- Email: [email protected]
Proton release by roots of Medicago murex and Medicago sativa growing in acidic conditions, and implications for rhizosphere pH changes and nodulation at low pH
Soil Biology & Biochemistry 36 (2004) 1357–1365 www.elsevier.com/locate/soilbio
Proton release by roots of Medicago murex and Medicago sativa growing in acidic conditions, and implications for rhizosphere pH changes and nodulation at low pH Y. Chenga,b, J.G. Howiesonb,*, G.W. O’Harab, E.L.J. Watkinb,c, G. Souchea, B. Jaillarda, P. Hinsingera a
b
INRA UMR Rhizosphe`re et Symbiose, Place Viala, 34060 Montpellier Cedex 01, France Division of Science and Engineering, Centre for Rhizobium Studies, Murdoch University, South Street, Murdoch Drive, Murdoch, WA 6150, Australia c School of Biomedical Sciences, Curtin University of Technology, G.P.O. Box U1987, Perth, WA 6845, Australia
Abstract Medicago murex nodulates faster and produces more nodules than Medicago sativa in acidic sandy soils. Experiments using a ‘root mat’ approach and videodensitometry examined pH changes in the rhizospheres of nitrate-fed plants of M. murex and M. sativa. Using the ‘root mat’ approach with soil disks of pH 4.49, M. sativa cv. Aquarius acidified its rhizosphere by approximately 0.2– 0.4 pH-units within 4 d, while M. murex cv. Zodiac did not acidify its rhizosphere. Rates of Hþ release were higher from M. sativa than from M. murex. Videodensitometry of roots embedded in agarose of pH 4.5 showed that the mature parts of the tap-root of both species exuded OH2 ions, but was approximately twofold more in M. murex than in M. sativa. Consequently, young parts of the M. sativa rhizosphere were less alkaline than that of M. murex. It is suggested that the difference in nodulation response between the two species at low pH may be related to the different patterns of rhizosphere acidification: the stronger rhizosphere acidification of M. sativa being less favourable for survival and growth of Sinorhizobium medicae. The higher rate of rhizosphere acidification by M. sativa roots may be related to its genetic characteristics including greater relative root growth rate and greater sensitivity to acidity in comparison to M. murex. q 2004 Published by Elsevier Ltd. Keywords: Acid production; Nodulation delay; Proton release; Rhizosphere; Rhizobia; Rhizosphere pH; Root mat; Videodensitometry
1. Introduction Medicago sativa (lucerne; alfalfa) has the capacity to transpire large volumes of water and is well suited for controlling rising water tables in south-western Australia (Cransberg and McFarlane, 1994; Ward et al., 2000). However, its widespread use is currently limited as soils in the region are too acidic (Bettenay and Hingston, 1964) for consistent nodulation with its microsymbiont Sinorhizobium medicae (Munns, 1968, 1970). Medicago spp. differ in their ability to nodulate in acidic soil and the annual species M. murex grows and nodulates well in soil of pH 4.9 (Howieson and Ewing, 1989). However, M. murex is not commercially attractive as a pasture legume and cannot fill the niche of M. sativa in lowering groundwater tables. The nodulation of M. murex and M. sativa is affected differently by soil acidity (Cheng et al., 2002). When grown * Corresponding author. Tel.: þ 61-8-9360-2231; fax: þ61-8-9360-6486. E-mail address: [email protected] (J.G. Howieson). 0038-0717/$ - see front matter q 2004 Published by Elsevier Ltd. doi:10.1016/j.soilbio.2004.04.017
in an acidic sandy soil of pH 4.3, M. murex produced fewer root nodules than when grown in soil of pH 7.0, but these nodules formed at a similar rate. In the same soil, M. sativa also produced fewer nodules when grown in the acidic soil, but the nodules appeared later compared to those on plants grown in soil of pH 7.0 (Cheng et al., 2002). These differences in nodulation response could be due to differential multiplication of S. medicae in the rhizosphere. After 24 d growth in soil of pH 4.3 inoculated with an acidtolerant strain of S. medicae, there were over 100-fold more S. medicae associated with the roots of M. murex than M. sativa (Cheng, unpublished). The ability of different pasture legumes to modify rhizosphere pH varies within and between species (Tang et al., 1997). The average rhizosphere pH of M. sativa seedlings growing in soil of pHH2 O 5.2 increased from pH 5.1 (1 d after sowing) to pH 5.7 (12 d after sowing), but nodulation was not affected (Pijnenborg et al., 1990). Localised changes in rhizosphere pH have also been observed along the roots of M. sativa (Blanchar and Lipton, 1986). In soil of pHH2 O 5.5, the rhizosphere pH of
1358
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
10 d old M. sativa seedlings at the older tap roots was more alkaline (pH 6.8), while the rhizosphere pH at the younger lateral roots was more acidic (pH 4.2) (Blanchar and Lipton, 1986). Modification of rhizosphere pH by M. murex roots has not yet been studied, and M. murex may modify the pH of the rhizosphere differently from M. sativa, thereby developing more favourable conditions for rhizobial survival and growth. The acidifying effect of N2-fixing legumes has been demonstrated in both solution culture (McLay et al., 1997; Tang et al., 1997) and soil (Jarvis and Robson, 1983a,b; Tang et al., 1998). To our knowledge, there is limited research on young plants prior to nodulation where rhizosphere pH changes may affect the pre-infection steps of the nodulation process. As recently reviewed by Hinsinger et al. (2003), root-induced pH changes in the rhizosphere are primarily caused by root and microbial respiration (Nye, 1981), unbalanced uptake of inorganic anions and cations (Haynes, 1990), exudation of organic anions (Hoffland et al., 1989; Neumann and Ro¨mheld, 2001) and oxidation of soil minerals (Hinsinger, 1998). The release of Hþ to counterbalance an excess uptake of cations over anions is often the major source of root-induced changes of rhizosphere pH (Hinsinger et al., 2003). Methods used to measure Hþ concentrations in the rhizosphere, and fluxes of Hþ released by roots include approaches investigating the whole root system or single roots (Gregory and Hinsinger, 1999). At the level of the whole root system, the ‘root mat’ approach has been used to measure pH gradients in the rhizosphere of legumes such as Trifolium repens (white clover) (Li et al., 1991) and T. subterraneum (subterranean clover) (Hinsinger and Gilkes, 1996). Backtitration of the nutrient solution can also measure fluxes of Hþ released by whole root systems in solution (McLay et al., 1997; Tang et al., 1997), but cannot provide localised pH values along a root. This localisation of pH changes along single roots may be semi-quantitatively measured by colorimetry coupled with the agar-indicator technique (Weisenseel et al., 1979), quantitatively by micropotentiometry (Marschner and Ro¨mheld, 1983; Yu et al., 2000), or both of these techniques combined (Pijnenborg et al., 1990). Videodensitometry (Jaillard et al., 1996), based on the principles of colorimetry, was developed to quantitatively measure Hþ flux using coloured pH indicator with agar or agarose gels. It provides maps of pH values along intact roots (Gregory and Hinsinger, 1999; Jaillard et al., 1996, 2003), and simultaneously calculates temporal and spatial profiles of Hþ flux (Plassard et al., 1999). Our aim was to determine whether M. murex and M. sativa induced pH changes in the rhizosphere differentially under acidic conditions, and to test the hypothesis that at low pH the rhizosphere of young parts of the M. murex root are less acidic than those of M. sativa. Relatively young and un-nodulated plants of M. murex and M. sativa were used as there has been limited research on rhizosphere pH changes in young plants prior to nodulation. Seeds of
M. murex cv. Zodiac and M. sativa cv. Aquarius were used in these experiments because the same genotypes had been studied for their nodulation in an acidic soil (Cheng et al., 2002).
2. Materials and methods 2.1. Experimental design Experiment 1 used the ‘root mat’ approach designed by Chaignon et al. (2002) to test changes in soil rhizosphere pH. The experiment consisted of two Medicago species (M. murex cv. Zodiac and M. sativa cv. Aquarius), two pH treatments (initial soil pHCaCl2 of 4.49 with de-ionised water or nutrient solution at pH 4.5, and initial soil pHCaCl2 of 6.71 with de-ionised water or nutrient solution at pH 6.0). Deionised water and nutrient solution were used to compare the responses of the two species in the absence or presence of a cation/anion supply from the solution medium. Six replicates were prepared for each treatment, and one additional replicate was used to assess shoot and root dry weight at the commencement of treatment. Root mats of the two Medicago species were grown in a specially designed cropping device (see Chaignon et al., 2002) in nutrient solution of pH 6.0 for 13 d. Roots were then placed in contact with soil disks. Soil disks of pH 4.49 were placed above solution media of pH 4.5, and soil disks of pH 6.71 were placed above solution media of pH 6.0, for a further 4 d. Control soil disks without root contact were also used as references. Bulk Hþ or OH2 release fluxes were calculated from pH changes in the soil disks between 13 and 17 d after sowing, relative to the soil without root contact. Experiment 2 used the videodensitometry technique designed by Jaillard et al. (1996). M. murex and M. sativa were grown in nutrient solution of pH 6.0 for 13 d. In each of six measurements, three to five plants of each species were embedded in nutrient gel film of pH 4.5 or 6.0 containing pH indicator. Values of proton flux were calculated from images acquired with a video camera. 2.2. Plant and soil material Seeds of M. murex cv. Zodiac and M. sativa cv. Aquarius were surface sterilised in 6% (v/v) H2O2 solution (10 min) to remove seed-borne microorganisms, and then thoroughly rinsed with de-ionised water. The nutrient solution for plant culture contained: 500 mM K2SO4, 200 mM MgSO4, 100 mM FeEDTA, 100 mM Ca(NO3)2, 10 mM KH2PO4, 3 mM H3BO3, 1 mM MnCl2, 30 nM Na2MoO4, 0.75 mM ZnSO4, 0.2 mM CuSO4. The pH was adjusted to 6.0 with 10 mM KOH. The composition of the nutrient solution was based approximately on the chemical properties of the soil used by Cheng et al. (2002). Nitrate was used as the main source of N in the nutrient solution to avoid extreme acidification of the rhizosphere and to prevent N deficiency
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
in plants. Plants were not inoculated with rhizobia. For both experiments, plants were placed in a growth cabinet with a 16 h, 25 8C/8 h, 20 8C day/night regime and a photon fluence rate of 600 mmol photons m22 s21. The soil used in experiment 1 was an acidic sand from Merredin, Western Australia (1188170 E 318290 S), mixed with coarse river sand (3:2) (Cheng et al., 2002). The pH values of 4.49 and 6.71 were achieved by liming the soil mix at two rates (0.12 and 1.5 g kg21). The two rates of lime were added to slightly dampened soil in a cement mixer. After liming, the soil was left then overnight at room temperature in covered plastic tote-boxes. For transportation purposes, the soil was dried at 70 8C for 48 h, and sealed in plastic bags. The soil mix consisted of approximately 88% sand (2.0 – 0.02 mm), and had available N (measured by KCl extraction; Rayment and Higginson, þ 2 1992) of 1 mg kg21 each of NO2 3 and NH4 , and NaHCO3 21 extractable P of 14– 15 mg kg (Colwell, 1963). The soil mix was wet to field capacity with de-ionised water (10%, w/v) and incubated in the dark for 24 h before being used.
1359
mat of roots within 13 d after sowing. Plant culture in this experiment was a two-step procedure. In the first step, root mats were made by growing plants of M. murex and M. sativa in an aerated nutrient solution (pH 6.0; replaced after 7 d) for a period of 13 d (Fig. 1A). In the second step, cropping devices containing 13 d old plants were randomised, taped securely in contact with soil disks (Fig. 1B) and placed over an aerated nutrient solution. Soil disks without plants were covered with an opaque plastic lid to minimise evaporation and algal growth. All soil disks were placed in the growth cabinet for a further 4 d. In experiment 2 using videodensitometry, plants were grown in a similar cropping device as described by Chaignon et al. (2002) for experiment 1, except the lower nylon mesh (30 mm) was not used, and the upper nylon mesh (900 mm) was replaced by a cloth mesh of 900 mm. Roots grew through the cloth mesh into 5 l of an aerated nutrient solution (pH 6.0), which was changed after 7 d, for a period of 13 d. 2.4. Measurements and calculations
2.3. Cropping device and procedure The cropping device for experiment 1 was that described by Chaignon et al. (2002). Soil (2.5 g wet soil) was placed into shallow PVC containers (32 mm i.d.) to produce disks approximately 1.5 mm thick. Soil disks of pH 4.49 were then placed on shelves above 5 l of aerated de-ionised water or nutrient solution of pH 4.5 (adjusted with 10 mM H2SO4), while soil disks of pH 6.71 were placed above 5 l of aerated de-ionised water or nutrient solution of pH 6.0. Soil disks were kept moist by capillary rise through a filter paper strip. To prevent the transfer of Hþ or OH2 ions between different treatments through the solution medium, the six replicates of the same treatment were placed in the same solution medium. Tests found (data not shown), 40 seeds/cropping device of both Medicago spp. were needed to form a dense, planar
2.4.1. Relative root growth rate In experiment 1 at the time of soil –plant contact (13 d after sowing), shoots and roots of plants from one additional replicate were separated, fresh weight recorded, dried at 105 8C for 24 h and re-weighed. After 4 d of treatment (at 17 d after sowing), plants placed on soil disks were removed from the soil disks. The shoots and roots were separated, and fresh and dry weights recorded. Relative root growth rate (RRGR, (mg g21 dry weight) d21) was calculated by the equation RRGR ¼
ðW2 2 W1 Þ W1 £ ðt2 2 t1 Þ
where W1 and W2 are the initial and final dry weight of roots at time t1 and t2 ; respectively. The pH of the wet soil at the commencement of root contact was measured
Fig. 1. Cropping device used to produce root mats of M. murex and M. sativa in a two-step process. (A) Step 1, growth of seedlings in aerated nutrient solution for 13 d; (B) step 2, roots placed in contact with soil disk, and plants cultured in aerated nutrient solution for a further 4 d. Experiment 1.
1360
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
by shaking soil for 1 h in 10 mM CaCl2 (1:5) (Metrohm 744 pH meter, Herisau, Switzerland). The pH of soil disks in contact with plants represents rhizosphere pH. As the moisture content of the soil disks may have changed during the course of the experiment, it was necessary to determine the exact volume of CaCl2 to add to the soil to constitute a 1:5 suspension for pH measurements. This was done by the following procedure. Four soil disks with root contact and four disks without root contact were randomly chosen. The moisture content of 0.5 g of soil from each randomly chosen disk was determined after drying at 105 8C for 1 h. A total of 5 ml of 20 mM CaCl2 was added to the remaining 2 g of soil from each soil disk. Depending on the moisture content of the 0.5 g of soil, a volume of de-ionised water added to the 2 g soil suspended in 20 mM CaCl2 such that its dilution resulted in a 1:5 suspension in 10 mM CaCl2. The soil suspension was then shaken for 1 h and pH measured. 2.4.2. Hþ flux To estimate Hþ flux released by the plant roots, pH buffer curves of the soils were first determined. Soil of both low and high pH were wet to 10% moisture (w/v) with either 10 mM CaCl2 or solutions of 10 mM CaCl2 containing various concentrations of HCl or NaOH. The concentrations of HCl and NaOH ranged from 0.4 to 8.0 mM. Soil was incubated in the dark at 22 8C for 3 d, suspended in 10 mM CaCl2 (1:5), shaken for 1 h and the pH measured. The pH buffer capacity of soil was calculated from the slope (b; in mmol (g soil21) (pH unit)21) of the linear regression of the amount of HCl or NaOH added versus the soil pH (Tang et al., 1998). The soil pH were 4.49 and 6.71 (10 mM CaCl2). The b values were 2 8.1 (g soil21) (pH unit)21 (soil of pH 4.49, r 2 ¼ 0:96), and 2 9.6 mmol (g soil21) (pH unit)21 (soil of pH 6.71, r 2 ¼ 0:98). Amount of proton released was calculated from total acid production, estimated as the difference in acidity between soils with plants and control soils without plants. Thus, flux of proton jH þ (nmol (g root)21 s21) released by roots was calculated as jHþ ¼
½bðpH0 2 pHÞ £ Msoil ðMroot £ tÞ
where pH0 is the pH of soil disks without plants; pH, pH of soil disk with plants; Msoil ; mass of the soil disk; Mroot ; root dry weight, averaged from plants at the beginning of soil – plant contact, and from plants 4 d after soil –plant contact; t is the time duration of soil – plant contact in seconds (4 d in this experiment, i.e. 345,600 s). In experiment 2, after 13 d of growth in nutrient solution of pH 6.0, plants were equilibrated in fresh nutrient solution. This was done 48 h before embedding in nutrient agarose gel. Plants for the pH 4.5 treatment were placed in fresh nutrient solution of pH 5.2 for 24 h, and then transferred to nutrient solution of pH 4.5, 24 h before embedding. This intermediate step decreased acid-shock and subsequent necrosis of root tips. Plants for the pH 6.0
treatment were transferred to fresh nutrient solution of pH 6.0, 24 h before embedding. By the time roots were embedded, plants were 15 d old. A separate volume (100 ml) of each nutrient solution (pH 4.5 and 6.0) that was used for equilibrating plants was also placed in the growth cabinet for 24 h before embedding. These solutions were used for embedding plants. Solutions of pH indicators bromocresol green (pK ¼ 4:69 for measurements at pH 4.5) and bromocresol purple (pK ¼ 6:40 for measurements at pH 6.0) were prepared first. Films were prepared by melting 1 g of agarose powder (Fluka Ref 05068, Buchs, Switzerland; see Calba et al., 1996), in 20 ml of pH indicator solution and 80 ml of the nutrient solution used for plant culture (pH 4.5 or 6.0), yielding a final pH indicator concentration of 90 mg l21. The agarose solutions were boiled for 30 min, cooled to 37 8C in a water bath and pH readjusted to 4.5 or 6.0 with 10 mM KOH or H2SO4. Plants were gently removed from the cloth mesh so not to damage roots, roots blotted dry with tissue, and carefully placed between two glass sheets (100 £ 200 mm2) separated by a 1 mm thick strip of PVC. The agarose solution was then syringed between the glass sheets. After approximately 5 min when the agarose solution had cooled and set, the upper glass sheet was removed and a 3 mm PVC strip was placed between the glass sheets so as not to confine the roots, and to enable respired CO2 to diffuse into the atmosphere. Plants were placed in the growth cabinet and images were acquired by videodensitometry 2 h after embedding. Two saturated calibration standards (pH 3.5 and 8.0 with bromocresol green, and pH 4.8 and 8.5 with bromocresol purple) were prepared for each series of measurements. Three to five plants of each species per treatment were embedded and measured in six separate experiments. Images were acquired using the methods described by Plassard et al. (1999). Hþ flux profile were calculated at three defined locations along the root: 3, 20 and 50 mm from the root tip. These locations correspond with the emerging root hair, growing root hair and mature root hair zones of the root, respectively, with decreasing susceptibility to infection by rhizobia further away from the root tip (Bhuvaneswari et al., 1981; Wood and Newcomb, 1988). 2.5. Statistical analysis Variability in the measured values among the various replicates is indicated by standard deviations. Significant differences between treatments were deduced from an analysis of variance (ANOVA) using Statistica for Windows 1999 (Kernal Release 5.5, StatSoft, Inc., Tulsa, Oklahoma, USA), and a t-test using Microsoft Excel 2000 (Microsoft Corporation, Washington, USA) at P , 0:05:
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
1361
Table 1 Root dry weights (mean value of six replicates ^ standard deviation) of Medicago murex and M. sativa cultured in de-ionised water or nutrient solution at pH 4.5 and 6.0 for 4 d, and their relative root growth rate (RRGR) between 13 and 17 d. Experiment 1 Solution media
pH
Root dry weight (mg plant21), 17 d after sowing
RRGR (mg g21 dry weight d21), 13–17 d after sowinga
M. murex
M. sativa
M. murex
M. sativa
Water
4.5 6.0
614 ^ 354 732 ^ 379
248 ^ 140 380 ^ 118
1.84 ^ 0.60 2.13 ^ 0.58
2.16 ^ 0.61 2.54 ^ 0.48
Nutrient solution
4.5 6.0
510 ^ 96 780 ^ 261
380 ^ 094 602 ^ 619
1.09 ^ 0.56* 1.94 ^ 0.77
2.60 ^ 0.31 2.65 ^ 0.70
RRGR ¼
ðW2 2 W1 Þ W1 £ ðt2 2 t1 Þ
where W1 and W2 are the initial and final dry weight of roots at time t1 and t2, respectively. *Significant difference in RRGR between M. murex and M. sativa at P , 0:05: a After 13 d of growth in nutrient solution of pH 6.0, root dry weight (mg plant21) for M. murex was 360 (^193), and for M. sativa was 128 (^67).
3. Results 3.1. Root growth In experiment 1, after 13 d of growth in nutrient solution of pH 6.0, root dry weight (mg plant21) for M. murex was 360 (^ 193), and for M. sativa was 128 (^ 67). The greater initial weight of M. murex seeds probably contributed to the greater root dry weight of M. murex compared to M. sativa at 13 d after sowing. Between 13 and 17 d after sowing, RRGR was generally higher in M. sativa than M. murex, despite M. sativa having lower root dry weights in all treatments (Table 1). In nutrient solution of pH 4.5, RRGR was significantly higher (approximately twice) in M. sativa than M. murex. 3.2. Soil pH and Hþ release For soil disks of pHCaCl2 4.49 moistened with de-ionised water or nutrient solution of pH 4.5, the pH of soil in contact with M. murex roots did not differ from the pH of the control
soil, while the pH of the soil in contact with M. sativa roots was significantly lower (Table 2). With soil of pHCaCl2 6.71 moistened with de-ionised water of pH 6.0, both Medicago spp. decreased soil pH. With soil of pHCaCl2 6.71 moistened with nutrient solution of pH 6.0, M. murex roots decreased soil pH while M. sativa roots did not significantly change soil pH relative to the control soil. For soil disks of pHCaCl2 6.71 without root contact, increases in pH after moistening for 4 d with de-ionised water (pH increased by 0.52 pH-unit) and nutrient solution (pH increased by 0.37 pH-unit) (Table 2) may be due to the short time (overnight) in which lime equilibrated in the soil before the limed soil was dried. For soil disks of pHCaCl2 4.49 moistened with nutrient solution of pH 4.5, proton release was significantly higher from M. sativa than M. murex (Table 2), with the rate of proton release from M. sativa being 1.9 nmol (g root)21 s21 compared to M. murex at 2 0.1 nmol (g root)21 s21. The negative value indicates net OH2 release from M. murex. In soil of pHCaCl2 6.71, there was no significant difference in proton release fluxes between the two species. Except for
Table 2 Soil pH and proton extrusion rates (mean value of six replicates ^ standard deviation) from roots of Medicago murex and M. sativa cultured in rhizosphere soil moistened with water or nutrient solution between 13 and 17 d after planting. Experiment 1 Solution Initial pH of soil disk and solution Mean control soil pH Mean rhizosphere soil pH
M. murex
Ratio M. sativa/M. murex
No plants
M. murex
4.49/4.5 6.71/6.0
4.49 ^ 0.02b 7.23 ^ 0.05c
4.45 ^ 0.08b 4.31 ^ 0.07a 6.55 ^ 0.19a 6.70 ^ 0.02b
Nutrient 4.49/4.5 6.71/6.0
4.49 ^ 0.05b 7.08 ^ 0.17b
4.51 ^ 0.04b 4.14 ^ 0.15a 20.1 ^ 0.1* 1.9 ^ 0.9 221.0 5.95 ^ 0.50a 6.78 ^ 0.04b 3.2 ^ 1.3 1.8 ^ 0.9 0.6
Water
M. sativa
Proton extrusion ratea (nmol (g root21 s21)
0.1 ^ 0.2 2.6 ^ 2.3
M. sativa 0.1 ^ 0.0 3.4 ^ 0.8
1.0 1.5
Mean values with different letters within a line are significantly different at P , 0:05: *Significant difference in proton extrusion rate between M. murex and M. sativa at P , 0:05: a Negative value represents net OH2 extrusion.
1362
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
Table 3 Proton efflux rates (mean value ^ standard deviation) from roots of 15 d-old Medicago murex and M. sativa 2 h after embedding in nutrient agarose gel of pH 4.5 and 6.0 containing 100 mM NO2 3 . Experiment 2 PH
Tap-root, distance from root tip (mm)
Proton efflux (pmole (m root)21 s21)a M. murex
M. sativa
Ratio M. murex/M. sativa
4.5
50 20 3
285 ^ 65a 297 ^ 46a* 25 ^ 37b
261 ^ 19a 248 ^ 22a 23 ^ 36b
1.39 2.02 1.67
6.0
50 20 3
28 ^ 3ab 25 ^ 7b 211 ^ 5a*
25 ^ 3a 24 ^ 4a 26 ^ 6a
1.60 1.25 1.83
Mean values with different letters within a column are significantly different at P , 0:05: Mean values from 7 to 11 plants (M. murex), and 8 and 7 plants (M. sativa) for pH 4.5 and 6.0, respectively. *Significant difference in proton efflux between M. murex and M. sativa at P , 0:05: a Negative values represent net OH2 efflux.
M. murex grown in nutrient solution of pH 4.5 releasing OH2, Hþ release was generally higher with plants grown in nutrient solution than in de-ionised water. In experiment 2, at 15 d after sowing, roots of both M. murex and M. sativa released OH2 (Table 3), with M. murex consistently exhibiting larger efflux than M. sativa. The rate of OH2 efflux was significantly higher (approximately twice) in M. murex than M. sativa at 20 mm from the root tip (at pH 4.5), and at 3 mm from the root tip (at pH 6.0). There was no significant difference in OH2 efflux rates between the two species at the other locations along the root. At pH 4.5, the highest rate of OH2 efflux occurred at 20 and 50 mm from the root tip of both species (Table 3). At pH 6.0, M. murex had a lower rate of OH2 efflux at 20 mm, compared to 3 and 50 mm from the root tip. In contrast, there was no significant difference in rates of OH2 efflux at the three locations of the M. sativa root.
4. Discussion 4.1. Changes in rhizosphere pH as affected by Medicago species Roots of M. sativa released more protons than those of M. murex into a sandy soil of pHCaCl2 4.49, while into a nutrient agarose gel of pH 4.5, M. sativa released fewer OH2 ions compared to M. murex. As a result, the rhizosphere of M. sativa was consistently more acidic than that surrounding M. murex. The differential ability of these species to acidify the root medium may be genetic, relating to species adaptation and tolerance to low pH. M. murex is commonly associated with acidic soils (Francis and Gillespie, 1981), nodulates consistently at low pH (Howieson and Ewing, 1989), and is more tolerant of low pH than other annual Medicago spp. (M.A. Ewing, unpub, PhD thesis, University of Western Australia, 1991). Differences in the rate of Hþ release between the two genotypes of M. murex and M. sativa may also be related to
their different relative growth rates of root growth. M. sativa had a greater relative rate of root growth and a higher rate of Hþ release compared to M. murex, which had a slower relative rate of root growth, released less Hþ, and acidified the rhizosphere less than M. sativa. In a study comparing Al-tolerant and Al-sensitive genotypes of Zea mays (maize), Calba et al. (1996) showed similarly that the genotype which grew the fastest, also exhibited the largest nutrient requirements, and hence released the largest fluxes of protons. Different growth rates also affected rhizosphere pH changes in different genotypes of T. subterraneum (Jarvis and Robson, 1983a). Root exudation may be directly associated with root growth since there was almost no root exudation from plants where almost no root growth was observed even though shoots were actively growing (Prikryl and Vancura, 1980). M. murex and M. sativa induced pH changes in the rhizosphere differentially, especially in younger parts of roots where the rhizosphere of M. murex was less acidic than that of M. sativa. In acidic soils where survival and growth of S. medicae (Coventry and Evans, 1989; Robson and Loneragan, 1970), and attachment to host roots (Caetano-Anolle´s et al., 1989; Howieson et al., 1993) are already challenged by low pH, root-induced acidification of the rhizosphere will further inhibit the nodulation process. The more acidic rhizosphere of M. sativa would certainly be less favourable for saprophytic growth and colonisation by S. medicae. Consequently, rhizobial populations of S. medicae on the roots of M. sativa may take longer to reach a sufficient population size to initiate infection (Hirsch, 1992). Results from our study support the hypothesis that slower rates of rhizobial growth caused the delayed nodule initiation and development in M. sativa at low pH (Cheng et al., 2002). Rhizosphere acidification may be an important factor contributing to this. It would be instructive to determine whether other genotypes of M. murex and M. sativa also induce similar pH changes in the rhizosphere. Rhizosphere alkalinisation by seedlings of M. sativa growing in an acidic soil (Pijnenborg et al., 1990)
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
compared to rhizosphere acidification by the same species observed in experiment 1 of our study was probably due to soil properties contributing to different nutrient uptake and hence pH changes. Pijnenborg et al. (1990) suggested that root-induced rhizosphere alkalinisation by M. sativa roots did not result in the formation of nodules on M. sativa since the 0.6 pH-unit increase occurred after the acid-sensitive period of the nodulation process. However, an equivalent decrease in pH would certainly affect nodulation in an acidic soil, especially during the acid-sensitive stages of rhizobial survival, growth and colonisation in the rhizosphere. For plants relying on N2-fixation, the acidity produced by roots during plant growth largely persists in the soil after plants have been removed and may lead to subsoil acidification (Tang et al., 1998). 4.2. Changes in rhizosphere pH as affected by growth conditions Alkalinisation of rhizosphere soil when cultured at low pH, and acidification of rhizosphere soil when cultured at higher pH as displayed by M. murex, has been observed in Hordeum vulgare (barley) and Glycine max (soybean) (Youssef and Chino, 1989) and in Vicia faba (faba bean) (Schubert et al., 1990). The reason for this ‘plasticity’ in response to external pH is unclear, but may be related to nutrient availability, and the subsequent imbalance of cation – anion uptake. In addition, the amount of Hþ released by roots of both genotypes of Medicago spp. was higher in plants grown at higher pH than those grown under acidic conditions. Furthermore, the higher overall rate of ion exchange may result from higher nutrient uptake when plants were less acid-stressed. Decreased plant growth caused by low pH in acid-sensitive species has been related to a build-up of Hþ ions in the cytoplasm, because of a lack of Hþ release (Schubert et al., 1990), or an influx of protons into root cells because of higher membrane permeability (Yan et al., 1992). This might have occurred to a larger extent for M. sativa than for M. murex in our experiment. Compared to plants grown in soil moistened with de-ionised water, plants grown in soil moistened with nutrient solution are likely to have higher rates of nutrient uptake due to access to nutrients from both the soil disk and nutrient solution. This is consistent with the larger proton release fluxes observed in these plants relative to those grown in soil moistened with de-ionised water. Finally, the availability of nutrients (which is likely to vary considerably from soil to soil) will have a considerable influence on both the direction (efflux or influx) and intensity of Hþ fluxes. Such considerations can help in understanding why plants grown in experiment 1 on soil disks decreased rhizosphere pH while plants of a similar age and past nutrient status exhibited net OH2 efflux when grown in nutrient solution. Further work is needed to estimate whether the observed differences
1363
between these and other genotypes of M. murex and M. sativa are consistently found in a range of nutrient þ solutions of varying compositions, especially NO2 3 /NH4 ratios, and in a range of acidic soils. In this respect, it should be stressed that plants were only relying on NO2 3 in experiment 2 (NHþ 4 was not supplied in the agarose gel) and hence exhibited OH2 efflux, while they might have derived some NHþ 4 from the soil in experiment 1, thereby resulting in rhizosphere acidification. 4.3. Changes in rhizosphere pH as affected by the location along the root axis Depending on plant species and N source, but also Fe and P status of the plant, root-induced changes in rhizosphere pH can vary considerably along the root (Marschner et al., 1982; Marschner and Ro¨mheld, 1983). Higher acidification at the root tips compared to basal parts of roots has often been reported, e.g. in Phleum pratense (timothy) (Zieschang et al., 1993) and Z. mays (Marschner and Ro¨mheld, 1983). In our study, the root apex did not behave differently at pH 6.0 (no significant differences were found with other locations along the root axis, see Table 3). However, at pH 4.5, root tips of both genotypes of Medicago spp. exhibited significantly lower efflux of OH2 and were thus more acidic than older regions of the roots. Different rates of ion release in various regions of a root may be related to age of roots cells, and hence affect the uneven pattern of cation and anion uptake along the root. It is possible that older parts of the root take up more NO2 3 , while younger parts take up more cations (Blanchar and Lipton, 1986). This would need to be verified by the means of electrophysiological measurements, i.e. micropotentiometry with NO2 3 selective microelectrodes (Plassard et al., 1999; Jaillard et al., 2003). Susceptibility of root hairs to infection by rhizobia decreases with root hair age. The highest frequency of susceptible root hairs is in the younger region of the root, although some root hairs are susceptible to infection in the more mature region (Bhuvaneswari et al., 1981). At low pH, the more mature regions of roots of both Medicago species had higher rates of OH2 release, were less acidic, and therefore more favourable for rhizobial survival and growth than younger parts of the root. This may explain why nodules formed lower down the tap-root of M. murex when grown in soil of pH 4.3 compared to plants grown in pH 7.0 (Cheng et al., 2002). The faster rate of root growth in M. sativa meant that rhizobia accumulated and attached further down the tap-root than M. murex, and this species variation may have contributed to the different location of the uppermost nodule even in soil of pH 7.0 (Cheng et al., 2002). However, at pH 4.3, rhizobia may have also taken longer to reach sufficient numbers to initiate infection, resulting in delayed nodule development in M. sativa.
1364
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
Acknowledgements This research was done at the Institut National de la Recherche Agronomique, Montpellier, France, funded by a French Government Scientific Fellowship (French Embassy, Canberra, Australia), and a travel grant from the Australian Grains Research and Development Corporation awarded to Yvonne Cheng. We thank the staff and students at the Laboratoire de Science du Sol for their support, especially Nicole Balsera for help with preparing the cropping device for plant culture. Dr Caixian Tang is acknowledged for helpful discussion on the manuscript. We also thank Regina Carr from Murdoch University for help with soil preparation. This research forms part of a PhD thesis by Yvonne Cheng, funded by the Australian Research Council’s SPIRT Program.
References Bettenay, E., Hingston, F.J., 1964. Development and distribution of soils in the Merredin area, Western Australia. Australian Journal of Soil Science 2, 173–186. Bhuvaneswari, T.V., Bhagwat, A.A., Bauer, W.D., 1981. Transient susceptibility of root cells in four common legumes to nodulation by rhizobia. Plant Physiology 68, 1144–1149. Blanchar, R.W., Lipton, D.S., 1986. The pe and pH in alfalfa seedling rhizosphere. Agronomy Journal 78, 216 –218. Caetano-Anolle´s, G., Lagares, A., Favelukes, G., 1989. Adsorption of Rhizobium meliloti to alfalfa roots: dependence on divalent cations and pH. Plant and Soil 117, 67–74. Calba, H., Jaillard, B., Fallavier, P., Arvieu, J.C., 1996. Agarose as a suitable substrate for use in the study of Al dynamics in the rhizosphere. Plant and Soil 178, 67–74. Chaignon, V., Di Malta, D., Hinsinger, P., 2002. Fe-deficiency increases Cu acquisition by wheat cropped in a Cu-contaminated vineyard. New Phytologist 154, 121 –130. Cheng, Y., Watkin, E.L.J., O’Hara, G.W., Howieson, J.G., 2002. Medicago sativa and Medicago murex differ in the nodulation response to soil acidity. Plant and Soil 238, 31–39. Colwell, J.C., 1963. The estimation of the phosphorus fertilizer requirements of wheat in southern New South Wales by soil analysis. Australian Journal of Experimental Agricultural and Animal Husbandry 3, 190 –198. Coventry, D.R., Evans, J., 1989. Symbiotic nitrogen fixation and soil acidity. In: Robson, A.D., (Ed.), Symbiotic Nitrogen Fixation and Soil Acidity, Academic Press, Sydney, pp. 103–137. Cransberg, L., McFarlane, D.J., 1994. Can perennial pastures provide the basis for a sustainable farming system in southern Australia? New Zealand Journal of Agricultural Research 37, 284–294. Francis, C.M., Gillespie, D.J., 1981. Ecology and distribution of subterranean clover and Medicago species in Sardinia. Plant Introduction Review 13, 15– 25. Gregory, P.J., Hinsinger, P., 1999. New approaches to studying chemical and physical changes in the rhizosphere: an overview. Plant and Soil 211, 1–9. Haynes, R.J., 1990. Active ion uptake and maintenance of cation–anion balance: a critical examination of their role in regulating rhizosphere pH. Plant and Soil 126, 247–264. Hinsinger, P., 1998. How do plant roots acquire mineral nutrients? Chemical processes involved in the rhizosphere. Advances in Agronomy 44, 225– 265.
Hinsinger, P., Gilkes, R.J., 1996. Mobilization of phosphate from phosphate rock and alumina-sorbed phosphate by the roots of ryegrass and clover as related to rhizosphere pH. European Journal of Soil Science 47, 533 –544. Hinsinger, P., Plassard, C., Tang, C., Jaillard, B., 2003. Origins of rootmediated pH changes in the rhizosphere and their responses to environmental constraints: a review. Plant and Soil 248, 43–59. Hirsch, A.M., 1992. Developmental biology of legume nodulation. New Phytologist 122, 211–237. Hoffland, E., Findenegg, G.R., Nelemans, J.A., 1989. Solubilization of rock phosphate by rape. II. Local root exudation of organic acids as a reponse to P-starvation. Plant and Soil 113, 161–165. Howieson, J.G., Ewing, M.A., 1989. Annual species of Medicago differ greatly in their ability to nodulate on acid soils. Australian Journal of Agricultural Research 40, 843 –850. Howieson, J.G., Robson, A.D., Ewing, M.A., 1993. External phosphate and calcium concentrations, and pH, but not the products of rhizobial nodulation genes, affect the attachment of Rhizobium meliloti to roots of annual medics. Soil Biology & Biochemistry 25, 567– 573. Jaillard, B., Ruiz, L., Arvieu, J.C., 1996. pH mapping in transparent gel using color indicator videodensitometry. Plant and Soil 183, 85– 95. Jaillard, B., Plassard, C., Hinsinger, P., 2003. Measurements of Hþ fluxes and concentrations in the rhizosphere. In: Rengel, Z., (Ed.), Measurements of Hþ Fluxes and Concentrations in the Rhizosphere, Marcel Dekker, New York, pp. 231–266. Jarvis, S.C., Robson, A.D., 1983a. A comparison of the cation/anion balance of ten cultivars of Trifolium subteranneum L., and their effects on soil acidity. Plant and Soil 75, 235 –243. Jarvis, S.C., Robson, A.D., 1983b. The effects of nitrogen nutrition of plants on the development of acidity in Western Australian soils. II. Effects of differences in cation/anion balance between plant species grown under non-leaching conditions. Australian Journal of Agricultural Research 34, 355–365. Li, X.-L., George, E., Marschner, H., 1991. Phosphorus depletion and pH increase at the root–soil and hyphae–soil interfaces of VA mycorrhizal white clover fertilized with ammonium. New Phytologist 119, 397 –404. Marschner, H., Ro¨mheld, V., 1983. In vivo measurement of rootinduced pH changes at the soil–root interface: effect of plant species and nitrogen source. Zeitschrift fu¨r Pflanzenphysiologie 111, 241 –251. Marschner, H., Ro¨mheld, V., Ossenberg-Neuhaus, H., 1982. Rapid method for measuring changes in pH and reducing processes along roots of intact plants. Zeitschrift fu¨r Pflanzenphysiologie 105, 407 –416. McLay, C.D.A., Barton, L., Tang, C., 1997. Acidification potential of ten grain legume species grown in nutrient solution. Australian Journal of Agricultural Research 48, 1025–1032. Munns, D.N., 1968. Nodulation of Medicago sativa in solution culture. I. Acid-sensitive steps. Plant and Soil 28, 129 –146. Munns, D.N., 1970. Nodulation of Medicago sativa in solution culture. V. Calcium and pH requirements during infection. Plant and Soil 32, 90 –102. Neumann, G., Ro¨mheld, V., 2001. The release of root exudates as affected by the plant’s physiological status. In: Pinton, R., Varanini, Z., Nannipieri, P. (Eds.), The Release of Root Exudates as Affected by the Plant’s Physiological Status, Marcel Dekker, New York, pp. 41–93. Nye, P.H., 1981. Changes of pH across the rhizosphere induced by roots. Plant and Soil 61, 7–26. Pijnenborg, J.W.M., Lie, T.A., Zehnder, A.J.B., 1990. Nodulation of lucerne (Medicago sativa L.) in an acid soil: pH dynamics in the rhizosphere of seedlings growing in rhizotrons. Plant and Soil 126, 161 –168. Plassard, C., Meslem, M., Souche, G., Jaillard, B., 1999. Localization and quantification of net fluxes of Hþ along maize roots by combined use of
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365 pH-indicator dye videodensitometry and Hþ-selective microelectrodes. Plant and Soil 211, 29 –39. Prikryl, Z., Vancura, V., 1980. Root exudates of plants. VI. Wheat root exudation as dependent on growth, concentration gradient of exudates and the presence of bacteria. Plant and Soil 57, 69– 83. Rayment, G.E., Higginson, F.R., 1992. Australian Laboratory Handbook of Soil and Water Chemical Methods, Inkata Press, Melbourne. Robson, A.D., Loneragan, J.F., 1970. Nodulation and growth of Medicago truncatula on acid soils. II. Colonization of acid soils by Rhizobium meliloti. Australian Journal of Agricultural Research 21, 427–434. Schubert, S., Schubert, E., Mengel, K., 1990. Effect of low pH of the root medium on proton release, growth, and nutrient uptake of field beans (Vicia faba). Plant and Soil 124, 239–244. Tang, C., McLay, C.D.A., Barton, L., 1997. A comparison of proton excretion of twelve pasture legumes grown in nutrient solution. Australian Journal of Experimental Agriculture 37, 563 –570. Tang, C., Barton, L., Raphael, C., 1998. Pasture legume species differ in their capacity to acidify soil. Australian Journal of Agricultural Research 49, 53–58.
1365
Ward, P.R., Dunin, F.X., Micin, S.F., 2000. Water balance of annual and perennial pastures on a duplex soil in a Mediterranean environment. Australian Journal of Agricultural Research 52, 203 –209. Weisenseel, M.H., Dorn, A., Jaffe, L.F., 1979. Natural Hþ currents traverse growing roots and root hairs of barley (Hordeum vulgare L.). Plant Physiology 64, 512–518. Wood, S.M., Newcomb, W., 1988. Nodule morphogenesis: the early infection of alfalfa (Medicago sativa) root hairs by Rhizobium meliloti. Canadian Journal of Botany 67, 3108–3122. Yan, F., Schubert, S., Mengel, K., 1992. Effect of low root medium pH on net proton release, root respiration and root growth of corn (Zea mays L.) and broad bean (Vicia faba L.). Plant Physiology 99, 415–424. Youssef, R.A., Chino, M., 1989. Root-induced changes in the rhizosphere of plants. I. pH changes in relation to bulk soil. Soil Science and Plant Nutrition 35, 461– 468. Yu, Q., Tang, C., Kuo, J., 2000. A critical review on methods to measure apoplastic pH in plants. Plant and Soil 219, 29 –40. Zieschang, H.E., Ko¨hler, K., Sievers, A., 1993. Changing proton concentrations at the surfaces of gravistimulated Phleum roots. Planta 190, 546–554.
Proton release by roots of Medicago murex and Medicago sativa growing in acidic conditions, and implications for rhizosphere pH changes and nodulation at low pH Y. Chenga,b, J.G. Howiesonb,*, G.W. O’Harab, E.L.J. Watkinb,c, G. Souchea, B. Jaillarda, P. Hinsingera a
b
INRA UMR Rhizosphe`re et Symbiose, Place Viala, 34060 Montpellier Cedex 01, France Division of Science and Engineering, Centre for Rhizobium Studies, Murdoch University, South Street, Murdoch Drive, Murdoch, WA 6150, Australia c School of Biomedical Sciences, Curtin University of Technology, G.P.O. Box U1987, Perth, WA 6845, Australia
Abstract Medicago murex nodulates faster and produces more nodules than Medicago sativa in acidic sandy soils. Experiments using a ‘root mat’ approach and videodensitometry examined pH changes in the rhizospheres of nitrate-fed plants of M. murex and M. sativa. Using the ‘root mat’ approach with soil disks of pH 4.49, M. sativa cv. Aquarius acidified its rhizosphere by approximately 0.2– 0.4 pH-units within 4 d, while M. murex cv. Zodiac did not acidify its rhizosphere. Rates of Hþ release were higher from M. sativa than from M. murex. Videodensitometry of roots embedded in agarose of pH 4.5 showed that the mature parts of the tap-root of both species exuded OH2 ions, but was approximately twofold more in M. murex than in M. sativa. Consequently, young parts of the M. sativa rhizosphere were less alkaline than that of M. murex. It is suggested that the difference in nodulation response between the two species at low pH may be related to the different patterns of rhizosphere acidification: the stronger rhizosphere acidification of M. sativa being less favourable for survival and growth of Sinorhizobium medicae. The higher rate of rhizosphere acidification by M. sativa roots may be related to its genetic characteristics including greater relative root growth rate and greater sensitivity to acidity in comparison to M. murex. q 2004 Published by Elsevier Ltd. Keywords: Acid production; Nodulation delay; Proton release; Rhizosphere; Rhizobia; Rhizosphere pH; Root mat; Videodensitometry
1. Introduction Medicago sativa (lucerne; alfalfa) has the capacity to transpire large volumes of water and is well suited for controlling rising water tables in south-western Australia (Cransberg and McFarlane, 1994; Ward et al., 2000). However, its widespread use is currently limited as soils in the region are too acidic (Bettenay and Hingston, 1964) for consistent nodulation with its microsymbiont Sinorhizobium medicae (Munns, 1968, 1970). Medicago spp. differ in their ability to nodulate in acidic soil and the annual species M. murex grows and nodulates well in soil of pH 4.9 (Howieson and Ewing, 1989). However, M. murex is not commercially attractive as a pasture legume and cannot fill the niche of M. sativa in lowering groundwater tables. The nodulation of M. murex and M. sativa is affected differently by soil acidity (Cheng et al., 2002). When grown * Corresponding author. Tel.: þ 61-8-9360-2231; fax: þ61-8-9360-6486. E-mail address: [email protected] (J.G. Howieson). 0038-0717/$ - see front matter q 2004 Published by Elsevier Ltd. doi:10.1016/j.soilbio.2004.04.017
in an acidic sandy soil of pH 4.3, M. murex produced fewer root nodules than when grown in soil of pH 7.0, but these nodules formed at a similar rate. In the same soil, M. sativa also produced fewer nodules when grown in the acidic soil, but the nodules appeared later compared to those on plants grown in soil of pH 7.0 (Cheng et al., 2002). These differences in nodulation response could be due to differential multiplication of S. medicae in the rhizosphere. After 24 d growth in soil of pH 4.3 inoculated with an acidtolerant strain of S. medicae, there were over 100-fold more S. medicae associated with the roots of M. murex than M. sativa (Cheng, unpublished). The ability of different pasture legumes to modify rhizosphere pH varies within and between species (Tang et al., 1997). The average rhizosphere pH of M. sativa seedlings growing in soil of pHH2 O 5.2 increased from pH 5.1 (1 d after sowing) to pH 5.7 (12 d after sowing), but nodulation was not affected (Pijnenborg et al., 1990). Localised changes in rhizosphere pH have also been observed along the roots of M. sativa (Blanchar and Lipton, 1986). In soil of pHH2 O 5.5, the rhizosphere pH of
1358
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
10 d old M. sativa seedlings at the older tap roots was more alkaline (pH 6.8), while the rhizosphere pH at the younger lateral roots was more acidic (pH 4.2) (Blanchar and Lipton, 1986). Modification of rhizosphere pH by M. murex roots has not yet been studied, and M. murex may modify the pH of the rhizosphere differently from M. sativa, thereby developing more favourable conditions for rhizobial survival and growth. The acidifying effect of N2-fixing legumes has been demonstrated in both solution culture (McLay et al., 1997; Tang et al., 1997) and soil (Jarvis and Robson, 1983a,b; Tang et al., 1998). To our knowledge, there is limited research on young plants prior to nodulation where rhizosphere pH changes may affect the pre-infection steps of the nodulation process. As recently reviewed by Hinsinger et al. (2003), root-induced pH changes in the rhizosphere are primarily caused by root and microbial respiration (Nye, 1981), unbalanced uptake of inorganic anions and cations (Haynes, 1990), exudation of organic anions (Hoffland et al., 1989; Neumann and Ro¨mheld, 2001) and oxidation of soil minerals (Hinsinger, 1998). The release of Hþ to counterbalance an excess uptake of cations over anions is often the major source of root-induced changes of rhizosphere pH (Hinsinger et al., 2003). Methods used to measure Hþ concentrations in the rhizosphere, and fluxes of Hþ released by roots include approaches investigating the whole root system or single roots (Gregory and Hinsinger, 1999). At the level of the whole root system, the ‘root mat’ approach has been used to measure pH gradients in the rhizosphere of legumes such as Trifolium repens (white clover) (Li et al., 1991) and T. subterraneum (subterranean clover) (Hinsinger and Gilkes, 1996). Backtitration of the nutrient solution can also measure fluxes of Hþ released by whole root systems in solution (McLay et al., 1997; Tang et al., 1997), but cannot provide localised pH values along a root. This localisation of pH changes along single roots may be semi-quantitatively measured by colorimetry coupled with the agar-indicator technique (Weisenseel et al., 1979), quantitatively by micropotentiometry (Marschner and Ro¨mheld, 1983; Yu et al., 2000), or both of these techniques combined (Pijnenborg et al., 1990). Videodensitometry (Jaillard et al., 1996), based on the principles of colorimetry, was developed to quantitatively measure Hþ flux using coloured pH indicator with agar or agarose gels. It provides maps of pH values along intact roots (Gregory and Hinsinger, 1999; Jaillard et al., 1996, 2003), and simultaneously calculates temporal and spatial profiles of Hþ flux (Plassard et al., 1999). Our aim was to determine whether M. murex and M. sativa induced pH changes in the rhizosphere differentially under acidic conditions, and to test the hypothesis that at low pH the rhizosphere of young parts of the M. murex root are less acidic than those of M. sativa. Relatively young and un-nodulated plants of M. murex and M. sativa were used as there has been limited research on rhizosphere pH changes in young plants prior to nodulation. Seeds of
M. murex cv. Zodiac and M. sativa cv. Aquarius were used in these experiments because the same genotypes had been studied for their nodulation in an acidic soil (Cheng et al., 2002).
2. Materials and methods 2.1. Experimental design Experiment 1 used the ‘root mat’ approach designed by Chaignon et al. (2002) to test changes in soil rhizosphere pH. The experiment consisted of two Medicago species (M. murex cv. Zodiac and M. sativa cv. Aquarius), two pH treatments (initial soil pHCaCl2 of 4.49 with de-ionised water or nutrient solution at pH 4.5, and initial soil pHCaCl2 of 6.71 with de-ionised water or nutrient solution at pH 6.0). Deionised water and nutrient solution were used to compare the responses of the two species in the absence or presence of a cation/anion supply from the solution medium. Six replicates were prepared for each treatment, and one additional replicate was used to assess shoot and root dry weight at the commencement of treatment. Root mats of the two Medicago species were grown in a specially designed cropping device (see Chaignon et al., 2002) in nutrient solution of pH 6.0 for 13 d. Roots were then placed in contact with soil disks. Soil disks of pH 4.49 were placed above solution media of pH 4.5, and soil disks of pH 6.71 were placed above solution media of pH 6.0, for a further 4 d. Control soil disks without root contact were also used as references. Bulk Hþ or OH2 release fluxes were calculated from pH changes in the soil disks between 13 and 17 d after sowing, relative to the soil without root contact. Experiment 2 used the videodensitometry technique designed by Jaillard et al. (1996). M. murex and M. sativa were grown in nutrient solution of pH 6.0 for 13 d. In each of six measurements, three to five plants of each species were embedded in nutrient gel film of pH 4.5 or 6.0 containing pH indicator. Values of proton flux were calculated from images acquired with a video camera. 2.2. Plant and soil material Seeds of M. murex cv. Zodiac and M. sativa cv. Aquarius were surface sterilised in 6% (v/v) H2O2 solution (10 min) to remove seed-borne microorganisms, and then thoroughly rinsed with de-ionised water. The nutrient solution for plant culture contained: 500 mM K2SO4, 200 mM MgSO4, 100 mM FeEDTA, 100 mM Ca(NO3)2, 10 mM KH2PO4, 3 mM H3BO3, 1 mM MnCl2, 30 nM Na2MoO4, 0.75 mM ZnSO4, 0.2 mM CuSO4. The pH was adjusted to 6.0 with 10 mM KOH. The composition of the nutrient solution was based approximately on the chemical properties of the soil used by Cheng et al. (2002). Nitrate was used as the main source of N in the nutrient solution to avoid extreme acidification of the rhizosphere and to prevent N deficiency
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
in plants. Plants were not inoculated with rhizobia. For both experiments, plants were placed in a growth cabinet with a 16 h, 25 8C/8 h, 20 8C day/night regime and a photon fluence rate of 600 mmol photons m22 s21. The soil used in experiment 1 was an acidic sand from Merredin, Western Australia (1188170 E 318290 S), mixed with coarse river sand (3:2) (Cheng et al., 2002). The pH values of 4.49 and 6.71 were achieved by liming the soil mix at two rates (0.12 and 1.5 g kg21). The two rates of lime were added to slightly dampened soil in a cement mixer. After liming, the soil was left then overnight at room temperature in covered plastic tote-boxes. For transportation purposes, the soil was dried at 70 8C for 48 h, and sealed in plastic bags. The soil mix consisted of approximately 88% sand (2.0 – 0.02 mm), and had available N (measured by KCl extraction; Rayment and Higginson, þ 2 1992) of 1 mg kg21 each of NO2 3 and NH4 , and NaHCO3 21 extractable P of 14– 15 mg kg (Colwell, 1963). The soil mix was wet to field capacity with de-ionised water (10%, w/v) and incubated in the dark for 24 h before being used.
1359
mat of roots within 13 d after sowing. Plant culture in this experiment was a two-step procedure. In the first step, root mats were made by growing plants of M. murex and M. sativa in an aerated nutrient solution (pH 6.0; replaced after 7 d) for a period of 13 d (Fig. 1A). In the second step, cropping devices containing 13 d old plants were randomised, taped securely in contact with soil disks (Fig. 1B) and placed over an aerated nutrient solution. Soil disks without plants were covered with an opaque plastic lid to minimise evaporation and algal growth. All soil disks were placed in the growth cabinet for a further 4 d. In experiment 2 using videodensitometry, plants were grown in a similar cropping device as described by Chaignon et al. (2002) for experiment 1, except the lower nylon mesh (30 mm) was not used, and the upper nylon mesh (900 mm) was replaced by a cloth mesh of 900 mm. Roots grew through the cloth mesh into 5 l of an aerated nutrient solution (pH 6.0), which was changed after 7 d, for a period of 13 d. 2.4. Measurements and calculations
2.3. Cropping device and procedure The cropping device for experiment 1 was that described by Chaignon et al. (2002). Soil (2.5 g wet soil) was placed into shallow PVC containers (32 mm i.d.) to produce disks approximately 1.5 mm thick. Soil disks of pH 4.49 were then placed on shelves above 5 l of aerated de-ionised water or nutrient solution of pH 4.5 (adjusted with 10 mM H2SO4), while soil disks of pH 6.71 were placed above 5 l of aerated de-ionised water or nutrient solution of pH 6.0. Soil disks were kept moist by capillary rise through a filter paper strip. To prevent the transfer of Hþ or OH2 ions between different treatments through the solution medium, the six replicates of the same treatment were placed in the same solution medium. Tests found (data not shown), 40 seeds/cropping device of both Medicago spp. were needed to form a dense, planar
2.4.1. Relative root growth rate In experiment 1 at the time of soil –plant contact (13 d after sowing), shoots and roots of plants from one additional replicate were separated, fresh weight recorded, dried at 105 8C for 24 h and re-weighed. After 4 d of treatment (at 17 d after sowing), plants placed on soil disks were removed from the soil disks. The shoots and roots were separated, and fresh and dry weights recorded. Relative root growth rate (RRGR, (mg g21 dry weight) d21) was calculated by the equation RRGR ¼
ðW2 2 W1 Þ W1 £ ðt2 2 t1 Þ
where W1 and W2 are the initial and final dry weight of roots at time t1 and t2 ; respectively. The pH of the wet soil at the commencement of root contact was measured
Fig. 1. Cropping device used to produce root mats of M. murex and M. sativa in a two-step process. (A) Step 1, growth of seedlings in aerated nutrient solution for 13 d; (B) step 2, roots placed in contact with soil disk, and plants cultured in aerated nutrient solution for a further 4 d. Experiment 1.
1360
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
by shaking soil for 1 h in 10 mM CaCl2 (1:5) (Metrohm 744 pH meter, Herisau, Switzerland). The pH of soil disks in contact with plants represents rhizosphere pH. As the moisture content of the soil disks may have changed during the course of the experiment, it was necessary to determine the exact volume of CaCl2 to add to the soil to constitute a 1:5 suspension for pH measurements. This was done by the following procedure. Four soil disks with root contact and four disks without root contact were randomly chosen. The moisture content of 0.5 g of soil from each randomly chosen disk was determined after drying at 105 8C for 1 h. A total of 5 ml of 20 mM CaCl2 was added to the remaining 2 g of soil from each soil disk. Depending on the moisture content of the 0.5 g of soil, a volume of de-ionised water added to the 2 g soil suspended in 20 mM CaCl2 such that its dilution resulted in a 1:5 suspension in 10 mM CaCl2. The soil suspension was then shaken for 1 h and pH measured. 2.4.2. Hþ flux To estimate Hþ flux released by the plant roots, pH buffer curves of the soils were first determined. Soil of both low and high pH were wet to 10% moisture (w/v) with either 10 mM CaCl2 or solutions of 10 mM CaCl2 containing various concentrations of HCl or NaOH. The concentrations of HCl and NaOH ranged from 0.4 to 8.0 mM. Soil was incubated in the dark at 22 8C for 3 d, suspended in 10 mM CaCl2 (1:5), shaken for 1 h and the pH measured. The pH buffer capacity of soil was calculated from the slope (b; in mmol (g soil21) (pH unit)21) of the linear regression of the amount of HCl or NaOH added versus the soil pH (Tang et al., 1998). The soil pH were 4.49 and 6.71 (10 mM CaCl2). The b values were 2 8.1 (g soil21) (pH unit)21 (soil of pH 4.49, r 2 ¼ 0:96), and 2 9.6 mmol (g soil21) (pH unit)21 (soil of pH 6.71, r 2 ¼ 0:98). Amount of proton released was calculated from total acid production, estimated as the difference in acidity between soils with plants and control soils without plants. Thus, flux of proton jH þ (nmol (g root)21 s21) released by roots was calculated as jHþ ¼
½bðpH0 2 pHÞ £ Msoil ðMroot £ tÞ
where pH0 is the pH of soil disks without plants; pH, pH of soil disk with plants; Msoil ; mass of the soil disk; Mroot ; root dry weight, averaged from plants at the beginning of soil – plant contact, and from plants 4 d after soil –plant contact; t is the time duration of soil – plant contact in seconds (4 d in this experiment, i.e. 345,600 s). In experiment 2, after 13 d of growth in nutrient solution of pH 6.0, plants were equilibrated in fresh nutrient solution. This was done 48 h before embedding in nutrient agarose gel. Plants for the pH 4.5 treatment were placed in fresh nutrient solution of pH 5.2 for 24 h, and then transferred to nutrient solution of pH 4.5, 24 h before embedding. This intermediate step decreased acid-shock and subsequent necrosis of root tips. Plants for the pH 6.0
treatment were transferred to fresh nutrient solution of pH 6.0, 24 h before embedding. By the time roots were embedded, plants were 15 d old. A separate volume (100 ml) of each nutrient solution (pH 4.5 and 6.0) that was used for equilibrating plants was also placed in the growth cabinet for 24 h before embedding. These solutions were used for embedding plants. Solutions of pH indicators bromocresol green (pK ¼ 4:69 for measurements at pH 4.5) and bromocresol purple (pK ¼ 6:40 for measurements at pH 6.0) were prepared first. Films were prepared by melting 1 g of agarose powder (Fluka Ref 05068, Buchs, Switzerland; see Calba et al., 1996), in 20 ml of pH indicator solution and 80 ml of the nutrient solution used for plant culture (pH 4.5 or 6.0), yielding a final pH indicator concentration of 90 mg l21. The agarose solutions were boiled for 30 min, cooled to 37 8C in a water bath and pH readjusted to 4.5 or 6.0 with 10 mM KOH or H2SO4. Plants were gently removed from the cloth mesh so not to damage roots, roots blotted dry with tissue, and carefully placed between two glass sheets (100 £ 200 mm2) separated by a 1 mm thick strip of PVC. The agarose solution was then syringed between the glass sheets. After approximately 5 min when the agarose solution had cooled and set, the upper glass sheet was removed and a 3 mm PVC strip was placed between the glass sheets so as not to confine the roots, and to enable respired CO2 to diffuse into the atmosphere. Plants were placed in the growth cabinet and images were acquired by videodensitometry 2 h after embedding. Two saturated calibration standards (pH 3.5 and 8.0 with bromocresol green, and pH 4.8 and 8.5 with bromocresol purple) were prepared for each series of measurements. Three to five plants of each species per treatment were embedded and measured in six separate experiments. Images were acquired using the methods described by Plassard et al. (1999). Hþ flux profile were calculated at three defined locations along the root: 3, 20 and 50 mm from the root tip. These locations correspond with the emerging root hair, growing root hair and mature root hair zones of the root, respectively, with decreasing susceptibility to infection by rhizobia further away from the root tip (Bhuvaneswari et al., 1981; Wood and Newcomb, 1988). 2.5. Statistical analysis Variability in the measured values among the various replicates is indicated by standard deviations. Significant differences between treatments were deduced from an analysis of variance (ANOVA) using Statistica for Windows 1999 (Kernal Release 5.5, StatSoft, Inc., Tulsa, Oklahoma, USA), and a t-test using Microsoft Excel 2000 (Microsoft Corporation, Washington, USA) at P , 0:05:
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
1361
Table 1 Root dry weights (mean value of six replicates ^ standard deviation) of Medicago murex and M. sativa cultured in de-ionised water or nutrient solution at pH 4.5 and 6.0 for 4 d, and their relative root growth rate (RRGR) between 13 and 17 d. Experiment 1 Solution media
pH
Root dry weight (mg plant21), 17 d after sowing
RRGR (mg g21 dry weight d21), 13–17 d after sowinga
M. murex
M. sativa
M. murex
M. sativa
Water
4.5 6.0
614 ^ 354 732 ^ 379
248 ^ 140 380 ^ 118
1.84 ^ 0.60 2.13 ^ 0.58
2.16 ^ 0.61 2.54 ^ 0.48
Nutrient solution
4.5 6.0
510 ^ 96 780 ^ 261
380 ^ 094 602 ^ 619
1.09 ^ 0.56* 1.94 ^ 0.77
2.60 ^ 0.31 2.65 ^ 0.70
RRGR ¼
ðW2 2 W1 Þ W1 £ ðt2 2 t1 Þ
where W1 and W2 are the initial and final dry weight of roots at time t1 and t2, respectively. *Significant difference in RRGR between M. murex and M. sativa at P , 0:05: a After 13 d of growth in nutrient solution of pH 6.0, root dry weight (mg plant21) for M. murex was 360 (^193), and for M. sativa was 128 (^67).
3. Results 3.1. Root growth In experiment 1, after 13 d of growth in nutrient solution of pH 6.0, root dry weight (mg plant21) for M. murex was 360 (^ 193), and for M. sativa was 128 (^ 67). The greater initial weight of M. murex seeds probably contributed to the greater root dry weight of M. murex compared to M. sativa at 13 d after sowing. Between 13 and 17 d after sowing, RRGR was generally higher in M. sativa than M. murex, despite M. sativa having lower root dry weights in all treatments (Table 1). In nutrient solution of pH 4.5, RRGR was significantly higher (approximately twice) in M. sativa than M. murex. 3.2. Soil pH and Hþ release For soil disks of pHCaCl2 4.49 moistened with de-ionised water or nutrient solution of pH 4.5, the pH of soil in contact with M. murex roots did not differ from the pH of the control
soil, while the pH of the soil in contact with M. sativa roots was significantly lower (Table 2). With soil of pHCaCl2 6.71 moistened with de-ionised water of pH 6.0, both Medicago spp. decreased soil pH. With soil of pHCaCl2 6.71 moistened with nutrient solution of pH 6.0, M. murex roots decreased soil pH while M. sativa roots did not significantly change soil pH relative to the control soil. For soil disks of pHCaCl2 6.71 without root contact, increases in pH after moistening for 4 d with de-ionised water (pH increased by 0.52 pH-unit) and nutrient solution (pH increased by 0.37 pH-unit) (Table 2) may be due to the short time (overnight) in which lime equilibrated in the soil before the limed soil was dried. For soil disks of pHCaCl2 4.49 moistened with nutrient solution of pH 4.5, proton release was significantly higher from M. sativa than M. murex (Table 2), with the rate of proton release from M. sativa being 1.9 nmol (g root)21 s21 compared to M. murex at 2 0.1 nmol (g root)21 s21. The negative value indicates net OH2 release from M. murex. In soil of pHCaCl2 6.71, there was no significant difference in proton release fluxes between the two species. Except for
Table 2 Soil pH and proton extrusion rates (mean value of six replicates ^ standard deviation) from roots of Medicago murex and M. sativa cultured in rhizosphere soil moistened with water or nutrient solution between 13 and 17 d after planting. Experiment 1 Solution Initial pH of soil disk and solution Mean control soil pH Mean rhizosphere soil pH
M. murex
Ratio M. sativa/M. murex
No plants
M. murex
4.49/4.5 6.71/6.0
4.49 ^ 0.02b 7.23 ^ 0.05c
4.45 ^ 0.08b 4.31 ^ 0.07a 6.55 ^ 0.19a 6.70 ^ 0.02b
Nutrient 4.49/4.5 6.71/6.0
4.49 ^ 0.05b 7.08 ^ 0.17b
4.51 ^ 0.04b 4.14 ^ 0.15a 20.1 ^ 0.1* 1.9 ^ 0.9 221.0 5.95 ^ 0.50a 6.78 ^ 0.04b 3.2 ^ 1.3 1.8 ^ 0.9 0.6
Water
M. sativa
Proton extrusion ratea (nmol (g root21 s21)
0.1 ^ 0.2 2.6 ^ 2.3
M. sativa 0.1 ^ 0.0 3.4 ^ 0.8
1.0 1.5
Mean values with different letters within a line are significantly different at P , 0:05: *Significant difference in proton extrusion rate between M. murex and M. sativa at P , 0:05: a Negative value represents net OH2 extrusion.
1362
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
Table 3 Proton efflux rates (mean value ^ standard deviation) from roots of 15 d-old Medicago murex and M. sativa 2 h after embedding in nutrient agarose gel of pH 4.5 and 6.0 containing 100 mM NO2 3 . Experiment 2 PH
Tap-root, distance from root tip (mm)
Proton efflux (pmole (m root)21 s21)a M. murex
M. sativa
Ratio M. murex/M. sativa
4.5
50 20 3
285 ^ 65a 297 ^ 46a* 25 ^ 37b
261 ^ 19a 248 ^ 22a 23 ^ 36b
1.39 2.02 1.67
6.0
50 20 3
28 ^ 3ab 25 ^ 7b 211 ^ 5a*
25 ^ 3a 24 ^ 4a 26 ^ 6a
1.60 1.25 1.83
Mean values with different letters within a column are significantly different at P , 0:05: Mean values from 7 to 11 plants (M. murex), and 8 and 7 plants (M. sativa) for pH 4.5 and 6.0, respectively. *Significant difference in proton efflux between M. murex and M. sativa at P , 0:05: a Negative values represent net OH2 efflux.
M. murex grown in nutrient solution of pH 4.5 releasing OH2, Hþ release was generally higher with plants grown in nutrient solution than in de-ionised water. In experiment 2, at 15 d after sowing, roots of both M. murex and M. sativa released OH2 (Table 3), with M. murex consistently exhibiting larger efflux than M. sativa. The rate of OH2 efflux was significantly higher (approximately twice) in M. murex than M. sativa at 20 mm from the root tip (at pH 4.5), and at 3 mm from the root tip (at pH 6.0). There was no significant difference in OH2 efflux rates between the two species at the other locations along the root. At pH 4.5, the highest rate of OH2 efflux occurred at 20 and 50 mm from the root tip of both species (Table 3). At pH 6.0, M. murex had a lower rate of OH2 efflux at 20 mm, compared to 3 and 50 mm from the root tip. In contrast, there was no significant difference in rates of OH2 efflux at the three locations of the M. sativa root.
4. Discussion 4.1. Changes in rhizosphere pH as affected by Medicago species Roots of M. sativa released more protons than those of M. murex into a sandy soil of pHCaCl2 4.49, while into a nutrient agarose gel of pH 4.5, M. sativa released fewer OH2 ions compared to M. murex. As a result, the rhizosphere of M. sativa was consistently more acidic than that surrounding M. murex. The differential ability of these species to acidify the root medium may be genetic, relating to species adaptation and tolerance to low pH. M. murex is commonly associated with acidic soils (Francis and Gillespie, 1981), nodulates consistently at low pH (Howieson and Ewing, 1989), and is more tolerant of low pH than other annual Medicago spp. (M.A. Ewing, unpub, PhD thesis, University of Western Australia, 1991). Differences in the rate of Hþ release between the two genotypes of M. murex and M. sativa may also be related to
their different relative growth rates of root growth. M. sativa had a greater relative rate of root growth and a higher rate of Hþ release compared to M. murex, which had a slower relative rate of root growth, released less Hþ, and acidified the rhizosphere less than M. sativa. In a study comparing Al-tolerant and Al-sensitive genotypes of Zea mays (maize), Calba et al. (1996) showed similarly that the genotype which grew the fastest, also exhibited the largest nutrient requirements, and hence released the largest fluxes of protons. Different growth rates also affected rhizosphere pH changes in different genotypes of T. subterraneum (Jarvis and Robson, 1983a). Root exudation may be directly associated with root growth since there was almost no root exudation from plants where almost no root growth was observed even though shoots were actively growing (Prikryl and Vancura, 1980). M. murex and M. sativa induced pH changes in the rhizosphere differentially, especially in younger parts of roots where the rhizosphere of M. murex was less acidic than that of M. sativa. In acidic soils where survival and growth of S. medicae (Coventry and Evans, 1989; Robson and Loneragan, 1970), and attachment to host roots (Caetano-Anolle´s et al., 1989; Howieson et al., 1993) are already challenged by low pH, root-induced acidification of the rhizosphere will further inhibit the nodulation process. The more acidic rhizosphere of M. sativa would certainly be less favourable for saprophytic growth and colonisation by S. medicae. Consequently, rhizobial populations of S. medicae on the roots of M. sativa may take longer to reach a sufficient population size to initiate infection (Hirsch, 1992). Results from our study support the hypothesis that slower rates of rhizobial growth caused the delayed nodule initiation and development in M. sativa at low pH (Cheng et al., 2002). Rhizosphere acidification may be an important factor contributing to this. It would be instructive to determine whether other genotypes of M. murex and M. sativa also induce similar pH changes in the rhizosphere. Rhizosphere alkalinisation by seedlings of M. sativa growing in an acidic soil (Pijnenborg et al., 1990)
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
compared to rhizosphere acidification by the same species observed in experiment 1 of our study was probably due to soil properties contributing to different nutrient uptake and hence pH changes. Pijnenborg et al. (1990) suggested that root-induced rhizosphere alkalinisation by M. sativa roots did not result in the formation of nodules on M. sativa since the 0.6 pH-unit increase occurred after the acid-sensitive period of the nodulation process. However, an equivalent decrease in pH would certainly affect nodulation in an acidic soil, especially during the acid-sensitive stages of rhizobial survival, growth and colonisation in the rhizosphere. For plants relying on N2-fixation, the acidity produced by roots during plant growth largely persists in the soil after plants have been removed and may lead to subsoil acidification (Tang et al., 1998). 4.2. Changes in rhizosphere pH as affected by growth conditions Alkalinisation of rhizosphere soil when cultured at low pH, and acidification of rhizosphere soil when cultured at higher pH as displayed by M. murex, has been observed in Hordeum vulgare (barley) and Glycine max (soybean) (Youssef and Chino, 1989) and in Vicia faba (faba bean) (Schubert et al., 1990). The reason for this ‘plasticity’ in response to external pH is unclear, but may be related to nutrient availability, and the subsequent imbalance of cation – anion uptake. In addition, the amount of Hþ released by roots of both genotypes of Medicago spp. was higher in plants grown at higher pH than those grown under acidic conditions. Furthermore, the higher overall rate of ion exchange may result from higher nutrient uptake when plants were less acid-stressed. Decreased plant growth caused by low pH in acid-sensitive species has been related to a build-up of Hþ ions in the cytoplasm, because of a lack of Hþ release (Schubert et al., 1990), or an influx of protons into root cells because of higher membrane permeability (Yan et al., 1992). This might have occurred to a larger extent for M. sativa than for M. murex in our experiment. Compared to plants grown in soil moistened with de-ionised water, plants grown in soil moistened with nutrient solution are likely to have higher rates of nutrient uptake due to access to nutrients from both the soil disk and nutrient solution. This is consistent with the larger proton release fluxes observed in these plants relative to those grown in soil moistened with de-ionised water. Finally, the availability of nutrients (which is likely to vary considerably from soil to soil) will have a considerable influence on both the direction (efflux or influx) and intensity of Hþ fluxes. Such considerations can help in understanding why plants grown in experiment 1 on soil disks decreased rhizosphere pH while plants of a similar age and past nutrient status exhibited net OH2 efflux when grown in nutrient solution. Further work is needed to estimate whether the observed differences
1363
between these and other genotypes of M. murex and M. sativa are consistently found in a range of nutrient þ solutions of varying compositions, especially NO2 3 /NH4 ratios, and in a range of acidic soils. In this respect, it should be stressed that plants were only relying on NO2 3 in experiment 2 (NHþ 4 was not supplied in the agarose gel) and hence exhibited OH2 efflux, while they might have derived some NHþ 4 from the soil in experiment 1, thereby resulting in rhizosphere acidification. 4.3. Changes in rhizosphere pH as affected by the location along the root axis Depending on plant species and N source, but also Fe and P status of the plant, root-induced changes in rhizosphere pH can vary considerably along the root (Marschner et al., 1982; Marschner and Ro¨mheld, 1983). Higher acidification at the root tips compared to basal parts of roots has often been reported, e.g. in Phleum pratense (timothy) (Zieschang et al., 1993) and Z. mays (Marschner and Ro¨mheld, 1983). In our study, the root apex did not behave differently at pH 6.0 (no significant differences were found with other locations along the root axis, see Table 3). However, at pH 4.5, root tips of both genotypes of Medicago spp. exhibited significantly lower efflux of OH2 and were thus more acidic than older regions of the roots. Different rates of ion release in various regions of a root may be related to age of roots cells, and hence affect the uneven pattern of cation and anion uptake along the root. It is possible that older parts of the root take up more NO2 3 , while younger parts take up more cations (Blanchar and Lipton, 1986). This would need to be verified by the means of electrophysiological measurements, i.e. micropotentiometry with NO2 3 selective microelectrodes (Plassard et al., 1999; Jaillard et al., 2003). Susceptibility of root hairs to infection by rhizobia decreases with root hair age. The highest frequency of susceptible root hairs is in the younger region of the root, although some root hairs are susceptible to infection in the more mature region (Bhuvaneswari et al., 1981). At low pH, the more mature regions of roots of both Medicago species had higher rates of OH2 release, were less acidic, and therefore more favourable for rhizobial survival and growth than younger parts of the root. This may explain why nodules formed lower down the tap-root of M. murex when grown in soil of pH 4.3 compared to plants grown in pH 7.0 (Cheng et al., 2002). The faster rate of root growth in M. sativa meant that rhizobia accumulated and attached further down the tap-root than M. murex, and this species variation may have contributed to the different location of the uppermost nodule even in soil of pH 7.0 (Cheng et al., 2002). However, at pH 4.3, rhizobia may have also taken longer to reach sufficient numbers to initiate infection, resulting in delayed nodule development in M. sativa.
1364
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365
Acknowledgements This research was done at the Institut National de la Recherche Agronomique, Montpellier, France, funded by a French Government Scientific Fellowship (French Embassy, Canberra, Australia), and a travel grant from the Australian Grains Research and Development Corporation awarded to Yvonne Cheng. We thank the staff and students at the Laboratoire de Science du Sol for their support, especially Nicole Balsera for help with preparing the cropping device for plant culture. Dr Caixian Tang is acknowledged for helpful discussion on the manuscript. We also thank Regina Carr from Murdoch University for help with soil preparation. This research forms part of a PhD thesis by Yvonne Cheng, funded by the Australian Research Council’s SPIRT Program.
References Bettenay, E., Hingston, F.J., 1964. Development and distribution of soils in the Merredin area, Western Australia. Australian Journal of Soil Science 2, 173–186. Bhuvaneswari, T.V., Bhagwat, A.A., Bauer, W.D., 1981. Transient susceptibility of root cells in four common legumes to nodulation by rhizobia. Plant Physiology 68, 1144–1149. Blanchar, R.W., Lipton, D.S., 1986. The pe and pH in alfalfa seedling rhizosphere. Agronomy Journal 78, 216 –218. Caetano-Anolle´s, G., Lagares, A., Favelukes, G., 1989. Adsorption of Rhizobium meliloti to alfalfa roots: dependence on divalent cations and pH. Plant and Soil 117, 67–74. Calba, H., Jaillard, B., Fallavier, P., Arvieu, J.C., 1996. Agarose as a suitable substrate for use in the study of Al dynamics in the rhizosphere. Plant and Soil 178, 67–74. Chaignon, V., Di Malta, D., Hinsinger, P., 2002. Fe-deficiency increases Cu acquisition by wheat cropped in a Cu-contaminated vineyard. New Phytologist 154, 121 –130. Cheng, Y., Watkin, E.L.J., O’Hara, G.W., Howieson, J.G., 2002. Medicago sativa and Medicago murex differ in the nodulation response to soil acidity. Plant and Soil 238, 31–39. Colwell, J.C., 1963. The estimation of the phosphorus fertilizer requirements of wheat in southern New South Wales by soil analysis. Australian Journal of Experimental Agricultural and Animal Husbandry 3, 190 –198. Coventry, D.R., Evans, J., 1989. Symbiotic nitrogen fixation and soil acidity. In: Robson, A.D., (Ed.), Symbiotic Nitrogen Fixation and Soil Acidity, Academic Press, Sydney, pp. 103–137. Cransberg, L., McFarlane, D.J., 1994. Can perennial pastures provide the basis for a sustainable farming system in southern Australia? New Zealand Journal of Agricultural Research 37, 284–294. Francis, C.M., Gillespie, D.J., 1981. Ecology and distribution of subterranean clover and Medicago species in Sardinia. Plant Introduction Review 13, 15– 25. Gregory, P.J., Hinsinger, P., 1999. New approaches to studying chemical and physical changes in the rhizosphere: an overview. Plant and Soil 211, 1–9. Haynes, R.J., 1990. Active ion uptake and maintenance of cation–anion balance: a critical examination of their role in regulating rhizosphere pH. Plant and Soil 126, 247–264. Hinsinger, P., 1998. How do plant roots acquire mineral nutrients? Chemical processes involved in the rhizosphere. Advances in Agronomy 44, 225– 265.
Hinsinger, P., Gilkes, R.J., 1996. Mobilization of phosphate from phosphate rock and alumina-sorbed phosphate by the roots of ryegrass and clover as related to rhizosphere pH. European Journal of Soil Science 47, 533 –544. Hinsinger, P., Plassard, C., Tang, C., Jaillard, B., 2003. Origins of rootmediated pH changes in the rhizosphere and their responses to environmental constraints: a review. Plant and Soil 248, 43–59. Hirsch, A.M., 1992. Developmental biology of legume nodulation. New Phytologist 122, 211–237. Hoffland, E., Findenegg, G.R., Nelemans, J.A., 1989. Solubilization of rock phosphate by rape. II. Local root exudation of organic acids as a reponse to P-starvation. Plant and Soil 113, 161–165. Howieson, J.G., Ewing, M.A., 1989. Annual species of Medicago differ greatly in their ability to nodulate on acid soils. Australian Journal of Agricultural Research 40, 843 –850. Howieson, J.G., Robson, A.D., Ewing, M.A., 1993. External phosphate and calcium concentrations, and pH, but not the products of rhizobial nodulation genes, affect the attachment of Rhizobium meliloti to roots of annual medics. Soil Biology & Biochemistry 25, 567– 573. Jaillard, B., Ruiz, L., Arvieu, J.C., 1996. pH mapping in transparent gel using color indicator videodensitometry. Plant and Soil 183, 85– 95. Jaillard, B., Plassard, C., Hinsinger, P., 2003. Measurements of Hþ fluxes and concentrations in the rhizosphere. In: Rengel, Z., (Ed.), Measurements of Hþ Fluxes and Concentrations in the Rhizosphere, Marcel Dekker, New York, pp. 231–266. Jarvis, S.C., Robson, A.D., 1983a. A comparison of the cation/anion balance of ten cultivars of Trifolium subteranneum L., and their effects on soil acidity. Plant and Soil 75, 235 –243. Jarvis, S.C., Robson, A.D., 1983b. The effects of nitrogen nutrition of plants on the development of acidity in Western Australian soils. II. Effects of differences in cation/anion balance between plant species grown under non-leaching conditions. Australian Journal of Agricultural Research 34, 355–365. Li, X.-L., George, E., Marschner, H., 1991. Phosphorus depletion and pH increase at the root–soil and hyphae–soil interfaces of VA mycorrhizal white clover fertilized with ammonium. New Phytologist 119, 397 –404. Marschner, H., Ro¨mheld, V., 1983. In vivo measurement of rootinduced pH changes at the soil–root interface: effect of plant species and nitrogen source. Zeitschrift fu¨r Pflanzenphysiologie 111, 241 –251. Marschner, H., Ro¨mheld, V., Ossenberg-Neuhaus, H., 1982. Rapid method for measuring changes in pH and reducing processes along roots of intact plants. Zeitschrift fu¨r Pflanzenphysiologie 105, 407 –416. McLay, C.D.A., Barton, L., Tang, C., 1997. Acidification potential of ten grain legume species grown in nutrient solution. Australian Journal of Agricultural Research 48, 1025–1032. Munns, D.N., 1968. Nodulation of Medicago sativa in solution culture. I. Acid-sensitive steps. Plant and Soil 28, 129 –146. Munns, D.N., 1970. Nodulation of Medicago sativa in solution culture. V. Calcium and pH requirements during infection. Plant and Soil 32, 90 –102. Neumann, G., Ro¨mheld, V., 2001. The release of root exudates as affected by the plant’s physiological status. In: Pinton, R., Varanini, Z., Nannipieri, P. (Eds.), The Release of Root Exudates as Affected by the Plant’s Physiological Status, Marcel Dekker, New York, pp. 41–93. Nye, P.H., 1981. Changes of pH across the rhizosphere induced by roots. Plant and Soil 61, 7–26. Pijnenborg, J.W.M., Lie, T.A., Zehnder, A.J.B., 1990. Nodulation of lucerne (Medicago sativa L.) in an acid soil: pH dynamics in the rhizosphere of seedlings growing in rhizotrons. Plant and Soil 126, 161 –168. Plassard, C., Meslem, M., Souche, G., Jaillard, B., 1999. Localization and quantification of net fluxes of Hþ along maize roots by combined use of
Y. Cheng et al. / Soil Biology & Biochemistry 36 (2004) 1357–1365 pH-indicator dye videodensitometry and Hþ-selective microelectrodes. Plant and Soil 211, 29 –39. Prikryl, Z., Vancura, V., 1980. Root exudates of plants. VI. Wheat root exudation as dependent on growth, concentration gradient of exudates and the presence of bacteria. Plant and Soil 57, 69– 83. Rayment, G.E., Higginson, F.R., 1992. Australian Laboratory Handbook of Soil and Water Chemical Methods, Inkata Press, Melbourne. Robson, A.D., Loneragan, J.F., 1970. Nodulation and growth of Medicago truncatula on acid soils. II. Colonization of acid soils by Rhizobium meliloti. Australian Journal of Agricultural Research 21, 427–434. Schubert, S., Schubert, E., Mengel, K., 1990. Effect of low pH of the root medium on proton release, growth, and nutrient uptake of field beans (Vicia faba). Plant and Soil 124, 239–244. Tang, C., McLay, C.D.A., Barton, L., 1997. A comparison of proton excretion of twelve pasture legumes grown in nutrient solution. Australian Journal of Experimental Agriculture 37, 563 –570. Tang, C., Barton, L., Raphael, C., 1998. Pasture legume species differ in their capacity to acidify soil. Australian Journal of Agricultural Research 49, 53–58.
1365
Ward, P.R., Dunin, F.X., Micin, S.F., 2000. Water balance of annual and perennial pastures on a duplex soil in a Mediterranean environment. Australian Journal of Agricultural Research 52, 203 –209. Weisenseel, M.H., Dorn, A., Jaffe, L.F., 1979. Natural Hþ currents traverse growing roots and root hairs of barley (Hordeum vulgare L.). Plant Physiology 64, 512–518. Wood, S.M., Newcomb, W., 1988. Nodule morphogenesis: the early infection of alfalfa (Medicago sativa) root hairs by Rhizobium meliloti. Canadian Journal of Botany 67, 3108–3122. Yan, F., Schubert, S., Mengel, K., 1992. Effect of low root medium pH on net proton release, root respiration and root growth of corn (Zea mays L.) and broad bean (Vicia faba L.). Plant Physiology 99, 415–424. Youssef, R.A., Chino, M., 1989. Root-induced changes in the rhizosphere of plants. I. pH changes in relation to bulk soil. Soil Science and Plant Nutrition 35, 461– 468. Yu, Q., Tang, C., Kuo, J., 2000. A critical review on methods to measure apoplastic pH in plants. Plant and Soil 219, 29 –40. Zieschang, H.E., Ko¨hler, K., Sievers, A., 1993. Changing proton concentrations at the surfaces of gravistimulated Phleum roots. Planta 190, 546–554.