JXB Advance Access originally published online on April 4, 2005
Journal of Experimental Botany 2005 56(415):1389-1395; doi:10.1093/jxb/eri140
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
RESEARCH PAPER |
Growth, fructan yield, and quality of chicory (Cichorium intybus L.) as related to photosynthetic capacity, harvest time, and water regime
Department of Agroenvironmental Science and Technologies, University of Bologna, Viale G Fanin 44/46, I-40127, Bologna, Italy
* To whom correspondence should be addressed. Fax: +39 051 2096241. E-mail: amonti{at}agrsci.unibo.it
Received 6 December 2004; Accepted 16 February 2005
| Abstract |
|---|
|
|
|---|
Fructans are polymers that are widely used in several industrial applications. In the last few years they have received increasing interest because of their positive effects on health. At present, fructans are mostly supplied by chicory, which is only grown and processed in The Netherlands, France, and Belgium. It would therefore be an attractive concept to expand its cultivation to the southern European countries, although water shortage and high temperatures may hinder its growth and yield. So far, few experiments have been carried out on the effects of water, so the present research was focused on the course of growth and fructan quality in rainfed (W0) and well-watered (W1) conditions. The positive effects of water restoration mostly concerned the above-ground dry weight (ADW), whereas the root dry weight (RDW) was less influenced. No significant differences on RDW were found in 1999, whereas it was 14% higher (P <0.01) in W1 in 2000. The effect of water was very clear on assimilate allocation: the overall priority at the whole plant scale seemed to be root structures, then storage reserves, and finally ADW. Therefore, the fructan content was higher in W0, and insignificant differences between W0 and W1 were found on fructan yield at the final harvests. The only significant effect of the water regime on fructans was to speed up their storage. The leaf photosynthetic capacity (A) was poorly affected by water availability, whereas it appeared consistently modulated by leaf temperature and leaf nitrogen content. Stomatal conductance appeared to be mostly affected by the soil water content and it was mostly related to A up to about 300 mmol m2 s1. The fructan chain length (DP) was not affected by water regime. Besides, DP classes showed a normal statistical distribution; skewness and kurtosis significantly changed only when the harvest was very late. Equally, a very late harvest time significantly lowered DP.
Key words: Chicory, fructan, inulin, photosynthesis
| Introduction |
|---|
|
|
|---|
Fructans are polydisperse polysaccharides consisting of ß(2,6) fructosyl-fructose units with one glucose unit at the reducing end (Fuchs, 1990
Despite the 36 000 fructan-containing species (Pignatelli, 1998
), curently only three crops are practically processed: blue agave (for the tequila-distilling industry), Jerusalem artichoke, and chicory. The greater interest in chicory compared with Jerusalem artichoke is mainly due to its high fructan yield and quality (Meijer and Mathijssen, 1992
). Chicory has been extensively grown in Europe since the beginning of the 19th century and today is by far the most commonly used, but nowadays it is only grown in The Netherlands, Belgium, and France on approximately 15 200 ha (De Bruyn et al., 1992
). It would therefore be an attractive possibility to expand its cultivation into southern European countries. However, Jerusalem artichoke was found to be quite tolerant to the water-stress conditions (Monti et al., 2005
), so chicory might be not the most cost-effective crop in southern Europe, especially when irrigation is not feasible. Moreover, Jerusalem artichokes accumulate fructans in their stems before their translocation into tubers, thus a summer harvest of stems, when soil conditions are generally more favourable, may be proposed for Jerusalem artichoke.
To the best of the authors' knowledge, many experiments have been performed on genotype screening, effects of harvest date, nitrogen supply, and plant density on root and fructan yields of chicory (Meijer and Mathijssen, 1992
; Baert, 1997
; Amaducci and Pritoni, 1998
; Demeulemeester et al., 1998
; Desprez et al., 1999
; Monti et al., 2002
); it was, however, surprising to find that little is known about the effects of water stress. Moreover, the majority of the experiments on chicory was carried out in North-Central Europe where water availability and temperature do not generally represent serious constraints for growth. By contrast, the severe and durable water stress, that commonly occurs during the summer in southern European countries, could be strongly detrimental to chicory yield (Schittenhelm, 1999
; Danuso, 2001
). In very interesting research on fructan-containing crops, Schittenhelm (1999)
found negative effects of water deficit, particularly on above-ground dry matter (leaves without crown). These results were confirmed by Skinner et al. (2002)
, although the authors never explored the effects of water deficit on fructan accumulation and degree of polymerization. The effect of high temperatures, which are known to inhibit photosynthesis and yield, was not examined (Salvucci and Crafts-Brandner, 2004
; Wise et al., 2004
).
This research was therefore focused on the effects of water stress on fructan yield and their degree of polymerization (DP, i.e. the fructan chain length), that is a basic parameter for the industrial applications. Parallel to DP, the pure index (PI) and free sugar index (FSI) of fructan were also determined in order to assess the course of depolymerization over the growing season. The depolymerization process is modulated by two enzymes: fructan-exo-hydrolase (FEH) and fructosyl-fructosyl transferase (FFT) (Van den Ende and Van Laere, 2002
) that remove terminal fructose residues from inulin chains resulting in shorter fructan chains. The activity of the two enzymes is mostly modulated by several factors such as temperature and sucrose availability (Van den Ende and Van Laere, 2002
). In particular, FEH was mostly found to be active after leaf senescence (Khuri and John, 2000
) which may be linked with sucrose availability. It is well known that water stress reduces photosynthetic rate and thus, possibly, the sucrose loading to the taproot; similarly, harvest time is related to leaf senescence.
The fructan average DP of a single chicory root (commonly 1020) is the average of short (fewer than 10 units) and long fructan chains (about 60, Roberfroid and Delzenne, 1998
). Therefore, the statistical distribution of DP classes is necessary to have a reliable DP profile into a root or sample. It is shown here that the shape of the statistical distribution of the DP classes depends on the harvest time and not the water regime.
| Materials and methods |
|---|
|
|
|---|
Experimental site and treatments
The experiments were carried out in a flat soil classified as Haplic Calcisol (FAO), in Bologna, Po Valley (Lat. 44° 03', Long. 11° 02', 33 masl) in 1999 and 2000. Rainfed (W0) and well-watered (W1) plants were compared according to a randomized block design with four replications. Sowing was carried out with a pneumatic seed drill placing seeds at 0.5 cm depth, with a row spacing of 45 cm. Thinning was carried out to a final plant population of 15 plants m2. A drip irrigation system was placed on every other row. Water restorations occurred whenever evapotranspiration (Et) reached 30 mm. Et was calculated daily by multiplying evaporation (class-A evaporation pan) by the specific crop coefficients (Allen et al., 1998
|
Radiation use efficiency (RUE) was determined as the slope of the regression of the accumulated dry matter and accumulated absorbed photosynthetic active radiation (PARa; Monteith, 1977
Leaf traits and gas-exchange measurements
Eight leaf gas-exchange measurements were performed from May to August of the second year using a portable infrared gas analyser (IRGA, CIRAS-1, PP Systems, Hertfordshire, UK). Leaf photosynthetic capacity (A), stomatal conductance (gs) and intercellular CO2 (Ci) were calculated as given by von Caemmerer and Farquhar (1981)
. The leaf gas exchange data referred to the most recently fully expanded leaves up to 11.30 h in order to escape the midday depression on A. During the measurements, the average PAR was 1600±48 µmol (quanta) m2 s1; the CO2 concentration in the air entering the chamber was 350±11 µmol mol1 (of air); the entering air flux was 250±4 cm3 min1. Some functional leaf traits were also measured on the same leaves chosen for gas-exchange measurements in order to investigate their influence upon A: nitrogen content (Nleaf, Kjeldahl, 1883
); the relative water content (RWC), i.e. the ratio of water contents in fresh to turgid leaves; the specific leaf area (SLA), i.e. the ratio of leaf area to leaf dry mass; the leaf dry matter content (LDMC).
Fructan analysis
Root fructan content was measured at each harvest time in 2000, but in 1999 it was only measured in the two final harvests. Root samples (about 500 g) were frozen (18 °C) and milled in dry ice. Fructans were extracted in hot water (80 °C) for 60 min. Fructan content (FC) was determined as given by Baert (1997)
: FC=(F+G)(f+g+s); where F and G, are the total fructose and glucose after acid hydrolysis (HCl) and f, g, and s are the reducing free sugars fructose, glucose, and sucrose before the acid hydrolysis. The amount of reducing sugars was determined by HPLC (high-performance-liquid-chromatography). The column was Rezex 8% Ca (30x0.78 cm); column temperature was 75 °C. Water with a flow rate of 0.6 ml min1 was used as the mobile phase. An analytical differential refractive index (RI) detector was used. The average fructan chain-length (DP), the pure fructan index (PI) and the free sugars index (FSI) were calculated as follow (Baert, 1997
): DP=F/G; PI=[(F+G)/RI]x100; FSI=[(f+g+s)/(F+G)]x100, where RI (Brix %) is the refractive index of solids measured on taproot sap (Palm Abbe 200, Misco, Cleveland, OH). To determine the DP classes, the samples were extracted in a hot (80 °C) waterethanol mix (33% v/v); afterwards a 40 mg subsample of the extract was diluted in 1 ml of distilled water and than analysed using a size-exclusion-chromatography. The DP classes were based on exclusion 30 s time-intervals.
Statistical analysis
All the measured and derived data were subjected to the analysis of variance (ANOVA) carried out with the Systat package (Systat Software Inc.). Bartlett's and KolmogorovSmirnov's tests were used, respectively, to verify the homogeneity of variance and the normal distribution of data. The relevance of leaf traits upon photosynthesis was tested by a sensitivity analysis using the adjusted multiple regression coefficient as the discriminating parameter. The variance inflation factor (VIF) was used to measure the multicollinearity, that is the inflation of a regression parameter for an independent variable due to the redundant information in other independent variables. Briefly, the higher the VIF, the more redundant are the variables in the regression model.
The statistical distribution of fructan DP was assessed using the KolmogorovSmirnov's test. Pearson's skewness (SK), i.e. the ratio of the difference between mean and mode to standard deviation, was considered as an indicator of the asymmetric distribution: if the ratio of SK to its standard error was greater than 2 the distribution was considered asymmetric. A positive SK indicates a left long tail. Kurtosis (K) was taken as a measure of the flat distribution: negative K indicates a flatter shape. Again, K was considered significant when the ratio of K to its standard error was higher than 2.
| Results |
|---|
|
|
|---|
Weather data
The seasonal rainfall distribution varied between the two years. It was higher from April to May in 1999 (+93 l m2) and from June to August in 2000 (+88 l m2). Air temperatures was generally higher (on average by 2 °C) in 2000, especially in the early part of the cycle. Water deficit, i.e. the difference between Et and rainfall, during the treatment time (10682858 GDD and 8822384 GDD, in 1999 and 2000, respectively) were similar in the two years: 374 l m2 (1999) and 358 l m2 (2000). Water table depth was also similar in the two years and was not affected by irrigation (from 1.4 to 2.3 m depth from the start to the end of treatment). Conversely, during the treatment the average soil water content of W0 was significantly lower than that of W1 (4.7% and 4.0% (v/v), respectively, in 1999 and 2000).
Dry matter accumulation and photosynthesis
The consistently higher biomass yield of 1999 (Fig. 1) was probably explained by the early rainfall which speeded up the growth of the young roots and leaves with a consequently higher intercepted solar radiation and water supply; water table being more superficial in the early growing cycle of 1999. The effects of water regime was clearly more visible on above-ground dry weight (ADW) and leaf area index (LAI) than on root dry weight (RDW). ADW was, on average, 50% higher in W1, whereas RDW was only 5% higher (Fig. 1). The early leaf area development allowed W1 to absorb only 5% more incident photosynthetic active radiation (PAR) than W0 (75% versus 80%). However, the radiation use efficiency (RUE), which ranged from 2.6 g MJ1 to 1.9 g MJ1, was always clearly higher in W1 (Fig. 2).
|
|
The leaf net photosynthesis (A) was poorly influenced by the water regime. Leaf photosynthesis appeared significantly correlated to nearly all the leaf traits (Table 2), thus a sensitivity analysis was performed to find the traits that had the most influence upon A. Tleaf appeared the most relevant factor, explaining 70% of the total A variation. Nleaf added 8% to the predicting capacity, whereas all the other variables considered had little influence upon A (Table 2). Tleaf showed a close linear relation with A within the explored temperatures (from 28 to 38 °C) (Fig. 3). Conversely, it was unexpected to find gs weakly correlated to A. However, the Pearson's correlation coefficient measures the linear predictability, while a clear non-linear association between A and gs was observed. Specifically, a strong correlation was found between A and gs up to 300 mmol m2 s1, whereas the correlation was much weaker above this threshold (Fig. 3). Finally, gs appeared not to be significantly related to the other leaf traits, whereas it was affected by soil drying (r=0.69**). Thus, soil drying somewhat modulated A as well (r=0.47*).
|
|
Fructan accumulation
The significant differences in fructan contents (FC) and fructan yields (FY) between W0 and W1 were limited to the three first irrigations, W1 showing a faster fructan accumulation than W0 (Fig. 4). At about 2000 thermal units, FC was 42% higher in W1, while FY was still higher (78%) due to the higher RDW of W1. However, a few days later (2500 GDD) W0 and W1 showed similar FY values, and at the last harvest no significant differences were found between treatments or years.
|
Water regime did not significantly modify the maximum DP that was 13.7 in W0 and 14.2 in W1. Since DP was strictly correlated with FC (r=0.92**), the maximum DP occurred 60 d earlier in W1.
Since DP merely represents the average value of fructan chain length, the statistical distribution of DP classes around the mean was also investigated. DP strongly varied from 2 to more than 100, showing a natural trend platykurtic distribution of the DP classes (negative kurtosis). Generally, the statistical distribution of DP classes did not deviate from the Gaussian curve (Fig. 5). Water regime did not affect the Pearson's skewness (SK) and K. Conversely, harvest time appeared to have a strong influence upon the fructan chain length and the statistical distribution of DP (Fig. 5). Specifically, during the final harvest of 1999, the average DP decreased from 12 to less than 4, while the DP distribution showed a significant low K (1.94), i.e. a long right tail. This only included 36% of the total DP values in the two central classes, compared with 4749% at the other harvest times in the same or in the following year.
|
The reducing free sugars were measured at each sampling date and represented as pure (PI) and free sugar indices (FSI). FSI and PI indices appeared neither correlated to each other nor changing over time and treatments. FSI ranged from 4% to 7%; PI from 85% to 89%. Similar values were found also by Baert (1997)
| Discussion |
|---|
|
|
|---|
According to Hsiao (1973)
Radiation use efficiency was higher in W1. However, despite the higher biomass yield of W1 the intercepted PAR of W1 was only 5% higher than W0 (+32 and +57 MJ m2 in 1999 and 2000, respectively). Therefore, as given by RUE, W1 was expected to accumulate 83 g m2 (1999) and 148 g m2 (2000) more dry biomass than W0, i.e. 20% and 50% less than that measured on total accumulated biomass. A higher photosynthetic capacity of W1 was therefore suspected, but this was only weakly demonstrated by the leaf gas exchange measurements.
The effect of water regime was more evident on marketable products (fructans) than on total dry matter. In fact, when RUE was weighted on fructans instead of the total dry matter, it appears very similar in W0 and W1 (1.0 and 1.1 g (fructan) MJ1). Similarly, the net assimilation rate (NAR) weighted on FY (i.e. by multiplying NAR per FC) appeared frequently higher in W0 (Fig. 6). Therefore, all things being equal, the crop effectiveness on fructan accumulation was not reduced by the water deficit.
|
Among the leaf traits, Tleaf showed the highest influence upon A. For every additional degree of the explored Tleaf the photosynthetic capacity was expected to be reduced by 1.4 µmol m2 s1, thus 56% less photosynthetic capacity was observed from 2838 °C. The negative effects of high Tleaf were found both in the early and late parts of the growing cycle, therefore the suspected combined effects of Tleaf and age were not confirmed. Conversely, a lower Nleaf was generally found in old leaves that showed a low photosynthetic capacity as well. Thus a linear regression model could be considered between A (dependent variable) and Tleaf (driving variable), whereas Nleaf seemed simply to run in parallel to A. However, although the effects of Nleaf on the carboxylation activity of Rubisco is largely known (Sage and Pearcy, 1987
Generally, the DP did not appear affected statistically by the water regime and the harvest time. The only exception concerned the final harvest of 1999 that significantly lowered the DP. This was not found in the final harvest of 2000 which, however, was performed one month earlier than in 1999. Since the depolymerization process is modulated by the enzymes FEH and FFT, that are mostly active when low temperatures occur and no more sucrose is provided (Van den Ende and Van Laere, 2002
; Denoroy, 1996
), it was not surprising to find a strong reduction of DP in 1999 since this harvest occurred when night temperatures were low and many days after total leaf senescence. This study's results were corroborated by those of Wilson et al. (2004)
that found the DP>20 category to decline from 23% to 13% a few days after the first frost.
To sum up, despite the fact that water availability consistently increased biomass accumulation, water restoration seems not to be worthwhile in this area. RUE and NAR weighed on fructan clearly showed the reserves to override the above-ground structural parts. Thus the dry biomass accumulation did not run in parallel to the fructan storage, that was not significantly lower in rainfed conditions. The only significant effect of water regime was to speed up the fructan accumulation, probably related to the sink-to-source ratio. The harvest time was found to be strongly detrimental of the quality, both on fructan chain length and normal distribution of chain classes. Among the leaf traits, temperature appeared to be the most influencing parameter upon photosynthetic capacity. Therefore, since durable high temperatures generally occur in this region during the growing cycle, they may represent a basic constraint for chicory yield, independently from the water regime.
| Acknowledgements |
|---|
This work was supported by the MURST Project Crops for Inulin Production: models of environmental crop response and cultural strategies. The assistance of Dr Roberta Piccaglia during fructan analysis is gratefully acknowledged.
| Footnotes |
|---|
Abbreviations: A, leaf net photosynthesis; ADW, above-ground dry weight; DP, degree of polymerization; Et, evapotranspiration; FC, fructan content; FSI, free sugars index (free sugar content over total carbohydrate content); FY, fructan yield; gs, stomatal conductance; GDD, growing degree days; K, kurtosis; LAI, leaf area index; LDMC, leaf dry matter content; NAR, net assimilation rate; Nleaf, leaf nitrogen content; PI, pure fructan index (total carbohydrate content expressed over the refractometer index); RDW, root dry weight; RI, refractometer index; RUE, radiation use efficiency; RWC, relative water content; SLA, specific leaf area; SK, Pearson's skewness; Tleaf, leaf temperature; TDR, time domain reflectometry; VIF, variance inflation factor; W0, rain-fed plants; W1, irrigated plants.
| References |
|---|
|
|
|---|
Allen RG, Pereira LS, Raes D, Smith M. 1998. Crop evapotranspiration. FAO irrigation and drainage paper 56. Rome: FAO, 103125.
Amaducci S, Pritoni G. 1998. Effect of harvest date and cultivar on Cichorium intybus yield components in north Italy. Industrial Crops and Products 7, 345349.[CrossRef]
Baert JRA. 1997. The effect of sowing and harvest date and cultivar on inulin yield and composition of chicory roots. Industrial Crops and Products 6, 195199.[CrossRef]
Danuso F. 2001. Crops for inulin production: state of the art and perspectives. Rivista di Agronomia 35, 176187.
De Leenheer L. 1994. Production and use of inulin: industrial reality with a promising future. In: Van Bekkum H, Röper H, Voragen F, eds. Carbohydrates as organic raw materials, III. Proceedings of the 3rd International Workshop on Carbohydrates as Organic Materials, Wageningen.
De Bruyn A, Alvarez AP, Sandra P, De Leenheer L. 1992. Isolation and identification of ß-D-fructofuranosyl-(2,1)-D-fructose, a product of enzymatic hydrolysis of the inulin from Cichorium intybus. Carbohydrate Research 235, 303308.[CrossRef][Web of Science][Medline]
Demeulemeester MAC, Verdoodt V, de Profit MP. 1998. Interaction between physiological age and cold treatment on the composition and concentration of carbohydrates in chicory roots (Cichorium intybus). Journal of Plant Physiology 34, 437445.
Denoroy P. 1996. The crop physiology of Helianthus tuberosus L.: a model orientated view. Biomass and Bioenergy 11, 1132.[CrossRef]
Desprez BF, Delesalle L, Dhellemmes C, Desprez MF, Rambaud C, Vasseur J. 1999. Genetics and breeding of industrial chicory, a historical review. In: Proceedings of the 8th seminar on inulin, Lille, France, 110.
Farquhar GD, Sharkey TD. 1982. Stomatal conductance and photosynthesis. Annual Review of Plant Physiology 33, 317345.[Web of Science]
Fuchs A. 1990. Perspectives of inulin and inulin-containing crops in the Netherlands in the early growth and in Europe. In: Fuchs A, ed. Third seminar on inulin. The Hague: National Council for Agricultural Research, 80102.
Hsiao TC. 1973. Plant responses to water stress. Annual Review of Plant Physiology 24, 519570.[Web of Science]
Khuri S, John P. 2000. Fructan exo-hydrolase activity in harvested chicory roots. In: Proceedings of the fourth international fructan conference, Basel, Switzerland, 4547.
Kjeldahl JZ. 1883. Neue Methode zur Bestimmung des Stickstoffs in organischen Körpern. Zeitschrift für Analytische Chemie 22, 366382.[CrossRef]
Meijer WJM, Mathijssen EWJM. 1992. Crop characteristics and inulin production in chicory. European Journal of Agronomy 2, 99108.
Monsi M, Saeki T. 1953. Uber den Lichtfaktor in den Pflanzengesellschaften und seine Bedeutung für die Stoffproduktion. Japanese Journal of Botany 14, 2252.
Monteith JL. 1977. Climate and the efficiency of crop production. Philosophical Transactions of the Royal Society of London Series B 281, 277294.
Monteith JL, Unsworth MH. 1990. Principles of environmental physics, 2nd edn. Edward Arnold.
Monti A, Amaducci MT, Venturi G. 2005. Growth response, leaf gas exchange and fructans accumulation of Jerusalem artichoke (Helianthus tuberosus, L.) as affected by different water regimes. European Journal of Agronomy (available on line).
Monti A, Pritoni G, Amaducci MT. 2002. Evaluation of chicory varieties in North Italy: inulin production and photosynthesis. In: Proceedings of the ninth seminar on inulin, 1819 April 2002, Budapest, Hungary, 23.
Pignatelli V. 1998. In: Amaducci MT, Marzi V, Venturi G: Le Colture di Nicchia. Edagricole (ed.), 8189.
Roberfroid MB, Delzenne NM. 1998. Dietary fructans. Annual Review of Nutrition 18, 117143.[CrossRef][Web of Science][Medline]
Sage RF, Pearcy RW. 1987. The nitrogen use efficiency of C3 and C4 plants. Plant Physiology 84, 959963.
Salvucci ME, Crafts-Brandner SJ. 2004. Mechanism for deactivation of Rubisco under moderate stress. Physiologia Plantarum 122, 513519.[CrossRef]
Schittenhelm S. 1999. Agronomic performance of root chicory, Jerusalem artichoke and sugarbeet in stress and non-stress environment. Crop Science 39, 18151823.
Skinner R, Hall M, Sanderson M. 2002. Water stress on Puna chicory and Lancelot Plantain. Morphological and physiological effects. Agricultural Research Service, USDA.
Tei F, Scaife A, Aikman DP. 1996. Growth of lettuce, onion, and red beet. 1. Growth analysis, light interception and radiation use efficiency. Annals of Botany 78, 633643.
Van den Ende W, Van Laere A. 2002. Induction of 1-FEH in mature chicory roots appears to be related to low temperatures rather than to leaf damage. Scientific World Journal 2, 17501761.
von Caemmerer S, Farquhar GD. 1981. Some relationship between the biochemistry of photosynthesis and the gas exchange of leaves. Planta 153, 376387.[CrossRef][Web of Science]
Wilson RG, Smith JA, Yonts CD. 2004. Chicory root yield and carbohydrate composition is influenced by cultivar selection, planting and harvest date. Crop Science 44, 748752.
Wise RR, Olson AJ, Schrader Sharkey TD. 2004. Electron transport is the functional limitation of the photosynthesis in field-grown Pima cotton plants at high temperature. Plant, Cell and Environment 27, 717724.[CrossRef]
![]()
CiteULike
Connotea
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||

0.05; * and ** mean significant differences between treatments for P 



