Mineral Composition of Red Clover under Rhizobium Inoculation and Lime Application in Acid Soil

In the present study the effects of Rhizobium inoculation and lime application on the mineral composition (N, P, K, Ca, Mg, Fe, Mn, Cu, Zn, B) of red clover (Trifolium pratense L.), in very acid soil were evaluated. Inoculation with Rhizobium leguminosarum bv. trifolii significantly increased shoot dry weight (SDW) of red clover plants (three times greater), as well as N, Mg, Fe, Mn and Cu contents in plants compared to the control. Application of lime and Rhizobium together, depending on the lime rate (3, 6 or 9 t ha of lime) and the cut, increased SDW significantly, but decreased the contents of N, P, K, Mg, Mn, Zn and B in plants. Regardless of the changes, in all treatments in both cuts, contents of N, K, Ca, Mg, Mn and Zn in plants were among sufficiency levels (Mg content was elevated in the second cut), while Fe content was mainly high, as well as Cu (in the second cut).


Introduction
Red clover (Trifolium pratense L.) is one of the most important perennial forage legumes owing to its high biomass yield and quality of forage (Vasiljević et al., 2010).In addition, it can tolerate soils of poorer quality, slightly higher acidity and shallow soils.Red clover can establish a nitrogen-fixing symbiosis with Rhizobium leguminosarum bv.trifolii and fix atmospheric nitrogen of the benefit to the plant in the wide range of 76-334 kg N ha -1 annually (Taylor and Quesenberry, 1996).For red clover, N values derived from the atmosphere went up above 90% of the total plant N (Warembourg et al., 1997;Thilakarathna et al., 2012).Consequently, red clover can satisfy most of its nitrogen needs and also contributes to the incorporation of N in soil, helping to reduce the application of mineral N fertilizers.Some studies indicated that rhizobia, besides N2 fixation, may have positive effect on certain macro and micronutrient contents in leguminous plants and soil and overall plant growth (Bambara and Ndakidemi, 2010;Ndakidemi et al., 2011).Rhizobial strains enhanced the uptake of many macro and micronutrients in peanut nodules and seeds (Howell, 1987), soybean (Tairo and Ndakidemi, 2014a,b) and common bean plants (Makoi et al., 2013;Ndakidemi et al., 2011).Inoculating chickpea with Mesorhizobium mediterraneum also increased the levels of K, Ca, Mg, P and N in plant parts (Peix et al., 2001).Belane et al. (2014) concluded that rhizobial strains possessing high nitrogen-fixing ability induce mineral accumulation in young cowpea plants (Vigna unguiculata L. Walp.).Rhizobia strains may facilitate the availability and uptake of certain nutrients such as P, Fe, Zn (Jin et al., 2006;Ndakidemi et al., 2011).Mechanisms of rhizobial action include: production of organic acids which convert insoluble phosphorus compounds to more soluble forms, production of siderophores that chelate iron and make it available to the plant, and the production of phytohormones (auxin, etc.) which change root architecture and favour nutrient uptake (Antoun et al., 1998).Therefore the use of rhizobia to increase soil fertility and improve growth and yield of important crops such as red clover could be an important alternative to chemical fertilizers in sustainable agriculture.
Besides nutrient availability, low soil pH is one of the most limiting factors for plant production.Acid soils take up around one third of the soil worldwide, while in Serbia, around 60% of arable land is of acid reaction (Sikiric et al., 2011).Soil pH of 5.5 is considered the lower limit for most forage legumes.In acid soil, plant growth was limited due to toxic concentrations of H, Al, Mn and Fe and deficiency of N, P, Ca, Mg, B and Mo (Von Uexküll and Mutert, 1995).The sensitivity of legume to acid soil is also the result of soil pH influence on the rhizobial bacteria.However, some effective R. trifolii strains can tolerate quite low organic C and N were determined with a elemental CNS analyser, Vario model EL III (Hanau, Germany).Available forms of Mn, Fe, Zn and Cu in soils were extracted by DTPA (Soltanpour et al., 1996) and determined by plasma emission spectrometry ICAP 6300 ICP (Cambridge, UK).The B contents in soil and plant was determined by colorimetry (curcumin method) after dry digestion (Dible et al., 1954).
For the determination of the plant K, P, Mg, Ca, Mn, Fe, Zn and Cu, leaves were burned to ash at 550°C and acid digestion with HCl was performed according to Chapman and Pratt (1961).Phosphorus was measured by the colorimetric ammonium vanadate method and K by flame photometry (Egner et al., 1960).Ca and Mg contents in plant samples were determined by atomic absorption spectroscopy after Chapman and Pratt (1961).Total plant N was determined with elemental CNS analyzer.Mn, Fe, Zn and Cu contents in plants were determined by ICAP 6300 ICP.

Statistical analysis
The effect of the treatments was evaluated using analysis of variance (SPSS 16.0 program, 2007), and significant differences between means were tested by Duncan's multiple range test.

Changes in soil properties
Rhizobium inoculation did not significantly change soil pH(H2O) or content of available macronutrients P, K, Ca and Mg in soil (Table 2).The application of lime increased soil pH significantly and the optimal soil pH range of 6-6.5 for red clover cultivation (Fairey, 1988) was reached with 3 t ha -1 of lime.Soil was at the medium level of total N supply in all treatments and there were no differences between treatments.In lime treatments NH4 + increased (½ NPK was added in these treatments), while NO3 - concentration decreased in all treatments (except NPK) (Table 2).Lime application increased initially very low soil P content to the medium, but did not change medium content of available K in soil.After liming, the Ca content in soil increased from medium to high, while high available Mg content decreased significantly only with 9 t ha -1 of lime (Mg content still remained high).In the treatment with zeolite, pH, K and Ca contents in soil increased significantly, while Mg content decreased.
In this research, high content of micronutrients Fe, Mn, Cu and Zn (medium) in control soil did not change significantly under Rhizobium inoculation, while low B content increased to the medium level.Lime application decreased Fe, Mn, Cu (except 9 t ha -1 ), Zn and B contents in soil, but Fe and Cu contents were still high.In the treatment with zeolite, Fe, Mn, and Zn decreased in soil, while B increased.

Yield of red clover
Inoculation with Rhizobium increased shoot dry weight (SDW), of red clover plants in both cuts in respect to the control Ø (Table 4).In the second cut, SDW was three times greater compared to the control and did not differ from the treatments with 3 t ha -1 of lime, nor with zeolite, but was higher than in NPK treatment.All lime rates together with inoculation increased SDW in both cuts due to an immediate rise in pH of very acid soil (Table 2), with no extra effect of the highest lime rate 9 t ha -1 on SDW increase or soil pH.The maximum yield in the first cut was obtained with the application of 3 t ha -1 lime, while in the second cut it was achieved with 6 t h -1 lime.soil pH (4.9) (Lindstrom and Millyniemi,l987).Red clover grows well on moderately acid soils, but a pH above 6.0 and adequate Ca is required for maximum yields (Fairey, 1988).Therefore, to obtain high and quality red clover yield, the application of lime is desirable.In contrast, the application of lime causes the decrease of Al, Mn and Fe toxic concentrations, but can also result in the lack of certain micronutrients (Fageria et al., 1995).
Although there are some researches showing the potential of Rhizobium inoculation to increase red clover yield in acid soils, little is known about Rhizobium effect on the nutrient accumulation in red clover plants.Therefore, the aim of the this work was to evaluate the effect of Rhizobium inoculation and lime on yield and macro and micronutrient contents (N, P, K, Ca, Mg, Fe, Mn, Cu, Zn, B) of red clover grown in an acid soil.

Soil and experiment design
To set up the experiment, top soil layer (30 cm) of an acid Pseudogley soil (FAO, 1998) (44°41'38''N and 19°39'10''E) from Varna village (near Valjevo, Serbia), with the following chemical characteristics (Table 1) was used.Each pot was filled with 2 kg of air-dried soil and 20 seeds of red clover were planted, and after two weeks thinned to 10 plants.The experiment was observed for a year, with 6 different treatments and one control with three repetitions: inoculation (R. leguminosarum bv.trifolii strain 459), inoculation and addition of 3, 6 or 9 t ha -1 of CaCO3 and NPK1/2 fertilization, inoculation and zeolite application, NPK fertilization without inoculation (NPK), control Ø (no inoculation and no fertilization).

Soil and plant samples analyses
Plant samples were taken and analysed in two cuts, 9 and 15 weeks after the experiment establishment, while soil samples were analysed 15 weeks after the experiment establishment.
Soil pH was determined with a glass electrode pH meter in 1N KCl and in H2O (in ratio soil: KCl or H2O 1:2.5).Available N forms in the soil (NH4 + , NO3 -) were determined by steam distillation method described by Bremner (1965).Available P and K in soil were determined by the AL-method of Egner-Riehm (Egner et al., 1960)

Mineral composition of red clover plants
The content of all macro and micronutrients (% of dry weight and mg kg -1 , respectively) varied significantly between treatments, and it was either decreased or increased in respect to control (Table 4 and 5).In contrast to nutrient contents, the total nutrient uptake of all macro and micronutrients (mg plant - 1 , data not shown) in all treatments was significantly increased in respect to control as the result of high shoot dry weight increase.

Macronutrient contents in red clover plants
In all treatments in both cuts, macronutrient contents, N, P, K, Ca and Mg in red clover plants (Table 4), were above the critical level for macronutrient deficiency (Fairey, 1988), and mainly between adequate ranges for red clover (Bergmann, 1992), except low P content and elevated Mg content in some treatments.
Nitrogen content in plants varied between 2.67 and 4.48% of dry matter and it was in the range (23.4-47.0g kg -1 dry matter) published by Spedding and Diekmahns (1972).In the first cut, the highest N content in plants was in the treatment with Rhizobium (4.48%), due to nitrogen fixation, while in the second cut it was in NPK treatment (4.2%), as a result of N mineral fertilization (Table 4).Nitrogen content in plants was in negative correlation with SDW in both cuts (r = -0.610;r = -0.887).In all treatments in the second cut (except NPK), N content was lower compared to the control.However, these treatments showed higher SDW and total N uptake in plants (mg plant -1 ) was always higher in respect to control, indicating N2 fixation in inoculated plants.
According to Bergmann (1992), P content in red clover of 0.17-0.27%was low (0.3-0.6% adequate range), but among or close to the values given by Fairey (1988) (0.2-0.4%), and above the critical level of 0.15%.Phosphorus content in plants in the first cut decreased in the Rhizobium treatment compared to control, while in the second cut, P content significantly decreased in all treatments (except in NPK treatment), regardless of the P increase in the soil in lime treatments (Table 2).
Content of K in plants was optimal (1.8-3%; Bergmann, 1992) except in the first cut in plants with NPK fertilization (3.6%) and control (3.1),where it was high.In the treatments with Rhizobium and zeolite + Rhizobium, there were no changes in K content in plants despite the K increase in soil.Lime   In all treatments Ca content was optimal, between 1-2% comparing to Bergmann (1992) findings.In the first cut in all treatments, including lime treatments, Ca content significantly decreased compared to the control, while there were no differences in the second cut.
Higher content of Mg in both cuts, considering adequate range being between 0.25-0.60%(Bergmann, 1992), in all treatments (except zeolite) was detected (Table 4).In the first cut, in the treatments with Rhizobium as well as with lime, Mg content significantly increased compared to control.However, Mg content was lower in lime treatments in respect to Rhizobium alone.In treatment with zeolite + Rhizobium, Ca and Mg contents were significantly lower compared to other treatments.
In general, the macronutrient contents did not follow the increase in SDW (plant growth) in most of the treatments.Moreover, there were negative correlations between contents of some macronutrients and SDW in one or both cuts: N and K in both cuts, P and Ca in the first cuts (data not shown).

Micronutrient contents in red clover plants
The high Fe contents in plants, above 250 mg kg -1 , or even toxic, >500 mg kg -1 , (Bergmann, 1992), were obtained mainly in the second cut in all treatments (Table 5).The most significant changes in Fe content were under Rhizobium inoculation, where Fe content increased 2.4 to 3.4 fold in respect to control in two cuts.In the treatments with lime and Rhizobium, Fe content was reduced compared to Rhizobium, but not completely and without a clear lime rate effect, in contrast to Fe content reduction in soil.
In all treatments in both cuts, Mn (except in Rhizobium treatment), and Zn contents in plants were among adequate ranges (35-100; 25-70 mg kg -1 respectively, Bergmann, 1992), while Cu content was elevated in the second cut (7-15 mg kg -1 , Bergmann, 1992) (Table 5).The B content was above the critical level (15 mg kg -1 ; Fairey, 1988) in all treatments, and in the first cut at sufficient level (20-60 mg kg -1 ; Neubert et al., 1970), but in the second cut it was somewhat lower.Rhizobium treatment increased Mn and Cu contents (in the first cut), but did not influence Zn and B contents in plants, compared to the control.Depending on the cut and the lime rate applied, in some treatments with lime and Rhizobium together, a decrease was noted in Mn, Cu, Zn and B contents (Table 5).

Discussion
Soil microorganisms are very important in the biogeochemical cycles of inorganic and organic nutrients in the 557 soil and in the maintenance of soil health and quality (Jeffries et al., 2003).It has been reported that plant growth promoting rhizobacteria including rhizobia can influence the chemistry of soil nutrients in many ways and enhance nutrient uptake by plants (Dey et al., 2004;Saharan and Nehra, 2001).Rhizobium inoculation increased the soil pH and the availability of Ca and Na (but not K and P) and micronutrients (Fe, Cu, Zn and Mn) in the rhizosphere of common bean (Bambara and Ndakidemi, 2010).In the presented study Rhizobium inoculation did not change soil pH or content of available macro and micronutrients, except the B content in the red clover rhizosphere.These differences can be the result of a quite higher initial soil pH (6.2 in CaCl2) in the previous study, compared to our research (pHKCl 4.38 or pHH2O 5.1).The changes in soil nutrient contents after liming were similar to the results published (Sikiric et al., 2011;Bailey, 1995;Dugalić et al., 2006).
Increased dry matter of red clover by Rhizobium inoculation and/or lime application obtained in this study corresponded to the previous researches (Jarak et al., 2006;Stevović et al., 2007).The positive effects of liming usually occur through amelioration of Al and Mn toxicity and/or alleviation of Ca deficiency (Haynes and Naidu, 1998).Lime-induced positive response in dry matter yield was greater and more immediate at sites where pH was below 5.3 (Fystro and Bakken, 2005) which was also the case in our study.Although higher yield was obtained with the application of 6 t h -1 lime and Rhizobium, it is cheaper to use inoculation only with satisfactory yield, particularly since the good lime effects fade in the second and third year after its application (Stevović et al., 2007).
As previously mentioned, Rhizobium inoculation increased nutrient uptake (P, K, Mg, S, Na, Fe, Cu, Zn, Mn and B) in numerous leguminous species; peanut, common bean, cowpea, chickpea and soybean (Howell 1987;Peix et al., 2001;Makoi et al., 2013;Ndakidemi et al., 2011;Belane et al., 2014;Tairo and Ndakidemi, 2014a).In these studies the total nutrient uptake in plants (mg plant -1 ) were mainly considered, but in the study by Belane et al. (2014) the nutrient concentrations were also increased after inoculation.In our study, the total nutrient uptake of all macro and microelements was also higher in all treatments in respect to control.
On the other hand, the inoculation with Rhizobium significantly increased contents of only two macronutrients in the first cut, N and Mg (compared to the control); contents of K and Ca did not change under inoculation, while P content even decreased in both cuts.Application of lime and Rhizobium together decreased the contents of P, K and Mg in plants (in respect to control or Rhizobium alone).The increased N content in plants (% of dry weight) in the first cut is the result of N2 fixation, and the total N content in plants (mg plant -1 ) of all treatments in both cuts was higher compared to control.Increased plant growth and high dry matter production (3 fold higher over control) caused the dilution effect of N and its decrease in plants of the treatments with higher SDW (Jarrell and Beverly, 1981;Timmer, 1991).Consequently, N content is negatively correlated with SDW, which was also found for some other nutrients in this and previous studies (Timmer, 1991;Imo, 2012).
After liming, the available P content in soil increased, since the increase in pH influenced the release of one part of bound P from Al-and Fe-phosphate in acid soils (Dugalić et al., 2006;Sikiric et al., 2011).Despite this increase, the P content in plants in the second cut decreased.Fageria et al. (1995) reported that increasing levels of lime tended to reduce P uptake, while lower P content in leaves could be caused by the antagonism between P and Ca (Jaravan and Poldma, 2004).In addition, as ryegrass developed and dry mass accumulation increased, phosphorus tissue content decreased due to the dilution effect (Mazza et al., 2012).Therefore the P fertilization should be considered in red clover cultivation in very acid soils.
Lime application in both cuts, decreased K content in plants, possibly due to the antagonism in Ca and K uptake from soil solution (Jakobsen, 1993;Weis, 2009) and significant increase in SDW.
The unexpected decrease in Ca in the first cut in lime treatments was probably influenced by the short period to the first cut, and very rapid plant growth, together with relatively slow Ca uptake and transport (Reid and Horvath, 1980;Sonneveld and Voogt, 2009).The lack of differences in Ca concentration in plants between limed and unlimed plots was also noticed in some researches (Fystro and Bakken, 2005;Viade et al., 2011).
Higher contents of Mg in both cuts could be the result of high Mg content in soil and the tendency of red clover to uptake Mg easily (Loide, 2004).In the first cut, in the treatments with Rhizobium as well as with lime, Mg content increased compared to control.However, Mg content was lower in lime treatments in respect to Rhizobium alone, which could be the result of the antagonism between Ca and Mg (Jakobsen, 1993).
In treatment with zeolite + Rhizobium, Ca and Mg contents were significantly lower compared to other treatments, possibly due to basic zeolite cations (Ca, Mg, Na and K) and the antagonistic relationship in Ca/K and Mg/K uptake, since high K content in soil was detected.
Rhizobium treatment increased Fe, Mn and Cu contents (in the first cut), but did not influence Zn and B contents in plants, compared to the control.The most visible increase was in Fe content.Significant increase of Fe after Rhizobium inoculation in common bean and red clover was detected in previous studies and it is considered that microbes may favour plant Fe acquisition by producing siderophores and auxins (Jin et al., 2006;Ndakidemi et al., 2011).Siderophores are chelators with a high affinity for Fe 3+ , and many Fe 3+ -siderophore chelates are source of available Fe for plants, while auxin generation triggers the enhancement of Fe uptake capacity in plants, by enhancing root acquisition area and the activity of ferric chelate reductase (Jin et al., 2006;Jin et al., 2014).The R. leguminosarum bv.trifolii strain 459 used in this study apart from nitrogen-fixing ability, showed significant auxin production, as well as phosphate solubilization in in vitro conditions, however it is difficult to connect these traits with the changes induced in red clover nutrient contents without additional researches.In our study, in the treatments with lime and Rhizobium, Fe content was reduced compared to Rhizobium, but not completely and without clear lime rate effect, in contrast to Fe content reduction in soil.In previous studies, a correlation between overall Fe content in plants and Fe concentration in the soil was not established (DeKock, 1981).
Depending on the cut and the lime rate applied, in some treatments with lime and Rhizobium together, a decrease was noted in Mn, Cu, Zn and B contents.This decrease in micronutrient contents (Mn, Cu, Zn and B), in plants after liming, is in agreement with previous reports (Fageria et al., 1995;Lalljee and Facknath, 2001).The decrease in micronutrient contents was the consequence of pH increase after liming or zeolite application, which decreases solubility of micronutrients in soil and the antagonism between particular ions (Ca and Mn) (Fageria et al., 1995).

Conclusions
Inoculation of red clover with Rhizobium increased plant growth in very acid soil, but lime application together with Rhizobium provided higher shoot dry weight.The uptake of all nutrients (N, P, K, Ca, Mg, Fe, Mn, Cu, Zn, B) in plants increased with Rhizobium and lime applications, while the contents of particular macro and micronutrients depended on the treatment applied, and were mainly in the adequate ranges for red clover.The P content in plants remained low, despite its increase in soil after lime application.In addition, content of B in plants was somewhat lower, although it was the only nutrient increased in soil by Rhizobium inoculation.In contrast, Rhizobium inoculation induced the biggest increase in Fe content, which was increased few times.To achieve proper growth and acceptable mineral composition of red clover under acid soils rhizobial inoculation may be useful.

Table 1 .
. Soil Ca and Mg were extracted by ammonium acetate and determined with a SensAA Dual atomic adsorption spectrophotometer (Dandenong, Australia).Soil 555 Soil characteristics at the study site

Table 2 .
Effects of inoculation and lime on macronutrient contents in soil Ø -control (no inoculation and no fertilization); ns -statistically not significant; values followed by the same letter in a column are not significantly different (Duncan test, P< 0.05)

Table 3 .
Effects of inoculation and lime on micronutrient contents in soil (no inoculation and no fertilization); values followed by the same letter in a column are not significantly different (Duncan test, P< 0.05)

Table 4 .
Effects of inoculation and lime on shoot dry weight (SDW) and macronutrient contents in red clover plants (% of dry weight) (no inoculation and no fertilization); values followed by the same letter in a column are not significantly different (Duncan test, P< 0.05); 1 st cut, 2 nd cut application in both cuts decreased K content in plants, while K content in plants significantly increased after NPK fertilization.

Table 5 .
Effects of inoculation and lime on micronutrient contents (mg kg -1 ) in red clover plants Control (no inoculation and no fertilization); ns -statistically not significant; values followed by the same letter in a column are not significantly different (P< 0.05); 1 st cut, 2 nd cut