Relation etween Cultivar and Keeping Quality for Batches of Cucumers R.E. Schouten, L.M.M. Tijsens and O. van Kooten Horticultural Production Chains, Wageningen University, Wageningen, The Netherlands G. Jongloed Division of Matheics and Computer Science, The Vrije Universiteit, The Netherlands Keywords: Biological variation, postharvest, preharvest, precursor, heical approach Astract This paper focuses on the characterisation of the cultivar effect in atches with respect to atch eeping quality ased on colour measurements only. To do so, a correction for the urity per atch had to e made for two atches per cultivar. This left the cultivar inforion with respect to the atch eeping quality. This method to extract cultivar inforion out of cucumer atches is shown for six atches from three cultivars ( Volcan, Beluga and Borja ) over two growing seasons. The method consist of descriing the atch eeping quality determining property on an individual level and then translating it to a atch level, therey introducing and descriing iological variation. It turned out that the est cultivar, with respect to atch eeping quality, was Borja, followed y Beluga and finally Volcan. This paper demonstrates how to use the iological variance present in all iological atches; to tae advantage of the iological variation instead of seeing and dealing with it as if it were a nuisance. INTRODUCTION On first sight, it seems impossile to recognise whether fresh cucumers are from the same cultivar. Still, cucumer-reeding companies have sometimes large programs to develop new populations with genetically distinguishale properties. Those properties may well e eeping quality or resistance against pests. This paper focuses on the characterisation of the cultivar effect with respect to eeping quality. The limiting quality attriute for cucumers is colour (Schouten et al., 1997). The eeping quality can e defined as the time it taes for an individual cucumer to reach a predefined colour limit. Long shelf life of cucumers has een associated with high chlorophyll content in the peel (Lin and Jolliffe, 1995). However, it is nown that cucumers of the same colour can exhiit large differences in colour upon reaching the customer. So, for a high eeping quality, not only the initial colour is of interest, ut also specifically the aility to stay green. For individual cucumers this is not possile, ecause of the unnown stage of urity (Schouten et al., 1997). However, this is different when atches (one grower, one harvest and of the same cultivar) of cucumers are considered. On a atch level, next to inforion on all individuals elonging to that atch, extra inforion is availale due to the shared harvest date, cultivar and grower. Batch eeping quality can e defined as the time it taes efore 5% of the cucumers in a atch reaches the predefined colour limit (Schouten and Van Kooten, 1998). This paper focuses on the characterisation of cultivar effects with respect to atch eeping quality from a modelling point of view. Models will e presented on two aggregation levels. The first level is on the level of the individual cucumer, the second is the atch level which descries the atch itself, ut also a cultivar dependant parameter. All models are ased on non-destructive, repeated measurements, of colour. The models will e applied to six atches of cucumers from three cultivars over two growing seasons. Proc. Postharvest Unltd Eds. B.E. Verlinden et al. Acta Hort. 599, ISHS 23 451
MATERIAL AND METHODS Cucumers Cucumers consisted of six atches of 8-1 cucumers each, elonging to either cultivar Volcan, Borja or Beluga, otained from the Almeria region in Spain. Three atches were harvested in Septemer 1998 (autumn season) and three atches were harvested in June 1999 (spring season). All atches were grown under equal, commercial growing conditions and were of maretale size and colour. After harvest, atches were placed in oxes with air-filled polystyrene and transported to the measuring facility in the Netherlands within 24 hours. Upon arrival cucumers were individually tagged on the lightest side and stored in the dar at 2 C and 1% RH. Colour Measurements Image analysis was used for colour measurements using a JVC KY-F3 3CCD colour video camera, with the same set-up as descried in Schouten et al. (1997). Colour measurements per cucumer too place twice per wee and were expressed as the ratios of the lue/red (B/R) values from the separate intensities of the lue and red values of the RGB colour scale. After a measurement, the light intensities for the red and lue colour were separately averaged over all pixels that elong to the cucumer image. Colour measurements started one day after arrival at the measuring facility and ended when yellowing was complete or decay of the cucumer was imminent. Colour Behaviour of Individual Cucumers The colour model is ased on nowledge otained from literature regarding the processes of synthesis and degradation of chlorophyll in terms of colour compounds. POR (protochlorophyllide oxidoreductase) is a photo-enzyme which catalyses the reduction from protochlorophyllide (Pchl) to chlorophyllide (chl), the direct precursor of chlorophyll (CHL) (Leedev and Timo, 1998). A ternary complex of POR:NADPH:Pchl has een oserved, which may e assumed to e a safe form of Pchl storage as to prevent photo-toxic events when illuminated during the initial greening of young tissues (Porra, 1997). Here the assumption is made that the concentration of the complex POR:NADPH:PChl formed during pre-harvest is restrictive for the concentration of chl and CHL formed during postharvest. During senescence in fruits and vegetales chlorophyll can e cleaved y chlorophyllase resulting in the forion of chl, which will e turned into colourless compounds (Heaton and Marangoni, 1996). From a modelling point of view, only compounds with colour and their precursors are of interest. Next to CHL itself (lue green), this is chl (lue green), and the colourless precursor Pchl. Colour is defined as the sum of CHL and chl. Fig. 1 shows the proposed mechanism for these compounds during synthesis and readown of CHL. chl holds a special position as it is an intermediate in oth synthesis and readown. The initial concentration of Pchl, as part of the ternary complex POR :NADPH:PChl, is depicted as crucial and governing the colour ehaviour. The mechanism shown in Fig. 1 can e expressed in heical form y coupled differential equations, one for each process (Schouten et al., 22). These equations were solved analytically which resulted in a model formulation descriing colour ehaviour from harvest time till complete yellowing of an individual cucumer in terms of Pchl and CHL, with Pchl and CHL eing the concentration of Pch and CHL present at harvest time (Schouten et al., 22).An indication of the colour ehaviour in time decay for three hypothetical cucumers, differing only in Pchl, is depicted in Fig 2. The colour model does not contain or need cultivar factors. Keeping Quality of Batches of Cucumers Batch eeping quality was determined for the six atches of cucumers. First the colour data per cucumer, otained y following the colour development in time y 452
repeatedly measuring the same cucumers with image analysis, were analysed with the colour model to estie the values of Pchl and CHL per cucumer. Then, the time it too for each cucumer to reach the colour limit was determined using the esties of Pchl and CHL. To otain the atch eeping quality the eeping qualities of all cucumers in a atch were sorted on time. When necessary, a simple linear interpolation procedure was used to estie the time at which 5% of individuals in that atch crossed the colour limit (Schouten et al., 22). Batch eeping quality depended on season and cultivar, Borja having the highest and Volcan having the lowest eeping quality (Tale 1). As expected, the average concentration of Pchl per atch was closely related to the atch eeping quality (Tale 1). Behaviour of Pchl during pre- and postharvest for Individual Cucumers During postharvest Pchl is roen down into chl according to the mechanism shown in Fig. 1. During preharvest Pchl is thought to e stored in the ternary complex POR:NADPH:Pchl and susequently released when exposed to light (Porra, 1997). The ehaviour of Pchl may e descried as a consecutive reaction where Pchl as part of the ternary complex (PT) is transformed into Pchl, and the susequent decay of Pchl. From this proposed mechanism, the ehaviour of Pchl over time can e extracted using the fundamental rules of chemical inetics. The set of differential equations is given in Eq. 1-2. PT = PT t T Pchl = t T PT - f Pchl where T is the reaction rate for the forion of Pchl from PT. This set of differential equations can e solved analytically for constant external conditions. Given that no Pchl was present at the start of the cucumer growth, Pchl can e expressed as follows (Eq. 3). where PT is the initial concentration of PT present at the start of cucumer growth. An Pchl(t) = T t PT ( e T + T - e f f indication of the ehaviour of Pchl in time may now e simulated for different cucumers, each differing in PT and assuming a constant temperature during pre- and t ln T = f T f postharvest. Surprisingly, the imum in Pchl occurs at t independently of PT (Eq. 4). t depends only on the reaction rate constants f and T. During the analysis of the colour data of cucumers calirating the postharvest colour model it was oserved that all reaction rate constants were identical for a numer of cultivars (Schouten et al., 22). It may e assumed that, next to f, T does not show much difference etween cultivars. In that case PT is determinant for the ehaviour of Pchl in time and t will e constant for all cucumers, irrespective of cultivar, when cucumers are grown at the same temperature. As very young cucumers already show considerale amounts of greening it may e assumed that the ternary complex POR:NADPH:Pchl can e quicly transformed into t ) (1) (2) (3) (4) 453
Pchl. Therefore, when full-grown cucumers are harvested it is liely that the concentration of Pchl is already decreasing. Harvest proaly taes place somewhere in the range etween t and t=+. When the ehaviour of Pchl is oserved for t t then the ehaviour shows similarities with a decreasing logistic function varying etween and P. This logistic function can e expressed as follows (Eq. 5): Pchl(t) P 1+ e P t (5) with t t where P is defined as the concentration of Pchl at t= t. P is assumed to e cultivar specific. By using the logistic function for Pchl(t) instead of Eq. 3 a time transforion is introduced. In Eq. 3, t stands for the time from the start of cucumer growth, ut in Eq. 5 it had een replaced y t, the urity of a cucumer. t varies etween - and. At t = - the optimal urity is reached as it is equivalent to the concentration of Pchl at t (P ) and at t = the minimal urity is reached as at that point no Pchl is left. When no Pchl is left the cucumer will start to lose colour quicly (Fig. 2). Behaviour of Logistic Batches in Time As Pchl plays also a decisive role in the atch eeping quality, it is of interest to now how the ehaviour of atches in time is with respect to the Pchl concentration. To do that the concept of iological variation has to e used. This might e defined as the composite of (iologically ased) properties that differentiate individual units of a product (Tijsens and Konopaci, 21). Here, it is proposed that iological variation can e applied on the level of t. This means that a atch can e characterised y variation in t, indicated y t with standard deviation σ. The connected open symols of Fig. 3 show the distriutions of all individual values of Pchl gathered per atch. Those distriutions are shown as function of a class of Pchl and expressed as frequency, the fraction of cucumers in that class. To descrie this heically, a description has to e given of the proaility that a faction of the atch, represented y Pchl, is within a specific class (Eq. 6). Pr(Pchl (t ) (q,q 1 ]) = Pr Pchl (t Pr Pchl (t where q and q 1 are used to descrie the class orders, expressed in Pchl values. The aim is to translate atch function Pchl in terms such as t, σ and P. This translation involves a small numer of heical steps which are omitted here for clarity (Eq. 7). Pr(Pchl (t ) (q,q ] 1 = ψ Pchl Ψ stands for the cumulative distriution function which descries the variation in t. The next step is to choose which heical description should e used for Ψ. In Fig. 3 the Pchl values are expressed as faction of all cucumers that elong to a atch, i.e. values etween and 1, so for Ψ a normalised function should e chosen. Out of convenience it was assumed that Ψ was normally distriuted. The proaility that a faction of the atch, represented y Pchl, is within a specific class can now e expressed in terms of Pchl, t, σ and P, the cultivar specific parameter, when the approxied (inverse) version of Pchl (Eq. 5) is sustituted (Eq. 8). ) q 1-1 (q ) - t -1 - ψ Pchl (q ) 1 ) q t (6) (7) 454
Pr(Pchl (t ) (q,q ] 1 q P ln q = Φ P - σ t q P ln- 1 q - Φ 1 P σ t (8) with Φ the normalised cumulative normal distriution function. An indication of the atch ehaviour for one atch varying in urity is shown in Fig. 4. RESULTS AND DISCUSION Shape of Distriutions The autumn atches of cultivars Beluga and Volcan show the sewed distriutions for atches which have almost the minimum Pchl concentration. On the other hand the distriution of the spring atch of cultivar Volcan shows also a sewed distriution (ut mirrored) typically for a atch which has almost the imum concentration of Pchl. The shape of the distriutions varies etween two limits in which vicinity they are sewed. Between those limits the distriution adopts almost the shape of the normal distriution, for instance for the autumn atch of cultivar Borja. The first limit is the one at a value of for Pchl and the second limit at a imum value of Pchl. This imum value, P, may e specific for each cultivar. Estiion of P Fig. 3 shows the Pchl distriution per atch estied on colour data (connected open symols) and analysed using the logistic atch model formulation of Eq. 8 (closed symols). Distriutions of all atches were analysed in one optimalisation to otain the logistic rate over all atches, P per cultivar, t and σ per atch (Tale 1). The nonlinear regression routine of Genstat (release 3.2, Lawes Agricultural trust, Rothamsted Experimental Station, UK.) was used. Genstat has a convenient standard function uild in for Φ. For cultivar Volcan, σ varied sustantially over the growing seasons, while this was not the case for oth other cultivars (Tale 1). Only during the spring season for cultivars Volcan and Borja positive t were encountered. This means that, compared to the other atches, these atches were of much etter urity. When only σ and t are concerned, the spring atch of cultivar Volcan is the est atch. However, the influence of P, which is the largest for cultivar Borja and the smallest for cultivar Volcan (Tale 1, Fig. 3) is sustantial. When a correction for urity and σ per atch is carried out, the est cultivar with respect to atch eeping quality is Borja, followed y Beluga and finally Volcan. Biological Variation Normally, iological variation is treated lie an ever present nuisance which should e minimised as much as possile. Most commonly used technique to deal with iological variation is sorting and grading on external quality attriutes (Tijsens and Konopaci, 21). Simple colour measurements on cucumer colour are, however, not enough for eeping quality predictions as the aility to stay green cannot e assessed. For cucumers it turned out that the precursor of colour components was determining atch eeping quality. Building in iological variation on the level of t of the precursor function might e possile for a host of other products, provided that process-ased models are availale or can e generated to descrie the ehaviour of the quality attriute and the precursor. In this paper the first practical application on how to tae advantage of the iological variation is presented. 455
One prolem exists applying this iological variation incorporation technique. The function descriing the ehaviour in time of the precursor (Eq. 3) needs to e continuously increasing or decreasing, otherwise the inverse does not exist. This inverse is needed when incorporating process ased inforion into the atch model (Eq. 7). Liely, most process-ased descriptions in physiology do not comply with this condition. Therefore a way to circumvent this condition had to e made y transforming the exact Pchl function (Eq. 3) to an approxied version (Eq. 5). Literature Cited Leedev, N., Timo, 1998. Protochlorophyllide photoreduction. Photosynthesis research 58, 5-23. Lin, W.C., Jolliffe, P.A. 1995. Canopy light effects shelf-life of long English cucumer. HortScience 26, 1299-. Porra, R.J. 1997. Recent progress in porphyrin and chlorophyll iosynthesis. Photochem. photoiol. 65, 492-516. Schouten, R.E., Otma, E.C., van Kooten, O., Tijsens, L.M.M. 1997. Keeping quality of cucumer fruits predicted y the iological age. Postharvest Biol. Technol. 12, 175-181. Schouten, R.E., van Kooten, O. 1998. Keeping quality of cucumer atches: Is it predictale? Acta Hort. 476, 349-355. Schouten, R.E., L.M.M. Tijsens, L.M.M. and van Kooten, O. Predicting eeping quality of atches of cucumer fruits ased on a physiological mechanism. Postharvest Biol. Technol. (in press). Tijsens, L.M.M. and Konapaci, P. 21. Biological variance in agricultural products. Theoretical considerations. Proc. 8 th International Controlled Atmosphere Research Conference CA21, Rotterdam, The Netherlands 8-13 July (in press). Tales Tale 1. Overview of parameters elonging to the six atches. and P are common for all atches and per cultivar, respectively. cultivar season atch eeping average sigma t P quality (days) Pchl estie s.e. estie s.e. estie s.e. Volcan autumn 3.8.11 23.34 2.12 3.9 1.6 Volcan spring 6.8.59 4.959 2.31-6.87 3.86 1.752.16 Borja autumn 8.4.69 13.77 1.12 7.99 1.17.24 Borja spring 13.1 1.54 15.59 1.2-6.76 3.8 2.869.194 Beluga autumn 4.3.12 7.466 1.55 28.52.633 Beluga spring 7.2.65 6.757 1.46 5.6 2.74 2.23.24 R 2 adj (%) 93.2 n 282 456
Figurese f fw Pchl chl CHL d w colourless Fig. 1. Model representation of the last part of the chlorophyll pathway for cucumers stored in the dar. Closed arrows indicate chlorophyll synthesis and open arrows cataolism. Indicated are the reaction rate constants. Colour value (lue/red) colour limit Pchl =2 Pchl =1 Storage time (days) Fig 2. Colour ehaviour for of three cucumers differing in Pchl. Indicated is the colour limit. The cucumer with the smallest concentration of Pchl has an intercept with the colour limit at, the medium concentration at 8and the highest concentration at 2 days. This intercept is defined as the eeping quality for individual cucumers. 457
.5 cultivar Volcan frequency/proaillity.4.3.2.1 P.5 1 1.5 2 2.5 3.5 cultivar Borja Pchl frequency/proaillity.4.3.2.1.5 1 1.5 2 2.5 3 Pchl P frequency/proaillity.5.4.3.2.1 cultivar Beluga P.5 1 1.5 2 2.5 3 Pchl Fig 3. Distriution of Pchl concentrations at harvest (Pchl ) for two atches per cultivar. Per cultivar one atch was harvested in the autumn season (, ) or the spring season (, ). Open symols indicate the Pchl distriution otained from colour data, closed symols the distriution from atch model estiions. 458
t = 2 t = -2 t = 16 t = -16 t = 12 t = -12 t = 8 t = -8 t = 4 t = -4 t = Fig. 4. Simulation of Pchl distriutions for one atch harvested at different urity stages, indicated y t. Simulation was carried out using P =1, =.15 and σ=5, applying Eq. 8. 459