Aspartic Acid Protease from Botrytis cinerea Removes Haze-Forming

Subscriber access provided by Libraries of the | University of Hawaii at Manoa

Article

An aspartic acid protease from Botrytis cinerea removes haze forming proteins during white winemaking Steven C Van Sluyter, Nicholas I Warnock, Simon Schmidt, Peter A. Anderson, Jan A.L. van Kan, Antony Bacic, and Elizabeth J Waters J. Agric. Food Chem., Just Accepted Manuscript • DOI: 10.1021/jf402762k • Publication Date (Web): 06 Sep 2013 Downloaded from http://pubs.acs.org on September 24, 2013

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 25

Journal of Agricultural and Food Chemistry

An aspartic acid protease from Botrytis cinerea removes haze forming proteins during white winemaking

Steven C. Van Sluytera,b,c,*, Nicholas I. Warnockd, Simon Schmidta, Peter Andersond, Jan A.L. van Kane, Antony Bacicb,f, Elizabeth J. Watersa,g

a

The Australian Wine Research Institute, P.O. Box 197, Glen Osmond, SA 5064,

Australia b

c

School of Botany, University of Melbourne, VIC 3010, Australia

Current address: Department of Biological Sciences, Macquarie University, NSW

2109, Australia d

School of Biological Sciences, Flinders University, SA 5042, Australia

e

Laboratory of Phytopathology, Wageningen University, P.O. Box 8025, 6700EE

Wageningen, The Netherlands f

ARC Centre of Excellence in Plant Cell Walls and Bio21 Molecular Science and

Biotechnology Institute, The University of Melbourne, VIC 3010, Australia g

Current address: Grape and Wine Research and Development Corporation, P.O.

Box 610, Kent Town, SA 5071, Australia

*Corresponding author (Tel:+61 412 653 966; Fax: +61 2 9850 8245; E-mail: [email protected]) 1 ACS Paragon Plus Environment


1

Abstract

2

White wines suffer from heat-induced protein hazes during transport and storage

3

unless the proteins are removed prior to bottling. Bentonite fining is by far the most

4

commonly used method, but it is inefficient and creates several other process

5

challenges. An alternative to bentonite is the enzymatic removal of haze-forming

6

grape pathogenesis-related proteins using added proteases. The major problem with

7

this approach is that grape pathogenesis-related proteins are highly protease

8

resistant unless they are heat denatured in combination with enzymatic treatment.

9

This paper demonstrates that the protease BcAP8, from the grape fungal pathogen

10

Botrytis cinerea, is capable of degrading chitinase, a major class of haze forming

11

proteins, without heat denaturation. Because BcAP8 effectively removes haze-

12

forming proteins under normal winemaking conditions, it could potentially benefit

13

winemakers by reducing bentonite requirements.

14

Keywords

15

Wine protein haze, Botrytis cinerea, aspartic protease, pathogenesis-related protein,

16

chitinase, Vitis vinifera, bentonite alternative

2 ACS Paragon Plus Environment

Page 2 of 25

Page 3 of 25

17


Introduction

18

Vitis vinifera grape berries contain high levels of pathogenesis-related (PR)

19

proteins that persist through winemaking and cause commercially unacceptable

20

heat-induced hazes if not removed before bottling.1 The majority of commercial

21

wineries use the clay cation exchanger bentonite to remove PR-proteins from wine,

22

but the process results in wine losses, is laborious, and can negatively affect wine

23

sensory properties.2

24

An alternative to protein removal by bentonite is the enzymatic hydrolysis of

25

PR-proteins by proteases. However, PR-proteins are highly resistant to enzymatic

26

proteolysis and no commercially viable enzymatic methods are currently available for

27

use at typical winemaking temperatures. A recent enzymatic alternative to bentonite

28

uses the combination of a heat-tolerant fungal protease preparation with flash

29

pasteurization.3 Combined heat denaturation of PR-proteins and enzymatic

30

hydrolysis effectively reduces or prevents wine protein haze, but incurs the costs of

31

specialized equipment and heating energy. An ideal enzymatic alternative to

32

bentonite would not require heating. The present study demonstrates the

33

effectiveness of a fungal protease in removing haze-forming proteins during

34

winemaking at typical temperatures.

35

The secretion of proteases by Botrytis during infection has been extensively

36

observed.4-6 More recent studies have identified the secreted proteases as primarily

37

aspartic proteases,6-9 a serine protease, and an unusual glutamic acid protease.10

38

Marchal et al.11 proposed that Botrytis proteases are responsible for negatively

39

affecting sparkling wine foam properties by removing proteins and subsequently

40

characterized wine protein changes induced by Botrytis infection, suggesting a role

41

of Botrytis proteases in wine protein degradation.12-14 However, a direct causal



42

relationship between Botrytis proteases and wine protein content was not

43

demonstrated.

44

The aspartic acid family of Botrytis protease has been characterized by

45

sequence analysis and expression studies,6,15 but without evidence for a biological

46

role of the enzymes. Ten Have et al.15 characterized the B. cinerea aspartic acid

47

proteases by sequence analysis and knock-out mutants. The mutants were grown on

48

several types of plant materials and in the presence of grape PR-proteins on agar

49

plates, but no phenotypic differences were observed between the wild-type and

50

mutants.

51

One of the findings presented in ten Have et al.15 was that a particular

52

aspartic protease, BcAP8, is the predominant protein secreted into liquid media.

53

Even though BcAP8 knock-out mutants did not display a different phenotype, the

54

high levels of production of the enzyme in the wild type were the motivation for the

55

current study.

56

In this study, BcAP8 was added to juice prior to inoculation with yeast and

57

successfully removed chitinases under normal winemaking temperatures and

58

conditions. Thaumatin-like proteins (TLPs) were mostly unaffected. Even though

59

BcAP8 only partially reduced PR-protein levels in the finished wines, the impact on

60

heat-induced haze was significant. Chitinases have recently been shown to

61

contribute to wine haze more so than TLPs.16-18 It is possible that by selectively

62

removing chitinases from wines, BcAP8 could either heat stabilize wines or at least

63

dramatically reduce the requirement for bentonite treatment.


Page 4 of 25

Page 5 of 25


64

Materials and Methods

65

Botrytis aspartic protease BcAP8 heterologous production in Pichia pastoris

66

BcAP8 was transformed into, and produced in Pichia pastoris according to the

67

methods of Schmidt et al.19 The predicted BcAP8 amino acid sequence from ten

68

Have et al.15 was used to create the corresponding cDNA, which was codon

69

optimized for expression in P. pastoris and synthesized by GeneArt (Regensburg,

70

Germany). Not I and Xho I restriction sites were included in the synthetic cDNA and

71

used by GeneArt for insertion in a pPICZα C vector (Invitrogen). The expression

72

vector and BcAP8 gene sequence are described in Supporting Information Figures

73

S1 and S2. A successful P. pastoris transformant was selected by zeocin resistance,

74

grown, and BcAP8 expression was induced with methanol addition, according to

75

Schmidt et al.19 Secreted BcAP8 in synthetic medium PNB120 was acidified to

76

promote autohydrolysis of the propeptide inhibitor domain. Acidified, activated

77

BcAP8 was 0.17 mg protein/mL and produced a single band by SDS-PAGE and was

78

not purified further. Culture supernatant from the same P. pastoris strain transformed

79

with the empty expression vector was used as a control on a volume basis.

80

Assessment of BcAP8 activity in grape juice

81

BcAP8 and pepsin (Sigma) were added at 5 mg/L to Australian Semillon and

82

Sauvignon blanc juices (200 µL per reaction + 1 mM sodium azide to prevent

83

microbial growth). Pepsin (Sigma-Aldrich, Australia, P6887) was from a 0.1 mg/mL

84

stock solution (A278, ε = 51,300 M-1 cm-1, 34,620 Da, values from Sigma product

85

information) in 50 mM KH2PO4, pH 2 with H3PO4. Culture supernatant from empty

86

vector transformed P. pastoris was used at the same rate as BcAP8 (34-fold dilution)

87

as a negative control. Pepstatin A (2 mM in DMSO) was added to inhibitor

88

treatments at 20 µM; 1% DMSO was added to treatments without pepstatin. Once all 5 ACS Paragon Plus Environment


89

components were added to the juices, 100 µL of each was used for reactions at 22

90

°C for 21 days and the remainder incubated at 40 °C for 18 hrs. Following

91

incubation, total protein was methanol/chloroform extracted according to Wessel and

92

Flügge21 and used for SDS-PAGE with NuPage gels according to manufacturer

93

instructions (Invitrogen). Gels were stained with Coomassie Brilliant Blue R-250

94

(Pierce Imperial Stain, Thermo Scientific) according to manufacturer instructions.

95

The lower of the two major bands in the 20-25 kDa range was considered putative

96

TLP, the higher of the two bands putative chitinase, based on previous studies in

97

which SDS-PAGE bands were subjected to protein identification by mass

98

spectrometry.3,18,22

99

Assessment of BcAP8 activity in fermentations

100

Yeast cultures were started with Zymaflore VL3 yeast (Laffort) added to 5 mL

101

Yeast Peptone Dextrose broth, followed by incubation with mixing at 30 °C for 24

102

hrs. The VL3 culture was added to 10 mL synthetic juice (1x Difco Yeast Nitrogen

103

Base (Becton Dickinson), 100 g/L glucose, 100 g/L fructose, 4 g/L tartaric acid, 4 g/L

104

malic acid, adjusted to pH 3.5 with KOH) supplemented with 10 g/L ammonium

105

sulphate and that mixture was incubated at 28 °C in a 50 mL flask on an orbital

106

shaker at 180 RPM overnight.

107

Two Australian Vitis vinifera juices, a Sauvignon blanc (21.8 Brix, pH 3.36,

108

titratable acidity to pH 8.2: 5.2 g/L as tartaric acid) and a Semillon (18.8 Brix, pH

109

3.30, TA: 5.3 g/L), were syringe filtered (32 mm 0.8/0.2 µm Acrodisc PF filters, Pall

110

Corporation, Australia) and used for BcAP8 ferments. To account for the possibility

111

that either phenolics or other components of grape juice might inhibit BcAP8, total

112

juice protein ammonium sulphate precipitated (80% saturation) from a second

113

Semillon juice was added to 300 mg/L in synthetic juice. BcAP8 was added at 5


Page 6 of 25

Page 7 of 25


114

mg/L to the three musts while they were on ice; culture filtrate from empty vector

115

transformed Pichia was added in equal volume (34-fold dilution) to controls. The

116

musts were then incubated at 17 °C for 17 hrs, and then inoculated with 106 cells/mL

117

from the yeast starter cultures. Diammonium phosphate was added to the two

118

authentic juices to increase available nitrogen by 150 mg/L. Inoculated musts were

119

split among three formats: 8 wells per treatment of 200 µL each in clear 96-well

120

plates for biomass assays (A600); 4 wells per treatment of 1.5 mL in a 96-well 2 mL

121

Masterblock (Greiner Bio One, Australia) for enzymatic sugar assays; 4 mL of

122

Sauvignon blanc and Semillon, 3 mL for the synthetic juice treatments, all in

123

triplicate, in 5 mL polypropylene screw cap sample tubes (Sarstedt).

124

The 96-well plates were sealed with Breathe-Easy membranes (Sigma-

125

Aldrich), the screw caps left loose, and the ferments incubated at 18 °C, 2.3-2.5%

126

O2, in a Cytomat incubator (Thermo Scientific) interfaced with a Freedom Evo liquid

127

handling robot (Tecan Australia). The robot was configured to measure A600 of the

128

200 µL ferments every 2 hrs for the first 48 hrs. Samples were taken manually from

129

the Masterblock ferments at 2, 3, 4 and 9 days, frozen, and glucose and fructose

130

assayed enzymatically with a hexokinase method enzyme kit (Roche Diagnostics,

131

Australia).23

132

Protein analysis by reversed-phase (RP) HPLC

133

Ferments conducted in the 5 mL sample vials were analysed, and TLPs and

134

chitinases quantified as cytochrome C equivalents,24 by RP-HPLC 14 days post-

135

inoculation according to Van Sluyter et al.,22 but with an increased flow rate of 0.5

136

mL/min. Following storage at 4 °C, samples were analysed again by RP-HPLC 360

137

days post-inoculation. Peaks eluting between 6 and 9 min were considered TLPs,



138

and peaks between 11-15 min considered chitinases, based on chromatograms of

139

purified proteins and well characterized juices and wines in previous studies.3,22,24

140

Heat stability tests

141

Heat tests were performed on 200 µL samples in a 96-well plate with silicon

142

mat lid at 55 °C in a humid chamber for 18 hrs, followed by cooling on ice for 5 hrs.

143

Haze was expressed in mAU as A540 following heating minus A540 measured prior to

144

heating. Samples with and without 20 µM pepstatin were incubated to test for

145

residual BcAP8 activity.

146

Statistical analysis

147

Haze and protein concentration data were analyzed by one-way ANOVA followed by

148

Tukey’s test using InStat 3.1 (GraphPad Software, Inc.). P values less than 0.01

149

were considered significant.

150

Results and Discussion

151

Recombinant BcAP8 effects on grape proteins

152

Several studies have demonstrated that wines from Botrytis infected grapes contain

153

lower levels of PR-proteins, raising the possibility of Botrytis proteases acting on

154

grape PR-protein substrates.11-14 Because BcAP8 is the most abundant B. cinerea

155

protein in liquid culture,15 the gene for BcAP8 was expressed in Pichia to investigate

156

its activity directly against grape proteins. BcAP8 was secreted in sufficient quantities

157

so that purification was not required. Because BcAP8 was added to juice as culture

158

supernatant, control treatments contained equivalent volumes of culture supernatant

159

from P. pastoris transformed with the empty expression vector. As a comparison to

160

BcAP8, pig pepsin was considered an example of a typical aspartic protease

161

because it is commercially available and its activity against wine haze proteins has 8 ACS Paragon Plus Environment

Page 8 of 25

Page 9 of 25


162

been previously studied.25,26 Although the pH optimum of pepsin is often considered

163

to be pH 2 or less, the actual optimum is pH 3.5,27 the approximate pH of grape juice

164

and wine.

165

Grape PR-proteins have been shown to be resistant to hydrolysis by fungal

166

proteases28 and pepsin has been demonstrated to be ineffective in preventing wine

167

haze, although it does remove some proteins at 37 °C.25 Pocock et al.26

168

demonstrated that pepsin and fungal protease are effective in removing PR-proteins

169

from wine at 90 °C. As demonstrated by Marangon et al.,3 denaturing PR-proteins

170

with heat treatment at 75 °C to make them susceptible to fungal proteases is an

171

effective way to remove them, but an ideal protease treatment should be effective at

172

winemaking temperatures.

173

In the present work, pepsin had no effect against grape PR-proteins at 22 °C,

174

but did hydrolyse an apparent chitinase at 40 °C; proteolysis that was inhibited by

175

pepstatin (Figure 1). In comparison, BcAP8 demonstrated pepstatin-sensitive

176

proteolysis of the same apparent chitinase at both 22 °C and 40 °C, demonstrating

177

that it could be an effective enzyme without heat denaturing PR-proteins, i.e. under

178

typical winemaking conditions. In both the 22 °C and 40 °C treatments, BcAP8

179

removed proteins associated with bands at 12 kDa and 60 kDa, possibly LTP and

180

invertase, respectively, although pepsin did not. Because LTP and invertase are

181

involved in promoting positive sparkling wine foam properties,12,14 it is possible that

182

the proteolytic activity of BcAP8 demonstrated in Figure 1 is responsible for the

183

reduced foam of sparkling wines from Botrytis infected grapes. In the gels in Figure

184

1, BcAP8 appeared as a band at 36 kDa, but pepsin did not, which is typical of

185

pepsin because it binds an unusually low amount of Coomassie stain per molecule.29



186

BcAP8 effects on grape proteins under winemaking conditions

187

Although the addition of BcAP8 to sparkling wine could be detrimental to foam

188

properties, it could be beneficial in still wines. To test the ability of BcAP8 to reduce

189

PR-proteins during winemaking, the protein was added to two grape juices and one

190

synthetic juice including crude grape juice protein. The juices were inoculated and

191

fermented at 17 °C. Figure 2 contains the RP-HPLC chromatograms of the resulting

192

wines. In both of the authentic juices BcAP8 effectively eliminated chitinase and very

193

slightly reduced the levels of TLPs. In the model juice a similar trend was evident,

194

but some of the major chitinase remained.

195

SDS-PAGE results for the same wines are shown in Figure 3. BcAP8

196

appeared as a band slightly lower than the 37 kDa standard. Below that, at

197

approximately 35 kDa, was a potential glucanase that was removed from Semillon

198

and Sauvignon Blanc by BcAP8, but was less affected in synthetic must. Reductions

199

in chitinases were apparent in the authentic musts, with slight reductions in TLPs.

200

Chitinase reduction in the synthetic must was more dramatic than in the authentic

201

juices and the reduction in what could be a TLP, the band slightly higher than the 20

202

kDa standard, was substantial.

203

Unlike the results in uninoculated juice, BcAP8 did not remove the band at 60

204

kDa. BcAP8 removed one potential LTP at approximately 13 kDa, which was

205

particularly evident in the Semillon and synthetic must treatments. It is possible that

206

either the higher temperatures of the juice experiments (22 and 40 °C) promoted

207

BcAP8 activity against invertase or that BcAP8 becomes less active as stabilizing

208

sugars are converted to denaturing alcohol. In that case, the longer exposure time of

209

BcAP8 to substrates in uninoculated juice (3 weeks at 22 °C), versus the shorter

210

exposure time at 17 °C before alcohol became inhibitory, could explain the presence


Page 10 of 25

Page 11 of 25


211

of potential invertase in the wines. In Botrytis infected fruit, however, BcAP8 is

212

possibly secreted into berries at high levels for substantial amounts of time before

213

the fruit is picked, let alone inoculated with yeast, and that exposure of grape

214

proteins to BcAP8 could potentially affect foaming properties in sparkling wines.

215

To test the long term, possibly protective effect of BcAP8 on protein instability,

216

wines were stored at 4 °C for one year and analysed a second time by RP-HPLC.

217

Results for TLPs and chitinases are shown in Figure 4. Among the controls, the

218

changes in TLPs and chitinases were minimal. In contrast, the BcAP8 treatments, all

219

of which have lower levels of TLPs and much lower chitinase levels shortly after

220

fermentation, showed additional reductions in both classes of proteins and in all

221

wines over the year of storage. The continued activity of BcAP8 after fermentation

222

demonstrates that the enzyme is still active in the absence of sugar and in the

223

presence of ethanol.

224

Effects of BcAP8 on wine protein heat stability

225

Because BcAP8 added before inoculation remains active in the resulting wine,

226

at least for a short period of time, it is possible that it could hydrolyse proteins either

227

as they denature naturally or in response to heat during transport or storage. To

228

assess that possibility, small volumes of each wine, with and without the addition of

229

pepstatin, were heated to 55 °C for 18 hrs in a 96-well plate and the resulting hazes

230

measured spectrophotometrically in a plate reader (Figure 5). Pepstatin was

231

included in the 55 °C tests to establish at what point the BcAP8 activity was

232

destroying haze forming proteins, before the heating or during the heating. The

233

results were very similar between the BcAP8 treatments with and without pepstatin,

234

suggesting that BcAP8 was not active at the end of the year of storage.



235

Thus this experiment also effectively served as a heat stability test, except at

236

conditions less stringent than normal heat stability tests,30 but more closely

237

resembling maximum temperatures during wine transport and storage.31 There were

238

dramatic differences in haze between the empty expression vector controls and

239

BcAP8 treated wines, particularly in the synthetic must. In all the control treatments

240

hazes were visible; in the BcAP8 treatments, no hazes were visible even though the

241

turbidity (A540) values of the wines after heating were slightly more than the wines

242

before. The 55 °C for 18 hrs treatment simulated a real world scenario more closely

243

than 80 °C for 6 hrs, and the significant reduction in haze under those circumstances

244

illustrates the effectiveness of BcAP8 in reducing wine haze.

245

In all wines, and in the synthetic must treatment in particular, the differences

246

in turbidity between control and BcAP8 treatments seem disproportionate compared

247

to differences in protein amounts. All three BcAP8 wines contained more than 60

248

mg/L TLP, but produced no visible haze. On the other hand, the three BcAP8 wines

249

had much lower levels of chitinases than the controls. It could be the case that

250

chitinases contributed more to the hazes than TLPs in the controls; a result that

251

would reflect the lower heat stability of chitinases.16,17 Similarly, the potential

252

glucanase at 35 kDa in Figure 3, removed from the authentic wines and reduced in

253

the synthetic one, could contribute more to heat instability than TLPs.

254

The contribution of different types of proteins to the hazes in the control wines

255

was assessed by collecting the hazes by centrifugation and analysing by SDS-PAGE

256

(Figure 6), along with the supernatants and total protein of the BcAP8 wines. (There

257

was insufficient precipitated protein in BcAP8 wines for analysing soluble and

258

precipitated fractions separately.) The most apparent differences between the

259

proteins isolated from heat tests versus those from wine are the apparent low


Page 12 of 25

Page 13 of 25


260

molecular weight ( 0.01). Full statistical results are available in Supporting Information Table S2.

Figure 6. SDS-PAGE analysis of proteins following 55 °C, 18 hr heat tests. Control wines containing no BcAP8 were centrifuged and the resulting protein pellets used for the insoluble samples; supernatants were used for soluble samples. The BcAP8 wines did not contain visible haze and were not separated into fractions prior to SDS-PAGE. Top arrows: chitinases; bottom arrows: TLPs.


Page 18 of 25

Page 19 of 25


Figures

Figure 1



Figure 2


Page 20 of 25

Page 21 of 25


Figure 3



Figure 4


Page 22 of 25

Page 23 of 25


Figure 5



Figure 6


Page 24 of 25

Page 25 of 25


Graphic for the table of contents


Aspartic Acid Protease from Botrytis cinerea Removes Haze-Forming

Recommend Documents