Effect of substitution of chemical fertilizer by bioinoculants on plant performance and rhizospheric bacterial community: case study with Cajanus cajan

Abstract

Improper nutrient management is one of the major limitations linked with cultivation of Cajanus cajan. This calls for an urgent need for a promising alternative, employing both bioinoculants and chemical fertilizer. Present study attempted to understand the impact of bioinoculants {Azotobacter chroococcum, Bacillus megaterium, and Pseudomonas fluorescens (ABP)} as their mono-inoculations, triple-inoculation, and their combination with different doses of fertilizer on (a) plant parameters, (b) soil nitrogen (N) economy, (c) resident bacterial community, (d) genes and transcripts involved in N cycle, and to evaluate the extent to which fertilizer could be replaced by ABP without compromising on grain yield. Bradyrhizobium sp. was used in all the treatments (as it was recommended for C. cajan). Combined application of bioinoculants and 75% of recommended dose of fertilizer (RDF) led to 1.28-fold enhancement in grain yield as compared to RDF alone. Apart from exerting a positive impact on grain yield, the combined application of ABP and fertilizer led to an improvement in soil fertility, and modified the culturable rhizospheric bacterial community involved in N cycle. Integrated use of bioinoculants and fertilizer led to better N substrate utilization and hence, metabolic diversity when compared with application of fertilizer alone. An increase in the transcripts of nifH gene at the harvest stage in the soil treated with ABP alone and its combination with fertilizer, over individual treatment with fertilizer was observed. The combined use of ABP and fertilizer shaped the resident bacterial community towards a more beneficial community, which helped in increasing soil nitrogen turnover and hence, soil fertility as a whole.

Supplementary Information

The online version contains supplementary material available at 10.1007/s42770-020-00418-7.

Introduction

India leads in terms of the production and consumption of Cajanus cajan (common name: pigeonpea), with about 90% of global production. It is the second most important legume crop in India after chickpea (Cicer arietinum) [1]. C. cajan is a source of food rich in nutrition; it also serves as fodder for domestic animals, and fuel wood. Owing to its high nitrogen fixation ability due to nodulation by Rhizobium, it improves soil nutrient status. Seeds of C. cajan contain about 20–22% protein and significant amounts of minerals and essential amino acids [2]. Due to its deep tap root system, the plant enhances soil fertility, prevents soil erosion, and is able to tolerate drought. However, stagnancy has been observed in its productivity over the last five decades [3]. Low yield of C. cajan is due to the restricted partitioning of nutrients (division of nutrients among various plant parts) like nitrogen, phosphorous, and potassium into pods [4–6]. Therefore, the role of nutrients in soil is of utmost importance in promoting plant productivity.

To enhance the crop productivity, there has been an over use of chemical fertilizers, which has led to an extreme dependence on them, because without their application, the already nutrient depleted soil is unable to sustain the crop. The repercussions from the overuse/misusage of chemical fertilizers, increasing food demand, and low purchasing power of Indian farmers have emphasized the urgent need for sustainable agronomy. Hence, the present trend in agriculture is directed towards exploring alternative ways to attain sustainability by minimizing the application of chemicals. Ecologically safe and environment-friendly solution to this problem is the use of microbial cultures as bioinoculants for enhanced plant growth and soil health in integrated plant management systems [7, 8]. Therefore, employment of integrated nutrient management by judicious application of combination of bioinoculants and fertilizers would be beneficial for the plant wherein, we have to assess what is the maximum amount of fertilizer that can be replaced by bioinoculants without affecting grain yield. This makes it imperative to gain an understanding of the role of rhizospheric bacterial communities in affecting soil N economy (optimum availability of soil nitrogen that can be utilized by plants).

Bioinoculants are widely used for sustainable agriculture due to their useful interactions with plant roots, improvement of grain yield, and protection against soil-borne pathogens [9–11]. Apart from these advantages, these eco-friendly alternatives like Azotobacter and Azospirillum help in saving about 20–30 kg ha⁻¹ inorganic N fertilizer, because of their potential in fixing nitrogen in leguminous as well as non-leguminous plants [12, 13] by enhancing soil NPK status. It was reported that the combination of organic and bio-fertilizers enhanced the soil nutrient status in terms of available N, P, K, and soil organic C as compared to inorganic fertilizer alone [14]. A huge amount of data is available supporting the potential of bioinoculants in promoting plant growth attributes [15–18]; but limited studies have been carried out on the ‘non-target’ effects of bioinoculants on soil resident bacterial community [19–22]. Besides, limited literature is available on determining the extent of replaceability of fertilizer by bioinoculants without compromising on grain yield as well as soil health [12, 17, 23, 24]. Additionally, most of the studies have been performed wherein a single microorganism has been employed for enhancement of plant growth [25–27]. However, these studies have reported inconsistencies in performances with respect to grain yield and plant growth. According to [28], this could be attributed to the fact that a single bioinoculant might not be able to satisfy all the nutritional demands of the plant, and might not be able to survive in different soil conditions. To answer these issues, the use of diverse and compatible strains is advocated. Harnessing different plant growth-promoting mechanisms of strains of bioinoculants is a means to guarantee efficacy in plant growth promotion.

The present study involves assessment of the impact of bioinoculants as a combination of Azotobacter chroococcum, Bacillus megaterium, and Pseudomonas fluorescens (ABP). A. chroococcum is a free-living diazotrophic microorganism. B. megaterium is a key organism for the biocontrol of plant disease. P. fluorescens exhibits biocontrol properties by producing enzymes like chitinase and β-1, 3-glucanase [29]. Bradyrhizobium sp. was used in all the treatments (treatments mentioned in materials and methods section), as it was recommended for C. cajan UPAS-120 variety used in the study.

Nitrogen is a pivotal element responsible for plant growth promotion. Hence, it becomes important to target microbes involved in nitrogen turnover in the soil [30]. This is achieved by targeting the genes and transcripts involved in several steps of N cycle, viz., nitrogen fixation, nitrification, and denitrification. Though other macro and micronutrients are also crucial, the present study focussed on steps involved in N cycle. The present study provided an insight into non-targets effect of bioinoculants on the soil resident bacterial community which is a much neglected aspect, along with the extent of replaceability of fertilizers by bioinoculants. We hypothesized that bioinoculants have the potential to replace fertilizer to certain extent without compromising on grain yield, thereby eventually leading to an overall beneficial impact on the crop owing to their impact on soil N economy and their ‘non-target effects,’ which they exert on resident bacterial community. Hence, to elucidate the hypothesis, our objectives were to gain an in-depth knowledge of the effect of bioinoculants (ABP) and their combination with chemical fertilizer (CF) on (i) plant growth parameters, (ii) soil fertility, (iii) culturable bacteria involved in N cycle, (iv) abundance of resident and active bacterial community (employing 16S rRNA as molecular marker), (v) community function by targeting the genes and transcripts involved in N cycle (nitrogen fixation, nifH; nitrification, bacterial amoA and denitrification, narG and nirK), and to evaluate the extent to which the fertilizer can be replaced by ABP without compromising on grain yield.

Materials and methods

Plant system and bacterial strains

The model crop for the study was C. cajan, cultivar UPAS-120, which is an early maturing variety. The seeds were procured from National Seed Corporation, Pusa, New Delhi, India. Bioinoculants used in the study were A. chroococcum A-41, B. megaterium MTCC 453, and P. fluorescens MTCC 9768. B. megaterium MTCC 453 and P. fluorescens MTCC 9768 were procured from Institute of Microbial Technology, Chandigarh, while A. chroococcum A-41 and Bradyrhizobium were obtained from Division of Microbiology, Indian Council of Agricultural Research-Indian Agricultural Research Institute, New Delhi, India. The three bioinoculants (A. chroococcum, B. megaterium, and P. fluorescens) were assessed for their compatibility with each other for the confirmation of antagonism, if any, by employing standard cross streak assay method [31]. The three bioinoculants were found to be compatible with each other [32].

Preparation of formulation, seed surface disinfection, and bacterization

Culture broths consisting of 1.0 × 10¹⁰ cfu mL⁻¹ of each bacterial strain were used for preparation of formulation. To prepare 100 g formulation, 80 g sterilized talcum powder (inorganic carrier) was blend with 18 mL of culture broth, 1 mL of 50% autoclaved glycerol, and 1 mL of 0.1 mg mL⁻¹ filter-sterilized carboxymethyl cellulose (served as an adhesive). Seed surface disinfection and bacterization was carried out as per [32]. The average cfu per seed was found to be in the range of 4.0–8.0 × 10⁸.

Plant growth experiment to assess the impact of bioinoculants in combination with fertilizer

Pot experiment was set in Indian Institute of Technology Delhi (IITD) nursery to determine the amount of fertilizer that can be replaced by mixed consortium of ABP without compromising on plant growth parameters. The experiment was carried out for 120 days. Soil used in the experiment was collected from IITD nursery, and had the following properties: clay loam (39% clay, 34% sand, and 27% silt), pH 6.89, EC 1.37 mS cm⁻¹, and organic carbon 5.44%. This soil was used to fill pots of ~ 40 cm diameter and ~ 60 cm height, and 5–6 seeds were sown at a depth of approx. 5 cm at regular spacing. The experiment was set up in completely randomized design (CRD), wherein the pots were randomly assigned to treatments (mentioned below). For each treatment, there were three pots in the experiment. After germination each pot had approx. 4–6 plants. One plant per pot was sampled at each sampling point so as to have three biological replicates. In total, the number of samples analyzed were 99 (11 treatments * 3 sampling points * 3 replicates). The treatments included monoinoculation [T1–A. chroococcum (A), T2–B. megaterium (B), and T3–P. fluorescens (P)], T4–treatment with Bradyrhizobium sp., triple inoculation [T5–ABP], combination of bioinoculants and varying RDF (RDF = recommended dose of fertilizer: 100 kg di-ammonium phosphate (DAP) ha⁻¹/67 ppm) [T6–ABP + 25%RDF, T7–ABP + 50%RDF, T8–ABP + 75%RDF], 100% RDF (T9–CF), control plant without any treatment (of bioinoculants or fertilizer) was used as negative control (T10–C), and bulk soil (T11- Bulk) to assess the seasonal variations (Table 1). Three sampling points were selected based on different plant growth stages: vegetative stage [30 days after sowing (DAS)], flowering stage (90 DAS) and harvest stage (120 DAS). At the time of each sampling, rhizospheric soil was collected by brushing and collecting the soil, which was tightly adhered to the roots. The samples were then stored at 4 °C and − 20 °C, for cultivation-dependent and cultivation-independent analysis, respectively. At each sampling point various plant parameters were measured viz., root length, fresh weight, number of branches per plant, and grain yield. Root length was measured from the tip of the tap root to the root crown.

Table 1.

Treatments used in the study

Treatments	Description	Abbreviation
T1	Monoinoculation with Azotobacter chroococcum	A
T2	Monoinoculation with Bacillus megaterium	B
T3	Monoinoculation with Pseudomonas fluorescens	P
T4	Treatment with Bradyrhizobium sp.	Bradyrhizobium sp.
T5	Triple inoculation of A, B and P	ABP
T6	Triple inoculation with 25% Recommended dose of fertilizer	ABP + 25%RDF
T7	Triple inoculation with 50% Recommended dose of fertilizer	ABP + 50%RDF (T50CF in Fig. 1)
T8	Triple inoculation with 75% Recommended dose of fertilizer	ABP + 75%RDF
T9	100% Recommended dose of fertilizer (100 kg di-ammonium phosphate ha-1)	CF
T10	Control plant without any treatment of bioinoculants or fertilizer	C
T11	Bulk soil	Bulk

Enumeration of bacterial groups involved in N cycle

Cultivation-dependent analysis was done on the rhizosphere soil samples for the quantification of bacterial groups involved in various steps of N cycle. Initial sampling was done at the time of sowing. Nitrogen fixers were enumerated on Jensen’s medium (containing 20.0 g sucrose; 1.0 g K₂HPO₄; 0.5 g MgSO₄.7H₂O; 0.001 g Na₂MoO₄; 0.01 g FeSO₄.7H₂O; 2.0 g CaCO₃; 18.0 g agar; volume made up to 1000 mL by distilled water) [33], denitrifiers on Asparagine nitrate medium (containing 1.0 g KNO₃; 1.0 g l-asparagine; 8.5 g Na₃C₆H₅O₇; 1.0 g KH₂PO₄; 1.0 g MgSO₄; 0.2 g CaCl₂; 0.0001 g FeCl₃; 15.0 g agar; volume made up to 1000 mL by distilled water) [34], Nitrosomonas and Nitrobacter species on Winogradsky phase I medium (containing 2.0 g (NH₄)₂SO₄; 1.0 g K₂HPO₄; 0.5 g MgSO₄.7H₂O; 2.0 g NaCl; 0.4 g FeSO₄.7H₂O; 0.01 g CaCO₃, 15.0 g agar; volume made up to 1000 mL by distilled water), and phase II medium (containing KNO₂ 0.1 g, Na₂Co₃ 1.0 g, NaCl 0.5 g, FeSO₄.7H₂O 0.4 g, agar 15.0 g; volume made up to 1000 mL with distilled water) [35], respectively. The samples were plated onto respective selective media post serial dilution. The plates were then incubated at 30 °C for 48 h. Plates containing colonies in the range from 30 to 300 were counted and the results were obtained in cfu g⁻¹ dry soil.

Soil N content

Soil N content was analyzed for the presence of nitrate, nitrite and ammonia by using HACH®kit (Model DR-EL, Hach Chemical Company, Iowa) and total N by Kjeldahl method. For the analysis of ammonia, nitrate, and nitrite, the soil samples were extracted with 1 M KCl.

Community-level physiological profiling

Metabolic profiling of both bulk and rhizospheric soil with different treatments was performed at flowering and harvest stages targeting Gram-positive and Gram-negative bacterial community using BIOLOG®PM3B Microplate™ (Biolog, Inc, CA, USA). The plate contained 95 different N sources and a negative control (devoid of any N source). Community-level physiological profiling (CLPP) assesses the metabolic potential of a single organism, or whole microbial community. The principle behind this technique relies on the reduction of a tetrazolium dye that changes to purple color when N substrates coated in the wells are oxidized. The samples were prepared as follows: 1 g of soil was mixed in sterile 10 mL PBS, and kept under shaking (180 rpm) at 25 °C overnight. The tubes were kept undisturbed at room temperature until the soil particles settled down, and then the supernatant was diluted a thousand times with phosphate buffered saline (PBS). The supernatant was then processed for Gram-positive (GP) and Gram-negative (GN) bacteria separately as per manufacturer’s protocol. PM3B plates were then inoculated with 150 μL of processed soil samples (for both GP and GN bacteria) and incubated in the dark at room temperature. Absorbance was measured after every 24 h at 590 nm for 7 days. Microbial activity was expressed as average well color development (AWCD = Σ ODi/31). Metabolic diversity (Shannon diversity) was calculated {H = - Σpi(lnpi)} (pi = proportional color development of the well over total color development of all wells of a plate) in accordance with [36, 37].

Total nucleic acid extraction and cDNA synthesis

Rhizospheric soil samples stored at – 20 °C were subjected to total nucleic acid extraction [38]. Extracted nucleic acid was divided into two parts, one part was treated with RNase-free DNase I enzyme (Thermo Scientific, USA), to eliminate genomic DNA which was co-extracted with RNA, prior to cDNA synthesis in accordance with manufacturer’s protocol. cDNA synthesis from RNA was conducted with random hexamers, in a final reaction volume of 20 μL using RevertAid First-strand cDNA synthesis Kit (Thermo Scientific, USA). Thermal cycling conditions were as follows: 25 °C for 5 min, 42 °C for 60 min, and termination by heating at 70 °C for 5 min. The product was stored at – 20 °C for further use.

Real-time PCR assays

Before quantitative analysis of resident (DNA) and active (cDNA) bacterial community, the presence of inhibitors in the nucleic acid extracts was examined by spiking a pre-determined quantity of pGEM-T plasmid (Promega, USA) with Enterococcus sp. specific gene, followed by amplification with T7 and SP6 primers. PCR conditions used were 94 °C for 15 min, 35 cycles at 94 °C for 45 s, annealing at 55 °C for 45 s, 72 °C for 90 s, and a final extension step at 72 °C for 10 min. In all the cases, no inhibition was detected since the measured cycle threshold (Ct) values obtained from DNA extracts were not significantly different from the control. No template controls (NTC) gave null or negligible values.

For the quantitative analysis of resident and active bacterial community in the rhizosphere, qPCR assay was performed. The total bacterial abundance was assessed by employing 16S rRNA primer-based qPCR assay using specific primers (1369F and 1492R) and PCR conditions [39]. Genes encoding catalytic enzymes involved in N cycle (nifH for nitrogen fixation, amoA for ammonia oxidation, and narG and nirK for denitrification) were employed as molecular markers to assess the abundances of resident and active N fixers, bacterial ammonia oxidizers, and denitrifiers, respectively, by using the PCR protocols (nifH: 95 °C-15 s, 60 °C-30 s, 72 °C-30 s, 80 °C-30 s for 35 cycles; amoA: 95 °C-60 s, 55 °C-60 s, 72 °C-30 s for 45 cycles; narG/nirK: 95 °C-15 s, 63 °C-1 °C, cy-30 s, 72 °C-30 s, 80 °C-30 s for 6 cycles; and 95 °C-15 s, 58 °C-30 s, 72 °C-30 s, 80 °C-30 s for 35 cycles) and primers (nifH: PolF and PolR, amoA: amoA1F and amoA2R, narG: narGF and narGR, nirK: nirK 876F and nirK 1040R) as per [18]. The gene copy number was expressed per g dry soil.

qPCR assays were performed in polypropylene 96-well plates with the CFX96 Touch Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA) by employing SYBR green as the detection dye in a reaction mixture of 10 μL containing 0.5 μL of 0.5 μM of each primer; 5 μL of 2X SsoFast EvaGreen Supermix (Bio-Rad, Hercules, CA, USA); and 0.5 μL of diluted template corresponding to 15 ng of DNA, nuclease-free water was then added to make the final volume to 15 μL. The explicitness of the amplicon was assessed by observing a unique single melt peak. Standard curves produced were linear and generated with serial dilutions of a known amount of plasmid DNA, containing targeted gene fragment (16S rRNA: Agrobacterium tumefaciens; nirK: Sinorhizobium meliloti; nifH: Frankia alni; narG: Pseudomonas aeruginosa; amoA: Nitrosomonas europaea) (r² > 0.988 for all assays), with PCR efficiency (E = 10^–1/slope) > 95%.

Statistical analysis

The experiments were performed in CRD. Standard deviation was calculated for each treatment. Two-way analysis of variance (ANOVA) and principal component analysis (PCA) was performed on the data using SPSS Statistics 16.0 for Windows® (SPSS Inc., Chicago, III., USA). Two-way ANOVA was conducted to determine if there were significant differences among various treatments, with ‘treatments’ and ‘time points’ as independent variables and various plant growth parameters, soil N content, abundance of rhizospheric bacterial community, and gene/transcript copy numbers as the dependent variables. Tukey’s HSD post-hoc test (t test examining mean differences between various treatments) was used for means’ comparison when ANOVA yielded p < 0.05 [40]. PCA was performed on various plant growth parameters (wherein each dot represented the value for the particular treatment in triplicates) at harvest stage to assess the impact of various treatments and to analyze possible relationships between them. Heat map generation and PCA for CLPP was performed by employing XLSTAT (Addinsoft, New York, USA). For the construction of heat map and PCA for data obtained from BIOLOG® plates, AWCD values were taken into consideration.

Results

Effect of bioinoculants and their combination with fertilizer on plant growth parameters

The plants harvested at each time point were assessed for their growth parameters, viz. root length, fresh weight, number of branches per plant, and grain yield (data for harvest stage compiled in Table 2). For simplicity, the impact on plant parameters has been presented at harvest stage (120 DAS) only. In the case of root length, ABP, ABP + 25%RDF, ABP + 50%RDF, ABP + 75%RDF, and fertilizer treatments showed a fold enhancement of 1.90, 1.89, 1.97, 1.79, and 1.80, respectively, over control. ABP exhibited a marked difference in fresh weight over control, along with the treatments ABP + 50%RDF, ABP + 75%RDF, and fertilizer. The grains were collected from the plants from all the treatments and then weighed to obtain the grain yield per plant. The treatments Bradyrhizobium sp., ABP, ABP + 25%RDF, ABP + 50%RDF, ABP + 75%RDF, and fertilizer showed a fold enhancement of 1.71, 4.26, 3.18, 4.83, 5.65, and 4.43, respectively, over control.

Table 2.

Effect of bioinoculants on various plant growth parameters at harvest stage (120 DAS)

Treatmentsa	Root length (cm)	Fresh wt. (g/plant)	No. of branches/plant**	Grain yield (g/plant)
T1 (A. chroococcum)	4.1a	46.7a	1.7abc	29.7a
T2 (B. megaterium)	4.1a	45.9a	1.0a	45.8ab
T3 (P. fluorescens)	4.5a	46.5a	1.7abc	34.4a
T4 (Bradyrhizobium sp.)	4.8ab	48.2a	2.3abc	64.8b
T5 (ABP)	11.9c	63.8cd	4.0bc	161.7d
T6 (ABP + 25% RDF)	11.9c	61.5bc	3.3abc	120.5c
T7 (ABP + 50% RDF)	12.3c	67.2cd	3.7abc	183.4d
T8 (ABP + 75% RDF)	11.2c	*80.7e	4.3c	*214.5e
T9 (CF)	11.3c	69.9d	4.0bc	168.1d
T10 (Control)	6.3b	49.2ab	1.3ab	37.9a

*Numbers in italics represent significantly higher values compared to the rest of the treatments

**The values are average values

^aA: A. chroococcum, B: B. megaterium, P: P. fluorescens, RDF: recommended dose of fertilizer, CF: chemical fertilizer

Distribution of plant growth attributes at harvest stage, viz., root length, grain yield, and fresh weight, were visualized by PCA (Fig. 1), which depicted that plant parameters clustered distinctly with respect to the treatments. Only selected treatments with most striking differences were included in the figure for clarity. The matrix generated clearly demonstrated a noticeable shift in the plant parameters upon application of ABP, their combination with fertilizer (ABP + 50%RDF) and fertilizer when compared to control.

Fig. 1.

Distribution of various plant growth parameters at harvest stage w.r.t. biometric observations within various treatments by principal component analysis. Black circles = root length; white circles = fresh weight and, grey circles = grain yield (suffixes 1, 2 and 3, respectively after the name of treatment). C: control, ABP: A. chroococcum + B. megaterium + P. fluorescens, T50CF: ABP + 50%RDF, CF: chemical fertilizer treatment

Effect of bioinoculants and their combination with fertilizer on the abundance of culturable fraction of specific rhizospheric bacterial groups

The rhizospheric bacterial groups targeted were those involved in various steps of N cycle (Table 3). The abundance of these bacterial groups was in the order of 10⁵ cfu g⁻¹ dry soil at the time of sowing. At vegetative stage (30 DAS), all the treatments, except mono-inoculation with Pseudomonas, showed an increase in the population of N fixers as compared to the control and bulk soil (Online Resource 1a). As the plant progressed to the pre-flowering stage (60 DAS), the treatment ABP, its combination with fertilizer, and CF exhibited significant enhancement with an increase in a range from 3.7- to 5.8-fold and 3.9- to 6.2-fold over control and bulk soil, respectively. The maximum abundance of N fixers was observed at flowering stage (90 DAS). Also, rhizosphere effect was detected at this stage, as the abundance in all the treatments, including control without inoculation, was markedly higher compared to bulk soil.

Table 3.

Effect of bioinoculants on culturable microbial fraction at harvest stage (120 DAS)

Treatments	Nitrogen fixers (cfu/g dry soil)	Nitrosomonas sp. (cfu/g dry soil)	Nitrobacter sp. (cfu/g dry soil)	Denitrifiers (cfu/g dry soil)
T1 (A. chroococcum)	2.31E+06ab	1.72E+05cd	1.17E+05bcd	6.17E+04a
T2 (B. megaterium)	2.05E+06a	1.19E+05b	1.02E+05b	5.77E+04a
T3 (P. fluorescens)	1.97E+06a	1.35E+05bc	1.08E+05bc	4.23E+04a
T4 (Bradyrhizobium sp.)	3.32E+06c	1.85E+05de	1.06E+05bc	4.10E+04a
T5 (ABP)	6.21E+06d	2.25E+05ef	1.43E+05cde	7.27E+04a
T6 (ABP + 25% RDF)	6.33E+06d	2.46E+05fg	1.55E+05de	6.90E+04a
T7 (ABP + 50% RDF)	6.56E+06d	2.65E+05fg	1.63E+05e	4.90E+04a
T8 (ABP + 75% RDF)	6.99E+06d	2.87E+05g	1.68E+05e	7.23E+04a
T9 (CF)	6.26E+06d	2.69E+05fg	1.35E+05b-e	6.63E+04a
T10 (Control)	3.13E+06bc	1.48E+05bcd	1.06E+05bc	4.50E+04a
T11 (Bulk soil)	1.54E+06a	6.77E+04a	6.10E+04a	3.93E+04a

Numbers in italics represent significantly higher values compared to the control treatment (T10)

Nitrosomonas spp. did not respond to the treatments with its abundance remaining comparable to control and bulk soil until flowering stage (Online Resource 1b). At harvest stage, however, treatments ABP + 75%RDF and fertilizer resulted in marked enhancement of Nitrosomonas spp. over bulk soil showing a fold increase of 4.25 and 3.97, respectively. The abundance of Nitrobacter spp. on the other hand was higher at vegetative stage for the treatments containing Azotobacter alone, ABP, ABP + 75%RDF, and CF over control (Online Resource 1c). At flowering stage, rhizosphere effect was prominent leading to enhanced abundance of Nitrobacter spp. in all the treatments. At harvest stage, however, the abundance was reduced being comparable in both rhizosperic soil as well as bulk soil.

Abundance of denitrifiers was affected by the treatments until pre-flowering stage, after which it reduced to the level of bulk soil in all the treatments (Online Resource 1d). At the vegetative stage, the mono-inoculation treatments with Azotobacter, Bradyrhizobium sp., ABP, ABP + 25%RDF, ABP + 50%RDF, ABP + 75%RDF, and CF exhibited marked increase over control. At the pre-flowering stage, all the treatments showed significant enhancement in the abundance of denitrifiers compared to the control and bulk soil.

Effect of bioinoculants and their combination with fertilizer on soil N content

Soil N content was assessed at all the stages of plant’s growth viz., vegetative stage, pre-flowering stage, flowering stage and harvest stage (Online Resource 2). Data on N content at harvest stage has been shown in Table 4. To assess soil N content, the treatments compared with respect to control and bulk soil were ABP, ABP + 50%RDF, and CF. As the plant progressed to harvest stage (120 DAS), ABP + 50%RDF showed the best result for soil nitrate resulting in 1.79-fold enhancement over control treatment. In the case of soil nitrite availability at harvest stage, there was a 2.18-fold increase for ABP + 50%RDF as compared to the control soil. In terms of ammonium availability in soil, CF exhibited highest values of ammonium in soil.

Table 4.

Effect of bioinoculants on soil N content at harvest stage (120 DAS)

Treatmentsa	Nitrate (mg/L)	Nitrite (mg/L)	Ammonium (mg/L)	TKNb (mg N/g soil)
ABP	20.3bc	26.2c	863.1b	5.7a
ABP + 50%RDF	27.2c	36.6d	1089.6c	7.3a
CF	21.3bc	23.6c	1184.7c	6.2a
C	15.2ab	16.7b	656.6a	4.8a
Bulk soil	11.0a	9.0a	608.9a	4.4a

Numbers in italics represent significantly higher values compared to the control treatment

^aA: A. chroococcum, B: B. megaterium, P: P. fluorescens, RDF: recommended dose of fertilizer, CF: chemical fertilizer, C: control

^bTKN total Kjeldahl nitrogen

Effect of bioinoculants and their combination with fertilizer on metabolic diversity of soil resident bacterial community

BIOLOG® plates were used to assess substrate utilization capability of microbes as indicated by AWCD. In general, the maximum number of substrates were utilized in flowering stage (90 DAS) compared to harvest stage (120 DAS) (Fig. 2). Hence, the metabolic potential was higher at flowering stage. Gram-negative bacteria were capable of utilizing more substrates compared to Gram-positive bacteria of the respective treatments at both the time points. At flowering stage, high values of AWCD were obtained for bulk soil and control treatment followed by ABP (p < 0.05) (Fig. 2a). The saturation in AWCD was obtained after 7 days. For both Gram-negative and Gram-positive bacteria, bulk soil and control formed a separate cluster. Most of the l-amino acids were not utilized by the treated soil (soil amended with the three inoculants, chemical fertilizer, and their combination), while the bacteria in bulk soil were easily able to utilize them. At harvest stage as well, maximum substrates were utilized by the Gram-negative bacteria of bulk soil, which was then trailed by Gram-negative bacteria in ABP treatment (Fig. 2b). Three major clusters were formed; the first cluster comprised of Gram-negative bacteria of bulk and ABP, which showed maximum substrate utilization. The second cluster was formed by Gram-positive and Gram-negative bacteria of CF and ABP + 50%RDF treatments, exhibiting minimum number of substrate utilization, and the third cluster was formed by control and Gram-positive bacteria of bulk and ABP treatments, showing intermediate substrate utilization capability.

Fig. 2.

Heat map from CLPP depicting metabolic potential of both rhizosphere and bulk soil bacterial community of various treatments in consuming variety of nitrogenous substrates at (a) flowering and (b) harvest stages. +ve and –ve denote Gram +ve and Gram –ve bacteria, respectively. ABP: A. chroococcum + B. megaterium + P. fluorescens, ABP+ 50%RDF: ABP + 50% recommended dose of fertilizer, CF: chemical fertilizer. Wells with different colors represent statistically different values

Shannon diversity index (H′) and evenness (E) were used to describe the metabolic richness and homogeneity of soil bacterial community for various treatments (Table 5). In general, higher values of Shannon diversity index was observed with Gram-negative bacteria compared to Gram-positive bacteria. The flowering stage depicted higher metabolic diversity compared to harvest stage. Within treatments, bulk soil exhibited maximum metabolic potential and evenness, which was then trailed by ABP treatment. CF showed the lowest metabolic diversity. PCA was employed to differentiate N substrate utilization patterns between various treatments compared to the control and bulk soil at both flowering (Online Resource 3a) and harvest (Online Resource 3b) stages.

Table 5.

Shannon diversity index (H′) and evenness (E) at flowering (90 DAS) and harvest (120 DAS) stages

	Shannon diversity (H′)				Shannon evenness (E)
	Flowering stage (90 DAS)		Harvest stage (120 DAS)		Flowering stage (90 DAS)		Harvest stage (120 DAS)
Treatmentsa	Gram positive	Gram negative	Gram positive	Gram negative	Gram positive	Gram negative	Gram positive	Gram negative
Bulk soil	4.52	4.09	4.26	4.04	1.82	1.06	1.22	1.07
C	3.40	4.37	3.71	4.22	1.05	1.67	1.06	1.12
ABP	2.77	4.16	3.71	4.17	1.04	1.49	1.04	1.08
CF	1.95	3.93	3.43	3.97	0.10	1.06	1.03	1.06
ABP + 50%RDF	2.48	4.08	3.43	3.99	1.00	1.04	1.03	1.07

^aA: A. chroococcum, B: B. megaterium, P: P. fluorescens, RDF: recommended dose of fertilizer, CF: chemical fertilizer, C: control

Effect of bioinoculants and their combination with fertilizer on the abundance of 16S rRNA and functional markers

The impact of various treatments when compared to control treatment on abundance of 16S rRNA and N cycle functional markers at harvest stage is presented in Table 6. The detailed information is provided in Online Resource 4 and 5.

Table 6.

Impact of various treatments on genes and transcripts of 16S rRNA and markers for N cycle at harvest stage (120 DAS)

Treatments	16S		nifH		amoA		narG		nirK
	Genes/g dry soil	Transcripts/g dry soil	Genes/g dry soil	Transcripts/g dry soil	Genes/g dry soil	Transcripts/g dry soil	Genes/g dry soil	Transcripts/g dry soil	Genes/g dry soil	Transcripts/g dry soil
T1 (A. chroococcum)	2.93E+10abc	5.17E+08c	3.59E+09e	6.09E+05d	3.67E+07ef	7.26E+02c	4.73E+08efg	2.13E+04d	4.07E+07de	1.87E+04d
T2 (B. megaterium)	3.13E+10bc	2.96E+08b	2.01E+09bcd	3.49E+05bc	3.27E+07e	4.15E+02b	4.90E+08fg	1.56E+04bc	3.79E+07d	1.36E+04bc
T3 (P. fluorescens)	3.22E+10bc	3.81E+08b	9.96E+08ab	4.48E+05c	1.02E+07b	5.34E+02b	2.07E+08b	1.82E+04cd	1.18E+07b	1.60E+04cd
T4 (Bradyrhizobium sp.)	5.73E+10cd	6.46E+08d	3.2E+09e	7.60E+05e	2.05E+07cd	9.06E+02de	2.70E+08bc	1.96E+04cd	2.38E+07c	1.72E+04cd
T5 (ABP)	1.15E+11g	6.35E+08cd	5.49E+09f	7.48E+05e	4.62E+07f	8.91E+02de	5.74E+08g	2.06E+04cd	4.94E+07e	1.81E+04cd
T6 (ABP + 25% RDF)	1.07E+11fg	5.97E+08cd	3.37E+09e	7.04E+05de	2.30E+07d	8.39E+02cd	3.98E+08def	2.09E+04cd	1.47E+07bc	1.83E+04cd
T7 (ABP + 50% RDF)	8.85E+10efg	9.62E+08e	2.96E+09de	9.67E+05f	2.02E+07cd	1.02E+03e	3.50E+08cde	3.00E+04e	1.16E+07b	2.64E+04e
T8 (ABP + 75% RDF)	7.13E+10de	3.52E+08b	2.56E+09cde	4.15E+05c	1.75E+07bcd	4.94E+02b	3.02E+08bcd	2.17E+04d	1.79E+07bc	1.91E+04d
T9 (CF)	8.22E+10def	5.86E+08cd	1.13E+09ab	2.60E+05b	1.14E+07bc	5.37E+02b	1.95E+08b	1.25E+04ab	2.07E+07bc	1.10E+04ab
T10 (Control)	1.25E+10ab	3.59E+08b	1.84E+09bc	4.23E+05c	1.26E+07bc	7.02E+02c	2.71E+08bcd	1.86E+04cd	2.36E+07c	1.64E+04cd
T11 (Bulk soil)	6.45E+08a	8.70E+07a	1.14E+08a	7.76E+04a	3.90E+05a	6.80E+01a	1.16E+07a	9.12E+03a	4.01E+05a	8.03E+03a

Numbers in italics represent significantly higher values compared to the control treatment (T10)

Effect on total resident and active bacterial community

The gene copy number of 16S rRNA varied from 3.9 × 10⁸ to 1.2 × 10¹¹ per gram dry soil, while that of transcripts were in the range of 2.1 × 10⁷ to 9.6 × 10⁸ per gram dry soil in all the treatments across the time points (Online Resource 4a, b). Both genes and transcripts followed a similar trend, wherein vegetative stage (30 DAS) did not reflect any effect on the resident and active bacterial abundance among different treatments and were comparable to control and bulk soil. At flowering stage (90 DAS), maximum bacterial abundance was observed as compared to other time points in all the treatments. At flowering and harvest stages (120 DAS), abundance of both resident and active bacterial population was highest in the treatment ABP + 50%RDF, over control and bulk soil.

Effect on resident and active nitrogen fixers

The number of nifH gene copies (1.0 × 10⁸ to 5.5 × 10⁹ per gram dry soil) was almost 10000-fold higher compared to nifH transcripts copies (2.3 × 10⁴ to 9.7 × 10⁵ per gram dry soil) (Online Resource 5a, b). The population of resident and active N fixers were similar in all the treatments at vegetative stage (30 DAS). As the plant entered into the flowering stage (90 DAS), a fold increase of 1.7 and 1.6 was exhibited by resident and active N fixers respectively, for the treatment ABP + 50%RDF over control. At the harvest stage (120 DAS), ABP exhibited highest abundance (3.0-fold) of resident N fixers over control soil.

Effect on resident and active nitrifiers

The range of copies of bacterial amoA gene was from 3.9 × 10⁵ to 4.6 × 10⁷ per gram dry soil (Online Resource 5c). No difference was seen among the treatments at vegetative stage (30 DAS). At flowering stage (90 DAS), ABP + 25%RDF resulted in significantly higher abundance of nitrifiers compared to the rest of the treatments. During the harvest stage of the plant (120 DAS), mono-inoculation with Azotobacter and Bacillus, ABP, and ABP + 25%RDF surpassed the abundance significantly over control treatment. Copies of bacterial amoA transcripts ranged from 6.8 × 10¹ to 1.0 × 10³ per gram dry soil (Online Resource 5d). At the vegetative stage, ABP + 50%RDF showed significant rise over control soil. During both flowering and harvest stages, ABP and ABP + 50%RDF had a marked increment over control.

Effect on resident and active denitrifiers

While the gene copy number of narG (1.2 × 10⁷ to 5.8 × 10⁸ per gram dry soil) was higher than that observed for nirK gene (4.0 × 10⁵ to 4.9 × 10⁷ per gram dry soil), similar trend was observed for both markers (Online Resource 5 e, f, g, h). The highest abundance of denitrifiers was observed at the harvest stage in mono-inoculation of Azotobacter and Bacillus and ABP resulting in over 1.5-fold increase over control. Similarly with respect to the transcripts copy number of narG and nirK, the active denitrifiers were similar for all the treatments at the vegetative stage. The active denitrifiers were significantly higher for mono-inoculation with Azotobacter and Bradyrhizobium sp., ABP, ABP + 25%RDF, and ABP + 50%RDF at flowering stage of the crop. At the harvest stage, however, ABP + 50%RDF showed a marked enhancement in the active denitrifiers over control soil.

Discussion

The comparable effects of a combination of bioinoculants on various plant growth parameters and grain yield, with RDF was established in field in our previous study [32]. Hence, it became imperative to evaluate the amount of fertilizer that could be replaced by these eco-friendly alternatives without compromising on grain yield and soil health.

Effect on plant growth parameters

The choice of bioinoculants was based on their reported beneficial impact on the plant, viz. A. chroococcum is a free-living diazotroph used as a biofertilizer and hence, helps in promoting plant growth. B. megaterium acts as PGPR by producing antibiotics against nematodes and by possible mechanisms involving cytokinin signaling, which helps in root growth [41]. P. fluorescens produces a wide spectrum of bioactive metabolites, which in turn help in competing with plant pathogens aggressively [29]. Assessing the impact of any treatment on growth parameters of plant becomes the first step while carrying out such experiments. Therefore, the effect of bioinoculants (mono-inoculation and ABP) and their combination with different doses of fertilizer was observed on various plant growth parameters. The best results were obtained with ABP + 75%RDF. However, the replacement of 50% of fertilizer with bioinoculants gave comparable results to RDF with respect to all the plant growth attributes measured. Direct effects of the bioinoculants on C. cajan clearly exhibited their positive impact on various plant growth attributes. This could be due to a combined effect of nutrient acquisition, protection against soil-borne diseases, and phytostimulation by plant hormone synthesis as discussed by others [42].

The enhancement in the grain yield by combination of ABP and CF can be explained by the integrated effects of both the bioinoculants (direct and indirect effects) as well as fertilizer (addition of nitrogen and phosphorus to the soil) in the present study. In a study conducted by [12], it was observed that the soil fertility and yield of onions had significantly increased by the application of biofertilizer combined with organic manure and inorganic fertilizer over control treatment containing fertilizer alone. Another study [24] reported a significant impact on plant growth and yield by the application of different combinations of biofertilizer and inorganic fertilizer when compared to inorganic fertilizer alone. In a study conducted by [17], it was shown that a combination of bioinoculants (Sinorhizobium fredii KCC5 and Pseudomonas fluorescens LPK2) with half of the recommended dose of chemical fertilizer resulted in better quality and grain yield in C. cajan. In addition to this, different studies conducted with combination of biofertilizers and inorganic fertilizer showed similar beneficial effects irrespective of plant type or experimental conditions [12, 17, 23, 24]. In the present study, PCA analysis showed that soil supplemented with different treatments at all the time points clustered separately from the untreated control highlighting the strong effects of treatments over plant growth stages on various plant growth parameters.

Effect on soil N content

Besides plant growth parameters, it is crucial to know the nutrient content of the soil too, as plant derives its nutrition from the soil. Soil also provides a platform where plant and microbes interact. The macronutrient focussed in the present study was nitrogen. As the crop under study was a leguminous crop, studying N economy of soil becomes all the more important. The effect of combination of bioinoculants with different doses of fertilizer was assessed on the soil N content. The N content in rhizospheric soil was low, and was comparable to the bulk soil. The result was in accordance with [43] who reported that though N concentration is enhanced in soil solution after application of nitrogen fertilizer, it stays lower in the rhizosphere. This is because of the absorption of the nutrient ions from the soil by the roots resulting in a difference in nutrient status in the rhizosphere as compared to the bulk soil. Decrease in ammonium content of the soil at flowering and harvest stages suggested an increased uptake by the plant because of enhancement in soil N mineralization due to microbial activities [44]. The increase in nitrite and nitrate content in the soil in the treatment containing bioinoculants and fertilizer could be due to the non-target effects of ABP, as well as direct nutrient (N and P) deposition by fertilizer. In a study conducted by [45], remarkable increment in soil nutrient status over control could be observed when fertilizer was combined with organic manure.

Effect on culturable fraction of specific rhizospheric bacterial groups

Soil resident microbial communities play a major role in affecting soil N economy. Hence, effect of different treatments on the soil bacterial community involved in various steps of N cycle was studied. It was found that effect on the culturable fraction of bacterial community involved in various steps of N cycle was affected by different treatments. A significant enhancement in the population of N fixers and nitrifiers at all the stages in treated plants over control emphasized the healthy interaction of the introduced bioinoculants with the resident N fixers. This observation can be attributed to the solubilization and fixation of certain substances like soluble form of phosphorous, fixation of nitrogen in usable form, recalcitrant carbon sources etc., which successively act as substrates for soil resident microbial community [46, 47]. An increase in the abundance of N fixers and nitrifiers in the treatments involving combination of bioinoculants and fertilizer can be attributed to their combined effect. In another study [48], it was found that an increase in the abundance of Azotobacter, spore-forming bacterial species, and pseudomonads in the rhizosphere of maize resulted after co-inoculation with Azotobacter, Bacillus, and Pseudomonas. Rhizospheric effect was clear at flowering stage, which is in line with the study [49] that stated the release of root exudates in high concentrations containing easily degradable compounds result in proliferation of resident bacterial community in the rhizosphere at this stage. Another study reported that organic fertilizer treatment in sweet corn plant led to an increase in the abundance of Gram-negative bacteria [50]. Other studies also showed similar effects on bacterial abundance by the introduction of organic fertilizers [51, 52]. The building up of denitrifiers during early stages of plant growth can be explained by the fact that for the benefit of slow growing K-strategic species, the r-strategic species that include Gram-negative bacteria proliferate during early stages and then decline [50, 51, 53–55].

Effect on metabolic diversity of soil resident bacterial community

Studying the metabolic potential of bacterial community present in soil provides a deeper understanding of microbial community function and functional adaptations over space and time. To determine the community catabolic responses upon various treatments, CLPP was performed. In the present study, different treatments were assessed for their impact on the resident bacterial community (Gram-positive as well as Gram-negative) as compared to the bulk soil, at both flowering and harvest stages. Bulk soil was included to check for seasonal variations and to evaluate the differences with respect to rhizospheric soil and hence, to assess the importance of plant root exudates on metabolic potential of the soil resident bacterial community. Metabolic diversity does not necessarily indicate species diversity, for example in the case of a community that consists of lesser number of generalist species compared to more number of specialists [56, 57]. It was found that the treated soil resulted in disturbances in the metabolic potential and evenness of soil resident community. The metabolic potential and diversity was higher for bulk soil compared to rhizosphere as has been reported earlier [58]. This can be attributed to the fact that the bulk soil has no selective pressure, and hence, is in an equilibrium state having evenly distributed community. In a study conducted by [59], it was found that the microbial diversity in terms of richness and evenness was higher in the bulk soil as compared to that of rhizosphere soil. The introduction of plant resulted in a selection pressure on the soil resident bacterial community in such a way that the community shifted towards the one that is more beneficial for the plant. The diversity with respect to substrate utilization was further decreased with the treatments containing bioinoculants, which had imposed an additional selection pressure [60]. The lowest metabolic diversity and evenness in case of fertilizer treated soil can be attributed to the fact that fertilizer resulted in an adverse effect on the resident bacterial community [61]. The metabolic potential of soil bacterial community in the case of bioinoculant treated soil was able to match with that of the bulk soil at harvest stage, suggesting the perturbations to be transient, and also revealed no negative impact of these agricultural amendments on the soil richness. The reason for higher richness and metabolic diversity of GN bacteria as compared to GP bacteria in the rhizospheric soil might be due to the fact that GN bacteria utilize C source that are plant-derived (labile), whereas GP bacteria mostly thrive on soil organic matter (recalcitrant) [62]. Comparatively higher metabolic potential was found at the flowering stage in the rhizospheric soil, suggesting higher number of microorganisms that are attracted towards the enhanced amount of root exudates at this stage [49].

Effect on abundance of 16S rRNA and other functional markers

To assess the shift in bacterial community structure, 16S rRNA was used as a marker. Maximum number of resident and active bacterial community was detected during flowering stage. It can be justified by the maximum deposition by root exudates at this stage, resulting in maintaining maximum number of rhizospheric microorganisms [63]. The positive impact of the treatments containing bioinoculants and fertilizer can be explained by various direct and indirect effects of the bioinoculants, and also direct nutrient deposition by the fertilizer. This effect was observed till the harvest stage, explaining the relatively long-term impact of these agricultural amendments by improving soil N economy. Analyzing soil nitrogen economy is crucial when the target crop is a leguminous crop. Hence, to attain a deeper insight into the soil nitrogen economy, quantitative assessment of the genes and transcripts involved in various steps of N cycle was done. Nitrogen fixation leads to an enhancement in the fixed nitrogen in the soil and hence, is considered to be a crucial process to be targeted. Though both nifH genes and transcripts did not respond much to the treatments at vegetative stage, at flowering stage their enrichment in the rhizosphere highlights the importance of rhizodeposits, which were maximum at this stage [63]. Our findings are in conformity with [30] who showed that at the late flowering stage, the nifH gene copy numbers were highest in the rhizospheric soil of alfalfa (Medicago sativa L.) treated with Sinorhizobium meliloti OS6. Enhancement in the active N fixers was observed with the treatments containing Azotobacter, Bradyrhizobium, ABP, and fertilizer. The result emphasized the role of both Azotobacter and Bradyrhizobium as free living and symbiotic N fixers, respectively, in addition to cumulative effects of bioinoculants in consortium. Fertilizer led to an enhancement in the abundance of N fixers at flowering stage, but it was not able to influence the active N fixers. This can be explained by their negative impact on the diazotrophs, as their application has been reported to reduce N fixers’ abundance [64, 65]. The process of ammonia oxidation holds importance as it leads to an increase in the nitrate availability in the soil and hence, contributes to the soil nitrogen economy. Our study showed an enhancement in the resident and active nitrifiers at flowering and harvest stages with respect to the treatment with ABP, signifying the synergistic roles of both the plant growth stages as well as treatments. It also highlighted the significance of the non-target effects of these agricultural amendments, as none of them carried amoA gene, and re-shaped the rhizosphere in a manner that was beneficial for the plant. The fertilizer treatment did not show any effect on the resident and active nitrifiers. He et al. [66] also found the highest abundance of ammonia-oxidizing bacteria (AOB) in the treatment with NPK + organic matter, which was trailed by NPK and NP treatments by employing qPCR. This can be explained by the higher nutrient availability in case of amendment with organic matter as opposed to inorganic fertilizer alone. Denitrification not only results in nitrogen loss from soil, it also leads to the emission of greenhouse gas resulting in climatic change, and therefore, considered to be deleterious. Genes and transcripts involved in denitrification were strongly influenced by the amendment with bioinoculants as well as plant growth stage. The enhanced denitrification process at the flowering stage can be attributed to the increased production of root exudates [63].

The present study focusses not only on the ‘non-target’ beneficial effects of consortium (ABP) on the soil resident bacterial community, but also on finding the most suitable replacement of fertilizer by eco-friendly means by using three bioinoculants (A. chroococcum A-41, B. megaterium MTCC 453, and P. fluorescens MTCC 9768) upon their release in the rhizosphere of C. cajan. It was observed that integrated system of nutrient supply by employing bioinoculants with different doses of fertilizer can lead to optimum and economical yield of C. cajan along with improvement in soil fertility as compared to application of fertilizer alone. Combination of bioinoculants with 75% RDF proved to be the best, whereas with 50% RDF the effects were still comparable to the application of fertilizer alone. The combination of bioinoculants with fertilizer not only significantly improved plant growth, but also transformed the soil resident bacterial community in a way that was plant favorable. Hence, the application of bioinoculants in combination with 50% RDF can be practiced to attain desired productivity.

Authors’ contributions

RS and SS conceived and designed research, and analyzed the data. RS and VLS conducted experiments. SS procured funding. RS wrote the manuscript. SS reviewed the manuscript. All authors read and approved the manuscript.

Funding information

RS wishes to acknowledge fellowship received from Council of Scientific and Industrial Research, India, towards her doctoral work. This work was supported by the Department of Biotechnology, Govt. of India (Grant No. BT/PR5499/AGR/21/355/2012).

Data availability

All data generated or analyzed during this study are included in this published article [and its supplementary information files].

Compliance with ethical standards

Conflicts of interest

The authors declare that they have no conflicts of interest.

Ethics approval

This article does not contain any studies with human participants or animals performed by any of the authors.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Code availability

Not applicable.