Discovery of a Bacterial Glycoside Hydrolase Family 3 (GH3

Home , Myrosinase

Subscriber access provided by Imperial College London | Library Article Discovery of a bacterial glycoside hydrolase family 3 (GH3) #-glucosidase with myrosinase activity from a Citrobacter strain isolated from soil Abdulhadi Albaser, Eleanna Kazana, Mark Bennett, Fatma Cebeci, Vijitra Luang-In, Pietro D. Spanu, and John T. Rossiter J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jafc.5b05381 • Publication Date (Web): 28 Jan 2016 Downloaded from http://pubs.acs.org on January 29, 2016

Just Accepted

“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of Agricultural and Food Chemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties. Page 1 of 39 Journal of Agricultural and Food Chemistry

1 Discovery of a bacterial glycoside hydrolase family 3 (GH3) βglucosidase with myrosinase activity

2 from a Citrobacter strain isolated from soil

4 Abdulhadi Albaser 1,5 , Eleanna Kazana 1,3 , Mark H Bennett 1, Fatma Cebeci 2, Vijitra LuangIn 1,4 Pietro

5 D Spanu 1 and John T Rossiter 1*

7 1 Imperial College London, Faculty of Life Sciences, London SW7 2AZ, UK

8 2 Food and Health Programme, Institute of Food Research, Norwich, NR4 7UA, UK

9 3Current address: School of Biosciences, University of Kent, Canterbury, Kent, CT2 7NJ

10 4Current Address: Mahasarakham University, Faculty of Technology, Department of Biotechnology,

11 Natural Antioxidant Research Unit, Mahasarakham 44150, Thailand

12 5Current address: Division of Microbiology, Faculty of Sciences, University of Sebha, Sebha Libya

13 *Corresponding author; John T Rossiter, Imperial College London, Faculty of Life Sciences, London

14 SW7 2AZ, UK; email: [email protected]

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 2 of 39

22 ABSTRACT

23 A Citrobacter strain (WYE1) was isolated from a UK soil by enrichment using the glucosinolate sinigrin

24 as a sole carbon source. The enzyme myrosinase was purified using a combination of ion exchange and

25 gel filtration to give a pure protein of approximately 66 kDa. The Nterminal amino acid and internal

26 peptide sequence of the purified protein were determined and used to identify the gene which, based on

27 InterPro sequence analysis, belongs to the family GH3, contains a signal peptide and is a periplasmic

28 protein with a predicted molecular mass of 71.8 kDa. A preliminary characterization was carried out

29 using protein extracts from cell free preparations. The apparent KM and Vmax were 0.46 mM and 4.91

30 mmol dm 3 min 1 mg 1 respectively with sinigrin as substrate. The optimum temperature and pH for

31 enzyme activity were 25 oC and 6.0 respectively. The enzyme was marginally activated with ascorbate by

32 a factor of 1.67.

34 Keywords: myrosinase, bacteria, isothiocyanate, sequence

ACS Paragon Plus Environment Page 3 of 39 Journal of Agricultural and Food Chemistry

44 INTRODUCTION

45 Bacterial myrosinases are not well understood and there has been little work on this topic. Bacterial

46 myrosinases were first purified some years ago as an active enzyme but without any characterisation in

47 terms of sequence.1, 2 Since then there have been various reports examining the metabolism of

48 glucosinolates by bacteria and attempts have been made to characterize the enzyme but with limited

49 success. 3-7 We have set out to isolate a bacterium that can grow on sinigrin, purify the myrosinase and

50 obtain a sequence. βGlucosidases (EC 3.2.1.21) catalyse the hydrolysis of the glucosidic linkages of a

51 variety of βglucosides and, depending on sequence and structure, can be placed in the glycoside

52 hydrolase (GH) families GH1, GH3, GH9, GH 30 and GH116.8-10 With the exception of GH9, the

53 remainder utilize a twostep mechanism that involves a catalytic nucleophile that displaces the aglycone

54 and is followed by protonation of the nucleofuge and attack by a water molecule to give a glycosyl

55 residue. Plant myrosinases (thioglucoside glucohydrolase EC 3.2.3.147) belong to the family GH1 and

56 their structure and mechanisms have been well studied.11-14 Myrosinases hydrolyse glucosinolates, a

57 group of plant secondary metabolites, to give a variety of products (Figure 1) depending on the presence

58 of specifier proteins and type of glucosinolate, but the default product is an isothiocyanate (ITC).15-18 The

59 mechanism of plant myrosinase does not utilize the typical catalytic acid/base employed by β

60 glucosidases but instead uses ascorbate as a catalytic base. The Sinapis alba myrosinase is a dimer linked

61 by a zinc atom and has a characteristic (β/α) 8barrel structure in common with GH1 enzymes. The insect

62 Brevicoryne brassicae (cabbage aphid) contains a myrosinase that has been studied in some detail and

63 also belongs to the glucose hydrolase family GH1 19-21 and its structure has been determined.22 In contrast

64 to the S. alba myrosinase, the aphid myrosinase does not require ascorbate for activity and its mechanism

65 is the same as for βglucosidases i.e . a nucleophile (glutamic acid residue) that forms a glycosylenzyme

66 intermediate together with a glutamic acid residue that acts as a catalytic base/acid typical of most GH1

67 glucosidases.

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 4 of 39

68 There is, however, increasing interest in bacteria that can metabolise glucosinolates, particularly those of

69 the human gut, where the chemoprotective nature of glucosinolate hydrolysis products is of dietary

70 importance.23 In addition, soil bacteria are also of importance in terms of soil ecology and in the

71 application of biofumigation, where plants belonging to the Brassicaceae, which are high in glucosinolate

72 content, are used to fumigate soil.24-27 In this work we describe the isolation of a soil bacterium that

73 possesses an inducible βglucosidase that transforms glucosinolates into isothiocyanates. The properties

74 and sequence of the enzyme are described.

ACS Paragon Plus Environment Page 5 of 39 Journal of Agricultural and Food Chemistry

89 METHODS

90 Glucosinolates

91 Glucosinolates were isolated from seed sources as previously described.28

92 Isolation of a glucosinolatemetabolizing bacterial strain by enrichment

93 Soil was taken from a field in Wye (Kent, UK), which had previously been used for growing oilseed rape,

94 and 1 gram was placed into 10 ml of M9 medium with sinigrin (10 mg) as the sole carbon source and

95 placed in the dark at room temperature for 2 weeks. After gentle agitation, 1 ml of solution was removed,

96 placed into fresh M9 media with sinigrin as the sole carbon source and incubated for a further 2 weeks.

97 This process was repeated a further 5 times. Finally, dilutions were made and a small amount of the

98 enrichment culture was streaked onto a nutrient agar plate for each dilution. The plates were incubated

99 overnight at 30 oC. Colonies were added to fresh M9 media containing sinigrin as the sole carbon source

o 100 and incubated at 30 C, and the OD 600nm (optical density at 600nm) was monitored. The culture was

101 centrifuged at 10000 g for 5 min and stored at 20 oC until analysis. Sinigrinmetabolizing colonies were

102 grown overnight in M9 media supplemented with sinigrin as the sole carbon source, and the following

103 day glycerol stocks (40%) were made and stored at 80 oC.

104 HPLC analysis of sinigrin fermentation

105 The fermentation of sinigrin was monitored by removing 1 mL of culture into an Eppendorf (1.5 mL) tube

106 and centrifuging at 10000 g for 10 min. The supernatant was removed and stored at 20 oC until analysis.

107 Samples were thawed out (1 mL), 100 L of a barium/lead acetate solution (equimolar, 1M) were added,

108 and samples were thoroughly mixed and allowed to stand for a few minutes. The samples were

109 centrifuged at 10000 g for 5 min and the supernatant was removed and transferred to a fresh Eppendorf

110 tube. DEAE SephadexA25 (Sigma, UK) was mixed with an excess of NaOAc buffer (1M, pH 5), which

111 was changed several times over 6 h and finally left overnight to equilibriate. The excess buffer was

112 decanted and the ion exchanger was mixed with deionized water, decanted (x 2) and finally stored in 20

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 6 of 39

113 % EtOH to give an approximately 50% settled gel. One mL of suspended DEAE SephadexA25 slurry

114 was pipetted into PolyPrep Chromatography Columns (BioRad, UK) and allowed to drain. The ion

115 exchange material was washed twice with 1 mL deionised water, and the sample was applied and

116 allowed to drain. The column was washed with 2 x 1 mL deionised water, followed by two washes of 0.5

117 mL sodium acetate buffer (0.05 M, pH 5.0). Sulfatase (Sigma, UK) was added (75 L, 0.3U/mL)

118 carefully to the top of the gel and the column was left overnight at room temperature. The desulfo

119 glucosinolate (DSGSL) was eluted with 3 x 0.5 mL of deionized water and stored at 20 oC until HPLC

120 analysis. The DSGSLs were analysed by HPLC on a Synergi HydroRP column (4 m, 15 x 0.2 cm,

121 Phenomenex Inc. Torrance, CA). The following conditions were used for the HPLC analysis: Water

122 (solvent A), ACN (solvent B); 2% B (15 min), 225% B (2 min), 2570% B (2 min), 70% (2 min), 702%

123 B (2 min), 702%B (15 min); flow rate 0.2 mL/min, column temperature 35 oC. HPLC was carried out

124 using an Agilent HP1100 system with a Waters UV detector (model 486) at a wavelength of 229 nm.

125 GCMS analysis of glucosinolate metabolites

126 One mL of culture or cell free extract was removed and centrifuged at 10000 g for 10 min. The

127 supernatant was removed and transferred to a 2 mL Eppendorf tube, 0.5 mL DCM was added and the tube

128 was vortexed for 30 s. The tubes were centrifuged at 10000 g for 5 min and the DCM was carefully

129 removed with a glass syringe. The sample was reextracted with a further 0.5 mL DCM and combined

130 with the first extraction. A small amount of dried MgSO 4 was added to the DCM extract to remove any

131 water and briefly vortexed. The extract was centrifuged and the DCM was carefully removed, transferred

132 to an Agilent 2 mL screw cap vial and stored at 4 oC until analysis.

133 An Agilent HP 6890 series system connected to an Agilent HP 5973 mass selective detector was used for

134 the GCMS analysis. A capillary column, Agilent HP5MS (5% Phenylmethylsiloxane, 30 m × 0.25 mm

135 i.d.; film thickness, 0.25 m), was used with helium as the carrier gas (split mode, 25:1; splitter inlet

136 pressure, 40 kPa). The temperature was kept at 40 °C for 10 min and ramped up to 150 °C at the rate of

137 4°C/min for 25 min and then to 250°C at the rate of 4°C/min for 15 min, with the flow at 1 mL/min,

ACS Paragon Plus Environment Page 7 of 39 Journal of Agricultural and Food Chemistry

138 average velocity of 36 cm/s, pressure at 7.56 kPa and injection volume of 1 L. Mass spectra were

139 obtained by electron ionization (EI) over a range of 50−550 atomic mass units. Ion source temperature

140 was 230 °C and the electron multiplier voltage was 70.1 eV. The analysis of allyl amine, allyl nitrile and

141 2propenoic acid (Sigma UK) were carried out on a MacheryNagel (UK) Optima Delta6 column (30 m

142 × 0.25 mm i.d.; film thickness, 0.25 m) using the same programme as previously described, with the

143 exception that the final oven temperature was 200 °C. The limit of detection for pure ITC/NIT standards

144 by GCMS analysis was 10 ng/mL.

145 16S rRNA Sequence determination, genomic DNA assembly and preparation of phylogenetic tree

146 An overnight culture was centrifuged at 13000 g for 5 min, and the pellet was washed with milliQ water

147 (x1), resuspended in 15 L milliQ water and boiled at 95 oC for 5 min. The sample was cooled on ice,

148 spun for 5 min at 13,000 g and the supernatant was collected. The gene was amplified by PCR using the

149 primers AMP_F: 5': GAG AGT TTG ATY CTG GCT CAG, Tm: 60.5 and AMP_R: 5': AAG GAG GTG

150 ATC CAR CCG CA, Tm: 68.9, GoTaq DNA polymerase, PCR buffer (Promega) and dNTPs (200 M

151 each). The PCR conditions were the following: initial denaturation 95 oC, 2 min (1 cycle); denaturation

152 95 oC, 30 s (20 cycles); annealing 55 oC, 30 s (20 cycles); extension 72 °C, 1 min 30 s (20 cycles); final

153 extension 72 °C, 1 min (1 cycle). The PCR product (1500 bp) was purified by QIAquick PCR

154 purification kit (Qiagen) and quantified by a Nanodrop 2000 spectrometer (Thermo scientific). The

155 product was sequenced with both primers (AMP_F and AMP_R; Eurofins Sequencing). The sequences

156 were assembled using Seqman to create a contig and this contig was searched on the RDP database

157 (https://rdp.cme.msu.edu/seqmatch/seqmatch_intro.jsp). The genomic DNA of the Citrobacte r WYE1

158 was sequenced at The Genome Analysis Centre, Norwich, UK, using the Illumina MiSeq platform as

159 previously reported (29). The 251bp length of DNA reads from Genomic DNA sequencing was analysed

160 and qualitytrimmed extended contigs were produced by using SPAdes program version 3.6.1. 29 These

161 contigs of Citrobacter WYE1 were used in construction of a phylogenetic tree. Draft genomes of different

162 Citrobacter species were obtained from Genbank whole genome shutgun projects. Core genome single

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 8 of 39

163 nucleotide polymorphisms (SNPs) in draft genomes and Citrobacter WYE1 were used to prepare the

164 phylogenetic tree using parSNP program Version 1.2 30 .

165 Myrosinase assay

166 Myrosinase activity determinations were based on the glucose oxidase/peroxidase enzyme assay of

167 glucose. 31 Preparation of the GODPERID (glucose oxidaseperoxidase) reagent (all reagents purchased

168 from Sigma, UK): 2,2'azinobis(3ethylbenzothiazoline6sulphonic acid) 250 mg, glucose oxidase 12.7

169 mg (157.5 U/mg) and peroxidase 4.7 mg (148U/mg) were dissolved in Tris buffer (3 g Tris, final

170 concentration 0.1M, pH 7.2) to a volume of 250 mL. The reagent was mixed well, wrapped with

171 aluminium foil and kept at 4°C. A calibration curve was obtained using glucose. Samples were assayed in

172 a total volume of 300 L containing the myrosinase extract (maximum volume 20 L), 2 mM sinigrin,

173 CPB (citrate phosphate buffer, 20 mM, pH 6.0), and incubated at 25 oC for 2 h or less depending on the

174 activity. The reaction was stopped by placing in a heating block (95 oC) for 5 min and cooled on ice. 1

175 mL of GODPERID reagent was added to the sample and incubated for 15 min at 37 °C. The absorbance

176 at 420 nm was measured and the glucose concentration was determined from a calibration graph.

177 General assay conditions for myrosinase activity

178 As there was insufficient pure protein for characterization studies, preliminary studies were carried out

179 using cellfree protein extracts. An overnight culture of Citrobacter WYE1 was grown on nutrient broth

180 (NB), inoculated into M9 medium (50 mL) with 50 mg of sinigrin and incubated aerobically overnight at

181 30 °C in a shaking incubator at 180 rpm. The culture was centrifuged, the pellet was washed as described

182 in the purification methods and the cells resuspended in 5 ml of extraction buffer (CPB, 50 L 100 xs

183 protease inhibitor cocktail (Melford Ltd) and 1 mM DTT) and lysed using a one shot cell disruptor

184 (CONSTANT SYSTEMS Ltd, Northants, UK). The lysate was centrifuged at 10,000 g for 15 min at 4oC.

185 The cellfree protein extract was desalted against CPB using EconoPac 10DG Desalting Columns

186 (BIORAD, UK) and stored at 4 oC. The protein concentration was determined using Bradford reagent

ACS Paragon Plus Environment Page 9 of 39 Journal of Agricultural and Food Chemistry

187 (BIORAD, UK). The activity of the myrosinase was initially determined to ensure linearity and an

188 appropriate amount of protein used for each assay. As the activity of the myrosinase markedly diminished

189 over time, the cellfree protein extracts were used within 72 h of preparation.

190 pH Optimisation

191 Citrate phosphate buffer (pH range of 3.67.6, 20 mM) was used to determine the optimum pH for

192 Citrobacter WYE1myrosinase. Crude protein extract (5 µg) was added to a mixture of sinigrin (2 mM) in

193 CPB to a final volume of 300 µL at specific pH values over the range 3.6 7.6 and incubated at 37 °C for

194 2 h. The reaction was stopped by immersing the tubes in boiling water for 5 min. Glucose release

195 measurements were then carried out using the GODPERID assay.

196 Temperature optimization

197 The effect of temperature on the activity of the Citrobacter WYE1 myrosinase was measured over a range

198 of temperatures (570 ºC). A total volume (300 µL) with protein (5µg), sinigrin (2 mM) and CPB was

199 incubated at the specified temperature for 30 min and glucose release was assayed by the GOD Perid

200 assay.

201 Substrate specificity

202 Myrosinase assays were carried out using CPB, 2 mM glucosinolate substrate, 60 min incubation and 20

203 g protein. Samples for qualitative GCMS analysis were incubated overnight and extracted with DCM as

204 previously described for glucosinolate metabolites.

205 Activation by ascorbate

206 A mixture of the protein (16 µg) and CPB was incubated (30 ºC, 1h) with ascorbate concentrations in the

207 range 0.02520 mM and gluconasturtiin (0.5 mM) to a volume of 300 µL. The reaction was stopped by

208 boiling (5 min). ITCs were extracted with dichloromethane and quantitated by GCMS using an Agilent

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 10 of 39

209 HP5MS column as previously described. Quantitation was carried out using a calibration graph of known

210 phenethylisothiocyanate concentrations versus total ion current.

211 Determination of apparent Km & Vmax for sinigrin

212 A range of sinigrin concentrations (0.15 mM) was incubated (25 °C, 10 min) with protein (2 µg) and

213 CPB in a total volume of 300 µL. The reaction was stopped by boiling (5 min) and cooled before the

214 GOD Perid reagent (1 mL) was added and incubated (15 min, 37 °C), followed by absorbance

215 measurement (420 nm). Assays were carried out in triplicate and the kinetic parameters evaluated using

216 GraphPad version 6.

217 Estimation of pI

218 A range of buffers at different pH ( Nmethyl piperazine pH 4.5 and 5; L histidine pH 5.5 and 6.0; bis

219 Tris pH 6.6; bisTris propane pH 7; triethanolamine pH 7.6; Tris pH 8) was used to determine the

220 approximate p I of the Citrobacter myrosinase. Mini anion Q columns (Thermo scientific) were

221 equilibrated with the appropriate buffer (400 µL) and centrifuged on an Eppendorf 5425 centrifuge

222 (2000 g, 4 ºC, 5 min). The protein extract was desalted using Zebra spin desalting columns 0.5 mL

223 (Thermo scientific) against the appropriate pH buffer and applied to the mini anion Q column and then

224 centrifuged (2000 x g, 4 ºC, 5 min). The column was washed twice with the appropriate buffer (400µL).

225 Finally the protein was eluted with buffer that contained NaCl (1 M) and desalted against CPB before

226 being assayed for activity.

227 Inducibility of myrosinase

228 Citrobacter WYE1 was grown overnight in M9 media and separately supplemented with sinigrin (10

229 mM) or glucose (10 mM). Cell protein extracts were prepared as previously described (general assay

230 conditions) and assayed for myrosinase activity using 5 µg of crude desalted protein extract/assay.

231 Purification of Citrobacter WYE1 myrosinase

ACS Paragon Plus Environment Page 11 of 39 Journal of Agricultural and Food Chemistry

232 An overnight culture of Citrobacter WYE1 was grown on nutrient broth (NB), inoculated into M9

233 medium (500 mL) with 0.5 g of sinigrin and incubated aerobically overnight at 30 °C in a shaking

234 incubator at 180 rpm. The bacteria were harvested by centrifugation (Avanti J26 XP, Beckman coulter,

235 USA) (8000 g, 15 min, 4 °C) and washed twice with CPB. The cell pellet was resuspended in extraction

236 buffer (CPB, 250 L 100 Xs protease inhibitor cocktail (Melford Ltd), 1 mM DTT) and lysed using a one

237 shot cell disruption system. The cell debris was pelleted by centrifugation (20,000 x g, 20 min, 4°C) and

238 the supernatant was desalted against CPB using a desalting column (EconoPac 10DG) and concentrated

239 to 11 mL using an ultrafiltration membrane tube (molecular weight cutoff 10 KDa, 4 mL UFC801024,

240 Amicon ultra, Millipore, USA). The protein concentration was measured using Bradford’s reagent

241 (BioRad, UK) and the activity of the extract using the GODPERID reagent.

242 Proteins were separated on a Mono Q HR 515 column (50 x 5 mm I.D., Pharmacia, Uppsala, Sweden).

243 The column was attached to a WATERS HPLC purification system (650 E, Millipore, USA) equipped

244 with a UV detector (ABI 757 absorbance detector, Applied Biosystems) and protein elution monitored at

245 A280nm . Both the initial starting buffer (Tris 20 mM, pH 8.0) and elution buffer (Tris 20 mM, plus NaCl

246 1M pH 8.0) were filtered using a Millipore membrane filter. The column was preequilibrated with

247 starting buffer. The sample (40 mg of protein) was desalted against the initial starting buffer using Econo

248 Pac 10DG desalting columns then filtered through a 0.2 µm miniSart filter prior to injection. Fractions

249 (1.5 mL) were collected every 2 min at flow rate of 0.75 mL/min. The bound protein was eluted by using

250 a linear buffer concentration gradient from 01M NaCl (Supplementary information S1).

251 Fractions with myrosinase activity were combined and desalted against starting buffer using EconoPac

252 10DG desalting columns and concentrated using an ultrafiltration membrane tube. The sample (0.77 mg)

253 was applied to an ion exchange column (Mono Q column) equilibrated with the same buffer. The

254 separation was carried out exactly as previously outlined except that 0.75mL fractions were collected.

255 The active fractions following the ion exchange purification second step were combined, desalted against

256 CPB (50 mM, pH 6.0) using EconoPac 10DG desalting columns and concentrated by ultrafiltration

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 12 of 39

257 (molecular weight cutoff 10 KDa, 4 mL UFC801024, Amicon ultra, Millipore, USA). The sample (50

258 µg) was loaded onto a Superdex 75 gel filtration column (Superdex 75, 75, 10/300 GL, GE Healthcare

259 Life Sciences, Sweden) equilibrated with CPB buffer containing NaCl (150 mM) and sodium azide

260 (0.02%). Fractions (0.35 mL) were collected every 1 min at flow rate of 0.35 mL/min and assayed for

261 myrosinase activity. Molecular weight markers 12 79 kDa (GE Healthcare Ltd) were used to calibrate

262 the column: cytochrome c (12.4 kDa), carbonic anhydrase (29 kDa), albumin (45 kDa), albumin bovine

263 serum (66 kDa) and transferrin (79 kDa).

264 SDS PAGE

265 Gels (13%) were prepared as previously described 32 and stained with SimplyBlue™ SafeStain (Life

266 Technologies) .

267 NTerminal sequencing and internal peptide sequencing

268 Pure protein was used for Nterminal sequencing (Edman degradation) which was carried out at the

269 Protein Sequence Analysis Facility, Department of Biochemistry, Cambridge (UK). Internal peptide

270 analysis was carried out by slicing out the pure protein from an SDSPAGE gel stained with

271 SimplyBlue™ SafeStain . The gel bands were processed and analysed by LCMS as previously

272 described. 33

273 The fulllength DNA coding sequence was obtained using Geneious Pro 5.6.6. Software.

274 GenBank accession numbers

275 Citrobacter _WYE1_Myr_KT821094; Citrobacter_ WYE1_16sRNA_KT825141

276

277

278

ACS Paragon Plus Environment Page 13 of 39 Journal of Agricultural and Food Chemistry

279

280

281

282

283 RESULTS

284 Strain isolation and identification

285 Serial dilutions were made from the enrichment culture and 100 L spread onto nutrient agar plates

286 followed by incubation at 30 oC. A single colony was picked and cultured in M9 medium with sinigrin as

287 the sole carbon source overnight. The pure isolate was identified by 16S rRNA sequence analysis. S_ab

288 scores of > 0.95 for twenty matches with known Citrobacter species were obtained and it was deduced

289 that the isolated organism belongs to the family Enterobacteriaceae and the genus Citrobacter and was

290 named Citrobacter WYE1 to indicate the origin of the isolate. It was not possible to assign Citrobacter

291 WYE1 to a specific species based on the 16S rRNA sequence. A phylogenetic tree was constructed to

292 show its relationship among selected Citrobacter strains (Figure 2). It is likely that Citrobacter WYE1 is

293 a new species.

294 Fermentation of sinigrin

295 The metabolism of sinigrin was monitored by HPLC and OD values were recorded over a 24 h period

296 (Figure 3). Sinigrin was completely utilized although no product could be identified. The potential

297 products allyisothiocyanate, or corresponding nitrile, carboxylic acid or amine could not be detected by

298 GCMS.

299 Properties of the myrosinase activity

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 14 of 39

300 Fresh cellfree protein extracts were used for all assays and were tested for activity prior to use. The

301 testing of activity was necessary as there were differences between preparations. An experiment was

302 carried out to determine whether the myrosinase was constitutive or inducible. Growth of Citrobacter

303 WYE1 in M9 media supplemented with glucose gave only low levels of myrosinase activity as did

304 nutrient broth. It was clear, however, that in the presence of sinigrin, myrosinase activity was

305 substantially induced (Figure 4). Sinigrin was stable over the 24 h period in M9 medium alone. The

306 approximate pI of the protein was between pH 67, based on binding of the enzyme to miniQ ion

307 exchange columns. Thus, for purification purposes, ion exchange buffers of pH 8 were used. The

308 temperature optimum was 25 oC for the myrosinase (Figure 5), which is unlike most plant and aphid

309 myrosinases where temperature optima are generally higher.17, 18 The pH optimum of the enzyme was

310 found to be 6, while at pH 7.6 the activity was markedly diminished (Figure 6). With this in mind, it was

311 necessary to desalt all purified fractions against CPB in order to obtain the full myrosinase activity. The

312 myrosinase was active towards three other glucosinolates tested: glucoerucin had similar specific activity

313 to sinigrin, while glucotropaeolin and glucoraphanin had lower values, suggesting that the nature of the

314 side chain is important (Figure 7). In addition to the GODPERID assays, the ITCs were identified by

315 GCMS. All of the substrates gave the expected ITCs and no other products such as nitriles were observed

316 (Supplementary information S2). GCMS analysis of the hydrolysis products was used to determine the

317 activation of myrosinase by ascorbate, as this compound interferes with the GODPERID assay. In

318 addition, gluconasturtiin, which results in phenethylisothiocyanate on hydrolysis, was used as the

319 substrate, since allylisothiocyanate produced by sinigrin is very volatile and can result in significant

320 losses during work up. The ascorbatetreated myrosinase assays resulted in an activation of 1.67 (Figure

321 8) which is very low in comparison to the plant enzymes. The apparent Km and Vmax were 0.46 mM and

322 4.91 mmol dm 3 min 1 mg 1 respectively (Figure 9). The Km value of the Citrobacter WYE1 myrosinase

323 was similar to those obtained for the aphid myrosinase 19, 20, 22 and some plant myrosinases.18

324 Purification of Citrobacter WYE1 myrosinase

ACS Paragon Plus Environment Page 15 of 39 Journal of Agricultural and Food Chemistry

325 The enzyme was purified by ion exchange chromatography (Supplementary information S1) on a high

326 resolution monoQ ion exchange column, followed by gel filtration The procedure gave a pure single band

327 with an approximate molecular mass of 66 kDa on analysis by SDSPAGE (Figure 10). The gel filtration

328 step using a high resolution Superdex 75 column gave a native molecular size of approximately 66 kDa,

329 based on GEHealth Care molecular weight markers. Thus it would appear that the bacterial myrosinase is

330 a monomeric protein. The myrosinase is however of low abundance, with a yield of 4.5 µg pure protein

331 from a fermentation of 500 ml (Table 1).

332 Sequence analysis

333 The Nterminal sequence (SIQSAQQPELGYDTV ) of the purified protein was determined (Figure 11a) as

334 well as peptides from tryptic digestion. The Nterminal peptide was used as a query to identify the

335 respective gene by tBLASTn against the genomic sequence reads obtained by Illumina. The DNA

336 sequences found were then assembled; the resulting consensus sequence was then used in iterative

337 searches of the nonassembled genome reads to extend the locus encoding the myrosinase gene, until a

338 fulllength coding sequence corresponding to a plausible ORF for myrosinase was obtained. No other

339 genes could be identified using the sequence data obtained, suggesting that there is only one myrosinase

340 gene. The validity of the DNA sequence was confirmed by carrying out PCR of the genomic DNA

341 (unpublished data).

342 BLAST analysis of the translated sequence of Citrobacter WYE1 myrosinase showed a high degree of

343 homology with a number of bacterial βOglucosidases (Figure 11b), the greatest being with a

344 hypothetical protein sequence from Citrobacter sp . 30_2. This bacterium was made available (Dr. Emma

345 AllenVercoe, University of Guelph) to us but showed no activity towards glucosinolates. There was no

346 significant homology with either plant or aphid myrosinases. Using InterProt 34 , the protein was shown to

347 belong to the GH 3 family of glucosidases. This is in contrast to known myrosinases which so far all

348 belong to GH 1.

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 16 of 39

349 The full length gene appears to code for an Nterminal signal peptide which presumably targets the

350 enzyme to the periplasm. This observation fits with the purified protein which appears to have lost a

351 peptide as evident from the actual Nterminal sequencing of the protein (Figure 11a).

352

353

354

355

356

357

358 DISCUSSION

359 The bacterial myrosinase is of low abundance in comparison with plant and aphid myrosinases 18, 20, 21 but

360 nonetheless is an active inducible enzyme (Figure 4). In contrast to both plant and insect myrosinases,

361 which appear to have a function in defence, the bacterial myrosinases are more likely to be involved in

362 scavenging glucose from glucosides. Citrobacter WYE1 is a soil microorganism which was isolated from

363 a field where oilseed rape had previously grown. The stubble and roots left in the field were ploughed

364 back into the soil which would consequently contain a significant amount of glucosinolate that would be

365 available for bacterial growth. It would seem therefore that bacteria can adapt 35 to changes in the

366 environment, and it is likely that other soil bacteria will also have the ability of the Citrobacter WYE1 to

367 metabolise glucosinolates. This is of importance in the field of biofumigation with Brassicaceae crops

368 where glucosinolate metabolism can yield the bioactive isothiocyanates, and is key to the success of this

369 technology. Thus bacteria that degrade glucosinolates can potentially counter the effects of biofumigation

370 by decreasing the available glucosinolate, although at this stage more work is required to establish the

ACS Paragon Plus Environment Page 17 of 39 Journal of Agricultural and Food Chemistry

371 products of hydrolysis in the soil.25, 36, 37 On the other hand, bacteria that can metabolise glucosinolates in

372 the human gut play a key role in the generation of the chemoprotective isothiocyanates.3, 28, 38

373 The Citrobacter WYE1 myrosinase belongs to the GH3 family of the glucosidases 9 and INTERPRO

374 analysis shows that the full length gene codes for a signal peptide, which presumably targets it to the

375 periplasm. In terms of sequence homology there is no resemblance to either plant or aphid myrosinases,

376 although there is high homology with other bacterial βOglucosidases (Figure 11b). A feature of the GH3

377 glucosidases is the conserved motif ‘SDW’ which contains aspartate as the nucleophile, as part of the

378 mechanism for hydrolysis.8, 39 The SWISSMODEL webpage tools 40 were used to generate models based

379 on available templates. The scores, however, were too low to obtain a meaningful model. With the GH3

380 glucosidases, the general acid/base catalytic component of the mechanism is much more variable and will

381 require structure determination to fully elucidate the mechanism, work that is currently underway. Recent

382 reports 41 have suggested that E. coli 0157:H7 possesses 6phosphoβglucosidases ( bglA and ascbB ) and,

383 following gene disruption, the sinigrin degradation capacity of the bacteria is substantially diminished.

384 This result 41 is supported by previous work 7 where it was not possible to isolate myrosinase activity in

385 cell free extracts of bacteria, despite the production of isothiocyanates and nitriles during fermentation. A

386 differential proteomic study of E. coli VL8 with or without sinigrin did not reveal any glucosidase

387 associated with glucosinolate hydrolysis.33 A glucosespecific phosphotransferase system was however

388 induced, which suggests that glucosinolates may undergo phosphorylation at the 6hydroxyl group of the

389 sugar residue. If this is the case, then it may well be that 6phosphoβglucosidases in cell free extracts

390 cannot recognize glucosinolates without prior phosphorylation or that they are membranebound and loss

391 of integrity results in loss of activity. This might explain why it has proven so difficult to show

392 myrosinase activity in cell free extracts of bacteria 4, 28 that can metabolise glucosinolates, although

393 enzyme instability cannot be ruled out. Thus a synthesis, either chemical or chemoenzymatic 42 , of a 6

394 phosphoglucosinolate would go some way to probing the mechanism of glucosinolate metabolism if 6

395 phosphoβglucosidases are involved. Clearly at this time there are several potential distinct mechanisms

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 18 of 39

396 for glucosinolate hydrolysis by bacteria that may involve phosphorylation, membranebound enzyme

397 systems and a more typical hydrolysis by a glucosidase as shown in our work.

398 It was not possible to identify the fermentation products at this time, although they were not any of the

399 usual glucosinolate hydrolysis products, i.e . nitrile, carboxylic acid, amine or thiocyanate (unpublished

400 data) that have been previously described.17, 18 Previous work 43 has shown isothiocyanates to be unstable

401 over a period of 24 h in phosphate buffers, although cell free extracts of the Citrobacter WYE1 produce

402 isothiocyanates in citrate phosphate buffer. Of course, it may also be possible that Citrobacter WYE1

403 possesses a detoxification pathway. M9 media contain not just phosphate but also ammonium ions which

404 likely add to the possible interactions with ITCs. Thus at this stage it is not clear what the actual

405 degradation metabolites are in vivo and more work is required on this topic. In conclusion, this is the first

406 report to describe a fully characterized bacterial myrosinase in terms of its properties and sequence.

407

408 ACKNOWLEDGEMENTS

409 Abdul Albaser was funded by the Libyan Government and Vijitra LuangIn by the Royal Thai

410 Government. We thank Dr Arnoud Van Vliet of the Institute of Food Research, Norwich, UK for help

411 with the bioinformatics.

412 SUPPORTING INFORMATION

413 A table describing the FPLC programme for the ion exchange chromatography is shown in S1.

414 Figures for the qualitative GCMS analysis of cell free extracts with sinigrin, glucoerucin, glucoraphanin,

415 glucotropaeolin, glucoraphanin and gluconasturtiin are shown in S2. This material is available free of

416 charge via the internet at http://pubs.acs.org.

417

418

ACS Paragon Plus Environment Page 19 of 39 Journal of Agricultural and Food Chemistry

419

420

421

422

423

424

425

426

427

428

429 REFERENCES

430 (1) Tani, N.; Ohtsuru, M.; Hata, T., Studies on bacterial myrosinase .1. Isolation of myrosinase producing

431 microorganism. Agric. Biol. Chem. 1974, 38 , 16171622.

432 (2) Tani, N.; Ohtsuru, M.; Hata, T., Studies on bacterial myrosinase .2. Purification and general

433 characteristics of bacterial myrosinase produced by enterobactercloacae. Agric. Biol. Chem. 1974, 38 ,

434 16231630.

435 (3) Saha, S.; Hollands, W.; Teucher, B.; Needs, P. W.; Narbad, A.; Ortori, C. A.; Barrett, D. A.; Rossiter,

436 J. T.; Mithen, R. F.; Kroon, P. A., Isothiocyanate concentrations and interconversion of sulforaphane to

437 erucin in human subjects after consumption of commercial frozen broccoli compared to fresh broccoli.

438 Mol. Nutr. Food Res. 2012, 56 , 19061916.

439 (4) Palop, M. L.; Smiths, J. P.; Tenbrink, B., Degradation of sinigrin by lactobacillusagilis strain r16. Int.

440 J. Food Microbiol. 1995, 26 , 219229.

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 20 of 39

441 (5) Rabot, S.; Nugonbaudon, L.; Raibaud, P.; Szylit, O., Rapeseed meal toxicity in gnotobioticrats

442 influence of a whole human fecal flora or single human strains of escherichiacoli and bacteroides

443 vulgatus. Br. J. Nutr. 1993, 70 , 323331.

444 (6) Nugonbaudon, L.; Rabot, S.; Wal, J. M.; Szylit, O., Interactions of the intestinal microflora with

445 glucosinolates in rapeseed meal toxicity 1st evidence of an intestinal lactobacillus possessing a

446 myrosinaselike activity invivo. J. Sci. Food Agric. 1990, 52 , 547559.

447 (7) LuangIn, V., Influence of Human Gut Microbiota on the Metabolic Fate of Glucosinolates. PhD

448 Thesis, Imperial College London 2013 .

449 (8) Ketudat Cairns, J. R.; Esen, A., betaGlucosidases. Cell. Mol. Life. Sci. 2010, 67 , 3389405.

450 (9) Cantarel, B. L.; Coutinho, P. M.; Rancurel, C.; Bernard, T.; Lombard, V.; Henrissat, B., The

451 CarbohydrateActive EnZymes database (CAZy): an expert resource for Glycogenomics. Nucleic Acids

452 Res. 2009, 37 , D233D238.

453 (10) Sansenya, S.; Opassiri, R.; Kuaprasert, B.; Chen, C.J.; Cairns, J. R. K., The crystal structure of rice

454 (Oryza sativa L.) Os4BGlu12, an oligosaccharide and tuberonic acid glucosidehydrolyzing beta

455 glucosidase with significant thioglucohydrolase activity. Arch. Biochem. Biophys. 2011, 510 , 6272.

456 (11) Mahn, A.; Angulo, A.; Cabanas, F., Purification and Characterization of Broccoli (Brassica oleracea

457 var. italica) Myrosinase (betaThioglucosidase Glucohydrolase). J. Agric. Food Chem. 2014, 62 , 11666

458 11671.

459 (12) Burmeister, W. P.; Cottaz, S.; Driguez, H.; Iori, R.; Palmieri, S.; Henrissat, B., The crystal structures

460 of Sinapis alba myrosinase and a covalent glycosylenzyme intermediate provide insights into the

461 substrate recognition and activesite machinery of an Sglycosidase. Structure. 1997, 5, 663675.

462 (13) Burmeister, W. P.; Cottaz, S.; Rollin, P.; Vasella, A.; Henrissat, B., High resolution xray

463 crystallography shows that ascorbate is a cofactor for myrosinase and substitutes for the function of the

464 catalytic base. J. Biol. Chem. 2000, 275 , 3938539393.

465 (14) Cottaz, S.; Henrissat, B.; Driguez, H., Mechanismbased inhibition and stereochemistry of

466 glucosinolate hydrolysis by myrosinase. Biochemistry. 1996, 35 , 1525615259.

ACS Paragon Plus Environment Page 21 of 39 Journal of Agricultural and Food Chemistry

467 (15) Dosz, E. B.; Ku, K.M.; Juvik, J. A.; Jeffery, E. H., Total Myrosinase Activity Estimates in Brassica

468 Vegetable Produce. J. Agric. Food Chem. 2014, 62 , 80948100.

469 (16) Hanschen, F. S.; Lamy, E.; Schreiner, M.; Rohn, S., Reactivity and Stability of Glucosinolates and

470 Their Breakdown Products in Foods. Angew. Chem., Int. Ed. 2014, 53 , 1143011450.

471 (17) Bones, A. M.; Rossiter, J. T., The enzymic and chemically induced decomposition of glucosinolates.

472 Phytochemistry. 2006, 67 , 10531067.

473 (18) Bones, A. M.; Rossiter, J. T., The myrosinaseglucosinolate system, its organisation and

474 biochemistry. Physiol. Plant. 1996, 97 , 194208.

475 (19) Pontoppidan, B.; Ekbom, B.; Eriksson, S.; Meijer, J., Purification and characterization of myrosinase

476 from the cabbage aphid (Brevicoryne brassicae), a brassica herbivore. Eur. J. Biochem. 2001, 268 , 1041

477 1048.

478 (20) Jones, A. M. E.; Bridges, M.; Bones, A. M.; Cole, R.; Rossiter, J. T., Purification and

479 characterisation of a nonplant myrosinase from the cabbage aphid Brevicoryne brassicae (L.). Insect

480 Biochem. Mol. Biol. 2001, 31 , 15.

481 (21) Jones, A. M. E.; Winge, P.; Bones, A. M.; Cole, R.; Rossiter, J. T., Characterization and evolution of

482 a myrosinase from the cabbage aphid Brevicoryne brassicae. Insect Biochem. Mol. Biol. 2002, 32 , 275

483 284.

484 (22) Husebye, H.; Arzt, S.; Burmeister, W. P.; Hartel, F. V.; Brandt, A.; Rossiter, J. T.; Bones, A. M.,

485 Crystal structure at 1.1 angstrom resolution of an insect myrosinase from Brevicoryne brassicae shows its

486 close relationship to betaglucosidases. Insect Biochem. Mol. Biol. 2005, 35 , 13111320.

487 (23) Higdon, J. V.; Delage, B.; Williams, D. E.; Dashwood, R. H., Cruciferous vegetables and human

488 cancer risk: epidemiologic evidence and mechanistic basis. Pharmacol. Res. 2007, 55 , 224236.

489 (24) Neubauer, C.; Heitmann, B.; Mueller, C., Biofumigation potential of Brassicaceae cultivars to

490 Verticillium dahliae. Eur. J. Plant Pathol. 2014, 140 , 341352.

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 22 of 39

491 (25) Ngala, B. M.; Haydock, P. P. J.; Woods, S.; Back, M. A., Biofumigation with brassica juncea,

492 raphanus sativus and eruca sativa for the management of field populations of globodera pallida. J.

493 Nematol. 2014, 46 , 210211.

494 (26) Sotelo, T.; Lema, M.; Soengas, P.; Cartea, M. E.; Velasco, P., In Vitro Activity of Glucosinolates

495 and Their Degradation Products against BrassicaPathogenic Bacteria and Fungi. Appl. Environ.

496 Microbiol. 2015, 81 , 432440.

497 (27) Taylor, F. I.; Kenyon, D.; Rosser, S., Isothiocyanates inhibit fungal pathogens of potato in in vitro

498 assays. Plant Soil. 2014, 382 , 281289.

499 (28) LuangIn, V.; Narbad, A.; NuenoPalop, C.; Mithen, R.; Bennett, M.; Rossiter, J. T., The metabolism

500 of methylsulfinylalkyl and methylthioalkyl glucosinolates by a selection of human gut bacteria. Mol.

501 Nutr. Food Res. 2014, 58 , 875883.

502 (29) Nurk, S.; Bankevich, A.; Antipov, D.; Gurevich, A. A.; Korobeynikov, A.; Lapidus, A.; Prjibelski,

503 A. D.; Pyshkin, A.; Sirotkin, A.; Sirotkin, Y.; Stepanauskas, R.; Clingenpeel, S. R.; Woyke, T.; McLean,

504 J. S.; Lasken, R.; Tesler, G.; Alekseyev, M. A.; Pevzner, P. A., Assembling SingleCell Genomes and

505 MiniMetagenomes From Chimeric MDA Products. J. Comput. Biol. 2013, 20 , 714737.

506 (30) Treangen, T. J.; Ondov, B. D.; Koren, S.; Phillippy, A. M., The Harvest suite for rapid coregenome

507 alignment and visualization of thousands of intraspecific microbial genomes. Genome biology 2014, 15 ,

508 524.

509 (31) Huggett, A. S.; Nixon, D. A., Use of glucose oxidase, peroxidase, and Odianisidine in determination

510 of blood and urinary glucose. Lancet (London, England). 1957, 273 , 36870.

511 (32) James, D. C.; Rossiter, J. T., Development and characteristics of myrosinase in brassicanapus during

512 early seedling growth. Physiol. Plant. 1991, 82 , 163170.

513 (33) LuangIn, V.; Narbad, A.; Cebeci, F.; Bennett, M.; Rossiter, J. T., Identification of Proteins Possibly

514 Involved in Glucosinolate Metabolism in L. agilis R16 and E. coli VL8. Protein J. 2015, 34 , 13546.

515 (34) Jones, P.; Binns, D.; Chang, H. Y.; Fraser, M.; Li, W.; McAnulla, C.; McWilliam, H.; Maslen, J.;

516 Mitchell, A.; Nuka, G.; Pesseat, S.; Quinn, A. F.; SangradorVegas, A.; Scheremetjew, M.; Yong, S. Y.;

ACS Paragon Plus Environment Page 23 of 39 Journal of Agricultural and Food Chemistry

517 Lopez, R.; Hunter, S., InterProScan 5: genomescale protein function classification. Bioinformatics. 2014,

518 30 , 123640.

519 (35) Diaz, E., Bacterial degradation of aromatic pollutants: a paradigm of metabolic versatility. Int.

520 Microbiol. 2004, 7, 173180.

521 (36) Gimsing, A. L.; Kirkegaard, J. A., Glucosinolate and isothiocyanate concentration in soil following

522 incorporation of Brassica biofumigants. Soil Biol. Biochem. 2006, 38 , 22552264.

523 (37) Gimsing, A. L.; Kirkegaard, J. A., Glucosinolates and biofumigation: fate of glucosinolates and their

524 hydrolysis products in soil. Phytochem. Rev. 2009, 8, 299310.

525 (38) Mullaney, J. A.; Kelly, W. J.; McGhie, T. K.; Ansell, J.; Heyes, J. A., Lactic Acid Bacteria Convert

526 Glucosinolates to Nitriles Efficiently Yet Differently from Enterobacteriaceae. J. Agric. Food Chem.

527 2013, 61 , 30393046.

528 (39) Withers, S. G., Mechanisms of glycosyl transferases and hydrolases. Carbohydrate Polymers 2001,

529 44 , 325337.

530 (40) Biasini, M.; Bienert, S.; Waterhouse, A.; Arnold, K.; Studer, G.; Schmidt, T.; Kiefer, F.; Cassarino,

531 T. G.; Bertoni, M.; Bordoli, L.; Schwede, T., SWISSMODEL: modelling protein tertiary and quaternary

532 structure using evolutionary information. Nucleic Acids Res. 2014, 42 , W252W258.

533 (41) Cordeiro, R. P.; Doria, J. H.; Zhanel, G. G.; Sparling, R.; Holley, R. A., Role of glycoside hydrolase

534 genes in sinigrin degradation by E. coli O157:H7. Int. J. Food Microbiol. 2015, 205 , 105111.

535 (42) Thompson, J.; Lichtenthaler, F. W.; Peters, S.; Pikis, A., betaglucoside kinase (BglK) from

536 Klebsiella pneumoniae Purification, properties, and preparative synthesis of 6phosphobetaD

537 glucosides. J. Biol. Chem. 2002, 277 , 3431034321.

538 (43) Mays, J. R.; Roska, R. L. W.; Sarfaraz, S.; Mukhtar, H.; Rajski, S. R., Identification, synthesis, and

539 enzymology of nonnatural glucosinolate chemopreventive candidates. ChemBioChem. 2008, 9, 729747.

540

541

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 24 of 39

542

543

544

545

546

547

548

549

550

551

552 Figure Legends.

553 Figure 1. Generalised scheme of the hydrolysis of glucosinolates by myrosinase. RNCS, isothiocyanate;

554 RCN, nitrile; RSCN, thiocyanate; ETN, epithionitriles.

555 Figure 2. Phylogenetic tree of Citrobacter WYE1 compared with related Citrobacter species. Draft

556 genomes of Citrobacter species were obtained from Genbank WGS projects (accession prefix

557 supplied). Core genome single nucleotide polymorphisms (SNPs) in draft genomes were used to

558 prepare the phylogenetic tree using the parSNP program.

559 Figure 3. The metabolism of sinigrin (10 mM) in M9 media by Citrobacter WYE1 over 24 h. The

560 metabolism of sinigrin was monitored by HPLC and OD600nm recorded.

ACS Paragon Plus Environment Page 25 of 39 Journal of Agricultural and Food Chemistry

561 Figure 4. The induction of myrosinase activity in cultures grown with sinigrin. Cell free protein extracts

562 were prepared as described in the methods and assayed for myrosinase activity with 5 µg of crude

563 desalted protein extract/assay.

564 Figure 5. The effect of temperature on the activity of the Citrobacter WYE1 myrosinase was measured

565 over a range of temperatures (570 ºC). Assays were carried out as described in the methods with 5µg

566 crude cell free protein.

567 Figure 6. The pH optimisation of the enzymatic degradation of sinigrin. Crude protein extract (5 µg) was

568 added to a mixture of sinigrin (2 mM) in CPB to a final volume of 300 µL at specific pH values over the

569 range 3.6 7.6 and incubated at 37 °C for 2 h. Activity was measured by determining the release of

570 glucose using the GODPERID assay.

571 Figure 7. The enzymatic degradation of a range of glucosinolates (10 mM) in CPB. Activity was

572 measured by determining the release of glucose using the GODPERID assay.

573 Figure 8. The effect of differing concentrations of ascorbate on the enzymatic activity of myrosinase.

574 Activity was determined by monitoring the production of phenethylisothiocyanate using GCMS. ● =

575 ascorbate, ■ = control.

576 Figure 9. The determination of the apparent Km and Vmax of the myrosinase.

577 Figure 10. SDS PAGE analysis of the purification steps of Citrobacter WYE1 myrosinase; M= marker;

578 IE1= Ion exchange first run; IE2= Ion exchange second run; GF= gel filtration fraction of pure

579 myrosinase at 66 kDa. Gels were washed three times with MilliQ H 2O (100 mL) and stained with

580 SimplyBlue SafeStain (Invitrogen LC6060) for an hour and destained in MilliQ H 2O.

581 Figure 11a. Full length protein based on the first ORF derived from the annotated gene. Putative signal

582 peptide is underlined, Nterminal sequence in bold and italics and internal peptide sequence in bold type.

583 Figure 11b. BLAST search results of the Citrobacter WYE1 showing the homology with a range of

584 bacterial βOglucosidases. The signature ‘ SDW’ of the GH3 glucosidases is highlighted.

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 26 of 39

585

586

587

588

589

590

591

592

593

594

595

596

597

ACS Paragon Plus Environment Page 27 of 39 Journal of Agricultural and Food Chemistry

Specific activity

Volume Protein (nmole/mg Total

Purification step (mL) (mg) protein/min) activity

Crude extract 10 40 611 24,444

Ion exchange 1st (Mono Q

column ) 2.5 0.775 1111 861

Ion exchange 2nd (Mono Q

column ) 0.5 0.05 593 29.6

Gel filtration (Superdex 75

column) 0.35 0.0045 2686 12.1

Table 1 Summary of the purification steps of Citrobacter myrosinase. Activity a ssays were carried out

with sinigrin (10 mM) at 25°C (pH 6.0, CPB 20 mM) and are based on glucose release (GODPerid assay).

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 28 of 39

Figure 1

ACS Paragon Plus Environment Page 29 of 39 Journal of Agricultural and Food Chemistry

Figure 2

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 30 of 39

100 0.8

80 0.6 60 0.4 40 0.2 20

0 0.0 0 4 8 12 16 20 24 28 h

Figure 3

ACS Paragon Plus Environment Page 31 of 39 Journal of Agricultural and Food Chemistry

Figure 4

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 32 of 39

Figure 5

ACS Paragon Plus Environment Page 33 of 39 Journal of Agricultural and Food Chemistry

Figure 6

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 34 of 39

Figure 7

ACS Paragon Plus Environment Page 35 of 39 Journal of Agricultural and Food Chemistry

Figure 8

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 36 of 39

Figure 9

ACS Paragon Plus Environment Page 37 of 39 Journal of Agricultural and Food Chemistry

Figure 10

ACS Paragon Plus Environment Journal of Agricultural and Food Chemistry Page 38 of 39

MLTAFKMNTVSLAVLVCLPLSVSA SIQSAQQPELGYDTV PLLHLSGLSFKDLNRDGKLNPYEDWRLSPQTRAADLVKRMSVAEKAGVM

MHGTAPAEGSTFGNGSVYDSEATRKMVVDAHVNSLITRLNGEEPARLAEQNNMIQKTAETTRLGIPVTISTDPR NSYQALVGISNPAGK

FTQWPEAIGLGAAGSEALAQEYADHVRREYR AVGITEALSPQADITTEPR WARISGTFGEDPELAKKLVRGYITGMQKGGQGLNPQSV

AAVVK HWVGYGAAEDGWDGHNAYGK HTVLSNESLQKHIIPFRGAFEANVAAVMPTYSVMKGVTWNGRETEQVAAGFSHFLLTDLL

RKQNNFSGVIISDWLITNDCDDECVNGSAPGKKPVAGGMPWGVESLSKERRFVKAVNAGIDQFGGVTDSAVLVTAVEKGLITQARLDA

SVERILQQKFELGLFEQPYVDAKLAEKIVGAPDTKKAADDAQFRTLVLLQNKNILPLKPGTKVWLYGADKSAAEKAGLEVVSEPEDAD

VALMRTSAPFEQPHYNYFFGRRHHEGSLEYREDNKDFAVLKRVSKHTPVIMTMYMERPAVLTNVTDKTSGFIANFGLSDEVFFSRLTSD

TPYTARLPFALPSSMASVLKQKSDEPDDLDTPLFQRGFGLTR

Figure 11a

361 420 Citrobacter WYE1 VII SDW LITNDCDDECVNGSAPGKKPVAGGMPWGVESLSKERRFVKAVNAGIDQFGGVTD Enterobacter cloacae βglucosidase(70%) LII SDW LITADCDEECVNGTTGDKKPVPGGMPWGVEHLSKEQRFVKAVNAGIDQFGGVTD Citrobacter sp.30_2 hypothetical protein(71%) VII SDW LITNDCDDECIHGAPGGKKPVTGGMPWGVESLTQEQRFVKAVQAGIDQFGGVTD Escherichia vulneris βglucosidase (72%) VII SDW LITNDCDNECLNGAPEGKKPVTGGMPWGVESLTKEQRFVKAIHAGIDQFGGVTD Rahnella aquatilis βglucosidase VIL SDW LITSNCEGECLNGSPEGKEPVPGGMPWGVENLTPQQRFVKAVLAGVDQFGGVTN Pectobacterium carotovorum βglucosidase(60%) VIL SDW LITNDCKDDCVTGAKPNEKPVPRGMPWGVENLTVEQRFIKAVEAGVDQFGGVTN Serratia sp 1,4 βxylosidase (61%) VIL SDW LITQDCKADCINGFKPGEKPVPRGMPWGVEHMTVDQRFIQAVKAGIDQFGGVTD Dickeya solani βxylosidase periplasmic (61%) VIL SDW LITNDCKGDCLTGVKPGEKPVPRGMPWGVENLTPAERFIKAVNAGVDQFGGVTD

Figure 11b

ACS Paragon Plus Environment Page 39 of 39 Journal of Agricultural and Food Chemistry

Table of Contents Graphic 334x202mm (96 x 96 DPI)

ACS Paragon Plus Environment