Postharvest Biology and Technology 136 (2018) 66–73

Contents lists available at ScienceDirect

Postharvest Biology and Technology

journal homepage: www.elsevier.com/locate/postharvbio

Effect of cold storage on stomatal functionality, water relations and flower MARK
performance in cut roses
⁎
Ernst J. Wolteringa,b, , Maxence J.M. Paillarta
a
Wageningen Food & Biobased Research, Bornse Weilanden 9, 6708 WG Wageningen, The Netherlands
b
Wageningen University, Horticulture & Product Physiology, Droevendaalsesteeg 1, 6708 PB Wageningen, The Netherlands

A R T I C L E I N F O A B S T R A C T

Keywords: Symptoms of water stress are the most frequent cause for the “end of vase life” in prior stored roses. It was
Cold storage hypothesized that dark storage may alter the stomatal functionality and may cause water balance problems
Cut flowers during the subsequent vase life period. The effect of short- and long-term storage on functionality of stomatal
Rose and subsequent flower performance was investigated in two rose cultivars (cvs) (‘Akito’ and ‘Grand Prix’) with
Stomatal functionality
presumed different sensitivity for development of water stress symptoms during the vase life.
Water relations
Compared to no storage, both short term storage (2.3 d at 6 °C) and long term storage (28 d at 0.5 °C) ne-
gatively affected the stomatal functionality in cultivar (cv) Akito. Stomatal functionality parameters such as the
rapidity of the closing response upon dehydration and the relative water content at which stomata are fully
closed showed good correlations with flower performance parameters (flower weight changes and vase life). This
indicates that in cv Akito, the decreased stomatal functionality is one of the factors involved in the poor vase life
of prior stored flowers. In cv Grand Prix, however, storage did not greatly affect the stomatal functionality but
storage negatively affected flower performance in a comparable way as in cv Akito.
A pre-treatment with abscisic acid prior to storage slightly improved stomatal functionality in both cvs, but no
clear effect on flower performance was observed. Addition of the bactericide 8-HQC to the vase water improved
flower performance in both cvs but could not alleviate the negative effect of cold storage on flower performance.
Results show that in roses cold storage may, depending on the cultivar, negatively affect stomatal func-
tionality and this may contribute to water stress and ultimately flower failure. In addition, cold storage may
negatively affect xylem water conducting properties through processes not related to bacterial contamination.

1. Introduction
The flower auction “FloraHolland” in the Netherlands, does regular
vase life tests on samples of the flowers that are supplied by the asso-
ciated growers and from other parties. These flowers may be from
batches of flowers that were grown in the Netherlands or from imported
flowers. It was found that symptoms of water stress are the most fre-
quent cause for the “end of vase life” in the majority of the tested rose
cultivars. Second important cause of the “end of vase life” is botrytis
and third cause is physiological senescence (Fanourakis et al., 2015).
Most prominent symptoms of water stress are loss of turgor of the petals
(flower wilting), bent neck and wilted leaves. Such symptoms are
generally associated with a negative water balance, when water de-
mand (transpiration) exceeds water uptake (van Doorn, 2012).
A negative water balance can be caused by either impaired water
uptake, excessive transpiration or by a combination of both. Impaired
uptake may be caused by an increased resistance of the xylem vessels to
water flow due to e.g. bacterial growth, air embolism or physiological
processes. Physiological blockage is related to the wound-induced
production of polyphenolic compounds and has been described in e.g.
Chrysanthemum and Syringa, but not in roses (van Doorn, 1997). Ex-
cessive transpiration may be caused by impaired functionality of the
stomata. If stomata do not adequately respond to water stress signals or
when stomata do not fully close e.g. when water uptake is limited, this
quickly leads to development of a negative water balance (van Doorn,
2012).
When no adequate biocides are added to the vase solution, the ac-
cumulation of bacteria in the vase solution and in the vascular system
will progressively limit the water uptake. In response to the developing
water stress, stomata will close in order to maintain a favourable water
balance. Even when the vase solution contains effective biocides and
bacterial numbers are maintained at relatively low levels, the xylem
vessels will develop an increasing resistance to water uptake due to
cavitation of vessel elements and possible physiological reactions

⁎
Corresponding author at: Wageningen Food & Biobased Research, Bornse Weilanden 9, 6708 WG Wageningen, The Netherlands.
E-mail address: ernst.woltering@wur.nl (E.J. Woltering).

http://dx.doi.org/10.1016/j.postharvbio.2017.10.009
Received 2 August 2017; Received in revised form 20 October 2017; Accepted 22 October 2017
0925-5214/ © 2017 Published by Elsevier B.V.
E.J. Woltering, M.J.M. Paillart
caused by wounding of the stem (Bleeksma and Van Doorn, 2003). As
most water loss in rose flowers is through the leaves and minor part
through the flower petals, an adequate response of the stomata to water
stress is of vital importance for maintaining a favourable water balance
especially when resistance of the xylem vessels to water flow develops.
When stomata are fully functional, up to two thirds of the vessels may
be blocked before the water balance becomes unfavourable and flowers
show symptoms of water stress (van Doorn, 1997).
Pre harvest conditions have been shown to affect the functionality
of stomata. Conditions of high humidity (low evaporative demand)
during cultivation tend to render stomata less responsive to water stress
and e.g. pre-harvest treatments with abscisic acid (ABA) may reverse
this effect (Fanourakis et al., 2012, 2013a,b; Nejad and van Meeteren,
2007). Similarly, continuous lighting to increase productivity, nega-
tively affects stomatal functionality. It has been shown that roses grown
at relatively high humidity and from continuous lighting have a con-
siderably reduced vase life (Fanourakis et al., 2013a).
Cold storage and handling of roses after harvest often negatively
affects the vase life and this effect is more pronounced after long sto-
rage times (van Doorn, 1997). Currently there is a trend toward ship-
ping flowers in refrigerated 40-feet containers (Reefer) over sea instead
of transport by airplane. According to a recent study of Rabobank (van
Rijswick, 2015), the most prominent cut flower Reefer transport routes
are from Colombia to UK (approximately 700 containers per year); from
Israel to EU (300 containers per year) and from Vietnam to Japan (400
containers per year). Although the journey by ship takes much longer
than by airplane (2–4 weeks by ship versus 2–4 d by airplane), much
better control of the storage conditions is possible in the Reefer. Due to
the higher energy efficiency of the Reefer on a ship compared to an
airplane, both the transport costs and CO2 footprint of Reefer trans-
ported product are considerably lower.
During the extended transport period in Reefer containers, the
product is held at a very low temperature (0.5–2 °C), densely packed in
boxes, sleeved or within plastic liners creating a high relative humidity
(> 95%). As symptoms of water stress are the most frequent cause for
the “end of vase life” of these roses, we hypothesized these storage
conditions may alter the stomatal functionally and in this way may
cause water balance problems during the subsequent vase life. Here we
report on the effect of short- and long term storage on functionality of
stomata in two rose cultivars (‘Akito’ and ‘Grand Prix’) with different
sensitivity for development of water stress symptoms during the vase
life. In addition, the effect of a pre-storage treatment with abscisic acid
(ABA) was investigated.
We show that in cv Akito the storage negatively affects stomatal
functionality (measured at start of the vase life) and that loss of sto-
matal functionality correlates with a lesser performance of the stored
roses. In cv Grand Prix, however, storage did not affect stomatal func-
tionality but stored roses did show a lesser performance than non-stored
roses. This indicates that apart from stomatal functionality additional
factors play a role in the lesser performance of stored roses.
2. Materials and methods
2.1. Plant material and experimental design
Cut roses of cvs Akito and Grand Prix were obtained from com-
mercial growers in the Netherlands. Flowers were transported dry at
4 °C to the laboratory within 3 h after harvest. Roses were graded for
homogeneity, recut (at least 5 cm) and placed overnight in either water
or water + 0.1 mM abscisic acid (Sigma-Aldrich, St Louis, MI, USA) at
4 °C and 90% relative humidity (RH). The last 6 h of treatment was in
the light (8–10 μmol m−2 s−1). Following the pre-treatment, stem
length was shortened to 45 cm and roses were placed in a climate room
(20 °C, 60% RH, 12 h light (13 μmol m−2 s−1) and 12 h dark) to study
the flower life and to perform measurements of physiological para-
meters. Remaining roses were stored dry, either at 6 °C and 90% RH for
67
Postharvest Biology and Technology 136 (2018) 66–73
2.3 d or at 0.5 °C and 90% RH for 28 d, simulating transport conditions
by plane and by ship, respectively. Storage times were chosen in such a
way that for both temperatures the total number of degrees-days was
14. Bunches of 10 flower stems were packed in poly-propylene sleeve
and placed in the storage rooms. Flower weight loss during the storage
period was monitored by weighing unpacked bunched before and just
after storage.
Following storage, flower stems were re-cut (3–5 cm) and rehy-
drated overnight in tap water at 4 °C and 90% RH. The last 6 h of
treatment were carried out under light (8–10 μmol m−2 s−1).
Thereafter, the flowers were recut to 45 cm and placed in the climate
room to study the flower life and to perform measurements of physio-
logical parameters.
2.2. Flower weight, water uptake and transpiration
Following the pre-treatment (in non-stored flowers) or following the
rehydration treatment (in stored flowers), stems were recut to 45 cm
and leaves from the lower end of the stem were removed (leaving about
4 fully grown leaves on the stem). Flowers were placed individually in
flasks containing water or water + 150 mg L−1 8-hydroxy quinoline
citrate (8-HQC; Sigma-Aldrich, St Louis, MI, USA) (5 flowers per
treatment). Flasks were randomly placed in the climate room. Flower
weight and water uptake of flowers was monitored by daily measure-
ment of the weight of the flowers and that of the flasks. This allowed to
calculate the effects of the treatments on flower performance (flower
weight change curves) and flower water balance parameters such as
transpiration and uptake. Flower transpiration was calculated as the
water uptake minus the flower weight change over a certain period of
time (typically 1 d).
At the end of the experiment total leaf area of each flower was
measured using LI-3100C Area meter (Li-Cor, Lincoln, NE, USA). In
order to be able to compare the transpiration rate and the cumulative
transpiration between the storage treatment, the results were corrected
to the specific leave area and expressed per 100 cm2 leave area. The
flower performance was established on basis of the flower weight gain
and loss. The daily flower weight was expressed in a percentage of the
initial flower weight (100% on day 0). Flower performance parameters
are derived from the flower weight curves i.e. the area under the curve
and the number of days where the flower weight is above its initial
weight (above 100% Calculated vase life).
2.3. Stomatal characteristics
Measurements of leaf stomatal characteristics were performed im-
mediately following the pre-treatment (in non-stored flowers) or im-
mediately following the rehydration treatment (in stored flowers). For
determination of stomatal characteristics, the first two fully grown
leaves under the flower head were removed from the stem (5 samples
per treatment were measured; each sample consisted of 4 leaves excised
from 2 stems). Leaves were weighed (measurement of fully hydrated
state) and placed abaxial side up on the table in the climate room (20 °C
and 60% relative humidity in light 13 μmol m−2 s−1). Thereafter, leaf
samples were weighed every 15 min till they had lost > 50% of their
initial weight. At the end of the experiment, leaf dry weight was de-
termined (drying at 70 °C for at least 24 h). This allowed to express the
water loss as a function of the relative water content (RWC).
RWC was calculated according to Formula (1) (Fanourakis et al.,
2013b).
Fresh weight − Dry weight
RWC =
Sturated Fresh Weight − Dry weight
× 100
(1)
Different stomatal characteristics were determined from curves de-
picting the water loss as a function of the RWC (described in Section
3.4).
E.J. Woltering, M.J.M. Paillart Postharvest Biology and Technology 136 (2018) 66–73

2.4. Statistical analyses

Results were compared applying a general analysis of variance

(ANOVA) using Genstat 18th ed. program (VSN international Ltd.,
Hemel Hempstead, UK). Means were compared by the least significant
difference (Fisher’s protected LSD) test at P < 0.05.

3. Results

3.1. Water loss during storage

Water loss during dry storage was approximately 1% for roses

stored for a short period at 6 °C and about 4–6% for roses stored for
long period at 0.5 °C. No effect of the ABA pre-treatment on water loss
during storage was observed (data not shown).

3.2. Flower weight changes and calculated vase life

Weight of individual flowers in water and in water + 8-HQC was

monitored during the vase life and expressed as a percentage of the rose
weight immediately following its rehydration (=100%) at low tem-
perature. The standard pattern showed that flower weight increased
during the first days of the vase life and decreased thereafter (Fig. 1).
The area under the curve (from time zero to the time the weight
drops below 100%) as a measure of flower performance was integrated
for each individual flower and averaged for each treatment. There was
no significant effect of the ABA pre-treatment on the measured areas
(data not shown) and therefore the data of the water and ABA pre-
treatments were combined (Fig. 2). Similar trends were observed for the
two cvs (Fig. 2A). When both cvs were pooled together (no statistical
difference between cvs), flowers on water + 8-HQC showed a sig-
nificant favourable water balance (higher area) than for flowers kept on
water, irrespective the storage duration. This indicates that bacterial
contamination may play a role in maintaining a favourable water bal-
ance during the vase life in both these cvs. Without exception, non-
stored flowers better maintained a favourable water balance than
stored flowers and this effect was independent on the vase solution
(water or water + 8-HQC).
From the curves depicting the change in flower weight, an ap-
proximate vase life can be calculated i.e. by taking the time till the
flower weight is back at the initial level again (Mayak et al., 1974).
Similar trends were observed in the two cvs (Fig. 2B). The statistical
Fig. 2. Flower performance parameters of non-stored, short- and long-term stored flowers
of cvs Akito and Grand Prix during the vase life in water and in water + 8-HQC. A. Area
under the flower weight curve (arbitrary units). B. Calculated vase life. Data from both
pre-treatments (ABA and Water) were taken together. Values are means of 10
flowers ± SEM.
analysis was performed on data from both cvs pooled together. In both
cvs, calculated vase life was longer when flowers were in water + 8-
HQC. In addition, calculated vase life was longest in non-stored flowers
and shortest in long-term stored flowers. As expected there was a cor-
relation between the area under the flower weight curves and the cal-
culated vase life (Supplement Fig. S1).
3.3. Water uptake and transpiration during the vase life
Water uptake, corrected for changes in flower weight, is a measure
of transpiration. As an example, daily transpiration expressed in kg per
m2 of leaf area of cv Akito is shown in Fig. 3A and cumulative tran-
spiration in Fig. 3B. Transpiration rate showed a gradual decrease over
time. This is supposed to be related closing of the stomata possibly as a
result of a developing resistance of the xylem system to water flow. As
the ABA pre-treatment did not significantly affect the transpiration
(data not shown), the data of the water and ABA pre-treatments were
taken together for overall comparison of the storage treatments. Cu-
mulative transpiration shows a logarithmic curve (Fig. 3B).
To better enable comparison of the cvs and treatments after one
week of vase life, the accumulated transpiration till day 7 was sum-
marized in Fig. 4. Transpiration behaviour was different between the
two cvs. Cumulative transpiration was higher in cv Akito than in cv
Grand Prix. No significant difference was found between water and
water + 8-HQC vase solutions. In both cvs the cumulative transpiration
was lower in stored flowers than in the non-stored flowers.

3.4. Stomatal functionality

Fig. 1. Weight changes (in % of initial weight) of non-stored (◊), short- (Δ) and long-term
(○) stored flowers of cv Akito during the vase life in water (open symbols) and in water
+ 8-HQC (closed symbols). Weight at time 0 (immediately after overnight hydration) was
set at 100% (horizontal line). Data from both pre-treatments (ABA and Water) were taken
together. Values are means of 10 flowers.
Stomatal functionality was measured shortly after the pre-treatment
(for non-stored flowers) or after the rehydration treatment (for stored
flowers). After 16 h of rehydration, it was assumed that leaves were
fully hydrated at the start of the measurement. Furthermore 6 h before
starting the stomatal functionality measurement, flower were exposed

68
E.J. Woltering, M.J.M. Paillart Postharvest Biology and Technology 136 (2018) 66–73

Fig. 3. Transpiration during the vase life in water (open symbols) and in water + 8-HQC (closed symbols) of non-stored (◊), short- (Δ) and long-term (○) stored flowers of cv Akito. A.
Daily transpiration rates. B. Cumulative transpiration. Data from both pre-treatments (ABA and Water) were taken together. Values are means of 10 flowers.

to light in order to make sure that stomata were fully open (Tallman
2004). Leaves were excised from the flower stem (4 leaves from 2
flowers were taken together) and weight loss was measured over time.
An example of water loss curves of leaves from long-term stored flowers
is shown in Fig. 5A (cv Akito) and B (cv Grand Prix). Based on the water
loss, transpiration rate over each interval was calculated. In this ex-
ample it can be seen that excised leaves of cv Grand Prix showed a
much more pronounced decrease in transpiration in response to the
developing water stress than the leaves of cv Akito, resulting in less
desiccation of the cv Grand Prix leaves.
To make the assay independent of the measurement conditions,
transpiration was expressed as a function of the RWC. The four
following stomatal characteristics were determined from curves de-
picting the water loss as a function of the RWC: a) RWC (in %) at which
stomata are closed (RWCclosed); RWCclosed represents the nick in the
Transpiration against RWC curve, calculated as the intersection be-
tween the fitted lines of the transpiration data before and after
RWCclosed. b) transpiration in dependence of the leaf RWC when sto-
mata are open (TRopen); TRopen is a value representing the slope of the
transpiration curve before RWCclosed is reached. c) transpiration in
dependence of the leaf RWC when stomata are closed (TRclosed);
TRclosed is a value representing the slope of the transpiration curve
after RWCclosed is reached. d) the ratio between TRclosed and TRopen.
In the graphs for cvs Akito and Grand Prix (Fig. 5C and D), two

Fig. 4. Cumulative transpiration (till day 7) in water and in water + 8-HQC of non-stored, short- and long-term stored flowers of cvs Akito and Grand Prix. Data from both pre-treatments
(ABA and Water) were taken together. Values are means of 10 flowers ± SEM.

69
E.J. Woltering, M.J.M. Paillart
Fig. 5. Weight loss and transpiration rates of excised leaves from flowers stored for 4 weeks at 0.5 °C. Weight loss (○) and transpiration (■) of excised leaves of cv Akito (A) and of cv
Grand Prix (B). Transpiration rate in dependence of relative water content cv Akito (C) and cv Grand Prix (D). Data are from representative samples containing 4 leaves derived from 2
flowers.
phases can be distinguished: the first phase reflecting the declining
transpiration when the stomata are still open, the second phase mostly
reflecting the transpiration when the stomata are closed. The tran-
spiration during the latter phase depends on the efficiency of the sto-
mata closure and of cuticular transpiration. The parameters TRopen
and TRclosed represent the calculated slopes of the linear regression
equations. Both measures characterize the stomatal functionality. The
interception between the regression lines indicates the RWC (in %) at
which the stomata are closed (RWCclosed) and can be considered an-
other stomatal functionality parameter. In this example of long-term
stored flowers, it can be seen that in cv Akito (Fig. 5C) the stomata are
closed at a RWC around 70%; in cv Grand Prix (Fig. 5D) around 85%. In
response to desiccation, stomata show a more pronounced closing re-
sponse (higher TRopen) in cv Grand Prix compared to cv Akito and
stomata are more tightly closed (lower TRclosed) in cv Grand Prix than
in cv Akito.
The calculated parameters (RWCclosed, TRopen, TRclosed and the
ratio TRclosed/TRopen) derived from these curves are shown for all
treatments in Fig. 6. Stomata in cv Grand Prix generally closed at higher
RWC than stomata in Akito (statistically significant in most of the
treatment combinations). In addition, a significantly higher TRopen
and a lower TRclosed (and therefore a lower TRclosed/TRopen ratio)
were observed in Grand Prix than in cv Akito. All measured stomatal
functionality parameters show that leaves of cv Grand Prix exhibit a
more adequate response to water stress than leaves of cv Akito. Stomata
in cv Grand Prix leaves are fully closed at lower levels of water stress.
Their closure occurs faster and stomata are more tightly closed com-
pared to stomata in cv Akito leaves. In both cvs, RWCclosed was higher
70
Postharvest Biology and Technology 136 (2018) 66–73
in ABA pre-treated flowers than in water pre-treated flowers (statisti-
cally significant in most of the treatment combinations). For the other
stomatal functionality parameters, differences between ABA and water
pre-treatments were not statistically significant. Storage negatively af-
fected RWCclosed, TRopen and the ratio TRclosed/TRopen in cv Akito.
In cv Grand Prix, only TRopen in water pre-treated flowers was nega-
tively affected by storage, the other stomatal functionality parameters
were not affected by storage. This shows that storage affects stomatal
functionality in cv Akito, but not, or much less in cv Grand Prix.
In cv Grand Prix the different treatments (water or ABA pre-treat-
ments) and storage scenarios had only minor effects on stomatal char-
acteristics and no clear relations between these parameters were found
(data not shown). In cv Akito a negative correlation was found between
RWCclosed and TRclosed and a positive relation was found between
RWCclosed and TRopen (data not shown). The ratio TRclosed/TRopen
as a measure of stomatal functionality showed a negative correlation
with RWCclosed (Fig. 7); these relations were expected. Fully func-
tional stomata will show a pronounced closing response already at re-
latively mild water stress and will have tightly closed stomata at higher
water stress levels.
3.5. Correlations between stomatal characteristics and flower performance
As no variation in the stomatal functionality parameters of cv Grand
Prix leaves with respect to the treatments was observed, the correla-
tions between stomata and water balance parameters were only ana-
lysed in cv Akito. In cv Akito there were consistent correlations between
the stomatall functionality parameters (RWCclosed, ratio TRclosed/
E.J. Woltering, M.J.M. Paillart Postharvest Biology and Technology 136 (2018) 66–73

Fig. 6. Overview of calculated stomatal characteristics of non-stored, short- and long-term stored flowers of cvs Akito and Grand Prix that were pre-treated with either water or ABA. A.
RWC at which stomata are closing (RWCclosed). B. Slope of the decrease in transpiration in dependence of the leaf RWC when stomata are open (TRopen). C. Slope. of the decrease in
transpiration in dependence of the leaf RWC when stomata are closed (TRclosed). D. Ratio between TRclosed and TRopen. Values are means of 5 flowers ± SEM.

Fig. 7. Correlation between RWCclosed and the ratio TRclosed/TRopen for non-stored,
short- and long-term stored roses of cvs Akito (◊) and Grand Prix (□) that were pre-
treated with either water or ABA. Values are means of 5 flowers ± SEM.
TRopen) and the flower water balance and performance parameters
(area under the flower weight curve and calculated vase life). Both in
flowers placed in water and in water + 8-HQC, the area under the
flower weight curve was positively correlated with the relative water
content when stomata closed (RWCclosed) (Supplemental Fig. S2). A
negative correlation was found between the area under the flower
weight curve and the ratio TRclosed/TRopen (Supplemental Fig. S3). In
flowers placed in water, a positive correlation was found between
RWCclosed and calculated vase life (Supplemental Fig. S4); a negative
correlation was found between the calculated vase life and the ratio
TRclosed/TRopen (Supplemental Fig. S5). These correlations confirm
that the flower performance in cv Akito correlates with the adequacy of
the stomatal response to desiccation.
There were no consistent correlations between the stomatal func-
tionality parameters and the cumulative transpiration (R2 < 0.1) and
no correlation was observed between the cumulative transpiration and
the area under the flower weight curve (R2 < 0.1) nor between the
cumulative transpiration and the calculated vase life (R2 < 0.2) (data
not shown). This indicates that during the vase life period there are no
direct relations between the amount of transpired water and the sto-
matal functionality (as measured at day 1). In addition, no correlation
exists between the amount of transpired water and the flower perfor-
mance (weight gain, vase life).

71
E.J. Woltering, M.J.M. Paillart
4. Discussion
The consequence of shipping flowers in refrigerated containers
(Reefers) is a long storage period (up to 5 weeks). Often the flower
performance after shipping is disappointing although there may be
differences between cultivars. Rose cv Akito is generally experienced as
having a short vase life after storage showing rapid wilting and bent
neck; cv Grand Prix often shows a satisfactory performance even after
long-term storage. These two cvs with supposedly contrasting beha-
viour with respect to (long-term) storage were chosen for this in-
vestigation. The occurrence of rapid wilting and bent neck following
storage is due to a negative water balance, indicating that stored
flowers either more rapidly develop blockage of the vascular system or
are unable to properly restrict transpiration (van Doorn, 1997).
We hypothesized that long-term storage under cold and humid
conditions may affect the functionality of the stomata. This could cause
rapid water balance problems once the flowers are placed in the vase at
much lower relative humidity (e.g. 40–60% RH). The effect of long-
term storage on flower performance, flower water balance and stomatal
functionality was studied in rose cvs Akito and Grand Prix. For com-
parison non-stored and short-term stored (with the same time-tem-
perature sum) were included.
Functionality of stomata is very sensitive to RH. Cultivation of roses
at high RH has severe effects on stomatal functionality and this has a
negative effect on the postharvest performance (Fanourakis et al., 2012,
2013a,b; Mortensen and Fjeld, 1998). Even short periods (a few days)
of high humidity in e.g. Tradescantia, faba bean or arabidopsis may
already attenuate the stomatal closure response to desiccation or ABA
(Aliniaeifard and Van Meeteren, 2013; Aliniaeifard and van Meeteren,
2014; Aliniaeifard et al., 2014). The decreased ABA levels in leaves
under high humidity are thought to be the primary cause of the loss of
stomatal response (Fanourakis et al., 2016a). The stomatal mal-
functioning following cultivation of plants at high humidity could be
partly overcome by spraying of the plants with ABA prior to harvest
(Nejad and van Meeteren, 2007). In different rose cvs cultivated at high
RH, the leaf ABA level was positively correlated with the stomatal re-
sponse to desiccation (Giday et al., 2013). By applying water stress to
roses cultivated at high RH, both the level of ABA and the stomatal
functionality was restored (Giday et al., 2014). Apparently under high
humidity either the production of ABA, its translocation to the stomata
closing cells or its breakdown are affected. In an attempt to improve
stomatal functioning, we therefore treated part of the flowers with ABA
prior to storage.
The vase life of the flowers was studied in tap water and in water to
which the biocide 8-HQC was added to create two levels of bacterial
load. Bacterial growth in the vase solution and in the lower end of the
stem is generally considerably less when roses are placed in water with
added 8-HQC and this leads to a slower development of the stem re-
sistance to water flow (van Doorn and Perik, 1990).
As a measure of flower performance the area under the flower
weight curves was calculated. Generally flower weight increases during
the first days in the vase due to the opening of the flower. When the
water balance becomes negative (more transpiration than uptake), the
weight decreases (Fig. 1) and this will be reflected in visual petal
wilting and eventually bent neck. From these curves the vase life was
calculated, being the number of days till the flower weight was back at
the initial (fully hydrated) level (100%).
For both cvs, the area under the flower weight curve and the cal-
culated vase life were highest in non-stored flowers and lowest in long-
term stored flowers; short-term stored flowers being intermediate
(Fig. 2). Especially in cv Akito, flower performance was much better in
water + 8-HQC than in water; in cv Grand Prix this difference was less
pronounced. The effect of 8-HQC was observed in both the non-stored
and the stored flowers. This indicates that, compared to cv Akito, cv
Grand Prix is better able to maintain a positive water balance even
under conditions that are favourable for bacterial growth in the vase
72
Postharvest Biology and Technology 136 (2018) 66–73
water. The relative contribution of 8-HQC to flower performance is not
changed by the storage, indicating that bacterial development in the
vase may not be dependent on prior storage. As we did not perform any
measurements of bacterial numbers in the vase water nor in the stems,
these statements need verification.
Measurement of the stomatal functionality parameters showed that
cv Grand Prix stomata were more responsive to water stress than cv
Akito stomata. In cv Grand Prix, stomata showed a more pronounced
closing response to the developing water stress (high TRopen), were
already completely closed at mild water stress (high RWCclosed) and
when closed they were tightly closed (low TRclosed) (Fig. 6). The sto-
matal functionality parameters showed a good correlation with each
other: when stomata close at relative high RWC, they also show a more
pronounced closing response, and once closed they are tightly closed.
Both the relative water content at which stomata are closed
(RWCclosed) and the ratio between the TRclosed and TRopen were
therefore considered suitable parameters to characterise the stomatal
functionality (Fig. 7).
Stomatal functionality was not, or very little, affected by storage
period in cv Grand Prix. In cv Akito, however, stomatal functionality
was negatively affected by both short- and long-term storage (Fig. 6).
The impaired stomatal functionality after storage showed a consistent
correlation with the lesser performance of the stored flowers (Supple-
mental Figs. S2–S5). The unfavourable water balance in cv Akito
flowers after storage may therefore partly be due to the impaired sto-
matal functionality. This implicates that improvement of stomatal
functionality through treatments before or after the storage period may
improve the performance of the stored flowers. The water balance is
determined both by the transpiration and by the uptake. The relative
contribution of both processes to the developing water stress in these
flowers, however, is not known and may depend on e.g. the relative
humidity.
In cv Grand Prix storage had a similar negative effect on water
balance (area under the flower weight curve) and the calculated vase
life as in cv Akito (Fig. 2). However, stomatal functionality in cv Grand
Prix was generally better than in cv Akito and not affected by the sto-
rage. The lesser flower performance in cv Grand Prix following storage
therefore cannot be ascribed to differences in stomatal functionality.
This indicates that in cv Grand Prix, an increased resistance to water
uptake following storage may be the cause of the lesser performance of
prior stored roses. Similarly, this may in addition to impairment of
stomatal functionality, also play a role in cv Akito. The general ob-
servations in commercial practice that cv Grand Prix is more tolerant to
long storage than cv Akito were not confirmed in our experiments. This
may be related to the RH in the flower observation room (60%), which
is higher than in most practical situations.
Increased resistance to water uptake can be caused by several pro-
cesses leading to blockage of the vascular system. Vascular occlusions
may be caused by either the growth of bacteria in the xylem vessels, by
formation of air embolisms in vessel elements or by physiological re-
sponses to wounding (van Doorn, 2012). An increased susceptibility of
stored flowers to vascular blockage could be due to the growth of
bacteria in the vascular system during the dry storage. Before stored
flowers were placed in the vase, their stems were two times recut: first
approximately 4 cm immediately after storage when flowers were re-
hydrated in water and thereafter again 4 cm when flowers were put in
the vase. As bacteria mostly accumulate in the lower part of the stem
(van Doorn and Perik, 1990) one would expect that excess bacteria may
have been removed by the re-cutting. If increased amounts of bacteria
in the stem after storage play an important role in the lesser perfor-
mance of stored flowers, the negative effects of storage on the water
balance should be less if flowers are placed in a bactericide. Judging
from the flower weight curves this was not the case (Figs. 1 and 2). Both
in water and in 8-HQC the area under the flower weight curves was
considerably, and to a similar extent, reduced in prior stored flowers
compared to non-stored flowers. These observations seem to indicate
E.J. Woltering, M.J.M. Paillart
that factors other than bacterial growth may determine any increased
susceptibility of stored flowers to vascular occlusion.
In our study we did not find any correlations between cumulative
transpiration and stomatal functionality as well as with flower perfor-
mance. A higher cumulative transpiration was expected in flower stems
with less functional stomata, but this was not the case. To explain this
discrepancy, more detailed measurements of water uptake and tran-
spiration and stomatal responses are necessary. It may be that stomatal
functionality changes during the vase life period but this was not
measured. We determined the stomatal functionality in response to
severe water stress in excised leaves. We have no information about the
closure of the stomata in response to darkness during the daily dark
period. Also the transpiration will be dependent on the availability of
water, that may be limited due to vascular occlusions.
Both storage periods, with same time-temperature sum, negatively
affected the performance of the flowers. However, the effect of long-
term storage on flower performance (area under the flower weight
curves, calculated vase life) was more severe than in short-term storage.
This shows that the long-term cold storage has more impact on the
flower quality than short-term at higher temperature. This is in line
with findings of Tromp et al. (2012) showing that the time-temperature
sum model is only viable within a certain temperature range and that
deviations occur below 2 °C. Storage at 0.5 °C may cause slight chilling
injury that increases the loss of quality on top of the combined effect of
time and temperature.
Stomatal functionality as affected by high RH during growth is re-
lated to ABA levels in leaves, as discussed above. Although slight, but
statistically significant effects on stomatal functionality parameters
were observed, the pre-treatment of the flowers with ABA generally had
no significant effect on flower performance parameters. Application of a
higher concentration of ABA or of other anti-transpirants (Fanourakis
et al., 2016b) or application of anti-transpirants after storage may have
a more pronounced effect on stomatal functionality and hence may
improve flower performance.
5. Conclusions
Long-term storage at 0.5 °C and short-term storage at 6 °C (both
storage regimes yielding 14°-days) negative affected the performance of
cv Akito and Grand Prix roses, both when placed in water and in water
+ 8-HQC. Stomata in cv Grand Prix leaves showed a more adequate
response to dehydration than stomata in cv Akito leaves. Stomatal
functionality following storage as measured at the start of the vase life
was negatively influenced by both storage regimes in cv Akito; stomata
in leaves from stored flowers closed at lower RWC, showed a relatively
slow closing response and were more leaky when closed compared to
stomata in leaves of non-stored flowers. In cv Akito there was a good
correlation between stomatal functionality and flower performance
parameters, indicating that stomatal functionality may be a factor de-
termining the negative impact of storage on flower performance in this
cv.
In cv Grand Prix, the storage negatively affected the vase life but no
effect of storage on stomatal functionality was observed. Apart from
stomatal functionality other factors related to the flower water balance
(e.g. xylem resistance to water uptake) likely play a role in the negative
effect of cold storage on flower performance. It was argued that bac-
terial development in stems during storage was not the causative factor
of the presumed increased xylem resistance.
Acknowledgements
The research was carried out within the GreenCHAINge project, a
cooperation between the Dutch Association of Wholesale Trade in
73
Postharvest Biology and Technology 136 (2018) 66–73
Horticultural Products (‘VGB’), the Dutch Federation of Agriculture and
Horticulture (‘LTO Glaskracht Nederland’) and Wageningen
Food & Biobased Research. Financial support came from the
Horticultural Marketing Board (‘Productschap Tuinbouw’) and the
Ministry of Economic Affairs via the research program ‘Topsector
Tuinbouw en Uitgangsmaterialen’. Authors are grateful to Harmannus
Harkema and Els Otma for technical assistance and to Bastiaan Brouwer
for critically reading the manuscript.
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in the
online version, at https://doi.org/10.1016/j.postharvbio.2017.10.009.
References
Aliniaeifard, S., Van Meeteren, U., 2013. Can prolonged exposure to low VPD disturb the
ABA signalling in stomatal guard cells? J. Exp. Bot. 64, 3551–3566.
Aliniaeifard, S., van Meeteren, U., 2014. Natural variation in stomatal response to closing
stimuli among Arabidopsis thaliana accessions after exposure to low VPD as a tool to
recognize the mechanism of disturbed stomatal functioning. J. Exp. Bot. 65,
6529–6542.
Aliniaeifard, S., Malcolm Matamoros, P., van Meeteren, U., 2014. Stomatal mal-
functioning under low VPD conditions: induced by alterations in stomatal mor-
phology and leaf anatomy or in the ABA signaling? Physiol. Plant. 152, 688–699.
Bleeksma, H.C., Van Doorn, W.G., 2003. Embolism in rose stems as a result of vascular
occlusion by bacteria. Postharvest Biol. Technol. 29, 335–341.
Fanourakis, D., Carvalho, S.M.P., Almeida, D.P.F., van Kooten, O., van Doorn, W.G.,
Heuvelink, E., 2012. Postharvest water relations in cut rose cultivars with contrasting
sensitivity to high relative air humidity during growth. Postharvest Biol. Technol. 64,
64–73.
Fanourakis, D., Heuvelink, E., Carvalho, S.M.P., 2013a. A comprehensive analysis of the
physiological and anatomical components involved in higher water loss rates after
leaf development at high humidity. J. Plant Physiol. 170, 890–898.
Fanourakis, D., Pieruschka, R., Savvides, A., Macnish, A.J., Sarlikioti, V., Woltering, E.J.,
2013b. Sources of vase life variation in cut roses: a review. Postharvest Biol. Technol.
78, 1–15.
Fanourakis, D., Velez-Ramirez, A.I., In, B.C., Barendse, H., van Meeteren, U., Woltering,
E.J., 2015. A survey of postharvest conditions affecting the regulation of water loss
during vase life. In: Acta Horticulturae, 1064 Ed. International Society for
Horticultural Science (ISHS). Leuven, Belgium. pp. 195–204.
Fanourakis, D., Bouranis, D., Giday, H., Carvalho, D.R.A., Rezaei Nejad, A., Ottosen, C.-
O., 2016a. Improving stomatal functioning at elevated growth air humidity: a review.
J. Plant Physiol. 207, 51–60.
Fanourakis, D., Giday, H., Li, T., Kambourakis, E., Ligoxigakis, E.K., Papadimitriou, M.,
Strataridaki, A., Bouranis, D., Fiorani, F., Heuvelink, E., Ottosen, C.-O., 2016b.
Antitranspirant compounds alleviate the mild-desiccation-induced reduction of vase
life in cut roses. Postharvest Biol. Technol. 117, 110–117.
Giday, H., Fanourakis, D., Kjaer, K.H., Fomsgaard, I.S., Ottosen, C.-O., 2013. Foliar ab-
scisic acid content underlies genotypic variation in stomatal responsiveness after
growth at high relative air humidity. Ann. Bot. 112, 1857–1867.
Giday, H., Fanourakis, D., Kjaer, K.H., Fomsgaard, I.S., Ottosen, C.-O., 2014. Threshold
response of stomatal closing ability to leaf abscisic acid concentration during growth.
J. Exp. Bot. 65, 4361–4370.
Mayak, S., Halevy, A.H., Sagie, S., Bar-Yoseph, A., Bravdo, B., 1974. The water balance of
cut rose flowers. Physiol. Plant. 31, 15–22.
Mortensen, L.M., Fjeld, T., 1998. Effects of air humidity, lighting period and lamp type on
growth and vase life of roses. Sci. Hortic. 73, 229–237.
Nejad, A.R., van Meeteren, U., 2007. The role of abscisic acid in disturbed stomatal re-
sponse characteristics of Tradescantia virginiana during growth at high relative air
humidity. J. Exp. Bot. 58, 627–636.
Tallman, G., 2004. Are diurnal patterns of stomatal movement the result of alternating
metabolism of endogenous guard cell ABA and accumulation of ABA delivered to the
apoplast around guard cells by transpiration? J. Exp. Bot. 55, 1963–1976.
Tromp, S.-O., van der Sman, R.G.M., Vollebregt, H.M., Woltering, E.J., 2012. On the
prediction of the remaining vase life of cut roses. Postharvest Biol. Technol. 70,
42–50.
van Doorn, W.G., 1997. Water relations of cut flowers. Hortic. Rev. 18, 1–68.
van Doorn, W.G., 2012. Water relations of cut flowers: an update. Hortic. Rev. 40,
55–106.
van Doorn, W.G., Perik, R.R.J., 1990. Hydroxyquinoline citrate and low pH prevent
vascular blockage in stems of cut rose flowers by reducing the number of bacteria. J.
Am. Soc. Hortic. Sci. 115, 979–981.
van Rijswick, C., 2015. World Floriculture Map 2015. Rabobank Industry Note 475.
Available from: https://www.rabobank.com/en/images/World_Floriculture_Map_
2015_vanRijswick_Jan2015.pdf (Accessed 28 June, 2017).