Background

The relevance of phosphorus (P) as the second most important macronutrient cannot be overemphasized (Gupta et al. 2018). This P stimulates plant growth and maturity (Wortmann et al. 2014), whose deficiency limits plant growth (Yu et al. 2011). Despite its importance, P is much less abundant in soil globally. This is because out of the reserve of total P (typically within 0.2-5 g P kg− 1 with an average of 0.6 g P kg− 1) in agricultural soil (Zhang et al. 2017), a significant fraction is locked in an insoluble form in various organic and inorganic substrates (Zhang et al. 2018; Tang et al. 2020). Consequently, only < 10% of total P enters the plant-animal cycle (Panhwar et al. 2014; Zhang et al. 2017), making it a limiting plant nutrient. Therefore, it is quite common to apply phosphatic fertilizers to agricultural soil to ensure substantial phosphorus is available for plants. Only a small portion of total P in phosphatic fertilizers applied on agricultural soil is utilizable by plants, as majority (75–90%) of it, rapidly form complexes with soil constituents like Al, Fe and/or Ca (Toro 2007; Cao et al. 2018), making it highly unavailable for plant. Meanwhile, apart from the cost, accumulation of complex compounds resulting from excessive application of phosphatic fertilizer pose several environmental threats (Zhang et al. 2018).

Natural agricultural practices such as application of organic fertilizer (Pare et al. 2000; Dittmar et al. 2009), plant health management system using antagonists of phytopathogens (Zaidi et al. 2014, 2016) and plant growth promoting microorganisms (PGPM) as biofertilizers (Rodrigues et al. 2016; Zhang et al. 2017) that promise environmentally sustainable development, have currently been considered among diverse biotechnological approaches to enhance plant growth (Khan et al. 2013; Rizvi et al. 2014; Tang et al. 2020). More interestingly, promoting growth and nutrient absorption in plant by biofertilizers has recently been reported to have potential of reducing the use of chemical fertilizers by half without loss (Pereg and McMillan 2015), and also increasing plant tolerance to abiotic and biotic stresses through biocontrol (Olanrewaju et al. 2017).

Many studies have documented the isolation of PSB from different natural environment, among which include; rhizosphere soil (Zaidi et al. 2009; Sadiq et al. 2013; Wang et al. 2017), acid sulfate soil (Panhwar et al. 2014) and mushroom residues (Zhang et al. 2017). Despite the complexity of rumen as a habitat, reports suggests that rumen of cattle could harbour PSM. Considering that Africa’s ruminants graze on grass (Cohen 1980) and the P constituent of grass has been estimated to range from 0.2 to 1.5 g P kg− 1 (Tenikecier and Ates 2018), phosphate-solubilizing microbes are hence required to hydrolyse or enhance the metabolism of phosphorus in their rumen. Meanwhile, most of the microorganisms associated with the plant rhizosphere soil, classified as PGPM, particularly PSB are not only capable of solubilizing mineral phosphates, among other nutrients, they also promote associative/atmospheric nitrogen fixation; synthesize or induce production of some plant hormones such as indole-3-acetic acid (IAA) and gibberellic acid; siderophores; hydrolytic enzymes; and antimicrobial agents (Goswani et al. 2014; Olanrewaju and Babalola 2018).

Since production of certain organic acids (such as acetic acid), hydrolysing enzymes and regulating molecules (such IAA, HCN and siderophore) by bacteria are responsible for solubilizing insoluble phosphate (Dutta et al. 2015; Wang et al. 2017), we hypothesized that PSB with such traits in this study, promise good PGP performance. The study was therefore design to isolate and characterize PSB with plant growth promoting (PGP) potentials from rumen content of White Fulani cattle.

Results

Isolation of microorganisms from rumen content of White Fulani cattle

The bacterial population in the rumen content in Table 1 ranged from 5.20 ± 0.36 × 105 CFU g− 1 to 4.0 ± 0.07 × 106 CFU g− 1 while the total fungal count (TFC) ranged between 0.14 ± 0.02 × 105 CFU g− 1 and 0.06 ± 0.01 × 106 CFU g− 1. A total of 31 microbes; 74% bacterial and 26% fungal isolates, were recovered from the rumen content.

Following cultural, morphological and biochemical characterization, 56% of the total bacterial isolates were identified as gram negative and the rest 44% were gram positive. Bacterial isolates were tentatively identified as belonging to either of six genera with varying distribution including; Pseudomonas (6), Staphylococcus (4), Escherichia (4), Bacillus (3), Micrococcus (3), Klebsiella (2) and Streptococcus (1).

Phosphate solubilizing potentials of bacterial isolates

The ability of bacterial isolates recovered from the rumen content to solubilize P was qualitatively enumerated solubilization index (SI) (Fig. 1). Of the 23 bacterial isolates, only 61% could solubilize P on NBRIP medium and were grouped as Phosphate Solubilizing Bacteria (PSB). The isolates exhibited varying SI ranging from 0.43 ± 0.02 by P. aeruginosa RC22 to 1.32 ± 0.43 by P. aeruginosa RC7. The solubilization index exhibited by P. aeruginosa RC7 was significantly higher than those from other isolates at p ≤ 0.05.

Fig. 1
figure 1

Estimation of Phosphate Solubilization Index (SI) of bacteria isolated from rumen content of White Fulani cattle. SI were estimated after 4 d of incubation in a triplicate experiment recorded as Mean ± SE; Error bars indicates significant differences at p ≤ 0.05 with One-way ONOVA by Duncan’s test

Full size image

Plant growth promoting trait of PSB isolates

The plant growth promoting substances released by PSB isolated from rumen content varied considerably, according to results obtained from in-vitro assessment, presented in Online Resource Table S1. The test for siderophore production ability (Fig. 2) revealed that isolates P. aeruginosa RC7, E. coli RC13, E. coli RC16 and P. aeruginosa RC22 produced three forms of siderophore including catecholate, hydroxymate and carboxylate, while isolates Staphylococcus sp.RC3, P. aeruginosa RC10 and Staphylococcus sp.RC21 were not able to produce siderophore.

Fig. 2
figure 2

Venn chart showing the distribution of siderophore producing PSB isolated from rumen content. Hydroxymate and catecholate siderophore determined by ferric chloride test and carboxylate siderophore determined by phenolphthalein test

Full size image

Molecular characterization of rumen PSB with PGP traits

Among the 14 PSB isolated from rumen content, 10 strains with prominent PGP traits were selected and tagged rumen PSB isolates for further identification based on 16 S rRNA gene sequence. The BLAST results show that eight (8) of the PSB have high similarities with different strains four of which are Pseudomonas aeruginosa and other four were Escherichia coli. The sequence data were deposited in public domain and the Accession Numbers allocated to these rumen PSB strains are presented in Table 2.

Table 1 Microbial composition of rumen content from White Fulani cattle
Full size table

Concentration of PGP metabolites produced by rumen PSB

The result of quantitative assay for organic phosphate production by rumen PSB for a period of 7 days is depicted in Fig. 3. During incubation, there was a progressive increase in the concentration of phosphate released by the 10 rumen PSB strains with the highest concentration of 687.75 µg mL− 1 produced by P. aeruginosa AKRC7 on day 4. E. coli AKRC16 liberated the least concentration of 207.75 µg mL− 1 P after day 7. E. coli AKRC16 however released maximum (345 µg mL− 1 P) phosphate among Escherichia spp. after 5 days of growth.

Fig. 3
figure 3

Concentration of soluble phosphate produced in vitro by rumen PSB for seven days of incubation. Quantity of soluble P in NBRIP broth supplemented with 0.5% Ca3(PO4)2, assessed by molybdenum blue method and measured at 650 nm

Full size image

The pH of media used to assess phosphate solubilizing potential of strains of rumen PSB showed constant decrease in value, indicating increasing acidity (Fig. 4). Furthermore, highest initial pH value for all cultured broth was 6.8 but decreased through day 7 of incubation to the least value of 4.2.

Fig. 4
figure 4

Change in pH during in vitro solubilization of phosphate by rumen PSB. Change in pH of NBRIP broth inoculated with PSB, determined at 24 h interval for 168 h incubation period

Full size image

The concentration of IAA produced by rumen PSB after 5 days of incubation period depicted in Fig. 5 shows that the 10 isolates were able to produce varying amount of IAA. E. coli AKRC15 produced the least concentration of 25.05 µg mL− 1 IAA while P. aeruginosa AKRC7 liberated the highest concentration of 39.27 µg mL− 1 IAA. E. coli AKRC16 however produced the highest concentration of 30.52 µg mL− 1 IAA among the genus of Escherichia.

Fig. 5
figure 5

Concentration of Indoleacetic acid produced in vitro by rumen PSB. Data were recorded as Mean ± SE of three independent replicates; Error bars indicates significant differences at p ≤ 0.05 with One-way ONOVA by Duncan’s test

Full size image

Moreover, strains of rumen PSB produced variable amount of ammonia after day 5 of incubation ranging from 2.86 µM mL− 1 to 3.89 µM mL− 1 (Fig. 6). Among the genera of Pseudomonas, P. aeruginosa AKRC7 secreted the highest concentration (3.89 µM mL− 1) of ammonia while E. coli AKRC16 liberated the highest concentration (3.73 µM mL− 1) of ammonia among the genus Escherichia.

Fig. 6
figure 6

Concentration of Ammonia produced in vitro rumen PSB. Data were recorded as Mean ± SE of three independent replicates; Error bars indicates significant differences at p ≤ 0.05 with One-way ONOVA by Duncan’s test

Full size image

All rumen PSB exhibited varying productivity of phosphate solubilizing enzymes; phosphatase (ranging from 4.51 to 35.08 mmol mL− 1) and phytase (ranging from 0.32 to 2.98 mmol mL− 1), under different optimal conditions (Table 3). Overall, P. aeruginosa AKRC7 consistently exhibited highest productivity for both enzymes at the same temperature (35oC), pH (6), incubation time (24 h), and optimal source of nitrogen and carbon for phosphatase (tryptone and starch) and phytase (peptone and glucose) correspondingly. Similarly, among the genera, E. coli AKRC16 and E. coli AKRC4 exhibited, in that order, highest enzyme activities at same conditions, except for the former with higher optimal temperature (60oC) and the latter with low optimal temperature (35oC). E. coli AKRC16 recorded high optimum temperature of 60oC with phosphatase activity and phytase activity of 25.20 mmol mL− 1 and 0.52 mmol mL− 1.

Table 2 Identity of rumen PSB based on 16 S rRNA gene with respective accession number in GenBank
Full size table
Table 3 Phosphate solubilizing enzymes produced by rumen PSB
Full size table

Growth promoting potential of rumen PSB on seeds of Solanum lycopersicum

Table 4 showed that 70% of rumen PSB in this study displayed promoting potential on the growth of tomato seeds through seed bacterization, in varying degree compare to the control. P. aeruginosa AKRC7 and E. coli AKRC16 enhanced 100% germination of tomato seeds with significantly high vigour index (2177 and 2114.33), fresh weight (1.42 ± 0.01 and 1.37 ± 0.02 g) and P content (309.04 ± 0.05 mg g− 1 and 301.01 ± 0.02 mg g− 1) of seedlings respectively. Only about 62, 63 and 65% of total P content (8.74% w/w) of the pot soil were respectively accumulated and solubilized by E. coli AKRC16, P. aeruginosa AKRC7 and E. coli AKRC16 for seed growth, compared to control treatment which resulted in seeds with 199.59 ± 0.04 mg P g− 1 after accumulating 69.34% of P content of the pot soil. The pH values of pot soils after the greenhouse experiment were not significantly different at p = 0.05 but decreased by 7–12% of the initial pH of the soil before sowing.

Table 4 Promoting effect of rumen PSB on growth response of Solanum lycopersicum seed and phosphorus content of soil in greenhouse experiment
Full size table

Discussion

Microorganisms with phosphate solubilizing potential from rumen content of White Fulani cattle

In this study, a high microbial load was recorded (Table 1) which is in accordance with the findings reported by Miron et al. (2001) and Preedaa et al. (2016). The high microbial composition of the rumen content in this study could be attributed to the conducive condition of rumen with high nutitional composition, relatively constant temperature range, water and food ingested and the exocrine secretion, and saliva Nagaraja (2016). Several researcher have also reported the predominance of bacteria in rumen belonging to different genera including Methanobacteriun, Methanobrevibacter, Ruminobacter, Ruminococcus, Butyrivibrio, Treponema, Pseudomonas, Micrococcus, Staphylococcus, Streptococcus, Klebsiella, Clostridium, Bacillus and Escherichia (Pukall et al. 2009; Kelly et al. 2010; Morgavi et al. 2013; Nagaraja 2016. These reports support the diversity of bacterial isolates recovered and identified as belonging to different genera (Pseudomonas, Staphylococcus, Escherichia, Bacillus, Micrococcus, Klebsiella and Streptococcus) in this study. It is a fact that the nutritional composition of every habitat determines the types of microbes that proliferate in them. Daily turn over of phosphorus (77%) in the rumen of cattle by their saliva and total P of the grass grazed by Africa’s ruminant contributes to the solubilizing ability of bacteria isolated from the rumen content of the cattle. The assay for phosphate solubilizing ability of rumen PSB was done using TCP as the inorganic phosphate. This is in support of the work by Son et al. (2006) that reported Ca3(PO4)2 as better inorganic phosphate (Pi) for PSB solubilization.

PGP traits of PSB from rumen content

Production of PGP metabolites that influence the physiology and productivity of plants by controlling some specific metabolic activities is one of the major attributes of PSB with great potentials as biofertilizers (Davona et al. 2012; Khan et al. 2014). The rumen PSB isolated here possess varying potentials to promote plant growth (Table 2). Particularly, the ability of PSB to produce siderophores of one or more forms, which can be used by the organisms to solubilize irons from minerals or organic compounds during iron starvation (Indiragandhi et al. 2008) is in support of the report of Khan et al. (2014). According to Dutta et al. (2015), protease, cellulase and chitinase production by the PSB isolates is also an indication of the ability to promote plant growth. While cellulase aid nutrient mineralization and degradation of organic matter; protease and chitinase hydrolyse the proteins in the cell wall of other organisms thereby serving as biocontrol. Hence, production of these enzymes (chitinase, proteins and cellulose) coupled with secreation of HCN and siderophore could confer the isolates the ability to inhibit the growth of plant pathogens (Olanrewaju and Babalola 2018). Furthermore, it has been shown by Goswani et al. (2014) that several PSB that produced IAA, HCN, chitinase and siderophores also possessed biocontrol ability.

Identity of PSB with PGP traits isolated from rumen content

Molecular identification of the PSB that showed consistent PGP traits, revealed that eight (8) of them were strains of either Pseudomonas aeruginosa or Escherichia coli (Table 4). Our finding is consistent with previous reports of bacteria found in the rumen (Nagaraja 2016). However, while we report Pseudomonas with PSP from rumen, most previous reports (Muleta et al. 2013; Dutta et al. 2015) documented Phosphate-solubilizing Pseudomonas from plant rhizosphere, soil and river. As regards E. coli, the authors are not aware of any study that has reported with phosphate solubilizing potential. It is therefore novel and very interesting that these strains were isolated from rumen content of cattle.

Quantitative estimate of PGP metabolite produced by PSB from rumen content

In this study, Pseudomonas aeruginosa AKRC7 and Escherichia coli AKRC16 exhibited the highest phosphate solubilizing efficiency amoung their respective genera with 687.75 µg mL− 1 and 345 µg mL− 1 soluble phosphate released after fourth and fifth day respectively. The efficiency exhibited by these isolates could be responsible for their better growth properties when compared to other isolates and consequently resulted in decrease in pH of the medium (Fig. 5) and increased productivity of phosphate solubilizing enzymes, phosphatase and phytase (Table 4). This is in support of the result of Panhwar et al. (2014) and Gen-fu and Xue-Ping (2005) who observed a linear relationship between soluble phosphorus and microbial growth. Soluble phosphate liberated by E. coli AKRC16 at 5th day of incubation (345 µg mL− 1) and by P. aeruginosa AKRC7 at 4th day of incubation (687.75 µg mL− 1) in this study was much higher than the highest produced by a strain of Bacillus megaterium reported by Zheng et al. (2018). This could be an indication that the P. aeruginosa AKRC7 isolated from rumen content of White Fulani cattle in this study appears more promising as a potential bioinoculant based on its attributes (Panhwar et al. 2014). Further corroborating this, is their production of significantly high concentration of IAA and ammonia both of which are among important metabolites that stimulate plant growth and yield (Wang et al. 2018; Zheng et al. 2018).

Secretion of enzymes phosphatase and phytase, by some of our rumen PSB strains especially under feasible optimal conditions, is an added advantage, as it enhances hydrolysis of insoluble P to a more available and utilizable P content for plant usage (Wang et al. 2017). The source of our PSB strains could have contributed to the difference in the activities of enzymes they produced, when compared with those reported by other studies of different sources (Singh et al. 2014; Behera et al. 2017; Obidi et al. 2018). Moreso, the prospective application of above said enzymes in agricultural, biotechnological, pharmaceutical and other industries (Vohra and Satyanarayana 2003; Obidi et al. 2018; Handa et al. 2020), has also scored our rumen PSB as prospective candidate for industrial exploration.

Phosphate solubilization by bacterial isolates was accompanied by decline in pH of media which is in line with findings documented by other researchers (Paul and Sinha 2016; Zheng et al. 2018; Tang et al. 2020). Production of organic acid which accounts for acidification of the medium (decrease in pH) is one of the major mechanisms used by PSB to mineralise available phosphorus, hence the negative correlation of pH with concentration of phosphate released (Chen et al. 2006; Khan et al. 2007; Zaidi et al. 2009; Zheng et al. 2018).

Greenhouse study of rumen PSB on seeds of Solanum lycopersicum

Seed bacterization has been tremendously used in the recent time, to evaluate the plant growth promoting potential of several microorganisms (Singh et al. 2014; Kumar et al. 2015; Qessaoui et al. 2019; Arkhipova et al. 2019). Growth parameters of Solanum lycopersicum seeds bacterized with our rumen PSB were better enhanced coupled with increased soluble P than those of control, which is apparently indicative of potential of our isolates to provide better nutrient uptake (Tahir et al. 2015; Majeed et al. 2015). Growth promoting effects recorded by Pseudomonas aeruginosa AKRC7 on tomato seed in our study corroborates those reported of rhizobacteria Pseudomonas isolated by (Qessaoui et al. 2019). Remarkable improvement in growth response of tomato seeds in this study further justifies the plant growth promoting traits exhibited in vitro by our rumen PSB.

Conclusion

This study demonstrated that the rumen content of White Fulani cattle, indigenous to Africa served as a reservoir of novel PSB which exhibited plant growth promoting traits making them potential bioinoculants. Furthermore, Pseudomonas aeruginosa AKRC7 showed promising potential which could be exploited in sustainable agricultural production systems for optimizing the crop production. Also, the discovery of new strains of Escherichia coli, AKRC16 which is thermophilic and AKRC4, both exhibiting plant growth promoting potential under feasible conditions, is promising. Further works like molecular delineation PGP potentials, evaluation for biocontrol and large scale insitu application of these isolates for plant growth promotion, are however required to further elucidate their potential as biofertilizers.

Materials and methods

Sample collection

The sample (rumen content) used in this study was aseptically collected from the rumen of a White Fulani cattle (Online Resource Figure S1), using a pre-sterilized metal hand trowel into a sterile 20 mL sample bottle and transported in ice-chest within 2 h, to the Laboratory for analysis.

Isolation of microbial isolates from rumen content

With slight modification, serial dilution technique described by Olutiola et al. (2000) and pour plate method of Oje et al. (2016) were adopted to culture bacteria and fungi from the rumen content. Ten grams (10 g) amount of rumen content sample was homogenized for about 1 h in 90 mL of sterile physiological saline solution using manual agitation. The homogenate was ten-fold serially diluted and 1 mL aliquot of appropriate dilutions were inoculated into petri dish and overlaid with sterile molten Nutrient Agar (Oxoid, Basingstoke Hampshire, England) and Potato Dextrose Agar (Hi-Media Lab, India). The plates were gently swirled and allowed to set before incubated aerobically at 37oC (for 24 h) and 45oC (for 48 h) for bacteria and fungi, respectively. After incubation, microbial colonies were examined and counted with illuminated colony counter (Gallenkamp, England). The microbial counts were expressed as colony forming unit per gram (CFU g− 1) of the sample homogenate (Olutiola et al. 2000). Discrete colonies were sub-cultured to obtain pure culture which were stored in sterile slanted agar (at 4oC), until subsequently identification and further studies (Oje et al. 2016).

Phenotypic identification of bacterial isolates from rumen content

Bacterial isolates were tentatively identified using cultural characteristics on solid media, cellular morphology through microscopy of Gram-stained smear and biochemical reactions. The results were interpreted according to Holt et al. (1994).

Qualitative screening of rumen bacteria for phosphate-solubilizing potential

The method of Zhang et al. (2017) with little modification was adopted to screen the bacteria isolated from rumen content for ability to solubilize phosphate. Isolates were separately streaked on a National Botanical Research Institute’s Phosphate (NBRIP) growth (agar) (g L− 1: glucose (10); MgCl2 · 6H2O (5); MgSO4 · 7H2O (0.25); KCl (0.2) and (NH4)2SO4 (0.1)) incorporated with 0.5% Ca3(PO4)2 as an insoluble source of phosphorus (P) (Nautiyal 1999; Chauhan et al. 2017) and incubated at 30oC for 96 h (4 d). Solubilization Index (SI) which indicated phosphate solubilizing ability of the isolates was calculated using the following formular. Bacterial isolates with better solubilizing potential tagged “rumen PSB” were subjected to further studies.

$$\mathrm{SI}=\frac{\mathrm{Diameter}\;\mathrm{of}\;\mathrm{CH}}{\mathrm{Diameter}\;\mathrm{of}\;\mathrm{colony}\;(\mathrm{mm})\;\times\;\mathrm{Number}\;\mathrm{of}\;\mathrm{IP}\;(\mathrm{days})}$$

Where

SI:

is the solubilization index

CH:

is the clear halo around colony on NBRIP

IP:

is the incubation period

Quantitative assay of phosphate solubilizing potential by rumen PSB

Tricalcium phosphate (TCP) solubilization assay

Phosphate solubilizing potentials of PSB isolates were quantified following the method of Zhang et al. (2017) with slight modification on the volume of the media and inoculum used. A single colony of 24 h old PSB was inoculated into 3 mL NBRIP broth containing 5 g L− 1 Ca3(PO4)2 and incubated under agitation (190 rpm) at 30 ± 1 °C for 20 h. One millilitre (1 mL) of 20 h old broth culture of the PSB was then introduced into a 250 mL Erlenmeyer flask containing 120 mL medium in triplicate, incubated in the dark in a shaker incubator (New Brunswick Scientific Co., Inc., USA) (working at 190 rpm) at 30 ± 1 °C for 168 h (7 d), using sterile NBRIP medium as control. Subsequently, the pH of 20 mL aliquot of cultured broth was measured using portable pH meter (Model HI-96,107, Hanna Instrument) and afterwards spun in a centrifuge (Beckman Coulter Massachusetts, USA) at 13,000 rpm for 10 min. Molybdenum-blue method described by Chen et al. (1956) was used to assess the soluble P in the supernatant as the absorbance of resultant blue solution was measured using UV/Visible spectrophotometer (WPA Linton Cambridge, UK) at 650 nm. The value of total soluble phosphorus was estimated and expressed in µg mL− 1 using the regression equation of standard curve described by (Adebayo 2019).

Phosphate solubilizing enzymes (phosphatase and phytase) production assay

Quantitative assessment of phosphatase and phytase production by rumen PSB were carried out following the method of Obidi et al. (2018) and Gontia-Mishra et al. (2013) with slight modification respectively. Isolates (24 h old) were aseptically introduced into sterile 100 mL nutrient broth and incubated under agitation (150 rpm) for 24 h at 28oC. Five millimeters (5 mL) of culture suspensions were mixed with 100ml of sterile basal medium (consisting (L− 1) of 0.9 g K2HPO4; 0.2 g KCl; 0.2 g MgSO4.7H2O; 1.0 g NH4NO3; 0.002 g ZnSO4; 0.002 g MnSO4; 0.002 g FeSO4.7H2O; 2 g yeast extract; incorporated with 1 g of specific substrate (sodium phosphate for phosphatase and sodium phytate for phytase)), and incubated for 48 h at 150 rpm and 28oC. Cell-free supernatants were obtained at 6 h interval through centrifugation of 5 mL of resultant suspension at 6000 rpm at 4oC for 20 min.

Determination of phosphatase activity

A reaction solution of modified universal buffer (consisting of 3.025 g tris-hydroxymethyl-aminomethane, 2.9 g maleic acid, 3.5 g citric acid, 1.57 g boric acid, 122 mL of 1 M sodium hydroxide; in distilled water to a final volume of 250 mL; pH 6.5), cell-free supernatant and 0.115 M p-nitrophenyl phosphate (p-NPP), was prepare at 1:4:1 v/v ratio. Toluene (0.1 mL) was added to the solution (to stop bacterial growth) before incubated at 37oC for 1 h. Mixture of 4 mL of 0.5 M NaOH and 1 mL of 0.5 M CaCl2 was added to the solution to discontinue the reaction. Absorbance of the solution indicative of P-nitrophenol (p-NP) constituent was then measured with UV-Visible spectrophotometer (WPA Linton Cambridge, UK) at 450 nm and the value was extrapolated from standard curve of p-NP (Behera et al. 2017).

Determination of phytase activity

Reaction mixture of 0.2 mL of cell (resultant) suspension from production assay and 0.8 mL of acetate buffer (0.2 M at pH 5.5, containing 1 mM sodium phytate) were incubated at 37oC for 30 min. The mixture was added with 0.5 mL of stop solution (5% (w/v) trichloroacetic acid) and colour solution (10% ascorbic acid; 2.5% ammonium molybdate; 5 N H2SO4; and distilled water, at 1:1:1:2 v/v ratio) and incubated for 10 min at room temperature (El-Toukhy et al. 2013). Concentration of phosphate released was measured by correlating absorbance value in a UV-Visible spectrophotometer (WPA Linton Cambridge, UK) at 650 nm with standard curve of KH2PO4 (Adebayo 2019).

Optimization of growth conditions for enzyme production

Optimal conditions for production of phosphatase and phytase by rumen PSB were determined by evaluating the activities of the enzyme under such different varietal conditions as; pH (3–11) temperature (25-70oC); incubation time (0–60 h) and carbon sources. A unit of enzyme activity is the amount of such enzyme required to liberate 1 mmol Pi per minute under the assay condition (Obidi et al. 2018).

In-vitro assessment of rumen PSB for plant growth promoting traits

Cellulase and protease production assay

Casein Yeast Extract Agar (containing g L− 1: casein 5.0; yeast extract 2.5; glucose 1.0; agar 15.0 dissolved in distilled water) amended with 1% carboxymethyl cellulose (CMC) as described by Teather and Wood (1982) was inoculated with a single colony of 24 h old pure PSB culture and incubated for 72 h at 28 °C. After incubation time, the cultured agar was flooded with aqueous solution of Congo red (1 mg mL− 1) and left on bench for 15 min. Clear halo around bacterial colonies after 15 min reaction with Congo red indicated cellulase production by PSB isolates.

As described by Kumar et al. (2005), Casein Yeast Extract Agar supplemented with 7% skimmed milked powder was inoculated with PSB and incubated at 28 °C for 72 h to test for their Proteolytic activity. Protease production was indicated by presence of clear halo zone around bacterial colonies.

Hydrogen cyanide (HCN) production assay

Picrate assay described by Castric (1975) was adopted with little modification to assess test isolates for HCN production. Bacterial isolates were streaked on Nutrient agar plates supplemented with 4.4 g/% glycine. Whatman filter paper No. 1 soaked in solution of 2% Na2CO3 and 0.5% picric acid was sealed with parafilm in-between base and lid of the cultured plate and incubated for 96 h at 27 ± 2oC. Change in colour of filter paper from yellow to orange-brown indicated HCN production.

Chitinase production assay

PSB isolates were spot-inoculated on Chitin Agar plate (containing in g L− 1: chitin (4), K2HPO4 (0.7); KH2PO (0.3); MgSO4-5H2O (0.5); FeSO4.7H2O (0.01); ZnSO4 (0.001); MnCl2 (0.001); and 15 g of agar) amended with 2% phenol red and incubated at 27 ± 2oC for 72 h. Presence of clear zones around culture spots (inoculants) indicated chitinase production (Wang et al. 2008).

Nitrogen fixation assay

Burk’s Nitrogen-free Medium (comprising g L− 1: glucose (10); KH2PO4 (0.41); K2HPO4 (0.52); Na2SO4 (0.05); CaCl2 (0.2); MgSO4.7H2O (0.1); FeSO4.7H2O (0.005); Na2MoO4.2H2O (0.0025); and agar (15); pH: 7 ± 1) described by Wilson and Knight (1952) was adopted to test nitrogen-fixing potential of isolated PSB. Media were inoculated with a 24 h old pure PSB and incubated at 37oC for 120 h. Only the PSB isolates with potential of nitrogen-fixation were able to grow after 120 h of incubation.

Indole Acetic Acid (IAA) production assay

Production of Indole acetic acid (IAA) was assayed using colorimetric method described by Glick (1995) with slight modification. The PSB isolates were cultured in IAA Minimal Medium (IAA-MM) (g L− 1: KH2PO4 (1.36); Na2HPO4 (2.13); MgSO4.7H2O (0.2) and trace elements) supplemented with 100 µg mL− 1 of L-tryptophan and incubated at 37oC for 72 h in a shaker incubator (New Brunswick Scientific Co., Inc., USA). After incubation, 1.5 mL of bacterial suspension was spun at 10,000 rpm for 10 min in a centrifuge (Beckman Coulter, Massachusetts, USA). The supernatant was added to Salkowski’s reagent (2% 0.5 M FeCl3 in 35% HClO4) at 1:2 proportion before incubating in dark for 30 min at 25 °C. Absorbance of the final solution was then measured with UV/Visible spectrophotometer (WPA Linton Cambridge, UK) at 530 nm and IAA produced by PSB isolates was determined against standard curve (Zhang et al. 2017; Adebayo 2019).

Ammonia production assay

PSB were cultivated in 10 mL Peptone Water Broth and incubated at 27 ± 2oC for 120 h, afterwards 5ml of bacterial suspension was spun at 10,000 rpm for 10 min in a centrifuge (Beckman Coulter, Massachusetts, USA). The supernatant (0.2 mL) of the cultured broth was then mixed with 1mL Nessler’s reagent (10 g Mercuric chloride, 7 g Potassium iodide and 16 g Sodium hydroxide in 100 mL distilled water (DW) (ammonia free); pH 13.2 ± 0.05) and the volume of this mixture was made up to 8.5 mL with ammonia free distilled water. Development of brown to yellow colour was indicative of ammonia production and the absorbance of the solution was measured with spectrophotometer (WPA Linton Cambridge, UK) at 450 nm. Concentration of ammonia was estimated with standard curve of ammonium sulphate (0.1-1 µM mL− 1 range) (Adebayo 2019).

Siderophore production assay

Production of siderophore was determined following the method described by Goswami et al. (2014) with little modification. The PSB isolates were inoculated into the Fiss-Glucose Mineral Medium and incubated in a shaking incubator (New Brunswick Scientific Co., Inc., USA) at 25 °C for 3 d at 174 rpm. After incubation, 5 mL of bacterial suspension was spun at 10,000 rpm for 10 min in a centrifuge (Beckman Coulter, Massachusetts, USA). The cell-free supernatants of the cultured broth were assayed for siderophore production using ferric chloride test for hydroxamate and catecholate nature of the siderophore following Neilands spectrophotometric assay (Neilands 1981); and phenolphthalein test described by Furniss et al. (1989) for carboxylate nature of the siderophore. Formation of red/pink colour indicated the presence of siderophore in the solution which was measured with a UV/Visible spectrophotometer (WPA Linton Cambridge, UK). The absorbance peak between 420 and 450 nm of ferrated siderophores indicated its nature as hydroxamate while absorption peak at 495 nm indicated the presence of catecholate siderophore. The disappearance of pink colour on the addition of phenolphthalein to the solution indicated its nature as carboxylate.

Molecular identification of PSB from rumen content

Bacterial isolates of interest were further characterised by amplification and sequencing of the 16S rRNA gene (~ 1500 bp). Firstly, genomic DNA was extracted using Bacterial Genomic DNA Isolation Kit (Jena Bioscience, Germany) following manufacturer’s instructions. Subsequently, the 16S rRNA gene was amplified from genomic DNA by polymerase chain reaction (PCR) using universal bacterial primers; 27F (5’-AGAGTTTGATYMTGGCTCAG-3’) and 1390R (5’-ACGGGCGGTGTGTRCAA-3’) (Mao et al. 2012). Each 30 µL PCR reaction contained 6 µL of RedLoad (Jena Bioscience, Germany) PCR mix, 0.3 µL of each primer, 18.4 µL of PCR grade water and 5 µL of DNA template. PCR was done with GeneAmp 9700 (Applied Biosystems, USA) thermal cycler as follows; 94 °C for 3 min, 30 cycles at 94 °C for 30 s, 54 °C for 30 s, 60 °C for 90 s and a final 7 min extension at 72 °C, after which product was held at 4oC till terminated. The PCR products were subsequently resolved on 1% electrophoresis gel stained with ethidium bromide and viewed using UV transilluminator (Fotodyne Incorporated, USA). Amplicons were shipped to Macrogen Inc., South-Korea, for purification and Sanger sequencing.

To identify the PSB isolates, the sequences were compared using the BLASTn programme (http://www.blast.ncbi.nlm.nih.gov) with other publicly available nucleotide sequences in GenBank (http://www.ncbi.nlm.gov/genbank/). The nucleotide sequences of the 16 S rDNA of PSB recovered from this study were submitted to GenBank and have been assigned Accession Numbers (Table 3).

Greenhouse evaluation of rumen PSB for growth promotion in tomato plant

Seed bacterization method described by Singh et al. (2014) was adopted with slight modification to evaluate plant growth promoting potential of rumen PSB on pre-authenticated tomato (Solanum lycopersicum L.). Briefly, 1 g of plant seeds, surface-sterilized according to Qessaoui et al. (2019), were treated with 10 mL bacterial suspension (1.0 × 108 CFU mL− 1) incorporated with 200 mg CMC for 2 h on a rotary shaker at 150 rpm. Control (non-bacterized) treatment were similarly prepared with sterile DW (10 mL) in place of bacterial suspension. Moisture content of seed preparation were drained off and air-dried for 12 h in a laminar hood. Bacterized and non-bacterized seeds (20 pieces) were separately sown in sterilized and triplicated pot soils (with 3:1 ratio of soil to decomposed cow dung manure), arranged (randomly) in a greenhouse under 18 − 6 h at 25-20oC (day-night) cycle and relative humidity of 70 ± 5%. Farmed pot soils were watered every 2 d for the period (21 d) of experiment. Growth responses such as; relative germination rate (determined at 5 d after sowing); height (shoot and root) and weight (both fresh and dried) after 21 d, of the seedlings were determined and used to calculate seedling vigour index (SVI) according to (Sankar et al. 2017).

Phosphorus content and pH of soil before the experiment and individual treatment soils after the experiment, were determined by Olsen’s method; and the phosphorus content of 21 d old seedlings were determined by ammonium-vanado-molybdate method (Estefan et al. 2013).

Statistical analysis

All experiments were performed in triplicate and data obtained was subjected to Analysis of Variance (ANOVA). Values were presented in Mean ± SD/SE (Standard Deviation/ Standard Error) computed with Duncan’s multiple range tests using SPSS package (version 22.0). Differences at P ≤ 0.05 level was considered significant.