There are numerous bacterial genera whose representatives can influence plant growth and production. Among these representatives, there are pathogens that can suppress plant diseases and they are used as biocontrol strains. Another group or bacterial species can contribute to increased plant growth by enhancing the availability of nutrients. These bacteria constitute the bio-fertilizers and are known as growth-promoting rhizobacteria (PGPR) as well. The term PGPR is associated with the ability of these bacteria to grow well at the interface between soil and plant root (the rhizosphere). PGPR can be applied either as seed coating or directly to soil. However, to exert their growth-promoting effect, sufficient numbers of the introduced PGPR have to survive in soil and rhizosphere, which not always happens. Consequently, the efficacy of PGPR is not always sufficient for commercial applications and there is a need to improve their performance. One of the possible approaches is to apply genetic modifications to facilitate the survival efficiency.
Any microbial cell introduced into the environment will encounter a large number of biotic and abiotic factors affecting its survival. Both biotic and abiotic factors are equally important. Thus, high clay content, high pH, and relatively high moisture content can have a positive effect on bacterial survival. On the contrary, dry periods, presence of competing microorganisms, predation by protozoa, and lysis by bacteriophages negatively affect the number of introduced bacteria. Speaking of biotic factors affecting the activity and survival of introduced bacteria, the presence of plant roots that provide nutrients to the microorganisms living in their vicinity is very important. Among the microorganisms that are well adapted to the rhizosphere, there are members of the genera Agrobacterium, Azospirillum, Azotobacter, Bacillus, Erwinia, Pseudomonas, Rhizobium, and Xanthomonas.
Microbial survival depends on the interrelation between the environmental conditions and the physiological state of the bacteria. As a result of these interactions, bacterial cells can switch their metabolism to different physiological states. For instance, cells can become more stress resistant or form dwarf cells, they can produce exopolysaccharides for protection, they can enter a viable but non-culturable state, and some are able to form spores or associations with plants.
One can speculate that the survival pattern of the GM bacteria will follow the one of their wild-type parents. In fact, this extrapolation should be applied with some precautions. Firstly, the expression of the inserted genes requires an extra amount of energy, which could reduce their environmental fitness. In addition, the insertion could have disrupted unknown functions weakening the competitiveness of the strains. Secondly, it is possible for the GMMs to evolve and adapt to the prevailing environmental conditions via natural selection. This last statement is supported by evidence for evolutionary adaptation of bacteria to degrade the herbicide 2,4-dichlorophenoxyacetic acid resulting in increased competitive fitness to use succinate as a substrate. Similarly, it has been reported that environmental stresses could alleviate the debilitating effects of mutations: organisms may become more tolerant to genetic perturbations under certain environmental stresses.
GMMs have been shown to survive even better than the wild-type strain in studies under artificial growth conditions. However, enhanced survival of GMMs has rarely been observed under field conditions. Often, the population of introduced bacterial cells declines rapidly in soil, and the GMM species survive in a mode similar to that of non-modified bacteria. There are a lot of experimental studies in which no difference in survival between GMM and the parent strain could be detected (for Pseudomonas chlororaphis, P. fiuorescens and Sinorhizobium meliloti). Furthermore, some GMMs were reported to be outcompeted by the parent strains. It is speculated that the presence of a number of constitutively expressed marker genes in a GMM had a negative effect on its survival in competition with the wildtype strain. Most probably, it is the metabolic load that is responsible for the decreased fitness, since this effect does not occur under nutrient-rich conditions.
To correctly interpret bacterial survival data, it is of crucial importance to use a reliable method for detection, since cells that enter a non-culturable state cannot be detected with standard cultivation-based techniques. Moreover, various studies have shown that GMMs introduced into soil become non-culturable. The presence of viable but non-culturable cells, dead cells or naked DNA detected using molecular techniques contributes to the complexity and the ecological significance of GMMs and their fitness in the context of the effect of the genetic modification introduced. The reliable way in which the effect of small differences in fitness will be measurable is to co-inoculate a GMM and its parental strain placing them in direct competition. However, results from such direct competition experiments have to be interpreted with care as well, since commercial application of GMMs does not include direct competition between GMMs and their wild-type strains.
All these data, contradictory to some extent, show that a conclusion regarding the survival of GMMs as compared to their parental strains cannot be definitely drawn. In each case where the colonizing ability and survival of the GMM are of importance, these parameters will have to be determined.
The possible effects of the release of GMMs in natural microbial ecosystems are quite diverse. The range encompasses events such as input of organic substrate, displacement of species, changes in population structure, and possible loss of certain functions; production of toxic metabolites, which might lead to disturbance of key ecological processes. It should be taken into consideration that small changes in community composition are difficult or even impossible to determine, and the relationship between microbial diversity and ecosystem functioning is not quite clear. Undoubtedly, the soil microbial diversity is enormous, with a high redundancy of functions. Disappearance of a few species with certain functions will be difficult to detect, since many functions can be performed by a large number of different microbes. Thus, only extreme disturbances might affect soil microbial communities to the extent that certain functions will be negatively influenced.
The limited culturability of indigenous soil microflora is one of the major problems in microbial ecology. DNA- and RNA-based techniques, which do not involve cultivation of microorganisms, are currently used to detect the impact of GMMs on the indigenous microbial community. The methods that are suitable to analyze shifts in community structures include denaturing gradient gel electrophoresis (DGGE), amplified ribosomal DNA restriction analysis (ARDRA), terminal restriction fragment length polymorphisms (T-RFLP), and single-strand conformation polymorphism (SSCP).
GM derivatives of bacteria contribute to an enhanced nutrient availability for plants, and thereby increase plant growth.
The most important bio-fertilizers are bacteria, such as Azospirillum and Rhizobium that can fix nitrogen. Rhizobium, Bradyrhizobium, and Sinorhizobium are plant symbionts, which form root nodules in leguminous plants and fix atmospheric nitrogen. These bacteria have been used widely as plant inoculants to increase the yield of leguminous crops. There is a long history of safe use of non-modified rhizobia as inoculants to increase crop yields. However, the yield increase is variable, and the success of inoculants seems to be dependent on competition with indigenous strains that are usually less effective. Rhizobium, Bradyrhizobium, and Sinorhizobium have been reported to survive in soil for years, in some cases even without the presence of their specific host. Rhizobium has been shown to be able to form nodules when its host plant is planted again after several years. This shows that presence of the host plant is not strictly necessary for their survival, but also characteristics of the strain not related to symbiosis play a role in its survival in bulk soil for years. Fast-growing Rhizobium species have been found to be more susceptible to desiccation than the slower-growing Bradyrhizobium.
Except for carbon dioxide (CO2) which plants obtain from the atmosphere, plants get all their nutrients from soil. Nature has developed various mechanisms to supply plant nutrients by means of renewable resources, and the best example of this principle is biological nitrogen fixation in leguminous plants. Nitrogen-fixing bacteria can be regarded as a self-propagating source of nitrogen for plants. Unfortunately, not all plants are able to perform such interaction with N2-fixing bacteria. That is why, at present, plant production yields still largely depend on input of chemical fertilizers. Most of these fertilizers are very mobile in the soil and are supplied in greater quantities than required for optimal plant growth. The loss of valuable compounds is not only of economic importance; this also causes serious problems for the environment, through leakage in surface and ground water and accumulation in the atmosphere.
Different strategies have been developed that aim at better uptake of fertilizers by plant roots. These include other formulations of fertilizer (e.g. slow-release fertilizer) and the use of plant-growth-promoting rhizobacteria (PGPR).
PGPR can exert their effect in both direct and indirect way. The indirect pattern comprises exercise of biocontrol of pathogens and deleterious microorganisms. The best documented example of PGPR acting in a direct plant-growth-promoting way is phytostimulation. Various bacterial genera are capable of producing plant-growth-stimulating factors (auxins, cytokinins, etc.) and when colonizing the roots of plants, they promote root growth. This assures a better uptake of water and nutrients by the plants and can result in higher crop yields.
GM Azospirillum increases nitrogen uptake
It is known that Azospirillum strains can promote plant root development and increase nitrogen uptake through the phytohormones produced by them. However, the mechanisms by which, and the conditions under which, these bacteria produce phytohormones as well as the interaction between bacteria and plant roots, are still not defined and require a better understanding.
To elucidate these mechanisms, several important questions/approaches should be addressed:
At present, GM Azospirillum strains with these basic features are available. Research with these strains is focused on their impact on resident microbial populations, plant growth and nitrogen uptake rates from soil. These studies are being conducted in lab experiments (i.e. growth cabinet and glasshouse studies) in order to gain vital information on the way GM strains are likely to behave under field conditions. The experiments are conducted with a range of crops, soil types and climate conditions, representing the existing agricultural parameters within Europe. Despite of the advancement of these research studies, extensive and careful testing under containment is required before the GM Azospirillum can be considered for field release.
GM Rhizobium strains with increased competitiveness
Legume inoculation with highly efficient nitrogen-fixing bacteria is a widely used approach to increase the productivity of leguminous crops. This inoculation is not always successful, since native soil bacteria with low nitrogen-fixing efficiency can out-compete the introduced strains in terms of nodulation initiation. What is critical for the successful use of rhizobial inoculants is their competitiveness, i.e. the ability to dominate nodulation. Thus, inoculant strains are modified in a way that they occupy a sufficient number of root nodules to provide high rates of nitrogen fixation for the plant host.
Experiments with Sinorhizobium meliloti strains from diverse geographical origins regarding their competitiveness for alfalfa roots have shown that, in all cases, this property has been enhanced by genetic manipulation. The said genetic manipulation comprises modification of the expression of the nifA gene which is responsible for the control of all the rest nitrogen-fixation (nif) genes. When GM S. meliloti strains were mixed with wild-type ones, the former occupied most of the nodules on the alfalfa roots. The precise mechanism of this improvement is not understood yet but it is speculated that nifA regulates the expression of genes different from the nif cluster, resulting in an advantage during nodule formation and development.
Another feature that contributes to the nodulation competitiveness of Rhizobium strains is their ability to efficiently recognize the plant root. This is very important because the efficient inoculation means lower doses of the bacterial strain. Furthermore, the movement of the inoculation strain towards the plant roots is another factor influencing competitiveness. Experiments with GM Rhizobium leguminosarum strains engineered to express the β-glucuronidase reporter gene (gusA) showed that the percentage of the nodules induced by the GM gusAlabeled strain is higher compared to the nodules induced by a flagella-deficient non-motile strain. In this way it was proven that the functional flagella are required for effective competition for nodulation.
All these data provide valuable information on the mechanism of root attraction allowing the development of Rhizobium strains with enhanced nodulation competitiveness and increased host specificity.
Impact of GM Rhizobium strains on arbuscular mycorrhizal fungi
Arbuscular micorrhizal fungi are an important group of fungi that form symbiotic relationships with plants. A major question is whether the application of GM Rhizobium strains with increased competitiveness leads to an increase in the colonization and nodulation of the plant root or it interferes with the beneficial symbiotic relationship.
In lab and green-house experiments, it has been shown that a GM Sinorhizobium meliloti strain with improved nodulation ability does not interfere with any aspect of mycorrhizal formation by the representative AM fungus Glomus mosseae. On the contrary, GM S. meliloti increased the number of AM colonization units and the nutrient acquisition ability of the mycorrhizal plant.
GM Rhizobium strains: field release
Several Rhizobium species have been genetically modified either to improve nitrogen fixation, or to study their survival, making use of marker genes through field trials.
Thus, a Tn5-marked R. leguminosarum strain introduced into a field as an inoculant for peas and cereals persisted for 5 years in the plots where peas were grown. The persistence of the strain was attributed to the soil type, the cultivation of the proper host plants, and the climate conditions. Potential non-target effects on the microbial ecosystem were not studied.
The use of an improved R. meliloti strain, with additional copies of nifA and dctABC, resulted in a 12.9% increase in alfalfa yield in a field study. However, at sites with high nitrogen concentrations or native rhizobial populations, the alfalfa yield did not increase.
The fate of a Tn903-marked R. meliloti strain introduced into alfalfa-planted field plots was studied and it was found that the cell numbers decreased rapidly after inoculation. One year after introduction, the numbers of introduced cells had dropped to below the numbers of indigenous rhizobia.
In a contained field experiment, a GM S. meliloti strain with enhanced competitiveness for nodule occupancy was released in the rhizosphere of alfalfa. Effects of the GMM and the wild type on the indigenous microbial communities were studied by restriction fragment length polymorphism (RFLP) and temperature gradient gel electrophoresis (TGGE). Inoculation of wild type and GMM had only limited effects. It appeared that alfalfa plants had a greater influence on the microbial community than the inoculated strains.
Both the fate and ecosystem effects of a Luc-marked S. meliloti in a field experiment with Medicago sativa were studied. The bacteria were detected up to 12 weeks after introduction. No effects of the strains on carbon and nitrogen concentrations in the soil could be detected, and there were no differences in the total number of colony-forming units of indigenous microorganisms. Over a thousand bacterial isolates obtained from the plots were further studied by ARDRA, and the dominant groups were identified by 16S rRNA sequencing. In the rhizosphere of M. sativa, the numbers of Alcaligenes and Pseudomonas were reduced as a result of the inoculation. Molecular analysis by studying the SSCP banding profiles revealed shifts confirming the effect of the inoculum on the native microbial population.
In China, wild-type and GM Alcaligenes faecalis isolates have been introduced into rice fields at a large scale to improve crop productivity. A. faecalis, a non-nodule-forming nitrogen-fixing isolate, was genetically modified by insertion of a constitutively expressed nifA regulatory gene. Nitrogen fixation appeared to be 15–20% higher and the yield was 5–12% higher compared to the non-treated fields. The possible ecosystem effects of the introduction of this GM strain was studied by DGGE of amplified 16S rDNA in a microcosm experiment. The introduced GM strain survived well in the rhizosphere. The DGGE banding profiles of samples treated with the modified strain closely resembled the profiles of untreated samples throughout the 40 days of the experiment, suggesting that there are no obvious effects on the bacterial community. Overall, the survival of the strain and the increase in crop yield indicate that this derivative of A. faecalis is a good candidate for commercial application, since its ecosystem effects seem very limited.
The impact and fate under field conditions of GM Rhizobium strains were investigated in a field trial with a model system comprising different GM Rhizobium leguminosarum v. viciae strains, marked with the lacZ gene and HgCb resistance genes (mer genes) inoculated in the rhizosphere of pea plants. Three modified strains were used:
These strains were monitored according to the reporter lacZ/mer system along with the soil metabolic activity plus nitrogen-transforming capacity.
The field experiments showed that all tested strains colonized the rhizosphere to the same extent; similar values were determined for the respiration rate and soil metabolic activity as well as for the nitrogen-transforming capacity of all tested strains. These results indicate that, although the presence of the plant had a considerable impact on carbon mineralization in soil, the impact of GM Rhizobium strains is indistinguishable from the impact of the wild-type strain, and also suggest that the impact of the plant on microbial activity is considerably greater than the impact of GM inoculants compared with wild-type strains.
In spite of the fact that the field trials with GM bio-fertilizers are limited, the initial results about their use are promising with respect to the improved performance in agricultural applications. GM bio-fertilizers have been introduced with an encouraging success in terms of the survival and activity of the inoculants, which is dependent on the environmental conditions. So far, non-target effects of GM bio-fertilizer strains that have been reported are small and insignificant compared to natural variations, such as differences between populations of different plant species.
However, our knowledge on the benefits, fate and effects of GM strains in the environment is still quite limited and partial.
Questions that have to be solved include: how and when (at what physiological state) bacteria survive best in soil; what their effect on the natural microflora is; how a mixed microbial community can be structured and optimized for use in agriculture. And last but not least, what the ecosystem effects of GM strains are, especially on non-target organisms.
The European Commission support for the production of this publication does not constitute endorsement of the contents which reflects the views only of the authors, and the Commission cannot be held responsi-ble for any use which may be made of the information contained therein.
