Is coconut coir dust an efficient biofertilizer carrier for promoting coffee seedling growth and nutrient uptake?

Bacterial isolation from soil and root tissue of coffee rhizosphere

Rhizosphere soil and root tissue samples of coffee (Coffea arabica L.) were collected from abandoned coffee plantations (N18O 92′37N99O 32′78E), referring to coffee plantations without farmer management such as chemical fertilizer, pesticides, herbicides, and soil preparation. The average age of coffee plants was 17 years. Land use history was received by the interview from farmers.

At each location, five coffee plants of comparable diameter, stem, height, and canopy were selected. The average values of stem diameter and plant height were 76.6 mm and 2.64 m, respectively. The average diameter of the coffee plant canopy was approximately 200 cm. Soil samples were directly collected around the rhizosphere of coffee plants using a shovel, and then soil adhering to the root after shacking was manually collected. At each soil sampling, three points were sampled and pooled as a representative sample. In every soil sampled, root samples were directly taken at 1–15 cm of depth from three sections of fine roots and combined. After sampling, soil and plant samples were immediately placed in an ice box before being transported to the laboratory for further soil microbial analyses.

Root samples were washed three times in sterile and distilled water, respectively. Subsequently, their surfaces were sterilized with 70% ethanol for 5 min and then in 2.6% sodium hypochlorite solution for 5 min. Finally, the samples were washed ten times with sterile distilled water to remove the NaOCl. The coffee roots were cut into one mm lengths and placed in a microcentrifuge tube contained 30 µL of sterile water. Root samples were softly ground by a stirring rod. Root suspension on the stirring rod tip was taken for streaking on tryptic soy agar (TSA) and incubated at room temperature for 7 days for bacterial colony formation. Colonies were selected and purified by subculturing three times on TSA media contained (g/L⁻¹) 3 g glucose, 2.5 g K₂HPO₄, 5 g NaCl, 3 g soy, 7 g trypton and 15 g agar with pH of 7.0 before their storage in a 40% glycerol solution at −80 °C until further analysis.

To isolate bacteria from rhizosphere soil, 10 g of soil was diluted in sterile water to concentrations of 10⁻¹, 10⁻², 10⁻³, 10⁻⁴, and 10⁻⁵, and 100 µL of the sample was pipetted onto TSA medium, using the sterile water as a blank control, and incubated at room temperature for 7 days with other plates under the same conditions. The streak plate method was used to purify the bacteria after incubation. Pure cultures were stored on slants at 4 °C and in a 25% glycerol stock at −80 °C for future experiments.

Screening and identification of bacterial isolates

All the bacterial isolates were qualitatively screened for nitrogen fixation on Burk’s N free medium (pH 7.2) contained (g L⁻¹): 10 g glucose, 0.52 g K₂HPO₄, 0.41 g KH₂PO₄, 0.05 g Na₂SO₄, 0.2 g CaCl₂, 0.1 g MgSO₄ 7H₂O, 0.0025 g Na₂MoO₄ ⋅2H₂O, 0.005 g FeSO₄7H₂O and 15 g agar, and incubated at 30 °C for 7 days (Weaver & Danso, 1994).

All nitrogen fixing bacterial isolates were qualitatively tested for phosphorus solubilization and Indole 3-acetic acid (IAA) production. To determine the phosphorus solubilizing ability, the isolate was inoculated into 25 mL of pikovskaya broth (PVKB) contained: 5 g Ca₃(PO₄)₂, for 72 h at 25 °C with constant shaking at 125 rpm. After 72 h, Pikovskaya (PVK) broth was centrifuged at 4,500 rpm for 10 mins, and supernatants were taken and checked for soluble phosphorus content in the culture broth using the molybdenum blue method (Murphy & Riley, 1962). As a control, autoclaved PVK broth without bacteria was incubated under the same conditions as the other samples.

IAA production was determined by a colorimetric method (Gordon & Weber, 1951). The bacterial isolates were cultured in NA liquid medium (with L-tryptophan) in the dark for 3 days at constant shaking at 125 rpm. After 3 days, the culture medium was centrifuged at 4,500 rpm for 10 min. One mL of supernatants (clear zone) were mixed with two mL of Salkovskii reagent to develop a color, and the IAA production was measured using the absorbance at 530 nm with a UV spectrophotometer (Ahmad, Ahmad & Khan, 2005; Bose, Shah & Keharia, 2013).

Small subunit ribosomal RNA (SrRNA) gene sequencing

S2-4a1 and R2-3b1 isolated from the coffee rhizosphere soil and root tissue, were chosen due to their ability to solubilize mineral phosphates and potassium. The genomic DNA of both bacterial isolates was extracted as described by Murray & Thompson (1980) and identified by amplifying and sequencing the 16S rRNA gene with the universal primers forward 785F (GGATTAGATACCCTGGTA) and reverse 907R (CCGTCAATTCMTTTRAGTTT). The nucleotide sequences were compared with those of the National Center for Biotechnology Information (NCBI) (http://www.ncbi.nlm.nih.gov) using the Basic Local Alignment Search Tool (BLAST) software.

Shelf-life determination of the inoculant

Carrier material and inoculum preparation

The carrier materials used in this study were perlite, vermiculite, diatomite and coconut coir dust. Perlite (∼2-4 mm, SiO₂ ≥75 wt%), vermiculite (∼1–3 mm, SiO₂ ≥41 wt%) and diatomite (∼1−1.5 mm, SiO₂ ≥90 wt%) were purchased from Shijiazhuang Huabang Mineral Products Co., Ltd. (China). While coconut coir dust (∼0.25 mm) was provided by the factory in Thailand. To prepare the carrier materials, 50 g of each carrier material was placed into a plastic bag, which was then sterilized by autoclave at 121 °C for 1 h. The carrier material was sterilized again 3 days after the first sterilization.

In this study, the S2-4a1 and R2-3b1 isolates were chosen for the development of inoculants due to their higher ability to solubilize mineral phosphates and potassium and produce IAA as compared to other isolates. These selected isolates were incubated in molasses low yeast (MLY) at 12 rpm at 28 °C for 7 days until reaching a concentration of 1 × 10⁸ CFU mL⁻¹. After incubation, 15 mL of bacterial supernatants were put into the sterile carrier materials and incubated at 25 °C for 90 days. To measure the survival of bacteria, each inoculated carrier was sampled at 0, 7, 15, 30, 45 and 60 days after incubation. The pH, EC and moisture content of the inoculated carrier were also determined at sampling.

Physico-chemical characteristics of carrier materials

Soil pH of all carriers were determined was measured by a pH meter using soil-water suspensions (1:5 v/v) and electrical conductivity (EC) by digital probe (Seven Excellence pH meter; Mettler Toledo, OH, USA).

Coffee seedling growth promotion by the S2-4a1 and R2-3b1 isolates

Local Arabica coffee (Chiang Mai 80) seeds used in this study were surface-sterilized with 3% hydrogen peroxide for 3 min and subsequently rinsed with autoclaved distilled water five times. The sterilized seeds were sown into sterilized sand for 60 days after sowing (DAS), and then coffee seedlings of equal height were transplanted into black polyethylene bags (one plant per pot) filled with 1 kg of autoclaved media containing a mixture of soil and coconut coir dust based bioformulations (3:1 (w/w)) for the tests. The chemical analysis of the media for seedlings is presented in Table 1. The experiment was arranged in a completely randomized design with three replications, involving four treatments inoculated with the liquid formulation (S2-4a1 and R2-3b1 isolates) and an uninoculated control: (i) control (no inoculation), (ii) soil inoculated with coconut coir dust based bioformulations of S2-4a1 and (iii) R2-3b1 isolate and (iv) a mixture of S2-4a1 and R2-3b1 isolates. The inoculum preparation for S2-4a1 and R2-3b1 isolates was described above. The media treated with uninoculated MLY broth were used as controls. According to the mixture treatment, these bacteria were grown separately and then mixed in equal proportions at the time of application. All the experiments were performed in a greenhouse with 75–80% humidity and temperatures between 25 and 30 °C. Irrigation was supplied manually to keep the media moisture at the field capacity.

Determination of plant nutrient uptake and media physico-chemical properties

The coffee plants were destructively harvested after 90 days of treatment to determine the dry mass of leaves, branches and roots, which were then dried in an oven with forced-air circulation at 55 °C until a constant mass was obtained. From each plant tissue, 0.5 g of dry biomass was subjected to nitric-perchloric acid digestion to analyze total nitrogen (N), phosphorus (P), potassium (K), calcium(Ca) and magnesium (Mg) content (Bremner, 1996). Total nutrient uptake was based on the nutrient content in the whole plants coffee biomass (roots, shoots, leaves) (Eq. (1)). (1)${\displaystyle \text{Total nutrient uptake}=\left(\text{biomass}\times \text{nutrient concentration}\left(\text{\%}\right)\right)/100.}$

Determination of nutrient transfer factor

In this study, the nutrient transfer factor (NTF) was used as an indicator of the ability of the inoculated bacteria to improve nutrient uptake efficiency by transferring available nutrients into soil that can be directly absorbed by plants. The NTF was calculated by dividing the concentration of nutrients in the whole plants by the nutrient concentration in the media after the harvested plants. To determine the physico-chemical properties of the experimental media after harvest, the available phosphorus (P) of the media were determined by UV–VIS Spectrophotomete and the available K, Ca and Mg detection by ICP-OES (ICAP 6000 SERIES, Thermo Fisher Scientific, Waltham, MA, USA).

Statistical analysis

Statistical analysis was conducted using analysis of variance (ANOVA). The least significant difference (LSD) was performed to compare the mean differences among treatments, and P value < 0.05 was considered for the significant level. Linear regressions were carried out to determine the spatial variations in the pH and EC and their changes with times. The interaction effects of carrier type and time on the bacterial population, pH, and EC were quantified using a two-way ANOVA. The statistical software used was SPSS 19.0 (SPSS, Inc., Chicago, IL, USA).

Isolation and characterization of mineral solubilizing microbes

The S2-4a1 and R2-3b1 isolated from rhizosphere soil and root tissue were chosen due to their high abilities to solubilize phosphate and produce IAA as compared to other isolates. The S2-4a1 and R2-3b1 showed their ability to solubilize phosphate with 407.1 and 301.7 mg L⁻¹, respectively (Table 1). However, negligible differences were observed for potassium solubilizing ability and IAA production between S2-4a1 and R2-3b1 isolates. Two isolates (S2-4a1 and R2-3b1) were identified using 16S rRNA gene sequencing for comparison with NCBI sequences using the BLASTn algorithm. The result found that S2-4a1 and R2-3b1 had a 99% similarity with Bacillus megaterium and Bacillus sp., respectively.

The survival of bacteria

A suitable carrier material should be able to stimulate and provide an environment for microbial development and the survival of the inoculated bacteria over long-term storage. In this study, the different carriers (coconut coir dust, diatomite, vermiculite, perlite) inoculated with both bacterial S2-4a1 and R2-3b1 isolates were evaluated. The population of S2-4a1 inoculated in coconut coir dust, vermiculite and perlite and R2-3b1 inoculated in coconut coir dust, diatomite and vermiculite continuously increased during the first 30 days after inoculation (DAI) (Fig. 1). The highest population of S2-4a1 was observed in diatomite carrier, but there was no significant difference as compared to coconut coir dust and vermiculite (P < 0.05). While the highest population of R2-3b1 was found in the coconut coir carrier. At 45 and 60 DAI, there was no significant difference in the population of both S2-4a1 and R2-3b1 among carriers (P < 0.05), indicating that coconut coir dust could be used as an alternative carrier for developing the formulation of biofertilizer.

Effect of carriers on pH and EC

The different carriers used in this study significantly influenced the pH and EC after bacterial inoculation (P < 0.01) (Table 2). The pH of diatomite, vermiculite and perlite carriers inoculated with both S2-4a1 and R2-3b1 was significantly higher than coconut coir dust during the incubation periods (Fig. 2). The pH ranges of coconut coir dust inoculated with S2-4a1 and R2-3b1 isolate were 6.25 to 6.99 and 6.29 to 6.90, respectively. While the highest electrical conductivity (EC) was observed in coconut carriers inoculated with both bacterial isolates as compared to other carriers (Fig. 3). The ranges of EC in coconut coir dust, diatomite, vermiculite and perlite inoculated with S2-4a1 isolate were 4.18 to 4.67, 0.72 to 0.99, 1.52 to 1.70 and 1.60 to 1.69 mS cm⁻¹, respectively. For the carrier inoculated with R2-3b1, the EC ranges were 4.14 to 4.68, 0.84 to 0.94, 1.36 to 1.60 and 1.70 to 1.78 dS cm⁻¹ for coconut coir dust, diatomite, vermiculite and perlite, respectively. During inoculation, only the coconut coir dust carrier inoculated with both bacterial isolates exhibited significant decreases in pH and EC (P < 0.001) in the present study (Figs. 2 and 3).

Interaction between carriers and inoculation time

The two-way ANOVA indicated that carrier and time revealed significant effects on both S2-4a1 and R2-3b1 populations, pH and EC (P < 0.05), as shown in Table 3. Based on the S2-4a1 isolate, the interaction of carrier and time had significant effects on the S2-4a1 population (P < 0.01), pH (P < 0.05) and EC (P < 0.01). While the interaction of carrier and time had no significant effects on R2-3b1 population and pH. The impact of environmental factors on both S2-4a1 and R2-3b1 populations varied with carrier types (Fig. 4). There were no any environmental factors (pH and EC) significantly correlated with both the S2-4a1 and R2-3b1 populations. However, the responses to carrier substrates of both S2-4a1 and R2-3b1 populations was significantly different. The S2-4a1 populations were significantly correlated with vermiculite and perlite, while the response of the R2-3b1 populations to carrier substrates was not observed in this study.

Influence of isolated bacteria on biomass and nutrient uptake of coffee seedling under greenhouse conditions

In this study, two selected isolates (S2-4a1 and R2-3b1) were tested for their growth enhancement effect on coffee seedlings. The results showed that inoculated treatments increased the total biomass of the seedlings compared to the uninoculated control (Fig. 5). However, the isolate S2-4a1 increased biomass significantly more than the R2-3b1 isolate (P < 0.05). In addition, the inoculation of the S2-4a1 isolate increased the nutrient uptake of coffee seedlings (Fig. 6), which increased the N, P, K, Ca and Mg uptake by 48%, 48%, 40%, 80% and 56%, respectively, while the R2-3b1 isolate increased 21% (N), 18% (P), 18% (K), 22% (Ca) and 24% (Mg) when compared to the inoculated control. Interestingly, inoculation with the mixture of both tested isolates did not show a significant difference in biomass and nutrient uptake as compared to sole inoculation with the R2-3b1 isolate.

Pearson’s correlation analysis was performed to determine the relationships between the plant P, K, Ca and Mg uptake, P, K, Ca, Mg transfer factor and plant biomass at 90 DAS (Fig. 7). The plant biomass was positively correlated with the plant P (R² = 0.59 P < 0.01), K (R² = 0.95 P < 0.01), Ca (R² = 0.75 P < 0.01) and Mg uptake (R² = 0.89 P < 0.01). The plant biomass attained a positive correlation with the K (R² = 0.85 P < 0.01), Ca (0.52 P < 0.01), and Mg (0.51, P < 0.01) transfer factor but was not significantly correlated with the P transfer factor (R² = 0.22, P > 0.05).