Dynamic changes of the ethylene biosynthesis in Jonagold apple

© 2013 Scandinavian Plant Physiology Society, ISSN 0031-9317

Physiologia Plantarum 150: 161–173. 2014

Dynamic changes of the ethylene biosynthesis in ‘Jonagold’ apple Inge Bulensa , Bram Van de Poela , Maarten L. A. T. M. Hertoga,∗ , Simona M. Cristescub , Frans J. M. Harrenb , Maurice P. De Proftc , Annemie H. Geeraerda and Bart M. Nicolaia a

Division of Mechatronics, Biostatistics and Sensors (MeBioS), Department of Biosystems (BIOSYST), Katholieke Universiteit Leuven, Willem de Croylaan 42, bus 2428, B-3001, Leuven, Belgium b Life Science Trace Gas Research Group, Molecular and Laser Physics, Institute for Molecules and Materials, Radboud University, P.O. Box 9010, NL-6500 GL, Nijmegen, the Netherlands c Division of Crop Biotechnics, Department of Biosystems (BIOSYST), Katholieke Universiteit Leuven, Willem de Croylaan 42, bus 2427, B-3001, Leuven, Belgium

Correspondence *Corresponding author, e-mail: [email protected] Received 22 October 2012; revised 21 March 2013 doi:10.1111/ppl.12084

In this study, the short-term and dynamic changes of the ethylene biosynthesis of Jonagold apple during and after application of controlled atmosphere (CA) storage conditions were quantified using a systems biology approach. Rapid responses to imposed temperature and atmospheric conditions were captured by continuous online photoacoustic ethylene measurements. Discrete destructive sampling was done to understand observed changes of ethylene biosynthesis at the transcriptional, translational and metabolic level. Application of the ethylene inhibitor 1-methylcyclopropene (1-MCP) allowed for the discrimination between ethylene-mediated changes and ethylene-independent changes related to the imposed conditions. Online ethylene measurements showed fast and slower responses during and after application of CA conditions. The changes in 1-aminocyclopropane1-carboxylate synthase (ACS) activity were most correlated with changes in ACS1 expression and regulated the cold-induced increase in ethylene production during the early chilling phase. Transcription of ACS3 was found ethylene independent and was triggered upon warming of CA-stored apples. Increased expression of ACO1 during shelf life led to a strong increase in 1-aminocyclopropane-1-carboxylate oxidase (ACO) activity, required for the exponential production of ethylene during system 2. Expression of ACO2 and ACO3 was upregulated in 1-MCP-treated fruit showing a negative correlation with ethylene production. ACO activity never became rate limiting.

Introduction Postharvest life of fruits and vegetables can be prolonged substantially if they are harvested and stored at optimal conditions. The latter depend on species, cultivar, season and ultimately even on the individual fruit.

The use of controlled atmosphere (CA) conditions (reduced O2 and elevated CO2 ) for prolonging the storage life of apple (Malus × domestica) has been introduced over 80 years ago (Smock and Van Doren 1941). The better the processes involved in fruit ripening and senescing are understood, the easier and more

Abbreviations – 1-MCP, 1-methylcyclopropene; ACC, 1-aminocyclopropane-1-carboxylic acid; ACO, 1-aminocyclo propane-1-carboxylate oxidase; ACS, 1-aminocyclopropane-1-carboxylate synthase; CA, controlled atmosphere; MACC, 1-(malonylamino)cyclopropane-1-carboxylic acid; MTA, 5 -methylthioadenosine; SAM, S-adenosyl-L-methionine.

Physiol. Plant. 150, 2014

161

accurate storage conditions can be optimized to retain high quality throughout the postharvest chain. The phytohormone ethylene plays a crucial role in the onset and regulation of ripening of climacteric fruits like apple. Two biosynthetic systems have been discovered. System 1 functions during normal vegetative growth, is autoinhibited by ethylene and is responsible for the production of basal levels of ethylene by all plant tissues, including non-climacteric fruit. System 2 occurs during ripening of climacteric fruit and is characterized by an autocatalytic ethylene response (McMurchie et al. 1972). The metabolite S -adenosyl-L-methionine (SAM) can be regarded as the starting point of the ethylene biosynthesis pathway (Adams and Yang 1977). SAM is converted into 1-aminocyclopropane-1-carboxylic acid (ACC) and 5 -methylthioadenosine (MTA) by the enzyme 1aminocyclopropane-1-carboxylate synthase (ACS). The MTA moiety can be recycled through the Yang cycle by an energy (ATP)-requiring process. It is assumed that the pool of SAM is non-limiting as long as the respiration rate remains sufficiently high (Barry and Giovannoni 2007), and was confirmed for Jonagold apple during CA storage and shelf life (Bulens et al. 2012). The protein ACS is a pyridoxal-5 -dependent enzyme encoded by a multigene family and is considered the main regulating enzyme of the ethylene biosynthesis (Argueso et al. 2007). Regulation can take place at the transcriptional as well as at the post-translational level (Tatsuki 2010). ACC can be converted into the biologically inactive 1-(malonylamino)cyclopropane-1-carboxylic acid (MACC) by ACC n-malonyl transferase (Hoffman and Yang 1982) or ethylene by 1-aminocyclopropane-1carboxylate oxidase (ACO; Adams and Yang 1979). For this final conversion, Fe2+ and ascorbate are required as a cofactor and a co-substrate, respectively (Lin et al. 2009). The regulation and production of ethylene is influenced by the action of harvesting and the drastic environmental changes when apples are put under CA storage conditions. The effect of harvest is known to differ between apple cultivars. For various early and mid-season cultivars, including Jonagold, it has been suggested that fruit detachment triggers the onset of autocatalytic ethylene production (system 2) related to ripening (Wilkinson 1963, Sfakiota and Dilley 1973, Lau et al. 1986, Lin and Walsh 2008, Bulens et al. 2012). Exposure of pome fruit to low temperature differentially affects the ethylene biosynthesis depending on the cultivar, perhaps owing to a genetically based difference in their sensitivity to cold acclimation (Larrigaudiere et al. 1997). In general, long-term cold storage inhibits ethylene biosynthesis, whereas shorter periods can delay (Royal Gala), depress (Starking Delicious) or stimulate 162

(Granny Smith) ethylene biosynthesis during subsequent warming up (Lelievre et al. 1995, Larrigaudiere et al. 1997). Extreme low O2 concentrations lead to reduced ACS abundance, whereas high CO2 concentration caused the enzyme activity to decrease (Gorny and Kader 1996). Additionally, CO2 is a competitive inhibitor of ethylene and thus high levels of CO2 suppress ethylene action (Burg and Burg 1967). The conversion of ACC to ethylene catalyzed by ACO is oxygen dependent, and under anaerobic conditions ethylene formation is completely suppressed (Yang 1985, Lin et al. 2009). Most published work on ethylene biosynthesis during apple storage focuses on the long-term global changes observed throughout the whole 4–6 months storage period (e.g. Bulens et al. 2012). To better understand the regulation behind the fruit’s adaptation to the imposed changes in terms of O2 , CO2 and temperature, attention has to be shifted from the timescale of months down to a timescale of hours to days. This is the first study focusing on the short-term and relative fast dynamic changes in response to the discrete changes imposed during the application of CA storage. The four events studied are (1) cooling the fruit while still at regular air conditions, (2) at low temperature changing the gas composition of the storage atmosphere to CA, and toward the end of storage the reversed process of (3) at low temperature changing the gas composition of the storage atmosphere back to regular air and (4) warming up the fruit to room temperature. The relative rapid changes in ethylene production occurring during these events were captured using continuous online ethylene detection methods allowing us to two-hourly monitor the real-time ethylene dynamics. In this manner, a clear distinction could be made between fast (within hours) and slower (extending over several days) responses to the changes in temperature and atmospheric composition. Additionally, discrete destructive sampling in time was done to gather information on the origin of the observed changes in ethylene biosynthesis at the transcriptional, translational and metabolic level by sampling fruit on 12-hourly intervals.

Materials and methods Experimental setup

Plant material Jonagold apples were harvested at the Research Station for Fruit Growing (pcfruit, Sint-Truiden, Belgium) 1 week after the commercial harvest window (October 11, 2010). Half of the fruit were treated within hours after harvest with 1-methylcyclopropene (1-MCP; concentration of 600 nl l−1 obtained by dissolving 1 g of


Smartfresh™ per cubic meter in 20 ml H2 O, at room temperature). To study the dynamic changes of the ethylene biosynthesis during the application of and subsequent release of CA storage conditions, Jonagold apples were subjected to sequential changes in temperature (18 to 1◦ C and vice versa) and atmosphere conditions (normal air to CA: 1% O2 , 3% CO2 and vice versa).

apples. Each apple was individually placed in a glass jar 24 h before connecting it to the valve controller and continuously flushed at the same rate (1 l h−1 ) as during the measurement to obtain equilibrium between the headspace and the apple.

Experimental design

To obtain information about the change of the intermediates and the enzymes of the ethylene biosynthesis pathway in time, destructive tissue samples were collected from the individual apples by the end of the 1-day monitoring period (see Figs S1B and S2B). To anticipate on the fact that biological variability might be larger than the changes in time, parallel to the experiment with the individually monitored apples, pooled samples were taken (see Figs S1C and S2C). For these pooled samples, six control apples and six 1-MCP-treated apples were cut up, mixed and crushed in liquid nitrogen. The apples for the pooled samples were from the same batch of fruit and were treated and sampled similarly and in parallel to the individually monitored apples.

The experiment consisted of two parts: (1) the application of the CA conditions and (2) after 92 days of storage, the release of the CA conditions. During and shortly after the application of CA storage conditions, dynamic changes of the ethylene biosynthesis were closely monitored during a period of 10 days as described in the next section. In the 20 days following this 10-day period, sampling continued at a reduced frequency. Subsequently, the remaining apples continued to be stored at CA conditions for 60 days without being sampled. After this storage period (91 days after harvest), sampling started again for a period of 14 days during which subsequently CA conditions were released and temperature was increased.

Destructive discrete sampling

Methods

Continuous ethylene measurements

Ethylene measurements

To measure the change in ethylene production rate as accurately as possible, ethylene production was measured online using a laser-based photoacoustic detector. The specific resonance frequency for ethylene in both air and CA was determined prior to measuring and adjusted when the gas conditions changed. The detector was connected to a gas handling system that provides the gas mixture to the samples and automatically switches between six channels. One of them was used as a reference leaving five available for online ethylene measurement of apples individually enclosed in glass jars and continuously flushed with a gas mixture of known composition (either air or CA). Two of the channels were used to continuously monitor one control and one 1-MCP-treated apple. This generated a complete picture for one individual apple per treatment ignoring biological variability. To include biological variability, the remaining channels were used to sequentially monitor individual apples for shorter 1-day period. By overlapping these shorter periods, two to three biological replicates were obtained for both the control and the 1-MCP-treated apples for each moment in time (see Figs S1A and S2A, Supporting Information). Until measurement all apples were kept in a large container at the same temperature and under the same CA conditions as the continuously monitored

Online ethylene measurements were performed with a laser-based photoacoustic ethylene detector (ETD-300, Sensor sense, Nijmegen, the Netherlands) (Cristescu et al. 2008). Prior to analysis, gas samples were scrubbed to remove moisture and CO2 . By using a valve controller six different channels (five connected to samples and one reference channel) were measured in a continuous cycle. Each channel was measured for 20 min. Depending on the difference in ethylene production between consecutive samples, a stable ethylene reading was obtained after 5–15 min. The ethylene concentration was calculated based on the measurements of the last 2 min to ensure a stable result in every situation. Ethylene measurements at discrete moments in time were performed by quantifying the ethylene concentration in the headspace of a sealed respiration jar with a compact GC equipped with FID detector (Interscience, Louvain la Neuve, Belgium) as described by Bulens et al. (2011).


Biochemical analyses The metabolites SAM, ACC and MACC and the in vitro activity of ACC synthase and ACC oxidase were quantified as described by Van de Poel et al. (2010) and Bulens et al. (2011). 163

Western blotting Polyclonal antibodies were developed against a consensus epitope for four ACO isoforms (ACO1, ACO2, ACO3 and ACO4: CQDDKVSGLQLLKDE). The total protein content of ACO extracts was determined by the Bradford assay (Bradford 1976). For western blotting, 15 μg of total protein was used as described by Van de Poel et al. (2012). Accumulation of ACO protein was determined in control and treated pooled samples taken at days 0, 3, 5, 6, 22, 35, 93, 95, 97, 100, 107 and 118 of the experiment.

RNA extraction and RT-qPCR Extraction of 4 g of frozen crushed apple tissue was performed as described by Van de Poel et al. (2012). Purification of the extract and DNA removal was carried out with the Qiagen RNeasy Plus Mini Kit (Qiagen GmbH, Hilden, Germany). After the RNA integrity was checked on an agarose gel and the RNA content was quantified, 1 μg of RNA was reverse transcribed into cDNA and the real time qPCR reaction was performed (Van de Poel et al. 2012). The selected primers are shown in Table S1. The results for each gene were normalized against the average expression of three reference genes (MdS26, Md-Ubiquitin and Md-Elongationfactor1; see Table S1).

Results Primary and secondary ethylene response due to changing conditions After harvest, when apples were kept at 18◦ C in air, ethylene production of the control fruit increased because of the onset of climacteric ripening (Fig. 1). Compared with the control apples, the ethylene production of the 1-MCP-treated apples was 10 times lower at time 0 and remained stable. When temperature was reduced to 1◦ C, ethylene production of both apples dropped. After 12 h the ethylene production of the control apple recovered and seemed to continue its climacteric rise. Four days after the change in temperature, the introduction of CA conditions caused an initial fast decrease, followed by a secondary recovery response. After 2 days of CA storage, the ethylene production rate of the control apple stabilized. Further storage at CA conditions led to a slow decrease in ethylene product rate, shown by the discrete measurements at days 13, 16 and 22. The 1-MCP-treated apple also showed two fast decreasing ethylene responses when temperature or gas conditions changed. But compared with the control apple the 1-MCP-treated apples did not fully recover; ethylene production stays suppressed because 164

Fig. 1. Ethylene production rate of an individual control and 1-MCPtreated apple during changing temperature and storage gas conditions. ( : 1-MCP-treated fruit, continuous measurement; : control fruit, continuous measurement; : control fruit, discrete measurements in time).

of the combination of low temperature, CA conditions and 1-MCP treatment. After almost 90 days of storage at CA conditions, ethylene production had stabilized in both control and 1-MCP-treated apples, with the latter showing a 100-fold lower ethylene production when compared with the former. When the CA gas conditions were removed, a fast increase was measured in both the control and treated apple (although of a different order of magnitude). Ethylene production rates then stabilized at the same level as measured under CA conditions. It was only after temperature was increased that not only a fast primary increase was observed but also a slower but lasting secondary response. For the control apple this resulted in a climacteric increase, whereas the 1-MCP-treated apple retained a low and constant ethylene production throughout shelf life, 10 times higher than the production level during CA storage but 1000 times less than that of the control fruit. Overall, we observed two distinct ethylene responses: a fast primary response to changing temperature or CA conditions and a long-term secondary response due to habituation to the new environment (CA storage or shelf life). Biological variability affects the magnitude of the ethylene response At the start of storage during the application of CA conditions large differences existed between the ethylene production rates of individual fruit (Fig. 2, left). Nevertheless, the fast primary and slower secondary


Fig. 2. Ethylene production rate during the application (left) and release (right) of CA conditions for various biological replicates (indicated by different symbols) of control (top) and 1-MCP-treated (bottom) apples.

ethylene responses, described in the previous section, were clearly visible in all replicates for the control fruit (Fig. 2, top) and the 1-MCP-treated fruit (Fig. 2, bottom). After 3 months of CA storage upon releasing of the CA conditions, the biological variability was strongly diminished for the control apples (Fig. 2, right), whereas for the 1-MCP-treated apples this reduction in biological variation occurred only during shelf life. This suggests that a long-term storage period conditions the fruit toward a similar steady-state level of ethylene production, at least for non-1-MCP-treated fruit. ACC synthase activity and expression is regulated by temperature The activity of ACS (Fig. 3) in the 1-MCP-treated apples was almost completely suppressed throughout the experiment except for the last 2 weeks at shelf life


conditions (days 104–118), indicating that the 1-MCP treatment effectively blocked all ACS activity. For the control apples, ACS activity first decreased but then increased again during the period of precooling and the introduction of the CA gas conditions (day 5.5) (Fig. 3, inset). ACS activity reached a maximum around day 12 after which the activity steadily went down during further CA storage. This initial increase in ACS activity was also observed in an independent experiment where the effect of harvest maturity was included (see Fig. S3A). In those fruit only late harvested apples showed the early increase in ACS activity. During the change from CA to air conditions (starting from day 92.5) the climacteric-related increase in ACS activity only commenced after the temperature elevation. Of the four known isoforms of ACS in Malus × domestica (MdACS1, MdACS2, MdACS3 and MdACS5) only ACS1 and ACS3 were found to be 165

Fig. 3. In vitro ACS activity of individual and pooled samples during the application and the subsequent removal of CA conditions (with the vertical lines indicating the transitions between temperature and gas conditions and : pooled samples of control fruit; : individual samples of control fruit; : pooled samples of 1-MCP-treated fruit; : individual samples of 1-MCP-treated fruit). The inset figure contains an enlarged view of the changes during the first 30 days.

expressed in fruit tissue. The expression profile of these two genes was analyzed for the 46 pooled samples taken throughout the experiment (Fig. 4; sampling scheme Figs S1 and S2). It is known that in ripening apple fruit, ethylene production is strictly related to the level of ACS1 expression (Wang et al. 2009). This relationship was also observed in our experiment. The global pattern of ACS1 expression (Fig. 4, top) matches the pattern of ACS activity (Fig. 3) as well as the fruit ethylene production rate (Figs 1 and 2). The small increase in ACS1 expression during cooling and during the application of CA conditions seemed to be sufficient to cause the observed increase in ACS activity and the related increase in ethylene production rate. During the shelf life period at day 100 and beyond, ACS1 expression augmented rapidly. A delay was observed in the increase in ACS1 expression between the 1-MCP-treated fruit and the control fruit during this stage. Surprisingly, the 1-MCP-treated fruit reached a similar expression level of ACS1 expression as the control fruit. Prior to the increase in ACS1 expression during the initiation of CA storage, ACS3 expression dropped regardless of the treatment applied (Fig. 4, bottom). Ten days after harvest (day 11 in Fig. 4) the expression of ACS3 was completely suppressed. This is in accordance with the findings that changes in ACS3 expression trigger the transition of system 1 to system 2 ethylene production and that ACS3 is regulated by a negative feedback system 166

Fig. 4. Relative gene expression of ACS1 and ACS3 in pooled apple samples (based on six replicates) during the application and the subsequent removal of CA conditions (with the vertical lines indicating the transitions between temperature and gas conditions and : 1-MCPtreated fruit; : control fruit).

(Varanasi et al. 2011). It is intriguing, though, to see that upon releasing the CA storage conditions ACS3 was upregulated again in parallel to ACS1 while at this moment ethylene production is still high. The increase in relative ACS3 expression peaks around day 100 and day 110 for the control and treated apples, respectively. Our results clearly demonstrated that the ACS expression is differentially regulated by temperature causing altered ethylene dynamics. The gas conditions do not seem to regulate ACS expression. Differential response in ACO activity and expression during CA storage and shelf life ACO activity slightly decreased after lowering the temperature from 18 to 1◦ C, but unlike for ethylene production no immediate recovery was observed (Fig. 5A). During the first 3 weeks of storage the activity was


Fig. 5. (A) In vitro ACO activity of individual and pooled samples during the application and the subsequent removal of CA conditions (with the vertical lines indicating the transitions between temperature and gas conditions and : pooled samples of control fruit; : individual samples of control fruit; : pooled samples of 1-MCP-treated fruit; : individual samples of 1-MCP-treated fruit). The inset figure contains an enlarged view of the changes during the first 30 days. (B) Western blot of ACO protein accumulation in pooled apple samples (control and 1-MCP treated) taken at days 107 and 118.

lower than at harvest. After 30 days a small increase in activity was observed for the control fruit, but this did not result in an increase in ethylene production which was stabilizing at this moment in time (Fig. 1). Until day 96, the effect of 1-MCP on the ACO activity was negligible compared with its effect on ethylene production rate. After increasing temperature (day 96.5) the effect of 1-MCP manifested. ACO activity of control apples increased rapidly, whereas the activity of the treated apples lags behind for about 10 days. This was also confirmed at the protein level, because western blots against ACO showed a reduced ACO abundance in 1-MCP-treated fruit during the final stages of shelf life (Fig. 5B; ACO protein levels were not detected during the previous storage periods). Untreated control fruit also showed a higher ACO abundance than the 1-MCP-treated fruit, matching the higher ACO activities observed. These large differences in ACO abundance are not always correlated with the actual in vitro activity. At the final time point (day 118) the 1-MCP-treated apple showed a comparable in vitro activity to the control apple, but contained less ACO protein. This might indicate the possibility of a post-translational modification of ACO during the final stages of shelf life. Additional information about the influence of harvest date on the changes in ACO activity during CA storage is added in Fig. S3C. It is clear from these results that the


later the fruits are harvested (more mature) the faster the increase in ACO activity in storage. From the three known ACO isoforms (MdACO1, MdACO2 and MdACO3) expressed during fruit ripening, ACO1 (Fig. 6 top) is believed to be the isoform contributing most to the activity of the enzyme (Binnie and McManus 2009). At harvest the expression of ACO1 was still low but it was transiently upregulated during CA storage. After 60 days of storage and upon releasing of CA storage and low temperature conditions, ACO1 expression increased strongly during the shelf life period, matching the observed increased ACO activity during these stages (Fig. 5A). Parallel to the activity, the ACO1 expression of the 1-MCP-treated apples lagged 10 days behind. The expression patterns of the other two isoforms, ACO2 and ACO3, differed strongly from ACO1 (Fig. 6). The expression of ACO2 was low during the first 20 days of the experiment, but then steadily increased during CA storage and reached its highest expression level after 85 days of CA storage. Surprisingly, the expression was higher in 1-MCP-treated apples than in the control fruit. When CA conditions were released ACO2 expression started to fall back and this decrease went on when the temperature was raised from 1 to 18◦ C. Expression of ACO3 increased rapidly as the temperature declined at the start of the experiment but decreased when the gas composition was changed as 167

Fig. 6. Relative gene expression of ACO1, ACO2 and ACO3 in pooled apple samples (based on six replicates) during the application and the subsequent removal of CA conditions (with the vertical lines indicating the transitions between temperature and gas conditions and : 1-MCPtreated fruit; : control fruit).

well. When the CA conditions were released toward the end of storage, ACO3 expression started to increase, which was further enhanced by the change in gas composition. ACO3 expression rapidly fell back again when temperature was increased to 18◦ C. 168

Fig. 7. SAM (A), ACC (B) and MACC (C) concentration of individual and pooled samples during the application and the subsequent removal of CA conditions (with the vertical lines indicating the transitions between temperature and gas conditions and : pooled samples of control fruit; : individual samples of control fruit; : pooled samples of 1-MCP-treated fruit; : individual samples of 1-MCP-treated fruit).

Quantification of the ethylene biosynthesis intermediates To better understand the observed changes in enzyme activity and their related expression patterns, we analyzed the intermediate metabolites (SAM, ACC and


Fig. 8. Overview of the changes in ethylene biosynthesis on both the transcriptional and translational level for the untreated apples. The 5 main phases of postharvest storage (cooling, introducing CA, ongoing CA storage, releasing CA and finally shelflife) are represented by 5 subsequent squares each coloured according the level of change observed during each phase (increase, no change, decrease).

MACC) of the ethylene pathway. In the first part of the experiment (implementation of low temperature and gas conditions) the SAM levels stayed more or less constant, whereas the ethylene production slightly increased. This might indicate that the substrate was not a rate-limiting factor in the formation of ACC at this stage (Fig. 7A). When the CA conditions were released at the end of storage, SAM sharply dropped. Upon increasing the temperature from 1 to 18◦ C, SAM levels quickly increased again. During further shelf life storage, SAM levels dropped again. This coincided with the bulk of climacteric ethylene (Fig. 2). These results confirm earlier published results where we also noticed a consistent drop in SAM during shelf life (Bulens et al. 2012). The reduced SAM pool did not seem to become rate limiting to sustain the high ethylene production rate. The ACC concentration (Fig. 7B) increased dramatically when CA conditions were applied at day 5.5. This corresponded to the increase in ACS activity observed at these stages. During ongoing CA storage ACC levels gradually dropped again matching the slight increase in ACO activity and ethylene production during storage. In the second part of the experiment when CA conditions were released and temperature was increased again, ACC levels changed in parallel to the observed changes in SAM (drop during CA removal and temporal increase when temperature increased). The 1-MCP treatment resulted in no ACC accumulation throughout the entire storage period, matching the inhibited ACS activity observed in Fig. 3. Only during the late shelf


life period, some increase in ACC was observed. During the induction of CA and during storage itself, the MACC concentration was low and constant for both the control and the 1-MCP-treated apples (Fig. 7C). This indicates that the enzyme malonyl transferase was not active yet and was unaffected by the changing temperature or gas conditions. During the second part of the experiment (release of the CA conditions) MACC levels started to accumulate. The 1-MCP-treated samples showed a delayed and reduced increase compared with the control apples. Whether this reduction in MACC is owing to the direct suppression of ACC-malonyl transferase activity by 1-MCP or owing to a lower availability of ACC caused by the ACS inhibition is not known. An overview of the changes in ethylene biosynthesis on both the transcriptional and translational level for the untreated apples is given in Fig. 8. When the pattern of changes upon changing temperature and gas conditions are compared, a clear correlation is observed between the expression of ACS1, the ACS activity and the ethylene production rate. The complete set of changes is complex and is discussed in more detail in the next section.

Discussion Long-term chilling and application of CA conditions lead to the suppression of ethylene biosynthesis in apple. In this experiment, it was shown that chilling and application of CA conditions actively suppressed the increase in the ethylene biosynthesis of Jonagold apples that occurred after harvest. After removal from storage 169

the climacteric increase in ethylene production resumed. The effect of the change in atmospheric composition was minor compared with the effect of temperature increase. This indicates that small interruptions in CA conditions during long-term storage do not influence the ethylene production rate as long as the temperature is being controlled. During and after storage, a distinction could be made between fast and slower responses. The ethylene production rate is the result of the turnover of the involved metabolites by the presented enzymes, which exert a certain activity. This activity is strongly influenced by the surrounding conditions (Lelievre et al. 1997). The drastic change in temperature and gas conditions throughout this experiment would definitely influence the in vivo activity of the enzymes determining the observed ethylene production rates and changes in the metabolite levels. It is reasonable to assume that this effect would only occur during a short transition period. This transition period is defined by the time it takes to reach and equilibrium between the internal temperature and gas composition of the fruit and the surrounding temperature and atmospheric composition. As a consequence, the fast (12–24 h) lasting response in ethylene production rate that is observed after changing the conditions might be caused mainly by changes in in vivo activity. This would also explain the parallel behavior between biological replicates but especially between control and treated fruit despite the large differences in ethylene production rate. The biological variation in ethylene production was reduced upon rewarming of control apples after CA storage, which is in accordance with previously published results (Larrigaudiere et al. 1999, Bulens et al. 2012). It seems that the biological variation is related to the ripeness stage. During the onset of climacteric ripening large differences can be observed between individual fruits, while once apples are fully climacteric they produce ethylene at a similar high level. The observed dynamic changes of the ethylene biosynthesis can be separated in three parts, the effect of application of CA storage, longterm effect during CA storage and the effect of release of CA conditions. ACS mediates the early cold- and CA-induced ethylene response Our results showed that ethylene production increased during the initial drop in temperature from 18 to 1◦ C. This cold-induced triggering of the ethylene production has previously been described for ‘Golden Delicious’ and ‘Golden Smoothee’ apples (Knee et al. 1983, Vilaplana et al. 2007). The cold activation of the ethylene production in apple is not only cultivar dependent but 170

is also influenced by the maturity stage upon ripening (Larrigaudiere et al. 1997). ‘Jonagold’ apples harvested 2 weeks after commercial harvest showed a strong coldinduced response, whereas apples harvested at the commercial harvest date and 2 weeks before did not. This leads to the assumption that the difference in response might be influenced by the ethylene production rate at the moment of harvest rather than by the difference in cultivar. Larrigaudiere et al. (1999) assumed that this maturity-based accumulation in ACC and subsequent increase in ethylene production were owing to a difference in ACO activity. Our results, however, showed that the ACS activity drastically increased when applying low temperature and CA conditions, whereas the ACO activity remained unaltered. Also, we observed that the initial ACS activity shortly after inducing CA conditions is maturity stage dependent (Bulens et al. 2012). On the basis of these results, we conclude that the maturity stage-dependent differences in ACO activity are negligible compared with the difference in ACS activity (especially in the early stages in storage). This pinpoints ACS as the crucial enzyme of the early coldand CA-induced ethylene response. ACS activity is known to be tightly regulated by a low transcript abundance and a rapid protein turnover rate (Kim and Yang 1992). A small increase in ACS1 transcription will thus almost immediately result in a considerable increase in ACS activity and ethylene biosynthesis. Of the four known ACS genes in apple, only ACS1 and ACS3 were expressed in the fruit tissue, which is in accordance with the finding of Wiersma et al. (2007). The activity of ACS3 is known to be 40 times less than that of ACS1 (Wang et al. 2009), as a consequence its contribution to the measured activity is negligible. Nonetheless, ACS3 is believed to be responsible for the system 1 to system 2 transition (Wang et al. 2009, Varanasi et al. 2011). Before ACS1 expression started to increase (day 3: 1 day after cold treatment) ACS3 expression was already steeply declining. This decline was similar for the control and the 1-MCP-treated apples although in the treated apples no increase in ACS1 expression was observed as a result of this decline. It was postulated that the increase in system 2 ethylene negatively feeds back to system 1 causing ACS3 to decline (Wang et al. 2009). This might be the case for the control apples, but the question arises what stimulus causes ACS3 expression to decline in the treated apples when ethylene production is still low. Perhaps, an ethylene-independent developmental cue might cause the decline in ACS3 transcript abundance or this gene might be sensitive to cold and/or CA conditions.


ACO expression and activity is differentially regulated during long-term CA storage When ‘Jonagold’ apples were kept in CA conditions for a longer time the ethylene production stabilized and leveled off after 4–5 days. This long-term decrease in ethylene production was probably accelerated compared with cold air storage by the inhibitory effect of low O2 and increased CO2 on the ACS activity, which has already been observed in ‘Golden Delicious’ (Gorny and Kader 1996). ACO activity in untreated apples is inhibited by the application of CA storage conditions but not completely suppressed. During cold storage and in lesser extent during CA storage, the ACO activity of ‘Golden Delicious’ and ‘Jonagold’ slowly increases (Gorny and Kader 1996, Bulens et al. 2012) while in Braeburn it was suppressed (Tian et al. 2002). The term activity can, however, be deceiving because it concerns an in vitro measured activity under standardized aerobic conditions. Although being a good measure of the amount of active ACO present, this does not resemble the actual suppressed in vivo activity under low oxygen conditions as incurred during CA (Lelievre et al. 1997). An accumulation of ACO protein during cold storage was also reported for ‘Granny Smith’ (Lelievre et al. 1995). It is generally accepted that the change in ACO activity during storage is of minor importance compared with the regulatory role of ACS (Gorny and Kader 1996, Tian et al. 2002). The isoform ACO1 shows the highest activity [almost nine and two times more for ACO2 and ACO3, respectively (Binnie and McManus 2009)]. This might explain why the increasing expression of ACO2 (which is known to be prominent in preclimacteric apples; Binnie and McManus 2009) observed during CA storage in 1-MCP-treated apples did not result in an increase in in vitro activity. The expression of ACO2 was found to be negatively correlated with ethylene production, because 1-MCP-treated apples showed the highest expression. The same is observed for ACO3, but here the change in expression seems to be more related to changing gas conditions. Removal of CA conditions instantly activates the ethylene pathway Cold storage accumulates the expression of ACS and ACO allowing a fast and large burst in ethylene production upon rewarming (Vilaplana et al. 2007). ACS1 gene expression of ‘Jonagold’ apples increases slightly during CA storage, but falls back upon removal of CA conditions. When the temperature was increased an almost parallel increase was seen for ACS1 and ACS3 expression. The function of this increase in ACS3 expression in control apples at this moment in


time remains rather unclear. ACS1 expression (and as a consequence ACS activity) preceded the ACS3 increase and the system 2 ethylene response was already triggered. For the 1-MCP-treated apples, the decline in ACS3 expression coincided with the increase of ACS1 expression and ethylene production, so ACS3 might still perform a triggering function. On the basis of this experiment, it seems that the expression of ACS3 is ethylene independent rather than negatively controlled and is triggered in CA-stored ‘Jonagold’ apples upon rewarming. As opposed to ACS1 expression, ACO1 gene expression increased rapidly once CA conditions were removed. At the end of shelf life there was a discrepancy between the ACO activity and the amount of protein, similar to the observation by Gorny and Kader (1996) that a higher amount of protein does not necessarily coincide with a higher activity level. This might indicate the presence of inactive ACO proteins or perhaps a post-translational modification causing ACO to become inactive.

Conclusions The use of online ethylene measurements has shown fast and slower responses of the ethylene biosynthesis because of application and subsequent release of CA storage conditions. The fast response within 12–24 h is believed to be directly related to the change in temperature and gas conditions. The slower response can be correlated with long-term changes in the expression and activity of the biosynthetic enzymes ACS and ACO. The changes in ACS activity were most correlated with changes in ACS1 expression and were regulating the cold-induced increase in ethylene production during the early chilling phase. Transcription of ACS3 was found ethylene independent and was triggered upon rewarming of CA-stored apples. Increased expression of ACO1 during shelf life led to a strong increase in ACO activity, needed for the exponential production of ethylene during system 2. Expression of ACO2 and ACO3 was upregulated in 1-MCP-treated fruit, showing a negative correlation with ethylene production. Under the experimental conditions, ACO never became rate limiting although we saw some indications for post-translational control at the end of shelf life. Acknowledgements – The research was funded by the Institute for the Promotion of Innovation through Science and Technology in Flanders (IWT-Vlaanderen) (SB-71435). Online ethylene measurements were performed at the Life

171

Science Trace Gas Facility (Nijmegen, the Netherlands) in the framework of the FP6-026183 project of the EU.

References Adams DO, Yang SF (1977) Methionine metabolism in apple tissue – implication of S-adenosylmethionine as an intermediate in conversion of methionine to ethylene. Plant Physiol 60: 892–896 Adams DO, Yang SF (1979) Ethylene biosynthesis: Identification of 1-aminocyclopropane-1-carboxylic acid as an intermediate in the conversion of methionine to ethylene. Proceedings of the National Academy of Sciences of the United States of America 76: 170–174 Argueso CT, Hansen M, Kieber JJ (2007) Regulation of ethylene biosynthesis. J Plant Growth Regul 26: 92–105 Barry CS, Giovannoni JJ (2007) Ethylene and fruit ripening. J Plant Growth Regul 26: 143–159 Binnie IE, McManus MT (2009) Characterization of the 1-aminocyclopropane-1-carboxylic acid (ACC) oxidase multigene family of Malus domestica Borkh. Phytochemistry 70: 348–360 Bradford MM (1976) A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein–dye binding. Anal Biochem 72: 248–254 Bulens I, Van de Poel B, Hertog MLAT, De Proft MP, Geeraerd AH, Nicolai BM (2011) Protocol: an updated integrated methodology for analysis of metabolites and enzyme activities of ethylene biosynthesis. Plant Methods 7: 17 Bulens I, Van de Poel B, Hertog MLAT, De Proft MP, Geeraerd AH, Nicolai BM (2012) Influence of harvest time and 1-MCP application on postharvest ripening and ethylene biosynthesis of ‘Jonagold’ apple. Postharvest Biol Technol 72: 11–19 Burg SP, Burg EA (1967) Molecular requirements for biological activity of ethylene. Plant Physiol 42: 144 Cristescu SM, Persijn ST, Hekkert STL, Harren FJM (2008) Laser-based systems for trace gas detection in life sciences. Appl Phys B 92: 343–349 Gorny JR, Kader AA (1996) Regulation of ethylene biosynthesis in climacteric apple fruit by elevated CO2 and reduced O2 atmospheres. Postharvest Biol Technol 9: 311–323 Hoffman NE, Yang SF (1982) Identification of 1-(malonylamino)cyclopropane-1-carboxylic acid as a major conjugate of 1-aminocyclopropane-1-carboxylic acid, an ethylene precursor in higher plants. Biochem Biophys Res Commun 104: 765–770 Kim WT, Yang SF (1992) Turnover of 1-aminocyclopropane-1-carboxylic acid synthase protein in wounded tomato fruit tissue. Plant Physiol 100: 1126–1131

172

Knee M, Looney NE, Hatfield SGS, Smith SM (1983) Initiation of rapid ethylene synthesis by apple and pear fruits in relation to storage temperature. J Exp Bot 34: 1207–1212 Larrigaudiere C, Graell J, Salas J, Vendrell M (1997) Cultivar differences in the influence of a short period of cold storage on ethylene biosynthesis in apples. Postharvest Biol Technol 10: 21–27 Larrigaudiere C, Recasens I, Graell J, Vendrell M (1999) Harvest maturity related changes in the cold-induced activation of 1-aminocyclopropane-1-carboxylic acid metabolism in Granny Smith apples. Food Sci Technol Int 5: 223–228 Lau OL, Liu Y, Yang SF (1986) Effects of fruit detachment on ethylene biosynthesis and loss of flesh firmness, skin color, and starch in ripening Golden Delicious apples. J Am Soc Hortic Sci 111: 731–734 Lelievre JM, Tichit L, Fillion L, Larrigaudiere C, Vendrell M, Pech JC (1995) Cold-induced accumulation of 1-aminocyclopropane 1-carboxylate oxidase protein in Granny-Smith apples. Postharvest Biol Technol 5: 11–17 Lelievre JM, Latche A, Jones B, Bouzayen M, Pech JC (1997) Ethylene and fruit ripening. Physiol Plant 101: 727–739 Lin SF, Walsh CS (2008) Studies of the ‘‘tree factor’’ and its role in the maturation and ripening of ‘Gala’ and ‘Fuji’ apples. Postharvest Biol Technol 48: 99–106 Lin ZF, Zhong SL, Grierson D (2009) Recent advances in ethylene research. J Exp Bot 60: 3311–3336 McMurchie EJ, McGlasson WB, Eaks IL (1972) Treatment of fruit with propylene gives information about biogenesis of ethylene. Nature 237: 235 Sfakiota EM, Dilley DR (1973) Internal ethylene concentrations in apple fruits attached to or detached from tree. J Am Soc Hortic Sci 98: 501–503 Smock RM, Van Doren A (1941) Controlled-Atmosphere Storage of Apples. Cornell University Agricultural Experiment Station, Ithaca, NY Tatsuki M (2010) Ethylene biosynthesis and perception in fruit. J Jpn Soc Hortic Sci 79: 315–326 Tian MS, Prakash S, Zhang NX, Ross GS (2002) Chilling-induced ethylene biosynthesis in Braeburn apples. Plant Growth Regul 38: 249–257 Van de Poel B, Bulens I, Lagrain P, Pollet J, Hertog MLAT, Lammertyn J, De Proft MP, Nicolai BM, Geeraerd AH (2010) Determination of S-adenosyl-L-methionine in fruits by capillary electrophoresis. Phytochem Anal 21: 602–608 Van de Poel B, Bulens I, Markoela A, Hertog MLATM, Dreesen R, Wirtz M, Vandoninck S, Oppermann Y, Keulemans J, Hell R, Waelkens E, De Proft MP, Sauter M, Nicolai BM, Geeraerd AH (2012) Targeted systems biology profiling of tomato fruit reveals coordination of the Yang cycle and a distinct regulation of ethylene


biosynthesis during post-climacteric ripening. Plant Physiol 160: 1498–1514. DOI: 10.1104/pp.112.206086 Varanasi V, Shin S, Mattheis J, Rudell D, Zhu YM (2011) Expression profiles of the MdACS3 gene suggest a function as an accelerator of apple (Malus × domestica) fruit ripening. Postharvest Biol Technol 62: 141–148 Vilaplana R, Soria Y, Valentines MC, Larrigaudiere C (2007) Specific response of apple skin and pulp tissues to cold stress and 1-MCP treatment. Postharvest Biol Technol 43: 215–220 Wang A, Yamakake J, Kudo H, Wakasa Y, Hatsuyama Y, Igarashi M, Kasai A, Li TZ, Harada T (2009) Null mutation of the MdACS3 gene, coding for a ripening-specific 1-aminocyclopropane-1-carboxylate synthase, leads to long shelf life in apple fruit. Plant Physiol 151: 391–399 Wiersma PA, Zhang HY, Lu CW, Quail A, Toivonen PMA (2007) Survey of the expression of genes for ethylene synthesis and perception during maturation and ripening of ‘Sunrise’ and ‘Golden Delicious’ apple fruit. Postharvest Biol Technol 44: 204–211 Wilkinson BG (1963) Effect of time of picking on ethylene production of apples. Nature 199: 715 Yang SF (1985) Biosynthesis and action of ethylene. HortScience 20: 41–45

Supporting Information

specificity was checked by BLAST-ing against all apples ESTs and known cDNA sequences. All primers were optimized toward an annealing temperature of 60◦ C. Fig. S1. Overview of the ethylene measurements (A), sampling for the intermediates and enzymes on individual replicates (B) and on pooled samples (C) during the first part of the experiment (the application of CA conditions). Fig. S2. Overview of the ethylene measurements (A), sampling for the intermediates and enzymes on individual replicates (B) and on pooled samples (C) during the second part of the experiment (the removal of CA conditions). Fig. S3. Overview of the changes in ethylene biosynthesis during CA storage of Jonagold, harvest at three different moments in time (early: September 5, 2008; optimal: September 19, 2008; late: October 3, 2008). Values at day 0 represent the measurements on the day of harvest for each group. On the short term application of the CA conditions caused the ACS activity of late harvested control apples to increase drastically (A). This led to an increased ACC concentration (B) as well as increased ethylene biosynthesis (D). ACO activity was not affected upon application of CA conditions but increased steadily throughout storage (C).

Additional Supporting Information may be found in the online version of this article: Table S1. List of primers used and their properties. Primers were designed with Primer3 software. Primer

Edited by V. Shulaev


173