Introduction

In the present day agriculture, intensive farming practices that achieve high yield require fertilizers. However, inappropriate agricultural intensification coupled with reckless use of fertilizers has deteriorated soil quality. Therefore, there is growing awareness on the use of environment friendly sustainable nutrient management practices that lay emphasis on restoration and maintenance of soil quality both in the short- and long-term. Thus, effective biological technologies like the use of plant growth-promoting rhizobacteria (PGPR) are being exploited for enhancing crop yields. PGPR represent a wide variety of rhizosphere-inhabiting bacteria which colonize the root systems of plants and can stimulate plant growth by direct or indirect mechanisms. Direct mechanisms of plant growth-promotion include biofertilization, stimulation of root growth, rhizoremediation, and plant stress control, while mechanisms of biological control include reducing the level of disease, antibiosis, induction of systemic resistance, and competition for nutrients and niches [19]. Common PGPR include genera Acinetobacter, Alcaligenes, Arthrobacter, Azospirillum, Azotobacter, Bacillus, Beijerinckia, Burkholderia, Enterobacter, Erwinia, Flavobacterium, Rhizobium, and Serratia [4]. At present, PGPR are being increasingly used in combination with manures and fertilizers for improving crop yields and have contributed to the development of sustainable agricultural systems.

Studies have shown that PGPR had positive effects on cereals [27], fruits [18], vegetables [15], and spices [5]. PGPR have also been found to improve plant uptake of nutrients, and thereby increase the use efficiency of applied fertilizers and manures thus allowing reduced application rates of fertilizers [1]. While studies on PGPR effects on growth and yield of crops is widely reported, very little information exists on the effects of PGPR applied alone or in combination with graded doses of fertilizers on sensitive biochemical and microbial indices reflecting soil quality.

Studies employing a host of physical and chemical parameters have been made to determine the effects of nutrient inputs on soil quality. However, the present focus is on biochemical and microbial parameters that reflect the size and activity of microbial processes. This is because the biochemical properties are more sensitive to environmental stress, play a major role in degradation, and provide rapid and accurate estimates on soil quality [14]. The biochemical parameters include variables directly related to microbial activity (microbial biomass C and N, respiration, etc.), and the activities of extracellular hydrolytic enzymes involved in the C, N, S, and P cycles in soil. These soil biochemical and microbiological parameters have been considered as potential indicators of soil quality and management impacts in numerous studies [10–12, 32, 33].

Exclusive use of only fertilizers have been found to cause a significant reduction in microbial biomass C [34, 35] as well as reductions in soil respiration (SR) and dehydrogenase (DH), acid phosphatase (AcP), and β-glucosidase (βG) activities [12]. We hypothesized that use of PGPR along with fertilizers would attenuate such negative or insignificant effects of sole NPK application on important soil microbial and biochemical processes. Hence, the primary objective of the study was to determine the effects of promising native strains of PGPR (Burkholderia cepacia, Klebsiella sp., Serratia marcescens, and Enterobacter sp.) isolated from ginger rhizosphere, applied alone and in combination with NPK fertilizers (urea, rock phosphate [RP], muriate of potash [MOP]) on sensitive biochemical and microbial parameters that reflect short-term changes in soil quality. A secondary objective was to determine the inter-relationships between these parameters in soils applied with PGPR and fertilizers.

Materials and Methods

Experiment Details

The experiment was conducted at Indian Institute of Spices Research, Kozhikode, Kerala and was repeated for 3 years from 2007 to 2009. For the study, earthen pots of 20 kg capacity were filled with 15 kg sieved soil (<2 mm). The soil was a clay loam Ustic Humitropept. The initial properties of the soil are pH 5.12, organic C 14.2 g kg−1, mineral N 42 mg kg−1, Bray P 5.4 mg kg−1, exchangeable K 84 mg kg−1. For planting of rhizomes, small shallow pits were made and the seed-rhizome (20–30 g) of ginger (var. Varada) with at least two sprouted buds after pre-treatment with the respective PGPR was placed 3.5–5.0 cm deep in the pits and the soil-pressed over it.

Treatments

Inorganic Fertilization

The recommended dose (RD) of NPK for ginger in Kerala State, India is 75–50–50 kg ha−1, respectively. The inorganic sources of NPK used were urea, RP and MOP, respectively. RP was applied as basal, urea in two splits (45th and 90th day after planting [DAP]) and MOP in two splits (45th and 90th DAP). This fertilization regime was considered as 100 % RD and each nutrient was reduced to either 75 or 50 % and was applied alone or in combination with PGPR (Tables 2, 3).

Isolation of PGPR

Rhizosphere soils were collected from healthy ginger growing sites in Calicut and Wayanad districts (Kerala State, India) and Kodagu district (Karnataka State, India). PGPR were isolated using dilution plate technique using Tryptic Soy Agar. The most suitable dilution was selected for estimating the population of the rhizobacteria and expressed as colony forming units per gram soil. Colonies on each plate were distinguished based on phenotypic characteristics such as shape, motility, color, growth rate, culture morphology, and Gram’s staining. The representative isolates (100 nos.) were selected and cryopreserved at −80 °C in glycerol (40 %) for further use.

Selection of Isolates

The isolates were then tested for nutrient mobilization in vitro using standard parameters viz., production of indole acetic acid, solubilization of phosphorus, potassium, silica, and zinc. Four best isolates were then shortlisted for the study and preliminary identification of the isolates was done using the Bergey’s Manual of Determinative Bacteriology. Identity of the isolates was confirmed using 16S rDNA sequence analysis and Biolog [7]. The four PGPR used in the experiment were identified as B. cepacia (GRB25), Klebsiella sp. (GRB36), S. marcescens (GRB38), and Enterobacter sp. (GRB70). The basic traits of the shortlisted PGPR for nutrient mobilization are given in Table 1.

Table 1 Basic information on shortlisted PGPR for nutrient mobilization traits in vitro
Full size table

Application of PGPR

The shortlisted PGPR were applied both as rhizome dip and soil application. Just before planting, the healthy ginger rhizome seeds were dipped in 1 % starch solution containing bacterial cells at the rate of 1 × 108 cells mL−1 for 1 h. Immediately after planting, 100 mL of 1 × 108 cells mL−1 suspension was applied to each pot. The soil application was repeated on 30th, 60th, and 90th day of planting. The PGPR were either applied alone or in combination with varying levels of NPK (Tables 2, 3). The study also consisted of an absolute control where no nutrients, whatsoever, were applied. The experiment consisted of six replications. The temperature in the greenhouse during the experiment hovered between 28 and 33 °C and relative humidity between 65 and 82 %.

Table 2 Effects of PGPR, inorganic NPK fertilizers and their combinations on pH, mineral N, Bray P, exchangeable K, soil organic C (SOC), dissolved organic C (DOC), and dissolved organic N (DON) levels in soil at 120 days after planting of ginger
Full size table
Table 3 Effects of PGPR, inorganic NPK fertilizers, and their combinations on microbial biomass-C, -N, -P, net N mineralized (N MIN), soil respiration, metabolic quotient, soil dehydrogenase (DH), urease (UR), acid phosphatase (AcP), and β-glucosidase (βG) activities in soil at 120 days after planting of ginger
Full size table

Soil Sampling

The crop duration is 270 days. However, we collected soil samples 1 month after the final application of PGPR and fertilizers treatments, i.e., at 120 DAP. Apparently, it was presumed that the effects of the treatments would be more distinct at 120 DAP than at harvest. Besides, 120–135 DAP also happens to be the active growth stage of ginger. The soils samples were taken from bottom two-thirds of each pot, cleared of any organic debris and transferred for storage in sealed plastic bags. The soils were then sieved (<2 mm), analyzed for their moisture content, and stored at 4 °C. Subsamples for the determination of soil organic carbon (SOC) and mineral N were sieved to pass a 0.5 mm mesh.

Soil Physico-chemical Properties

SOC, NH4 + NO3-N, Bray P, and exchangeable K were estimated using standard methods [29]. Soil pH was measured in 1:2.5 soil:water suspension.

Soil Biochemical and Microbiological Analyses

Nitrogen mineralization (N MIN) capacity was determined by extracting 10 g soil with 50 mL of 2 M KCl for 30 min before and after incubation for 10 days at 30 °C. The NH4-N and total inorganic N were determined by steam distillation [20]. The difference between the values obtained before and after incubation indicates N mineralization capacity. Dissolved organic carbon (DOC) and dissolved organic nitrogen (DON) were determined by the method described by Smolander and Kitunen [28]. The fumigation–extraction method [33] was used to determine soil microbial biomass-C (C MIC), -N (N MIC), and -P (P MIC) as described by Heinze et al. [16]. SR was measured as the CO2 evolved from moist soil, adjusted to 55 % water holding capacity, and pre-incubated for 7 days at 20 °C in the dark. The CO2 production was then measured for the next 7 days using NaOH traps and titration with HCl. The metabolic quotient (\( q_{{CO_{2} }} \)) was calculated as follows: (μg CO2-C evolved in 7 days g−1 soil)/((μg biomass C g−1 soil)/7 days) × 1,000 = mg CO2-C g−1 biomass C day−1 [24]. Soil enzyme activities viz., dehydrogenase (DHA), urease (UR), AcP, and βG were estimated using standard methods [31].

Statistics

All values reported are average of results across the 3 years and are expressed on an oven-dried soil basis (105 °C). The significance of treatment effects was determined by ANOVA. Where the F values were significant, post-hoc comparisons of means were made using the Least Significance Test (lsd) at the 0.05 probability level. Pearson’s correlation coefficients (n = 25; P < 0.05) were calculated to determine the strength of the relationship between various parameters. Principal component analysis (PCA) was performed for reflection of any intrinsic pattern in the multidimensional data swarm. PCA often reveals previously unexpected associations among variables and thereby allows interpretation that would not be possible otherwise. Only PCs with Eigen values >1 and that explain >10 % of the total variance were retained.

Results

Soil pH, Mineral N, Bray P, and Exchangeable K

Soil pH ranged from 4.10 to 5.49 (Table 2) and among the treatments, mineral N, and Bray P levels were greatest in the treatment GRB38 + 100 % NPK (98.6 and 8.8 mg kg−1, respectively) and exchangeable K level was greatest in the treatment with GRB25 + 75 % N + 100 % PK (224 mg kg−1).

SOC, DOC, and DON

The levels of SOC did not exhibit marked variations across treatments, albeit showing a slight increase in treatments with NPK and PGPR applied alone or in combination (Table 2). However, the DOC level was lowest in control and increased marginally by 9.0–19.0 % in treatments with only NPK and 9.0–17 %, respectively in treatments with only PGPR. Greatest levels of DOC were, however, registered in the treatments involving PGPR + NPK. Mean level DOC in these treatments (165.4 μg g−1) indicated a 24 % increase compared to control. In contrast, DON levels increased significantly (Table 2) with increasing N levels indicating a 17.5 % increase in treatments with 75 % N (16.8 μg g−1) and a significant 48–71 % increase in treatments with 100 % N relative to control. No significant variation with respect to DON existed among NPK and PGPR + NPK treatments.

Microbial Biomass-C (C MIC), -N (N MIC), and -P (P MIC)

C MIC level under sole NPK application (mean 115 μg g−1) was almost identical to control (114 μg g−1) but lower than the treatments with sole application of PGPR (mean 128 μg g−1) and PGPR + NPK (mean 146.0 μg g−1; Table 3). PGPR application also positively influenced P MIC. Mean level in PGPR alone treatments was 12.2 μg g−1 indicating an increase of 45 % relative to control (Table 3). In treatments involving only NPK, the P MIC levels were, in general, greater in treatments with 100 % P suggesting positive effects of P fertilization. Relatively greater levels of P MIC were, however, registered in the treatments involving 100 % P + PGPR. Contrary to C MIC and P MIC, N MIC accumulated at markedly greater level in the treatments with NPK fertilization. The N MIC level was lowest in the control (6.3 μg g−1), while in treatments with only PGPR it averaged 11.9 μg g−1 and in treatments with sole application of NPK it averaged 15.6 μg g−1 (Table 3). This suggested large average increases of 89 and 148 %, respectively in these treatments relative to control. The degree of increase in N MIC levels was, however, greatest in treatments involving PGPR + NPK (mean 26.7 μg g−1) indicating a significant increase of 124 % compared to PGPR alone treatments and 71.0 % compared to treatments with only NPK.

Total N Mineralized, SR, and Metabolic Quotient

Sole application of PGPR had very little effects on N MIN rates (Table 3). Contrarily, NPK fertilization markedly enhanced N MIN rates; mean value across these treatments (76.8 mg N kg−1 10 day−1) being greater by 60 % compared to control. The results also showed a clear effect of combined application of PGPR and NPK on N MIN. Mean level across these treatments (86.0 mg N kg−1 10 day−1) suggested an increase of 79 % relative to control. Among the treatments, the greatest levels of N MIN were registered by the treatments with combined application of either of the PGPR with 100 % N. SR (Table 3) showed little variation among the treatments including control. The average \( q_{{CO_{2} }} \) levels in NPK alone and PGPR alone treatments were 181 and 192 mg CO2-C (g biomass C)−1 day−1, respectively (Table 3). Across PGPR + NPK treatments, \( q_{{CO_{2} }} \) levels (mean 153.0 mg CO2-C (g biomass C)−1) decreased by 15 % relative to NPK alone treatments and by 20 % relative to PGPR alone treatments.

Soil Enzyme Activities

Control recorded lowest activity of all enzymes (Table 3) and across treatments greatest activities were always recorded by those involving NPK + PGPR. DHA activity was on an average greater by 49.0 and 23.2 % in PGPR + NPK treatments compared to NPK and PGPR alone treatments, respectively. In NPK alone treatments, UR activity was on an average greater by 43 % compared to PGPR alone treatments, while AcP activity was on an average greater by 46 % in PGPR + NPK treatments relative to NPK alone treatments. The activity of βG was almost identical in NPK alone and PGPR alone treatments (3.5 and 3.3 μmol p-nitrophenol g−1 h−1, respectively) but lower by 29–33 % compared to PGPR + NPK treatments.

Discussion

Unlike soil physico-chemical properties, the treatments effects were more evident on soil microbial properties (Table 3). Our results suggested little or no change in C MIC levels in treatments with only NPK. This is most likely due to application of N fertilizer as urea at rates equivalent to 100 or 75 % N. No changes or reductions in soil C MIC that are attributable to N fertilization have been reported earlier [12, 22, 27]. The supply of readily metabolisable C (DOC) in the treatments with PGPR + NPK is likely to have been the most influential factor contributing to the C MIC and P MIC increases. This confirmed that the levels of C MIC is strongly related to the steady-state substrate availability in soils as reflected by the existence of strong correlations between C MIC and DOC (r = 0.80). Also, P MIC showed strong correlations with DOC (r = 0.74). Interestingly, C MIC levels in treatments with PGPR + NPK increased significantly even though it involved inorganic fertilization. This suggested that the application of PGPR attenuated the insignificant effects of inorganic fertilization on C MIC. In this study, other nutrients (P and K) besides N were utilized, although the influence of applied nutrients other than N on soil microbial biomass has largely been regarded as inconsequential [2].

The results also suggested a positive effect of N application on N MIC. After N application, N availability increased and consequently microbes immobilized N, which subsequently increased N MIC [35]. Positive correlation between N MIC and DON (r = 0.72) suggested that inorganic fertilization especially N enhanced DON levels. Earlier reports also indicate significantly positive effects of N fertilization on DON [12, 23]. A clear positive effect of combined application of PGPR and NPK on N MIN was observed. This suggested that increased soil microbial pool is often associated with high net N MIN rates [11] as indicated by the positive correlation between N MIC and N MIN (r = 0.87). This is further supported by positive correlations reported between N MIC and N MIN with UR activity (r = 0.81 and 0.84, respectively), which might be explained by the role played by microbial extracellular enzymes in the depolymerisation of N-containing polymers [25].

SR did not vary considerably among the treatments including control (Table 3), while the metabolic quotient (\( q_{{CO_{2} }} \)) in PGPR + NPK treatments, were considerably lower compared to NPK alone and PGPR alone treatments. The metabolic quotient (SR per unit of microbial biomass) reflects the maintenance energy requirement of soil microbes and can be a relative measure of how efficiently the soil microbial biomass is utilizing C resources, as well as the degree of substrate limitation for soil microbes [36]. The results in this study did show changes in the DOC pool that would suggest changes in labile C. It is, therefore, possible that this explained the difference in \( q_{{CO_{2} }} \) rates across treatments. Apparently, relatively greater \( q_{{CO_{2} }} \) levels in NPK and PGPR alone treatments and control suggested lower substrate quality, resulting in a lower C use efficiency or high maintenance C demand [6]. In contrast, lower \( q_{{CO_{2} }} \) under PGPR + NPK treatments indicated relatively more efficient microbial community and better use of the available organic substrates. Our results suggested that inorganic fertilization did not enhance SR. Similar effects on SR due to inorganic fertilization has been found when N fertilizer was added [21], possibly from the suppression of the decomposition of native SOC due to decrease in microbial biomass [13].

Similar to soil microbial biomass, the soil enzymes were activated to varying degrees by PGPR and NPK applied alone or in combination (Table 3). The stronger effects of PGPR + NPK application on DHA might be due to the greater metabolism by soil microorganisms and the poor influence of NPK fertilization, especially N on DHA is consistent with the results of Kautz et al. [17]. The study also indicated positive effects of N fertilization on UR and is in conformity with the results of Allison et al. [3]. Conversely, we observed an inhibition in AcP activity due to inorganic fertilization. This may be partly explained by the high P-fixing capacity of our soils [30] and partly by the negative effects of inorganic P application [26]. Similarly, the activity of βG was lowered in NPK alone and PGPR alone treatments compared to PGPR + NPK treatments. Our finding is supported by the results of Debosz et al. [8] who reported that the activities of βG was markedly lowered due to exclusive mineral N fertilization. The lower values of βG in NPK and PGPR alone treatments indicated that the potential to mineralize organic matter, and so the activity of the C-cycle, is reduced.

The inter-relationships between various soil parameters were studied using PCA (Table 4). Factor 1 (PC1) explaining 53 % of the total variance was defined mainly by mineral N, Bray P, exchangeable K, SOC, DOC, DON, N MIC, P MIC, N MIN, UR, and βG. This possibly reflected the strong relationship between the labile and easily mineralisable organic matter and microbial activity and the logical dependence of microbial biomass on soil nutrients. The negative loading of \( q_{{CO_{2} }} \) in this factor suggested a stressed microbial community with reduced substrate use efficiency in treatments with only PGPR or NPK or the control. Conversely, the strong loading of βG in this factor indicated a decrease of the microbial community maintenance energy requirement and higher C efficiency [9] in soil with PGPR + NPK. Also, positive loadings of UR, N MIN, and DON in PC1 suggested positive correlations between UR activity, N dynamics, and organic matter mineralization. Factor 2 (PC2) explaining 22 % of the total variance was defined mainly by DOC, C MIC, SR, DH, and AcP activities which suggested that enhanced levels of labile organic substrates positively influenced soil microbial and enzyme activities.

Table 4 Principal component loadings after Varimax rotation
Full size table

Conclusions

Results showed that PGPR or NPK applied alone consistently registered markedly lower levels of microbial biomass and activity. Contrarily, application of PGPR in combination with inorganic NPK promoted soil biological quality as evidenced by enhanced soil microbial biomass and enzyme activities. The non-significant effects of chemical fertilization on soil microbial properties were overcome to some extent by combined application of NPK and PGPR. Though the study did not indicate any reduction in the dose of NPK fertilizers due to application of PGPR, we found that the use of PGPR along with fertilizers attenuated the negative or insignificant effects of sole NPK application on important soil microbial and biochemical processes. However, considering that S. marcescens, Enterobacter sp., and Klebsiella sp. are opportunistic human pathogens, we recommend the use of B. cepaceae for enhanced soil quality under ginger. Overall, the study suggested that in cropping systems that overly depend on inorganic fertilization for enhanced yields, it would be ideal to combine crop specific native strain of PGPR to maintain or enhance soil quality even in the short-term.