Diversity and Phosphate Solubilization Efficiency of Phosphate Solubilizing Bacteria Isolated from Semi-Arid Agroecosystems of Eastern Kenya

Abstract

Phosphorus (P) is a major nutrient required for plant growth but it forms complexes with other elements in soil upon application. A cost-effective way of availing P to plants is by use of Phosphate Solubilizing Bacteria (PSB). There is a wide range of PSB suited for diverse agro-ecologies. A large part of Eastern Kenya especially the lower altitude regions are semi-arid with nutrient depleted soils and predominated by low-income smallholders farmers who do not afford costly inorganic fertilizers. To alleviate poor soil nutrition in this agroecosystem, we sought to study the diversity of phosphate solubilizing bacteria and their phosphate solubilization efficiency. The bacteria were selectively isolated in Pikovskaya’s agar media. Bacterial colonies were enumerated as Colony Forming Units and morphological characterization determined by analyzing morphological characteristics. Genetic characterization was determined based on sequencing of 16S rRNA. A total of 71 PSB were isolated and they were placed into 23 morphological groups. Correlation analysis showed a negative correlation between phosphate solubilizing bacteria and the levels of phosphorus, iron, calcium, magnesium and soil pH. Analysis of 16S rRNA sequences revealed that the genetic sequences of the isolates matched the strains from the genera Burkholderia, Pseudomonas, Bacillus, Enterobacter, Pantoea, Paraburkholderia, Cronobacter, Ralstonia, Curtobacterium, and Massilia deposited in NCBI Database. Analysis of Molecular Variance showed that variation within populations was higher than that of among populations. Phosphate solubilization index values ranged between 1.143 and 5.883. Findings on biodiversity of phosphate solubilizing bacteria led to identification of 10 candidate isolates for plant growth improvement and subsequently, bio-fertilizer development.

Introduction

Phosphorus (P) is the second most essential nutrient after nitrogen that is required by plants for growth and development. ¹ It plays a vital role in various physiological and biochemical activities including respiration, photosynthesis, transduction, cell division, biosynthesis of macromolecules and tissue development. ² Lack of Phosphorus is characterized by formation of brown leaves, and its deficiency leads to poor plant development and delayed maturity. ³ P is found as mineral deposits in the earth’s crust as a finite supply. ² Most of it is found as apatite, oxyapatite or hydroxyapatite which is insoluble. ⁴ It occurs in the soil in 2 forms; organic and inorganic. The proportions of these forms found in the soil differ in different places due to soil fertility management methods, soil types and soil use. ³ Soluble organic phosphate is in form of orthophosphate and organic polyphosphates. Plants mainly absorb P in the form of phosphate anions, most of which are HPO₄²⁻ and H₂PO₄⁻. ⁵

In most soil, P is deficient making it one of the plant growth limiting nutrient. ⁶ This is because it forms complexes with other elements upon application to the soil hence becoming unavailable for plant use. ⁷ Formation of the complexes depends on soil type and the pH. ⁸ In acidic soil, P form complexes with Al and Fe oxides, while in alkaline soil, it forms complexes with calcium. ⁹ Since phosphorus supplies are finite, ways of reclaiming this chemically bound P in the soil are continuously being exploited.^4,10 This promotes sustainable agriculture which in the long run will lead to mitigation of negative climate change effects.

Soil microorganisms have been shown to have abilities of utilizing the natural reservoir to siphon out scarce nutrients and thereby enriching the soil with important nutrients. ¹¹ Plant Growth Promoting Rhizobacteria (PGPR) is a group of soil bacteria associated with the plant rhizosphere that have the ability of promoting growth in plants.^12,13 Plant growth promotion is mainly by production of important metabolites required by plants, including phytohormones and nutrients. ¹⁰ A number of bacteria, fungi, actinomycetes and algae have the capabilities of improving plant growth through various mechanisms. ¹⁴ Particularly, bacteria have proven to be the most effective and their population is higher in the plant rhizosphere. ¹³ These bacteria enhance plant growth through the provision of phosphorus and other important plant growth-promoting metabolites. ¹⁵ Several mechanisms in which micro-organisms solubilize phosphates have been reported. ¹⁶ PSB are believed to solubilize P through secretion of organic acid which lowers the pH, chelation reaction of ions bound to P and by competing with P for adsorption sites in the soil. ¹⁷

Soil bacteria which have been demonstrated to be powerful phosphate solubilizers are from genera Bacillus, Pseudomonas, Rhizobium, Enterobacter, and Burkholderia.^18-22 Other reported phosphate solubilizers include species from genera Rhodococcus, Arthrobacter, Serratia, Chryseobacterium, Xanthomonas, Klebsiella, Agrobacterium, Azotobacter, Erwinia, Kushneria, and Pantoea.^12,19,23-26 The occurrence, abundance, diversity and bioactivity of PSB vary in different soils. The variation is attributed to the different soil properties including the nutritional conditions and physiochemical properties. ²⁷ Phosphate solubilization is influenced by several factors including interactions with other microorganisms, agronomic activities, ecological conditions, and soil types. ²⁸

Plant growth promoting microorganisms including phosphate solubilizers have been harnessed and used to develop biofertilizers. Biofertilizers are microbial inoculants applied to soil to improve fertility and enhance crop growth. ²⁹ Their use is recommended because they are non-toxic, cost-effective and eco-friendly. Microbial inoculants influence soil fertility through mineralization, decomposition and release of plant growth-promoting metabolites. ³⁰ To develop efficient microbial inoculants, continuous screening of natural biodiversity of soil microorganisms is undertaken and potential organisms screened for growth improvement potential. Development of effective molecular techniques has aided in identification and classification of a wide variety of PSB.^31,32 This study was aimed at determining the biodiversity of phosphate solubilizing bacteria in smallholder agroecosystems in Eastern Kenya and to assay their solubilization efficiency.

Materials and Methods

Study site and sample collection

Soil samples were collected from the dry regions of Tharaka-Nithi (S 0°9′42″ E 37°50′44″), Embu (S 0°29′8″ E 37°41′19″) and Kitui (S 1°11′36″ E37°51′29″) counties in Eastern Kenya from smallholder farmers. Tharaka-Nithi lies at 882 m above sea level and receives annual rainfall of 860 mm per year with average temperature of 22.3°C. Embu lies at 1137 m above sea level with annual rainfall of 1120 mm and average temperature of 20.2°C. Kitui is 1141 m above sea level with an annual rainfall of 1068 mm and average temperature of 21.4°C (Source: http://www.en.climate-data.org/Africa/Kenya). These regions experience 2 rainy seasons per year in the months of March to June and Ocober to December. ³³ The selected farms in which the samples were taken were under cultivation and had no history of bioaugmentation with PSB or other biofertilizers. Sampling was done after crop harvesting at the end of planting season on maize, cowpea, beans, millet, and green-gram fields. The roots of plants and the immediate rhizospheric soil were collected in a clean sterile khaki bag at various sampling points from the chosen farms. Samples from each farm were air-dried then mixed to obtain homogenous composite sample which was then sieved through 2 mm diameter sieve. Soil Physio-chemical analysis were performed for total C, total N, exchangeable cations (Mg, K, Ca), micronutrients (Fe, Cu, Zn, and Mn) and soil pH using established protocols. ³⁴

Selective isolation of Phosphate solubilizing bacteria

Ten grams of the soil sample was suspended in 90 ml of sterile distilled water and continuously agitated for 1 hour in a shaker. Serial dilution was prepared up to 10⁻⁵ and aliquots of 100 µl were plated evenly on Pikovskaya’s agar (0.5 g yeast extract, 10 g glucose, 5 g Ca₃(PO₄)₂, 0.5 g (NH₄)₂SO₄, 0.1 g MgSO₄.7H₂O, 0.002 g MnSO₄.2H₂O, 0.2 g KCl, 0.002 g FeSO₄.7H₂O, and 15 g agar per L). ³⁵ The samples were incubated for 6 days at 28°C. The bacterial colonies which formed a clear halo zone in the plates were selected and purified by streaking single colonies in fresh Pikovskaya;s agar medium. PSB and total bacteria population were enumerated and expressed as colony-forming units (CFU) ²⁴ :

$$CFU/gsoil=\frac{numberofcolonies\times dilutionfactor}{Volumeofthealiquot}$$

Phosphate solubilization index assay

Pure single colonies were streaked into the middle of the Pikovskaya’s agar plate and then incubated at 28°C for 7 days. The diameter of the bacteria and the cleared zone was measured on the fourth day and used to calculate the Phosphate Solubilization Index (PSI) using the equation below ³⁶ :

$$Phosphatesolubilisingindex(PSI)=\frac{\begin{array}{l}halozonediameter\\ includingcolonydiameter\end{array}}{colonydiameter}$$

Morphological characterization

The isolates were morphologically characterized by streaking on agar plates and observing the colony characteristics based on Bergey’s Manual of Systemic Bacteriology. ³⁷ The observable characteristics used to distinguish them included the colony shape, color, opacity, size, elevation surface texture, and surface form. Gram staining was also done to confirm gram reaction of the bacterial isolates. Bacterial isolates with similar gram reaction and morphological characteristics were grouped together.

Molecular characterization

Pure colonies of isolated bacterial cells were grown on PVK agar media for 2 days and used for DNA extraction. They were transferred to 400 µl of sterile normal saline and mixed thoroughly. The mixture was centrifuged at 13 000 rpm for 10 minutes to obtain a pellet. DNA was extracted using Zymo Research Quick-DNA™ miniprep kit as per the manufacturer’s protocol. The quality of the isolated DNA was checked by running gel electrophoresis using agarose gel and visualizing on UV trans-illuminator.

PCR was performed using universal primers, 27f (5′AGAGTTTGATCCTGGCTCAG 3′) and 1492r (5′GGTTACCTTGTTACGACTT 3′) which are complimentary to the highly conserved regions of the bacterial 16S rRNA gene. ³² PCR master mix was prepared by mixing 1 µl of 10 µM dNTPs, 0.5 µl of both 27f and 1492r primers, 0.5 µl of taq polymerase, 2.5 µl of 10X dream taq buffer and DNase, RNase free PCR water for top up to a total of 24 µl. 1 µl DNA template was then added.

Amplification was performed on Techgene Thermal Cycler (Techne) programed as follows: an initial denaturation step of 3 minutes at 94°C followed by 35 cycles of denaturation at 94°C for 45 seconds, annealing for 45 seconds at 51.8°C and extension for 2 minutes at 72°C. The final extension was at 72°C for 5 minutes. PCR products were stained with SYBR green stain and resolved in 1.4% agarose gel in 0.5X TBE buffer at 80 V for 30 minutes and then visualized on a UV trans-illuminator. Thereafter, PCR products were sequenced from both ends using both forward (27f) and reverse (1492r) primers.

Data analysis

Morphological diversity indices were calculated using PAST software version 3. Redundancy analysis (RDA) was used to show relationship between soil properties and colony-forming units using Canoco software version 5. The phosphate solubilization index was analyzed using non-parametric Kruskal Wallis test H using the Minitab software version 17. Consensus sequences were prepared from the sequenced data using BioEdit software version 7.2.5. Basic Local Alignment Test (BLAST) was used to draw comparison from the available bacterial standard sequences in NCBI GeneBank (http://www.ncbi.nlm.nih.gov/) for bacterial strain identities. Genetic Phylogenetic tree was constructed using MEGA X software. Sequences were aligned using ClustalW and the evolutionary history was inferred using the Neighbor-Joining method computed with the p-distances. Nucleotide diversity was calculated using DnaSP 6 software. Sequenced data was converted to haplotypes and used to calculate Analysis of Molecular variance (AMOVA) and genetic differentiation using Arlequin software version 3.5.2.2.

Results

Morphological characterization

A total of 71 isolates that formed a clear halozone in Pikovskaya’s agar media were isolated. Based on their morphological characteristic, they were placed into 23 groups (Table 1). The isolates exhibited varied morphological characteristics. Colony sizes varied from medium to large while their texture was either glistening, dull or mucoid. Colony shape was either circular, spindle or punctiform while their elevation varied from convex and flat to raised. The color of the isolates varied from yellow to white. All the isolates except those in group I and K were Gram negative.

Table 1.

Morphological characteristics of the isolates.

GROUP	SHAPE/ FORM	COLOR	COLONY SIZE	OPACITY	TEXTURE	ELEVATION	GRAM STAIN
A	Circular	Cream Yellow	Medium	Opaque	Glistening	Convex	Neg
B	Circular	Cream white	Small	Opaque	Glistening	Flat	Neg
C	Circular	White	Medium	Opaque	Dull	Convex	Neg
D	Spindle	White	Small	Opaque	Dull	Raised	Neg
E	Circular	Cream white	Small	Opaque	Glistening	Convex	Neg
F	Circular	Yellow	Medium	Opaque	Glistening	Convex	Neg
G	Circular	Cream white	Medium	Opaque	Glistening	Convex	Neg
H	Punctiform	White	Small	Opaque	Glistening	Flat	Neg
I	Circular	White	medium	Transluscent	Glistening	Flat	Pos
J	Circular	White	Large	Opaque	Mucoid	Convex	Neg
K	Circular	Cream white	Medium	Opaque	Glistening mucoid	Convex	Pos
L	Spindle	White	Small	Transluscent, Opaque center	Glistening	Convex	Neg
M	Spindle	White	Medium	Opaque	Glistening	raised	Neg
N	Circular	White	Medium	Transluscent	Glistening	Convex	Neg
O	Circular	White	Large	Transluscent	Glistening	Convex	Neg
P	Circular	Yellow	Medium	Transluscent	Glistening	Convex	Neg
Q	Circular	White	Small	Opaque	Dull	Flat	Neg
R	Circular	White	Small	Transluscent	Glistening	Flat	Neg
S	Circular	White	Medium	Transluscent	Glistening	Convex	Neg
T	Punctiform	Cream Yellow	Small	Opaque	Glistening	Convex	Neg
U	Circular	White	Medium	Transluscent, Opaque center	Glistening mucoid	Convex	Neg
V	Circular	White	Small	Opaque	Glistening	Convex	Neg
W	Circular	White	Small	Opaque	Dull	Convex	Neg

Based on morphological diversity Embu had the highest number of individuals at 30 while Kitui had the lowest at 17, (Table 2, Figure 1). According to Shannon H diversity and Dominance D, the variation among the 3 regions was low. Kitui had the highest diversity evenness while Embu had the lowest.

Table 2.

Diversity indices of isolates based on morphological characteristics.

	THARAKA-NITHI	EMBU	KITUI
Taxa_S	17	17	14
Individuals	24	30	17
Dominance_D	0.07292	0.08	0.07958
Simpson_1-D	0.9271	0.92	0.9204
Shannon_H	2.73	2.682	2.589
Evenness_eH/S	0.902	0.8598	0.9508
Brillouin	2.047	2.094	1.849
Equitability_J	0.9636	0.9467	0.9809

Figure 1.

Individual rarefaction diagram showing isolates per region and taxa at 95% confidence.

Relationship between population of phosphate solubilizing bacteria and soil properties

In all the regions, the proportion of total microorganisms that were able to grow in the plates were higher than the phosphate solubilizing bacteria (Table 3). PSB formed colony units ranging from 1.3 × 10⁴ to 3.63 × 10⁴ per gram of soil. On the other hand, total microorganism colony forming units ranged from 2.327 × 10⁵ to 3.507 × 10⁵ per gram of soil. Sample T4 had the lowest percentage of PSB to total microorganism at 3.45% while sample E1 had the highest at 12.93%.

Table 3.

Tukey’s Studentized Range (HSD) Test for CFU. %PSB is the proportion of Phosphate solubilizers to the total bacteria. T1-T5; Tharaka-Nithi samples, K1-K5; Kitui samples, E1-E5; Embu samples.

SAMPLE	PSB ×104 CFU/g SOIL	TOTAL 104 CFU/g SOIL	% PSB TO TOTAL
T1	1.67 ± 0.067e	35.07 ± 0.706b	4.75
T2	3.30 ± 0.058ab	28.60 ± 0.173de	11.54
T3	3.63 ± 0.088a	25.70 ± 0.252g	14.14
T4	1.30 ± 0.058e	37.63 ± 0.186a	3.45
T5	1.60 ± 0.058e	25.83 ± 0.273g	6.19
K1	1.43 ± 0.033e	29.60 ± 0.100d	4.84
K2	1.33 ± 0.067e	34.90 ± 0.265b	3.82
K3	3.50 ± 0.058a	27.17 ± 0.555efg	12.88
K4	2.47 ± 0.088d	23.47 ± 0.203h	10.51
K5	2.77 ± 0.088cd	32.70 ± 0.208c	8.46
E1	3.50 ± 0.058a	27.07 ± 0.176efg	12.93
E2	3.40 ± 0.115a	29.33 ± 0.240d	11.59
E3	3.53 ± 0.067a	27.50 ± 0.321ef	12.85
E4	2.90 ± 0.058c	25.97 ± 0.203fg	11.17
E5	2.93 ± 0.088bc	23.27 ± 0.145h	12.61

^*Means with the same superscript letter are not significantly different. P < .05

Redundancy analysis (RDA) between colony forming units and select soil properties are displayed in Figure 2. There was a positive correlation between PSB colony forming unit and K, Zn and Na. On the hand, there was a negative correlation between PSB colony forming unit and the available P, soil pH, Ca, Mg, and Fe. Total bacteria colony forming unit was positively correlated to available P but was negatively correlated to nitrogen, calcium, carbon, and magnesium.

Figure 2.

Redundancy analysis (RDA) showing relationship between Colony Forming Units (CFU) and soil properties. CFU PSB- Phosphate solubilizing bacteria, Total BAC- CFU of total bacteria.

Molecular characterization

Amplification of 16S rRNA gene of the isolates resulted in single bands of approximately 1550 base pairs (Figure 3). Genetic sequencing of the amplified 16S rRNA gene revealed 94% to 100% similarity of the isolates to other DNA sequences deposited at NCBI database (Table 4). These isolates belong to the genera Burkholderia, Bacillus, Pseudomonas, Pantoea, Enterobacter, Cronobacter, Massilia, Curtobacterium, Caballeronia, Paraburkholderia, Ralstonia, Erwinia and Citrobacter. Sequences of all the isolates were deposited in NCBI GenBank and were assigned Accession numbers as indicated in Table 4.

Figure 3.

Gel electrophoresis of amplified 16S rRNA on 1.4% agarose gel. Lane M – Quick-load^® 2-log DNA ladder (0.1-10 kb), Lanes 5 to 37- amplified samples of the isolates, Lanes A1 to A3- non-amplified samples.

Table 4.

Isolates reference matches based on 16s rRNA sequencing.

Isolate	Group	Total no. of isolates	Match	% Gene similarity	Accession no.
12	A	8	Burkholderia cenocepacia	99.86	MW013466
22	B	5	Burkholderia cepacia	100	MW013470
51	C	7	Massilia sp.	99.93	MW013481
15	D	5	Ralstonia pickettii	94.90	MW013467
5	E	4	Burkholderia ambifaria	100	MW013462
42	F	4	Pseudomonas oryzihabitans	99.93	MW013476
55	G	3	Erwinia sp.	99.72	MW013482
60	H	4	Pantoea eucrina	100	MW013485
41	I	3	Curtobacterium citreum	99.57	MW013475
29	J	3	Cronobacter sakazakii	99.86	MW013471
9	K	3	Bacillus amyloliquefaciens	98.39	MW013465
7	L	3	Citrobacter sp.	97.95	MW013464
18	M	3	Enterobacter sp.	99.86	MW013469
58	M	1	Enterobacter sp.	99.93	MW013484
47	N	2	Cronobacter turicensis	99.93	MW013478
57	O	1	Pantoea stewartii	99.79	MW013483
36	P	2	Pseudomonas psychrotolerans	99.95	MW013473
33	Q	2	Burkholderia contaminans	100	MW013472
17	R	2	Pseudomonas putida	99.86	MW013468
46	S	1	Cronobacter dublinensis	99.93	MW013477
50	T	2	Caballeronia calidae	99.64	MW013480
6	U	1	Pantoea septica	99.63	MW013463
37	V	1	Burkholderia territorii	99.93	MW013474
49	W	1	Paraburkholderia phenoliruptrix	99.93	MW013479

Phylogenetic analysis

The evolutionary history was inferred using the Neighbor-Joining method and the evolutionary distances were computed using the p-distance method (Figure 4). The sum of branch length in the optimal tree shown is =0.776. Bootstrap test of 1000 replicates were used to cluster associated taxa and the percentage of replicate trees are shown next to the branches. Based on the sequence data, phylogenetic analysis clustered the isolates into 2 main clusters (I and II) (Figure 4). The first cluster (cluster I) grouped together isolate 9 and isolate 41 with strains from the genera Bacillus and Curtobacterium supported by bootstrap values of 100. Isolate 9 had 98% match with Bacillus amyloliquefaciens strain while isolate 41 had 99.57% match with Curtobacterium citreum (Table 4). Cluster II had the highest number of isolates and formed 2 main sub-clusters (A and B). The first main sub-cluster (sub-cluster A) had 2 sub-clusters; sub-clusters A1 and A2. Sub-cluster A2 grouped isolate 15 with strains from the genus Massilia with 99.93% match and supported by bootstrap value of 100. Sub-cluster A1 formed 2 minor sub-clusters; A1.1 and A1.2. Sub-cluster A1.1 grouped together Isolates 5, 12, 33, 22, 37, and 50 with strains from the genera Burkholderia, Caballeronia and Paraburkholderia with sequence similarity of between 99.64% and 100%. Sub-cluster A1.2 grouped Isolate 15 with strains from the genus Ralstonia with sequence match of 94.9% supported by bootstrap value of 100. The second main sub-cluster B formed 2 sub-clusters; B1 and B2. Sub-cluster B1 grouped together Isolates 17, 36, and 42 with isolates from the genus Pseudomonas having sequence matches of 99.86%to 99.95%. Sub-cluster B2 further formed 2 minor sub-clusters; B2.1 and B2.2. Sub-cluster B2.1 grouped isolates 29, 47, and 46 with isolates from the genus Cronobacter with sequence similarity of 99.86% to 99.93%. Sub-cluster B2.2 grouped Isolates 55, 60, 7, 58, 18, 57, and 6 with isolates from the genera Pantoea, Enterobacter, Erwinia and Citrobacter with sequence homology of between 99.79% and 100%.

Figure 4.

Phylogenetic tree showing genetic relationship between isolates based on the Neighbor-Joining method. Evolutionary distances were computed by p-distance method and are in the units of the number of base differences per site. Cluster levels are indicated by labels in red at the branches.

Genetic diversity

Molecular diversity of the isolates is displayed in Table 5. Generally, isolates from Tharaka-Nithi were more diverse compared to those from other regions. Isolates from Tharaka-Nithi had the highest number of Segregating sites- S at 1266 followed by Kitui at 1219 while Embu had the lowest at 880. The number of haplotypes- h ranged from 5 (Tharaka-Nithi) to 11 (Embu). Isolates from all the region had high haplotype diversity- Hd of 1. In contrast, nucleotide diversity was relatively lower in all the regions, with isolates from Tharaka-Nithi having nucleotide diversity- Pi of 0.651 while those from Kitui and Embu had 0.456to 0.329 respectively. The highest values of nucleotide diversity based on Juke’s Cantor- PiJC was obtained from Tharaka-Nithi isolates at 2.47 while the lowest was from Embu (0.48).

Table 5.

Molecular diversity of PSB isolates from the 3 regions.

	Number of segregating sites, S	Number of haplotypes, h	Haplotype diversity, Hd	Average number of differences, K	Nucleotide diversity, Pi	Nucleotide diversity with Juke’s Cantor, PiJC
THARAKA-NITHI	1266	5	1.00	854.90	0.651	2.47
EMBU	880	11	1.00	432.84	0.329	0.48
KITUI	1219	8	1.00	598.93	0.456	0.81

Genetic differentiation

Analysis of variance showed that there was significant variation (P < .05) between isolates from various populations. The variation was high within population (92.05%) but low among population (7.95%) (Table 6).

Table 6.

Analysis of Molecular Variance (AMOVA) for 23 isolates from 3 populations based on 16S rDNA sequences.

Source	Df	SS	VC	% Mol var.	P-Value
Among populations	2	947.83	24.67 Va	7.95	.039
Within populations	21	6002.04	285.81 Vb	92.05	<.001
Total	23	6949.88	310.48
Fixation index	FST :	0.07946

Abbreviations: Df, Degrees of freedom; SS, sum of squares; VC, variance components; % Mol var., percentage molecular variance.

Based on distance method, a significant population pairwise differentiation was observed between Tharaka-Nithi and Embu population at p < .05 (Table 7). However, pairwise differentiation analysis demonstrated that there was no significant differentiation (p < .05) between Tharaka-Nithi and Embu populations and Kitui and Embu populations.

Table 7.

Population pairwise F_ST difference based on distance method.

	THARAKA-Nithi	EMBU	KITUI
THARAKA-Nithi	0.000
	*
EMBU	+0.1699	0.000
	0.00488 ± 0.0020	*
KITUI	−0.025	−0.0434	0.000
	0.23438 ± 0.0131	0.15723 ± 0.0120	*

Matrix of significant F_ST P values, Significance Level = 0.0500.

Screening for Phosphate solubilization

The isolates had different phosphate solubilization index (PSI) (Table 8). Non-parametric Kruskal-Wallis test was used to analyze Phosphate Solubilization Index (PSI) because the values did not fulfil the assumptions of ANOVA. Bigger clearance zone in relation to colony size is an indication of greater solubilization. In this study, only the bacteria that were able to retain their ability to solubilize P were characterized. The lowest PSI value was 1.143 (isolate 7) while the highest was 5.883 (isolate 22). A total of 15 isolates had PSI values greater than 4 representing 21.1% of all the isolates. Fourteen isolates had PSI values of between 3 and 4 making up 19.7% of the total isolates. Twenty-eight isolates had PSI value of between 2 and 3 representing 39.5% of the total isolates. The number of isolates with PSI value of less than 2 was 14 which represents 19.7% of the total.

Table 8.

Kruskal-Wallis test on Phosphate Solubilization Index (PSI).

ISOLATE	MEDIAN	AVE RANK	ISOLATE	MEDIAN	AVE RANK
1	3.125	136.8	37	4.833	204.2
2	4.500	188.5	38	1.857	35.3
3	3.333	148.0	39	4.286	180.5
4	4.000	170.0	40	2.200	65.0
5	3.375	152.3	41	3.200	139.2
6	2.500	86.5	42	3.200	142.0
7	1.143	2.0	43	4.600	193.2
8	1.545	9.7	44	1.800	26.5
9	2.500	86.2	45	4.800	199.7
10	2.571	101.8	46	4.167	177.5
11	2.667	108.8	47	2.500	89.3
12	3.333	150.3	48	2.000	48.0
13	3.000	128.3	49	1.800	34.2
14	1.714	19.7	50	1.667	17.3
15	2.000	45.2	51	1.500	7.5
16	2.167	63.2	52	3.000	126.0
17	2.500	89.3	53	3.143	137.8
18	2.714	112.7	54	3.750	162.8
19	1.857	41.7	55	3.750	162.8
20	1.833	31.7	56	2.857	125.3
21	2.500	89.0	57	3.333	149.5
22	5.833	212.0	58	4.200	179.0
23	4.800	199.7	59	2.500	84.8
24	2.333	73.5	60	3.750	162.8
25	3.714	161.5	61	2.250	66.0
26	4.833	204.2	62	1.667	16.2
27	2.000	45.2	63	1.500	7.5
28	2.600	108.3	64	4.167	176.3
29	4.600	192.0	65	2.000	43.0
30	1.833	37.7	66	2.600	101.5
31	2.500	89.3	67	2.750	113.2
32	2.833	123.5	68	4.286	181.7
33	4.857	206.7	69	2.000	45.2
34	2.833	120.0	70	2.333	73.5
35	2.143	59.8	71	2.500	90.0
36	1.833	37.7	Overall		107.0

H = 210.29 DF = 70 P = .000.

H = 210.43 DF = 70 P = .000 (adjusted for ties).

Discussion

We sought to determine the diversity and efficiency of phosphate solubilizing microorganisms with the goal of identifying suitable strains that can be used to improve soil fertility and productivity in the semi-arid regions of Eastern Kenya. As the world population grows, so does the demand for food and in order to meet this demand, there is need to develop eco-friendly and sustainable farming techniques. Plant growth nutrients are getting depleted from the soil due to a number of factors including finite supply and soil erosion. ³⁸ To guarantee crop yields, farmers have to replenish the soil with fertilizers and this increases the production cost. Nitrogen, phosphorus and potassium are the most important nutrients needed by plants for growth. ¹ Use of phosphate solubilizing microorganisms as a way of providing P to plants is gaining momentum since it is safe and effective.

The population and diversity of phosphate solubilizers were determined. All the isolates were able to solubilize insoluble phosphate and are therefore potential plant growth promoters. It is imperative to note that only those bacteria which are culturable in selective media were characterized in this study. Using morphological data, Diversity indices were calculated to check the diversity of the isolates. Results on diversity analysis indicated that Embu had higher number of isolates than the other region. This can be attributed to the fact that the region receives higher amount of the rainfall compared to others and therefore there is increased farming activities. ³³ Increased land tillage would probably favor proliferation of soil microorganisms. ³⁹

The population of culturable PSB was in the range 1.3 × 10⁴ and 3.63 × 10⁴ per gram of soil. These values was within the ranges of a similar study conducted in Morocco which reported values of between 0.0021 and 7.24 × 10⁵ CFU g-1 soil. ⁴⁰ However, the percentage of culturable PSB to total culturable bacteria was lower than the finding of the same study. In their study, Nannipieri et al, ⁴¹ showed that soil properties and land use affect the population of soil bacteria. The isolated PSB were negatively correlated to P level in the current study. Soil pH have also been shown to have an impact on the abundance of PSB. ⁴⁰ This finding is in agreement with the finding of Ndung’u-Magiroi et al, ²⁴ who showed that the population of PSB is higher in soils with low phosphorus levels. One mechanism in which PSB work is believed to be through production of organic acid which lowers the pH. ⁴² Al, Ca, Fe, and Mg are the main ions which adsorb the phosphate ions and therefore their level in the soil is correlated to PSB population. The action of PSB will lead to liberation of these elements from phosphate complex. In the current study, the levels of these elements were negatively correlated to the population of PSB as was also demonstrated by Zheng et al, ⁴² in their study.

Genetic sequencing using 16S rRNA is a popular and reliable method of identifying bacteria according to Yang et al. ⁴³ In this study, all the isolates were identified using the highly conserved 16S rRNA gene. Majority of the isolates had sequences that matched strains from the genus Burkholderia. Several members which showed great genetic similarity to Burkholderia cepacia complex were isolated and this is in agreement with similar study by Draghi et al, ⁴⁴ who isolated several members of the genus Burkholderia in Argentinean soil. The genus Paraburkholderia is a group of Burkholderia with nitrogen fixing ability. ⁴⁵ The family Burkholderiaceae also consists of the genus Caballeronia and Ralstonia which were isolated in this study. In their similar study, Kailasan and Vamanrao ⁴⁶ reported isolation of Ralstonia pickettii which was an effective phosphate solubilizer from Pomegranate Rhizosphere. To the best of our knowledge, there is little literature about the isolated species of Caballeronia calidae being a plant growth promoter and therefore its isolation in this study provides an area of interest for further characterization. The genus Massilia belong to the family Oxalobacteraceae and can inhabit broad range of niches. Several studies have classified Massilia as rhizosperic and endorhizal colonizers and they have also been shown to have plant growth promoting properties as demonstrated by Kong et al. ⁴⁷

The genera Pantoea, Enterobacter, Citrobacter, and Cronobacter belong to the family Enterobacteriaceae. ⁴⁸ They were reported as plant growth promoters by Büyükcam et al. ⁴⁹ In most instances, Pantoea have been isolated from the environment as demonstrated by Chakdar et al, ⁵⁰ in their study. Cronobacter have been isolated from food, environment and clinical samples and a study by Zeng et al, ⁵¹ revealed that they can cause rare ailments in immune compromised individuals. Citrobacter are found in water, soil and human intestines and it utilizes citrate as sole carbon source. ⁵² The genus Pseudomonas is one of the diverse groups of bacteria found in water, soil, plants and animal tissues. It belongs to the family Pseudomonadaceae and can tolerate a range of physical environment. Isolates with great gene match to Pseudomonas oryzihabitans and Pseudomonas putida from this study had the ability to solubilize phosphate as was also demonstrated by Leontidou et al, ⁵³ in their study.

Bacillus is one of the most studied and diverse bacteria of the family Bacillaceae. Some members from this genus have been proven to be among the best phosphate solubilizers. In this study, an isolate with a sequence match to Bacillus amyloliquefaciens was isolated and Fan et al,. ⁵⁴ in their study showed that it has the ability of improving plant growth. Curtobacterium belongs to the family Microbacteriaceae and their primary habitat is soil and water. ⁵⁵ Curtobacterium citreum was isolated in the current study and it was shown to be a great phosphate solubilizer.

Analysis of molecular variance showed a low variability among the populations but the variability within the populations was high. This suggests that there is limited physical barriers to gene flow in the populations as shown by Muthini ⁵⁶ in his study. Nucleotide and haplotype diversity are commonly used in populations as a measure of genetic diversity. Haplotype diversity is the representation of probability that 2 alleles which are randomly sampled from genetic sequence are different while nucleotide diversity is the number of nucleotide differences for very site in pairwise comparisons of genetic sequence. Isolates from the 3 population displayed the maximum Haplotype diversity which is an indication that their haplotypes were highly diverse. Low nucleotide diversity shown by the population is an indication that they share common haplotypes with relatively small divergence. The nucleotide diversity based on Juke’s Cantor varied across the population. This confirms the fact that the isolates were highly diverse. This large variation is probably due to a wide range of bacteria families that have the ability of solubilizing the phosphate.

Burkholderia cepacia (Isolate 22) had the highest solubilization index in the current study. This solubilization index was slightly higher than the value obtained by Pande et al, ⁵⁷ in their similar study. This isolate form a good potential plant growth promoter and further analysis are ongoing on the same. A total of 15 isolates representing 21.1% of total isolates had PSI value greater than 4. These isolates are from the following groups; Burkholderia cenocepacia, Burkholderia cepacia, Burkholderia ambifaria, Curtobacterium citreum, Cronobacter sakazakii, Bacillus amyloliquefaciens, Enterobacter sp., Burkholderia contaminans, Cronobacter dublinensis, Caballeronia calidae, and Burkholderia territorii. It is worth noting however that some of the isolates from the above groups had PSI values less than 4. This could be due to the different soil properties and also different microbial interactions at their isolation points. Availability of adequate P in the soil may make the P solubilizers less active and therefore reduce their solubilization potentials. All the isolates belonging to Pseudomonas had PSI values less than 4. However, their efficiency in improving plant growth may be better due to production of other growth improving metabolites. ⁵⁸

The ability of microorganisms to solubilize P ensures that there is adequate P in the soil for plant uptake. Once the ideal effective phosphate solubilizing bacteria have been isolated, they can be inoculated into the soil where they will convert insoluble P to a form that can be utilized by plants. This will lead to establishment of sustainable cropping systems with reduced application of costly chemical fertilizers.

Conclusion

Based on morphological analysis, 71 isolated PSB bacteria were placed into 23 groups. Representative of these groups were identified by genetic sequencing and they showed high genetic variability. This variability provides a good gene-pool of capturing organisms that are able to solubilize phosphates. Ten isolates which showed effectiveness in solubilizing the phosphate were identified and further analysis are on-going to assay their effectiveness in plant growth improvement. The end goal of this process is the development of effective PSB that will be used in the development of low-cost biofertilizers.