of a reductant agent corresponded with dimensions in the 4–7 nm range. These parameters are compatible with those for the monomeric CaFADS crystal structure (dimensions for a CaFADS protomer according to the pdb are 7.4 × 3.5 × 4.0 nm [10]) (Fig. 1a), despite that the heights detected by AFM are slightly affected by electrostatic tip–sample interactions and by volume surface increase, due to insertion of buffer molecules among the enzyme side chains. The quality of the data also allowed to identify oligomeric CaFADS species with different association stoichiometries, including overlaps in two planes, as well as to estimate percentages for the different oligomeric states. Visualization of the purified CaFADS samples unravels, together with 41% of the enzyme molecules corresponding to monomeric protein, 29% of the enzyme molecules associated as dimers, and 30% associated as amorphous and/or well organized tetramers (Fig. 2a, Table 1). This represents 65, 23 and 12%, respectively, of the features observed. Higher stoichiometries are not detected. The presence of oligomers in solution was previously reported by using analytical gel filtration, but at that point attempts to identify their stoichiometry and relative amounts were unsuccessful [10]. Incubation of CaFADS with the reductant agent DTT considerably altered the oligomerization pattern (Fig. 2b). Tetramers were not detected under these conditions and the percentage of monomers increased up to 70% of the total protein, while the percentage of dimers was similar to that in the absence of DTT (Table 1). This later observation suggests that although under our experimental conditions disulfide bridges are not

P

266 267

thermostated multimode plate reader (Bioteck Instruments) using 6 well plates and 96U well plates for measurements of samples immobilized on the mica and in solution, respectively. In a first set of experiments, the 2cmx2cm mica pieces were placed in the wells and activities were measured in a final volume of 1.5 mL in 20 mM PIPES, pH 6.0, including 10 mM MgCl2, 50 μM ATP and either 5 μM RF (RFK plus FMNAT activities) or 5 μM FMN (FMNAT activity). Changes in fluorescence intensity upon conversion of RF/FMN into FAD, with excitation and emission wavelengths at 420 nm and 530 nm, respectively, were used to monitor the activity. In a second set of experiments immobilized CaFADS molecules were desorbed from the mica surface by incubation with a 0.5 M NaCl solutions during 10 min at room temperature. Concentration of desorbed samples was then spectroscopically quantified and activity measured on 96U plates in a final volume of 100 μl following the same protocol as for the ones deposed on the mica. The activity of a fresh CaFADSm sample of known concentration was similarly measured as a control. FAD and RF/FMN fluorescence were individually calibrated using standard solutions. In all cases, the rate of FAD formation was calculated from the rate of fluorescence decrease (ΔF/Δt), taken as the tangent to the initial part of the experimental curve by applying the equation:

D

264 265

C. Marcuello et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

E

4

Fig. 2. AFM topography images of CaFADS. (a) 3D map of a CaFADS sample in 20 mM PIPES, pH 6.0. Different stoichiometries are observed for the detected features. In most of the monomers, modules are individually resolved. (b) 3D map of a CaFADS sample in 20 mM PIPES, 4 mM DTT, pH 6.0. A majority of monomers and a dimer can be observed. The image in panel 2b was the original scanned image, while the image in panel 2a is a zoomed image of an original AFM scan of a 500 nm2 area; this aspect slightly affects the roughness of the surface and also the lateral resolution expanding the size of XY features. Monomers, dimers and tetramers are shown in black, red and brown circles, respectively.

Please cite this article as: C. Marcuello, et al., Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule..., Biochim. Biophys. Acta (2013), http://dx.doi.org/10.1016/j.bbapap.2012.12.013

304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349

C. Marcuello et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

t1:8 t1:9 t1:10 t1:11

Features Molecules Features Molecules

350

DTT 4 mM

Monomers (%)

Dimers (%)

Tetramers (%)

65 41 82 70

23 29 18 30

12 30 – –

367

3.2. Association of CaFADS induced by ligand binding

368 369

To investigate the influence of ligands binding on the CaFADS organization we analyzed by AFM the effects produced upon incubation of its monomeric form with its different ligands in 20 mM PIPES, 2 mM DTT, pH 6.0. Results are summarized in Table 3. Incubation of CaFADSm with FMN did not produce any effect and the enzyme was mainly maintained as monomer. On the other hand, incubation with FAD or RF induced the appearance of dimers and well organized tetrameric ensembles (Fig. 4a). Incubation with ATP or ADP also induced the formation of several oligomeric species, including the formerly observed dimers and tetramers, but particularly of a new aggregation stoichiometry in the form of hexamers that accounts for up to 50% of the molecules (Fig. 4b).

370 371 372 373 374 375 376 377 378 379

t2:1 t2:2 t2:3 t2:4 t2:5 t2:6 t2:7

C

363 364

E

361 362

R

359 360

R

357 358

N C O

355 356

Table 2 Oligomeric ratios detected for purified CaFADSm under different control conditions by AFM. CaFADSm (0.5 μM) was prepared in 20 mM PIPES, pH 6.0. Features correspond to image units, while molecules refer to the amount of individual protein monomers in the corresponding image units. Error associated to percentage determination was ± 5–15%.

U

353 354

Conditions

Units

t2:8

t2:9 t2:10 t2:11 t2:12 t2:13 t2:14 t2:15 t2:16

T

365 366

responsible of the stabilization of CaFADS dimers, they might contribute to stabilize tetramers. To carry out a comparative analysis with a homogeneous sample, subsequent analyses were carried out by using the isolated monomeric fraction of the enzyme, CaFADSm. Under reducing conditions CaFADSm showed almost 100% of the molecules as monomers (Table 2). The high resolution of the image features obtained with this sample and the use of the WSxM zoom allowed distinguishing between both CaFADS modules, what permitted determining the type of domain contacts between monomers (Fig. 3). This allowed identifying a head-to-tail interaction within the dimer, where the N-terminal FMNAT-module of one protomer contacts with the C-terminal RFK-module of the second protomer with an angle of ~ 90° (Fig. 3b). The prevalence of this angle in most of the dimeric features suggests that interaction between protomers must be specific. This head-to-tail interaction resembles that described in the CaFADS dimer of trimers suggested by X-ray crystallography (Fig. 1) [10].

351 352

Association state Monomers Dimers Tetramers Amorphous Dimer of (%) (%) (%) hexamer trimers (%) (%)

No agent

Features Molecules 2 mM DTT Features Molecules 8 mM MgCl2 Features 2 mM DTT Molecules 0.25 mM ATP Features 2 mM DTT Molecules 0 MgCl2

85 71 97 94 77 58 65 38

13 22 3 6 18 28 23 27

2 7 – – 5 14 7 16

– – – – – – 2 7

– – – – – – 3 12

3.3. Association of CaFADS induced by catalysis

413

To evaluate the degree of CaFADS association along the catalytic cycle of this enzyme, CaFADSm was also incubated with the substrates of the activities it is able to catalyze [40], RFK and FMNAT (in both the forward and backward reactions, namely FMNAT and FADpp), at different times in the presence of DTT and Mg2+. AFM was again used to evaluate the formation of oligomeric species along the reactions. The results are summarized in Table 4. Simultaneous incubation of CaFADSm with RF and ATP:Mg2+ activates its RFK activity, a process that, according to the AFM images, induced the formation of dimers and organized tetramers immediately after mixing. Medium incubation times of the reaction mixture induced the conversion of some of the dimers and tetramers in hexamers, which include both amorphous and the dimer of trimers. Finally, incubations long enough to assure substrates consumption show CaFADS is stabilized mainly as a monomer with a small proportion of dimers. Mixing of CaFADSm with the substrates of the FMNAT reaction, FMN and ATP, show a quick production of dimers and tetramers in proportions similar to those obtained at similar incubation times for the RFK activity. However, increasing the incubation time up to 10 min induces the formation of the compact dimer of trimers. Longer incubation times still retain half of the proportion of the hexamer obtained along the reaction, but in this case both amorphous and the dimer of trimers are identified. Finally, we similarly analyzed the effect of the FADpp activity of CaFADSm in the oligomerization state. Incubation with FAD and PPi produced images resembling very much those obtained for the FMNAT reaction: short incubation times produced mainly monomers with some dimers and tetramers, 10 min incubations stabilized an important percentage of the dimer of trimers, and the equilibrium was finally reached with the dimer of trimers disappearing in favor of the amorphous hexamer and the monomer

414

O

No agent

t1:7

F

Association state

R O

Units

380 381

P

Conditions

Nevertheless, the morphology of the hexamers induced by these two adenine nucleotides resulted considerably different. While the dimer of trimers induced by incubation with ADP has an amorphous but reproducible, topology, ATP:Mg 2+ induced the formation of an organized dimer of trimers that resembles the oligomeric CaFADS ensemble reported by X-ray crystallography (Fig. 4b and c). To further clarify the role exerted by ATP a control was also carried out in the absence of MgCl2. Lack of the cation prevented ATP to induce formation of the large amounts of the organized dimer of trimers observed in its presence, but instead induced the appearance of the amorphous hexamer that was absent in the presence of the cation. This points at the ATP:Mg 2+ complex as responsible for the organization of the dimer of trimers. Control experiments also indicated that the single addition of an excess of the cation in the absence of the adenine nucleotide produced a similar pattern than that associated to the addition of RF or FAD (Tables 2 and 3). Simultaneous incubation of CaFADSm with the products of the RFK reaction, FMN and ADP also induce the formation of a large proportion of the organized dimer of trimers that in this case reaches up to 72% of the total amount of the protein (Table 3, Fig. 5). Functionality of the different CaFADS forms upon immobilization was proven by detecting the enzyme ability to transform both RF and FMN into FAD directly on the mica surface (Figure SP1a) with similar rates for CaFADSm and CaFADSm preincubated with ATP: Mg 2+, 5.6∙10 −6 and 7.8∙10 −6 μmol substrate/min. cm 2, respectively. In addition, samples desorbed from the mica also showed rates for the synthesis of FAD in the same order than fresh CaFADSm; as example 0.24 (from RF) or 0.38 (from FMN) min −1 for a desorbed CaFADSm sample preincubated with ATP:Mg 2 + and exposed to the mica surface versus 0.38 min −1 for fresh CaFADSm in the transformation of RF into FAD (Figure SP1b), when assayed under the same experimental conditions. Therefore, immobilization on the mica does not appear to compromise the RFK and FMNAT activities of CaFADS.

D

t1:6

Table 1 Percentages of CaFADS oligomeric species detected by AFM in a purified CaFADS sample under two redox conditions in 20 mM PIPES, pH 6.0. Features correspond to image units, while molecules refer to the amount of individual protein monomers in the corresponding image units. Error associated to percentage determination was ±5–15%.

E

t1:1 t1:2 t1:3 t1:4 t1:5

5

Please cite this article as: C. Marcuello, et al., Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule..., Biochim. Biophys. Acta (2013), http://dx.doi.org/10.1016/j.bbapap.2012.12.013

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

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

C. Marcuello et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

E

D

P

R O

O

F

6

444

C

T

Fig. 3. AFM topography images of a CaFADSm sample. (a) 2D image of a monomeric feature in 20 mM PIPES, pH 6.0 and calculated height profile (corresponding to the blue line). (b) 2D image of a representative dimer in 20 mM PIPES, 2 mM DTT, pH 6.0 and calculated height profiles for the two protomers of the dimer. Profiles corresponding to the blue lines for protomers 1 and 2 are shown in black and red, respectively. The feature shows a head-tail interaction between the N-terminal and C-terminal modules of the two different protomers (N-2:C-1, shown in the red line), making a ~90° angle. Heights in the Z axis for the N- and C-terminal modules of each protomer are represented by N and C, respectively.

forms. These results for the FMNAT and FADpp activities are in agreement with the reversibility of these reactions and the equilibrium between them.

447

3.4. Oligomeric states of the FMNAT- and RFK- isolated CaFADS modules

448 449

The individually expressed FMNAT- and RFK-modules of CaFADS were similarly analyzed by AFM. Previous to immobilization, samples

t3:1 t3:2 t3:3 t3:4 t3:5 t3:6 t3:7

Table 3 Oligomeric ratios detected upon incubation of purified CaFADSm with different ligands by AFM. CaFADSm (0.5 μM) was prepared in 20 mM PIPES, pH 6.0, with 0.8 mM MgCl2 and 2 mM DTT. When ADP, ATP or PPi were used they were added at 250 μM, while FAD, FMN and RF were added at 50 μM. Features correspond to image units, while molecules refer to the amount of individual protein monomers in the corresponding image units. Error associated to percentage determination was ±5–15%.

R

R

O

C

N

Ligands

t3:9

t3:10 t3:11 t3:12 t3:13 t3:14 t3:15 t3:16 t3:17 t3:18 t3:19 t3:20 t3:21

FAD FMN RF ADP ATP ADP, FMN

Units

U

t3:8

E

445 446

Features Molecules Features Molecules Features Molecules Features Molecules Features Molecules Features Molecules

Association state Monomers (%)

Dimers (%)

Tetramers (%)

Amorphous hexamer (%)

Dimer of trimers (%)

87 73 100 100 82 66 52 21 50 18 30 9

10 17 – – 15 24 20 16 12 9 28 17

3 10 – – 3 10 8 13 13 19 2 2

– – – – – – 20 50 – – – –

– – – – – – – – 25 54 40 72

of the FMNAT-module were centrifuged at 3800 g to remove the observed turbidity in order to avoid aggregation of unfolded molecules. Nevertheless, AFM images of the clarified FMNAT-module samples exhibited the formation of large aggregates of up to 30–50 nm, independently of the dilution (Fig. 6a). This indicated that the FMNAT-module of CaFADS tents to form aggregates when independently expressed. The RFK-module presented an initial problem for its absorption on the mica surface, due to its lower isoelectric point (theoretical pI 4.8), that was solved by lowering the pH of the buffer up to 5.0 to neutralize the protein negative charges and to promote its adsorption on the mica negative surface. Previous to immobilization, samples of the RFK-module were centrifuged to remove turbidity. AFM images for the RFK-module showed features that correspond to monomers and dimers, being both species in similar percentage (Fig. 6b). Circular dichroism spectra confirmed that the supernatant consisted of folded RFK-module, although no transformation of RF into FMN was observed at pH 5.0. All together, these data indicate that under the assayed conditions the FMNAT-module cannot exists as a single soluble protein in solution, while folding of the RFK-module is stable under such conditions.

450

3.5. In vitro detection of CaFADS oligomeric forms

470

Additional methodologies were also used to further analyze the in vitro formation of CaFADS ensembles in solution. As shown in Lane 1 of Fig. 7a, the just purified CaFADS sample showed three broad bands when analyzed by N-PAGE, while CaFADSm (Lane 2 in Fig. 7a) exhibited a single sharp band that apparently corresponds to the one migrating faster in the CaFADS sample. Therefore, this band was related with CaFADSm, while the other two bands observed in the CaFADS

471 472

Please cite this article as: C. Marcuello, et al., Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule..., Biochim. Biophys. Acta (2013), http://dx.doi.org/10.1016/j.bbapap.2012.12.013

451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469

473 474 475 476 477

C. Marcuello et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

7

R O

P

D

E

T

490

C

488 489

E

486 487

R

484 485

R

482 483

mobility appear more intense in the presence of Mg2+ (Lane 5 Fig. 7a). Incubation with ATP, either in the absence or presence of Mg2+, induced the appearance of a higher MW band (Lanes 4 and 6 in Fig. 7a). Under these conditions the band apparently corresponding to the monomer was wider, suggesting it might include more than one conformation of this state. It is also worth mentioning that all these bands also show different migration speed upon incubation with FMN and ADP or only with ATP:Mg2+, suggesting the conformations of these oligomeric ensembles despite having similar stoichiometries might be different. Similarly, SDS-PAGE cannot be directly applied to the analysis of the MW of these transient oligomeric species, since forces involved in their production disappear under its denaturing conditions. Thus, for SDS-PAGE approaches similar samples were produced including in the incubation

N C O

480 481

sample were putatively assigned to higher association states. Unfortunately, since in N-PAGE migration of species depends on both size and shape, it cannot be directly related to molecular weight (MW) and identification of species with CaFADS dimers, trimers, tetramers or hexamers is not possible. Nevertheless, this result qualitatively agrees with different aggregation states also observed by AFM for this sample (Table 2). CaFADSm was also incubated with those ligands inducing larger formation of hexamers according to the AFM analysis: the mixture of FMN and ADP (products of the RFK activity) and ATP (substrate for both the RFK and FMNAT activities). Simultaneous incubation with ADP and FMN, either in the absence or presence of Mg2+, (Lanes 3 and 5 in Fig. 7a, respectively), produced a N-PAGE profile quite similar to that for the just purified CaFADS mixture and the bands with the slower

U

478 479

O

F

Fig. 4. AFM topography images of a CaFADSm sample upon incubation with different ligands. (a) 3D image for an organized tetramer observed upon incubation with RF. (b) 3D image for the amorphous hexamer observed upon incubation with ADP:Mg2+. (c) Representative 3D image of the features observed upon incubation with ATP:Mg2+; a monomer rounded in black, a dimer in red, a tetramer in brown and a dimer of trimers in yellow. All measurements in 20 mM PIPES, 2 mM DTT, pH 6.0.

Fig. 5. AFM topography images of a CaFADSm sample upon simultaneous incubation with the products of the RFK reaction, ADP:Mg2+ and FMN. (a) 2D and 3D images of a single organized hexamer. (b) Topographic profiles numbered from 1 to 3 for the indicated blue lines. 1, 2 and 3 correspond to the black, red and blue profiles, respectively. (c) Superposition of the crystallographic CaFADS hexamer on the 2D AFM topography image. Measurements in 20 mM PIPES, 2 mM DTT, pH 6.0.

Please cite this article as: C. Marcuello, et al., Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule..., Biochim. Biophys. Acta (2013), http://dx.doi.org/10.1016/j.bbapap.2012.12.013

491 492 493 494 495 496 497 498 499 500 501 502 503

8

Association state Monomers (%)

Dimers (%)

Tetramers (%)

Amorphous hexamer (%)

Dimer of trimers (%)

Features Molecules Features Molecules Features Molecules Features Molecules Features Molecules Features Molecules Features Molecules Features Molecules Features Molecules

62 40 70 42 90 82 60 38 52 20 75 43 80 62 65 32 72 38

28 35 18 21 10 18 30 37 10 7 10 11 15 23 13 12 10 10

10 25 8 19 – – 10 25 15 22 5 11 5 15 7 13 3 6

– – 2 7 – – – – – – 3 10 – – – –3 9

– – 2 11 – – – – 23 51 7 25 – – 15 43 12 37

517 518 519 520 521 522 523 524 525 526

FAD PPi

b30 s 10 min 16 h

of Gaussians indicated a main component with a retention volume of 14.7 ml, and considerably smaller amounts of three additional peaks at 10.5 ml, 12.0 ml and 13.5 ml (Figure SP1 and Table SP1). The estimated MW for the main component was ~ 50 kDa, a value slightly higher than the theoretic one for the CaFADS monomer and to the estimated by SDS-PAGE. Such differences might be explained by the fact that the protein is elongated, as well as to the cross-linker stabilizing different conformers. MW estimated for the peaks eluting at 12.0 ml and 10.5 ml correspond to 3- and 6-folds, respectively, that assigned to the monomer: therefore, they may be correlated with the trimer and the hexamer, respectively. The peak at 13.5 ml corresponds to a MW slightly lower than twice the estimated for the monomer, being this MW close to the real expected value for the dimer. This might be easily explained thinking in the production of a cross-linked dimer whose overall shape will be more globular than those of the rest of species, and therefore behaving more similar to the standard proteins used in the calibration of the column. Similar analyses were carried out on CaFADSm samples pre-incubated with FMN, ADP and BS3, as well as with ATP and BS3. Fitting of the chromatograms produced the same components as for CaFADSm, but changes in the relative proportions of the peaks were observed. The presence of ligands decreased the proportion of monomer, particularly for incubation with ADP and FMN, in favor of the trimer and hexamer species (Figure SP1 and Table SP1). It

T

mixture also the BS3 cross-linker to capture the oligomers formed in each particular case. Again, these experiments cannot be conclusive regarding MW, since cross-linking also produced intra-molecular bonds that prevent full unfolding of some regions of the protein, but they can provide an estimation of the number of species formed and of their sizes. Analysis of the CaFADSm sample by SDS-PAGE yielded a single band with an apparent MW of ~40–41 kDa (Lane 1 Fig. 7b), only slightly higher than the theoretically calculated from the aminoacid sequence (~38 kDa). Incubation of CaFADSm with the BS3 cross-linker produced lost in the definition of this band and its displacement to slightly lower MW values (~36–39 kDa) (Lane 2 Fig. 7b), suggesting formation of intramolecular covalent links of different nature within the monomer that prevent full protein denaturation. As a control, a SDS-PAGE of a ready purified CaFADS sample incubated with BS3, produced, additionally to the wide band related to the monomer, the appearance of at least three diffuse bands with apparent MW ~78, ~100 and ~140 kDa (arrows in Lane 3 Fig. 7b). All of them were also detected after incubation of CaFADSm either with FMN and ADP, or with ATP, in the presence of BS3, although with different relative proportions and slight displacements (Lanes 5 and 6 Fig. 7b). Some of these protein samples were further analyzed by gel filtration chromatography (Fig. 7e). Fitting of the chromatogram obtained for a sample of CaFADSm incubated with the BS3 cross-linker to a set

O

16 h

C

515 516

10 min

E

513 514

b30 s

R

511 512

ATP FMN

R

509 510

16 h

O

507 508

10 min

C

505 506

b30 s

N

504

ATP RF

U

t4:8 t4:9 t4:10 t4:11 t4:12 t4:13 t4:14 t4:15 t4:16 t4:17 t4:18 t4:19 t4:20 t4:21 t4:22 t4:23 t4:24 t4:25

R O

t4:7

F

Units

P

Ligands

D

t4:6

Table 4 Oligomeric ratios detected upon incubation of purified CaFADSm with the substrates for it RFK, FMNAT and FADpp activities at different reaction times. CaFADSm (0.5 μM) was prepared in 20 mM PIPES, pH 6.0, with 0.8 mM MgCl2 and 2 mM DTT. When ADP, ATP or PPi were used they were added at 250 μM, while FAD, FMN and RF were added at 50 μM. Features correspond to image units, while molecules refer to the amount of individual protein monomers in the corresponding image units. Error associated to percentage determination was ±5–15%.

E

t4:1 t4:2 t4:3 t4:4 t4:5

C. Marcuello et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

Fig. 6. 3D AFM topography images of the independently produced modules of CaFADS. (a) N-terminal FMNAT-module showing large aggregates and (b) C-terminal RFK module exhibiting a mixture of monomers and dimers. Measurements in 20 mM PIPES, 2 mM DTT, pH 6.0.

Please cite this article as: C. Marcuello, et al., Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule..., Biochim. Biophys. Acta (2013), http://dx.doi.org/10.1016/j.bbapap.2012.12.013

527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549

C. Marcuello et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

a

b

9

140 kDa 100

e

2

3

4

5

1

6

2

3

4

5

6

d

c

100 80 60 40 20 0 0

2

4

6

F

1

Absorbance (mAU)

78

8

10 12 14 16

2

3

1

2

3

4

5

R O

1

O

Volume (ml)

550 551

T

E

D

P

Fig. 7. In vitro and In vivo detection of CaFADS oligomeric species. (a) CaFADS species detected by N-PAGE upon incubation of 50 μM CaFADS or CaFADSm with the indicated combination of ligands in PIPES 20 mM, pH 7.0, during 30 min at 37 °C. When indicated ATP or ADP were used at 250 μM, FMN at 50 μM and Mg2+ at 10 mM. From left to right, Lane 1: ready purified CaFADS; Lane 2: CaFADSm; Lane 3: CaFADSm incubated with FMN and ADP; Lane 4: CaFADSm incubated with ATP; Lane 5: CaFADSm incubated with FMN and ADP: Mg2+; Lane 6: CaFADSm incubated with ATP:Mg2+. (b) CaFADS species detected by SDS-PAGE after incubation in phosphate buffer 50 mM, pH 8.0 with the same methodology as in (a) and followed by subsequent exposition to the BS3 cross-linker at 2 mM during 30 min at room temperature. Lane 1: CaFADSm; Lane 2: CaFADSm incubated with BS3; Lane 3: ready purified CaFADS incubated with BS3; Lane 4: MW markers (from bottom 30, 45, 66 and 97 kDa); Lane 5: CaFADSm incubated with FMN, ADP and BS3; Lane 6: CaFADSm incubated with ATP and BS3. (c) CaFADS species detected by WB analysis upon incubation with BS3 of crude extracts from E. coli over-expressing CaFADS and from C. ammoniagenes cells. Lane 1: E. coli cells over-expressing CaFADS; Lane 2: C. ammoniagenes cells; Lane 3: control ready purified CaFADS (500 nM) incubated with BS3. (d) CaFADS species detected by WB analysis upon incubation E. coli over-expressing CaFADS and C. ammoniagenes cells with the BSOCOES cross-linker. Lane 1: control of a ready purified CaFADS (500 nM) sample similarly incubated with BSOCOES; Lane 2: control of a CaFADSm sample (500 nM) similarly incubated with BSOCOES; Lane 3: control of the gel filtration purified CaFADS oligomer (500 nM) incubated with BSOCOES; Lane 4: E. coli cell over-expressing CaFADS upon incubation with BSOCOES; Lane 5: C. ammoniagenes incubated with BSOCOES. (e) Gel filtration chromatography of CaFADSm (at 50 μM) before (solid line) and after incubation with ATP (250 μM) (dotted line) and FMN (50 μM) plus ADP (250 μM) (striped line). All samples were incubated with the BS3 cross-linker before chromatography. Reactions involving cross-linkers were prepared in phosphate buffer 50 mM, pH 8.0 and stopped by the addition of TRIS-HCL 500 mM (10×), pH 8.0 up to 50 mM (quenching solution) and the incubation of additional 15 min. Control samples in the absence of cross-linkers were similarly treated.

559

3.6. In cell detection of CaFADS oligomeric forms

560

Our final goal was to identify whether in vivo the presence of some of the above mentioned CaFADS quaternary organizations was relevant by using two different approaches. On one side, crude extracts of E. coli over-expressing CaFADS and of C. ammoniagenes were immediately treated upon production with BS3. Then different components were separated by SDS-PAGE and the migration bands corresponding to the different CaFADS species identified by WB. E. coli over-expressing CaFADS and C. ammoniagenes extracts (Line 1 and Line 2, respectively, Fig. 7c) exhibited essentially the same bands identified in a purified CaFADS sample similarly incubated with the BS3 cross-linker (Line 3 Fig. 7c). Extracts from E. coli showed some additional weak bands that might be attributed to the fact that CaFADS antibodies were produced using a CaFADS sample produced in E. coli. The C. ammoniagenes extract also shows a weak additional band, with apparent MW below 78 kDa, which might be related with a protein dimer. According to the visual relative intensity of each band, the monomeric form appears as the main species in all the samples. To corroborate that the preparation of crude extracts did not alter the relative population of oligomeric CaFADS species within the living

561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579

E

R

R

556 557

N C O

554 555

U

552 553

C

558

is also worth noting than incubation with ATP slightly displaces the maximum of the monomer. This suggests conformational changes in the protein affecting its shape upon ATP binding, in agreement with unpublished studies that show reorganization of different superficial protein loops upon binding of adenine nucleotides. All together, and despite the limitations of these techniques in the identification of transient complexes (as the oligomers here produced), these results also confirm that the protein ligands induce the formation of trimers and hexamers in solution and agree with the AFM results.

cells a second approach was used. Both E. coli cells over-expressing CaFADS and C. ammoniagenes cells were incubated with a crosslinker able to cross membranes and get introduced in vivo into the cells, BSOCOES. Then, crude extracts were produced and analyzed by WB. Similar results to those of the previous experiment were observed (Fig. 7d). Again, extracts (Lines 4 and 5, Fig. 7d) exhibited essentially the same bands identified in different samples of purified CaFADS incubated with the BSOCOES cross-linker (Lines 1, 2 and 3 Fig. 7d). All together these results show the presence in vivo of CaFADS oligomeric species of similar nature than those found in vitro.

580 581

4. Discussion

591

Liquid JM AFM permits the visualization of CaFADS single isolated molecules, as well as the estimation of their molecular dimensions, opening new possibilities for the study of the association state of this enzyme. AFM and N-PAGE experiments confirmed that the ready purified enzyme was a mixture of species with different stoichiometries, while further purification by gel filtration produced a sample that was mainly in monomeric state (fully monomeric under reduction conditions). Therefore, this CaFADSm appeared as a good sample for comparative analyses, and even allowed to independently distinguishing between the two protein modules, the RFK- and FMNAT-modules (Figs. 2 and 3). When dimers were observed, AFM allowed the identification of a preferred head-to-tail contact between the C-terminal RFK-module of one protomer and the N-terminal FMNAT-module of the second protomer that apparently approaches these modules from different protomers in an orientation similar to that found in the trimers constituting the reported crystallographic hexamer [10]. These observations suggest dimers might allow incorporation of new molecules, being a possible starting association or a

592

Please cite this article as: C. Marcuello, et al., Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule..., Biochim. Biophys. Acta (2013), http://dx.doi.org/10.1016/j.bbapap.2012.12.013

582 583 584 585 586 587 588 589 590

593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609

633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651

F

O

R O

631 632

P

629 630

D

627 628

that bind at the RFK domain, event only produced for ATP/ADP when Mg2+ is present, must be the key event in the oligomerization process. For the vast majority of enzymes, the oligomerization processes and the role of the different oligomers have not been unraveled yet. In particular, the possibility of oligomerization has not been reported, suggested nor predicted in any of the prokaryotic FADSs so far characterized from other species. Our analysis of the available crystal structures for the FADSs from T. maritima (TmFADS, pdb 1MRZ, [13,41]) and Streptococcus pneumoniae (SpnFADS, pdb 3PO1) does not predict formation of any oligomeric association. Main differences of the CaFADS structure with other proteins of its family are located in the β1n strand of the FMNAT module (absent in TmFADS) and, particularly, in an insertion of 12 residues in the L3c (L232-V246) at the RFK module that is only present in the Corynebacterium and Mycobacterium species (Fig. 8) [14]. The position of this insertion, also lacking in monofunctional RFKs, is stabilized in the structure of the CaFADS monomer by H-bonds connecting L3c and L5c within the RFK module of the protein. In addition, L3c reinforces the interface between trimers in the formation of the hexamer. Thus, besides the electrostatic complementarity, the 14 H-bonds and 4 salt bridges stabilizing the dimer of trimers, the loop L3c additionally provides with 3 hydrophobic contacts (involving P237) per protomer between each one of the RFK modules of one trimer and residues from β1n (M1 and I3) and α3n (F78) at the FMNAT module of the protomer in contact from the second trimer (Fig. 8). The lack of L3c in other FADSs might prevent such stabilization, so this might be a particular feature contributing to the oligomerization of CaFADS. The organization of CaFADS into a dimer of trimers during catalysis is also supported by a recent site-directed mutagenesis study concluding that conformational changes at the RFK-module induced by ligand binding modulate the binding parameters and the catalytic efficiency at the FMNAT active site of CaFADS [9]. Additionally, it indicates that the Thr and Asn residues at the consensus active site 207-PTAN-210 motif of the RFK module are involved in binding of RF and ATP in the catalytic competent conformation to produce RF phosphorylation by CaFADS. Noticeably, while the conformation of the PTAN motif leaves an open cavity for ATP binding in the structures of monofunctional RFKs, TmFADS and SpnFADS, the organization of the Pro-Thr peptide bond at the RFK module of CaFADS closes the ATP binding site preventing its allocation [9,10]. Thus, in CaFADS conformational changes, not required in other members of the family, have been claimed, making the CaFADS mechanism for this activity structurally more complex than in other RFKs. The mutagenesis analysis of CaFADS additionally points to the

T

625 626

C

623 624

E

621 622

R

619 620

R

617 618

O

616

a

C

614 615

N

612 613

previous step, leading to the formation of the higher association stoichiometries detected by AFM, X-ray diffraction, solution assays and in cell extracts. Additionally we have here proven that neither oligomerization nor immobilization on the mica surface prevent the enzymatic activities of CaFADS. Investigation of the influence of ligands on the protein organization provided several interesting observations. FAD, but not FMN, induced formation of dimeric and well organized tetrameric ensembles (Fig. 4), despite both flavins preferentially bind at the N-terminal FMNAT under the assay conditions [9,12]. This might indicate that occupation of the adenine binding site at the FMNAT-module somehow induced oligomerization. On the other hand RF, lacking the adenine and phosphates of FAD but binding at the RFK- and FMNAT-modules, produces similar effects than FAD. Regarding adenine nucleotides, either ADP:Mg2+ or ATP induced formation of amorphous hexamers, while only ATP:Mg2+ or FMN:ADP:Mg2+ produced organized dimer of trimers resembling in shape and volume the CaFADS ensemble reported by X-ray crystallography (Table 3, Figs. 4, and 5) [10]. In general the in vitro techniques here presented agree with the AFM analysis (Fig. 7), despite differences in methodology and resolution. All together these data suggest that: i) some of the substrates of the RFK (RF or ATP:Mg 2 +), the FMNAT (ATP:Mg2+) and the FADpp (FAD) activities induce the organization of the enzyme in oligomeric ensembles; ii) the dimer and the amorphous hexamer might represent intermediate steps during the production or dissociation of the organized dimer of trimers; and iii) ATP: Mg 2+ and the products of the RFK activity are the main responsible to induce the organization of the packed dimer of trimers. Mixing of CaFADSm with the substrates for each one of its activities indicates oligomeric species dynamics upon turnover (Table 4), demonstrating association-dissociation processes taking place and, therefore, the formation of transient quaternary organizations. Maximal, and similar, amounts of the organized dimer of trimers were particularly observed upon mixing the enzyme with the substrates of its FMNAT and FADpp activities (Table 4). This is consistent with an equilibrium being established between both processes as forward and backward reactions at the same active site, and with the presence of ATP:Mg2+ either as substrate or product [11]. Therefore, along the catalytic processes the quaternary organizations might be structured by the previous formation of dimers, supported by RFK-FMNAT intermodule contacts of different protomers, followed by packing of additional protomers to finally build up the packed dimer of trimers. Careful analysis of the data additionally indicates that the stabilization of the phosphate groups of the ligands

U

610 611

C. Marcuello et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

E

10

F78 I3 P237

L3c

b

Protomer A

P237 I3 F78

Protomer B

Fig. 8. Cartoon representation of the interaction between protomers of different trimers in the CaFADS dimer of trimers (PDB code 2X0K). (a) Overall view of the interaction. One protomer from the upper trimer is coloured in pink and the one interacting with it in the lower trimer is in green. The N-terminal FMNAT and C-terminal RFK modules of these protomers are coloured in light and pale colours, respectively. The other four protomers from the hexamer are shown in grey. Residues of the loop L232-V246 from each protomer are shown in sticks and CPK coloured with carbons in yellow (for the upper protomer) and in blue (for the lower). Residues interacting with L3c from the FMNAT module of the neighbouring protomer are shown in sticks. (b) Detail of the interaction surfaces. L3c from each of the interacting protomers is shown in surface and coloured as in (a). Surfaces are also shown for the hydrophobic residues from the FMNAT modules at the interacting regions with L3c.

Please cite this article as: C. Marcuello, et al., Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule..., Biochim. Biophys. Acta (2013), http://dx.doi.org/10.1016/j.bbapap.2012.12.013

652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693

C. Marcuello et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

728

JM AFM in liquid has been used to further investigate the mechanism of the bifunctional FADS, allowing visualization in real time and at the single-molecule level of the association steps induced by ligand binding and catalysis without loss of the enzyme functionality. The images here presented support the dynamic transient formation of a compact dimer of trimers for CaFADS during catalysis, in agreement with previous hypothesis suggested by solution studies and X-ray crystallography. Besides, the presence of CaFADS oligomeric associations is here we also proven within the living cell. All these results indicate that the oligomeric states of CaFADS must have a physiological significance that will deserve further study.

729

Acknowledgements

730 731

734 735

This work has been supported by the Spanish Ministry of Science and Innovation [BIO2010-14983 to M.M.] and the Government of Aragón [MICINN-FEDER B-18]. A.L. thanks ARAID for financial support. C.M. and S. A.-L. are indebted to DGA and Spanish Ministry of Science, respectively, for their predoctoral fellowships. The authors thank Iñigo Echániz for technical support.

736

Appendix A. Supplementary data

737 738

Supplementary data to this article can be found online at http:// dx.doi.org/10.1016/j.bbapap.2012.12.013.

739

References

722 723 724 725 726 727

732 733

740 741 742 743 744 745 746 747 748 749 750 751 752 753 754

D

720 721

T

713 714

C

711 712

E

709 710

R

707 708

R

705 706

N C O

703 704

U

701 702

F

718 719

700

O

5. Conclusions

698 699

R O

717

696 697

[7] E.M. Torchetti, C. Brizio, M. Colella, M. Galluccio, T.A. Giancaspero, C. Indiveri, M. Roberti, M. Barile, Mitochondrial localization of human FAD synthetase isoform 1, Mitochondrion 10 (2010) 263–273. [8] A. Serrano, P. Ferreira, M. Martínez-Júlvez, M. Medina, The prokaryotic FAD Synthetase family: a potential drug target, Curr. Pharm. Des. (in press). [9] A. Serrano, S. Frago, B. Herguedas, M. Martínez-Júlvez, A. Velázquez-Campoy, M. Medina, Key residues at the riboflavin kinase catalytic site of the bifunctional Riboflavin kinase/FMN adenylyltransferase from Corynebacterium ammoniagenes, Chem. Biochem. Biophys. (2012), http://dx.doi.org/10.1007/s12013-012-9403-9. [10] B. Herguedas, M. Martinez-Julvez, S. Frago, M. Medina, J.A. Hermoso, Oligomeric state in the crystal structure of modular FAD synthetase provides insights into its sequential catalysis in prokaryotes, J. Mol. Biol. 400 (2010) 218–230. [11] I. Efimov, V. Kuusk, X. Zhang, W.S. McIntire, Proposed steady-state kinetic mechanism for Corynebacterium ammoniagenes FAD synthetase produced by Escherichia coli, Biochemistry 37 (1998) 9716–9723. [12] S. Frago, A. Velázquez-Campoy, M. Medina, The puzzle of ligand binding to Corynebacterium ammoniagenes FAD synthetase, J. Biol. Chem. 284 (2009) 6610–6619. [13] W. Wang, R. Kim, H. Yokota, S.H. Kim, Crystal structure of flavin binding to FAD synthetase of Thermotoga maritima, Proteins 58 (2005) 246–248. [14] S. Frago, M. Martínez-Júlvez, A. Serrano, M. Medina, Structural analysis of FAD synthetase from Corynebacterium ammoniagenes, BMC Microbiol. 8 (2008) 160. [15] S. Karthikeyan, Q. Zhou, A.L. Osterman, H. Zhang, Ligand binding-induced conformational changes in riboflavin kinase: structural basis for the ordered mechanism, Biochemistry 42 (2003) 12532–12538. [16] S. Bauer, K. Kemter, A. Bacher, R. Huber, M. Fischer, S. Steinbacher, Crystal structure of Schizosaccharomyces pombe riboflavin kinase reveals a novel ATP and riboflavin-binding fold, J. Mol. Biol. 326 (2003) 1463–1473. [17] V. Bafunno, T.A. Giancaspero, C. Brizio, D. Bufano, S. Passarella, E. Boles, M. Barile, Riboflavin uptake and FAD synthesis in Saccharomyces cerevisiae mitochondria: involvement of the Flx1p carrier in FAD export, J. Biol. Chem. 279 (2004) 95–102. [18] S. Garavaglia, I. D'Angelo, M. Emanuelli, F. Carnevali, F. Pierella, G. Magni, M. Rizzi, Structure of human NMN adenylyltransferase. A key nuclear enzyme for NAD homeostasis, J. Biol. Chem. 277 (2002) 8524–8530. [19] R.G. Zhai, M. Rizzi, S. Garavaglia, Nicotinamide/nicotinic acid mononucleotide adenylyltransferase, new insights into an ancient enzyme, Cell. Mol. Life Sci. 66 (2009) 2805–2818. [20] T. Izard, A. Geerlof, The crystal structure of a novel bacterial adenylyltransferase reveals half of sites reactivity, EMBO J. 18 (1999) 2021–2030. [21] G. Binnig, C.F. Quate, C. Gerber, Atomic force microscope, Phys. Rev. Lett. 56 (1986) 930–933. [22] D. Fotiadis, S. Scheuring, S.A. Muller, A. Engel, D.J. Muller, Imaging and manipulation of biological structures with the AFM, Micron 33 (2002) 385–397. [23] R. Seidel, C. Dekker, Single-molecule studies of nucleic acid motors, Curr. Opin. Struct. Biol. 17 (2007) 80–86. [24] D.J. Muller, Y.F. Dufrene, Atomic force microscopy as a multifunctional molecular toolbox in nanobiotechnology, Nat. Nanotechnol. 3 (2008) 261–269. [25] H. Jensenius, D.C. Klein, M. van Hecke, T.H. Oosterkamp, T. Schmidt, J.C. Jensenius, Mannan-binding lectin: structure, oligomerization, and flexibility studied by atomic force microscopy, J. Mol. Biol. 391 (2009) 246–259. [26] N.P. Barrera, J. Betts, H. You, R.M. Henderson, I.L. Martin, S.M. Dunn, J.M. Edwardson, Atomic force microscopy reveals the stoichiometry and subunit arrangement of the alpha4beta3delta GABA(A) receptor, Mol. Pharmacol. 73 (2008) 960–967. [27] S. Scheuring, J.N. Sturgis, Atomic force microscopy of the bacterial photosynthetic apparatus: plain pictures of an elaborate machinery, Photosynth. Res. 102 (2009) 197–211. [28] M.J. Martinez-Perez, R. de Miguel, C. Carbonera, M. Martinez-Julvez, A. Lostao, C. Piquer, C. Gomez-Moreno, J. Bartolome, F. Luis, Size-dependent properties of magnetoferritin, Nanotechnology 21 (2010) 465707. [29] A. Lostao, M.L. Peleato, C. Gomez-Moreno, M.F. Fillat, Oligomerization properties of FurA from the cyanobacterium Anabaena sp. PCC 7120: direct visualization by in situ atomic force microscopy under different redox conditions, Biochim. Biophys. Acta 1804 (2010) 1723–1729. [30] S.M. Carnally, J.M. Edwardson, N.P. Barrera, Imaging the spatial orientation of subunits within membrane receptors by atomic force microscopy, Methods Mol. Biol. 736 (2012) 47–60. [31] S.M. Carnally, H.S. Dev, A.P. Stewart, N.P. Barrera, M.X. Van Bemmelen, L. Schild, R.M. Henderson, J.M. Edwardson, Direct visualization of the trimeric structure of the ASIC1a channel, using AFM imaging, Biochem. Biophys. Res. Commun. 372 (2008) 752–755. [32] N.P. Barrera, H. Ge, R.M. Henderson, W.J. Fitzgerald, J.M. Edwardson, Automated analysis of the architecture of receptors, imaged by atomic force microscopy, Micron 39 (2008) 101–110. [33] C. Marcuello, R. de Miguel, C. Gomez-Moreno, M. Martinez-Julvez, A. Lostao, An efficient method for enzyme immobilization evidenced by atomic force microscopy, Protein Eng. Des. Sel. 25 (2012) 715–723. [34] R.C. Rodrigues, C. Ortiz, A. Berenguer-Murcia, R. Torres, R. Fernandez-Lafuente, Modifying enzyme activity and selectivity by immobilization, Chem. Soc. Rev. (2012), http://dx.doi.org/10.1039/C2CS35231A. [35] P. de Pablo, J. Colchero, J. Gómez-Herrero, A.M. Baró, Jumping mode scanning force microscopy, Appl. Phys. Lett. 73 (1998) 3300. [36] J. Sotres, A. Lostao, C. Gomez-Moreno, A.M. Baro, Jumping mode AFM imaging of biomolecules in the repulsive electrical double layer, Ultramicroscopy 107 (2007) 1207–1212. [37] I. Horcas, R. Fernandez, J.M. Gomez-Rodriguez, J. Colchero, J. Gomez-Herrero, A.M. Baro, WSXM: a software for scanning probe microscopy and a tool for nanotechnology, Rev. Sci. Instrum. 78 (2007) 013705.

P

715 716

side-chain of T208 as a key residue that needs to turn to coordinate the metal ion bound to the adenine nucleotide phosphates [9,42,43]. Lack of Mg2+ and replacement of T208 have been reported to prevent ATP binding at the RFK [9]. Since our AFM results also indicate that the presence of the metal is critical to produce the dimer of trimers, a relationship might be envisage among both the L3c loop and the conformational changes required at the PTAN motif with the production of oligomeric forms along the CaFADS catalytic cycle. The observations here presented, together with previous studies [9,10], suggest a highly regulated oligomerization mechanism. This supports the hypothesis that, at least in CaFADS, the dimer of trimers must play a functional role during catalysis and that channeling between the RFK site and the FMNAT site of different protomers might take place. As described for several NTs that also stabilize hexameric organizations [18–20], the CaFADS dimer of trimers might contribute to regulation of flavin homeostasis. This might be an alternative regulation way to the cellular compartmentalization of different isoforms found in eukaryotes to maintain flavin homeostasis and flavoprotein biogenesis [7,17,44]. Nevertheless, it remains for future investigations to describe the functional role of the CaFADS quaternary organization, as well as to describe whether this might be a general behavior for prokaryotic FADSs or, on the contrary, a particular feature for the FADSs from Corynebacterium and, also probably, Mycobacterium species.

E

694 695

11

[1] S.S. Lee, D.B. McCormick, Effect of riboflavin status on hepatic activities of flavin-metabolizing enzymes in rats, J. Nutr. 113 (1983) 2274–2279. [2] I. Yruela, S. Arilla-Luna, M. Medina, B. Contreras-Moreira, Evolutionary divergence of chloroplasts FAD synthetase proteins, BMC Evol. Biol. 10 (2010) 311. [3] D.J. Manstein, E.F. Pai, Purification and characterization of FAD synthetase from Brevibacterium ammoniagenes, J. Biol. Chem. 261 (1986) 16169–16173. [4] C. Huerta, D. Borek, M. Machius, N.V. Grishin, H. Zhang, Structure and mechanism of a eukaryotic FMN adenylyltransferase, J. Mol. Biol. 389 (2009) 388–400. [5] C. Brizio, M. Galluccio, R. Wait, E.M. Torchetti, V. Bafunno, R. Accardi, E. Gianazza, C. Indiveri, M. Barile, Over-expression in Escherichia coli and characterization of two recombinant isoforms of human FAD synthetase, Biochem. Biophys. Res. Commun. 344 (2006) 1008–1016. [6] E.M. Torchetti, F. Bonomi, M. Galluccio, E. Gianazza, T.A. Giancaspero, S. Lametti, C. Indiveri, M. Barile, Human FAD synthase (isoform 2): a component of the machinery that delivers FAD to apo-flavoproteins, FEBS J. 278 (2011) 4434–4449.

Please cite this article as: C. Marcuello, et al., Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule..., Biochim. Biophys. Acta (2013), http://dx.doi.org/10.1016/j.bbapap.2012.12.013

755 756 757 758 Q2 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840

12

841 842 843 844 845 846 847 848 849 850

C. Marcuello et al. / Biochimica et Biophysica Acta xxx (2013) xxx–xxx

[38] D.J. Muller, A. Engel, M. Amrein, Preparation techniques for the observation of native biological systems with the atomic force microscope, Biosens. Bioelectron. 12 (1997) 867–877. [39] K. Meinander, T.N. Jensen, S.B. Simonsen, S. Helveg, J.V. Lauritsen, Quantification of tip-broadening in non-contact atomic force microscopy with carbon nanotube tips, Nanotechnology 23 (2012) 405705. [40] A. Serrano, S. Frago, A. Velázquez-Campoy, M. Medina, Role of key residues at the flavin mononucleotide (FMN): adenylyltransferase catalytic site of the bifunctional riboflavin kinase/flavin adenine dinucleotide (FAD) synthetase from Corynebacterium ammoniagenes, Int. J. Mol. Sci. 13 (2012) 14492–14517.

[41] W. Wang, R. Kim, J. Jancarik, H. Yokota, S.H. Kim, Crystal structure of a flavin-binding protein from Thermotoga maritima, Proteins 52 (2003) 633–635. [42] S. Karthikeyan, Q. Zhou, F. Mseeh, N.V. Grishin, A.L. Osterman, H. Zhang, Crystal structure of human riboflavin kinase reveals a beta barrel fold and a novel active site arch, Structure 11 (2003) 265–273. [43] S. Cheek, K. Ginalski, H. Zhang, N.V. Grishin, A comprehensive update of the sequence and structure classification of kinases, BMC Struct. Biol. 5 (2005) 6. [44] M.L. Pallotta, C. Brizio, A. Fratianni, C. De Virgilio, M. Barile, S. Passarella, Saccharomyces cerevisiae mitochondria can synthesise FMN and FAD from externally added riboflavin and export them to the extramitochondrial phase, FEBS Lett. 428 (1998) 245–249.

U

N

C

O

R

R

E

C

T

E

D

P

R O

O

F

862

Please cite this article as: C. Marcuello, et al., Detection of a quaternary organization into dimer of trimers of Corynebacterium ammoniagenes FAD synthetase at the single-molecule..., Biochim. Biophys. Acta (2013), http://dx.doi.org/10.1016/j.bbapap.2012.12.013

851 852 853 854 855 856 857 858 859 860 861