Does nature make provision for backups in the modification of bacterial community structures?

Abstract

Self-balancing is an inherent character in nature in response to community structure modification pressure and modern biotechnology has revolutionized the way such detections are made. Presented here is an overview of the forces and process interactions between released bacteria and indigenous microflora which encompass soil bacterial diversity, community structure, indigenous endorhizosphere micro-organisms, molecular detection methodologies, and transgenic plants and microbes. Issues of soil bacterial diversity and community structure as well as the interpretation of results from various findings are highlighted and discussed as inferred from research articles. An understanding of the factors influencing bio-inoculant modification of bacterial community structure in the colonization of the rhizosphere is essential for improved establishment of biocontrol agents, and is critically reviewed.

Keywords:

• bacterial community structure
• indigenous
• microbial inoculant
• resident
• rhizosphere

Introduction

Organic materials have a positive influence on soil microbial biomass and the improvement of soil productivity, thus it may be reasonable to argue that nature makes provision for backups in the modification of bacterial community structure. In a previous paper (Babalola & Glick, 2012), emphasis is laid on the prevalent agricultural practices and their magnified effects on soil microbe compared with relatively undisturbed agroecosystems supported by agroecology especially where there are perennial plants and nominal tillage. For an example, in one study, a combined bio-inoculation of diacetyl-phloroglucinol-producing plant growth-promoting rhizobacteria (PGPR) strains and arbuscular mycorrhizal fungi (AMF) synergistically improve the nutritional quality of wheat grain without negatively affecting rnycorrhizal growth (Roesti et al., 2006).

Another well-documented example is Bradyrhizobium japonicum strain CPAC 15 (which is equivalent to SEMIA 5079, the same serogroup as USDA 123), it is a highly competitive commercial strain and SEMIA 566 inoculations in soil cropped to soybean disperses and establish with vigor (Batista, Hungria, Barcellos, Ferreira, & Mendes, 2007). Some other commercial inoculants are Rhizobium leguminosarum bv. trifolii strain U204 (Blanco, Sicardi, & Frioni, 2010), R. leguminosarum bv. trifolii WSM1325 (Drew & Ballard, 2010) and Citrobacter freundii. There is a possibility of horizontal gene transfer events in B. japonicum CPAC and SEMIA prior to its introduction to Brazil (Godoy et al., 2008), and transfer of the GM plasmid between Sinorhizobium meliloti strains (Morrissey, Walsh, O'Donnell, Moenne-Loccoz, & O'GARA, 2002), as well as a residual effect of chromosomal information from previously used inoculant strains on indigenous microbes (Vessey & Chemining’wa, 2006).

In a study, it was shown that 19 days after sowing, the seed inoculant Pseudomonas fluorescens F113Rif had replaced some of the resident culturable fluorescent pseudomonads at the rhizoplane but had no effect on the number of these bacteria in the rhizosphere. The introduced pseudomonad induced a major shift in the composition of the resident culturable fluorescent Pseudomonas community (Moenne-Loccoz, Tichy, O'Donnell, Simon, & O'Gara, 2001), these interactions did not result in a modification of the structure or of the high level of strain diversity of the resident community.

Literature of the past few years illustrate that 1-aminocyclopropane-1-carboxylate (ACC) deaminase has been detected in a number of bacteria (Babalola, Osir, Sanni, Odhiambo, & Bulimo, 2003; Hontzeas et al., 2005; Khantsi, Adegboye, & Babalola, 2013). When a construct that contained a disrupted ACC deaminase gene was introduced into wild-type Enterobacter cloacae UW4 and cells where homologous recombination has occurred were selected for; it was established that the transconjugant had integrated the mutated acds gene in place of the wild-type version (Li & Kremer, 2000). The authors showed that the ACC deaminase mutant strain was unable to promote the elongation of canola roots.

Different strains of Bacillus thuringiensis are being used as bioinsecticide in today’s agroecosystem because the bacteria contain insecticidal proteins known as ‘Cry’ proteins. ‘Cry’ proteins are active against specific insect species and they exert no adverse effects on humans or domestic animals (Kleter, Groot, Poelman, Kok, & Marvin, 2009).

Underestimation of the bioluminescent derivative by counting bioluminescent colonies was attributed to competition for nutriments between the marked derivative and the soil bacteria, which, by decreasing the metabolic activity, affected the light production, hence the detection of the bioluminescent colonies. Underestimation of the antibiotic resistant derivative appeared to be related to the presence of the antibiotic in the selective medium used for counting antibiotic resistant cells. The use of marker systems in monitoring released bacteria which may therefore underestimate the survival and colonization of these genetically modified bacteria. Nevertheless, when the survival of a parent strain and of its genetically lacZY, lux or xylE marked derivative(s) were compared using the same method, the modified bacteria behave similar to the respective parent strains (Mahaffee & Kloepper, 1997; VanOverbeek, Vanveen, & Vanelsas, 1997). These marker genes had therefore no detectable effect on the environmental fitness of the marked bacteria.

These ecological considerations are being taken seriously and the success of a rhizobial inoculant in the soil depends to a large extent on its capacity to compete against indigenous strains. In a study to determine the effects of rhizobial inoculant on the indigenous microbial community, 16S rRNA genes were analyzed and compared by restriction fragment length polymorphism (RFLP) and temperature gradient gel electrophoresis (TGGE) molecular fingerprinting techniques. The results differentiate between alterations in the microbial community apparently caused by inoculation and by the rhizosphere effect and seasonal fluctuations induced by the alfalfa plants and by the environment. Only moderate inoculation-dependent effects could be detected, while the alfalfa plants appeared to have a much stronger influence on the microbial community (van Dillewijn, Villadas, & Toro, 2002).

There is increasing evidence from the literature that some plants enhance the proliferation of certain resident microbes, a term called the rhizosphere effect (Table 1); different plants tend to have different effects on the indigenous microbial populations. It is worth perusing the overview to see if nature makes provision for backups in soil microbial community structure. The purpose of this review is to gain insight in the modification of bacterial community structure, if any. The above studies underline the importance of this review article in ecological considerations.

Table 1. Fluctuations and effects of microbial populations in the soil on the indigenous microbial community associated with different plants.

Plant [Location]	Effect on the surrounding soil/ecological impact	Resident/indigenous population	Rhizobial inoculant	Incorporated gene	Primer set and/or detection method	References
Alfafa [Contained field experiment]	Enhance nodule occupancy	Sinorhizobium meliloti	S. meliloti M403	Proline dehydrogenase gene (putA)	RFLP, TGGE universal and specific primers	van Dillewijn et al. (2002)
	Minor sensational shifts in the rhizosphere but no systematic community shifts in bulk soil
Sugar beet seedling [Seed inoculant field grown]	Enhanced resistance to 2,4-diacetylphloro-glucinol	Fluorescent Pseudomonads	Pseudomonas fluorescens F113Rif	2,4-diacetylphloroglucinol (Phl)	ARDRA (16S r DNA primers) and RAPD (DAF-4)	Moenne-Loccoz et al. (2001)
	No modification of the structure or of the high level of strain diversity of the resident community
Legume crops (pea, lentil, or faba bean [Canadian Prairie]	Does it add plasmids into the soil’s rhizobial population?	Rhizobium leguminosarum bv. viciae (Rlv)	Commercial rhizobial inoculant. Rlv Strain PBC108	PBC108	N/A	Chemining’wa and Vessey (2006)
	There is tendency for increase d symbiotic efficiency with the use of commercial inoculant
Maize [Field-grown]	Phytostimulatory PGPR A. lipoferum CRTI caused a shift in the structure of the indigenous rhizobacteria community at 7 and 35 days	Diverse rhizobacteria	Azospirillum lipoferum CRT1	N/A	FGPS1490-72 and FGPL132-38a ARISA	Baudoin, Nazaret, Mougel, Ranjard, and Moënne-Loccoz (2009)
Maize [Pots]	Antagonistic activity against the tomato pathogen Ralstonia solanacearum	Heterotrophic bacteria	P. chlororaphis IDV1 and P. putida RA2	gfp	FAME analyses	Kozdroj, Trevors, and Van Elsas (2004)
Gerbera and pepper plants [Hydroponic]	N-Serve (R) selectively increases P. putida and reduced bacteria diversity 72 h after application	aerobic heterotrophic bacteria	Pseudomonas	N/A	T-RFLP	Pagliaccia et al. (2008)
Cucumber [Soil microcosms]	Antibiotic overproductionCHAO-Rif and CHAO-Rif/pME3424 had a small and transient impact on the population of resident Pseudomonads	Pseudomonads	P. fluorescens CHAO-Rif and CHAO-Rif/pME3424	rpoD	Spacer-ARDRA	Natsch, Keel, Hebecker, Laasik, and DEFAGO (1997)
Wheat (cultivar Lewjain) [pot]	As seed inoculant, strains Q2-87 (genotype) and LR3-A28 persisted in the rhizosphere over a 30-day period.	Heterogenous bacteria	P. fluorescens LR3-A28 and Q2-87 and 2,4-DAPG producing fluorescent Pseudomonad	phlD	RFLP	Mazzola, Funnell, and Raaijmakers (2004)
Soybean [Field]	Genes acquisition by the inoculant to establish effective symbioses with exotic host legume	Sinorhizobium (Ensifer) fredii	B. japonicum	nodY-nodA, nodC, and nifH genes	rep PCR (REP and Box AIR primers) RFLP	Barcellos et al. (2007)
Pepper [Phyllosphere]	Bt inoculation did change the phyllosphere microbial community structure significantly	Firmicutes	Bacillus thurigiensis	N/A	PLFA, DGGE	Zhang et al. (2008)
Tomato [Pot]	R. solanacearum depends on taxis to locate and infect plant hosts in its natural niches	Heterogenous	Ralstonia solanacearum K60	cheA and cheW	Southern blot and PCR, primers cheA-L and cheA-R	Yao and Allen (2006)
	The genes encoding are required for chemotaxis
Lettuce [Field]	Negligible, short-term effects resulting from bacteria treatments	Heterogenous	Serratia plymuthica 3Re4-18 Pseudomonas trivialis 3Re2-7	N/A	SSCP	Scherwinski, Grosch, and Berg (2008)
[mesocosms]	The variation observed was less important than the ‘non intentional’ variations due to the experimental conditions.	Trichoderma spp	Trichoderma atroviride 1-1237	N/A	T-RFLP	Cordier and Alabouvette (2009)
	The structure of the microbial communities evolved with time independently of the introduction of the biological control agent
Pepper [Field]	Strains CCR80 and ISE14 increase microbial activities and community composition, and affected bacterial and fungal community structure	Phytophthora capsici	P. corrugata CCR80 and Chryseobacterium indologenes ISE14	N/A	DGGE	Sang and Kim (2012)

Note: SSCP = Single-strand conformation polymorphism analysis; T-RFLP = Terminal Restriction Fragment Length Polymorphism; ARISA = Automated Ribosomal Intergenic Spacer Analysis; and PLFA = Phospholipid Fatty Acid.

Soil bacterial diversity and community structure

Rhizobacterial community structure is highly dynamic and influenced by different factors such as the plant’s age, the fertilizer input, and the type of bio-inoculant and a distance-related effect of the root on the bacterial community (Roesti et al., 2006). This can be inferred by the several cases hereafter presented.

Plant communities, growth stage and genotype

Illustrating how plant communities modulate ecosystem function, structural equation modeling was used to explore the direct and indirect effects of the plant community on soil diversity and potential function. Plant communities influenced archaea and bacteria via different pathways. Species richness and evenness had significant direct effects on soil microbial community structure, but the mechanisms driving these effects did not include either root biomass or the pools of carbon and nitrogen available to the soil microbial community. Species richness had direct positive effects on archaeal amoA prevalence, but only indirect impacts on bacterial communities through modulation of plant evenness (Lamb, Kennedy, & Siciliano, 2011). Increased plant evenness increased bacterial abundance which in turn increased bacterial amoA abundance. These results suggest that plant community evenness may have a strong impact on some aspects of soil ecosystem function. It was shown that a more even plant community increased bacterial abundance, which then increased the potential for bacterial nitrification. A more even plant community also increased total dissolved nitrogen in the soil, which decreased the potential for archaeal nitrification. The role of plant evenness in structuring the soil community suggests mechanisms including complementarities in root exudate profiles or root foraging patterns (Lamb et al., 2011). Findings showed that rotation and interplanting systems beneficially altered community structures of dominant soil bacteria and soil bacterial diversity (Li, Wu, Yang, & Wang, 2009).

Validation of microbial community structure showed a gradual positive increase of total microbial biomass between the intensive management reference site and the six grassland soils, and that soil organic matter storage is associated with changes in microbial biomass. There was a large increase in fungal and bacterial populations in the permanent grassland, but bacteria were more weakly affected by agricultural management practices than the fungi. Although potential functional diversity shifts in the bacterial community seemed to be related to the aging grassland gradient, this review was not able to highlight any significant difference in bacterial genetic diversity between the sites. There was, however, a strong relationship between fungal genetic diversity and the aging grassland. Finally, an increase in microbial activities (% mineralization) was observed according to the age of the meadow. Among agricultural management practices, grassland restoration may have a positive impact in maintaining the soil status (Plassart, et al., 2008). In addition, the age of the plant (Lottmann et al., 2000) was implicated in changes observed in the bacterial community structure. In other studies however, no change in bacterial community composition was observed in relation to plant age (Baudoin, Benizri, & Guckert, 2002; Normander & Prosser, 2000).

Plant genotype is becoming a growing importance as a determinant of the species and genetic composition of soil microbial community (Babalola, Sanni, Odhiambo, & Torto, 2007). Bacterial type significantly influenced the cowpea varieties with better performance over the non-inoculated control. The study indicated that plant genotype has differential responses to bacterial inocula because of the microbial activity.

Fertiliser, fallow and herbicide management practices

It has been reported (Ferreira, Andrade, Chueire, Takemur, & Hungria, 2000) that Brazilian soils were originally free of soybean bradyrhizobia but inoculation has established populations of some strains in most of the 12 million ha which are today cultivated with this crop. Those three strains, as well as most of the strains established in Brazilian soils, have been classified as Bradyrhizobium elkanii (Boddey & Hungria, 1997). Members of the genera Rhizobium, Azorhizobium, Allorhizobium, Mesorhizobium, Sinorhizobium, and Bradyrhizobium, have been used on a broad scale as inoculants to improve the nitrogen status of soils during crop rotation (Lugtenberg, Dekkers, & Bloemberg, 2001; Miethling, Wieland, Backhaus, & Tebbe, 2000; Schwieger & Tebbe, 2000). In 2001, 65% of the nitrogen supply to crops originated from PGPR. In the study, an inadequate soil management, crop rotation exclusively with grasses, reduced bradyrhizobia diversity, and a lower diversity of bean rhizobia was also reported as an effect of addition of N fertilizers (Ferreira et al., 2000).

However, the overall actinobacterial community structure in wheat rhizospheres was influenced by the sulfur fertilization regime, as shown by specific denaturing gradient gel electrophoresis of PCR-amplified 16S rRNA gene fragments, and asfAB clone library analysis identified nine different asfAB genotypes closely affiliated to the Rhodococcus isolates. However, asfAB-based multiplex RFLP/terminal-RFLP analysis of wheat rhizosphere communities revealed only slight differences between the fertilization regimes, suggesting that the desulfonating Rhodococcus community does not specifically respond to changes in sulfate supply (Schmalenberger, Hodge, Hawkesford, & Kertesz, 2009).

It has been shown that different fallow management practices influence the microbial biomass carbon and microbial activity as well as the distribution of the major microbial grouping in soil which is influenced by carbon stored in soil (Asuming-Brempong et al., 2008).

In some instances, pre-seed application of both 2, 4-D, and glyphosate altered the functional structure and reduced the functional diversity of soil bacteria, but increased microbial biomass C. These effects were not necessarily concurrent. The reduction in functional diversity was due to reduction in evenness, which means that the soil where both pre-seed herbicides had been applied was dominated by only few functional groups. In first year, two in-crop applications of glyphosate also reduced the functional diversity of soil bacteria when applied after pre-seed 2,4-D, as did in-crop sethoxydim + ethametsulfuron following pre-seed glyphosate. Even though significant differences between herbicides were fewer than non-significant differences, i.e. there were no changes in soil microbial community structure, diversity, or biomass in response to glyphosate or alternative herbicides applied to glyphosate-resistant canola in most cases, the observed changes in soil microbial communities could affect soil food webs and biological processes (Lupwayi, Harker, Clayton, O'Donovan, & Blackshaw, 2009).

Another explanation to the aforementioned trend is that no significant pesticide (fungicide vinclozolin and insecticide X-cyhalothrin) effects on soil microbial biomass or functional bacterial diversity were observed, but the functional structures of soil bacteria were altered. In 1 of 12 cases, the control treatment had a different soil bacterial community structure from the three pesticide treatments. The fungicide treatment had different bacterial community structures from the control or insecticide treatments in 3 of 12 cases, the insecticide treatment had different community structures from the control or fungicide treatments in 4 of 12 cases, and the combined fungicide and insecticide treatment had different community structures from the other treatments in 3 of 12 cases. Therefore, evaluating soil bacterial functional structures revealed pesticide effects that were not detected when bacterial diversity or microbial biomass were measured in canola rhizosphere or bulk soil (Lupwayi et al., 2009).

Bio-inoculation

In the presence of large rhizobial populations, typical of higher pH soils in western Victoria, effective granular or slurry inoculants did not increase nodulation in un-inoculated treatments (Denton et al., 2009). It is therefore reasonable to expect that nodulation effects attenuated as background nodulation from soil rhizobia increased.

When phosphate-solubilizing bacteria abundance and diversity were examined during two consecutive years, 2007 and 2008, in a crop/pasture rotation experiment in Uruguay; both phosphate-solubilizing bacteria and culturable heterotrophic bacteria populations were affected by crop rotations only in one of the two years sampled (Azziz et al., 2012). In 2007, they had a tendency towards being more abundant and diverse in those treatments with less intensive use of soil (natural prairie and permanent pasture). It was concluded that there is a need for multi-year studies to more fully assess the interaction of environmental variation and soil management (Azziz et al., 2012).

Experimental results showed that soil fungistasis (the inhibition of the growth of fungi) capacity was closely correlated with soil bacterial community composition and diversity, such that soil fungistasis declined with the decrease of soil bacterial diversity. Meanwhile, the bacterial community composition and structure were significantly different along the gradient of soil fungistasis tested. alpha-Proteobacteria, beta-Proteobacteria, Flexibacter, and some uncultured soil bacteria contributed to soil fungistasis in combination with some other special bacteria (Pseudomonas and Acidobacteria) which were know to be key species in suppression of fungal growth (Wu et al., 2008).

In the structure and composition of bacterial communities in sandy loam and silty clay soils; unlike the diversity in the sandy loam blends, the diversity in all silty clay blends was greater than the control. Phylogenetic analysis of bacterial isolates showed the addition of spent foundry sand to soil does bring about bacterial community-level changes; these changes are similar to that of blending soil with clean silica sand (Dungan, Kim, Weon, & Leytem, 2009).

It has also been suggested that the presence of plants for three years had direct impacts on soil function in terms of total heterotrophic respiration and on microbial biomass and microbial community structure (Singh, Dawson, Macdonald, & Buckland, 2009). However, the plant species-specific impact on bacterial community structure was weak, and differences were mainly driven by sample field location (Singh et al., 2009). Among the abiotic factors, moisture had a comparatively higher impact on bacterial communities compared to soil N and C. However, the fungal community was not affected by the soil moisture but soil N and C had a stronger impact than any on the bacterial community. These results indicate that the microbial community structure in the natural environment is influenced by interactions between both biotic and abiotic factors (Singh et al., 2009).

Another study is on the microbial communities in rice rhizosphere cultivated on arsenic-contaminated soils which might alter soil microbial populations that may impact arsenic chemistry (Somenahally, Hollister, Loeppert, Yan, & Gentry, 2011). The relative abundance of bacteria increased over the course of the growing season, while archaeal and fungal gene abundances decreased (Somenahally et al., 2011). Bacterial community structure and composition were significantly different between arsenic amended and unamended plots, as well as between the flooding treatments. Proteobacteria was the predominant phylum detected in most treatments with a relative abundance of 24–29%. The relative abundance of iron-reducing bacteria was higher with the continuous flood compared to the intermittent flood treatment, implying greater relative iron reduction and possibly arsenic release from the iron oxides under the continuously flooded conditions (Somenahally et al., 2011).

Microbial ecology of indigenous endorhizosphere micro-organism

Although genetic engineering in agriculture is in its 16th year of employment in commercial cultivation, issues of risk and fear of the unknown is yet to be completely abated. But even before the advent of genetically modified organism (GMO) there have been mild issues on the prevalence of introduced microbial inoculant on the indigenous resident microbes in the soil. Cases of displacement and temporary shift in favor of the introduced microbes have been reported (Miethling et al., 2000).

Elevated atmospheric CO₂ content at the rhizospheres of two grassland plants, Lolium perenne (ryegrass) and Trifolium repens (white clover) increased the most probable numbers of heterotrophic bacteria in the rhizosphere of Lolium perenne (ryegrass). This effect lasted only at the beginning of the vegetation period for Trifolium repens (white clover). Thus providing evidence for a CO₂-induced alteration in the structure of the rhizosphere bacterial populations and suggesting a possible alteration of the PGPR effect (Marilley, Hartwig, & Aragno, 1999).

Pseudomonas fluorescens strain BR-5 stimulated the growth of maize in natural soil and inhibited fungal root pathogens in vitro. Strain BR-5 was detected inside plant cells, indicating that it is able to colonize the endorhizosphere. No significant effect was detected on soil or ectorhizosphere microbial population after inoculation of strain BR-5 onto seeds (Botelho et al., 1998).

The non-gall-forming phytopathogen Erwinia chrysanthemi 3937 possesses an Rsm system, which plays a critical role in gene expression and has a profound effect on bacterial metabolism and behavior in many prokaryotic species. The Rsm system has been reported to control the production of different extracellular enzymes like pectinases, proteases, and cellulases; and secondary metabolites such as phytohormones, antibiotics, pigments, and polysaccharides (Yang et al., 2007). Genetically engineered micro-organisms altered for superior function, have been employed in biological control of plant pathogens in agricultural crops (Paoletti & Pimentel, 1996).

Methods

Rhizosphere bacterial community structure depends on the plant species (Table 1). The DNA sequences of bands that appeared as a result of a possible rhizosphere effect exhibited similarities with the sequences of bacteria which could be expected to be found in soil (see Table 1). DNA isolation from soil, PCR amplification, and RFLP and TGGE analyses of 16S rDNA genes each has biases which could affect the outcome of the patterns presented. Hence, the limitations of the techniques should be taken into account when results presented (Table 1) are interpreted.

Amplified ribosomal DNA restriction analysis

Community structure of two grassland plants was assessed after isolation of DNA, PCR amplification, and construction of cloned 16S rDNA libraries. Amplified ribosomal DNA restriction analysis (ARDRA) and colony hybridization with an oligonucleotide probe designed to detect Pseudomonas spp. showed, under elevated atmospheric CO₂ content, an increased dominance of pseudomonads in the rhizosphere of L. perenne and a decreased dominance in the rhizosphere of T. repens (Marilley et al., 1999). Other authors investigated how Phi-producing Pseudomonas biocontrol inoculants can influence nontarget community. The size of indigenous isolates within the ARDRA groups (Moenne-Loccoz et al., 2001) was neither influenced by the presence of roots nor inoculation with P. fluorescens F113Rif.

Polymerase chain reaction with denaturing gradient gel electrophoresis (PCR-DGGE)

Although molecular detection methods also come with certain limitations, its sensitivity over traditional culture technique cannot be overemphasized. One can envision a number of additional findings in an assessment for the presence of metal-resistant sulfate-reducing bacteria: isolated bacterial consortium reduced 91% of the added 3.9 g/L of sulfate after 28 days, precipitating 100% of the Fe, 96% of the Zn, and 97% of the Cu at the same time. However, even as total bacterial numbers and numbers of culturable sulfate-reducing bacteria showed low or absence of inhibitory effects of the metals on the bacterial consortium, directed PCR-DGGE profiles of the sulfate-reducing bacterial communities obtained from the Icelandic fumaroles soil sample showed that bacterial diversity decreased significantly after metal addition. Phylogenetic analysis of 16S rRNA gene sequences revealed that the remaining two out of 12 ribotypes were affiliated with the genera Clostridium and Desulfovibrio, with C. subterminale, C pascui, C. mesophilum and C. peptidovorans, and D. desulfuricans identified as their closest relatives (Alexandrino et al., 2011).

In another study, bacterial community fingerprints were analyzed from tubers of seven field-grown potato genotypes. The plant genotype significantly affected the number of culturable bacteria only at one field site. PCR-DGGE fingerprints of the 16S rRNA genes of bacterial communities from the tuber surfaces revealed that the potato genotype significantly affected the Pseudomonas community structure at one site. However, the genotypes showed nearly identical fingerprints for Bacteria – Actinobacteria, Alphaproteobacteria, Betaproteobacteria, Bacillus, and Streptomycetaceae. In conclusion, tuber-associated bacteria were weakly affected only by the plant genotype (Weinert et al., 2010).

Inoculation of a P. Putida strain did not alter the culture-independent 16S rDNA denaturing gradient gel electrophoretic patterns of bacteria in the rhizosphere of transgenic potatoes (Lottmann et al., 2000). The origin and composition of bacterial communities in developing barley (Hordeum vulgare) phytosphere were determined by PCR-denaturing gradient gel electrophoresis, the study showed that rhizoplane bacteria primarily originated from the surrounding soil (Normander & Prosser, 2000). The bacterial community structure of the rhizosphere soil and the rhizoplane/endorhizosphere was similarly determined by PCR-DGGE (Roesti et al., 2006). Besides, inoculation with Azospirillum brasilense exerts beneficial effects on plant growth and crop yields. A DGGE fingerprint analysis revealed that plant inoculation with A. brasilense had no effect on the structural composition of the bacterial communities, which was also found to be very similar at the root tip and at zones of root branching (Herschkovitz, Lerner, Davidov, Okon, & Jurkevitch, 2005). They concluded that inoculation with the A. brasilense (PGPR) causes little disturbance in the rhizosphere and rhizoplane of maize.

Assessments on plant microbial community were made using culture-independent techniques including phospholipid fatty acid analysis (PLFA) and 16S rRNA gene-directed PCR-DGGE. PLFA results indicated that both total and bacterial biomass increased after application of cypermethrin pesticide and Gram-negative bacteria became predominant (Zhang et al., 2009). DGGE analysis confirmed a significant change in bacterial community structure within pepper plant phyllosphere following cypermethrin pesticide application where different dendrogram clusters were observed between control and treated samples (Zhang et al., 2009). Phylogenetic analysis also suggested a change in bacterial phyla following treatment, where bands sequenced within control cultures were predominantly members of the Firmicutes phylum, but those bands sequenced in the treated samples were predominantly members of the Bacteroidetes and gamma-Proteobacteria phyla (Zhang et al., 2009). In conclusion, this study revealed an increase in bacterial abundance and a shift in community composition within the pepper plant phyllosphere following the pesticide application, and highlighted the effective use of PLFA and PCR-DGGE for studying the effect of pesticides upon indigenous phyllosphere microbes (Zhang et al., 2009).

In a different experiment, the structure and composition of bacterial communities in sandy loam and silty clay soils amended with 30% spent foundry sand grown with or without perennial ryegrass and samples were studied with PCR-DGGE (Dungan et al., 2009).

RFLP and TGGE

Neither the patterns obtained by RFLP analysis with universal or β- and γ- proteobacterial primers nor the patterns obtained by TGGE analysis with β-proteobacterial or α-proteobacterial primers revealed differences between M403-inoculated plots and the non-inoculated or M401-inoculated plots (van Dillewijn et al., 2002).

RFLP and TGGE results (van Dillewijn et al., 2002) allowed the detection of alterations in the microbial community structure of alfalfa plants. Changes in microbial community structure were reported to be the result of inoculation, rhizosphere effect, and seasonal fluctuations brought about by the alfalfa plants and by the environment.

It has been shown using principal component analysis and fatty acid signatures of soil microbes that sown and naturally colonized plots diverged from the initial soil microbial community of the field in two opposite directions. However, the author detected no effect of soil inoculations (Hedlund, 2002). Inoculation of legume crops with nitrogen-fixing soil bacteria, collectively known as rhizobia, has been used widely to improve legume productivity in the field. However, the introduced inoculants often fail to become established in soils with indigenous rhizobial populations (see van Dillewijn et al., 2002). In such soils, the better-adapted indigenous populations tend to perform better than the inoculants for nodule formation and saprophytic survival. Thus, for an inoculant to be effective, the strains used must not only fix nitrogen efficiently but also be highly competitive (van Dillewijn et al., 2002).

Data concerning the persistence of M403 in soil and the plasmid stability in this strain have shown that the modified putA gene tends to disappear with time. This suggests that the probability of gene transfer to natural populations is low. However, a rhizobial strain with a putative metabolic advantage, such as M403, may displace unrelated indigenous proline-dependent microbial populations from the rhizosphere. In most field release studies to date, non-rhizobial genetically modified micro-organisms were shown to cause only slight, transient perturbations in the microbial communities indigenous to the rhizospheres of maize and wheat. Similarly, lysimeter experiments (field experiments with polyvinyl chloride cylinders containing soil columns) with rhizobial luciferase gene-tagged S. meliloti strain L33 and a RecA derivative, L1, showed that these inoculants did not affect the indigenous microbial community (Schwieger & Tebbe, 2000).

Root exudates are an important source of nutrition for many rhizosphere micro-organisms and changes in their composition may affect the patterns and activities of rhizobacterial populations. Fluctuations in the rhizosphere microbial populations associated with maize and potatoes have been studied by several authors. These authors observed that some plants enrich certain populations from the surrounding soil in what is known as the rhizosphere effect. For instance, development of the maize plant specifically affected populations of Burkholderia cepacia (Dicello et al., 1997), Paenibacillus azotofixans (Seldin et al., 1998), and Pseudomonas (Picard, di Cello, Ventura, Fani, & Guckert, 2000). It is well known that the addition of non-indigenous microbes to soil can affect the indigenous rhizosphere population (Whipps, 2001).

The authors, who used TGGE with universal primers, detected minor seasonal shifts in the rhizosphere but did not detect systematic community shifts in bulk soil (van Dillewijn et al., 2002). The DNA sequences of bands that appeared as a result of a possible rhizosphere effect exhibited similarities with the sequences of bacteria which could be expected to be found in soil.

The rhizosphere microbial communities can be affected by a wide range of factors including agronomic practices (Alvey, Yang, Buerkert, & Crowley, 2003; Kennedy, Choudhury, & Kecskes, 2004; Lupwayi, Rice, & Clayton, 1998). It is, therefore, necessary to study the microbial community dynamics in the field before modifying the agricultural practice, especially when using bio-inoculants to improve soil health or crop yield (Roesti et al., 2006), and to study the effects of PGPR/AMF bio-inoculations on the bacterial community structure (Roesti et al., 2006).

As suggested by (Ciccillo et al., 2002), this result could mean that the negative effect caused by a modification of the bacterial community equilibrium was overcome by the beneficial effects of the bio-inoculants (Roesti et al., 2006). Our bio-inoculants may therefore have modified the bacterial community equilibrium towards the selection of beneficial populations. Plants responded better if the PGPR strains were isolated from the native rhizosphere (Roesti et al., 2006).

Transgenic plants and microbes so far so good

Transgenic plants are plants carrying genetic traits that are not naturally present in or that have been altered artificially outside the plant (Lilley, Bailey, Cartwright, Turner, & Hirsch, 2006). These plants are modified to have long lifespan, pest and diseases resistance, nutritional improvement, herbicide tolerance, and resistance to abiotic factors such as drought and nitrogen deficiency (Icoz & Stotzky, 2008). However, the possible effects of genetically modified plants on human health and ecological functioning are a call for concern (Sessitsch, Reiter, & Berg, 2004). The artificial altering of genetic traits is engineered by a process called horizontal gene transfer (HGT). Horizontal gene transfer is defined as a movement of non-sexual genetic traits or information from one organism to another (Keeling, 2009). These traits include resistance to bacterial, viral, and fungal plant diseases, herbicide tolerance, insect diseases resistance, tolerance to abiotic factors, and production of industrial chemicals (Sessitsch et al., 2004). Some genetically modified plants have been successful to effect targeted changes in microbial community composition by inhibiting plant pathogenic organisms (Ahrenholtz, Harms, de Vries, & Wackernagel, 2000). However, most studies show either minor non-target effects (Dunfield & Germida, 2001; Oger, Mansouri, & Dessaux, 2000) or no detectable non-target effects (Bumunang, Babalola, & Barros, 2013; Heuer, Kroppenstedt, Lottmann, Berg, & Smalla, 2002; Lottmann & Berg, 2001).

Plants being major drivers of the soil ecosystems provide fundamental services like the regulation of water quality and quantity, nutrient cycling, carbon sequestration, and the bioremediation of waste that supports plant growth (Lilley et al., 2006). Soil anchors plants and harbors diverse range of micro-organisms (bacteria, protista and fungi). Microbial interactions in the soil are influenced by factors such as water, oxygen, available nutrients, temperatures, and soil pH. These micro-organisms in the soil can be classified as beneficial, deleterious, or neutral with respect to root and plant health. The region of soil influenced by plant roots is called the rhizosphere. The rhizosphere is a biological active zone of soil where micro-organisms and plant roots interact, and is of major importance for plant growth as well as for nutrient cycles and ecosystem functions. Root exudates in root–soil interface create a unique microbial microenvironment, differentiating it from bulk soil not influenced by root. Furthermore, root exudates quality and quantity vary with plant developmental stage, plant species and determines microbial community structure.

Bacillus thuringiensis (Bt) corn is a genetically modified corn. Bt corn contains insecticidal toxic proteins (Bt endotoxins) having many advantages for crop production such as increase in insect resistance, increased grain yield, and plant growth. Fang, Chao, Roberts, and Ehlers (2007) stated that ‘research on the effects of altered chemicals and physical properties of Bt corn exudates and residues on decomposition have yield conflicting results.’ However, a good knowledge of the effects of GMPs residues and exudates on soil micro-organisms is essential for understanding the long-term environmental and agronomic impacts of GMPs, and for developing appropriate management practices for minimizing potential negative effects (Fang et al., 2007). Genetically modified plants not only affect the actual target organisms but also endanger numerous agriculturally important beneficial organisms (www.NABU.de/gentechnik).

Based on observations with a number of different micro-organisms, the sorts of physiological impairments that can result from a metabolic load being placed on transformed cells expressing high levels of foreign protein include alterations in cells size and growth rate (Hong, Pasternak, & Glick, 1995). Although the percentage of cell survival obtained with the yellow melanin is less than that previously obtained with the salmon/red melanin from the parent cell, the yellow pigmented cells seem to better tolerate higher UV dose than the parent cells (Le Curieux-Belfond, Vandelac, Caron, & Séralini, 2009; Lupwayi et al., 2007). These elements might be responsible for the experimental evidence of high rates of HGT of symbiotic genes from strains belonging to this serogroup to other indigenous or naturalized rhizobia under field conditions in Brazil (Batista et al., 2007; Barcellos, Menna, Batista, & Hungria, 2007; Godoy et al., 2008). Some of those CDS had highest similarity with USDA 110 using BLASTP, indicating HGT events prior to the introduction to Brazil (Godoy et al., 2008).

The effects of glyphosate-resistant (GR) crops on non-target soil organisms in agroecosystems were evaluated in the form of soil microbial biomass C, bacterial functional diversity and community structure, and dehydrogenase enzyme activity in a field experiment (Lupwayi et al., 2007). In the study, microbial biomass C increased with increasing frequency of GR crops in the rhizosphere, and had no effects in bulk soil.

The shifts in the structures of bacterial communities related to GR crop frequency detected were few and inconsistent over a wide range of growing conditions and crop management.

Prominent factors to consider before production and commercialization of aquatic GMOs: the case of transgenic salmon (Le Curieux-Belfond et al., 2009). The real number of viable cells introduced into the soil in a microgranular formulation was likely to have been smaller than the theoretical number. Growth-promoting effect due to liberation of nutrients or to the eradication of a part of the indigenous microflora may otherwise enter into conflict with the inoculant strain. Most rhizobacterial communities are drawn from the soil and fewer originate from seed-associated micro-organisms but all are selected for growth in the rhizosphere, reaching a high density in that milieu (Curl & Truelove, 1986).

Another study identified, in the case of plants producing attacin/cecropin or T4-lysozyme, a change of the exudation pattern could have been due either to a qualitatively and/or a quantitatively altered excretion of organic compounds, or to the release of the produced antibacterial substances (Rasche et al., 2006). It is assumed that the infection with the blackleg pathogen E. carotovora ssp. atroseptica caused alterations in plant physiology leading to qualitative and quantitative changes of root exudation patterns. These changes were probably responsible for the slightly altered microbial communities (Rasche et al., 2006). The estimates of microbial biomass show that bacteria and saprophytic fungi increase in soil where seed mixtures have been sown.

Main considerations and trade-offs

To date, the general, inherent uncertainties about the duration of changes incurred due to modification will not go away. It is open to serious doubt whether the molecular detection methods and experimental approaches are all consistent with each other.

Irrespective of all the above-mentioned considerations on bacterial diversity, community structure, equilibrium, microbial inoculant, resident microbe, rhizoplane microbe and soil rhizosphere, efforts must be directed towards the advancement of each of the latter. Experimental approaches must take account of overall trends, rather than focusing on isolated findings that have only limited prospects of being repeated.

Questions

1. Do non-target resident bacterial communities always have the capacity to buffer the ecological impact to which they are subjected following the introduction of taxonomically related bacterial inoculants?

2. If the indigenous populations in a plant rhizosphere originated from previously used commercial inoculants, how genetically close or divergent are they compared to the original strains?

Conclusions

This paper addresses a number of primary concerns about microbial inoculants, GMO, and biosafety. The developing roots generally support fast-growing micro-organisms like bacteria, whereas mature roots support slower-growing micro-organisms like fungi and actinomycetous bacteria. As far as the dynamics of the rhizosphere effect are concerned, it is necessary to obtain all the above-mentioned findings. From the technical point of view, there is need for multi-year studies to more fully assess the interaction of Feedback between plants and micro-organisms in order to answer the questions (1) Does GMO crops interfere with indigenous rhizosphere micro-organisms? (2) Any capacity to compete against indigenous strains?

Acknowledgments

The author thanks the reviewers for critically reviewing the article. She was a recipient of NWU grant through the Faculty of Agriculture Science and Technology. The National Research Foundation, South Africa is acknowledged for funds that have supported research in her laboratory.