Skip to main content
NCBI home page
3 Biotech logoLink to 3 Biotech
. 2019 Feb 27;9(3):108. doi: 10.1007/s13205-019-1643-6

High-throughput sequence analysis of bacterial communities in commercial biofertiliser products marketed in South Africa: an independent snapshot quality assessment

Adekunle R Raimi 1,2, Obinna T Ezeokoli 2,3, Rasheed A Adeleke 2,3,✉
PMCID: PMC6391514  PMID: 30863692

Abstract

The genetic and predicted functional diversity of bacterial communities in 12 commercial biofertiliser products were evaluated using high-throughput sequencing of the 16S rRNA gene. Proteobacteria, Firmicutes and Bacteroides dominated the bacterial communities, with the genera Pseudomonas, Lactobacillus, Bacillus, Bradyrhizobium and Rhizobium being prevalent. The manufacturer-specified species were detected in relatively high abundance in two of the products while a few or none of the specified species were detected in some products. A number of unspecified microbes were detected, including potential human and crop pathogens such as Alcaligenes, Clostridium, Escherichia–Shigella and Proteus. The functional prediction unravelled high prevalence of enzyme-coding genes such as nitrogenase, NifT, alkaline phosphatase and reductases of nitric oxide, nitrate and nitrite which contribute to nitrogen-fixation, phosphorus solubilisation and degradation of nitrates and nitrites. In addition, toxins such as leukocidin/hemolysin and colicin V protein that cause product quality damage were highly predicted in over 67% of the products. Overall, high-throughput sequence analysis of bacterial communities in biofertiliser products revealed that majority of the products were of poor quality. This result justifies the need for regular quality assessment and improvement in quality control systems during biofertiliser formulation.

Electronic supplementary material

The online version of this article (10.1007/s13205-019-1643-6) contains supplementary material, which is available to authorized users.

Keywords: Bacterial diversity, Biofertiliser, High-throughput sequencing, Predicted functional diversity, Quality assessment

Introduction

The increasing demand for a safer environment, cleaner technologies and organic farm produce has necessitated a shift from the use of chemical fertilisers to a more sustainable and ecologically friendly soil nutrient management system (Bashan 1998; Malusà et al. 2016). Lately, the application and development of biofertilisers, as a sustainable alternative to chemical fertilisers, have gained prominence (Bhardwaj et al. 2014).

Biofertilisers are composites of various functionally beneficial microbes, including nitrogen-fixing Rhizobium, Bradyrhizobium and Sinorhizobium, as well as phosphate-solubilising Pseudomonas, Bacillus and Enterobacter (Raimi et al. 2017). Species of Azotobacter, Azospirillum, Serratia and Alcaligenes have also been incorporated into several biofertiliser products. The microbial species in biofertilisers improve the availability of soil nutrients, including phosphorous, zinc, iron and copper, thereby improving plant-nutrient uptake (Han and Lee 2005). Some of these microbial species also produce phytohormones such as indole acetic acid (IAA), cytokinin, and gibberellin, which enhance plant growth and development (Adesemoye et al. 2008; Mehta et al. 2016). In addition, the microbial species in biofertiliser produce antimicrobial metabolites, toxins and siderophore that suppress soil-borne pathogens which compete for soil nutrients or niches (Herrmann et al. 2015; Sayyed et al. 2010).

However, product quality parameters such as high level of contamination, low shelf life and low-quality microbial content have contributed to the low economic advantage and poor perception of biofertilisers by end users—the farmers (Malusá and Vassilev 2014). The quality of biofertiliser products is a crucial determinant of its effectiveness when applied in the field. To a great extent, field effectiveness of a biofertiliser product is a function of the genetic diversity and functional attributes of its microbial consortium. According to Herridge et al. (2002), if the biofertiliser quality is poor, then everything about its application is inconsequential. Several biofertiliser-quality-assessment studies have observed that the specified microbial contents of biofertiliser products are either not present or when present, the viable cell counts are below the industry standard (Lupwayi et al. 2000; Olsen et al. 1995). For example, Olsen et al. (1995) reported about 98% of the biofertiliser tested in North America had more microbial contaminants than the stipulated rhizobia cells; in particular, three of the products contained approximately 1000 times more microbial contaminants than the rhizobia contents. Unfortunately, most microbial contaminants found in biofertiliser products are known pathogens of humans and plants. In a study by Herrmann et al. (2015), over 84% of contaminated biofertilisers had microbial contaminants that were either human, animal or plant pathogens. Some of the pathogenic contaminants such as Agrobacterium sp., Staphylococcus sp., Stenotrophomonas sp., Cuprivadus sp., Enterobacter sp., have been widely reported in different studies (Olsen et al. 1996; Herrmann et al. 2015). Therefore, the assessment of microbial communities of biofertiliser products is important for ascertaining quality and potential field efficiency.

The assessment of bacterial communities of biofertilisers has been carried out in different studies using culture-based techniques (Deaker et al. 2011; Herrmann and Lesueur 2013) that are well-known to underestimate community diversity. Given the limitations of microbial cultivation and the estimated 99% of yet uncultivable microbial species, culture-based techniques are highly inadequate for analysing and processing complex environmental samples, especially for large-scale studies (Christine 2004; Sanger et al. 1977). In recent years, the advent of next-generation sequencing (NGS) technologies has enabled the in-depth analysis of microbial communities without a need for a cultivation step (Ezeokoli et al. 2018; Mashiane et al. 2017). Next-generation sequencing is a reliable and widely used technique in analysing microbial communities of several environments (Mashiane et al. 2017; Van Wyk et al. 2017; Tyx et al. 2016).

Presently, several biofertiliser products are available commercially in South Africa, including imported and locally manufactured products. Although regulatory frameworks and quality guidelines are still being drafted by the South African department of Agriculture, Forestry and Fisheries (DAFF)—the organisaztion with a mandate to regulate agro-based products—there are presently no comprehensive quality control guidelines to ascertain the integrity of commercial biofertiliser products in South Africa. As an independent assessment of biofertiliser quality, this study was supported by DAFF to investigate in a snapshot (one-time), the quality of biofertiliser products in South Africa. For the assessment, high-throughput sequencing of the 16S rRNA gene on the Illumina MiSeq next-generation sequencer was employed. It is envisaged that information obtained from this study will contribute to the growing empirical evidence for a need to improve quality control systems for commercial biofertiliser production in South Africa and beyond.

Materials and methods

Biofertiliser samples

Biofertiliser products commercially available in the South African market were randomly sampled (N = 12) for this study. The sample pool included both single-strain and mixed-consortium biofertiliser types. Information on the types and form of the biofertiliser products are provided in supplementary materials (Table S1). Samples were collected without compromising the integrity of the contents and all analyses were performed aseptically. Samples were stored at 4 °C or under room temperature as stipulated by the manufacturer prior to high-throughput sequence analysis of bacterial communities. Samples were analysed within a week of collection.

DNA extraction

Total genomic DNA of microorganisms in biofertiliser samples was extracted from 0.25 g (for solids) or 200 µL (for liquids, after centrifugation at 10000 rpm for 5 min) aliquots using the PowerSoil DNA Isolation Kit (MO BIO Laboratories, Inc., California, USA) according to the manufacturer’s protocol. Integrity and quantification of extracted DNA were verified by agarose gel electrophoresis and fluorometric quantification (Qubit 2.0 fluorometer, Invitrogen, CA, USA), respectively.

16S rRNA gene library preparation and high-throughput sequencing

The hypervariable V3–V4 region (approximately 460 bp) of the 16S rRNA gene was amplified in a polymerase chain reaction (PCR) using region-specific primers 341F (5′-CCTAC GGGNGGCWGCAG-3′) and 805R (5′-GACTACHVGGGTATCTAATCC-3′) (Klindworth et al. 2013) and following the method described by Van Wyk et al. (2017). The 5′-end of forward and reverse primer sequences contained Illumina forward and the reverse adapters (Illumina Inc., California, USA), respectively. Partial 16S rRNA amplicons were further processed and sequenced on an Illumina MiSeq sequencer by following the steps described in the MiSeq 16S library preparation guide (Illumina n.d.).

Bioinformatics analyses

Demultiplexed sequence reads were analysed for quality using FastQC (Babraham Bioinformatics, UK; https://www.bioinformatics.babraham.ac.uk/index.html) and PANDAseq was used to assemble forward and reverse reads, and eliminate reads with ambiguous bases and spurious read length (Masella et al. 2012). Thereafter, assembled reads were binned into operational taxonomic units (OTUs) at 97% 16S rRNA gene sequence similarities using the open-reference OTU-picking strategy in the Quantitative Insight into Microbial Ecology (QIIME) software (Caporaso et al. 2010). For OTU picking and taxonomic assignment, sequences were aligned against the SILVANGS rRNA database (SILVA 128 release) (Quast et al. 2012) with usearch61 (Edgar et al. 2011). Because sample characteristics varied for each biofertiliser product and comparisons were meaningless, the OTU count table was not rarefied (normalised) to even depth prior to subsequent analyses. The generated OTU count table was taxonomically summarised in QIIME, while the computation of the alpha (α) and beta (β) diversity measures was done using a combination of the vegan and ape packages of R software version 3.4.4 (R Core Team 2013).

Predicted functional analysis of bacterial communities

The functional abilities of the bacterial community in biofertiliser samples were predicted using the Tax4Fun package (Aßhauer et al. 2015) in R software (R Core Team 2013). In summary, the obtained OTU table from the SILVA rRNA database alignment was transformed into a functional or metabolic profiles by the Tax4Fun package with the aid of a pre-computed association matrix of prokaryotic Kyoto Encyclopaedia for Genes and Genomes (KEGG) database (https://www.genome.jp/kegg/) (Aßhauer et al. 2015; Kanehisa et al. 2013). BRENDA (https://www.brenda-enzymes.org/) (Placzek et al. 2016) and KEGG orthology expression (Kanehisa et al. 2013) were used to obtain information on predicted enzyme-coding genes. For this study, the focus was on selected genes involved in the cycling and mineralization of nutrients in the soil.

Data accessibility

The raw sequence reads obtained in the present study have been deposited in the Sequence Read Archive of the National Centre for Biotechnology Information. (https://www.ncbi.nlm.nih.gov/sra) under the BioProject accession number PRJNA483344.

Results

Operational taxonomic units diversity

The number of sequence reads generated and selected for alpha-diversity measures are summarised in Supplementary Table S2. A total of 2,647,214 sequence reads were initially obtained from all samples (Table S2). After quality filtering, 2,186,464 high-quality reads were obtained and subsequently assigned to operational taxonomic units (OTUs). The rarefaction analysis of detected sequences showed sufficient sampling efforts (Supplementary Fig. S1). In total, 5791 OTUs were detected from all samples (data not shown). Amongst samples, CB1L had the highest number of observed OTUs, while CB2S had the least number of observed OTUs (Table S2). Similar trends (as with observed OTUs) in estimated richness (chao1), Shannon–Weiner and Simpson’s indices were observed (Table S2). In general, OTU abundances detected in all the biofertiliser samples were a close estimation of the true OTUs diversity as indicated by the calculated goods coverage of approximately 1 (100%).

The diversity of bacterial phylotypes

Bacterial phylotypes detected in all samples spanned 35 phyla, 92 classes, 222 orders, 453 families and 1030 genera with a classifiable read (at each taxa level) of 100%, 88%, 79%, 69% and 66%, respectively. The relative abundance of phyla taxa in the biofertiliser samples differed across the different products with Proteobacteria being the major dominant phylum in all the products except in samples CB3L and CB4L where Firmicutes occurred at 98% and 97%, respectively (Fig. 1). Other dominant phyla above 1% relative abundance included Acidobacteria, Actinobacteria, Bacteroides and Chloroflexi. Proteobacteria was more predominant in the single-strain products than the consortium products, occurring with the highest abundance of 96% in CB12L (Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Relative abundance of bacterial phyla in biofertiliser products

At the family taxa level, consortium-strain products were more diverse than the single-strain products, with an average of 11 families per consortium product and 6 families per single-strain products. The most dominant families were Lactobacillaceae, Pseudomonadaceae, Rhizobiaceae, Bradyrhizobiaceae, Enterobacteriaceae and Clostridiaceae (Fig. S2). A high dominance of Lactobacillaceae was observed in samples CB3L and CB4L, while Enterobacteriaceae was the most abundant family detected in CB10L, CB11L and CB12L. Samples CB8L, CB9L and CB1L were dominated by Pseudomonadaceae, Clostridiaceae and Methylocystaceae, respectively, while samples CB5S and CB2S, which are carrier-based products predominantly contain Bradyrhizobiaceae and Rhizobiaceae, respectively (Fig. S2).

Of the total 1030 detectable genera, only 63 genera occurred at a minimum of 1% relative abundance in at least one sample (Fig. 2). Of these, the genera Methylocystis, Parvibaculum, Pusillimonas, Pelagibacterium, Paucimonas, Salinihabitans, Pigmentiphaga, Devosia, Thioalkalispira, Proteiniphilum, Flavobacterium, Sphingomonas, Chryseobacterium and Petrimonas were detectable only in sample CB1L, Cellulosimicrobium and Rhizobium occurred in CB2S, Brevibacillus and Stenotrophomonas were observed in CB4L while Providencia, Serratia, Rahnella and Ewingella occurred in only CB12L (Fig. 2). Other genera detected in a single product included Pediococcus (sample CB3L), Bradyrhizobium (sample CB5S), Lactococcus (sample CB8L), Kluyvera (sample CB10L) and Enterococcus (sample CB8L). Similarly, Cronobacter, Pantoea and Tyzzerella 3 were observed only in sample CB11L. Of all the genera, Bradyrhizobium had the highest relative abundance of 94%, followed by Lactobacillus with 91% and 73% relative abundance in samples CB3L and CB4L, respectively (Fig. 2). Similarly, Rhizobium had a high relative abundance of 76% in sample CB2S while Pseudomonas, though occurring in all the samples, had high relative abundance in only samples CB8L (69%) and CB12L (30%), respectively (Fig. 2).

Fig. 2.

Fig. 2

Open in a new tab

Relative abundance of bacterial genera in biofertiliser products

Quality assessment: manufacturer-specified composition versus detectable phylotypes

The detectability of manufacturer-specified and -unspecified bacteria in biofertiliser samples was used to infer product quality. The manufacturer specified species and those detected (based on ≥ 1% relative abundance per sample) in biofertiliser products are presented in Table 1. Taking into cognizance the criteria for biofertiliser quality and limitations of the NGS technique, we devised a somewhat relaxed operational scoring system in which products were categorised as high quality (all the specified species and total relative abundance of unspecified species ≤ 10% detected), medium quality (≥ 50% of the specified species and a total relative abundance of unspecified species ≤ 50% detected), low quality (< 50% of specified species and a relative abundance of unspecified species > 50% detected) and poor quality (specified species not detected). The genus taxa level was used to define a “species” because the taxonomic assignment of NGS analysis beyond this taxonomic rank may not be absolutely reliable. The products quality ranged from high (8.3%), medium (8.3%), low (25%) and poor-quality (58.3%) (Table 1). In addition, one or more unspecified bacteria herein referred to as microbial contaminants were detected in all the samples (Table 1). For instance, sample CB5S, had the specified species, Bradyrhizobium at 94% relative abundance and only one unspecified species, Nosocomiicoccus, at 1.2% relative abundance. Similarly, CB2S had the specified species, Rhizobium, at approximately 76% relative abundance with unspecified microbes: Cellulosimicrobium, Nocardioides and Promicromonospora detected at a relative abundance of 11.4%, 7.2% and 1.0%, respectively. Therefore, CB5S and CB2S (rhizobial products, see Table S1) were regarded as high and medium quality products, respectively (as defined in the present study), because they contained the specified bacteria at higher relative abundance but with a relatively lower abundance of unspecified species. Aside from being carrier-based products, samples CB5S and CB2S are also imported products. On the other hand, sample CB4L a plant growth promoting rhizobacteria (PGPR) product was regarded as a low-quality product because it had the expected microbe, Bacillus, at a lower relative abundance (14.7%) than the unspecified species (83.8%) (Lactobacillus, Brevibacillus and Stenotromonas).

Table 1.

Expected and detected bacterial species in biofertiliser products based on partial 16S rRNA gene analysis

Sample code Manufacturer-specified bacterial genusa No. of genus expecteda Observed NGS sequences (OTUs) at above 1% relative abundance No. of unspecified genera Quality scorec
CB1L Enterobacter, Stenotromonas, Bacillus, Rhizobium, Pseudomonas 5 Pseudomonas b, Chryseobacterium, Pigmentiphaga, Proteiniphilum, Devosia, Pusillimonas, Dysgonomonas, Salinihabitans, Flavobacterium, Paucimonas, Sphigomonas, Methylocystis, Thioalkalispira, Parvibaculum, Pertimonas Pelagibacterium 15 Low
CB2S Rhizobium 1 Rhizobium b, Cellulosimicrobium, Nocardioides, Promicromonospora 3 Medium
CB3L Bacillus, Lactobacillus, Pseudomonas, Azotobacter 4 Lactobacillus b, Nosocomiicoccus, Pediococcus 2 Low
CB4L Bacillus 1 Bacillus b, Brevibacillus, Lactobacillus, Stenotromonas 4 Low
CB5S Bradyrhizobium 1 Bradyrhizobium b, Nosocomiicoccus 1 High
CB6S Azospirillum 1 Macellibacteroides, Alcaligenes, Pseudomonas, Hafnia, Clostridium sensu stricto 1, 12 & 5, Proteus, Dysgonomonas, Microvirgula, Morganella 11 Poor
CB7S Azospirillum, Azotobacter 2 Microvirgula, Clostridium sensu stricto 1, 12, 2 & 5, Macellibacteroides, Morganella, Phascolarctobacterium, Alcaligenes, Ruminiclostridium 5, Pseudomonas, Proteus, Dysgonomonas 13 Poor
CB8S Bradyrhizobium 1 Alcaligenes, Pseudomonas, Enterobacter, Enterococcus, Proteus, Lactobacillus, Lactococcus, Microvirgula 8 Poor
CB9L Pseudomonas 1 Alcaligenes, Clostridium sensu stricto 1, 12, 2 & 5, Dysgonomonas, Macellibacteroides, Microvirgula, Morganella, Enterobacter, Ruminiclostridium, Hafnia, Escherichia–Shigella, Desulfovibrio, Proteus 15 Poor
CB10L Brevibacillus, Paenibacillus, Lysinibacillus, Sporolactobacillus 4 Citrobacter, Clostridium sensu stricto 2 & 5, Desulfovibrio, Dysgonomonas, Escherichia–Shigella, Proteus, Kluyvera, Enterobacter, Macellibacteroides, Morganella, Microvirgula, Phascolarctobacterium, Ruminiclostridium 14 Poor
CB11L Bacillus 1 Cronobacter, Clostridium sensu stricto 1&5, Alcaligenes, Pantoea, Ruminiclostridium 5, Microvirgula, Dysgonomonas, Enterobacter, Tyzzerella, Escherichia–Shigella, Citrobacter, Morganella, Desulfovibrio, Desulfovibrio, Proteus, Macellibacteroides 16 Poor
CB12L Rhizobium phaseolus 1 Enterobacter, Ewingella, Providencia, Morganella, Proteus, Pseudomonas, Serratia, Rahnella, Hafnia 9 Poor
Open in a new tab

aBased on information provided by the manufacturer. Only genus classification was used

bSpecified species

cQuality score: high quality, all the specified species and total relative abundance of unspecified species ≤ 10% detected; medium quality, ≥ 50% of the specified species and a total relative abundance of unspecified species ≤ 50% detected; low quality, < 50% of specified species and a relative abundance of unspecified species > 50% detected; poor quality, specified species not detected

Also, consortium (mixed) products CB1L and CB3L were categorised as low-quality biofertilisers. In sample CB1L with five specified bacteria genera, only phylotypes Pseudomonas (1.4%) was detectable at ≥ 1% relative abundance while the unspecified species were observed at a much higher prevalence. Similarly, of the four bacterial species specified in CB3L, only one species including (relative abundance in parenthesis) Lactobacillus spp. (91%) were observed, while unspecified species including Nosocomiicoccus spp. (1.7%) and Pediococcus spp. (4.3%) were also detected (Table 1). Other samples including CB6L, CB7L, CB8L, CB9L, CB10L, CB11L and CB12L, which have none of the specified species detected but containing only the unspecified microorganisms were regarded as poor-quality products.

Generally, the results showed the detection of unspecified microbes at different levels of abundance in the samples. Of all the 125 unspecified phylotypes recorded at ≥ 1% relative abundance (at the genus taxa level), an average of 9 and 11 species occurred in single-strain and consortium-strain products, respectively. The high and medium-quality products being a rhizobia single-strain product had less of the unspecified species while the poor-quality products, especially the other PGPR and the free-living nitrogen-fixing bacteria products recorded high unspecified species.

Predicted functional profile of the biofertiliser bacterial community

A total of 6564 KEGG Orthology groups (KO funtional orthologs) were obtained from the imputed metagenomes of all the biofertiliser samples (data not shown). Of these, only genes coding for enzymes directly involved with microbial growth and metabolism as well as potential product field functions were focused on. Generally, predicted N-fixation genes such as Nif-specific regulatory protein, nitrogenase molybdenum-iron protein alpha- and beta-chain, and nitrogenase iron protein NifH were relatively high in most of the samples except in samples CB3L and CB4L, while nitrogenase fixation protein NifT and nitrogenase were less abundant in all the samples (Fig. 3). In addition, denitrification genes including nitric oxide reductase subunit B and nitrous-oxide reductase had medium predicted occurrence amongst the samples while nitric oxide subunit C was less abundant across the samples. Other predicted genes which are important to N mineralization included the nitrate and nitrite reductase genes. The nitrite reductase (NAD(P)H) large subunit and small subunit had a high prevalence in all the samples while nitrite reductase (NADH) and ferredoxin-nitrite reductases had a relatively low abundance. Similarly, all the predicted nitrate reductases (alpha, beta, delta and gamma subunits) were in high abundance in the samples. However, alpha and beta subunits had higher abundance than other nitrate reductase genes.

Fig. 3.

Fig. 3

Open in a new tab

Heatmap showing several enzyme-coding genes of interest in biofertiliser products. Predicted metagenomic functional profile was performed using Tax4Fun package in R software. Heatmap was constructed using the gplot package of R software. KEGG orthology of genes are in parenthesis

Based on functional prediction, alkaline phosphatase was more abundant than acid phosphatases across all the samples. In addition, phosphoserine phosphatase and 4-nitrophenylphosphatase were predicted in all the samples. Sulphur is as necessary as phosphorous for the formation of important enzymes in microorganisms and plants. Of the four sulphur-degrading enzymes predicted, arylsulphatase was the most prevalent followed by sulphate adenylyltransferase subunit 1 and subunit 2, while sulphate adenylyltransferase occurred in low abundance. Other essential genes investigated included the iron uptake and glucose-degrading genes. Ferric uptake regulator catecholate siderophore receptor and high-affinity iron transporter were relatively dominant in all the samples, with the highest abundance in samples CB4L and CB8L. The ferric-chelate reductase was observed to be less abundant across all the samples. Furthermore, toxin genes such as Shiga, leucocidin and haemolysin were predicted, while the leukocidin/haemolysin toxin family protein had a higher prevalence amongst five samples (CB2L, CB3L, CB4L, CB8L and CB12L), Shiga toxin genes had a very low abundance in all the samples. Similarly, membrane protein for colicin V production was less prevalent while gene encoding for glucosidase was widely predicted in high abundance in all the samples.

Discussion

Bacterial diversity in biofertiliser products

The quality of a biofertiliser product is determined by its microbial consortium, function and field effectiveness (Malusá et al. 2012; Balume 2013). Invariably, if bacterial species in a biofertiliser are of low potency or missing, its (biofertiliser) effectiveness is affected. Therefore, towards improving crop productivity, the quality assessment of microbial components of biofertilisers has become increasingly necessary. This concern is born of the need to improve the quality of products available to farmers, especially smallholder farmers. Few studies have assessed the quality of commercial biofertiliser products using Sanger sequencing technique (Herrmann et al. 2015; Olsen et al. 1995). However, recent high-throughput next-generation sequencing (HT-NGS) offers a more (compared to culture-dependent and pre-HT-NGS techniques) in-depth and accurate analysis of the microbial diversity of any given environment or matrix (Tyx et al. 2016; Zhang et al. 2017). Therefore, this study utilised high-throughput sequencing of the partial 16S rRNA gene in biofertiliser products available to farmers in South Africa.

In the present study, the relative abundance of OTUs revealed major bacterial genera in the biofertiliser samples to be Lactobacillus, Bradyrhizobium, Pseudomonas, Alcaligenes, Bacillus and Rhizobium. These are predominantly efficient PGPR that have been previously identified in different biofertiliser samples (Hasan et al. 2014; Hatayama et al. 2005; Malusá et al. 2012). Rhizobium and Bradyrhizobium belong to the group; rhizobia, which contributes significantly to biological nitrogen fixation (BNF), especially in legume-cultivated agricultural soils (Raimi et al. 2017; Sutherland et al. 2000). Similarly, Pseudomonas, Bacillus and Alcaligenes have been widely used as biofertilisers, which make phosphorous available for crop use by solubilising insoluble phosphorous (Behera et al. 2017; Fitriatin et al. 2011). Some of these bacteria also produce plant-growth-promoting substances such as indole acetic acid (IAA), gibberellin, 1-aminocyclopropane-1-carboxylic acid (ACC) and siderophores (Raimi et al. 2017; Khan et al. 2016).

Several factors contribute to microbial diversity in biofertiliser products. Some of these factors include the microbial species formulation by the manufacturers (either single or consortium) and a good quality control system that reduces or eliminates unwanted bacterial species in the products (Lupwayi et al. 2000; Olsen et al. 1995; Yadav and Chandra 2014). Formulation of microbial consortium with a better competitive advantage over native species, in diverse ecological conditions, is crucial for high field efficiency (Faye et al. 2013; Kyei-Boahen et al. 2002). This is because the diverse indigenous microbes have heterogeneous functionalities and adaptive traits to metabolise a broad range of compounds and survive under different field conditions, respectively. Results obtained in this study suggest microbial formulation may have influenced the OTU richness and diversity of the products. However, consortium product formulation is very challenging because of the less selective carrier material which must support an array of microbial species. Consequently, this less selective carrier material encourages the growth of unwanted species in the biofertiliser products (Stephens and Rask 2000). On the other hand, single-species products are selectively formulated to support a particular species. Perhaps, the foregoing explanation contributed to the higher species diversity of unwanted species observed in consortium products as opposed to the single-species products. Similarly, the influence of formulation type may explain the low diversity of unwanted species and thus, ‘good quality’ observed in the rhizobial products (CB2S and CB5S). Most rhizobial biofertilisers had high rhizobial density and less unwanted bacterial species. This observation could be due to carrier materials that selectively supported the optimal growth of rhizobia while preventing the proliferation of unwanted species (Balume 2013; Olsen et al. 1995). Nonetheless, with respect to efficiency and crop productivity, the application of single-species products in the field may be economically risky. This is because the likelihood of bacteria to establish an effective plant–microbe relationship in the rhizosphere is subject to different ecological factors; and no species is effective under all ecological conditions (Herrmann et al. 2015; Sutherland et al. 2000).

Higher OTU diversities (based on observed OTUs, Shannon–Weiner and Simpson diversity indices) were observed in liquid biofertiliser products compared to the carrier-based counterparts. Previous studies have reported that product form—liquid or carrier-based—influences the growth and development of bacteria in biofertiliser products (Herrmann et al. 2015; Pindi and Satyanarayana 2012). The observation of higher OTUs’ diversities in liquid products may be attributed to the readily available nutrients and water contents in the liquid products, which aid substantial bacterial growth (Pindi and Satyanarayana 2012). Additionally, samples CB9L and CB11L containing other PGPR such as Bacillus and Pseudomonas, as well as samples CB6L and CB7L containing Azotobacter and Azospirillum were found to have higher species diversities compared to other products. This observation may also be due to the formulation type. Similarly, Herrmann et al. (2015) reported a higher species diversity in free-living nitrogen-fixing biofertilisers containing species of Azotobacter and Azospirillum.

Comparing specified and observed bacterial species in biofertiliser products

The disparities between the observed bacterial OTUs and the specified bacterial species (as informed by the manufacturers’ labelling information) showed that the products had varying levels of unspecified bacteria. Similar to other reported studies, observed bacterial diversity not listed by the manufacturers are referred to as contaminants (Herrmann et al. 2015; Olsen et al. 1995; Deaker et al. 2011). Different levels of contamination and the consequential effects on product quality have been reported (Herridge et al. 2002; Lupwayi et al. 2000; Olsen et al. 1995). In a study by El-Fattah et al. (2013), it was observed that diverse contaminants adversely affected the quality of biofertiliser products. Similarly, the present study showed different levels of contaminants with major ones found in CB1L, CB6L, CB7L, CB8L, CB9L, CB10L CB11L, and CB12L. Interestingly, five of these products were produced by the same manufacturer. The analogous level of contamination between products from the same manufacturer could imply the use of similar sources of materials or production processes. This observation is similar to the work of Herrmann et al. (2015) who reported the same levels of contamination in products from the same manufacturer.

Several biofertiliser-associated contaminants have been reported to be potentially dangerous to the environment, humans and crop health (Herridge et al. 2002; Olsen et al. 1996). Akin to what was obtained in this study, Herrmann et al. (2015) reported that 53% of tested inoculants had contaminants with diverse opportunistic pathogenic effect. In this study, contaminants such as Acinetobacter, Arthrobacter and Alcaligenes have been reported in various studies to cause diseases in humans and animals (Goodfellow et al. 2012; Liu et al. 2002; Saffarian et al. 2017; Tille 2013). Similarly, several species of Brevibacillus, including B. laterosprus, are pathogens of invertebrates (Ruiu 2013) while the genera Proteus, Clostridium, Escherichia–Shigella, Enterococcus, Staphylococcus and Pediococcus have many species that are known pathogens of humans and animals (Chaves et al. 2005; Corcoran et al. 1991; Kim et al. 1981; Rivas et al. 2015; Teixeira et al. 2001). Some of these pathogenic contaminants also inhibit the growth of beneficial microbes such as rhizobia in biofertiliser products (Gomez et al. 1997). Considering these adverse effects, the incidence of contaminants in biofertilisers should be considered more seriously beyond being mere contaminants (Catroux et al. 2001).

Predictive metagenomics profiling of 16S rRNA gene

The predictive functional profiling analysis revealed significantly diverse enzyme-coding genes amongst the bacterial communities. It is important to note that the presence of these genes do not mean that they will be expressed under field conditions. Hence, the predicted functions are discussed herein on the premise of mere potential functions. The functions of some of these extracellular enzymes in microbial metabolism and biogeochemical cycles have been widely studied (Bhardwaj et al. 2014; Malusà et al. 2016; De Bruijn 2015). For example, the predicted N-fixation genes are involved in biological nitrogen fixation (BNF). BNF is a natural process whereby essential atmospheric nitrogen is converted to usable forms of ammonia (NH3), nitrite (NO2−) and nitrate (NO3−) in the soil for rhizosphere microflora and plant use (De Bruijn 2015). The N-fixation genes predicted in the PGPR-products containing Pseudomonas, Enterobacter and Citrobacter suggest that the bacterial communities can possibly participate in BNF (Desnoues et al. 2003; Hatayama et al. 2005; Neilson and Sparell 1976). Additionally, N-cycling pathways such as denitrification, which involves nitrate and nitrite degradation, are essential in converting nitrate, nitrite or ammonia to N gas. The presence of these genes suggests the prominence of denitrification ability among the bacterial (Šimek and Cooper 2002).

Furthermore, in oxygen-deficient conditions, such as during long-term product storage, respiratory nitrate reductase is often expressed where nitrate is present. Hence, the predicted nitrate reductase genes in the bacterial community suggest the use of nitrate as an alternative to oxygen to gain electrons for maintaining the proton-motive force in microbial cells (Tyx et al. 2016). However, the build up of extracellular nitrite during the respiratory process is toxic to microbial cells. To overcome this situation, nitrite-exporting enzymes are expressed, which establish the reason for the presence of nitrite genes in some of the bacterial communities. Most bacteria with assimilation and denitrification ability can further use the produced nitrite (Kraft et al. 2014; Luque-Almagro et al. 2011). If the aforementioned genes are absent, microbial cell death may occur after a long period of storage. This may have contributed to the low-quality or total product damage observed in some of the products.

Furthermore, the alkaline phosphatase gene, an important enzyme in phosphate mineralisation was predicted with a high abundance amongst some of the bacterial communities. This suggests the ability of bacterial communities to potentially mineralise phosphate when applied in the field, especially in alkaline pH environments (Behera et al. 2017; Fitriatin et al. 2011). Similarly, the predicted sulphur-degrading genes, especially arylsulphatase, may indicate the potential role of biofertiliser communities in sulphur cycling and mineralisation in the soil. Likewise, the presence of the ferric uptake regulator suggests potentials for the uptake of iron in microorganisms. The ferric uptake regulator gene improves crop growth and development through direct or indirect mechanisms of iron uptake by the bacterial communities (Sayyed et al. 2010). The presence of gene encoding for glucosidase suggests microbial communities could use different sources of carbohydrates, such as glucose. Other important predicted genes such as Shiga, leucocidin and haemolysin toxin genes have great potential in causing human diseases and product damage. These toxins are frequently cytotoxic, destroying cells by creating unregulated pores in the membranes of the host cells (Gouaux et al. 1997; Laohachai et al. 2003). Escherichia–Shigella, a major genus responsible for the production of Shiga toxin was observed in some of the products (Laohachai et al. 2003). Microbial communities producing these toxins may suffer from loss of viable cells due to the damaging effect of the toxins on the beneficial bacterial cells, thereby causing poor-quality products (Gomez et al. 1997).

Conclusions

Of concern in the present study is the abundant population of undesired microbes and the presence of toxins that contribute to biofertiliser product damage. These contaminants and toxins are dangerous to human and plant health. Therefore, eliminating or reducing microbial contaminants and their damaging effects in biofertiliser products should be pursued through improved quality control management involving regular product-quality evaluation. A rapid technique of evaluation such as NGS technology used in this study is, therefore, suggested as a suitable technique for assessing the microbial quality of biofertiliser products to establish product quality and potential efficiency before field application.

Moreover, the development of an appropriate quality control system is essential for a successful production of a quality product that will improve crop productivity and farmers profitability. This is also a step towards environmental sustainability. Nevertheless, further studies are required to investigate other factors such as physicochemical properties and storage conditions of products that may influence the microbial quality of biofertiliser products. Toxicity and pathogenicity of the microbes contained in biofertiliser formulations should be further investigated as well as gene expression studies to validate the activities of the predicted functional genes.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary material 1 (DOCX 23 KB) (23.7KB, docx)
13205_2019_1643_MOESM2_ESM.jpg (70.1KB, jpg)

Fig. S1 Rarefaction curve showing sufficient sampling of sequences analysed for all samples. Plot represents the estimated OTU abundance using non-normalised data and the Alpha-rarefaction curves were made in R (R Core Team 2013) using the observed-specie metric for estimating alpha (within sample) community diversity (JPG 70 KB)

13205_2019_1643_MOESM3_ESM.png (89.1KB, png)

Fig. S2 Relative abundance of family taxa in biofertiliser products (PNG 89 KB)

Acknowledgements

This work was supported by the Department of Agriculture, Forestry and Fisheries through the National Research Foundation (NRF) project (Grant number 98692). Obinna T. Ezeokoli is financially supported by NRF Grant (UID 102249). We acknowledge the staff of Gauteng Department of Agriculture and Rural Development, particularly Arthur Madaka and Lesego Phakedi for their support.

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical statement

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

References

  1. Adesemoye A, Torbert H, Kloepper J. Enhanced plant nutrient use efficiency with PGPR and AMF in an integrated nutrient management system. Can J Microbiol. 2008;54(10):876–886. doi: 10.1139/w08-081. [DOI] [PubMed] [Google Scholar]
  2. Aßhauer KP, Wemheuer B, Daniel R, Meinicke P. Tax4Fun: predicting functional profiles from metagenomic 16S rRNA data. Bioinformatics. 2015;31(17):2882–2884. doi: 10.1093/bioinformatics/btv287. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Balume IKM (2013) Assessment of quality control of inoculants used on Bean and Soybean in Eastern and central Africa. Dissertation, University of Nairobi
  4. Bashan Y. Inoculants of plant growth-promoting bacteria for use in agriculture. Biotechnol Adv. 1998;16(4):729–770. [Google Scholar]
  5. Behera B, Yadav H, Singh S, Mishra R, Sethi B, Dutta S, Thatoi H. Phosphate solubilization and acid phosphatase activity of Serratia sp. isolated from mangrove soil of Mahanadi river delta, Odisha, India. J Genet Eng Biotechnol. 2017;15(1):169–178. doi: 10.1016/j.jgeb.2017.01.003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bhardwaj D, Ansari MW, Sahoo RK, Tuteja N. Biofertilizers function as key player in sustainable agriculture by improving soil fertility, plant tolerance and crop productivity. Microb Cell Fact. 2014;13(1):1–10. doi: 10.1186/1475-2859-13-66. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Peña AG, Goodrich JK, Gordon JI. QIIME allows analysis of high-throughput community sequencing data. Nat Methods. 2010;7(5):335–336. doi: 10.1038/nmeth.f.303. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Catroux G, Hartmann A, Revellin C. Trends in rhizobial inoculant production and use. Plant Soil. 2001;230(1):21–30. [Google Scholar]
  9. Chaves F, García-Álvarez M, Sanz F, Alba C, Otero JR. Nosocomial spread of a Staphylococcus hominis subsp. novobiosepticus strain causing sepsis in a neonatal intensive care unit. J Clin Microbiol. 2005;43(9):4877–4879. doi: 10.1128/JCM.43.9.4877-4879.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Christine M. Microbial diversity unbound: what DNA-based techniques are revealing about the planet’s hidden biodiversity. Bioscience. 2004;54(12):1064–1068. [Google Scholar]
  11. Corcoran G, Gibbons N, Mulvihill T. Septicaemia caused by Pediococcus pentosaceus: a new opportunistic pathogen. J Infect. 1991;23(2):179–182. doi: 10.1016/0163-4453(91)92190-g. [DOI] [PubMed] [Google Scholar]
  12. De Bruijn FJ. Biological nitrogen fixation. In: Lugtenberg B, editor. Principles of plant-microbe interactions. New York: Springer; 2015. pp. 215–224. [Google Scholar]
  13. Deaker R, Kecskés M, Rose M, Amprayn K, Krishnen G, Cue TTK, Nga VT, Cong PT, Hien NT, Kennedy I. Practical methods for the quality control of inoculant biofertilisers. Canbera: Australian Centre for International Agricultural Research (ACIAR); 2011. [Google Scholar]
  14. Desnoues N, Lin M, Guo X, Ma L, Carreño-Lopez R, Elmerich C. Nitrogen fixation genetics and regulation in a Pseudomonas stutzeri strain associated with rice. Microbiology. 2003;149(8):2251–2262. doi: 10.1099/mic.0.26270-0. [DOI] [PubMed] [Google Scholar]
  15. Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics. 2011;27(16):2194–2200. doi: 10.1093/bioinformatics/btr381. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. El-Fattah DaA, Eweda WE, Zayed MS, Hassanein MK. Effect of carrier materials, sterilization method, and storage temperature on survival and biological activities of Azotobacter chroococcum inoculant. Ann Agric Sci. 2013;58(2):111–118. [Google Scholar]
  17. Ezeokoli OT, Adeleke RA, Bezuidenhout CC. Core bacterial community of soy-daddawa: insights from high-throughput DNA metabarcoding. LWT Food Sci Technol. 2018;97:61–66. [Google Scholar]
  18. Faye A, Dalpé Y, Ndung’u-Magiroi K, Jefwa J, Ndoye I, Diouf M, Lesueur D. Evaluation of commercial arbuscular mycorrhizal inoculants. Can J Plant Sci. 2013;93(6):1201–1208. [Google Scholar]
  19. Fitriatin BN, Arief DH, Simarmata T, Santosa DA, Joy B. Phosphatase-producing bacteria isolated from Sanggabuana forest and their capability to hydrolyze organic phosphate. J Soil Sci Environ Manag. 2011;2(10):299–303. [Google Scholar]
  20. Gomez M, Silva N, Hartmann A, Sagardoy M, Catroux G. Evaluation of commercial soybean inoculants from Argentina. World J Microbiol Biotechnol. 1997;13(2):167–173. [Google Scholar]
  21. Goodfellow M, Kämpfer P, Busse H-J, Trujillo ME, Suzuki K-I, Ludwig W, Whitman WB. Bergey’s Manual® of systematic bacteriology: volume five the Actinobacteria, part A. New York: Springer; 2012. [Google Scholar]
  22. Gouaux E, Hobaugh M, Song L. α-Hemolysin, γ-hemolysin, and leukocidin from Staphylococcus aureus: distant in sequence but similar in structure. Protein Sci. 1997;6(12):2631–2635. doi: 10.1002/pro.5560061216. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Han H, Lee K. Phosphate and potassium solubilizing bacteria effect on mineral uptake, soil availability and growth of eggplant. Res J Agric Biol Sci. 2005;1(2):176–180. [Google Scholar]
  24. Hasan M, Bano A, Hassan SG, Iqbal J, Awan U, Rong-Ji D, Khan KA. Enhancement of rice growth and production of growth-promoting phytohormones by inoculation with Rhizobium and other rhizobacteria. World Appl Sci J. 2014;31(10):1734–1743. [Google Scholar]
  25. Hatayama K, Kawai S, Shoun H, Ueda Y, Nakamura A. Pseudomonas azotifigens sp. nov., a novel nitrogen-fixing bacterium isolated from a compost pile. Int J Syst Evol Microbiol. 2005;55(4):1539–1544. doi: 10.1099/ijs.0.63586-0. [DOI] [PubMed] [Google Scholar]
  26. Herridge D, Gemell G, Hartley E. Legume inoculants and quality control. In: Herridge D, editor. Inoculants and nitrogen fixation of legumes in Vietnam. Canberra: Australian Centre for International Agricultural Research Proceedings 109c; 2002. pp. 105–115. [Google Scholar]
  27. Herrmann L, Lesueur D. Challenges of formulation and quality of biofertilizers for successful inoculation. Appl Microbiol Biotechnol. 2013;97(20):8859–8873. doi: 10.1007/s00253-013-5228-8. [DOI] [PubMed] [Google Scholar]
  28. Herrmann L, Atieno M, Brau L, Lesueur D. Microbial quality of commercial inoculants to increase BNF and nutrient use efficiency. In: de Bruin FJ, editor. Biological nitrogen fixation. New York: Wiley; 2015. pp. 1031–1040. [Google Scholar]
  29. Illumina (n.d.) 16S Metagenomic sequencing library preparation: preparing 16S ribosomal RNA gene amplicons for the Illumina MiSeq System. https://www.illumina.com/content/dam/illumina-support/documents/documentation/chemistry_documentation/16.s/16.s-metagenomic-library-prep-guide-15044223-b.pdf. Accessed 10 Aug 2017
  30. Kanehisa M, Goto S, Sato Y, Kawashima M, Furumichi M, Tanabe M. Data, information, knowledge and principle: back to metabolism in KEGG. Nucleic Acids Res. 2013;42(D1):D199–D205. doi: 10.1093/nar/gkt1076. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Khan AL, Halo BA, Elyassi A, Ali S, Al-Hosni K, Hussain J, et al. Indole acetic acid and ACC deaminase from endophytic bacteria improve the growth of Solanum lycopersicum. Electron J Biotechnol. 2016;21:58–64. [Google Scholar]
  32. Kim KH, Fekety R, Batts DH, Brown D, Cudmore M, Silva J, Jr, Waters D. Isolation of Clostridium difficile from the environment and contacts of patients with antibiotic-associated colitis. J Infect Dis. 1981;143(1):42–50. doi: 10.1093/infdis/143.1.42. [DOI] [PubMed] [Google Scholar]
  33. Klindworth A, Pruesse E, Schweer T, Peplies J, Quast C, Horn M, Glöckner FO. Evaluation of general 16S ribosomal RNA gene PCR primers for classical and next-generation sequencing-based diversity studies. Nucleic Acids Res. 2013;41(1):e1. doi: 10.1093/nar/gks808. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Kraft B, Tegetmeyer HE, Sharma R, Klotz MG, Ferdelman TG, Hettich RL, Geelhoed JS, Strous M. The environmental controls that govern the end product of bacterial nitrate respiration. Science. 2014;345(6197):676–679. doi: 10.1126/science.1254070. [DOI] [PubMed] [Google Scholar]
  35. Kyei-Boahen S, Slinkard AE, Walley FL. Evaluation of rhizobial inoculation methods for chickpea. Agron J. 2002;94(4):851–859. [Google Scholar]
  36. Laohachai KN, Bahadi R, Hardo MB, Hardo PG, Kourie JI. The role of bacterial and non-bacterial toxins in the induction of changes in membrane transport: implications for diarrhea. Toxicon. 2003;42(7):687–707. doi: 10.1016/j.toxicon.2003.08.010. [DOI] [PubMed] [Google Scholar]
  37. Liu L, Coenye T, Burns J, Whitby PW, Stull TL, LiPuma JJ. Ribosomal DNA-directed PCR for identification of Achromobacter (Alcaligenes) xylosoxidans recovered from sputum samples from cystic fibrosis patients. J Clin Microbiol. 2002;40(4):1210–1213. doi: 10.1128/JCM.40.4.1210-1213.2002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Lupwayi N, Olsen P, Sande E, Keyser H, Collins M, Singleton P, Rice W. Inoculant quality and its evaluation. Field Crops Res. 2000;65(2):259–270. [Google Scholar]
  39. Luque-Almagro VM, Gates AJ, Moreno-Vivián C, Ferguson SJ, Richardson DJ, Roldán MD. Bacterial nitrate assimilation: gene distribution and regulation. Biochem Soc Trans. 2011;39(6):1838–1843. doi: 10.1042/BST20110688. [DOI] [PubMed] [Google Scholar]
  40. Malusá E, Vassilev N. A contribution to set a legal framework for biofertilisers. Appl Microbiol Biotechnol. 2014;98(15):6599–6607. doi: 10.1007/s00253-014-5828-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Malusá E, Sas-Paszt L, Ciesielska J. Technologies for beneficial microorganisms inocula used as biofertilizers. Sci World J. 2012;2012:1–12. doi: 10.1100/2012/491206. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Malusà E, Pinzari F, Canfora L. Efficacy of biofertilizers: challenges to improve crop production. In: Singh D, Singh H, Prabha R, editors. Microbial inoculants in sustainable agricultural productivity. New Delhi: Springer; 2016. pp. 17–40. [Google Scholar]
  43. Masella AP, Bartram AK, Truszkowski JM, Brown DG, Neufeld JD. PANDAseq: paired-end assembler for illumina sequences. BMC Bioinform. 2012;13(1):31. doi: 10.1186/1471-2105-13-31. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Mashiane RA, Ezeokoli OT, Adeleke RA, Bezuidenhout CC. Metagenomic analyses of bacterial endophytes associated with the phyllosphere of a Bt maize cultivar and its isogenic parental line from South Africa. World J Microbiol Biotechnol. 2017;33(4):80. doi: 10.1007/s11274-017-2249-y. [DOI] [PubMed] [Google Scholar]
  45. Mehta C, Emmanuel B, Kesarwani A, Sirari K, Sharma AK. Nutrient management strategies based on microbial functions. In: Singh D, Singh H, Prabha R, editors. Microbial inoculants in sustainable agricultural productivity. New Delhi: Springer; 2016. pp. 143–163. [Google Scholar]
  46. Neilson A, Sparell L. Acetylene reduction (nitrogen fixation) by Enterobacteriaceae isolated from paper mill process waters. Appl Environ Microbiol. 1976;32(2):197–205. doi: 10.1128/aem.32.2.197-205.1976. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Olsen PE, Rice WA, Collins M. Biological contaminants in North American legume inoculants. Soil Biol Biochem. 1995;27(4):699–701. [Google Scholar]
  48. Olsen PE, Rice WA, Bordeleau LM, Demidoff A, Collins MM. Levels and identities of nonrhizobial microorganisms found in commercial legume inoculant made with nonsterile peat carrier. Can J Microbiol. 1996;42(1):72–75. doi: 10.1139/m96-010. [DOI] [PubMed] [Google Scholar]
  49. Pindi PK, Satyanarayana S. Liquid microbial consortium-a potential tool for sustainable soil health. J Biofertil Biopestic. 2012;3:124. [Google Scholar]
  50. Placzek S, Schomburg I, Chang A, Jeske L, Ulbrich M, Tillack J, Schomburg D. BRENDA in 2017: new perspectives and new tools in BRENDA. Nucleic Acids Res. 2016;45(D1):D380–D388. doi: 10.1093/nar/gkw952. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Peplies J, Glöckner FO. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res. 2012;41(D1):D590–D596. doi: 10.1093/nar/gks1219. [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. R Core Team . A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing; 2013. [Google Scholar]
  53. Raimi A, Adeleke R, Roopnarain A. Soil fertility challenges and biofertiliser as a viable alternative for increasing smallholder farmer crop productivity in sub-Saharan Africa. Cogent Food Agric. 2017;3:1–26. [Google Scholar]
  54. Rivas L, Mellor GE, Gobius K, Fegan N. Detection and typing strategies for pathogenic Escherichia coli. New York: Springer; 2015. [Google Scholar]
  55. Ruiu L. Brevibacillus laterosporus, a pathogen of invertebrates and a broad-spectrum antimicrobial species. Insects. 2013;4(3):476–492. doi: 10.3390/insects4030476. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Saffarian A, Touchon M, Mulet C, Tournebize R, Passet V, Brisse S, Rocha EP, Sansonetti PJ, Pédron T. Comparative genomic analysis of Acinetobacter strains isolated from murine colonic crypts. BMC Genom. 2017;18(1):525. doi: 10.1186/s12864-017-3925-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  57. Sanger F, Nicklen S, Coulson AR. DNA sequencing with chain-terminating inhibitors. Proc Natl Acad Sci. 1977;74(12):5463–5467. doi: 10.1073/pnas.74.12.5463. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Sayyed R, Gangurde N, Patel P, Joshi S, Chincholkar S. Siderophore production by Alcaligenes faecalis and its application for growth promotion in Arachis hypogaea. Indian J Biotechnol. 2010;9(3):302–307. [Google Scholar]
  59. Šimek M, Cooper J. The influence of soil pH on denitrification: progress towards the understanding of this interaction over the last 50 years. Eur J Soil Sci. 2002;53(3):345–354. [Google Scholar]
  60. Stephens J, Rask H. Inoculant production and formulation. Field Crops Res. 2000;65(2):249–258. [Google Scholar]
  61. Sutherland J, Odee D, Muluvi G, Mcinroy S, Patel A. Single and multi-strain rhizobial inoculation of African acacias in nursery conditions. Soil Biol Biochem. 2000;32(3):323–333. [Google Scholar]
  62. Teixeira LM, Carvalho M, Espinola M, Steigerwalt AG, Douglas MP, Brenner DJ, Facklam RR. Enterococcus porcinus sp. nov. and Enterococcus ratti sp. nov., associated with enteric disorders in animals. Int J Syst Evol Microbiol. 2001;51(5):1737–1743. doi: 10.1099/00207713-51-5-1737. [DOI] [PubMed] [Google Scholar]
  63. Tille P. Bailey & Scott’s diagnostic microbiology-e-book. Amsterdam: Elsevier Health Sciences; 2013. [Google Scholar]
  64. Tyx RE, Stanfill SB, Keong LM, Rivera AJ, Satten GA, Watson CH. Characterization of bacterial communities in selected smokeless tobacco products using 16S rDNA analysis. PLoS One. 2016;11(1):e0146939. doi: 10.1371/journal.pone.0146939. [DOI] [PMC free article] [PubMed] [Google Scholar]
  65. Van Wyk DA, Adeleke R, Rhode OH, Bezuidenhout CC, Mienie C. Ecological guild and enzyme activities of rhizosphere soil microbial communities associated with Bt-maize cultivation under field conditions in North West Province of South Africa. J Basic Microbiol. 2017;57(9):781–792. doi: 10.1002/jobm.201700043. [DOI] [PubMed] [Google Scholar]
  66. Yadav AK, Chandra K. Mass production and quality control of microbial inoculants. Proc Indian Natl Acad Sci. 2014;80(2):483–489. [Google Scholar]
  67. Zhang Y, Yang Q, Ling J, Van Nostrand JD, Shi Z, Zhou J, Dong J. Diversity and structure of diazotrophic communities in mangrove rhizosphere, revealed by high-throughput sequencing. Front Microbiol. 2017;8:2032. doi: 10.3389/fmicb.2017.02032. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplementary material 1 (DOCX 23 KB) (23.7KB, docx)
13205_2019_1643_MOESM2_ESM.jpg (70.1KB, jpg)

Fig. S1 Rarefaction curve showing sufficient sampling of sequences analysed for all samples. Plot represents the estimated OTU abundance using non-normalised data and the Alpha-rarefaction curves were made in R (R Core Team 2013) using the observed-specie metric for estimating alpha (within sample) community diversity (JPG 70 KB)

13205_2019_1643_MOESM3_ESM.png (89.1KB, png)

Fig. S2 Relative abundance of family taxa in biofertiliser products (PNG 89 KB)

Data Availability Statement

The raw sequence reads obtained in the present study have been deposited in the Sequence Read Archive of the National Centre for Biotechnology Information. (https://www.ncbi.nlm.nih.gov/sra) under the BioProject accession number PRJNA483344.

Articles from 3 Biotech are provided here courtesy of Springer

RESOURCES