Introduction
Soil pH is a major factor affecting plant growth, development, and reproductive output (Gentili et al., Reference Gentili, Ambrosini, Montagnani, Caronni and Citterio2018). Lupins have been domesticated independently as a crop (food, green manure, and forage) in both the Mediterranean region and the Andes of South America (Gladstones, Reference Gladstones1970). Their successful use in modern agriculture is based on their ability to grow on infertile, sandy, well-drained, and neutral to acidic soils (Lambers et al., Reference Lambers, Clements and Nelson2013). However, lupins are generally intolerant to alkaline soil (Tang et al., Reference Tang, Buirchell, Longnecker and Robson1993; Tang and Robson, Reference Tang, Robson, Longnecker and Buirchell1995). Interestingly, cultivars of Lupinus albus in Egypt are tolerant of calcareous soils (Christiansen et al., Reference Christiansen, Raza and Ortiz1999) and have tolerance to limed soils similar to the level of tolerance of the calcareous soil-tolerant Lupinus pilosus (Kerley et al., Reference Kerley, Norgaard, Leach, Christiansen, Huyghe and Römer2002). The high soil pH affects plant establishment and growth, since it affects the availability and uptake of mineral elements, especially phosphorus (P) (Lambers et al., Reference Lambers, Chapin and Pons2008).
Lupins have the ability to release carboxylates (mainly, citrate and malate) that mobilize poorly available soil P (Dinkelaker et al., Reference Dinkelaker, Römheld and Marschner1989; Gardner et al., Reference Gardner, Barber and Parbery1983). The release of carboxylates in L. albus L. has been extensively researched (Kidd et al., Reference Kidd, Ryan, Haling, Lambers, Sandral, Yang, Culvenor, Cawthray, Stefanski and Simpson2016; Pearse et al., Reference Pearse, Veneklaas, Cawthray, Bolland and Lambers2007; Zhu et al., Reference Zhu, Yan, Zörb and Schubert2005); however, there have been limited studies on the effect of soil alkalinity on plant growth, such as Ding et al. (Reference Ding, Clode and Lambers2019), or on the role of seed size in alkalinity tolerance in lupin cultivars. Large-seeded seedlings may perform better in tolerance to abiotic stress. For example, Benard and Toft (Reference Benard and Toft2007) reported that large-seeded seedlings of the sand dune plant Ericameria nauseosa perform better than small-seeded seedlings during summer drought in an ecosystem near Mono Lake, California. In this study, we hypothesized that two cultivars of L. albus L. (an Egyptian cultivar P27734, with larger seeds, and an Australian cultivar Kiev, with smaller seeds) differ in their physiological responses when grown at different soil pH. Testing this hypothesis should provide critical insights on how white lupin tolerates high soil pH. Therefore, the main objectives of the following experiment were to compare plant growth, root mass ratio, amount of rhizosheath carboxylates, and carboxylate composition of two cultivars of L. albus L. – an Egyptian cultivar, P27734, and an Australian cultivar, Kiev Mutant grown in sand culture with three pH treatments (5.1, 6.7, and 7.8), thus placing the results into a wider perspective relative to the Australian cultivar.
Material and Methods
Experiment 1: Assessing differences between seedlings of two cultivars
To test the germination rate of 2 cultivars of L. albus L. (white lupin), 10 seeds of each cultivar were planted in sand with neutral pH (6.7), without addition of nutrient solution. All seeds were obtained from the Australian Lupin Collection located at the Department of Primary industries and Regional Development, Western Australia. Germinated seeds were scored every 2 days until 14 days after sowing when seedlings were harvested. At harvest, roots were washed carefully. The number of germinated seeds, seedling height, root depth, and fresh and dry weight of shoots and roots were measured.
Experiment 2: Effects of pH levels on two L. albus L. cultivars
A glasshouse experiment was a combination of two L. albus L. cultivars (Egyptian cultivar P27734 and Australian cultivar Kiev Mutant); there were seven replicates in each of three pH treatments (5.1, 6.7, and 7.8, determined in 0.01-mM CaCl2). Target pH values were obtained by mixing 10 g CaCO3 and 0.008 g NaOH kg−1 river sand for the pH 7.8 treatment, 0.6 g FeSO4 kg−1 river sand for the pH 5.1 treatment; the unamended river sand had a pH of 6.7. Seeds of the two cultivars were surface sterilized in 10% (v/v) sodium hypochlorite for 15 min, rinsed thoroughly with deionized water, and germinated on water-moistened filter paper in the dark for 3 days. Three germinated seedlings were planted in each pot and thinned to one plant after 1 week. All seedlings were inoculated with the appropriate strain of rhizobium bacteria (New-Edge Microbials Pty. Ltd., North Albury, NSW, Australia). The experiment was carried out in a temperature-controlled glasshouse at the University of Western Australia (Perth, WA, Australia) with an average daytime temperature of 20 °C and an average nighttime temperature of 15 °C and natural daylight. Pots (8.5 × 8.5 × 18 cm) were filled with 1.2-kg washed river sand; all nutrients were provided with final concentrations being (in µg g−1 dry soil): N 30, S 50, Ca 33, Mg 10, Cu 0.5, Zn 2, Mn 4, B0.119, Mo 0.4, Fe 5, Cl 23, and P 50 (P source was KH2PO4). The germination started on 10 February 2017 and harvesting was on 18 April 2017. The pots were watered to 70% of field capacity by weighing every second day for the first 6 weeks and then daily until the end of the experiment. The plants were harvested 8 weeks after sowing when visual differences in plant size among the pH treatments could be observed.
Plant measurements
At harvest, the plants were removed from the pots, and the root systems were shaken gently. The sand that was attached to the roots was defined as the rhizosheath (Pang et al., Reference Pang, Ryan, Siddique and Simpson2017). Rhizosheath extracts for the analyses of carboxylates were collected according to Pang et al. (Reference Pang, Ryan, Tibbett, Cawthray, Siddique, Bolland, Denton and Lambers2010). The entire root system was submerged in 0.2-mM CaCl2 and shaken to remove the rhizosheath sand. The volume of CaCl2 needed to cover the whole root system varied according to the root size from 20 to 100 ml. Subsamples of the extracts were taken and filtered through a 0.22-μm syringe filter. One millimeter of the filtered extract was transferred into a 1-ml high-performance liquid chromatography (HPLC) vial. HPLC analysis of the extracts was performed as described by Cawthray (Reference Cawthray2003) using working standards of malic, malonic, lactic, acetic, maleic, citric, succinic, cis-aconitic, and trans-aconitic acid (ICN Biomedicals Inc., Aurora, OH, USA) to identify and quantify carboxylates.
Immediately after collection of carboxylates, plants were separated into leaves (mature and young), stems, roots, and flowers (if present). Total plant leaf area was measured using a leaf area meter (LI-3100; LI-COR, Lincoln, NE, USA). Subsequently, cleaned roots were scanned with a transparent adapted tray (acrylic box) positioned in the scanner (Epson Expression Scan 1680, Long Beach, ON, Canada). Roots were spread out in the acrylic box (size A4) containing tap water to minimize the number of overlaps. Based on the scanned images of the root systems, total length, mean diameter, and total surface area were measured using the WinRHIZO™ regular V.2009 software (Regent Instruments Inc., Quebec, QC, Canada). Plant parts were dried at 70 °C for 72 h, and dry mass (DM), i.e., total (DMtot), leaf (DMl), root (DMr), shoot (DMsh) and stem (DMst) DM, was measured. Root mass ratio was calculated as the ratio of root DM to total plant DM.
To determine the leaf P concentrations, dried leaf samples were ground to a fine powder using Geno/Grinder 2010 (Spex SamplePrep, Metuchen, NJ, USA). Weighed subsamples of approximately 200 mg were digested using a hot concentrated nitric-perchloric (3:1) acid mixture. Phosphorus concentration in mature leaves was determined by the malachite green method (Motomizu et al., Reference Motomizu, Wakimoto and Toei1983) using a UV-VIS spectrophotometer (Shimadzu Corporation, Kyoto, Japan) analyzed by inductively coupled plasma optical emission spectrometry (ICP-OES; ChemCentre, Perth, WA, Australia).
Statistical analysis
Generalized least square models were used (Zuur et al., Reference Zuur, Ieno, Walker, Saveliev and Smith2009) to test the responses of the two cultivars (Egyptian cultivar P27734 and Australian cultivar Kiev Mutant) at three soil pH levels and interaction between these two factors. Cultivars and pH were used as nonrandom/‘fixed’ factors of plant functional groups. Normality of data was checked, and transformations of log10 were undertaken if required. Statistical analyses were performed in ‘R’ using the ‘NLME’ package (Pinheiro et al., Reference Pinheiro, Bates, DebRoy and Sarkar2017). All figures present the cultivar × pH interaction (mean ± confidence intervals). If this interaction was significant (p < 0.0001), then post hoc Tukey tests were also performed (Hothorn et al., Reference Hothorn, Bretz and Westfall2008). If there was no significant interaction, while pH or cultivar had a significant effect (p < 0.0001), then marginal means for pH are presented in the results. t-test was performed to assess differences between cultivars’ seeds and seedlings.
Results
Shoot and root growth response
DMtot, DMst, DMsh, and DMr were significantly affected by soil pH, while mature leaf (DMl) was significantly different between cultivars, and mature leaf area, mean root diameter, and shoot P concentration were significantly affected by cultivar × pH interaction (Table 1). The Egyptian cultivar produced more DM in all pH treatments, especially at pH 7.8 (Figure 1, Supplementary Figure S1 available online at https://doi.org/10.1017/S0014479719000383). Increasing soil pH resulted in increased DMtot for both cultivars (p ≤ 0.0001), with values approximately 1.4- and 1.3-fold higher for the Egyptian and the Australian cultivars DM at alkaline than at acidic pH, respectively (Figure 1A). Similar to the total DM, DMsh (p ≤ 0.0001, Figure 1B) and DMst (Figure 1C) also differed significantly under different soil pH. The two cultivars differed significantly in DMl, with significantly higher values at alkaline and acidic pH for P27734 than for Kiev Mutant (p ≤ 0.0001), while no difference was observed between the two cultivars at neutral pH (Figure 1D). For mean root diameter, a significant interaction between cultivar and pH was found (Table 1, Figure 1E). The mean root diameter differed significantly between cultivars under each pH treatment; that of P27734 in the alkaline treatment was similar to that in the acidic pH treatment, whereas that of Kiev Mutant was 18% greater in the alkaline treatment than in the acidic treatment (Figure 1E).
Significance of characters under three pH levels (5.1, 6.7, and 8.3) in two Lupinus albus cultivars (P27734 and Kiev Mutant) and cultivar × pH interaction (detected by generalized least square models)
DM, dry mass; ns, nonsignificant difference.
The marginal means are presented here when there was no significant interaction of pH × cultivar.
*** p ≤ 0.0001.
Biomass of two Lupinus albus cultivars (Egyptian genotype P27734 and Australian cv. Kiev Mutant) shown at Y-axis grown at soil pH of 5.1, 6.7, and 7.8 shown at X-axis. Bars are the means, and error bars are the standard errors of the means, derived from generalized least square models (A) total dry mass (DM), (B) shoot DM, (C) stem DM (pH, p ≤ 0.0001); (D) mature leaf DM (Cultivar, p ≤ 0.0001), and (E) mean root diameter (Cultivar × pH interaction, p ≤ 0.0001). Different letters among cultivars indicate significant differences within each panel (post hoc Turkey test, p ≤ 0.05). (L. albus Kiev: L. albus Kiev Mutant).
The alkaline treatment caused a minor shift in the total leaf area of both cultivars; there was no cultivar and pH interaction and no significant difference was found between cultivars or pH treatments (Supplementary Figure S2). Root surface area, specific root length, and root mass fraction were not significantly affected by soil pH, cultivar, or cultivar × pH interaction (p > 0.05) (Supplementary Figure S3).
Rhizosheath pH and carboxylate response
Both cultivars showed the greatest amount of exuded malate and citrate per root DM at pH 7.8 (p < 0.001; Table 1, Figure 2.). The proportion of exuded carboxylates varied greatly among pH treatments, being ∼45–77% of malate for P27734 and 30–73% for Kiev Mutant. Citrate comprised 20–55% for P27734 and 25–60% for Kiev Mutant, with values of >18 µmol g−1 root DM (Supplementary Figure S4). The proportion of both fumarate and cis-aconitate was <1% for both cultivars (Supplementary Figure S4).
The most abundant carboxylates: (a) malate and (b) citrate relative to root dry mass (DM) of two Lupinus albus cultivars (Egyptian genotype P27734 and Australian cv. Kiev Mutant) shown at Y-axis grown at soil pH of 5.1, 6.7 and 7.8 shown at X-axis. Bars are the means, and error bars are the standard errors of the means, derived from generalized least square models (pH, p ≤ 0.0001). Different letters among cultivars indicate significant differences within each panel (post hoc Turkey test, p ≤ 0.05).
The pH of the rhizosheath soil extracted with 0.2-mM CaCl2 varied significantly among pH treatments; for P27734 at a bulk soil pH of 7.8, the rhizosheath pH was 6.7, whereas Kiev Mutant lowered the pH to only 7.1 (Supplementary Figure S5). The rhizosheath pH did not differ between the two cultivars (Table 1, Supplementary Figure S5).
Leaf phosphorus concentrations
For shoot P concentration, a significant interaction of cultivar × pH treatment was found (Table 1, Supplementary Figure S6). It showed a reduction by 9.8% for P27734 and 43% for Kiev Mutant cultivars due to the alkaline treatment compared with the acidic treatment. The leaf P concentration was significantly reduced by 33% for P27734, and by 40% for Kiev Mutant at pH 7.8 compared with pH 5.1 and 6.7 (Table 1, Supplementary Figure S6).
Seed characters and assessing differences between cultivars’ seedlings
Seed P content of P27734 was less than that of Kiev Mutant (2.4 and 2.7 mg P seed−1, respectively) (Table 2). The seed size of P27734 cultivar was larger than that of Kiev Mutant cultivar, being 0.32 and 0.24 g seed−1, respectively (p < 0.0001; Table 2). Root DMs of 14-day-old seedlings grown in river sand (neutral pH 6.7) were 0.13 and 0.11 g plant−1; total DMs were 0.46 and 0.33 g plant−1 for P27734 and Kiev Mutant cultivars, respectively. The seedling height showed a major difference with 8.75 cm for P27734 and 3.50 cm for Kiev Mutant. The germination percentages of P27734 and Kiev Mutant cultivars on the second day of germination were 100 and 70%, respectively; the larger seeds of the Egyptian cultivar germinated faster than the smaller ones of the Australian cultivar.
Seed and seedling characters of two Lupinus albus cultivars (P27734 and Kiev Mutant) at 14 days after germination grown in a glasshouse
DM, dry mass.
Data are means ± SE (n = 10).
Discussion
Most lupin species prefer acidic to neutral soils, with pH 4.5–7.5 (Huyghe, Reference Huyghe1997; Jansen, Reference Jansen2006; Lim, Reference Lim2012; Nigussie, Reference Nigussie2012; Wolko et al., Reference Wolko, Clements, Naganowska, Nelson, Yang and Kole2011). However, cultivars of L. albus L. in Egypt are tolerant of calcareous soils when cultivated in soils with pH 7.5 to 9.4 (Christiansen et al., Reference Christiansen, Raza and Ortiz1999). In Australia, some researchers reported that L. albus L. grows relatively well in alkaline soil (Tang et al., Reference Tang, Buirchell, Longnecker and Robson1993).
The most important difference between the two investigated cultivars was that the alkaline-tolerant Egyptian cultivar had larger seeds and faster germination. This difference alone accounted for its greater biomass production at pH 7.8 than that of Kiev Mutant. There were interesting physiological responses to soil pH, but these were essentially the same for both cultivars. Physiological traits that are involved in a plant’s tolerance to abiotic stress, that is, alkalinity stress (pH = 7.8), of both cultivars were similar, suggesting that the Australian cultivar had an alkalinity tolerance similar to that of the Egyptian cultivar (P27734). Of all the analysed physiological traits, soil pH significantly affected rhizosheath carboxylates, mainly citrate and malate.
Our results show that the Egyptian cultivar had larger seeds than the Australian cultivar and germinated more quickly. Two weeks after germination, the Egyptian cultivar had a greater seedling height, so the difference in seed size was likely a factor of the greater alkalinity tolerance of the Egyptian cultivar. The seed size gives an indication of the amount of reserves in the embryo which allow it to establish itself independent of resources to be acquired from the environment (Lambers et al., Reference Lambers, Chapin and Pons2008). Typically, small-seeded species are more affected by stress than large-seeded ones (Hewitt, Reference Hewitt1998; Leishman and Westoby, Reference Leishman and Westoby1994; Osunkoya et al., Reference Osunkoya, Ash, Hopkins and Graham1994). Liu and Tang (Reference Liu and Tang1999) compared several white lupin genotypes and noted that poor growth under alkaline conditions was associated with small-seeded plants; they also suggested that seed size might contribute to alkalinity tolerance.
The Australian cultivar exhibited a greater shoot P concentration when grown under neutral pH, while the Egyptian cultivar did so at the acidic pH. Moir et al. (Reference Moir, Jordan, Moot and Lucas2016) found that as the increase of lime in soil increases the phosphorus in plant tissue and that agree with our results in which both had the greatest DM at alkaline pH 7.8. The P concentrations for both cultivars were above the critical P concentration in the shoot; this critical P concentration, defined as the leaf concentration required to gaining 90% of the maximum grain yield, is 2.2 mg P g–1 DM for L. albus (Bolland, Reference Bolland1997). This supports the notion that both cultivars are equally alkaline-tolerant. A greater ability to produce DM combined with a lower leaf P concentration at pH 7.8 may partly explain why the Egyptian cultivar is considered alkaline-tolerant in southern Egypt.
Interestingly, the specific root length was lower at pH 7.8 for both cultivars than that at pH 5.1 and 6.7, while there was no significant difference in root mass ratio due to pH treatments for both cultivars. The observed slight increase in root surface area at alkaline conditions may contribute to higher P-uptake efficiency, suggesting that the P availability was sufficient for both cultivars (Brouwer, Reference Brouwer1963, Reference Brouwer1983; Kemp and Blair, Reference Kemp and Blair1994; Niklas, Reference Niklas1994). However, our results are inconsistent with studies that reported Kiev Mutant as alkaline sensitive. Tang et al. (Reference Tang, Buirchell, Longnecker and Robson1993, Reference Tang and Robson1995) compared the growth of lupin genotypes in alkaline soils and found that L. albus was as sensitive as Lupinus angustifolius. Tang and Thomson (Reference Tang and Thomson1996) investigated a range of legume adaptations to soil pH and biocarbonate and found that the growth of L. angustifolius and L. albus L. is sensitive to pH ≥ 6. They found that the most alkaline-tolerant lupin species is L. pilosus.
Lupins have the ability to release carboxylates (mainly, citrate and malate) that mobilize poorly available soil P (Dinkelaker et al., Reference Dinkelaker, Römheld and Marschner1989; Gardner et al., Reference Gardner, Barber and Parbery1983). The greatest carboxylate release for both cultivars was at pH 7.8; at this soil pH, P and several micronutrients are less available for the plant than at a neutral pH (Lambers et al., Reference Lambers, Chapin and Pons2008). Our finding that the greatest release of rhizosheath carboxylates was detected at pH 7.8 for both lupin cultivars is consistent with the study of Pearse et al. (Reference Pearse, Veneklaas, Cawthray, Bolland and Lambers2006b), who reported that Lupinus species acidified their rhizosheath; acidification enhances the solubility of acid-soluble Ca-phosphates (Dinkelaker et al., Reference Dinkelaker, Römheld and Marschner1989; Gahoonia et al., Reference Gahoonia, Claassen and Jungk1992; Gerke et al., Reference Gerke, Römer and Jungk1994). To acquire micronutrients, L. albus L. depends on carboxylate release and cluster-root production, instead of a large root system (Johnson et al., Reference Johnson, Allan, Vance and Weiblen1996; Pearse et al., Reference Pearse, Veneklaas, Cawthray, Bolland and Lambers2007). Both cultivars released carboxylate at pH 7.8, mainly citrate and malate; carboxylates increase not only the availability of P, but also that of Fe and Mn (Johnson et al., Reference Johnson, Allan and Vance1994). The large proportion of citrate in rhizosheath carboxylates agrees with previous reports on L. albus L. (e.g., Dinkelaker et al. Reference Dinkelaker, Römheld and Marschner1989; Gardner et al. Reference Gardner, Barber and Parbery1983; Kidd et al. Reference Kidd, Ryan, Haling, Lambers, Sandral, Yang, Culvenor, Cawthray, Stefanski and Simpson2016; Pearse et al. Reference Pearse, Veneklaas, Cawthray, Bolland and Lambers2006b). Malate comprises a significant fraction of exudates in L. albus L. as well (Neumann et al., Reference Neumann, Massonneau, Martinoia and Römheld1999; Pearse et al., Reference Pearse, Veneklaas, Cawthray, Bolland and Lambers2006a). Both citrate and malate are important for nutrient acquisition at alkaline pH for both cultivars. Jones and Darrah (Reference Jones and Darrah1994) demonstrated that both citrate and malate are capable of mobilizing calcium, potassium, and magnesium from alkaline soils.
Our results of the germination test show that the Egyptian cultivar had larger seeds than the Australian cultivar and germinated more quickly. Two weeks after germination, the Egyptian cultivar had a greater seedling height. Liu and Tang (Reference Liu and Tang1999) compared several white lupin genotypes and noted that poor growth under alkaline conditions was associated with small-seeded plants. They suggested that seed size might contribute to alkalinity tolerance. The results from this study fully support this suggestion, and the physiological responses of both cultivars were the same at the alkaline pH. Therefore, our study shows that both cultivars are equally alkaline-tolerant.
Concluding Remarks
Dry matter production of both cultivars was greatest at alkaline pH. From a physiological perspective, the Australian cultivar was equally alkaline-tolerant as the Egyptian cultivar. Our findings show no physiological differences in traits that might account for alkalinity tolerance. A pivotal difference between the cultivars was seed size and initial seedling weight, and that difference accounted entirely for the lime tolerance of the Egyptian cultivar. Future field tests should assess the effect of alkalinity on the growth of the cultivars under field conditions.
Supplementary materials
For supplementary material for this article, please visit https://doi.org/10.1017/S0014479719000383
Acknowledgements
The first author gratefully acknowledges the financial support provided by The Egyptian Mission Sector that supported OMA with funding to conduct this work at the University of Western Australia. The authors gratefully thank lab colleagues Wenli Ding, Patrick Hayes, and Rodrigo Pires for helping with statistical analyses. Thanks to Michael Smirk for assisting with ICP-OES analysis, and to Greg Cawthray for assisting with HPLC analyses. Deep thanks for editor of Experimental Agriculture and anonymous reviewers who improve the manuscript to be in this good form.
Financial support
None.
