Content uploaded by Christine Böttcher
Author content
All content in this area was uploaded by Christine Böttcher
Content may be subject to copyright.
Delaying Riesling grape berry ripening with a synthetic
auxin affects malic acid metabolism and sugar
accumulation, and alters wine sensory characters
Christine Böttcher
A
, Paul K. Boss
A
and Christopher Davies
A,B
A
CSIRO Plant Industry, PO Box 350, Glen Osmond, SA 5064, Australia.
B
Corresponding author. Email: christopher.davies@csiro.au
Abstract. An improved understanding of the hormonal control of grape (Vitis vinifera L.) berry ripening and the
ability to manipulate it are of interest scientifically and commercially. Grapes are nonclimacteric fruit with ethylene
unlikely to have a principal role in berry ripening but there are several other hormones thought to be involved. In this
work, a significant delay in Riesling berry ripening was achieved through preripening treatments with the synthetic
auxin 1-naphthaleneacetic acid (NAA). The initiation of sugar accumulation was delayed and the rate of sugar
accumulation was lower in NAA-treated fruit, resulting in a 15-day delay in harvest. NAA treatments also reduced the
rate of decline in malic acid levels that occurs during ripening, and increased the synchronicity of malic acid and
berry sugar accumulation. Sensory panel assessment revealed a significant difference between wine made from
control and NAA-treated fruit. Analysis of the volatile composition of the wines’headspace showed that the
concentration of several compounds was altered significantly by the NAA treatment. These data provide further
support for the involvement of auxins in inhibiting ripening and suggest that auxin treatments may be useful in
controlling both winery intake, and fruit and wine composition.
Additional keywords: fruit ripening, phytohormones, secondary metabolism.
Received 27 April 2012, accepted 10 July 2012, published online 20 August 2012
Introduction
Grape (Vitis vinifera L.) berry ripening is thought to be controlled
by several hormones, some of which promote ripening and
others that appear to inhibit ripening (Davies and Böttcher
2009). Increases in concentration coincident with ripening and
the promotion of ripening through their application provides
evidence that ABA, brassinosteroids and ethylene have a role
in promoting grape berry ripening (Hale et al. 1970; Chervin et al.
2004; Symons et al. 2006; Wheeler et al. 2009). For
brassinosteroids and ethylene, there is also evidence that
inhibitors of their synthesis or perception delay ripening
(Chervin et al. 2004; Symons et al. 2006). In contrast, the
concentration of auxins and cytokinins are high early in
development then decrease to be low before ripening; their
application has been shown to delay ripening (Zhang et al.
2003; Peppi and Fidelibus 2008; Böttcher et al. 2010,2011b).
There also appears to be considerable crosstalk between
hormone pathways, such as the induction of ethylene
biosynthesis by auxin application (Coombe and Hale 1973).
There is considerable interest in discovering how auxins act
in the control of berry ripening, what effects auxin treatments
may have on berry growth and metabolism, and how this may
be used to the advantage of the grape and wine industries.
The most abundant naturally occurring auxin in grape berries
is indole-3-acetic acid (IAA). Its level in berries decreases from a
maximum in flowers and young berries to be low at veraison
(‘veraison’is the word used by viticulturists meaning the
initiation of ripening) (e.g. Böttcher et al.2010). The high
levels of auxin early in development are thought to be
involved in cell division and expansion. The decrease in IAA
levels is thought to be a prerequisite for the ripening of a range
of fruit, including grapes (e.g. Given et al.1988; Buta and
Spaulding 1994; Davies et al.1997; Chen et al.1999;
Purgatto et al.2002; Böttcher et al.2010,2011a,2011b).
In agreement with the proposition that low auxin levels are a
requirement for the commencement of ripening, their
application during the preripening phase has been shown to
delay ripening in both climacteric and nonclimacteric fruit
(Vendrell 1969; Frenkel and Dyck 1973; Tingwa and Young
1975; Given et al.1988; Cohen 1996; Purgatto et al.2002;
Amorós et al.2004; Kondo et al.2004; Fabbroni et al.2006;
Villarreal et al.2009). Ripening of grape berries is also delayed
by the preveraison application of IAA or synthetic auxin
analogues such as 1-naphthaleneacetic acid (NAA) and
benzothiazole-2-oxyacetic acid (BTOA) (Weaver 1962; Hale
et al.1970; Davies et al.1997; Böttcher et al.2011a,2011b).
IAA in grapes is sequestered into inactive forms by
conjugation with amino acids. The family of IAA-amido
synthetase enzymes that catalyse these reactions has been
studied in detail in grapes (Böttcher et al.2010,2011a). The
CSIRO PUBLISHING
Functional Plant Biology,2012, 39, 745–753
http://dx.doi.org/10.1071/FP12132
Journal compilation CSIRO 2012 www.publish.csiro.au/journals/fpb
observed decrease in IAA during the early stages of berry
development is likely to be due, at least in part, to this
mechanism. IAA-amido synthetases also appear to play a role
in determining the effectiveness of applied auxins. A study
using Shiraz vines showed that the in vivo activity of auxins
was inversely proportional to their ability to act as the substrates
of two IAA-amido synthetases expressed in berries (Böttcher
et al.2011a). IAA was the least effective in delaying ripening
and was the most readily conjugated. BTOA was the most
effective auxin in delaying ripening and was very poorly
conjugated by the two enzymes tested. NAA had intermediate
properties (Böttcher et al.2011a).
Delaying ripening in order to control harvest date and
perhaps berry composition may have considerable benefits to
the wine industry in particular. Delaying the harvest date could
be beneficial in terms of improving winery intake scheduling,
which, in turn, would result in reduced labour and infrastructure
costs, and the ability to harvest fruit at the most desirable stage.
Compressed harvesting seasons over recent years, perhaps a
result of climate change, have greatly increased pressure on
wineries to process fruit in a timely manner (Webb et al.
2011). In addition, delaying or slowing the rate of ripening
may also allow some manipulation of grape composition in the
field and therefore act as useful tool in managing wine style.
Previous work on Shiraz berries (Böttcher et al.2011a,2011b)
has shown that NAA treatment significantly delayed ripening
but wines made from this fruit were not distinguishable by
sensory analysis from wines made from control fruit, and the
wines’volatile profile was quite similar. This paper seeks to
expand this work to an aromatic white variety that has a
different flavour profile to Shiraz and to assess potential of the
measurement of malic acid accumulation as a further indicator
of possible changes to primary metabolism and berry ripening
synchronicity.
Materials and methods
NAA treatment of field-grown Riesling berries
Vitis vinifera L. cv. Riesling vines, on own roots, were grown
on well drained soil near Charleston in the Adelaide Hills
(Nepenthe Wines, 34890S, 138910E). Natural rainfall was
supplemented by limited drip irrigation as required. Bunches
were sprayed three times to run-off during the preveraison period
(30 December 2009, 13 January 2010 and 28 January 2010), with
50 mg L
–1
NAA in 0.1% (v/v) Chemwet 1000 (treatment; Nufarm
Australia Limited, Laverton, Vic., Australia). Control fruit were
sprayed with a 0.1% (v/v) Chemwet 1000 solution. Veraison for
the control fruit was determined to be 3 February 2010. The trial
was of a randomised triplicate design with control and NAA
treatments randomised over three adjacent rows. Each replicate
consisted of eight treated vines (~200 bunches) with an untreated
vine separating the different treatments. Berry development was
followed by frequent sampling of 100 randomly harvested berries
per replicate from each of the treatments from random positions
on the bunches.
Berry analysis
Berry weight was measured for each of the three replicates
(100 berries per replicate). Individual Brix measurements were
taken for all the berries sampled (n= 298–300) using an
RFM710 digital refractometer (Bellingham Stanley, Tunbridge
Wells, Kent, United Kingdom (UK)). Malic acid content was
measured for individual berries (n= 228–270) using a microtitre
plate format enzyme-based assay as follows.
Juice from individual berries was squeezed into a 1.5 mL
Eppendorf tube, spun 1 min at 16 000gand the supernatant was
diluted as appropriate with water (1 : 30 for most developmental
stages) to ensure that the sample was in the linear range of the
assay. An L-malic acid standard curve of 20, 40, 60, 80, 100 nmol
in 5 mL was prepared in water. Reactions were conducted in a
96-well flat-bottomed, straight-sided microtitre plate. Five
microlitres of either standard solution, water (blank) or
diluted grape test solution was pipetted into each well. To
each well, 150 mL of reaction buffer (500 mM glycine, 68 mM
L-glutamic acid, pH 9.8) and 5 mL of 240 mM nicotinamide
adenine dinucleotide solution was added. The reactions were
mixed and optical density (OD) measured at 340 nm in a
FLUOstar Omega microtitre plate reader (BMG Labtechplate,
Ortenberg, Germany). Two microlitres of enzyme solution
(each mL contained 0.78 U of malate dehydrogenase
(M1567–5KU, Sigma-Aldrich, St Louis, MA, USA) and
0.31 U of glutamate-oxaloacetate transaminase (G2751–2KU,
Sigma) were added to each well, followed by mixing. The
plates were incubated for 30 min at 25C with readings at
340 nm taken every 5 min. The data for each well were plotted
and malic acid concentrations were calculated by subtracting
the initial reading from the final reading and comparing this to
the malic acid standard curve.
Small-scale wine making
Small-scale wine making was conducted in triplicate by the
Wine Industry Cluster winemaking services (Adelaide, South
Australia (SA)) using the following protocol. Control and NAA-
treated fruit (20 kg for each replicate) were harvested at 20.4
Brix (9 March 2010) and 19.1Brix (24 March 2010)
respectively. Harvested fruit was placed at 0C for 12 h, the
SO
2
levels were adjusted to 80 parts per million (ppm) during
crushing and destemming using K
2
S
2
O
5
. The grapes were
immediately drained and pressed, and SO
2
levels readjusted to
25 ppm with K
2
S
2
O
5
and pectinase added at a rate of 30 mLL
–1
(Ultrazyme-CPL; Novozymes, Bagsvaerd, Denmark). Yeast
strain EC1118 (Lallemand Australia, Adelaide, SA, Australia)
was added to 250 ppm, 150 ppm diammonium phosphate was
added and tartaric acid added where required to normalise the
must pH to 3.1–3.3. The fermentations were conducted at
1216C, and diammonium phosphate added to a maximum
of 500 ppm as required. When the must was fermented to dryness,
it was racked off gross lees, K
2
S
2
O
5
was added to 60 ppm and
the wine was cold stabilised at 0C for 21 days. The wine
was again racked and SO
2
levels adjusted to 25 ppm with
K
2
S
2
O
5
before filtering, followed by bottling with Stelvin
closures.
Headspace volatile compound analysis
Solid phase microextraction –gas chromatography –mass
spectrometry (SPME-GC-MS) was used to analyse the
volatile constituents of the small-scale wines produced from
746 Functional Plant Biology C. Böttcher et al.
the control and NAA-treated berries. The sample preparation and
analytical methods were identical to that reported by Böttcher
et al.(2010).
Identification of volatile compounds was carried out in three
ways: first, based on mass spectrometric data obtained from the
National Institute of Standards and Technology (Gaithersburg,
Maryland, USA; ver. 2008) or Wiley Registry 9th edition
mass spectral libraries (Wiley, Hoboken, NJ, USA); second by
comparison of mass spectrometric and chromatographic
retention data (as N-alkane standards) reported in the literature;
and third, when possible, comparison of the mass spectrometric
and chromatographic retention data (as N-alkane standards) of
reference compounds.
The components of the samples were quantified using
AnalyzerPro (SpectralWorks, Runcorn, UK) relative to the
internal standard (D
13
-hexanol) using the peak area of an
extracted ion. The effect of applying NAA to the grape
bunches on the concentration of volatiles in the headspace of
the wines was analysed by Student’st-tests and the effect of
wine ethanol concentration by general linear model using SPSS
ver. 16.0 (SPSS Inc., Chicago, IL, USA).
Sensory analysis of wine
Difference testing was conducted to determine if the wine made
from NAA-treated grapes had different sensory characteristics
to that made from control fruit. The replicates from the two
treatments were informally assessed and were deemed to be not
significantly different from each other. A consumer panel of 40
tasters was recruited from the Waite Campus (Adelaide, SA) and
the differences between the wines from NAA-treated and control
fruit were assessed through a duo-trio test. For each sample,
20 mL of wine was served to consumers in blind-coded 215-
mL tasting glasses (ISO 3591:1977; http://www.iso.org/iso/
catalogue_detail.htm?csnumber=9002) at room temperature
(21 2C). Wine glasses were covered with plastic Petri dish
lids to retain the headspace aromas. The testing was performed in
accordance with the Australian Standard for duo-trio testing
(Australian Standard 2542.2.4:2005; http://infostore.saiglobal.
com/store/Details.aspx?ProductID=275922&gclid=CPKQ39a1
wLECFUWHpAodDScAhw). A pretest assessed the wine
colour across the two treatments and concluded that there were
no visual differences between the samples. As a result, the testing
was conducted under normal lighting conditions. The
evaluations were carried out in the sensory booths of the
sensory laboratory at CSIRO (Adelaide, SA). The combination
of samples within the triangle test was randomised over the
subjects to ensure all combinations were presented an equal
number of times. The number of correct responses (when the
different one of the three samples was correctly identified) was
determined and compared against the critical value required for
significant differences.
Results
NAA applied to preveraison Riesling berries significantly
delayed ripening but did not affect berry weight
The berry weight profiles for control and NAA-treated fruit
were similar throughout development (Fig. 1a). There were no
Average berry weight (g)TSS (degrees Brix)Malic acid (mM)
Days after initial spray
(a)
(b)
(c)
Fig. 1. Changes in berry development due to preveraison treatment of grape
Riesling berries with 1-naphthaleneacetic acid (NAA). Control berries (solid
grey line, grey circles) were treated with 0.1% (v/v) Chemwet 1000 (Nufarm
Australia Limited, Laverton, Vic., Australia); NAA-treated berries (dotted
black line, black circles) were treated with 50 mg L
–1
NAA in 0.1% (v/v)
Chemwet 1000. Treatments were done at 0, 22 and 37 days after initial
spraying (DAIS). Samples were taken at 7, 13, 20, 28, 35, 42, 48, 56, 63 and
70 DAIS for both control and NAA-treated berries, and 77 and 85 DAIS for
NAA-treated berries only. Standard error bars are shown. (a) Changes in berry
weight (g). (b) Changes in total soluble solids (TSS), measured as degrees
Brix. (c) Changes in malic acid level (mM). The asterisks denote those times
when there was a statistical difference, as determined by Student’st-test
(*P<0.05, **P<0.01), between treatments. V
C
and V
N
indicate the timing
of veraison in control and NAA-treated berries respectively.
Auxin delayed ripening alters wine composition Functional Plant Biology 747
sampling time points where there was a significant difference
in mean berry weight between the treatments as measured using
a two-tailed unpaired t-test. The mean weight of NAA-treated
berries (1.51 g) was higher compared with the control (1.37g) at
harvest, but this difference was not significant as determined
by the two-tailed unpaired t-test. In contrast, berry Brix levels
were significantly different between control and NAA-treated
fruit at all time points (Fig. 1b). Berry ripening was delayed
by NAA treatment as measured by Brix (Fig. 1b). Statistical
analysis (ANOVA) was used to determine veraison, which was
defined as the last sample time before a significant increase in
Brix level. By this definition, veraison was determined to be at
28 days after initial spraying (DAIS) for control berries and 35
DAIS for NAA-treated berries. Not only was veraison delayed
but the rate of Brix increase –and therefore, by inference, the
rate of sugar accumulation –was lower in NAA-treated fruit.
This difference can be seen from the difference in slope between
the two Brix curves (Fig. 1b). The control fruit was harvested on
9 March 2010; the NAA-treated fruit on 24 March 2010, which
is a 15-day delay despite the NAA-treated fruit not attaining the
same final Brix as the control fruit (Table 1). The plot of Brix
measurements against berry weights (Fig. 2a) shows quite
clearly that for any given Brix level, the weight of the NAA-
treated berries was higher.
Malic acid levels were measured from 21 DAIS until harvest
(Fig. 1c). At 21 DAIS and 42 DAIS there was no significant
difference in malic acid levels between the control and NAA-
treated fruit. However, at all other time points, the levels were
significantly different. Malic acid levels in the berries from the
two treatments were very similar in their respective veraison
samples despite veraison being a week apart. Interestingly, the
malic acid levels for both treatments peaked on the same date
and the peak level was significantly lower in NAA-treated fruit.
After peaking, malic acid levels declined in both samples but
the rate of decline was lower in NAA-treated berries (Fig. 1c).
By plotting Brix measurements against malic acid levels, it
can be seen that the curves for the two treatments are similar
except between ~4.3Brix and 10.5Brix, where the curve
for NAA lies below that for the control (Fig. 2b). At harvest,
the pH of juice from NAA-treated berries was significantly
higher than the juice from the control fruit, but the levels of
titratable acid were not significantly different (Table 1).
NAA treatment increased the synchronicity of berry
development as measured by malic acid and Brix levels
A previous study using fruit from the Shiraz cultivar has shown
that preveraison treatment with NAA can increase the
synchronicity of Brix accumulation during the ripening phase
(Böttcher et al.2011b). Similar results were seen in Riesling,
where the s.d. of the NAA-treated berry population was lower
than that for the control population at six out of nine sampling
times (Fig. 3). As previously observed, the Brix distribution in the
two populations was narrow before veraison, became broader as
the fruit accumulated sugar and became more synchronous again
as harvest approached.
A similar pattern was seen for berry malic acid levels, which
were more narrowly distributed before ripening, broadened
during ripening and again became more tightly distributed
towards harvest time (Fig. 4). The standard deviations of the
Table 1. Basic biochemical analysis of berries (at harvest) and wine
from control and 1-naphthaleneacetic acid (NAA) treated Riesling grape
bunches
Values represent means s.e. (n= 3). Significant differences were determined
by Student’st-test. *, NAA treatment mean was significantly different from
the control at P= 0.05; **, NAA treatment mean was significantly different
from the control at P= 0.01
Total soluble
solids (Brix)
pH Titratable acid
(g L
–1
H
2
T)
Berry
Control 20.4 ± 0.40 3.17 ± 0.01 8.8 ± 0.29
NAA-treated 19.1 ± 0.15* 3.27 ± 0.01** 8.5 ± 0.15
% alcohol (v/v)
Wine
Control 12.7 ± 0.27 3.13 ± 0.03 8.6 ± 0.26
NAA-treated 12.0 ± 0.15 3.26 ± 0.03* 8.4 ± 0.26
Average berry weight (g)
0.4
0.6
0.8
1.0
1.2
1.4
1.6
(a)
(b)
TSS (de
g
rees Brix)
2 4 6 8 10 12 14 16 18 20 22
Malic acid (mM)
20
40
60
80
100
120
140
160
180
200
Fig. 2. Effect of 1-naphthaleneacetic acid (NAA) treatment on Riesling
grape berry weight and malic acid levels in relation to Brix levels. (a) Total
soluble solids (TSS) (degrees Brix) plotted against berry weight (g), (b) TSS
(degrees Brix) plotted against malic acid concentration (mM). Control berries,
solid grey line and grey circles; NAA-treated berries, dotted black line and
black circles.
748 Functional Plant Biology C. Böttcher et al.
malic acid populations were smaller in NAA-treated fruit than
in control fruit at four out of the seven sampling points.
Biochemical analysis of the small-scale wine lots
Table 1shows that although the Brix levels of the berry juice
were significantly different, the percent alcohol of the wines was
statistically similar. The pH was slightly higher in wine made
from NAA-treated fruit (as was the juice pH) but the titratable
acidity was not significantly different. A more detailed analysis
of wine volatile compounds was also undertaken.
Significant differences were observed in the volatile profiles
of the wines as determined using headspace SPME-GC-MS
analysis. Of the 105 compounds identified in the volatile
profiles, 19 were found at higher concentrations in the
headspace of the wines made from NAA-treated grapes and
16 were more abundant in the control wines (Table 2).
Of the 19 compounds higher in the NAA-treatment wines,
six were acetate esters and three were octanoate esters
(Table 2). Three of these acetate esters (hexyl acetate, (Z)-3-
hexenyl acetate and (E)-2-hexenyl acetate) are derived from
alcohols produced by the oxidation of fatty acids in the grape
berries. The monoterpenes linalool and trans-geraniol,
compounds known to be important for the flavour and aroma
of Riesling wine (Smyth et al.2004), were also found in
significantly higher concentrations in the wine made from the
NAA-treated berries than the control wines. Other isoprenoid
compounds that were present at greater levels in the NAA-treated
Number of berries
0
40
80
120
160
0
30
60
90
120
150
180
0
30
60
90
120
150
180
0
40
80
120
160
200
240
TSS (de
g
rees Brix)
0 5 10 15 20 25
0
50
100
150
200
0
20
40
60
80
100
120
0
10
20
30
40
50
60
0
10
20
30
40
50
0
10
20
30
40
50
0 5 10 15 20 25
0
10
20
30
40
50
60
06-01-2010
7 DAIS
C = 0.48
N = 0.42
12-01-2010
13 DAIS
C = 0.46
N = 0.45
19-01-2010
20 DAIS
C = 0.34
N = 0.35
27-01-2010
28 DAIS
C = 0.44
N = 0.29
02-02-2010
35 DAIS
C = 0.65
N = 0.48
09-02-2010
42 DAIS
C = 2.94
N = 0.97
15-02-2010
48 DAIS
C = 2.90
N = 2.49
23-02-201056 DAIS
C = 2.45
N = 3.05
02-03-2010
63 DAIS
C = 1.84
N = 2.67
16-03-2010
77 DAIS
N = 1.41
Fig. 3. Total soluble solids (TSS) (degrees Brix) distribution profiles for
control (solid line), and 1-naphthaleneacetic acid (NAA) treated (dotted
line) Riesling grape fruit. Fruit were treated as described for Fig. 1.The
Brix measurements of 298–300 individual berries were taken for each
treatment except the final sampling time point, where only NAA-treated
fruit were sampled. Brix values were separated into classes with a range
of 0.5Brix and the number of berries per class for each sample time was
plotted against the Brix class. The date of sampling and days after initial
spraying (DAIS) are indicated on the individual graphs. The s.d. for each
of the berry populations is given for each time point. Control, C; NAA-
treated fruit, N.
Number of berries
0
10
20
30
40
50
60
0
10
20
30
40
50
0
10
20
30
40
50
Malic acid (mM)
0 100 200 300
0
10
20
30
40
50
0
10
20
30
40
50
60
0
15
30
45
60
75
0
20
40
60
80
100
0 100 200 300
0
20
40
60
80
100
19-01-2010
20 DAIS
C = 25.1
N = 20.4
27-01-2010
28 DAIS
C = 35.6
N = 28.7
02-02-2010
35 DAIS
C = 38.3
N = 24.5
09-02-2010
42 DAIS
C = 37.3
N = 26.9
15-02-2010
48 DAIS
C = 26.8
N = 27.5
23-02-2010
56 DAIS
C = 23.5
N = 36.9
02-03-2010
63 DAIS
C = 17.4
N = 27.9
16-03-2010
77 DAIS
N = 13.8
Fig. 4. Malic acid concentration distribution profiles for control (solid line)
and 1-naphthaleneacetic acid (NAA) treated (dotted line) Riesling grape
berries. Fruit were treated as described for Fig. 1. The malic acid levels of
228–270 individual berries were measured for each treatment except for the
final sampling time point where only NAA-treated fruit were sampled. Malic
acid levels were separated into classes with a range of 10 mM and the number
of berries per class for each sample time was plotted against the malic acid
class. The date of sampling and days after initial spraying (DAIS) are
indicated on the individual graphs. The s.d. for each of the berry
populations is given for each time point. Control, C; NAA-treated fruit, N.
Auxin delayed ripening alters wine composition Functional Plant Biology 749
wines compared with the control wines include b-damascenone,
neryl propanoate and b-(E)-farnesene (Table 2). Of the
aromatic compounds present in the wines, phenyl ethyl
alcohol and its corresponding acetate ester were more
abundant in the wine produced using NAA-treated fruit, as
was benzyl alcohol. Ten of the 16 compounds that were
found to be higher in the headspace of the control wines were
either ethyl or diethyl esters (Table 2). Interestingly, the ethyl
esters of isobutyric acid, isovaleric acid, (Z)-3-hexenoic acid and
(E)-2-hexenoic acid were more abundant in the control wines than
the NAA-treated wines, whereas esters derived from the
corresponding alcohols were more concentrated in wines made
from the NAA-treated fruit. Other compounds higher in
concentration in the control wines than the NAA-treated wines
included the esters ethyl propanoate and ethyl heptanoate, and
the alcohols 1-octanol and 1-decanol. As the matrix can affect
headspace partitioning of volatiles, with ethanol having the
greatest effect (Robinson et al.2009), the relationship between
the ethanol content of the wines and the concentration of the
volatile compounds was also statistically tested. Only two of
the compounds that were significantly different between the
treatments also showed a significant association between their
abundance and the ethanol content of the replicated wine
samples (Table 2).
Table 2. Volatile compounds found to be significantly different in the headspace of the wines produced from 1-naphthaleneacetic
acid (NAA) treated berries and control berries
Compound RI
A
Compound ID
B
Control
C
NAA
More abundant in headspace of NAA-treated wines
Ethyl acetate
D
<1000 A 1.74 ± 0.07 2.07 ± 0.07*
3-Methylbutyl acetate 1112 A 47.19 ± 4.75 76.03 ± 8.57*
1-Butanol 1141 A 0.12 ± 0.01 0.19 ± 0.01**
Hexyl acetate 1269 A 0.31 ± 0.02 0.59 ± 0.08*
(Z)-3-Hexenyl acetate 1296 A 0.65 ± 0.1 1.13 ± 0.1*
(E)-2-Hexenyl acetate 1340 A 0.015 ± 0.001 0.031 ± 0.003*
Vinyl octanoate 1526 B 0.02 ± 0.01 0.13 ± 0.02*
Linalool 1546 A 0.35 ± 0.03 0.52 ± 0.02**
2-Methylpropyl octanoate 1552 A 0.060 ± 0.005 0.079 ± 0.004*
3-Methylbutyl octanoate 1656 A 0.83 ± 0.05 1.03 ± 0.03*
b-(E)-farnesene 1659 A 0.010 ± 0.001 0.018± 0.001*
Neryl propanoate 1774 B 0.020 ± 0.005 0.037 ± 0.002*
Phenyl ethyl acetate 1808 A 13.61 ± 0.55 19.08 ± 0.32***
b-Damascenone 1811 A 0.93 ± 0.05 1.19 ± 0.04*
Trans-geraniol 1825 A 0.013 ± 0.001 0.031 ± 0.002**
Ethyl dodecanoate 1842 A 1.13 ± 0.21 1.78 ± 0.03*
Hexanoic acid 1863 A 17.97 ± 0.17 18.93 ± 0.26*
Benzyl alcohol 1892 A 0.013 ± 0.001 0.019 ± 0.001*
Phenyl ethyl alcohol 1905 A 13.71 ± 0.56 19.20 ± 0.33***
More abundant in headspace of control wines
Ethyl propanoate 951 A 0.17 ± 0.001*** 0.12 ± 0.001
Ethyl 2-methylpropanoate 955 A 0.14 ± 0.01** 0.093 ± 0.001
Decane 1003 A 0.15 ± 0.003* 0.11 ± 0.007
Ethyl 3-methylbutanoate 1057 A 0.19 ± 0.03* 0.11 ± 0.001
Ethyl (Z)-3-hexenoate 1274 A 0.038 ± 0.005* 0.020 ± 0.001
Ethyl heptanoate 1342 A 0.14 ± 0.02* 0.077 ± 0.004
Ethyl (E)-2-hexenoate 1345 A 0.75 ± 0.1** 0.17 ± 0.04
(E)-3-hexen-1-ol 1359 A 0.25 ± 0.02* 0.18 ± 0.004
Heptyl acetate
E
1372 A 0.037 ± 0.002* 0.027 ± 0.002
Nerol oxide 1466 A 0.71 ± 0.03* 0.54 ± 0.04
Ethyl 2-hydroxyhexanoate 1541 B 0.18 ± 0.02* 0.11 ± 0.006
1-Octanol 1556 A 0.31 ± 0.009** 0.23 ± 0.004
Ethyl 2-furoate
E
1620 A 0.24 ± 0.002** 0.12 ± 0.02
Diethyl succinate 1674 A 7.84 ± 1.20* 3.249 ± 0.10
1-Decanol 1760 A 0.084 ± 0.001** 0.071 ± 0.002
Diethyl glutarate 1777 A 0.053 ± 0.002* 0.032 ± 0.006
A
RI (retention index) is calculated from retention of the compound relative to the retention of a series of N-alkanes (C
8
–C
26
).
B
A: Identity confirmed by matching mass spectra andthe linear retention index (RI) with that of authentic standards; B: tentative assignment
based upon comparison with mass spectral libraries and published RIs.
C
Values represent means s.e. (n= 3). *, ** and *** denote significant differences between treatments at P<0.05, 0.01 and 0.001
respectively.
D
Compounds quantified in the 1 : 100 dilution.
E
Also significantly different (P<0.05) when tested against the alcohol percentage of the wines.
750 Functional Plant Biology C. Böttcher et al.
Sensory testing of small-scale wines
Difference testing by sensory analysis showed that tasters could
distinguish between wines made from NAA-treated (delayed
ripening) and control fruit. Sixty-three percent of the
participants (25 out of 40) correctly identifiedthesamewineas
the control in a duo–trio test, (P= 0.04). Although the wines could
be distinguished by a majority of tasters, their informal tasting
comments did not suggest a particular liking for either wine.
Discussion
This work detailing the delay of ripening of Riesling grapes
by NAA application confirms and extends earlier work (Böttcher
et al.2011a;2011b) using Shiraz vines. In both cases, the timing
of veraison and harvest were delayed and there appeared to be
some increase in the synchronicity of ripening during certain
periods of development.
The similar effects of the treatment in the two cultivars
confirmed that the delaying of ripening by an auxin might be a
general property of grapevines and not dependent on a particular
genetic background.
There were, however, some relatively subtle differences in
the outcomes of the treatments in the two experiments. In the
Shiraz study, harvest was delayed by 10 days and the wines
could not be distinguished by sensory analysis even though
there were some differences detected by metabolite analysis
(Böttcher et al.2011b). Contrastingly, in this study using
Riesling, harvest was delayed by 15 days and even then, the
NAA-treated grapes had a significantly lower Brix value
(Table 1, Fig. 1). The long delay in harvest date for the NAA-
treated fruit indicates that it is possible to alter harvest time to
an extent that would be useful in commercial vineyards, where
staggering the fruit harvest times may be beneficial for winery
intake scheduling. This would allow fruit to be harvested at an
optimal time rather than have it left on the vine and become
over-ripe while waiting for winemaking to proceed. Another
advantage in delaying ripening may be the ability to ripen
berries during cooler times of the year. The delay of 15 days
observed in this study could be further extended through the use
of other viticultural techniques if desired. Wines made from
Riesling grapes grown in cooler climates frequently achieve a
higher price than those made from fruit grown in warmer climates.
Indeed, for several varieties there is a linear relationship
between mean January temperature and wine grape price
(Webb et al.2008).
There were some differences between the two experiments,
including different varieties, seasons and locations, which may
have contributed to the slightly different outcomes described
above. One important difference is the frequency and timing of
NAA application. The Shiraz fruit was treated twice, 19 and
14 days before veraison, whereas the Riesling fruit was treated
three times, 34, 20 and 5 days before veraison. The spray 5 days
before veraison, in particular, may have been responsible for
extending the delay in Riesling ripening over that seen
previously for Shiraz.
There was also a difference in growth between Shiraz and
Riesling berries in response to NAA treatment. In Riesling, the
berry weight profiles were similar for control and NAA-treated
fruit (Fig. 1a), which is in contrast to the work on Shiraz, where the
increase in berry weight was delayed in NAA-treated fruit but
once they started enlarging, these fruit had a significantly higher
mean weight (Böttcher et al.2011a,2011b). Again, this
difference could be due to the range of factors described above.
It is interesting to note that the rate of Brix increase during
ripening was lower for NAA-treated fruit than control fruit
(Fig. 1b). This indicates that there has been a change in the
relationship between elapsed time and the increase in total
soluble solids (TSS). If the characters desired in Riesling
wines are time-dependent rather than related directly to sugar
levels, it may mean that the use of NAA to delay ripening may
produce wine of a style normally associated with high Brix
levels from grapes with lower sugar content (and therefore
wine with a lower alcohol concentration). This may be an
advantage to the wine industry in regard to concerns about the
production of overly alcoholic table wines and indicates that the
rate of sugar increase can be controlled.
The microtitre plate malic acid assay used in this study may
be useful as a high throughput method for assaying berry
development from flowering to harvest. Measuring the
increase in sugar accumulation that occurs after veraison is a
useful measure of berry ripening but cannot be used for
determining berry stage during the preripening phase. The
levels of malic acid change throughout berry development
beginning at a low level, increasing to a maximum level at
veraison and then decreasing during ripening (Fig. 1c). Like
the Brix accumulation profiles, the profiles of malic acid
accumulation were quite distinct for the control and NAA-
treated fruit (Fig. 1c). Interestingly, malic acid levels peaked at
the same time, 1 week after veraison for control fruit and at
veraison for NAA-treated fruit. One possible explanation for
this is that the control of malic acid accumulation has more to do
with flowering or fruit set than it has to do with the timing of
veraison. Why the levels are significantly lower in NAA-treated
fruit at their peak is not understood. The slower rate of malic
acid decline after the maximal level is reached in NAA-treated
fruit is clear-cut and it could be proposed that it is related to
the reduced rate of sugar accumulation in NAA-treated fruit
(Fig. 1b,c). However, the relationship between malic acid and
sugar is not the same in control and NAA fruit. The pattern of
Brix versus malic acid accumulation is virtually identical in
control and NAA-treated fruit except for a period between ~5
and 10Brix (the second (13 DAIS) and fifth (35 DAIS) sample
points for the control fruit and the third (20 DAIS) and sixth
(42 DAIS) sample points for the NAA-treated fruit) (Fig. 2b).
For any given Brix level during this period, NAA-treated berries
had a lower malic acid concentration than the control berries. The
reason for this difference is unknown but it is interesting that
the rate of malic acid decline was similar later in development
(Fig. 1c).
In contrast to the wines from the Shiraz study (Böttcher et al.
2011b), the Riesling wines in this study were distinguishable
by sensory analysis. This may not be a disadvantage, as
informal written comments from the tasters did not indicate
the presence of any negative characters in the wines made
from the NAA-treated berries.
Significant differences in the concentration of volatile
compounds in the headspace above the wines produced
from the NAA-treated and control berries were observed
Auxin delayed ripening alters wine composition Functional Plant Biology 751
(Table 2). These may be responsible for the significant sensory
differences between the wines. Some of the compounds that
differed in concentration between the wines have previously
been shown to be influenced by the amount of grape juice
present in model fermentations and are therefore influenced by
berry composition (Keyzers and Boss 2010). These volatile
compounds include the acetate esters of hexanol, (Z)-3-
hexenol and (E)-2-hexenol, which were more abundant in the
wines made from the NAA-treated berries (Table 2). Grapes
produce C
6
alcohols and aldehydes via the lipoxygenase
degradation of linolenic and linoleic acid, and these
compounds have been shown to be precursors to the acetate
esters formed during fermentation (Dennis et al.2012). The
production of the C
6
compounds changes during the berry
ripening process (Kalua and Boss 2009,2010) so it is possible
that the NAA treatment has altered the rate of C
6
compound
production at harvest through the control of the lipoxygenase
pathway genes or enzyme activity. Alternatively, the
concentration of unsaturated fatty acid precursors could be
altered in the berries by the auxin treatment. However, in
contrast to the three acetate esters mentioned above, the ethyl
esters of (Z)-3-hexenoic acid and (E)-2-hexenoic acid were
higher in the headspace of the control wines than those made
with NAA-treated berries. This suggests that oxidative reactions
were more prevalent in the control berries, whereby there was a
greater tendency for the aldehydes produced by the lipoxygenase
pathway to be metabolised to carboxylic acids rather than
alcohols either in the berries or during fermentation. The same
was true for esters formed from leucine and valine degradation
products. For example, it was observed that the acetate and
octanoate esters of the alcohols 3-methylpropanol and 3-
methylbutanol were more abundant in the NAA-treated wines,
but the corresponding ethyl esters of the carboxylic acids of
these alcohols were more abundant in the control wines. This
suggests that there was a difference in the oxidative processes
occurring in the berry or during fermentation (or both), resulting
in alcohol production in the wines from the auxin-treated fruit
and carboxylic acid production in the control wines. This broad
change in the wines’volatile profiles could impact on the sensory
properties of the wines.
Also of note were the higher concentrations of the
monoterpenes linalool and trans-geraniol in the NAA-treated
wines. These compounds impart important varietal characters
onto Riesling-like and Muscat aromatic wines (Rapp 1998). b-
Damascenone was more abundant in the wines produced from
the NAA-treated berries, and this compound has been shown to
be associated with increased aroma impact or fruit characters
in some wines (Pineau et al.2007; Forde et al.2011).
Furthermore, in a study on Cabernet Sauvignon wines, ethyl
3-methylbutanoate concentration was found to be associated
with green flavour and ethyl (E)-2-hexenoate with pungent
aroma. Both of these compounds were significantly more
abundant in the control wines. In contrast, compounds
associated with fruity or sweet characters, such as hexyl
acetate, b-damascenone and ethyl dodecanoate (Forde et al.
2011), were more abundant in the NAA-treated wines
(Table 2). The wines were significantly different in several
compounds that have previously been shown to be important
for varietal sensory attributes of wine or have been associated
with the particular sensory properties of a wine, which probably
explains the significant result in the difference testing. This
means that the NAA treatment not only delays fruit ripening
but can also produce wines with different sensory characteristics
to the fruit that ripens earlier. In this regard, it should be noted
that the NAA-treated fruit may have been slightly less mature at
harvest that the control fruit as judged by the Brix values
(Table 1). Nevertheless, only two of the compounds found to
be present in significantly different concentrations between the
NAA-treated wines and the controls were found to be correlated
with the alcohol content of the wines (Table 2). The observed
changes in wine composition could be the result of NAA acting
as an auxinic hormone, thereby affecting a considerable range of
metabolic processes in treated berries. For example, it has been
shown that auxin treatment of preveraison berries alters the
metabolism of, or the expression of genes involved in the
metabolism of, flavonoids, general phenylpropanoids, sugars,
organic acids, chlorophyll, ABA and other ripening-associated
processes, including the volatile described in Table 2(Davies
and Böttcher 2009).
A comparison of the compounds that are significantly
different between the Riesling wines reported here and the
Shiraz wines from a similar experiment reported previously
(Böttcher et al.2011b) reveals that the abundance of only four
volatile compounds (ethyl acetate, 2-methylpropyl acetate, hexyl
acetate and (E)-hexen-1-ol) were altered by the NAA treatments
consistently in both experiments. More experimentation is
required to determine if these differences are varietal, or
determined by treatment timing, vineyard location or
management. Nonetheless, it would appear that varietal factors
do play a role in the response, as the abundance of linalool in the
headspace of the Shiraz wines was reduced by the NAA treatment
(Böttcher et al.2011b) but increased by the same growth regulator
in the Riesling wines (Table 2).
The greater synchronisation of grape berry development
resulting from preveraison treatment with NAA appears to be
reproducible, having now been demonstrated in both Riesling
(this work) and Shiraz (Böttcher et al.2011b), but this
phenomenon is not understood as yet. The ability to alter
ripening synchronicity using growth regulators raises the
possibility that severe levels of asynchronous development
could be ameliorated.
This work confirms the potential for NAA as a ripening
delaying agent, and suggests that such treatments might be used
to manipulate grape and wine composition. Exactly how the
ripening delay occurs is currently unknown but it seems that
auxin treatment ‘reinforces’the preveraison state, thus delaying
ripening.
A detailed understanding of the action of NAA on berry
development will require further work but this study shows
that the pattern of berry development can be altered with
consequent changes to metabolism.
Acknowledgements
The authors thank Nepenthe Wines for their generosity in providing
experimental material, Ms Katie Harvey and Ms Inna Mazonka for
technical assistance, and Ms Emily Nicholson for analysis of the wine
volatiles. We gratefully acknowledge financial support from the Grape and
Wine Research Development Corporation and CSIRO.
752 Functional Plant Biology C. Böttcher et al.
References
Amorós A, Zapata P, Pretel MT, Botella MA, Almansa MS, Serrano M (2004)
Role of naphthalene acetic acid and phenothiol treatments on increasing
fruit size and advancing fruit maturity in loquat. Scientia Horticulturae
101, 387–398. doi:10.1016/j.scienta.2003.11.010
Böttcher C, Keyzers RA, Boss PK, Davies C (2010) Sequestration of auxin by
the indole-3-acetic acid-amido synthetase GH3–1 in grape berry (Vitis
vinifera L.) and the proposed role of auxin conjugation during ripening.
Journal of Experimental Botany 61, 3615–3625. doi:10.1093/jxb/erq174
Böttcher C, Boss PK, Davies C (2011a) Acyl substrate preferences of an IAA-
amido synthetase account for variations in grape (Vitis vinifera L.) berry
ripening caused by different auxinic compounds indicating the importance
of auxin conjugation in plant development. Journal of Experimental
Botany 62, 4267–4280. doi:10.1093/jxb/err134
BöttcherC, Harvey K, FordeC, Boss PK, Davies C (2011b) Auxin treatment of
pre-veraison grape (Vitis vinifera L.) berries both delays ripening and
increases the synchronicity of sugar accumulation. Australian Journal of
Grapeand Wine Research17,1–8.doi:10.1111/j.1755-0238.2010.00110.x
Buta JG, Spaulding DW (1994) Changes in indole-3-acetic-acid and abscisic-
acid levels during tomato (Lycopersicon-esculentum Mill.) fruit
development and ripening. Journal of Plant Growth Regulation 13,
163–166. doi:10.1007/BF00196382
Chen K, Li F, Zhang S, Ross GS (1999) Role of abscisic acid and indole-3-
acetic acid in kiwifruit ripening. Acta Horticulturae Sinica 26,81–86.
Chervin C, El-Kereamy A, Roustan JP, Latché A, Lamon J, Bouzayen M
(2004) Ethylene seems required for the berry development and ripening in
grape, a non-climacteric fruit. Plant Science 167, 1301–1305.
doi:10.1016/j.plantsci.2004.06.026
Cohen JD (1996) In vitro tomato fruit cultures demonstrate a role for indole-3-
acetic acid in regulating fruit ripening. Journal of the American Society for
Horticultural Science 121, 520–524.
Coombe BG, Hale CR (1973) Hormone content of ripening grape berries and
effects of growth substance treatments. Plant Physiology 51, 629–634.
doi:10.1104/pp.51.4.629
Davies C, Böttcher C (2009) Hormonal control of grape berry ripening. In
‘Grapevine molecular physiology and biotechnology’. (Ed. KA
Roubelakis-Angelakis) pp. 229–261. (Springer: Dordrecht).
Davies C, Boss PK, Robinson SP (1997) Treatment of grape berries, a
nonclimacteric fruit with a synthetic auxin, retards ripening and alters
the expression of developmentally regulated genes. Plant Physiology 115,
1155–1161.
Dennis EG, Keyzers RA, Kalua CM, Maffei SM, Nicholson EL, Boss PK
(2012) Grape contribution to wine aroma. Production of hexyl acetate,
octyl acetate and benzyl acetate during yeast fermentation is dependent
upon precursors in the must. Journal of Agricultural and Food Chemistry
60, 2638–2646. doi:10.1021/jf2042517
Fabbroni C, Costa F, Bregoli AM, Costa G (2006) Effect of auxin on fruit
morphogenesis: physiological and molecular aspects in kiwifruit ripening.
Acta Horticulturae 753, 541–547.
Forde C, Cox A, Williams E, Boss PK (2011) Associations between the
sensory attributes and volatile profiles of Cabernet Sauvignon wine and
the volatile composition of the grapes used for wine production. Journal of
Agricultural and Food Chemistry 59, 2573–2583. doi:10.1021/jf103584u
Frenkel C, Dyck R (1973) Auxin inhibition of ripening in Bartlett pears. Plant
Physiology 51,6–9. doi:10.1104/pp.51.1.6
Given NK, Venis MA, Grierson D (1988) Hormonal regulation of ripening in
the strawberry, a non-climacteric fruit. Planta 174, 402–406. doi:10.1007/
BF00959527
Hale CR, Coombe BG, Hawker JS (1970) Effects of ethylene and
2-chloroethylphosphonic acid on ripening of grapes. Plant Physiology
45, 620–623. doi:10.1104/pp.45.5.620
Kalua CM, Boss PK (2009) Evolution of volatile compounds during the
development of Cabernet Sauvignon grapes (Vitis vinifera L.). Journal of
Agricultural and Food Chemistry 57, 3818–3830. doi:10.1021/jf803471n
Kalua CM, Boss PK (2010) Comparison of major volatile compounds for
Riesling and Cabernet Sauvignon grapes (Vitis vinifera L.) from fruitset
to harvest. Australian Journal of Grape and Wine Research 16, 337–348.
doi:10.1111/j.1755-0238.2010.00096.x
Keyzers RA, Boss PK (2010) Changes in volatile production in fermentations
made from musts with increasing grape content. Journal of Agricultural
and Food Chemistry 58, 1153–1164. doi:10.1021/jf9023646
Kondo S, Settsu K, Jitratham A (2004) How application times of 2,4-DP
influence the ripening capacity of ‘La France’pears. HortScience 39,
101–104.
Peppi MC, Fidelibus MW (2008) Effects of forchlorfenuron and abscisic acid
on the quality of ‘Flame Seedless’grapes. HortScience 43, 173–176.
Pineau B, Barbe JC, Van Leeuwen C, Dubourdieu D (2007) Which impact for
b-damascenone on red wines aroma? Journal of Agricultural and Food
Chemistry 55, 4103–4108. doi:10.1021/jf070120r
Purgatto E, do Nascimento JRO, Lajolo FM, Cordenunsi BR (2002) The onset
of starch degradation during banana ripening is concomitant to changes in
the content of free and conjugated forms of indole-3-acetic acid. Journal of
Plant Physiology 159, 1105–1111. doi:10.1078/0176-1617-00875
Rapp A (1998) Volatile flavour of wine: correlation between instrumental
analysis and sensory perception. Food –Nahrung 42,351–363. doi:10.
1002/(SICI)1521-3803(199812)42:06<351::AID-FOOD351>3.3.CO;2-U
Robinson AL, Ebeler SE, Heymann H, Boss PK, Solomon PS, Trengove RD
(2009) Interactions between wine volatile compounds and grape and wine
matrix components influence aroma compound headspace partitioning.
Journal of Agricultural and Food Chemistry 57, 10313–10322.
doi:10.1021/jf902586n
Smyth HE, Cozzolino D, Herderich M, Sefton MA, Francis IL (2004)
Identification of key aroma compounds in Australian Riesling and
unwooded Chardonnay wines. In ‘State-of-the-art in flavour chemistry
and biology. Proceedings of the 7th Wartburg Symposium’. (Eds T
Hofman, M Rothe, P Schieberle) pp. 508–513. (Eisenach, Deutsche
Forschungsanstalt für Lebensmittelchemie)
Symons GM, Davies C, Shavrukov Y, Dry IB, Reid JB, Thomas MR (2006)
Grapes on steroids. Brassinosteroids are involved in grape berry ripening.
Plant Physiology 140, 150–158. doi:10.1104/pp.105.070706
Tingwa PO, Young RE (1975) Effect of indole-3-acetic acid and other growth
regulators on ripening of avocado fruits. Plant Physiology 55, 937–940.
doi:10.1104/pp.55.5.937
Vendrell M (1969) Reversion of senescence: effects of 2,4-
dichlorophenoxyacetic acid and indoleacetic acid on respiration
ethylene production and ripening of banana fruit slices. Australian
Journal of Biological Sciences 22, 601–610.
Villarreal NM, Martinez GA, Civello PM (2009) Influence of plant growth
regulators on polygalacturonase expression in strawberry fruit. Plant
Science 176, 749–757. doi:10.1016/j.plantsci.2009.02.019
Weaver RJ (1962) The effect of benzo-thiazole-2-oxyacetic acid on
maturation of seeded varieties of grape. American Journal of Enology
and Viticulture 13, 141–149.
Webb LB, Whetton PH, Barlow EWR (2008) Modelling the relationship
between climate, winegrape price and winegrape quality in Australia.
Climate Research 36,89–98.
Webb LB, Whetton PH, Barlow EWR (2011) Observed trends in winegrape
maturity in Australia. Global Change Biology 17, 2707–2719.
doi:10.1111/j.1365-2486.2011.02434.x
Wheeler S, Loveys B, Ford C, Davies C (2009) The relationship between the
expression of ABA biosynthesis genes, accumulation of ABA and the
promotion of Vitis vinifera L. berry ripening by ABA. Australian Journal
of Grape and Wine Research 15, 195–204. doi:10.1111/j.1755-
0238.2008.00045.x
Zhang XR, Luo GG, Wang RH, Wang J, Himelrick DG (2003) Growth and
developmental responses of seeded and seedless grape berries to shoot
girdling. Journal of the American Society for Horticultural Science 128,
316–323.
Auxin delayed ripening alters wine composition Functional Plant Biology 753
www.publish.csiro.au/journals/fpb
