downloaded from //aem.asm.org/content/aem/early/2013/10/14/aem.03075-13.full.pdf · 102 the succd...
TRANSCRIPT
1
1
2
Novel characteristics of succinate-CoA ligases: conversion of malate to malyl-CoA and 3
CoA-thioester formation of succinate analogues in vitro 4
5
6
Johannes Christoph Noltea, Marc Schürmann
a, Catherine-Louise Schepers
a, Elvira 7
Vogela, Jan Hendrik Wübbeler
a, and Alexander Steinbüchel
a,b* 8
9
Institut für Molekulare Mikrobiologie und Biotechnologie, Westfälische Wilhelms-Universität 10
Münster, D-48149 Münster, Germanya, and Environmental Sciences Department, King 11
Abdulaziz University, Jeddah, Saudi Arabiab 12
13
14
Running title: 15
Characterization of succinate-CoA ligases 16
17
18
19
20
21
22
*Corresponding author. Mailing address: Institut für Molekulare Mikrobiologie und 23
Biotechnologie, Westfälische Wilhelms-Universität, Corrensstrasse 3, D-48149 Münster, 24
Germany. Phone: +49-251-8339821. Fax: +49-251-8338388. E-mail: steinbu@uni-25
muenster.de. 26
27
AEM Accepts, published online ahead of print on 18 October 2013Appl. Environ. Microbiol. doi:10.1128/AEM.03075-13Copyright © 2013, American Society for Microbiology. All Rights Reserved.
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
2
Abstract 28
29
Three succinate-CoA ligases (SucCD) from Escherichia coli, Advenella 30
mimigardefordensis DPN7T and Alcanivorax borkumensis SK2 were characterized regarding 31
their substrate specificity concerning succinate analogues. Previous studies had suggested that 32
SucCD enzymes might be promiscuous towards succinate analogues such as itaconate and 3-33
sulfinopropionate (3SP). The latter is an intermediate of the degradation pathway of 3,3’-34
dithiodipropionate (DTDP), a precursor for biotechnical production of polythioesters (PTE) in 35
bacteria. The sucCD genes were expressed in E. coli BL21 (DE3) pLysS. SucCD of E. coli 36
and of A. mimigardefordensis DPN7T were purified in native state using stepwise purification 37
protocols while SucCD from A. borkumensis SK2 was equipped with a C-terminal 38
hexahistidine tag at the SucD subunit. Beside the preference for the physiological substrates 39
succinate, itaconate, ATP and CoA, high enzyme activity was additionally determined for 40
both enantiomeric forms of malate amounting to 10-21% of the activity with succinate. Km-41
values ranged from 2.5 to 3.6 mM for L-malate and from 3.6 to 4.2 mM for D-malate for the 42
SucCDs investigated in this study. As L-malate-CoA ligase is present in the serine cycle for 43
assimilation of C1-compounds in methylotrophs, structural comparison of these two enzymes 44
as members of the same sub-subclass suggested a strong resemblance of SucCD with L-45
malate-CoA ligase and gave rise to the speculation that malate-CoA ligases and succinate-46
CoA ligases have the same evolutionary origin. Although enzyme activities were very low for 47
the additional substrates investigated, LC/ESI-MS analyses proved the ability of SucCDs to 48
form CoA-thioesters of adipate, glutarate and fumarate. Since all SucCDs were able to 49
activate 3SP to 3SP-CoA, we consequently demonstrate that the activation of 3SP is not a 50
unique characteristic of the SucCD from A. mimigardefordensis DPN7T. The essential role of 51
sucCD in the activation of 3SP in vivo was proved by genetic complementation. 52
53
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
3
Keywords: Advenella mimigardefordensis DPN7T, Alcanivorax borkumensis SK2, L-malate, 54
malate-CoA ligase, malyl-CoA, malyl-CoA synthetase, succinate-CoA ligase, succinyl-CoA, 55
succinyl-CoA synthetase 56
57
Abbreviations: Abo, Alcanivorax borkumensis SK2; Am, Advenella mimigardefordensis 58
DPN7T; Ap, Ampicillin; ATCC, American Type Culture Collection; BL21, Escherichia coli 59
BL21; Cm, chloramphenicol; CoA, coenzyme A; CAS, Chemical Abstracts Service number; 60
EC, enzyme commission; DTDP, 3,3’-dithiodipropionate; GC/MS, gas 61
chromatography / mass spectrometry; Gm, gentamicin; (HP)LC/ESI-MS, high performance 62
liquid chromatography - electro spray ionisation mass spectrometry; His, hexahistidine tag; 63
kcat, turnover number; Km, Michaelis-Menten constant; LB, Lysogeny broth medium; MSM, 64
mineral salt medium; MtkAB, malate thiokinase/malate-CoA ligase/malyl-CoA synthetase; 65
m/z, mass to charge ratio; Ni-NTA, nitrilotriacetic acid; PTE, polythioester; SDS-PAGE, 66
sodium dodecylsulfate acrylamide gel electrophoresis; SucC, subunit ȕ of SucCD; SucD, 67
subunit Į of SucCD; SucCD, succinate thiokinase/succinate-CoA ligase/succinyl-CoA 68
synthetase; 3SP, 3-sulfinopropionate 69
70
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
4
INTRODUCTION 71
Succinyl-CoA synthetases (succinate-CoA ligase; SucCD, EC 6.2.1.4 and 6.2.1.5) 72
catalyze the reversible conversion of succinyl-CoA to succinate under the concomitant 73
formation of a nucleoside triphosphate (NTP) in the citric acid cycle (7, 26). The enzyme 74
consists of two different subunits forming a heterodimer or a heterotetramer structure (8, 53). 75
The Į-subunit (SucD) and the ȕ -subunit (SucC) have molecular masses of 29-34 kDa and 41-76
45 kDa, respectively (48). The ȕ-subunit is responsible for the binding of the NTP, whereas 77
the Į-subunit binds CoA (23). So far, the binding site for the substrate succinate has not been 78
located however, it is assumed that it occurs at the dimer interface (23, 30, 37, 51). A 79
conserved histidine residue of the Į-subunit is phosphorylated during catalysis. The phosphate 80
moiety is conferred to a NDP to yield a NTP. Substitution of the histidine residue with any 81
other amino acid residue results in an inactive enzyme (8, 30). The reverse reaction leading 82
from succinate to succinyl-CoA is important in the reductive citric acid cycle in many 83
bacteria, as well as part of heme biosynthesis and ketone body activation in higher organisms 84
(20). 85
Although much interest was devoted to the structure (15, 23, 52), function, regulation 86
(4) and nucleotide specificity (25, 34), only little is known about the substrate range 87
concerning carbon acids. Among the more than 30 members of the sub-subclass acid CoA 88
ligases (EC 6.2.1) miscellaneous descriptions about extended ranges concerning the acid 89
substrates for the enzymes have been published. The greatest flexibility show long-chain fatty 90
acid CoA ligases (EC 6.2.1.3, (36)), but also acetate-CoA ligases (6, 29), a propionate-CoA 91
ligase (38) and a butyrate-CoA ligase (45) were shown to react with more than one carbon 92
acid substrate. In 2007 the succinate-CoA ligase from the hyperthermophilic archaeon 93
Thermococcus kodakaraensis was described which structurally resembles the acetate-CoA 94
ligase from Pyrococcus furiosus (44). This enzyme exhibits a subunit domain distribution 95
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
5
which is different from the Įȕ-heterodimer/-tetramer structure as typical for SucCDs. This 96
enzyme showed an extended substrate range and was also active with isovalerate, 3-methyl 97
thiopropionate, glutarate, adipate and butyrate. However, for succinate-CoA ligases with 98
classical domain structure relevant investigations are still missing. In 1957 it was reported 99
about the formation of itaconyl-CoA from itaconate, a structural analogue to succinate, in 100
mammalian liver mitochondria catalyzed by SucCD (1). Later, this reaction was proved for 101
the SucCD from Micrococcus sp. and Pseudomonas fluorescens (1, 12, 35). 102
In 2011, Schürmann et al. showed the participation of the SucCD of Advenella 103
mimigardefordensis DPN7T in the degradation pathway of 3,3´-dithiodipropionic acid 104
(DTDP), a precursor for the production of polythioesters in bacteria (28, 43, 55, 58). In this 105
strain DTDP is metabolized via the intermediate product 3-sulfinopropionate (3SP) (54, 56, 106
57). This xenobiotic structural analogue to succinate carries a sulfino group instead of the 107
carboxyl group in succinate, and it is converted to 3-SP-CoA in vivo (43). The authors also 108
showed that the mutant strain A. mimigardefordensis DPN7T ǻsucCD was no longer able to 109
grow on DTDP and 3SP. Furthermore, Schürmann et al. demonstrated the conversion of 110
itaconate to itaconyl-CoA by SucCDAm in vitro. In addition to that, authors observed the 111
formation of 3SP-CoA by crude extract of the expression strain E. coli BL21 (DE3) pLysS 112
not harbouring genes for SucCDAm (43). These findings suggested that the formation of 3SP-113
CoA from 3SP is not a unique characteristic of the A. mimigardefordensis DPN7T SucCD, and 114
it raised the question, whether SucCDs in general have an extended substrate range like other 115
members in this enzyme sub-subclass 6.2.1. and whether other SucCD enzymes are also able 116
to form itaconyl-CoA and 3SP-CoA. 117
In this study, we purified three homologous SucCDs and characterized these enzymes 118
with regard to their ability to convert different carbon acid substrates as analogues of 119
succinate to their corresponding CoA-thioesters in vitro. This included the SucCDs from 120
Escherichia coli BL21 (SucCDBL21), A. mimigardefordensis DPN7T (SucCDAm) and 121
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
6
Alcanivorax borkumensis SK2 (SucCDAboHis). The SucCDBL21 is identical to the E. coli K-12 122
enzyme (9) (ATCC® 47076™) and has been of scientific interest for the last 60 years. The 123
enzyme is able to use ATP as well as GTP as cosubstrate (34). Several crystallographic 124
structures proved the location of binding domains for the nucleotide and the CoA involved in 125
catalysis (15, 24). A. mimigardefordensis is a bacterium that can grow on the sulphur-126
containing precursor DTDP and is a suitable host for polythioester production (58). As 127
mentioned above, the activation of itaconate, as well as of 3SP, was shown for SucCDAm (43). 128
A. borkumensis SK2 is a marine Ȗ-proteobacterium whose genome was published in 2006 129
(41). The SucCDAbo has not been under investigation, yet. Alignments of primary structures 130
show high sequence similarity among the SucCDs investigated in this study. 131
As there are no data on extended substrate ranges of SucCD except from itaconate for 132
mammalian SucCDs and itaconate and 3SP for SucCDAm, we aimed at using substrate 133
compounds that differ from succinate with respect to their carbon acid backbone, their chain-134
length or their side chain; in addition, we analyzed monocarboxylate compounds (Fig. 1). We 135
also investigated the SucCD-dependent activation of the side-chain variant malate to malyl-136
CoA, and we analyzed the relation between our findings of malate-CoA ligase activity 137
exhibited by succinate-CoA ligases in this study and further reports about the participation of 138
a L-malate-CoA ligase in the serine cycle in some methylotrophic bacteria. 139
140
141
MATERIALS AND METHODS 142
Bacterial strains and cultivation conditions. All strains used in this study are listed 143
in Table 1. Cells of A. mimigardefordensis DPN7T and A. mimigardefordensis DPN7T 144
ǻsucCD were cultivated at 30 °C in mineral salt medium (MSM) (40) containing 20 mM 145
gluconate, 50 mM DTDP or 20 mM 3SP as sole source of carbon and energy. Carbon sources 146
were added to MSM from filter-sterilized stock solutions as indicated in the text. For 147
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
7
maintenance of plasmids, solutions of antibiotics were prepared according to the method of 148
Sambrook et al. (39) at the following concentrations: 100 µg/ml ampicillin (cultivation of 149
strains harbouring pET-23a(+) vector system), 150 µg/ml (cultivation of strains harbouring 150
pBluescriptSK(-) vector system), and 34 µg/ml chloramphenicol. 151
Strains of E. coli were cultivated in Lysogeny Broth (LB) medium (39) or ZYP-5052 152
complex medium for autoinduction according to the method of Studier (49). The latter was 153
used for homo- and heterologous expression of genes in E. coli under the control of the lac 154
promoter and T7-promotor, respectively. For expression of sucCD a single colony of the 155
expression strain harbouring expression plasmid was used to inoculate a pre-culture of 50 ml 156
LB medium in a baffled flask containing antibiotics which was then incubated at 105 rpm at 157
30 °C. After 6 to 9 hours, the main culture containing 500 ml ZYP autoinduction medium in a 158
2.5 l baffled flask with antibiotics, was inoculated with 2% (vol/vol) of the pre-culture and 159
incubated for 36-48 h at 105 rpm and 30 °C. For complementation experiments with A. 160
mimigardefordensis DPN7T strains, pre-cultures grown in presence of 20 mM gluconate were 161
used to inoculate the main cultures (50 mM DTDP) resulting in an optical density (600 nm) of 162
0.1. 163
Chemicals. D-malic acid of high purity grade (CAS: 636-61-3) was purchased from 164
Alfa Aeser (Karlsruhe, Germany), L-malic acid of high purity grade (CAS: 97-67-6) was 165
purchased from Applichem (Darmstadt, Germany). DTDP of high purity grade was purchased 166
from Sigma Aldrich (Steinheim, Germany). 3SP was synthesized according to Jollès-Bergeret 167
(21); the procedure was modified by one repetition of the step for alkaline cleavage of the 168
intermediate bis-(2-carboxyethyl)sulfone as described earlier (54). Synthesis and purity of the 169
substance were confirmed by GC and GC/MS. According to GC/MS the purity of the used 170
3SP was at 98.7% for qualitative and quantitative enzyme assay and at least 95% when used 171
as carbon and energy source in MSM. 172
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
8
Analysis of 3SP by GC or GC/MS. For purity analysis the synthesized 3SP was 173
analyzed by GC or GC/MS as described previously (43). For this, 3SP was subjected to 174
methylation in presence of 1 ml chloroform, 0.850 ml methanol, and 0.150 ml sulfuric acid 175
for 2 h at 100 °C. Upon methylation, 2 ml H2O were added, and the samples were vigorously 176
shaken for 30 s. After phase separation, the organic layer containing the resulting methyl 177
esters of the organic acids was analyzed in an HP6850 gas chromatograph equipped with a 178
BP21 capillary column (50 m by 0.22 mm; film thickness, 250 nm; SGE, Darmstadt, 179
Germany) and a flame ionization detector. Helium was used as carrier gas at a flow-rate of 180
0.6 ml/min. The temperatures of the injector and detector were 250 and 240°C, respectively. 181
Identification of peaks was performed by using the AMDIS software in combination with the 182
NIST database (47). 183
Analysis of CoA-thioester formation by LC/ESI-MS. CoA-thioesters, which were 184
formed during enzyme assays, were monitored by HPLC in combination with mass 185
spectrometry (LC/ESI-MS) as described previously (43). LC/ESI-MS analysis was carried out 186
with an UltiMate 3000 HPLC apparatus (Dionex GmbH, Idstein, Germany) connected 187
directly to a LXQ Finnigan (Thermo Scientific, Dreieich, Germany) mass spectrometer. An 188
Acclaim 120 C18 Reversed-Phase LC column (4.6 x 250 mm, 5 µm, with 120 Å pores served 189
for the separation of CoA-thioesters at 30 °C. The eluents used were an ammonium acetate 190
buffer (50 mM, pH 5.0) adjusted with acetic acid (eluent A) and 100% (vol/vol) methanol 191
(eluent B). Elution occurred at a flow rate of 0.3 ml/min. Ramping was performed as follows: 192
equilibration with 90% eluent A for 2 min before injection and 90 to 45% eluent A for 20 193
min, followed by holding for 2 min, and then a return to 90% eluent A within 5 min after 194
injection. Detection of CoA-thioesters occurred at 259 nm with a photodiode array detector. A 195
solution of 0.4 mM CoA served for tuning the instrument by direct infusion at a flow-rate of 196
10 µl/min into the ion source of the mass spectrometer to optimize the ESI-MS system for 197
maximum generation of protonated molecular ions (parents) of CoA derivatives. The 198
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
9
following tuning parameters were retained for optimum detection of CoA-thioesters: capillary 199
temperature, 300 °C; sheat gas flow, 12 liters/h; auxiliary gas flow, 6 liters/h; and sweep gas 200
flow, 1 liter/h. The mass range was set to m/z 50 to 1,000 Da when run in the scan mode. The 201
collision energy in the MS mode was set to 30 V and yielded fragmentation patterns that were 202
in good accordance with those found in other publications (13, 43). 203
Isolation and manipulation of DNA. Chromosomal DNA of A. mimigardefordensis 204
strain DPN7T was isolated according to the method of Marmur (32). Plasmid DNA was 205
isolated from E. coli using the peqGOLD plasmid miniprep kit I from PEQLAB 206
Biotechnologie GmbH (Erlangen, Germany) according to the manufacturer’s manual. DNA 207
was digested with restriction endonucleases (Fermentas GmbH, St. Leon-Rot, Germany) 208
under conditions described by the manufacturer. PCR were carried out in an Omnigene 209
HBTR3CM DNA thermal cycler (Hybaid, Heidelberg, Germany) or an PeqSTAR 2X 210
Gradient thermal cycler (PEQLAB Biotechnologie GmbH, Erlangen, Germany) using 211
Platinum© Taq DNA polymerase (Invitrogen, Carlsbad, USA) and Phusion High-Fidelity 212
DNA polymerase (Fermentas GmbH, St. Leon-Rot, Germany). T4-DNA-Ligase was 213
purchased from Fermentas (Fermentas GmbH, St. Leon-Rot, Germany). Primers were 214
synthesized by MWG-Biotech AG (Ebersberg, Germany) and are listed in Table 1. 215
Transfer of DNA. Competent cells of E. coli strains were prepared and transformed 216
by the CaCl2 procedure (39). Plasmids were transferred to A. mimigardefordensis DPN7T cells 217
by conjugation (46). 218
DNA sequencing and sequence data analysis. Sequence analysis was performed by 219
Seqlab (Göttingen, Germany). Sequences were analyzed using the program BLAST (National 220
Center for Biotechnology Information; http://www.ncbi.nlm.nih.gov/BLAST/) (2). 221
Cloning of sucCD genes for expression in E. coli BL21 (DE3) pLysS. The 222
corresponding sucCD genes were amplified from total genomic DNA of 223
A. mimigardefordensis strain DPN7T, E. coli BL21 and of A. borkumensis SK2 by PCR using 224
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
10
Phusion® High-Fidelity DNA Polymerase (New England Biolabs® GmbH, Frankfurt am 225
Main, Germany) or Biomix™ containing Taq DNA polymerase (Bioline GmbH, 226
Luckenwalde, Germany). Information on genomic sequences was obtained from the 227
Integrated Microbial Genomes nameplate (https://img.jgi.doe.gov/cgi-bin/er/main.cgi, (31)). 228
Oligonucleotides are listed in Table 1. Amplification of sucCDAm from genomic DNA of 229
A. mimigardefordensis DPN7T had been performed previously (43) using oligonucleotides 230
sucCDforward_PstI and sucCDreverse_XhoIstopp. Amplification of sucCDBL21(Genbank No. 231
P0A836 and P0AGE9), from genomic DNA of E. coli BL21 was performed by use of 232
oligonucleotides sucCDBL21_forward_EcoRI and sucCDBL21_reverse_HindIII and yielded 233
a fragment of 2171 bp. Amplification of the sucCD genes for the native form of SucCD from 234
A. borkumensis SK2 (locus tags ABO_1493 and ABO_1492) was performed by using 235
oligonucleotides sucCDAbo_forward_NdeI and sucCDAbo_reverse_SalI and gave a fragment 236
of 2041 bp. 237
PCR products were isolated from agarose gels using the peqGOLD GelExtraction Kit 238
(PEQLAB Biotechnologie GmbH, Erlangen, Germany), digested with appropriate restriction 239
enzymes provided in the primer name and ligated with digested pET-23a(+) (Novagen, 240
Madison, USA) or pBluescriptSK(-) (Stratagene, San Diego, USA) yielding 241
pBluescriptSK(-)::sucCDAm, pBluescriptSK(-)::sucCDBL21 and pET-23a(+)::sucCDAbo. 242
Ligation products were used for transformation of CaCl2 competent cells of E. coli 243
Top10, and transformants were selected on LB agar plates containing ampicillin. After that 244
the hybrid plasmids were isolated, analyzed by sequencing and used for transformation of 245
CaCl2 competent cells of E. coli BL21 (DE3) pLysS (New England Biolabs® Inc., Ipswich, 246
USA). Plasmid pET-23a(+)::sucCDAboHis was generated by PCR-based mutagenesis using 5´-247
phosphorylated oligonucleotides P_forward_XhoI_Histag_Abo and P_Abo_rev_mutagenesis 248
and pET-23a(+)::sucCDAbo as template. This PCR led to the deletion of the terminal stop 249
codon of sucD gene. After amplification, a ligation reaction was performed within the buffer 250
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
11
used for PCR reaction and the sample was used for transformation of CaCl2 competent cells 251
of E. coli Top10. 252
Construction of plasmid for complementation experiments. For complementation 253
studies in the broad host range vector pBBR1MCS-5 (27), sucCDAm and a 478-bp upstream 254
region were amplified by PCR using Phusion® High-Fidelity DNA Polymerase (New 255
England Biolabs® GmbH, Frankfurt am Main, Germany) and applying oligonucleotides 256
sucCDAm_Prom_fw and sucCDreverse_XhoI_stop (Table 1) (43). The PCR-product was 257
ligated into pJet1.2 blunt vector. After digestion using XhoI the gene fragment of 2541 bp was 258
extracted from agarose gel using peqGOLD GelExtraction Kit (PEQLAB Biotechnologie 259
GmbH, Erlangen, Germany) and ligated with pBBR1MCS5 which was before linearized with 260
XhoI restriction enzyme. The ligation products were transferred to CaCl2-competent cells of 261
E. coli S17-1 and E. coli Top10. The hybrid plasmid pBBR1MCS-5::sucCDAm was then 262
transferred into A. mimigardefordensis DPN7T ǻsucCD by conjugation (46). 263
Preparation of crude extracts. Cells from 50 to 500 ml cultures were harvested by 264
centrifugation (20 min, 4 °C, 4,000 x g) and stored at -20 °C until usage. Cells were 265
resuspended in 50 mM Tris-HCl buffer (pH 7.4) for purification of native SucCD or in 266
50 mM Tris-HCl, 500 mM NaCl and 20 mM imidazole (pH 7.4) for purification of the 267
hexahistine-tagged variant of SucCD. Cells were subsequently disrupted by applying a three-268
fold passage through a cooled French press (100 x 106 Pa) (Aminco, Silver Spring, MD) (33) 269
or a Sonoplus GM200 sonication apparatus (Bandelin, Berlin, Germany) equipped with a SH 270
213G boosterhorn and MS 72 or MS 73 microtip probes. The amplitude was 16 µm 271
(1 min/ml), while cooling was performed in a NaCl-ice bath. Soluble protein fractions of 272
crude extracts were obtained in the supernatants after 1.5 h of centrifugation at 100,000 x g 273
and 4 °C and were used for enzyme purifications. 274
Coupling of succinic acid anhydride to EAH-SepharoseTM
4B matrix. In order to 275
functionalize EAH-Sepharose 4B matrix with succinate as ligand, EAH-Sepharose 4B (GE 276
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
12
Healthcare, Munich, Germany) was dissolved in water and incubated under slight shaking at 277
4 °C. Solid succinic acid anhydride was added until insolubility was reached. The pH was 278
kept at 6 by addition of HCl during the reaction. Succinic acid anhydride was added stepwise. 279
The functionalization was performed over a period of three days. This method represents a 280
modification of the carbodiimide method according to manufacturer´s manual. 281
Purification of homo- and heterologously expressed sucCDs in native state. After 282
expression of sucCD in E. coli BL21 (DE3) pLysS, the cells were harvested and stored at 283
-20 °C until use. Cells were disrupted, and the soluble fraction was generated as described 284
above. All purification steps were performed at 4 °C. In case of SucCDAm and SucCDBL21 the 285
soluble fraction was applied to Q-Sepharose FF chromatography (32 ml; GE Healthcare, 286
Munich, Germany) according to Schürmann et al. (43). The matrix was equilibrated with 50 287
mM Tris-HCl (pH 7.4) and 0 mM NaCl at a flow-rate of 4 ml/min. The proteins were eluted 288
by a step gradient with increasing sodium chloride concentrations at a flow-rate of 4 ml/min 289
as follows: 0 to 20 min, 0 mM NaCl; 20 to 65 min, 50 mM NaCl; 65 to 110 min, 75 mM 290
NaCl; 110 to 155 min, 100 mM NaCl; and 155 to 210 min, 150 mM NaCl. The majority of 291
SucCDAm and SucCDBL21 eluted at 150 mM NaCl. In case of SucCDBL21 and SucCDAm the 292
eluted fractions were concentrated and buffered to binding conditions using ultrafiltration 293
(stirred cell model 8200, Amicon, Millipore Corporation, Billerica, Massachusetts, USA). An 294
ultracel regenerated cellulose membrane with a nominal molecular weight limit of 10 kDa 295
was used for protein concentration and buffer exchange to binding conditions for 296
chromatography. 297
Purification of SucCDAm. A protein solution containing enriched SucCDAm after Q-298
Sepharose chromatography (see previous section) was then applied to DEAE-Sepharose 299
column (27 ml; GE Healthcare, Munich, Germany) equilibrated to binding conditions (50 mM 300
Tris-HCl (pH 7.4)–0 mM NaCl at a flow-rate of 1 ml/min). Proteins were eluted by a linear 301
gradient with increasing NaCl concentration at a flow-rate of 1 ml/min as follows: 0 to 40 302
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
13
min, 0 mM NaCl, 40 to 110 min, linear gradient of 0 – 1 M NaCl (ǻ14.3 mmol/min), 110 to 303
160 min, 1 M NaCl. SucCDAm eluted at a concentration between 60 to 350 mM NaCl. The 304
eluted protein was concentrated, buffered to binding conditions and applied to a modified 305
EAH-Sepharose 4B column (25 ml; GE Healthcare, Munich, Germany) carrying a succinate 306
functionalization. Equilibration to binding condition was performed with 50 mM Tris-HCl 307
(pH 7.4) and 0 mM NaCl at a flow-rate of 1 ml/min. Proteins were eluted with the same 308
protocol as for DEAE-Sepharose chromatography. Impurities were eluted from the matrix, 309
pure SucCDAm was located in the flow-through. The latter was concentrated, buffered to 310
storage conditions according to Gibson et al. (16) (100 mM Tris-HCl, 150 mM NaCl, pH 7.5, 311
in 50% (vol/vol) glycerol) and used for enzyme assays. SucCD concentration was determined 312
using the specific molar extinction coefficient at 280 nm calculated for the Įȕ-dimer by 313
Protparam (50). 314
Purification of SucCDBL21. A protein solution containing enriched SucCDBL21 after 315
Q-Sepharose chromatography (see section Purification of homo- and heterologously 316
expressed sucCDs) was applied to Cibacron Blue F3GA column (50 ml; GE Healthcare, 317
Munich, Germany) equilibrated to binding conditions (50 mM Tris-HCl (pH 7.4) and 0 mM 318
NaCl at a flow-rate of 3 ml/min. The proteins were eluted by a linear gradient of sodium 319
chloride concentration at a flow-rate of 3 ml/min as follows: 0 to 40 min, 0 mM NaCl, 40 to 320
240 min, 0 to 1 M NaCl (ǻ5 mmol/min). SucCDBl21 eluted after 101 to 141 min according to 321
the purification protocol. Pooled samples were concentrated, buffered to storage conditions 322
(100 mM Tris-HCl, 150 mM NaCl, pH 7.5 in 50% (vol/vol) glycerol) and used for enzyme 323
assays. SucCD concentration was determined using the specific molar extinction coefficient at 324
280 nm calculated for the Įȕ-dimer by Protparam (50). 325
Purification of SucCDAboHis. In an initial attempt to purify SucCDAbo, it was 326
expressed from the vector pET-23a(+)::sucCDAbo (Table 1). Both subunits were expressed in 327
equimolar amounts as judged by SDS-PAGE but were repeatedly separated during 328
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
14
purification with Q-Sepharose. Hence, the vector pET-23a(+)::sucCDAboHis, encoding a C-329
terminal hexahistine-tag on SucD subunit, was generated (Table 1). After expression of 330
sucCDAboHis in E. coli BL21 (DE3) pLysS, cell harvest and cell disruption, the soluble fraction 331
was applied to Ni-NTA Sepharose column (HisTrap™ HP, 1 ml; GE Healthcare, Munich, 332
Germany). The SucD subunit was provided with a hexahistidine tag at the C-terminus. The 333
SucC subunit co-eluted from the column matrix. Binding conditions were 50 mM Tris-HCl, 334
500 mM NaCl and 20 mM imidazol (pH 7.4). Equilibration was performed following 335
manufacturer´s instruction. The flow-rate was 1ml/min. For elution purposes a gradient of 336
imidazol was applied. The elution program was the following: 0 to 5 min, 20 mM imidazol, 5 337
to 10 min, 40 mM imidazol, 10 to 50 min, 40 to 500 mM imidazole (ǻ11.5 mmol/min). 338
SucCD elution started at a concentration of 50 ± 5 mM imidazol. SucCD concentration was 339
determined as described previously by Bradford (5). 340
Enzyme assays. Standard in vitro activity of succinate-CoA ligase in direction of 341
ADP formation was assayed by a continuous spectrophotometric assay according to the 342
method of Cha (10). Measurements of succinate, itaconate, malate and 3SP were carried out at 343
30 °C in the presence of 50 mM Tris-HCl (pH 7.4), 0.4 mM ATP, 0.1 mM CoA, 344
7 mM MgCl2, 2 mM phosphoenolpyruvate, and 0.1 mM NADH, together with 6 U of 345
pyruvate kinase and 6 U of lactate dehydrogenase from rabbit muscle (Sigma) as auxiliary 346
enzymes. Concentrations of the organic acids were assayed in the range of 0.1 to 10 mM 347
(succinate), 0.3 to 10 mM (itaconate), 0.625 to 15 mM (malate) and 0.625 to 25 mM (3SP). 348
ATP and CoA measurements were carried out as described above in presence of 10 mM 349
succinate. The formation of ADP and a concomitant equimolar formation of the CoA-350
thioester were measured as a decrease in NADH absorption at 340 nm. The auxiliary enzymes 351
were tested to ensure that they were not rate limiting. The formation of the expected CoA-352
thioesters was verified by liquid chromatography-electrospray ionization-mass spectrometry 353
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
15
(LC/ESI-MS). For this analysis, the reactions were stopped by the addition of 30 µl of 15% 354
(w/v) trifluoroacetic acid. The samples were subsequently analyzed as described above. 355
The following compounds (each 10 mM) were investigated: succinate, sulfosuccinate, 356
mercaptosuccinate, itaconate, D-malate, L-malate, tartrate, acetate, butyrate, propionate, 357
levulinate, valerate, malonate, glutarate, adipate, 3SP, fumarate, maleate and 2,2´-358
thiodiacetate. 500 mM stock solutions of the respective compounds were prepared in 50 mM 359
Tris/HCl and were neutralized prior to application. 360
361
RESULTS 362
In silico analysis of SucCD enzymes. In this study, we characterized three different 363
bacterial SucCDs with regard to their substrate ranges concerning structural analogues to 364
succinate. An amino acid sequence alignment showed 100% sequence identity of the 365
SucCDBL21 and the E. coli K-12 enzyme. A multiple sequence alignment of SucC subunits 366
revealed the following: SucCBL21/SucCAm, 53% identical (72% similar) amino acid residues; 367
SucCBL21/SucCAbo, 74% identical (87% similar) amino acid residues and SucCAm/SucCAbo, 368
52% identical (71% similar) amino acid residues. For SucD subunits the following sequence 369
similarities were determined: SucDBL21/SucDAm, 55% identical (71% similar) amino acid 370
residues, SucDBL21/SucDAbo, 84% identical (90% similar) amino acid residues and 371
SucDAm/SucDAbo, 54% identical (73% similar) amino acid residues. The theoretical molecular 372
weights for SucCDBL21 is 41.4 kDa for SucC and 29.7 kDa for SucD subunit. Calculated 373
molecular weights of the A. mimigardefordensis DPN7T enzyme is 41.3 kDa for SucC and 374
30.9 kDa for SucD subunit. The A. borkumensis SK2 SucCD molecular weights were 375
calculated as 41.4 kDa (SucC) and 29.9 kDa (SucD), respectively. 376
Purification of SucCDs. The sucCD genes from E. coli BL21 and A. borkumensis 377
SK2 were amplified from genomic DNA and cloned into expression vectors yielding 378
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
16
pBluescriptSK(-)::sucCDBL21 and pET-23a(+)::sucCDAbo(His) as described in the material and 379
methods section. Optimal expression of all sucCDs in this study was achieved using 380
expression host E. coli BL21 (DE3) pLysS in ZYP 5052 autoinduction medium (49). Despite 381
a structural relation of at least 50% sequence identity, the SucCDs showed a quite distinct 382
binding behaviour on chromatography matrices resulting in three different purification 383
protocols (Table 2). Provision of a terminal hexahistidine tag for a more efficient purification 384
protocol for SucCDAm led to the formation of inclusion bodies. The amplified fragment of 385
sucCDAm, performed by Schürmann et al. (43), contained 68 bp of the sucC upstream region. 386
The fragment amplified from E. coli BL21 genomic DNA contained the bicistronic operon for 387
sucC plus sucD as well as the 135-bp region upstream of the sucC initiation codon. In case of 388
sucCDAboHis, the relevant fragment contained only the gene loci for sucC and sucD. The 389
sucCD genes investigated in this study were expressed independently in E. coli BL21 (DE3) 390
pLysS as soluble proteins. SucC and SucD were synthesized in almost equimolar amounts 391
according to SDS-PAGE analysis applying crude extracts and soluble protein fraction (not 392
shown). SucCDAm and SucCDBL21 were purified as native proteins using Q-Sepharose anion-393
exchange chromatography as a capture step. SucCDBL21 was further purified by Cibacron 394
Blue F3GA Sepharose affinity chromatography to electrophoretic homogeneity. Enriched 395
SucCDAm in Q-Sepharose eluate was further purified to electrophoretic homogeneity by 396
DEAE-Sepharose anion-exchange chromatography and by a final polishing step using 397
modified EAH-Sepharose 4B chromatography. SucCDAboHis, unlike SucCDBL21 and SucCDAm, 398
carried a hexahistidine tag at the C-terminus of SucD. The subunit SucC co-eluted with SucD 399
from the Ni-NTA matrix. The purity of the proteins was confirmed by applying 10 µg of 400
protein to SDS-PAGE (Fig. 2). 401
Carbon acid specificity of SucCDBL21, SucCDAm and SucCDAboHis. After expression 402
and purification of SucCDs, the enzyme activity was determined by using a continuous 403
spectrophotometric assay as described in the material and methods section. Several longterm-404
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
17
storage conditions were investigated for each SucCD. Optimal stability was observed with 405
100 mM Tris, 150 mM NaCl and storage at 4 °C or after the addition of 50% (vol/vol) 406
glycerol and storage at -20 °C. For verification of the formation of the expected CoA-407
thioesters the samples obtained by in vitro catalysis were subjected to LC/ESI-MS. Dalluge et 408
al. established a LC/ESI-MS-based method for detection and verification of CoA-thioesters 409
(13). By use of electrospray ionization the authors showed that CoA-thioesters from various 410
organic acids showed a specific parental ion mass spectrum. The mass of the parental ion 411
from a various CoA-thioester followed the equation (13): 412
768 Da (mass free CoA) + x Da (mass various organic acid) – 18 Da (mass H2O). 413
Cleavage of this parental ion results in two distinct daughter ions with m/z = 428 Da and 414
m/z = 261 Da + x Da (mass organic acid) – 18 Da (mass H2O). 415
For the identification of CoA-thioesters it was therefore essential to detect the masses 416
of the parental ion and of the organic acid covalently bound to 4-phosphopantetheine. The 417
SucCDs investigated in this study showed almost identical properties regarding the activation 418
of substrate analogues as well as regarding kinetic properties with substrates that showed 419
highest activity (Table 3). The SucCDs employed in this study were able to activate succinate, 420
itaconate, 3SP, L-malate, D-malate, glutarate and fumarate to their corresponding CoA-421
thioester. Typical fragmentation of malyl-CoA (Fig. 3) as observed in ESI/MS is exemplarily 422
shown in Fig. 4. SucCDBL21 was also able to activate adipate to adipyl-CoA; however, in 423
corresponding samples containing SucCDAm and SucCDAboHis only typical parent ions were 424
detected. No clear evidence for the formation of tartryl-CoA from tartrate was obtained. 425
Either only parental ions (in samples containing SucCDAm or SucCDAboHis) or typical 426
fragments from daughter ions (428 Da and 393 Da) were observed (when SucCDBL21 was 427
applied in enzyme assay). Nonetheless, the detected amounts were very low. No formation of 428
CoA-thioesters could be demonstrated for monocarboxylic variants such as propionate, 429
butyrate, acetate, levulinate and valerate. Kinetic data suggest that all enzymes strongly 430
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
18
preferred the physiological substrates CoA, ATP and succinate. Comparably high enzyme 431
activity was determined for itaconate, L-malate and D-malate (Fig. 5; Table 3). A common 432
characteristic was the reduced SucCD-activity with 3SP to only 1.0 to 2.5% or even less of 433
the activity with succinate. The Km-values calculated for itaconate were 2- to 10-fold higher 434
than the Km-values calculated for succinate. L-malate showed 18- to 20-fold and D-malate 20- 435
to 27-fold higher Km-values in comparison to the corresponding values for succinate. The 436
SucCDs investigated here exhibited also comparably low affinity for 3SP (10- to 16-fold 437
higher Km-values in comparison to succinate). 438
The kcat-values calculated for the substrates L-malate and D-malate were similar to the 439
corresponding value for itaconate (Table 3). According to Shikata et al. activity levels for 440
different substrates were obtained and normalized to activity with the substrate succinate at a 441
final concentration of 10 mM (44) (Fig. 5). Vmax-values for both enantiomeric forms of malate 442
were in the same order of magnitude and comparable to Vmax of the physiological substrate 443
itaconate (Fig. 5; Table 3). 444
Complementation studies applying pBBR1MCS-5::sucCDAm. In an attempt to 445
complement the A. mimigardefordensis 〉sucCD mutant, which exhibited a negative 446
phenotype on MSM agar plates containing either DTDP or 3SP as sole carbon and energy 447
source in comparison to the wild type, the hybrid plasmid pBBR1MCS-5::sucCDAm was 448
transferred to the deletion mutant by conjugation. Transconjugants were selected on MSM 449
containing 0.5% (wt/vol) gluconate and gentamicin. As expected, growth of the 450
complemented mutant was observed in liquid MSM containing 50 mM DTDP (and 451
gentamicin for plasmid stability). While the wild type grew normally, the deletion mutant 452
showed no growth at all. Growth of the complemented mutant was delayed in comparison to 453
the wild type but reached at least 56% of the wild type´s cell density in the given time range 454
(compare Fig. S1). 455
456
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
19
DISCUSSION 457
In this study, we purified and analyzed three different SucCDs with respect to their substrate 458
range. Additionally, growth of the mutant strain A. mimigardefordensis DPN7T ∆sucCD on 459
DTDP was restored by pBBR1MCS5::sucCDAm. The plasmid was constructed with a 478-bp 460
upstream region to apply the endogenous promotor region for expression of sucCD. Xia et al. 461
showed that both orientations of the gene to be expressed with the endogenous promotor 462
resulted in an efficient gene expression (58). These results complete the findings of 463
Schürmann et al. who described the essential role of SucCD in A. mimigardefordensis DPN7T 464
in the degradation of DTDP (43). Expression of the relevant genes sucC and sucD in about 465
equimolar amounts as essential for a successful purification protocol was observed by 466
including 68 or 135 bp of the corresponding sucC upstream regions into the expression vector 467
pBluescriptSK(-) for sucCDAm and sucCDBL21, respectively. During experiments, best 468
expression was obtained when the sucCDs of A. mimigardefordensis DPN7T and of E. coli 469
BL21 were each applied in one bicistronic operon including the strain-specific Shine-470
Dalgarno sequence upstream of sucC. The genes encoding SucCDAboHis were also expressed 471
in one bicistronic operon. In this case the vector-specific Shine-Dalgarno sequence was used. 472
Efficient purification of SucCDAboHis was achieved with a hexahistidine-tagged variant of 473
SucD. Both subunits co-eluted from the column matrix. All SucCD enzymes were isolated in 474
an active state, SucCDAboHis shows a reduced specific activity in comparison to the other 475
SucCDs investigated in this study. The lysate obtained from cells of E. coli BL21 (DE3) 476
pLysS expressing either sucCDAbo or sucCDAboHis led to similar results concerning expression 477
level of sucCD as well as specific SucCD activity and indicate that SucCD from A. 478
borkumensis SK2 might in general have a comparably low activity. 479
Kinetic parameters of all three SucCDs indicate a preference for the physiological 480
substrates CoA, ATP and succinate; therefore, this clearly allocated this enzyme to the citric 481
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
20
acid cycle. Low levels of activity were determined for SucCDBL21 and for SucCDAm with 3SP 482
as a substrate. Due to the generally reduced activity of SucCDAboHis, kinetic parameters 483
concerning 3SP were not determined as a strong excess of enzyme within the photometric 484
assay would have been necessary. SucCDAm showed an about 3.6-fold higher Km-value for 485
3SP in comparison to the data published (43). This could be explained by use of a different 486
method with altered concentrations of Mg2+ and ATP which had been adapted for the 487
determination of enzyme activity with SucCDBL21 in this study. Nonetheless, the other kinetic 488
parameters in this study were in good accordance with published data for SucCDAm (43). 489
Activity was also proven for itaconate which has been described as substrate for A. 490
mimigardefordensis DPN7T and for mammalian SucCD (1, 43). The ability to convert both 491
enantiomers of malate, with a slight preference for L-malate, is a general feature of the 492
SucCDs investigated in this study. The Km-values for these substrates are in the same range as 493
those for 3SP in this study. 494
D-malate has no relevance in vivo as no metabolic pathways involving this compound 495
are known, yet. The conversion of the stereoisomer L-malate to L-malyl-CoA is a crucial 496
catalytic step in the serine cycle in one group of the methylotrophic bacteria. This pathway 497
serves for the efficient assimilation of C1-compounds such as methanol or methylamine (11, 498
22). The genes responsible for the catalytic step from L-malate to L-malyl-CoA in 499
Methylobacterium extorquens strain AM1 have been identified as mtkA and mtkB encoding 500
the malate-CoA ligases, also known as malate thiokinase (11). The malate-CoA ligase of 501
Aminobacter aminovorans, formerly known as Pseudomonas sp. strain MA (ATCC® 502
23819™) and also a member of microorganisms able to assimilate C1-compounds, was 503
biochemically characterized in the past (14, 17-19). Surprisingly, the author determined a 5% 504
higher activity with succinate when compared to the activity with L-malate in vitro (18). 505
Beside these kinetic data obtained by Hersh (18), malate-CoA ligases show similarities to 506
SucCD enzymes concerning amino acid sequence (11), subunit distribution and molecular 507
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
21
weight (14). Although M. extorquens AM1 possesses both, mtkAB and sucCD genes, in its 508
genome (31), it was shown that mtkAB is essential for growth on C1- and C2-compounds, 509
because an insertion mutant lacking intact mtkA did not grow on these compounds. Growth 510
was restored by applying a rescue vector for mtkAB in complementation experiments (11). All 511
these data suggest that these two enzyme sub-subclasses, succinate-CoA ligase (EC 6.2.1.4/ 512
6.2.1.5) and malate-CoA ligase (EC 6.2.1.9) share the same evolutionary origin. Since SucCD 513
from M. extorquens AM1 is not able to compensate the MtkAB-deficiency in the mutant 514
strain (11), an in vitro characterization of corresponding SucCD and MtkAB might elucidate 515
mechanistic and kinetic differences to the SucCDs in this study. 516
In addition to the compounds succinate, itaconate, 3SP, L-malate and D-malate, CoA-517
thioester formation with fumarate, glutarate and adipate was verified with the same LC/ESI-518
MS method established by Dalluge et al. (13). However, SucCD activity with the latter 519
compounds was below 1% of that with succinate; therefore, it is assumed that these activities 520
do not have any relevance in vivo. Nonetheless, the ability to form CoA-thioesters of these 521
dicarboxylic acids might be used for the analysis of substrate specificities in further enzyme 522
characterization experiments (42). 523
No activation to CoA-thioesters was observed with monocarboxylic acids such as 524
acetate, propionate, butyrate, valerate and levulinate. Thus, a second carboxyl group appears 525
to be mandatory for proper binding within the active site of SucCD. Although maleate carries 526
a second carboxyl group, maleyl-CoA was not detected during LC/ESI-MS analyses. Hence, 527
the cis double bond of maleate (in contrast to trans in fumarate) might also impair proper 528
binding to the active site. With regard to chain length, succinate (C4) was found to be the best 529
substrate whereas CoA-thioesters of glutarate (C5) and adipate (C6) were only formed in 530
traces; malonate (C3) was not activated at all. While D- and L-malate were activated to the 531
corresponding CoA-thioester, the structural analogue mercaptosuccinate (Fig. 1) could not be 532
utilized by any of the investigated SucCDs. This might be due to the higher acidicity of 533
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
22
sulfhydryl groups in comparison to hydroxyl groups which consequently results in an 534
additional negative charge at the sulfur atom. The lack to activate sulfosuccinate might be 535
explainable due to steric hindrance by the sulfonic acid group in comparison to the smaller 536
hydroxyl group of malate. 537
This study proved the ability of different SucCDs to form CoA-thioesters also with 538
succinate analogues such as malate and 3SP. Concomittantly, this study showed that 539
activation of the latter is not a unique characteristic of SucCDAm in the degradation of DTDP. 540
541
ACKNOWLEDGMENT 542
The LC/ESI-MS device used in this study was provided by funds of the DFG (Deutsche 543
Forschungsgemeinschaft, Grant No.: INST 211/415-1 FUGG) which we gratefully 544
acknowledge. 545
546
REFERENCES 547
1. Adler J, Wang SF, Lardy HA. 1957. The metabolism of itaconic acid by liver mitochondria. 548
J. Biol. Chem. 229:865-879. 549
2. Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. 1997. 550
Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. 551
Nucleic Acids Res. 25:3389-3402. 552
3. Bachmann BJ. 1990. Linkage map of Escherichia coli K-12, edition 8. Microbiol. Rev. 553
54:130-197. 554
4. Birney M, Um HD, Klein C. 1996. Novel mechanisms of Escherichia coli succinyl-555
coenzyme A synthetase regulation. J. Bacteriol. 178:2883-2889. 556
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
23
5. Bradford MM. 1976. A rapid and sensitive method for the quantitation of microgram 557
quantities of protein utilizing the principle of protein-dye binding. Anal. Biochem. 72:248-558
254. 559
6. Bräsen C, Schönheit P. 2005. AMP-forming acetyl-CoA synthetase from the extremely 560
halophilic archaeon Haloarcula marismortui: purification, identification and expression of the 561
encoding gene, and phylogenetic affiliation. Extremophiles. 9:355-365. 562
7. Bridger, W. A. 1974. Succinyl-CoA synthetase. P. D. Boyer, Academic Press, Inc., New 563
York. 564
8. Bridger WA. 1971. Evidence for two types of subunits in succinyl coenzyme A synthetase. 565
Biochem. Biophys. Res. Commun. 42:948-954. 566
9. Buck D, Spencer ME, Guest JR. 1985. Primary structure of the succinyl-CoA synthetase of 567
Escherichia coli. Biochemistry. 24:6245-6252. 568
10. Cha S, Parks RE. 1964. Succinic Thiokinase. Ii. Kinetic Studies: Initial velocity, product 569
inhibition, and effect of arsenate. J. Biol. Chem. 239:1968-1977. 570
11. Chistoserdova LV, Lidstrom ME. 1994. Genetics of the serine cycle in Methylobacterium 571
extorquens AM1: identification, sequence, and mutation of three new genes involved in C1 572
assimilation, orf4, mtkA, and mtkB. J. Bacteriol. 176:7398-7404. 573
12. Cooper RA, Itiaba K, Kornberg HL. 1965. The utilization of aconate and itaconate by 574
Micrococcus sp. Biochem. J. 94:25-31. 575
13. Dalluge JJ, Gort S, Hobson R, Selifonova O, Amore F, Gokarn R. 2002. Separation and 576
identification of organic acid-coenzyme A thioesters using liquid chromatography/electrospray 577
ionization-mass spectrometry. Anal. Bioanal Chem. 374:835-840. 578
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
24
14. Elwell M, Hersh LB. 1979. Substrate-dependent dissociation of malate thiokinase. J. Biol. 579
Chem. 254:2434-2438. 580
15. Fraser ME, James MN, Bridger WA, Wolodko JWT. 1999. A detailed structural 581
description of Escherichia coli succinyl-CoA synthetase. J. Mol. Biol. 285:1633-1653. 582
16. Gibson J, Upper CD, Gunsalus IC. 1967. Succinyl coenzyme A synthetase from Escherichia 583
coli. I. Purification and properties. J. Biol. Chem. 242:2474-2477. 584
17. Hersh LB. 1974. Malate thiokinase. The reaction mechanism as determined by initial rate 585
studies. J. Biol. Chem. 249:6264-6271. 586
18. Hersh LB. 1973. Malate adenosine triphosphate lyase. Separation of the reaction into a malate 587
thiokinase and malyl coenzyme A lyase. J. Biol. Chem. 248:7295-7303. 588
19. Hersh LB, Peet M. 1981. Half-of-the-sites reactivity in the malate thiokinase reaction. J. Biol. 589
Chem. 256:1732-1737. 590
20. Jenkins TM, Weitzman PD. 1986. Distinct physiological roles of animal succinate 591
thiokinases. Association of guanine nucleotide-linked succinate thiokinase with ketone body 592
utilization. FEBS Lett. 205:215-218. 593
21. Jollès-Bergeret B. 1974. Enzymatic and chemical synthesis of 3-sulfinopropionic acid, an 594
analog of succinic acid. Eur. J. Biochem. 42:349-353. 595
22. Jones JG, Bellion E. 1991. In vivo 13C and 15N NMR studies of methylamine metabolism in 596
Pseudomonas species MA. J. Biol. Chem. 266:11705-11713. 597
23. Joyce MA, Fraser ME, Brownie ER, James MN, Bridger WA, Wolodko WT. 1999. 598
Probing the nucleotide-binding site of Escherichia coli succinyl-CoA synthetase. 599
Biochemistry. 38:7273-7283. 600
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
25
24. Joyce MA, Fraser ME, James MN, Bridger WA, Wolodko WT. 2000. ADP-binding site of 601
Escherichia coli succinyl-CoA synthetase revealed by x-ray crystallography. Biochemistry. 602
39:17-25. 603
25. Kapatral V, Bina X, Chakrabarty AM. 2000. Succinyl coenzyme A synthetase of 604
Pseudomonas aeruginosa with a broad specificity for nucleoside triphosphate (NTP) synthesis 605
modulates specificity for NTP synthesis by the 12-kilodalton form of nucleoside diphosphate 606
kinase. J. Bacteriol. 182:1333-1339. 607
26. Kaufman S. 1955. Studies on the mechanism of the reaction catalyzed by the phosphorylating 608
enzyme. J. Biol. Chem. 216:153-164. 609
27. Kovach ME, Elzer PH, Hill DS, Robertson GT, Farris MA, Roop RM,2nd, Peterson KM. 610
1995. Four new derivatives of the broad-host-range cloning vector pBBR1MCS, carrying 611
different antibiotic-resistance cassettes. Gene. 166:175-176. 612
28. Lütke-Eversloh T, Steinbüchel A. 2003. Novel precursor substrates for polythioesters (PTE) 613
and limits of PTE biosynthesis in Ralstonia eutropha. FEMS Microbiol. Lett. 221:191-196. 614
29. Mai X, Adams MW. 1996. Purification and characterization of two reversible and ADP-615
dependent acetyl coenzyme A synthetases from the hyperthermophilic archaeon Pyrococcus 616
furiosus. J. Bacteriol. 178:5897-5903. 617
30. Majumdar R, Guest JR, Bridger WA. 1991. Functional consequences of substitution of the 618
active site (phospho)histidine residue of Escherichia coli succinyl-CoA synthetase. Biochim. 619
Biophys. Acta. 1076:86-90. 620
31. Markowitz VM, Chen IM, Palaniappan K, Chu K, Szeto E, Grechkin Y, Ratner A, Jacob 621
B, Huang J, Williams P, Huntemann M, Anderson I, Mavromatis K, Ivanova NN, 622
Kyrpides NC. 2012. IMG: the Integrated Microbial Genomes database and comparative 623
analysis system. Nucleic Acids Res. 40:115-122. 624
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
26
32. Marmur J. 1961. A procedure for the isolation of deoxyribonucleic acid from micro-625
organisms. J. Mol. Biol. 3:208-218. 626
33. Milner HW, Lawrence NS, French CS. 1950. Colloidal dispersion of chloroplast material. 627
Science. 111:633-634. 628
34. Murakami K, Mitchell T, Nishimura JS. 1972. Nucleotide specificity of Escherichia coli 629
succinic thiokinase. Succinyl coenzyme A-stimulated nucleoside diphosphate kinase activity 630
of the enzyme. J. Biol. Chem. 247:6247-6252. 631
35. Nagai J. 1963. Studies on itaconate metabolism - I. Itaconyl-CoA synthesizing reaction in 632
cell-free extract. The Journal of Biochemistry. 53:181-187. 633
36. Noy N, Zakim D. 1985. Fatty acids bound to unilamellar lipid vesicles as substrates for 634
microsomal acyl-CoA ligase. Biochemistry. 24:3521-3525. 635
37. Pearson PH, Bridger WA. 1975. Catalysis of a step of the overall reaction by the alpha 636
subunit of Escherichia coli succinyl-coenzyme A synthetase. J. Biol. Chem. 250:8524-8529. 637
38. Rajashekhara E, Watanabe K. 2004. Propionyl-coenzyme A synthetases of Ralstonia 638
solanacearum and Salmonella choleraesuis display atypical kinetics. FEBS Lett. 556:143-147. 639
39. Sambrook J, Fritsch EF, Maniatis T. 1998. Molecular cloning: a laboratory manual. 2nd 640
edition. Cold Spring Harbor Laboratory Press. Cold Spring Harbor, NY. . 641
40. Schlegel HG, Kaltwasser H, Gottschalk G. 1961. A submersion method for culture of 642
hydrogen-oxidizing bacteria: growth physiological studies. Arch. Mikrobiol. 38:209-222. 643
41. Schneiker S, Martins dos Santos VA, Bartels D, Bekel T, Brecht M, Buhrmester J, 644
Chernikova TN, Denaro R, Ferrer M, Gertler C, Goesmann A, Golyshina OV, Kaminski 645
F, Khachane AN, Lang S, Linke B, McHardy AC, Meyer F, Nechitaylo T, Puhler A, 646
Regenhardt D, Rupp O, Sabirova JS, Selbitschka W, Yakimov MM, Timmis KN, 647
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
27
Vorholter FJ, Weidner S, Kaiser O, Golyshin PN. 2006. Genome sequence of the 648
ubiquitous hydrocarbon-degrading marine bacterium Alcanivorax borkumensis. Nat. 649
Biotechnol. 24:997-1004. 650
42. Schürmann M, Deters A, Wübbeler JH, Steinbüchel A. 2013. A novel 3-sulfinopropionyl 651
coenzyme A (3SP-CoA) desulfinase from Advenella mimigardefordensis strain DPN7T acting 652
as a key enzyme during catabolism of 3,3'-dithiodipropionic acid is a member of the Acyl-653
CoA dehydrogenase superfamily. J. Bacteriol. 195:1538-1551. 654
43. Schürmann M, Wübbeler JH, Grote J, Steinbüchel A. 2011. Novel reaction of succinyl 655
coenzyme A (Succinyl-CoA) synthetase: activation of 3-sulfinopropionate to 3-656
sulfinopropionyl-CoA in Advenella mimigardefordensis strain DPN7T during degradation of 657
3,3'-dithiodipropionic acid. J. Bacteriol. 193:3078-3089. 658
44. Shikata K, Fukui T, Atomi H, Imanaka T. 2007. A novel ADP-forming succinyl-CoA 659
synthetase in Thermococcus kodakaraensis structurally related to the archaeal nucleoside 660
diphosphate-forming acetyl-CoA synthetases. J. Biol. Chem. 282:26963-26970. 661
45. Shimizu S, Inoue K, Tani Y, Yamada H. 1981. Butyryl-CoA synthetase of Pseudomonas 662
aeruginosa - purification and characterization. Biochem. Biophys. Res. Commun. 103:1231-663
1237. 664
46. Simon R, Priefer U, Pühler A. 1983. A broad host range mobilization system for in vivo 665
genetic engineering: transposon mutagenesis in Gram negative bacteria. Biotechnology. 666
1:784-791. 667
47. Stein S, Levitsky A, Fateev O, Mallard G. 1998. The NIST Mass Spectral Search Program, 668
Windows software version 1.6d. National Institute of Standards and Technology, 669
Gaithersburg, MD. . 670
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
28
48. Studart-Guimaraes C, Gibon Y, Frankel N, Wood CC, Zanor MI, Fernie AR, Carrari F. 671
2005. Identification and characterisation of the alpha and beta subunits of succinyl CoA ligase 672
of tomato. Plant Mol. Biol. 59:781-791. 673
49. Studier FW. 2005. Protein production by auto-induction in high density shaking cultures. 674
Protein Expr. Purif. 41:207-234. 675
50. Wilkins MR, Gasteiger E, Bairoch A, Sanchez JC, Williams KL, Appel RD, Hochstrasser 676
DF. 1999. Protein identification and analysis tools in the ExPASy server. Methods Mol. Biol. 677
112:531-552. 678
51. Wolodko WT, Fraser ME, James MN, Bridger WA. 1994. The crystal structure of 679
succinyl-CoA synthetase from Escherichia coli at 2.5-A resolution. J. Biol. Chem. 269:10883-680
10890. 681
52. Wolodko WT, James MN, Bridger WA. 1984. Crystallization of succinyl-CoA synthetase 682
from Escherichia coli. J. Biol. Chem. 259:5316-5320. 683
53. Wolodko WT, Kay CM, Bridger WA. 1986. Active enzyme sedimentation, sedimentation 684
velocity, and sedimentation equilibrium studies of succinyl-CoA synthetases of porcine heart 685
and Escherichia coli. Biochemistry. 25:5420-5425. 686
54. Wübbeler JH, Bruland N, Kretschmer K, Steinbüchel A. 2008. Novel pathway for 687
catabolism of the organic sulfur compound 3,3'-dithiodipropionic acid via 3-688
mercaptopropionic acid and 3-sulfinopropionic acid to propionyl-coenzyme A by the aerobic 689
bacterium Tetrathiobacter mimigardefordensis strain DPN7. Appl. Environ. Microbiol. 690
74:4028-4035. 691
55. Wübbeler JH, Bruland N, Wozniczka M, Steinbüchel A. 2010. Biodegradation of the 692
xenobiotic organic disulphide 4,4'-dithiodibutyric acid by Rhodococcus erythropolis strain 693
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
29
MI2 and comparison with the microbial utilization of 3,3'-dithiodipropionic acid and 3,3'-694
thiodipropionic acid. Microbiology. 156:1221-1233. 695
56. Wübbeler JH, Lütke-Eversloh T, Van Trappen S, Vandamme P, Steinbüchel A. 2006. 696
Tetrathiobacter mimigardefordensis sp. nov., isolated from compost, a betaproteobacterium 697
capable of utilizing the organic disulfide 3,3'-dithiodipropionic acid. Int. J. Syst. Evol. 698
Microbiol. 56:1305-1310. 699
57. Wübbeler JH, Raberg M, Brandt U, Steinbüchel A. 2010. Dihydrolipoamide 700
dehydrogenases of Advenella mimigardefordensis and Ralstonia eutropha catalyze cleavage of 701
3,3'-dithiodipropionic acid into 3-mercaptopropionic acid. Appl. Environ. Microbiol. 76:7023-702
7028. 703
58. Xia Y, Wübbeler JH, Qi Q, Steinbüchel A. 2012. Employing a recombinant strain of 704
Advenella mimigardefordensis for biotechnical production of homopolythioesters from 3,3'-705
dithiodipropionic acid. Appl. Environ. Microbiol. 78:3286-3297. 706
707
708
709
710
711
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
30
Legends to figures 712
713
FIG 1 Succinate and analogous compounds investigated as potential substrates for SucCD 714
enzymes. 715
716
FIG 2 Purification of SucCDs from E. coli (1), A. mimigardefordensis DPN7T (2) and A. 717
borkumensis SK2 (3) as revealed by SDS-PAGE (11.5%, wt/vol, acrylamide). Lane 1, 10 µg 718
purified SucCD from E. coli; lane 2: 10 µg purified SucCD from A. mimigardefordensis 719
DPN7T; lane 3: 10 µg of purified SucCD from A. borkumensis SK2. The gel was stained with 720
Coomassie Brilliant blue R. Molecular mass standard was PageRuler Prestained Protein 721
Ladder (Thermo Fisher Scientific, Rockford, USA). 722
723
FIG 3 Structural formula of malyl-CoA and fragmentation pattern of parent ion into several 724
daughter ions in ESI-MS. 725
726
FIG 4 Analyses of malyl-CoA. On top an ESI spectrum of malyl-CoA in the positive mode 727
is shown. The central part shows an MS spectrum of the parent ion (m/z = 884 Da). Two main 728
fragments (m/z = 428 Da and m/z = 377 Da) were obtained. At the bottom, further 729
fragmentation of the parent m/z = 377 Da yielded the daughter ions m/z = 275 kDa and 730
m/z = 261 kDa. 731
732
FIG 5 Levels of acyl-CoA forming activity (ADP formation) of SucCDBL21, SucCDAm and 733
SucCDAboHis. Activity values were normalized to activity with succinate. Each 10 mM of the 734
various organic acid were applied to the assay. Activity was determined in duplicate 735
experiments. Black bars indicate the standard deviation. 736
737
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
31
FIG S1 Growth of A. mimigardefordensis DPN7T (wild type) ( ), A. mimigardefordensis 738
DPN7T ǻsucCD ( ) and A. mimigardefordensis DPN7T ǻsucCD (pBBR1MCS5::sucCDAm) 739
( ) in minimal medium containing 50 mM DTDP as sole carbon and energy source. 740
741
742
743
744
745
746
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
32
TABLE 1 Strains, plasmids and oligonucleotides used in this study. 747
Strain, plasmid or
oligonucleotide
Description or sequence (5´-3´)a Source or
reference
Strains
A. mimigardefordensis DPN7 Wild type, DTDP-degrading bacterium
(56), (DSM 17166T, LMG 22922T)
A. mimigardefordensis DPN7 ǻsucCD
sucCD deletion mutant of corresponding strain (43)
E. coli Top10 F- mcrA ǻ(mrr-hsdRMS-mcrBC) ij80lacZǻM15 ǻlacX74 nupG deoR recA1 araD139 ǻ(ara-leu)7697 galU galK rpsL(StrR) endA1 Ȝ-
Invitrogen, Carlsbad, CA
E. coli S17-1 thi-1 proA hsdR17 (rK- mK
+) recA1; tra gene of plasmid RP4
integrated into genome(46)
E. coli BL21(DE3) pLysS F- ompT hsdSB (rB- mB
-) gal dcm (DE3), pLysS (Cmr) Novagen, Madison, WI, USA
Alcanivorax borkumensis SK2
Wild type DSM 11573
Plasmids pBluescriptSK(-) Apr, lacPOZ´ Stratagene,
San Diego, CA, USA
pET-23a(+) pBR322 ori, f1 ori His6; Apr, T7lac Novagen, Madison, WI, USA
pBBR1MCS5 Broad host-range expression vector, Gmr (27)pBluescriptSK(-)::sucCDAm Apr
, (43)
pBluescriptSK(-)::sucCDBL21 Apr This study pET-23a(+)::sucCDAbo Apr This studypET-23a(+)::sucCDAboHis Apr This study pBBR1MCS5::sucCDAm Gmr This study
Oligonucleotides sucCDforward_PstI CTGCAGCAGTCTCAATTCGTGTGCTCGC (43) sucCDreverse_XhoI_stop CTCGAGTTACAGTACTGATTTGAGCAGTTTG (43)sucCDBL21_forward_EcoRI AAAAGAATTCTCCGACAAGCGATGCCTGATG suCDBL21_reverse_HindIII AAAAAAGCTTTTATTTCAGAACAGTTTTCAGTGCTTCACC sucCDAbo_forward_NdeI AAAACATATGAATCTCCATGAATATCAGTCAAAACAGC sucCDAbo_reverse_SalI AAAAGTCGACTTACCAGCCAGTCGCTTCTTTCAC P_Abo_rev_mutagenesis CCAGCCAGTCGCTTCTTTCAC (5´-phosporylation) P_forward_XhoI_Histag_Abo CTCGAGCACCACCACCACC (5´-phosphorylation) SKF2 (Abo seq) TCTGAGCGCAGTGCTGG T7 terminator pET-23a GCTAGTTATTGCTCAGCGG T7 promotor pET-23a TAATACGACTCACTATAGGG For3 (BL21 seq) ACTTTTTGCCCGCTATGG For2 (BL21 seq) GTATCAAACAGATGCCAATGG Rev2 (BL21 seq) GTAGTGCCGCCTTTACCTG SucCDAm_Prom_fw_XhoI CTCGAGTTGCTGGTCACGCAGGAAGG a For the abbreviations used in the E. coli genotypes, see reference 3. 748
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
33
TABLE 2 Purification of SucCDs from E. coli BL21, A. mimigardefordensis DPN7T, and A. borkumensis SK2. 749
SucCD origin Purification
step
Total protein
[mg]
Total enzyme activity
[U]
Spec. activity
[U/mg]
Yield
[%]
Level
[-fold]
E. coli BL21 Solube protein fraction
2961 37365 12.6 100.0 1.00
Q-Sepharose 1050 12573 12.0 33.6 0.95
Cibacron Blue F3GA- Sepharose
147 2734 18.6 7.3 1.47
A. mimigardefordensis DPN7T Solube protein fraction
2047 22829 11.2 100.0 1.00
Q-Sepharose 277 3194 11.5 14.0 1.03
DEAE-Sepharose 111 1437 12.9 6.3 1.16
EAH-Sepharose 4B 57 682 12.0 3.0 1.07
A. borkumensis SK2 Soluble protein fraction
279 744 2.67 100.0 1.00
Ni-NTA-Sepharose 42 173 4.14 23.3 1.55
Enzyme activity was determined at 30 °C in the direction of succinyl-CoA/ ADP formation as described in the materials and methods section. 750
1 Unit (U) corresponds to the formation of the formation of 1 µmol ADP per minute. 751
752
753
754
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from
34
TABLE 3 Kinetic parameters determined for SucCDBL21, SucCDAm and SucCDAboHis. 755
Enzyme activities were determined in the direction of CoA-thioester/ADP formation. 756
1 U corresponds to the formation of 1 µmol ADP per minute. kcat-values are given for 757
the number of active sites (Įȕ-dimer). 758
Enzyme Substrate Vmax [U/mg] Km [mM] kcat [s-1
] kcat/Km [s-1
mM-1
]
SucCDBL21 Succinate 12.06 ± 0.03 0.141 ± 0.003 14.3 101.4
Itaconate 1.30 ± 0.01 0.475 ± 0.019 1.5 3.3
L-Malate 1.51 ± 0.04 2.558 ± 0.106 1.8 0.7
D-Malate 0.98 ± 0.02 3.635 ± 0.223 1.2 0.3
3SP 0.15 ± 0.00 1.520 ± 0.081 0.2 0.1
CoA 22.48 ± 0.92 0.058 ± 0.005 26.7 461.3
ATP 16.49 ± 0.10 0.055 ± 0.002 19.6 354.4
SucCDAm Succinate 25.86 ± 0.06 0.182 ± 0.003 31.1 171.1
Itaconate 4.42 ± 0.02 0.351 ± 0.003 5.3 15.1
L-Malate 2.15 ± 0.09 3.095 ± 0.354 2.6 0.8
D-Malate 1.77 ± 0.02 3.588 ± 0.111 2.1 0.6
3SP 0.14 ± 0.01 2.964 ± 0.275 0.2 0.1
CoA 33.47 ± 0.23 0.037 ± 0.001 40.3 1099.6
ATP 48.89 ± 0.41 0.201 ± 0.002 58.8 292.5
SucCDAboHis Succinate 2.23 ± 0.01 0.157 ± 0.003 2.7 17.2
Itaconate 0.39 ± 0.00 1.509 ± 0.048 0.5 3.1
L-Malate 0.88 ± 0.00 3.270 ± 0.017 1.1 0.3
D-Malate 1.38 ± 0.02 4.243 ± 0.089 1.7 0.4
3SP n. d. n. d. n. d. n. d.
CoA 1.33 ± 0.03 0.004 ± 0.001 1.6 372.3
ATP 2.96 ± 0.02 0.241 ± 0.050 3.6 14.8
on March 13, 2020 by guest
http://aem.asm
.org/D
ownloaded from