<Articletype>Chapter

1

<doi>

<en>

<articletype>chapter

<producttype>review series

<subject>life sciences

<header>

<title>11Gene Detection, Expression and Related Enzyme Activity in Soil

<author>M. Krsek, W.H.Gaze, N.Z. Morris, E.M.H. Wellington

<authorinfo>M. Krsek, W.H.Gaze, N.Z. Morris, E.M.H. Wellington (e-mail: )

Department of Biological Sciences, University of Warwick, Coventry CV4 7AL, UK</authorinfo>

</header>

<heading1>11.1Introduction

<p1a>Molecular techniques have been used to detect functional genes present within soil bacterial communities, to establish the potential for key microbial functions in soil such as plant growth promotion and disease suppression, nitrification, nitrogen fixation, denitrification and decomposition. The use of reverse transcription polymerase chain reaction (RT-PCR) has enabled estimation of active functional groups, which, with other techniques such as real-time PCR, has enabled major developments in our understanding of how soil treatments and the environment affect soil microorganisms. There have been attempts to unravel the interrelationship between functional and taxonomic diversity and activity (ammonia-oxidisers, Okano et al. 2004; 2,4-dichlorophenoxyacetic acid degraders, Kamagata et al. 1997; streptomycin-producers, Huddleston et al. 1997; chitinolytic bacteria, Metcalfe et al. 2003) but the extent of functional redundancy in soil still needs to be explored for key processes involved in decomposition and mineral recycling. It will be a considerable challenge to relate function to community structure given recent findings concerning the need for representative sampling to estimate habitat diversity (Venter et al. 2004) and the predicted extensive diversity within soil microbial populations (Torsvik and Ovreas 2002). The relationship between microbial diversity and soil functions has been reviewed (Nannipieri et al. 2003) with consideration of the relationship between functioning of the soil in terms of decomposition, resilience and general activity as measured by soil respiration. This holistic approach has not provided any clear relationship between microbial diversity and activity in soil. Soil is a heterogeneous habitat providing a very large surface area for colonisation; the extent of the soil surface colonised by bacteria appears to be very small (<10%; Clewlow et al. 1990). This adds to the sampling problem where representative sampling may require intensive analysis of large sample sets in order to estimate prokaryotic diversity. Thus a targeted approach may well be required to focus on specific aspects of microbial activity in soil, as discussed by Curtis and Sloan (2004), who emphasised the problems of sampling microbial population diversity. A range of functional gene targets are available for detection of key microbial activities in soil in addition to taxonomic signature sequences linking structure to function such as 16S rRNA targets for nitrifiers. Phylogenetic analysis of functional gene sequences can give an indication of the host background if there is conservation within the target sequences. However, many activities of soil bacteria may be adaptations to the soil environment and subject to horizontal gene transfer (HGT) thus obscuring phylogenies. Advances in metagenomics of the soil microbial population may assist in resolving the link between structure and function by analysis of large DNA fragments (HMW DNA) linking highly conserved genes, such as 16S rRNA, with adaptive features, e.g. extracellular enzyme secretion or energy generation pathways.

<p1>Soil samples can be screened for the presence of specific genes allowing the analysis of many environmental samples, although the level of sensitivity will vary with soil type. Methods have been developed to increase the number of genes targeted such as the use of DNA microarrays as 'genosensors' for the detection of specific microorganisms in environmental samples (for more details, see Chap.9). The major advantage of this technique is that many sequences can be placed on one chip and the sample can be simultaneously analysed for the presence of many diverse target groups.

<p1>The highly complex trophic interactions that occur amongst microorganisms present a challenge to unravel the relationship between metabolic and genotypic diversity (Zak et al. 1994). Microorganisms do play a key role in the environment, playing a major part in the decomposition of polymers and degradation of environmental contaminants (Lamar and Dietrich 1990; Aelion and Bradley 1991). One of the better-understood processes is the impact of bacteria on nitrogen cycling in soil and the role of key genotypes in nitrification (Prosser and Embley 2002).

<p1>Methods for studying specific activities, such as an enzyme activity or the ability to degrade a toxic compound, give little information on the key active groups. Conversely, selective gene analysis of presence and expression of specific genes can only indicate identity of the host population, but can never confirm it unless metagenomics is used to link taxonomic and functional gene targets. However, the analysis of mRNA extracted from soil can provide useful data on activity of certain genotypes. Proteomics also promise to contribute to our understanding in this field (for more details, see Chaps.4 and 5). Despite the complexity of the soil microbiota there has been significant progress in the analysis of functional genes in relation to treatments and traditional measurements of soil sample activity. In addition, surveys of target gene prevalence and diversity have provided indications of impacts of agricultural practice on soil microbial communities such as the association between ergotrophic use of antibiotics and the prevalence of antibiotic resistance genes in manured soils and farm microflora (Seveno et al. 2002). Molecular detection of genes involved in degradation of pollutants in soil has provided data on prevalence and diversity of key enzymes involved, such as the in situ detection of mmoX genes coding for part of the soluble methane monoxygenase enzyme complex involved in the oxidation of trichloroethylene (Shigematsu et al. 1999).

<p1>This review aims to summarise selected recent developments in attempts to define microbial populations responsible for a defined activity, enzyme or process in soil by targeting specific genes. The direct molecular analysis of gene presence in soil via analysis of total community DNA (TC-DNA) and measurement of activity by targeting mRNA can be illustrated by a number of case studies. Examples have been selected on the basis of relevance to soil processes and attempts to evaluate the impacts of agricultural practice, pollution and climate on soil. Selected studies have attempted to quantify gene copies, which can be done by quantitative PCR (QPCR) by measuring the concentration of PCR product using densitometry, competitive PCR, and quantitative real-time PCR. Quantity of targetsequences in TC-DNA extracts, whatever method of estimation, must be compared to a calibration curve, either as enumerated cells in an extract of chromosomal or plasmid origin or by addition of known number of targets to a TC-DNA extract that does not contain the target gene of interest. Few studies have attempted to determine if QPCR with chromosomal or plasmid DNA is directly equivalent to the same reaction using TC-DNA. It is preferable to use seeded TC-DNA for calibration. Quantifying molecular targets in TC-DNA will only serve to estimate the potential activity for a given specific population, even when derived from mRNA. The latter is more likely to indicate actual activity in a soil extract but transcripts can give rise to inactive enzymes which require further processing. Thus it is important that any gene being used as a target to identify a potential functionally active population will have been well studied in vitro and molecular mechanisms of regulation understood. All molecular data should be used in conjunction with studies of enzyme activity and measurements of general microbial populations in a soil such as ATP, AEC, biomass and cell counts.

<heading1>11.2Molecular Detection of Functional Genes in Soil

<heading2>11.2.1Introduction

<p1a>A wide range of genes have served as targets for studies of microbial activity in soil, but some trends emerge for the study of a range of microbiallydriven processes. These include antagonism and biological control and the ability of the soil microbiota to respond to various impacts such as pollution and agricultural practice. The latter involves genes expressed in bioremediation studies, heavy metal and antibiotic resistance. Further important microbial activities relate directly to soil fertility and the cycling of key nutrients such as the genes implicated in the carbon and nitrogen cycles where specific activities can be attributed to key genes such as those involved in nitrification or biodegradation of recalcitrant polymers. Detection of functional gene signatures of pathogenic microorganisms in the soil environment represents another important issue, the significance of which has been acknowledged during recent years.

<heading3>11.2.1.1Extraction of Microbial Biomass from Soil

<p1a>Detection of genes in TC-DNA extracted from soil poses technical problems associated with the difficulty in extracting DNA of sufficient quality for use with PCR and other methods. Co-isolation of substances which inhibit PCR amplification occurs, particularly humic acids, and a variety of extraction protocols and purification methods have been developed in an effort to provide PCR quality TC-DNA (Frostegård et al. 1999; Krsek and Wellington 1999; Griffiths et al. 2000; Courtois et al. 2001; Sessitsch et al. 2002). Nucleic acid extraction from soil is discussed in detail in Chapter3.

<p1>A range of methods is available to concentrate soil microbes, and remove the soil particles, which may inhibit down-stream applications such as PCR. Where microbial biomass is extracted from large volumes of soil, sensitivity is also increased. Microbial cells extracted directly from soil without cultivation provide an opportunity to examine the physiological state of cells and identify individual genes using fluorescence in situ hybridisation (FISH) or using fluorescence activated cell sorting (FACS), where specific genes are labelled with fluorescent oligonucleotide probes.

<p1>Extraction of microbial cells requires physical or chemical dispersion of soil to separate cells from the soil matrix. Cells adhere to soil particles in many ways including electrostatic and hydrogen bonding. Cells are also physically entrapped within the soil aggregates (Stotzky 1986; Robert and Chenu 1992). The type and strength of microbial attachment will depend on the soil type and the identity and activity of the microbial community. Once the soil matrix is dispersed a further concentration step is used to separate the biomass fraction from soil debris.

<p1>Shaking, blending using a Waring blender or stomacher and ultrasonication are used to physically disperse soil samples. Gram-negative organisms and those sensitive to lysis using aggressive methods such as homogenisation using a Waring blender or ultrasonication may be better suited to dispersal by stomaching or shaking. Salmonella spp. were efficiently recovered from soil using a stomacher for dispersion combined with a chemical dispersion and centrifugation (Turpin et al. 1993). It has been demonstrated that ultrasonication may cause more lysis than other methods and this has been avoided by using mild ultrasonication (Hopkins and O'Donnell 1992).

<p1>Chemical dispersion using the ion-exchange resin Chelex 100 (reduces the electrostatic attraction between cells and soil particles) has been widely used for extraction of microbial biomass although Tris buffer and sodium hexametaphosphate have also been used (Niepold et al. 1979; Herron and Wellington 1990).

<p1>Buoyant density has been used to separate soil particles and microbial cells; this relies on the assumption that most cells are not aggregated and are detached from soil particles. The soil type is important in the success of this method; soils with a high organic content, for example, will have less dense particles and may be impossible to separate from the cells. A number of gradient media have been used but Percoll has been used extensively and allowed the recovery of a clean bacterial fraction for biochemical studies (Bakken 1985). A single step method for purification may be possible with the density material Nycodenz (Bakken and Lindahl 1995). A recent protocol (Berry et al. 2003) blended soil with disruption buffer and then soil homogenate was transferred to an ultracentrifuge tube and Nycodenz pipetted to form a layer below the homogenate. After centrifugation a faint whitish band containing bacterial cells appeared at the interface between the Nycodenz and the aqueous layer. Nycodenz extraction can be used to separate cell fractions of sufficient purity to allow direct flow cytometric analysis (Unge et al. 1999).

<p1>Immunomagnetic capture (IMC) has been used to recover cells from environmental samples (Morgan et al.1991; Fluit et al. 1993). Microscopic magnetic beads (such as those manufactured by Dynal) coated with monoclonal or polyclonal antibodies are used to extract target bacteria. Their major use to date has been in the detection of medically important species in food and water, but methodological modifications for retrieval of salmonellae from soil are currently in development by the authors. The feasibility of this approach is supported by studies conducted by Schmidt et al. (1968) who used fluorescent-antibody techniques to study free living soil bacteria such as rhizobia. Autofluorescence was not a problem but non-specific absorption restricted observation of rhizobia to microscopic fields free of soil particles. Frederickson et al. (1988) used fluorescent antibodies to enumerate Tn5 mutant bacteria in soil; the fluorescent antibody method tended to overestimate the viable population. The IMC approach provides a useful alternative to the more traditional isolation procedures and does not require the cultivation of the target cells. In addition a highly purified cell extract is produced as the captured magnetic beads can be washed and all soil particles removed. The majority of capture techniques have used a direct IMC approach where either monoclonal antibodies or specific polyclonal antibodies were used to coat magnetic beads. Both direct (Wipat et al. 1994) and indirect methods (Mullins et al.1995)have been applied to soil extracts; the direct method may be preferable if monoclonal antibodies are used as the polyclonal purified IgG fraction may contain <30% specific antibodies. IMC techniques have recovered very pure fractions of specific cells from dispersed soil extracts but the recovery efficiency can vary from 4--90% depending on the type of soil and technique used. Porter et al.(1998) reported isolation of E. coli O157 from soil using commercially anti-Escherichia coli O157-labelled magnetic beads.

<p1>Magnetic capture systems can also be used to obtain specific genes directly from TC-DNA. Jacobsen (1995) used a magnetic capture-hybridisation PCR technique (MCH-PCR) to eliminate the inhibitory effect of humic acids and other contaminants in PCRs targeting specific soil DNA. A single-stranded DNA probe, which was complementary to an internal part of the target gene, was used to coat magnetic beads. After hybridisation in a suspension of soil DNA, magnetic extraction of the beads separated the hybrid DNA from all other soil DNA and humic acids.

<heading2>11.2.2Antibiotic Biosynthesis Genes

<p1a>The production of antibiotics by bacteria, a vast majority of which were isolated from soil, has prompted much research and speculation into the possible production and role of antimicrobial secondary metabolites in soil (Anukool et al. 2004). Although antibiotics have not been isolated from natural untreated soils, production has been detected in non-sterile, amended and sterile unamended and amended soils when inoculated with producing strains of streptomycetes and fungi (Williams 1982; Wellington et al. 1993). There is evidence to support the occurrence of antibiosis in soil, e.g. survival of a Salmonella sp. in soil was inhibited by the presence of streptomycin-producing Streptomyces bikiniensis (Turpin et al. 1992). Evidence for antibiotic production in soil can be achieved by either detection of the active product extracted from soil or by detecting activity of specific promoters associated with antibiotic production using reporter genes such as lux. These approaches are also discussed in Section11.3.2.2.

<p1>Degenerate PCR primers designed to detect antibiotic biosynthesis genes have been used to study both diversity and distribution of potentially bioactive bacteria in soil. Key enzymes involved in the biosynthesis of certain chemical classes of antibiotic such as the type II polyketide antibiotics have been targeted.

<p1>Despite their apparent structural diversity, polyketides share a common mechanism of biosynthesis. The chemical synthesis of polyketides is centred around the reactive groups of the ketone and its adjacent -methylene carbon (Kramer and Khosla 1996). Therefore, three opportunities exist for structural diversity, the nature of the starter unit, the extender unit and the stereochemistry resulting from their condensation. The carbon backbone of a polyketide results from sequential condensation of short fatty acids, such as acetate, propionate, or butyrate, in a manner resembling fatty acid biosynthesis but catalysed by polyketide synthases (Hopwood 1997). The phylogenetic relationship of amino acid sequences from type II PKS, KS and KS genes was examined by Hopwood (1997) and were found to cluster separately suggesting different functions of these genes. Within these individual clusters the genes encoding polyketide spore pigments were found to be separate from those encoding the antibiotic (and other bioactive) polyketides. This suggests a separate line of descent for these two groups.

Fig. 1

Table 1

<p1>A detailed study of polyketide synthase (type II ketosynthase KS genes) diversity in soil was undertaken by sequence analysis of PCR products derived from use of consensus primers for a conserved region in type II KS genes covering both KS and KS (Morris 2000). The majority of sequences recovered did not match to known genes but were recovered in distinct clades, as shown in Fig.11.1. The list of known genes given in Table11.1 was used to design the primers and compare cloned sequences recovered from TC-DNA. This study showed that bacteria with the potential to produce polyketide antibiotic-like metabolites and also pigments were widespread in soil and there was an extensive metabolic diversity not yet explored.

<p1>Seow et al.(1997) used PCR primers to amplify KS genes from total community DNA extracted from soil. These products were cloned and two clones sequenced. Both sequences were found to cluster within known type II PKS gene clusters but again did not match any existing genes. These sequences were used to complement mutants for design of novel polyketide antibiotics.

<heading2>11.2.3Detection of Antibiotic and Heavy Metal Resistance Genes

<p1a>There have been relatively few studies focusing on detection of antibiotic or heavy metal resistance genes in situ, in environments such as soil. Most research has concentrated on detection of resistance genes in clinical and environmental isolates. Resistance genes have been detected in soil using conventional DNA extraction and PCR methods. Chee-Sanford et al. (2001) screened for eight tet genes in groundwater associated with swine production facilities. Tetracycline resistance genes were found as far as 250m downstream from waste lagoons, highlighting the danger posed by use of antibiotics in agriculture and the risk of contamination of drinking water with antibiotic resistant bacteria. In a different study a detection limit of 102--103 copies of the tet(M) gene per gram was achieved using a nested PCR method with TC-DNA (Agersø et al.2004). The gene was detected in farmland soil previously amended with pig slurry containing resistant bacteria; the number of positive samples from farmland soils 1year after manure treatment was significantly higher than in samples of garden soil not treated with manure.