JXB Advance Access originally published online on June 4, 2004
Journal of Experimental Botany, Vol. 55, No. 402, pp. 1557-1567, July 2004
Journal of Experimental Botany, Vol. 55, No. 402, © Society for Experimental Biology 2004; all rights reserved
RESEARCH PAPER |
Maintaining exponential growth, solution conductivity, and solution pH in low-ionic-strength solution culture using a computer-controlled nutrient delivery system
Department of Biological Sciences, University of Alberta, Edmonton, Alberta T6G 2E9, Canada
To whom correspondence should be addressed. Fax: +1 780 492 9234. E-mail: gregory.taylor{at}ualberta.ca
Received 14 November 2003; Accepted 14 April 2004
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResults and discussionReferences |
|---|
Studies of plant nutrient requirements in solution culture have often used nutrient concentrations many-fold higher than levels found in fertile soils, creating an artificial rooting environment that can alter patterns of nutrient acquisition. The relative addition rate (RAR) technique addresses this problem by providing nutrients in exponentially increasing quantities to plant roots in solution culture. A computer-controlled RAR nutrient delivery system has been developed to reduce workload and to facilitate more frequent nutrient additions (4x daily) than is possible with manual additions. In initial experiments, a minimum background solution containing 500 µM nitrogen and all other essential nutrients in optimal proportions was required for the healthy growth of Triticum aestivum. This requirement was reduced to 50 µM nitrogen when calcium in the background solutions was increased to 400 µM. Varying the abundance of ammonium and nitrate in both background and delivery solutions provided a means of controlling plant-induced pH changes in growth solutions. In optimized solutions, plant relative growth rates (RGR) in the order of 0.2 g g–1 plant d–1 were maintained over a 22 d experimental period. Variation in RARs provided a means of growing plants with varying RGRs under relatively constant conditions of solution electrical conductivity and pH.
Key words: Ammonium, nitrate, nitrogen, relative addition rate, RAR, Triticum aestivum
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Traditional solution culture methods often bear little resemblance to the conditions observed in natural ecosystems (Ingestad, 1982
). To begin with, plants deplete nutrients from solution until supply is exhausted and deficiency sets in. Thus, the experimental conditions change over time. In addition, studies have frequently used nutrient concentrations many-fold higher than levels found in fertile soils. Use of such high nutrient concentrations creates an artificial rooting environment that can potentially alter patterns of nutrient acquisition and uptake (Ingestad and Lund, 1979; Asher and Blamey, 1987). While efforts to mimic the nutrient availability of natural environments are desirable, the use of low nutrient concentrations in solution culture introduces technical difficulties. The most challenging of these difficulties has been the development of techniques that provide plants with access to exponentially increasing amounts of essential nutrients, while maintaining the low nutrient concentrations and solution conductivities that are typically observed in soil solutions. An additional challenge includes the maintenance of steady-state conditions (nutrient supply, pH, conductivity) in poorly buffered growth systems that lack a solid phase.
A number of techniques have been developed to address these challenges. The flowing solution culture technique (Asher et al., 1965) and the relative addition rate (RAR) technique (Ingestad, 1981, 1982) have received the most attention to date. The flowing solution culture technique provides plants with nutrient concentrations that resemble soil solution concentrations and facilitates the maintenance of a constant test ion concentration (with frequent analysis and additions; Asher et al., 1965). Several problems have been encountered using this technique, including its labour-intensive nature and the requirement for high volumes of solutions. In addition, high solution flow rates are required to meet the demands of exponential growth (Ingestad, 1982) and prevent depletion of all essential nutrients including the test nutrient(s) (Asher et al., 1965; Asher, 1981).
Exponential growth can also be maintained by frequent nutrient additions to experimental solutions (the RAR technique (Ingestad, 1981, 1982) or programmed nutrient additions (Asher and Blamey, 1987)). Since solutions are not changed during the experimental period, this technique requires lower volumes of water and fewer nutrients than flowing solution culture systems. If the exponential supply of nutrients (RAR) is less than that required to maintain the maximum relative growth rate (RGRmax), plant RGR will decline to approximately equal the RAR (Ingestad and Lund, 1986). Thus, this technique provides a means of growing nutrient-sufficient or nutrient-deficient plants with steady internal nutrient status.
Historically, the RAR technique has involved the use of aeroponic delivery systems. The RAR technique was recently adapted for use in solution culture systems where plant roots are continuously immersed in experimental solutions (Stadt et al., 1992). The modified technique shares many of the advantages available when using aeroponic systems, including the ability to support plant growth under conditions of low ionic strength and electrical conductivity (EC <150 µS cm–1). It is also possible to reduce pH fluctuations in growth solutions (without the addition of buffers) by balancing cation/anion uptake. Cation and anion uptake are normally dominated by the need to acquire nitrogen (N). By varying the relative abundance of ammonium (
) and nitrate (
) in growth solutions, cation and anion consumption can be balanced without changing the overall nutrient proportions (Ingestad and Lund, 1986; Stadt et al., 1992).
It has previously been shown that the modified RAR technique can be successfully used with manual nutrient additions (Stadt et al., 1992; Taylor et al., 1998). In this work, a computer-controlled nutrient delivery system has been developed to reduce workload and to facilitate more frequent nutrient additions (4x daily) than is possible with manual additions. It is shown how the manipulation of background nutrient supply and cation/anion balance can be used to support high rates of growth under relatively constant conditions of solution electrical conductivity (EC) and pH.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Theory for exponential growth
A nutrient that remains at a stable concentration in exponentially growing plants will increase as:
 | (1) |
). The increasing nutrient demand of exponentially growing plants can be met by supplying plants at a constant relative addition rate (RAR; g nutrient (g plant nutrient)–1 d–1; Ingestad and Lund, 1986). The amount of nutrient required to support growth for a given time interval (At) is determined by:
 | (2) |
 | (3) |
). Nutrient additions can be calculated for N alone and the remainder of the nutrients supplied in proportion to N (Ingestad, 1981; Stadt et al., 1992). Note that units for RAR are correctly described as g nutrient (g plant nutrient)–1 d–1, and units for RGR are correctly described as g g–1 plant d–1. For simplicity, subsequent citation of RARs and RGRs will use the numerically identical d–1 units.
Computer-controlled nutrient delivery system
A computer-controlled nutrient-delivery system was designed and built to provide frequent (1–4 times d–1) and accurate supply of nutrients to experimental containers
. The hardware portion of the system consists of an hydraulic pathway and an electronic and analogue signalling system (Fig. 1). The hydraulic pathway includes two Watson Marlow multi-channel peristaltic pumps (model 202S), each equipped with 15 drive cassettes. Each of the 30 drive cassettes control the flow of one solution to a series of six out of a total of 180 electric, three-way pinch valves (grouped into three sets of 60). Ten such series (60 valves in total) control the flow of one nutrient solution to the 60 experimental containers. Figure 1 is a diagram that illustrates the flow of nutrient solution from one of three nutrient reservoirs, through the first drive cassette on each pump, to two (out of 30) sets of six valves, then on to experimental containers. When open (shown in bold arrowheads on Fig. 1), one valve controls the flow of one solution to one growth container. When all six valves in a series of valves are closed (solid line) the nutrient solution circulates back to the nutrient reservoir. The system allows independent delivery of three separate solutions to each growth container.
|
The RAR software controls the opening and closing of electric valves through the electronic/analogue signalling system. The program calculates nutrient additions (At as in equation 3) based on user-supplied information specific for each experiment (including W0, plant nutrient content, RGR, concentration of nutrients in delivery solutions, and the flow rate of each pump cassette) and calculates open and closing times for each of the 180 values. When valves are required to open, the computer sends digital signals to a series of 180 relay switches, which convert the digital signals to analogue signals. The relay switches then energize specific valves. When valves are energized, they open, allowing nutrient solutions to circulate to growth containers. Nutrient additions are made once daily, unless valve open times exceed 20 min, at which time additions are split into four aliquots over 24 h.
Experimental design
In all the experiments reported here, two basic experimental designs were used: (i) time-course analysis and (ii) factorial treatments. In all but the final experiment, nutrients were added at a RAR of 0.20 d–1. Previous experiments indicated that this RAR was slightly lower than RGRmax under these experimental conditions. Thus, plant growth should be limited by the RAR. All experiments utilized a randomized block design with three or four statistically independent replicates, for a total of 60 containers. Each experiment was performed at least twice. Time-course experiments were conducted for 21–30 d. To determine RGRs accurately, plants were harvested every second day during the experimental period and dried to a constant weight. Dry weights (g pot–1), were converted to natural logarithms and plotted versus time. The slope of the best-fit regression line represents the average RGR for the plotted points.
General growth techniques
Nutrient solutions: The proportions of nutrients provided in delivery and growth solutions were based on solutions used by Ingestad and Stoy (1982), Pettersson and Strid (1989), and Stadt et al. (1992). In the text that follows, nutrient levels in solution are expressed as a concentration of N. These solutions contain that concentration of N, plus additional nutrients in the mole proportions specified in Table 1. The composition of three nutrient-delivery solutions (delivered by the computer-controlled, nutrient delivery system) varied with the experimental design and were used to make up all pretreatment solutions and experimental solutions. Delivery solution 1 contained NH4NO3, Mg(NO3)2.6H2O, Ca(NO3)2.4H2O, and 0.0025 g l–1 Vitavax (to limit fungal growth). Delivery solution 2 contained KH2PO4, KNO3, K2SO4, MnSO4.H2O, H3BO3, ZnSO4.7H2O, CuSO4.5H2O, and Na2MoO4.2H2O. Delivery solution 3 contained FeCl3.6H2O. In experiments where the relative supply of and was varied, NH4Cl, MgCl2 (delivery solution 1), and HNO3 (delivery solution 2) provided alternative salts for maintaining nutrient ratios.
|
Preparation of plant material (the pretreatment period): Seeds of Triticum aestivum L. cv. Katepwa were surface-sterilized in a 1.1% solution of sodium hypochlorite (v/v) for 20 min and germinated overnight in an aerated solution containing 0.005 g l–1 Vitavax (Uniroyal Chemical Ltd., Calgary, AB, Canada) to limit fungal growth. Seeds were then grown in aquaria (20 l Plexiglas containers; 300 seeds each) on nylon mesh suspended over an aerated solution containing background nutrients (50, 200, or 500 µM N, depending on experimental design) adjusted to pH 4.3 with 1.0 or 0.1 M HCl. Seedlings were thinned after 3 d to 150 plants per aquarium. At this time, 12 seedlings were dried for 2 h (to constant weight) at 60 °C to determine initial dry weight (W0) for calculation of nutrient additions for the remainder of the pretreatment period. Nutrient additions for each day (At) were calculated using equation 3. Calculated nutrient additions for the pretreatment period were reduced by half in the experiments reported later in Figs 6–10
, as seed reserves were providing seedlings with sufficient nutrients to accelerate the RGR beyond desired levels. Beginning 
3–4 d after germination (as soon as seedlings were large enough to separate from the seed), subsamples (n=8) of plants were collected each day, dried, and weighed (without the seed) to determine plant growth during the pretreatment period. The dry weight of plants for the final day of the pretreatment period was used to estimate W0 for the calculation of nutrient additions for the experimental period.
|
|
|
|
|
Experimental period: After 9 d, spent seeds were removed from all plants and eight uniform seedlings were transferred to each of 60 polyethylene containers filled with 10 l of aerated background solution. Solutions were adjusted to pH 4.3. These background solutions were prepared by computer-controlled delivery of an appropriate volume of nutrient-delivery solutions to 10 l of distilled water to achieve the desired background concentration (50–1000 µM N). Thereafter, nutrients were supplied in exponentially increasing quantities, 1–4 times d–1. Seedlings were mounted on opaque Plexiglas covers, which were placed over the containers to limit algal growth. Distilled water was added periodically to the nutrient solutions to compensate for water loss by evaporation and transpiration. Containers were suspended in a common water bath to limit temperature fluctuations and to maintain a constant temperature across all experimental solutions.
Solution pH and EC were monitored periodically with a Radiometer PHM80 pH meter and a Radiometer CDM80 electrical conductivity meter with a CDC104 probe. Measurements were taken prior to planting, three times per week (just before nutrient additions), and immediately after harvest.
Experiments were conducted in two controlled-environment growth chambers, with 16 h light and 8 h darkness. Temperatures for the light period ranged from 20–24 °C and from 16.7–19.5 °C during darkness. Relative humidity varied between 50% and 84% during the light period and 75% and 100% during the dark period. Solution temperatures varied between 19 °C and 23 °C during the light period and 18 °C and 21 °C during darkness. The growth chamber was illuminated by 103 cool white fluorescent lamps (25 W), and 16 incandescent lamps (150 W), located 1.3 m above the plant bases. The average photosynthetic photon flux density (PPFD) for single experiments conducted in both chambers ranged from 332 to 471 µmol m–2 s–1. Plants were harvested at the end of the experimental period, rinsed in distilled water, separated into roots and shoots, and dried to a constant weight at 60 °C.
Growth techniques by experiment
Time-course of growth (200 µM N background): This 30 d time-course experiment utilized a 0.20 d–1 RAR and a 200 µM N background with all essential nutrients supplied in the proportions outlined in Table 1 (initial solution, 25% ). These nutrient proportions were similar to those used by Ingestad and Stoy (1982), Pettersson and Strid (1989), and Stadt et al. (1992).
Background nutrient concentrations: This experiment was designed to determine if plant growth rate would be affected by the initial background concentration of nutrients. The experiment utilized a 200 µM N background for the pretreatment period and 15 background concentrations (50, 100, 150, 200, 250, 300, 350, 400, 450, 500, 600, 700, 800, 900, and 1000 µM N) for the 21 d experimental period. All nutrients were supplied in the same weight proportions as in the previous experiment (Table 1) at an RAR of 0.20 d–1.
Time-course of growth (500 µM N background): This experiment was used to determine if a 0.20 d–1 RGR could be maintained with a 500 µM N background and a 0.20 d–1 RAR over a 22 d experimental period. No changes were made to the nutrient proportions.
Calcium background: This experiment was designed to determine if increasing the calcium concentration in background solutions would allow the background levels of other nutrients to be reduced (i.e. using a N background of 200 instead of 500 µM). Calcium was added from a 1.0 M CaCl2 stock to a 200 µM N background to bring the final Ca concentration to 14, 95, 195, 400, 1000, and 2000 µM. Plants were grown for 9 d at an RAR of 0.20 d–1.
Background nutrient concentrations (400 µM total calcium): This 21 d experiment was designed to determine if a total concentration of 400 µM Ca in the pretreatment (200 µM N) and experimental background solutions (50–1000 µM N) would allow the levels of other background nutrients to be reduced. With the exception of Ca and Cl, nutrient proportions were the same as in previous experiments.
Nitrogen source: Variation in the relative abundance of and in background and delivery solutions provides a means of manipulating cation/anion balance and plant-induced pH changes in growth solutions. This experiment was designed to determine what /N ratio would produce the smallest plant-induced pH change without causing reductions in plant growth arising from toxicity. A 0.20 d–1 RAR, and a 50 µM N background with a total concentration of 400 µM Ca was used for the pretreatment and experimental periods. Ten / treatments (5, 10, 15, 20, 25, 30, 35, 40, 45, and 50% as a percentage of total nitrogen) plus five sodium chloride (NaCl) control treatments (delivery solution concentrations of 0.7, 1.4, 2.1, 2.8, and 3.5 M NaCl superimposed over a 25% /N solution) were used for this 21 d experiment. The NaCl treatments were included to control for the expected changes in the Na and Cl counter ions and to confirm that any observed growth reductions could not be attributed to increased concentrations of these ions in growth solutions. This was necessary as variations in NH4Cl, NaNO3, and MgCl2 were used in the treatment solutions to achieve appropriate /N ratios. Other stock solutions were not altered.
For all nitrogen source (/N ratio) experiments, delivery solution 1 was replaced with separate nutrient solutions for each N or NaCl treatment. Daily additions of the N and NaCl solutions were fed by hand as the nutrient delivery system is only capable of delivering three different nutrient solutions.
The effects of nitrogen source on an 
resistant cultivar (Atlas-66): Control of plant-induced pH changes in growth solutions could not be fully achieved without leading to toxicity in cv. Katepwa. Therefore, a second nitrogen source experiment was conducted with the more -resistant cultivar Atlas-66. Preliminary experiments (data not shown) indicated that the response of cv. Atlas-66 to changes in background nutrient concentrations was similar to that observed in cv. Katepwa; thus, the experimental conditions used in this experiment were the same as those in the previous nitrogen source experiment (5, 10, 15, 20, 25, 30, 35, 40, 45, and 50% as a percentage of total nitrogen). Salt controls were not included in this experiment.
Time-course of growth (Atlas-66): Based upon the results of previous experiments, this time-course experiment (22 d) was conducted with a 50 µM N background (plus 400 µM CaCl2). This experiment used 36% /N with no other changes to the nutrient proportions (Table 1; optimized solution, 36% ).
Time-course for various RARs: This 24 d experiment was used to determine if altering the RAR of nutrients would provide control of the RGR and affect plant-induced changes in solution EC and pH. A 50 µM N background plus 400 µM CaCl2 was used for the pretreatment and experimental periods. Nutrients were supplied at RARs of 0.09, 0.12, 0.15, 0.18, and 0.21 d–1 in both pretreatment and experimental solutions.
|
Results and discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Time-course of growth (200 µM N background)
The RGR of plants during the pretreatment period (0.305 d–1; Fig. 2A, open symbols) exceeded the 0.20 d–1 RAR. This higher rate of growth was presumably due to the mobilization of nutrient reserves from seeds, which provided plants with nutrients above those supplied by the nutrient additions. Growth was log-linear during the first 14 d of the experimental period, but the RGR (0.084 d–1) was lower than the RAR (0.20 d–1). Note the differences between the expected growth (dashed line) and the observed growth (solid line) in Fig. 2A. The RGR subsequently accelerated to 0.171 d–1 for the remainder of the experimental period (Fig. 2A). Restricted growth during the first 14 days of the experiment was accompanied by visible signs of stress. Roots took on a light brown colouration with restricted development of lateral roots. Growth of lateral roots resumed after day 14 and new growth appeared white and healthy for the remainder of the experimental period. The reduction in visible signs of stress coincided with the >2-fold increase in the RGR (Fig. 2A).
|
Since the RAR technique uses preprogrammed nutrient additions, reduced growth (RGR<RAR) results in nutrient additions exceeding nutrient uptake. This leads to the accumulation of nutrients in the growth solutions and increases in solution EC, which were observed over the experimental period (Fig. 2B). Increased nutrient availability can also affect plant-induced pH changes in solution. This is particularly true if plants have access to alternative sources of N in solution (
and ), since the form of N assimilated frequently dominates the overall cation/anion balance. If RGR=RAR, plants must assimilate N in the proportion supplied. However, if RGR<RAR, N builds up in the experimental solutions and plants can select from different sources of N. Ammonium () is preferentially assimilated by many plants (including wheat) and assimilation of is accompanied by a release of H+ (Loneragan, 1979; Taylor, 1988; Taylor and Foy, 1985). Thus, an accumulation of N in the experimental solutions leads to the preferential assimilation of (relative to ) and a decrease in solution pH. As would be expected, the major decline in solution pH observed in this experiment occurred during the second half of the experimental period (Fig. 2C), coincident with major increases in nutrient availability and solution conductivity (Fig. 2B).
Visible signs of stress in roots that are similar to those reported here have previously been interpreted as reflecting a period of plant adjustment to changes in internal nutrient status that follow the onset of nutrient addition with the RAR technique (Ingestad, 1981). The low growth rates (RGR<RAR) and visible symptoms of stress during the first 2 weeks of the experimental period have been interpreted as symptoms of nutrient deficiency. It is interesting to note that Stadt et al. (1992) observed symptoms of nutrient stress in plants grown with low background levels of nutrients (0–180 µM N), but these symptoms were not accompanied by reductions in RGR below the RAR. This may have reflected the low RARs used in their study (0.05 and 0.15 d–1). At the high RAR used in this study (0.20 d–1), an inadequate supply of nutrients in the background solutions may have placed real constraints on growth during the experimental period. Note that an inadequate background concentration in the pretreatment period would not necessarily reduce growth, since the mobilization of seed reserves could provide sufficient nutrients required for healthy growth. This was supported by the 0.305 d–1 growth rate observed during the pretreatment period.
Background nutrient concentrations
In this experiment, background nutrient concentrations were manipulated in an effort to overcome a possible nutrient limitation. When background concentrations were increased from 50 µM N to 500 µM N, the accumulation of plant biomass increased (Fig. 3) and the overall RGR increased from 0.08 to 0.165 d–1. Little additional growth was observed at concentrations greater than 500 µM N (Fig. 3). Ingestad (1972) found that growth of Cucumis sativus L. (cucumber) was reduced when nutrient concentrations were reduced below 14 mM N. In another study, Ingestad and Lund (1979) concluded that total N concentration must be at least 1.8 mM but not greater than 19.3 mM for maximum growth of Betula verrucosa Ehrh. (birch). These higher optimal concentrations may have reflected differences in experimental species (wheat versus cucumber or birch) or techniques (solution culture versus nutrient mist culture). Stadt et al. (1992) reported a linear growth response of wheat with background concentrations ranging from 0 to 360 µM N and a RAR of 0.15 d–1. Since maximum growth in this experiment was achieved at background concentrations of 500–1000 µM N, a single backgound of 500 µM N was used for the following time-course study.
|
Time-course of growth (500 µM N background)
When the background nutrient concentration was increased to 500 µM N, plant RGR remained close to the 0.20 d–1 RAR throughout the 22 d experimental period (Fig. 4A). Visible signs of stress (reduced lateral growth, brown colouration) were still observed during the first week of growth, but these signs disappeared after day 8 as lateral roots resumed growth. In contrast to the previous time-course experiment, where RGR<RAR and solution EC increased to
1300 µS cm–1 (Fig. 2), solution EC remained virtually constant (40–90 µS cm–1, Fig. 4B). Thus, it appeared that nutrient uptake was in balance with the nutrient supply. Nonetheless, the pH of the growth solution increased to over pH 6.5 after 2 weeks (Fig. 4C), suggesting that anion uptake exceeded cation uptake.
|
Even though log linear growth was achieved using a 500 µM N background (Fig. 4A), symptoms of stress observed during the initial week of the experiment suggested a continuing deficiency of some essential nutrient. The brown discoloration of roots and reduced growth of laterals resembled symptoms associated with calcium deficiency (Loneragan et al., 1969
).
Calcium background
To determine if symptoms of stress observed in the previous experiment could be alleviated and if the levels of background nutrients could be reduced by increasing Ca supply, total Ca concentrations in 200 µM N background solutions were varied from 14 to 2000 µM. Root growth increased 2-fold (0.036 to 0.073 g pot–1) and symptoms of stress gradually disappeared as Ca was increased to 400 µM. Above this concentration, little additional growth was observed (Fig. 5). This suggests a minimum concentration of 400 µM Ca is required for healthy root growth under these experimental conditions.
|
Background nutrients plus 400 µM total calcium
This experiment was designed to determine if the background concentrations of other nutrients could be reduced when a higher background Ca level (400 µM) was used. When other background nutrients were varied over the range between 50 and 1000 µM N, growth increased linearly (Fig. 6A), and no visible signs of root stress were observed in any treatment. Stadt et al. (1992)
also observed a linear response of growth when wheat was supplied with a more restricted range of background nutrient concentrations. The EC of growth solutions changed little during the experimental period for all background treatments. Nonetheless, EC values in the lower background treatments (50–200 µM N), showed less variation with time than values in the higher background treatments. For the higher background treatments (700–1000 µM N), there was a greater tendency for EC readings to decline between days 9 and 19 (Fig. 6B), suggesting the RGR of plants in these high-background treatments was greater than in the low-background N treatments and greater than the RAR. This was confirmed when the overall growth rates were compared. The relative growth rates of the high background treatments were 0.005 to 0.015 g g–1 plant d–1 greater than in the 50–200 µM N treatments (rates ranged from 0.176 to 0.181 g g–1 plant d–1). In all background treatments, pH increased from 4.3 to over 7.0 (Fig. 6C), once again suggesting that anion uptake exceeded cation uptake.
While plant-induced pH fluctuations indicated an imbalance in cation/anion consumption, this experiment confirmed that healthy growth could be achieved with a 50 µM N background with a total concentration of 400 µM Ca. Under these conditions, total nutrient consumption was in balance with nutrient supply and EC values remained within a narrow range. The remainder of the experiments in this study were conducted using the 50 µM N background + 400 µM CaCl2 combination.
Nitrogen source (cv. Katepwa)
The objective of this experiment was to control plant-induced changes in pH by identifying an /N ratio that minimizes the observed pH changes without growth reductions due to toxicity. The highest levels of growth were attained when was supplied between 5% and 25% of total N. Growth declined as levels were increased above 25% (Fig. 7A). Growth reductions at higher /N ratios were similar to reductions observed in cucumber by Ingestad (1972) when exceeded 80% of the N supply. By contrast, Ingestad and Stoy (1982) found that the growth of wheat, Hordeum vulgare L. (barley), and Avena sativa L. (oats) was inhibited when was as low as 10% of the nitrogen supply, suggesting that sensitivity varies between and within species. The decline in growth observed in this experiment was accompanied by visible signs of stress, including the reduced growth of lateral roots, yellowing of roots, interveinal chlorosis, and tip necrosis of shoots. The severity of these symptoms increased as the /N ratio increased. Similar toxicity symptoms have been observed in previous studies with Phaseolus vulgaris L. (bean), cucumber and Pisum sativum L. (pea; Maynard and Barker, 1969), barley, Zea mays L. (maize), and oats (Findenegg, 1987). Electrical conductivity was relatively stable for all treatments where growth rates were not reduced by toxicity. By contrast, in treatments where growth was reduced (Fig. 7A), EC increased (see for example, =50%, Fig. 7B), presumably as a result of the reduced accumulation of nutrients by the slower growing plants. No differences in growth rate were observed in any of the NaCl treatments compared with the maximum rate observed in the nitrogen alone treatments (data not shown).
The extent of plant-induced pH fluctuations decreased with increasing /N ratio, with little change in pH being observed with an /N ratio of 50% (Fig. 7C). Unfortunately, severe growth reductions were observed at this level, suggesting that pH control could not be achieved with this cultivar by increasing levels. In experiments with a different wheat cultivar (cv. Neepawa), Stadt et al. (1992) did not observe growth reductions until exceeded 40% of the total nitrogen. These studies suggest that pH control may be achieved with a wheat cultivar less sensitive to .
Nitrogen source (cv. Atlas 66)
In the previous experiment, growth reductions in cv. Katepwa were observed when exceeded 25% of total N, but higher levels were required to control changes in plant induced pH. In this experiment, it was tested whether changes in /N ratio could be used to control plant-induced pH changes in a more resistant cultivar, Atlas-66. By contrast with experiments with cv. Katepwa (Fig. 7), where maximal growth was observed at low levels (<25% of total N), maximal growth in Atlas 66 was observed at intermediate levels (20–40%; Fig. 8A). Sensitivity to low (5–15% ) and high (over 40% ) was also suggested as visible signs of stress (shoot chlorosis and brown root colour) were observed in these treatments. A relatively stable EC and reduced pH fluctuations were achieved at 35% (Fig. 8B, C), suggesting that a balance between cation and anion uptake, and a balance between nutrient uptake and supply had been achieved under these conditions.
Time-course of growth (Atlas 66)
A 0.20 d–1 RAR under the optimal conditions described above (50 µM N + 400 µM CaCl2; () increased slightly to 36%) produced close control of RGR. Growth was log-linear with time throughout the experimental period, with an RGR of 0.187 d–1 over the 22 d experiment (Fig. 9A). Closer analysis indicated a slight reduction in growth rate at the end of the experimental period. The RGR for the first 12 d of the experiment was 0.206 d–1, but declined to 0.158 d–1 for the remainder of the experimental period. This reduction in RGR was accompanied by a relatively minor increase in EC and a small decline in pH of growth solutions (Fig. 9B, C). The increases in EC observed at the end of the experiment presumably reflected a higher rate of nutrient addition (RAR=0.20 d–1) than nutrient uptake (to support an RGR of 0.158 d–1). Accumulation of nutrients in the growth solution (caused by declining growth) would result in greater availability of () for uptake, resulting in reduced pH.
Time-course for various RARs
In this final experiment, the effect of varying RARs (0.09–0.21 d–1) on plant growth and plant-induced changes in nutrient solution were investigated over a 24 d experimental period. As expected, plant growth increased with increasing rates of nutrient addition (Fig. 10A), confirming previous reports that the modified RAR technique provides control over plant RGR (Stadt et al., 1992). In this experiment, it was also shown that different RGRs can be maintained with relatively modest plant-induced changes in growth solutions. In the low RAR treatments (0.09–0.15 d–1), solution EC fluctuated between 125 and 185 µS cm–1 and solution pH varied by 0.5 pH units over the 24 d growth period. This contrasts with pH changes of more than three orders of magnitude in several of these optimization experiments (see for example, Figs 6C, 7C) or in experiments with conventional solution culture (see for example, Taylor and Foy, 1985). Solution EC showed greater increases in the 0.18 and 0.21 d–1 RAR treatments. The 0.21 d–1 RAR treatment was selected to provide conditions where RAR>RGRmax, thus the rise in EC was expected. As in previous experiments where RAR>RGR and solution EC increased, a parallel decline in solution pH was observed. The 0.18 d–1 RAR treatment was below the estimated RGRmax for these experimental conditions. Nonetheless, growth in the 0.18 d–1 RAR treatment was not as rapid as expected (RGR<RAR). Since nutrients began to accumulate in solution during the last 8–10 d of the experiment, a change in solution EC and pH were observed. Visible signs of stress (stubby, yellow roots) were observed in the 0.21 d–1 RAR treatment during this period.
These experiments demonstrate that a computer-controlled nutrient-delivery system can be used to manipulate background nutrient supply, daily nutrient additions, and cation/anion balance to support high rates of plant growth (0.2 g g–1 plant d–1). While optimal nutrient supply may be genotype specific, balancing nutrient supply and consumption provides a means of maintaining log-linear growth under relatively constant conditions of solution EC and pH.
|
Acknowledgements |
|---|
We would like to thank Ken Stadt for his valuable technical assistance with the computer-controlled delivery system. This research was supported by funds provided by the Natural Sciences and Engineering Research Council of Canada (NSERC) research grants program (Discovery Grant-individual to GJT). Funding for the computer-controlled nutrient-delivery system was provided by funds from the NSERC's Research Tools and Instruments Grant Program to GJT.
|
Footnotes |
|---|
* Present address: Alberta Environment, 4th Floor Oxbridge Place, 9820 106 Street, Edmonton, Alberta T5K 2J6, Canada.
Abbreviations: RAR, relative addition rate; RGR, relative growth rate; EC, electrical conductivity; , nitrate; , ammonium.
Investigators interested in additional technical details of the system are welcome to contact the authors.
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResults and discussionReferences |
|---|
Asher CJ. 1981. Limiting external concentrations of trace elements for plant growth: use of flowing solution culture techniques. Journal of Plant Nutrition 3, 163–180.
Asher CJ, Blamey FP. 1987. Experimental control of plant nutrient status using programmed nutrient addition. Journal of Plant Nutrition 10, 1371–1380.
Asher CJ, Ozanne PG, Loneragan JF. 1965. A method for controlling the ionic environment of plant roots. Soil Science 100, 149–156.
Findenegg GR. 1987. A comparative study of ammonium toxicity at different constant pH of the nutrient solution. Plant and Soil 103, 239–243.
Ingestad T. 1972. Mineral nutrient requirements of cucumber seedlings. Plant Physiology 52, 332–338.
Ingestad T. 1981. Nutrition and growth of birch and grey alder seedlings in low conductivity solutions and at varied relative rates of nutrient addition. Physiologia Plantarum 52, 454–466.[CrossRef]
Ingestad T. 1982. Relative addition rate and external concentration; driving variables used in plant nutrition research. Plant, Cell and Environment 5, 443–453.
Ingestad T, Lund A. 1979. Nitrogen stress in birch seedlings. Physiologia Plantarum 45, 137–148.[CrossRef]
Ingestad T, Lund A. 1986. Theory and techniques for steady-state mineral nutrition and growth of plants. Scandinavian Journal of Forest Research 1, 439–453.
Ingestad T, Stoy V. 1982. Mineral nutrition of wheat, rye, barley, and oat seedlings in nutrient solutions. Swedish Journal of Agricultural Research 12, 185–192.
Loneragan JF. 1979. The interface in relation to root function and growth. In: Harley JH, Russell RS, eds. The soil–root interface. New York: Academic Press, 351–367.
Loneragan JF, Snowball K, Simmons WJ. 1969. Response of plants to calcium concentration in solution culture. Australian Journal of Agricultural Research 19, 845–857.
Maynard DN, Barker AV. 1969. Studies on the tolerance of plants to ammonium nutrition. Journal of the American Society for Horticultural Science 94, 235–239.
Pettersson S, Strid H. 1989. Aluminum toxicity in two cultivars of wheat (Triticum aestivum L.) with different sensitivity to Al as affected by the level of nutrient supply. Swedish Journal of Agricultural Research 19, 183–191.
Stadt K, Taylor GJ, Dale MRT. 1992. Control of relative growth rate by application of the relative addition rate technique to a traditional solution culture system. Plant and Soil 142, 113–122.[CrossRef]
Taylor GJ. 1988. Mechanisms of aluminum tolerance in Triticum aestivum L. (wheat). V. Nitrogen nutrition, plant-induced pH, and tolerance to aluminum; correlation without causality? Canadian Journal of Botany 66, 694–699.
Taylor GJ, Blamey FPC, Edwards DG. 1998. Antagonistic and synergistic interactions between aluminum and manganese on growth of Vigna unguiculata at low ionic strength. Physiologia Plantarum 104, 183–194.[CrossRef]
Taylor GJ, Foy CD. 1985. Mechanisms of aluminum tolerance in Triticum aestivum L. (wheat). IV. The role of ammonium and nitrate nutrition. Canadian Journal of Botany 63, 2181–2186.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||