Review
Methods of studying soil microbial diversity
Jennifer L.Kirk a ,Lee A.Beaudette a,1,Miranda Hart b ,Peter Moutoglis c ,
John N.Klironomos b ,Hung Lee a ,Jack T.Trevors a,*
a
Department of Environmental Biology,University of Guelph,Ontario Agricultural College,Guelph,Ontario,Canada,N1G 2W1
b
Department of Botany,University of Guelph,Guelph,Ontario,Canada,N1G 2W1
c
PremierTech,Riviere-du-Loup,Que
´bec,Canada G5R 6C1Received 1December 2003;accepted 7April 2004
Available online 15June 2004
Abstract
Soil microorganisms,such as bacteria and fungi,play central roles in soil fertility and promoting plant health.This review examines and compares the various methods ud to study microbial diversity in soil.D 2004Elvier B.V .All rights rerved.
Keywords:Arbuscular mycorrhizal fungi (AMF);Bacteria;Biology;Diversity;DNA;Ecology;Fungi;Molecular ecology;Soil
1.Introduction
面积最大的咸水湖Soil bacteria and fungi play pivotal roles in various biogeochemical cycles (BGC)(Molin and Molin,1997;Trevors,1998b;Wall and Virginia,1999)and are responsible for the cycling of organic compounds.Soil microorganisms also influence above-ground ecosystems by contributing to plant nutrition (George et al.,1995;Timonen et al.,1996),plant health (Srivastava et al.,1996;Filion et al.,1999;Smith and Goodman,1999),soil structure (Wright and Upadhyaya,1998;Dodd et al.,2000)and soil fertility (Yao et al.,2000;O’Donnell et al.,2001).
Our knowledge of soil microbial diversity is limited in part by our inability to study soil microorganisms.Torsvik et al.(1990a,b)estimated that in 1g of soil there are 4000different bacterial ‘‘genomic units’’bad on DNA–DNA reassociation.It has also been estimated that about 5000bacterial species have been described (Pace,1997,1999).Approximately 1%of the soil bacterial population can be cultured by standard laboratory practices.It is not known if this 1%is reprentative of the bacterial population (Torsvik et al.,1998).An estimated 1,500,000species of fungi exist in the world (Giller et al.,1997).But unlike bacteria,many fungi cannot be cultured by current standard laboratory methods (Thorn,1997;van Elsas et al.,2000).Although molecular methods have been ud to study soil bacterial communities,very little rearch has been undertaken for soil fungi (van Elsas et al.,2000).
All organisms in the biosphere depend on microbial activity (Pace,1997).Soil microorganisms are vital for
0167-7012/$-e front matter D 2004Elvier B.V .All rights rerved.doi:10.1016/j.mimet.2004.04.006
泼妇骂街*Corresponding author.Tel.:+1-519-824-4120x53367;fax:+519-837-0442.
E-mail address:jtrevors@uoguelph.ca (J.T.Trevors).1
Current address:Environment Canada,Wastewater Technol-ogy Center,Burlington,Ontario,Canada L7R.
/locate/jmicmeth
Journal of Microbiological Methods 58(2004)169–
188
the continuing cycling of nutrients and for driving above-ground ecosystems(van der Heijden et al., 1998;Cairney,2000;Klironomos et al.,2000;Ovreas, 2000).While many anthropogenic activities,such as city development,agriculture,u of pesticides and pollution can potentially affect soil microbial diversity, it is unknown how changes in microbial diversity can influence below-ground and above-ground ecosys-tems.Before we can address how changes in microbial community structure influences ecosystem functions, there is the need for reliable and accurate mechanisms of studying soil microorganisms.This article will review the current methods,and their advantages and disadvantages,for studying microbial diversity in soil.
2.General limitations in studying microbial diversity
There are problems associated with studying bacte-rial and fungal diversity in soil.The ari not only from methodological limitations,but also from a lack of taxonomic knowledge.It is difficult to study the diver-sity of a group of microorganisms when it is not under-stood how to categorize or identify the species prent.
2.1.Spatial heterogeneity女友生日祝福
When studying microbial diversity,replicates of1 to5g of soil are often ud to measure diversity and then conclusions about the community are made.There are numerous problems with this ap-proach.One is the innate heterogeneity of soil and thus of spatial distribution of the microorganisms (Trevors,1998b).Franklin and Mills(2003)ud multiple spatial scales,with sampling intervals rang-ing from 2.5cm to11m,to study the spatial heterogeneity of soil microbial communities in an agricultural soil.They reported that microbial com-munities may have veral nested levels of organi-zation,and that they could be dependent on different soil properties or groups of properties. Microbial communities exist on such a small scale, that possibly1to5g of soil could bias results and favour detection of dominant populations(Grund-mann and Gourbiere,1999).Grundmann and Gour-biere(1999)suggested that sampling of soil has to be done on a smaller scale with more samples to asss the diversity of microorganisms in the micro-habitats in soil.Another problem with this approa
ch is that soil is heterogeneous,containing many microhabitats that are suitable for microbial growth. As a result,bacteria are highly aggregated in soil existing in clumps or‘‘hot spots’’.Plants also influence the spatial distribution of soil bacteria (Wall and Virginia,1999)and fungi,as shown by an approximately two-fold increa in bacterial numbers in the rhizosphere over bulk soil(Curl and Truelove,1986).Arbuscular mycorrhizal fungi (AMF)require a plant host to survive.Therefore, their distribution in soil is also clustered around plant species.
Much of what is known about soil fungal diversity has resulted from field studies of sporocarps or by morphological descriptions of below-ground fungal structures,especially for mycorrhizal fungi(Horton, 2002).Often,fungal diversity studies using data from above-ground fruiting bodies do not correspond to tho using below-ground fungal structures(Horton, 2002).This discrepancy can be caud by the sporadic production of sporocarps as well as the lack of infor-mation about below-ground structures.
Very little is known about spatial and temporal variability of microorganisms in soil(Sanders and Fitter,1992;Johnson et al.,1997;Trevors,1998b).If rearchers sample soil in a traditional random fash-ion,microbial diversity and population size could be grossly underestimated resulting in high variability between replicates and low statistical power(Kliro-nomos et al.,1999).Klironomos et al.(1999)sug-gested using a combination of geostatistical analys to describe spatial distribution of su
bsurface micro-organisms together with power analys to asss the required sample size.This approach should reduce variability in sampling and provide a more repren-tative sampling regime.
2.2.Inability to culture soil microorganisms
The immen phenotypic and genetic diversity found in soil bacterial and fungal communities makes it one of the most difficult communities to study(Ovreas et al.,1998).It has been suggested that at least99%of bacteria obrved under a microscope are not cultured by common laboratory techniques(Borneman et al.,1996;Giller et al.,
J.L.Kirk et al./Journal of Microbiological Methods58(2004)169–188 170
1997;Pace,1997;Torsvik et al.,1998;Trevors, 1998b).It is possible this1%of culturable bacteria is reprentative of the entire population and that the other99%are simply in a physiological state that eludes our ability to culture them(Rondon et al., 1999).However,it is also likely that the99%are phenotypically and genetically different from the1% and only the minority of the population is repre-nted(Rondon et al.,1999,2000).Many fungal species also elude culturing in the laboratory(van Elsas et al.,2000).To overcome problems associated with non-culturable bacteria and fungi,various meth-ods have been developed to identify and study the microorganisms including fatty acid analy
sis and numerous DNA-and RNA-bad methods.
2.3.Limitations of molecular-bad methods
Molecular techniques bad on PCR have been ud to overcome the limitations of culture-bad methods;however,they are not without their own limitations.
Lysis efficiency of cells or fungal structures varies between and within microbial groups(Prosr, 2002).Bacteria exist in or on the surface of soil aggregates;therefore,the ability to parate the cells from soil components is vital for studying biodiversity(Trevors,1998a).If the method of cell extraction ud is too gentle,Gram-negative,but not Gram-positive bacterial cells would be lyd.If the method is too harsh,both Gram-negative and Gram-positive cells may be lyd but their DNA may become sheared(Wintzingerode et al.,1997).Lysis efficiency also varies for different fungal cells. Spores will ly differently than mycelia and mycelia of different ages will also have different lysing efficiency(Prosr,2002).The variation in the ability to break open cells or fungal structures can lead to bias in molecular-bad diversity studies.
The method of DNA or RNA extraction ud can also bias diversity studies.Harsh extraction methods, such as bead beating,can shear the nucleic acids, leading to problems in subquent PCR
detection (Wintzingerode et al.,1997).Different methods of nucleic acid extractions will result in different yields of product(Wintzingerode et al.,1997).With environ-mental samples,it is necessary to remove inhibitory substances such as humic acids,which can be co-extracted and interfere with subquent PCR analysis. Subquent purification steps can lead to loss of DNA or RNA,again potentially biasing molecular diversity analysis.
Differential amplification of target genes can also bias PCR-bad diversity studies.Typically,16S rRNA,18S rRNA or ITS regions are targeted by primers for diversity studies becau the genes/frag-ments are prent in all organisms,they have well defined regions for taxonomic classification that are not subject to horizontal transfer and have quence databas available to rearchers.Wintzingerode et al. (1997)discusd some issues surrounding differential PCR amplification including different affinities of primers to templates,different copy numbers of target genes,hybridization efficiency and primer specificity. In addition,quences with lower G+C content are thought to parate more efficiently in the denaturing step of PCR and,therefore,could be preferentially amplified(Wintzingerode et al.,1997).
The above discuss a few limitations of molec-ular-bad methods,which can influence the analysis and interpretation of molecular-bad microbial com-munity analysis.Molecular-bad meth
ods provide valuable information about the microbial community as oppod to only culture-bad techniques.
2.4.Taxonomic ambiguity of microbes
Another problem associated with measuring mi-crobial diversity in soil is the problem of defining microbial species(Torsvik et al.,1998;Trevors, 1998b;Ovreas,2000).There is no official definition of a bacterial(Colwell et al.,1995)or AMF species. Moreover,Hey(2001)listed over24definitions of species,all of which were different.The traditional species definition was bad on higher plants and animals and does not readily apply to prokaryotes (Godfray and Lawton,2001)or axual organisms. The genetic plasticity of bacteria,allowing DNA transfer through plasmids,bacteriophages and trans-posons,complicates the concept of bacterial species.
Fungal taxonomy has similar problems in identi-fying vegetative structures.Most of the current tax-onomy is bad on fungal xual states,for instance mushrooms and truffles,and problems exist when trying to identify the below-ground vegetative struc-tures(Horton,2002).Molecular methods such as
J.L.Kirk et al./Journal of Microbiological Methods58(2004)169–188171
restriction analysis of the internal transcribed spacer (ITS)region,18S rDNA and restriction fragment length polymorphism(RFLP)analysis has been ud to identify fungi;however,the databas are still not sufficiently developed to prevent many‘‘unknowns’’in a community analysis.
AMF taxonomy is also transient.Traditionally, AMF were identified using spore morphology and differential staining.This technique made it difficult to identify morphologically similar spores,to deter-mine phylogenetic relatedness and to accurately clas-sify AMF.Recently,Redecker(2000)and Redecker et al.(1997,1999,2000)have put considerable effort into determining the phylogenetic relationships of AMF and this led to the identification of two new families. As more rearch is conducted,the specificity and nsitivity of molecular techniques to identify AMF species increa and so does the ability to study the microorganisms.
Another major limitation of the u of molecular techniques with AMF is our lack of understanding of genetic polymorphism in AMF.One single spore can contain considerable variation in ITS and5.8S rDNA quences(Redecker et al.,1999).Therefore,the single-quence,single-species hypothesis may not fit with AMF and diversity could be overestimated (Dodd et al.,2000;Schubler et al.,2001).
3.Methods of studying microbial diversity in soil
Species diversity consists of species richness,the total number of species prent,species evenness,and the distribution of species(Trevors,1998b;Ovreas, 2000).Methods to measure microbial diversity in soil can be categorized into two groups:biochemical-bad techniques(Table1)and molecular-bad techniques (Table2).Typically,diversity studies include the rela-tive diversities of communities across a gradient of stress,disturbance or other biotic or abiotic difference (Hughes et al.,2001).It is difficult with current techniques to study true diversity since we do not know what is prent and we have no way of determining the accuracy of our extraction or detection methods.Often rearchers will attempt to reduce the information gathered by diversity studies into discrete,numerical measurements such as diversity indices(Atlas and Bartha,1993).4.Biochemical-bad techniques to study microbial diversity
国际象棋象的走法Most methods described below can be ud for either bacteria or fungi,although some are specific to one or the other.
4.1.Plate counts
Traditionally,diversity was assd using lec-tive plating and direct viable counts.The methods are fast,inexpensive and can provide information on the active,heterotrophic component of the popul
a-tion.Limitations include the difficulty in dislodging bacteria or spores from soil particles or biofilms, growth medium lections(Tabacchioni et al.,2000), growth conditions(temperature,pH,light),the in-ability to culture a large number of bacterial and fungal species with current techniques and the po-tential for colony–colony inhibition or of colony spreading(Trevors,1998b).In addition,plate growth favours tho microorganisms with fast growth rates and tho fungi that produce large numbers of spores (Dix and Webster,1995).All of the limitations can influence the apparent diversity of the microbial community.
4.2.Sole carbon source utilization patterns/commu-nity level physiological profiling for measuring microbial diversity
Garland and Mills(1991)developed a technique using a commercially available96-well microtitre plate to asss the potential functional diversity of the bacterial population through sole source carbon utilization(SSCU)patterns.Gram-negative(GN)and gram-positive(GP)plates are available from Biolog (Hayward,CA,USA,)and each contains95different carbon sources and one control well without a substrate.GN and GP plates were developed originally for characterization of clinical bacterial isolates and not for community analysis. Subquently,Biolog introduced an Eco-plate(Choi and Dobbs,1999)containing3replicates of31 different environmentally relevant carbon sou
rces and one control well per replicate.Carbon sources not found in GN plates include D-cellobio,D-xylo,D-malic acid,L-arginine,2-hydroxybenzoic
J.L.Kirk et al./Journal of Microbiological Methods58(2004)169–188 172
acid and4-hydroxybenzoic acid.For a list of all31 carbon sources,plea refer to Choi and Dobbs (1999).Alternatively,rearchers can u plates containing the growth medium and a tetrazolium salt from Biolog,and add site-specific carbon sources to analyze their samples(Campbell et al.,1997;Becker and Stottmeister,1998).The tetrazolium salt changes colour as the substrate is metabolized.Since many fungal species are not capable of reducing the tetrazolium salt,Biolog developed fungal specific plates SFN2and SFP2,which have the same sub-strates as GN and GP plates but without the tetra-zolium salt(Clasn et al.,2003).Inoculated populations are monitored over time for their ability to utilize substrates and the speed at which the substrates are utilized.Multivariate analysis is ap-plied to the data and relative differences between soil functional diversity can be assd.
This method has been ud successfully to asss potential metabolic diversity of microbial communities in contaminated sites(Derry et al., 1998;Konopka et al.,1998),plant rhizospheres (Ellis e
t al.,1995;Garland,1996a;Grayston and Campbell,1996;Grayston et al.,1998),arctic soils (Derry et al.,1999),soil treated with herbicides(el Fantroussi et al.,1999)or inoculation of micro-organisms(Bej et al.,1991).For example,Roling et al.(2000)ud the Biolog system together with DGGE to study the anaerobic microbial community in an aquifer.They found that both the anaerobic community level physiological profiling(CLPP)and denaturing gradient gel electrophoresis(DGGE) were able to parate microbial communities from the polluted aquifer below a landfill site from tho of aquifers located up or downstream of the land-fill.Derry et al.(1999)ud GN Biolog plates to asss the functional diversity of microorganisms in three different Arctic soils incubated at different temperatures.They found significant differences in Shannon indices,substrate utilization richness and evenness at incubation temperatures that Arctic soil would be expod to.el Fantroussi et al.(1999) ud Biolog plates in conjunction with DGGE to asss the impact of three different phenylurea
Table1
Advantages and disadvantages of biochemical-bad methods to study soil microbial diversity
Method Advantages Disadvantages Selected references
Plate counts Fast Unculturable microorganisms Tabacchioni et al.(2000),
Inexpensive not detected
Bias towards fast growing
individuals
Trevors(1998b)
Bias towards fungal species
that produce large quantities of如何练肩
spores
Community level physiological Fast Only reprents culturable Clasn et al.(2003),Garland profiling(CLPP)Highly reproducible fraction of community(1996a),Garland and Mills
Relatively inexpensive Favours fast growing(1991)
Differentiate between microbial communities Generates large amount of data organisms
Only reprents tho organisms capable of utilizing
Option of using bacterial, fungal plates or site specific carbon sources(Biolog)available carbon sources Potential metabolic diversity, not in situ diversity Sensitive to inoculum density
Fatty acid methyl ester analysis (FAME)
No culturing of
解约合同
microorganisms,direct
extraction from soil
If using fungal spores,a lot of
material is needed
Can be influenced by external
Graham et al.(1995),Siciliano
and Germida(1998),Zelles
(1999)
Follow specific organisms or真酒
communities
factors
Possibility results can be
confounded by other
microorganisms
环卫所J.L.Kirk et al./Journal of Microbiological Methods58(2004)169–188173
