tin

ate

ilia

wan, 5

d form

onent

accessing otherwise unavailable (organic) forms of soil recovered spores are often unidentifiable and require trap

Soil Biology & BiochemistryE-mail address: germida@sask.usask.ca (J.J. Germida).low-input agriculture. Molecular techniques are used to investigate fungal community composition in uncultivated, disturbed, or

contaminated soils, but this approach to community analysis of AMF in agricultural soils has not been reported. In this study, a polymerase

chain reaction and denaturing gradient gel electrophoresis (PCR-DGGE) procedure for the detection of fungal 18S ribosomal RNA gene was

developed with reference cultures of seven isolates (representing five AMF species). These reference cultures were chosen because isolates

of their species were putatively identified in a previous survey of farm field soils in the province of Saskatchewan, Canada. A reference PCR-

DGGE profile was generated using DNA extracted and amplified from the spores of these cultures. The effectiveness of the procedure was

tested by its application to soil samples from 38 farms. Prominent bands from the PCR-DGGE profiles of these samples were excised for

sequence analysis. The total number of species recovered was low in comparison to other AMF community surveys of temperate climate

locations. The majority of the sequences recovered were Glomus species. Scutellospora calospora, a previously undetected AM fungus in

Saskatchewan was found. Though not without its drawbacks, this approach to community composition analysis of AMF was faster than

conventional trap cultivation methods.

q 2005 Elsevier Ltd. All rights reserved.

Keywords: 18S rRNA gene; AMF; Arbuscular mycorrhizal fungi; DGGE; Operon heterogeneity; PCR

1. Introduction

Arbuscular mycorrhizal (AM) fungi (AMF) are a critical

component in agricultural systems because of their ability to

increase plant growth (Smith and Read, 1997), reproductive

capacity (Lu and Koide, 1994), water stress resistance

(Gupta and Kumar, 2000), and plant health through

antagonistic and competitive effects on pests and pathogens

(Gange and West, 1994). The main benefit to the host plant

in the mycorrhizal symbiosis is the enhanced uptake of

immobile soil nutrients, in particular phosphorus (Jakobsen,

1999). Arbuscular mycorrhizal associations increase nitro-

gen accumulation in plant tissues as a result of the hyphae

fungi also interact with other soil organisms involved in

important nutrient cycles. For example, biological nitrogen

fixation by Rhizobium in legume hosts can be enhanced

through co-infection with AMF (Xavier and Germida,

2002). Such ecological roles are of special importance in

low-input farm management systems because these systems

rely on natural nutrient cycles to provide the nutrients

required for plant production.

Evidence of the ecological importance of AMF is

abundant, but understanding of the distinct roles of

individual fungal species is limited. Spore morphology

and spore enumeration are the traditional methods for

taxonomic identification and AMF diversity studies. FieldA PCR-DGGE method for detec

in cultiv

W.K. Ma, S.D. Sic

Department of Soil Science, University of Saskatche

Received 13 October 2003; received in revise

Abstract

Arbuscular mycorrhizal (AM) fungi (AMF) are important compg arbuscular mycorrhizal fungi

d soils

no, J.J. Germida*

1 Campus Drive, Saskatoon, SK, Canada S7N 5A8

12 January 2005; accepted 27 January 2005

s of agro-ecosystems and are especially significant for productive

37 (2005) 15891597

www.elsevier.com/locate/soilbiogathered by this approach is incomplete because sporulation

is dependent on the species, host, seasonality, growth

conditions, and other environmental factors (Oehl et al.,

2003). Molecular techniques that assess the AMF diversity

directly present in soil avoid many of the challenges0038-0717/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.

doi:10.1016/j.soilbio.2005.01.020

* Corresponding author. Tel.: C1 306 966 6836; fax: C1 306 966 6881.nitrogen (Ibijbijen et al., 1996). Arbuscular mycrorrhizal cultivation on host plants to produce identifiable spores

(Bever et al., 2001). However, fungal diversity information

applied to soil samples from a survey of 38 farm field soils. collected by wet sieving and sucrose density centrifugation

Thirty-eight soil samples from organic farm fields were

collected across Saskatchewan, Canada, during a weed

survey conducted in May, 2002. A 5.0 cm diameter

hydraulic soil probe was used to collect the samples. In

each field 16 soil cores were taken to a 45 cm depth. The soil

cores were divided into 15 cm depths and bulked to form

three composite samples of each field. Where the field

appeared level and uniform, the sampling was done

following a W-pattern (Thomas, 1985). Irregular fields

were sampled avoiding irregularities like foot or shoulder

regions, roads or paths and ditches, power lines, oil wells,

saline areas, etc. Collecting a representative sample was

considered to be more important than following a rigid

collection regime. The 015 cm sample from each site was

used for DNA extraction.

2.3. DNA extraction from spore and soil samples

The extraction method of Griffiths et al. (2000) was the

original protocol used for spore and soil DNA extraction,

but, in our hands, it was inconsistent. The final, optimized

DNA extraction method used on spores included elements

from Griffiths et al. (2000) and Kowalchuk et al. (2002).

SA101 Glomus luteum INVAM

BiocTo our knowledge, this is the first use of PCR-DGGE to

evaluate AMF community composition in agricultural soils.

2. Materials and methods

2.1. Reference AMF species

Seven reference AMF isolates (representing five species)

were used for the development of the procedures (Table 1).

They were selected because isolates of their species were

putatively identified in a previous survey of farm field soils

in Saskatchewan (Talukdar and Germida, 1993). Five of the

isolates were procured from the International Culture

Collection of Arbuscular and VesicularArbuscular Mycor-

rhizal Fungi (INVAM, http://invam.caf.wvu.edu/). They

were harvested by workers at INVAM approximately 2

weeks prior to reception. Approximately 150 g of each

culture (containing soil, infected roots, and spores) was

received and kept refrigerated at 4 8C. Two isolates wereregenerated in pot cultures at the University of Saskatch-

ewans Soil Microbiology Laboratory. Briefly, 1.5 kg of

each culture from 1998 (stored at 4 8C) was planted withcorn (Zea mays var. Golden Bantam) seeds that were surface

sterilized for 3 min in 10% (v/v) Javexw bleach and rinsed in

autoclave sterilized distilled water five times (Jarstfer and

Sylvia, 2002). Corn was planted in 2 kg pots on June 10,

2002 and harvested on September 10, 2002. After harvest,

plants were placed in plastic bags and stored at 4 8C.associated with spore production and cultivation and could

potentially provide information on the vegetative/active

phase of the fungal community (Kowalchuk et al., 2002).

Muyzer et al. (1993) was first to use PCR-DGGE to

profile microbial communities. The first use of this

technique for fungal community analysis was by Kowalchuk

et al. (1997). Since then, PCR-DGGE has proven to be a

powerful technique for the culture-independent detection

and characterization of fungal populations in plant

material and soil (Kowalchuk et al., 2002; Smit et al.,

1999; Vainio and Hantula, 2000). PCR-DGGE was demon-

strated by Smit et al. (1999) to be complimentary to cloning

strategies for fungal community studies by tentatively

identifying cloned 18S rDNA fragments by comparison to

community DGGE banding patterns. Vainio and Hantula

(2000) showed DGGE detected more fungal species from

environmental samples than culturing techniques. Kowal-

chuk et al. (2002) noted discrepancies observed between the

AMF-like groups detected in spore populations versus direct

18S rDNA analysis of root material by DGGE, corroborating

previous suggestions that spore inspection alone may poorly

represent actual AMF population structure.

The objective of this study was to develop a PCR-DGGE

procedure to detect AMF in cultivated soils of Saskatch-

ewan, Canada. To prove its efficacy, the technique was

W.K. Ma et al. / Soil Biology &1590All reference cultures were separated into soil (containingof 5.0 g soil aliquots (Clapp et al., 1996). Root samples were

collected during the wet sieving step of spore collection.

2.2. Sample collection from organic farmssoil, infected roots, and spores), root, and extracted spore

samples prior to use in DNA extraction. Spore samples were

WY110 Glomus mosseae INVAM

IT104 Glomus versiforme INVAM

NT4a,b Glomus luteum University of Saskatchewan,

Soil Microbiology Lab

Culture Collection

NT7b Glomus versiforme University of Saskatchewan,

Soil Microbiology Lab

Culture Collection

a NT4 was the voucher specimen submitted by Talukdar and Germida

(1993) to INVAM for classification and archive. It was given the INVAM

accession no. SA101 and have been maintained in successive cultures for

research use since submission. It was originally classified as G. clarum

based on spore morphology (Talukdar and Germida, 1993) but

subsequently reclassified as G. luteum (Kennedy and Morton, 1999).b The designations NT4 and NT7 are not INVAM accession numbers.

These are the University of Saskatchewan Soil Microbiology Lab Culture

Collections isolate numbers.Table 1

Reference arbuscular mycorrhizal fungi species used as controls in this

study

INVAM

accession no.

Species Source

AU102 Gigaspora decipiens INVAM

UT316 Glomus etunicatum INVAMa

hemistry 37 (2005) 15891597Selected spores (1040 per isolate) were vortexed at

Biocmaximum speed two times for 30 s each in 100 mL 1% SDS(w/v), and washed with 100 mL of 10 mM TrisHCl, pH 8.0,1 mM ethylenediamine tetra-acetic acid (EDTA) buffer

(TE) between vortex steps. Excess TE was removed after

final wash. Three cycles of crush/freeze/thaw were per-

formed on the spores using a flame sterilized glass

micropestle and liquid N2. Raw lysates were suspended in

60 mL TE. An equal volume of phenol:chloroform:isoamy-lalcohol (24:24:1) was added to each lysate and vortexed

twice for 30 s at maximum speed. The tubes were

centrifuged for 3 min at 3000g. Aqueous layers (containing

the extracted DNA) were removed, placed in new tubes, and

kept on ice. Another volume of TE was added to the raw

lysate/phenol:chloroform:isoamylalcohol mixture, and the

extraction procedure was repeated to increase yield. To

remove phenol, an equal volume of chloroform:isoamylal-

cohol (24:1) was added to the collected aqueous phase and

the tube was inverted gently for 10 s. The aqueous phase

was placed in a new tube with two volumes of 30% (w/v)

polyethylene glycol 40001.6 M NaCl and incubated at

room temperature for 2 h to precipitate the DNA. Pre-

cipitated DNA was spun at 14,000g for 10 min to pellet. The

supernatant was removed and the pellet washed with 100 mLK20 8C 70% (v/v) ethanol. Ethanol was drained and thepellet allowed to air dry for 10 min. Finally, the pellet was

suspended in 30 mL autoclaved distilled and deionized water(ddH2O). DNA extraction from soil followed these steps

except a 0.5 g sample was crushed in 750 mL TE (in three250 mL increments) using a flame sterilized mortar andpestle during the preparation of the raw lysate.

2.4. Nested PCR strategy and conditions

DNA isolated from the spore and soil samples was

subjected to a first PCR using primers (0.5 mM each) GeoA2and Geo11 to amplify an approximately 1.8 kb fragment of

the 18S rRNA gene (Schwarzott and Schussler, 2001).

These primers are universal 18S rDNA fungal primers

(based on sequence match to representatives from all phyla

of Fungi) that amplified all fungal DNA. PCR was done in

20 mL volume with 2.0 mL template DNA (w10 ng mLK1)using the Taq PCR Master Mix system (Qiagen; Hilden,

Germany) with the manufactures recommended buffer,

enzyme, and nucleotide conditions (1! Qiagen PCR buffercontains 1.5 mM MgCl2, 2.5 units Taq DNA polymerase,

and 200 mM of each dNTP). Product was amplified on aRobocycler Gradient 96 (Stratagene; California, USA)

using the following conditions: 948C for 2 min; 30x(94 8C,30 s; 59 8C, 60 s; 72 8C, 2.5 min); 72 8C, 10 min. PCRproduct was analyzed by agarose gel electrophoresis (1.0%

(w/v) agarose; 100 V, 2030 min) and ethidium bromide

staining.

First stage PCR product with a visible band was diluted

1:100 (PCR product without a visible band was undiluted)

and used as template in subsequent nested PCR using

W.K. Ma et al. / Soil Biology &the reaction mixture described above except for primers.The second stage primers (AM1 (Helgason et al., 1998) and

NS31-GC (which corresponds to NS31 described by Simon

et al. (1992) plus a 5 0 GC clamp sequence described byKowalchuk et al. (1997)) produce an approximately 550 bp

fragment. NS31 is a universal fungal primer (based on

sequences matched from GenBank) while AM1 is specific

to the AMF orders of Glomerales and Diversisporales but

not Archaeosporales and Paraglomerales (subsequent

results indicate AM1 is not specific to AMF). Thermo-

cycling used the following condition: 94 8C for 2 min; 30!(94 8C, 30 s; 67 8C, 60 s; 72 8C, 60 s); 72 8C, 10 min. NestedPCR product was analyzed as described for products of the

first PCR.

2.5. DGGE analysis

Ten micro-litres of PCR product were used for DGGE

analysis. Gels contained 4% (w/v) polyacrylamide (37:1

acrylamide/bis-acrylamide) 1! Tris/acetic acid/EDTAbuffer (TAE), and were 1.5 mm thick (20!20 cm). Thelinear gradient used was from 32 to 50% denaturant, where

100% denaturing acrylamide was defined as containing 7 M

urea and 40% (v/v) formamide. A 10 mL stacking gel

containing no denaturants was added before polymerization

was complete (w2 h). All DGGE analysis was run inDCode system (Bio-Rad Laboratories, Hercules, CA, USA)

at a constant temperature of 60 8C. Electrophoresis was for10 min at 75 V, after which the voltage was lowered to 45 V

for an additional 16 h. Gels were stained in 1! TAEcontaining 4 mL Sybr Green per 20 mL TAE and visualizedby UV illumination. Gel images were digitally captured by a

Nikon CoolPix 4500 digital camera with a Sybr Green filter.

2.6. Sequence analysis of DGGE bands and partial 18S

rDNA sequences from spores and soil

Prominent DGGE bands were excised from the UV

illuminated acrylamide gels and DNA eluted from the

excised gel by incubation in 30 mL ddH2O at 28 8Covernight. Eluted DNA was used for PCR amplification as

described above, and products again analysed by DGGE

using a narrower gradient (3446%). PCR products with

single bands on the second DGGE were purified for

sequence analysis using the QiaQuick PCR purification kit

(Qiagen; Hilden, Germany) with a final elution volume of

30 mL. The National Research CouncilPlant Biotechnol-ogy Institute DNA Sequencing Lab (Saskatoon, Canada)

performed the sequencing reactions using the primer NS31

(without GC clamp). Similarity comparisons of the partial

18S rDNA sequences were performed using the National

Centre for Biotechnology Information (NCBI) online

standard BLAST (Basic Local Alignment Search Tool)

program (http://www.ncbi.nlm.nih.gov/). Screening for

possible chimeric sequences was done using the Ribosomal

Database Project (RDP) online Chimera Check program

hemistry 37 (2005) 15891597 1591(http://rdp.cme.msu.edu/html/analyses.html). All sequences

accessions AY641811 to AY641828.

2.7. Detection limit of optimized procedures

3. Results

between INVAMs morphological classification of the

reference isolates and the GenBank database.

3.2. DGGE band analysis of reference spore samples

Most related isolate(s) from GenBank

(% sequence similarity by BLAST)aGenBank accession no. for

most related sequences

Gigaspora decipiens isolate BEG45 (98%) U96146, GI:2073578

Gigaspora decipiens isolate BEG45 (99%) U96146, GI:2073578

Glomus luteum (99%) AJ276089, GI:14270359

Glomus etunicatum isolate UT316 (99%) Y17639, GI:14275537

Glomus mosseae isolate BEG124 (100%) AJ505618, GI:22293519

Glomus versiforme isolate BEG47 (99%) X86687, GI:14018352

Glomus versiforme isolate BEG47 (100%) X86687, GI:14018352

Verticillium psalliotae strain CBS 639.85 (98%) AF339610, GI:15022605

Phialophora verrucosa (99%) AJ232945, GI:15865216

dt and Goebel (1994) demonstrated that at sequence similarity values below 97%,

ciation after complete denaturation (the standard for species identity), and, hence,

Biochemistry 37 (2005) 158915973.1. DNA extraction from and PCR-DGGE resultsThe detection limit of the optimized procedures for

INVAM culture Gigaspora decipiens AU102 was deter-

mined. Zero, two, four, six and eight spores were spiked into

each of five 0.5 g soil samples and subjected to the

optimized extraction and PCR-DGGE procedures pre-

viously described. The detection limit is expressed as the

number of spores required for detection by DGGE per gram

of soil sample.from this work was submitted to GenBank and given the

Table 2

Sequences recovered from DGGE bands of reference spores and soils

INVAM

accession no.

Species classification by

INVAM

Sequence designation

AU102 Gigaspora decipiens AU102-4b

AU102 Gigaspora decipiens AU102-5b

SA101 Glomus luteum SA101-1b

WY110 Glomus mosseae WY110-6b/WY110-7b

IT104 Glomus versiforme IT104-2b

IT104 Glomus versiforme IT104-3b

N/A N/A AU102-Bc

N/A N/A SA101-Ac

a 97% sequence similarity is minimum requirement for identity. Stackbran

it is unlikely that two organisms will have more than 70% DNADNA reasso

they are related at no more than the species level.b Sequence designations are as labelled in Fig. 1.c Sequence designations are as labelled in Fig. 2.

W.K. Ma et al. / Soil Biology &1592of reference spore samples

Originally, the extraction method as defined by Griffiths

et al. (2000) was used to extract DNA from spore samples.

No PCR amplifiable template was produced by this method.

The optimized method developed required the use of a

mortar and pestle to consistently extract amplifiable DNA

from reference fungal spores. Upon sequential amplification

with the GeoA2/Geo11 and NS31-GC/AM1 primer pairs,

spore PCR products of the expected size (w1.8 kbp andw550 bp, respectively) were observed for all the referenceisolates tested. Partial 18S rDNA sequences were obtained

from the excised DGGE bands for the reference spore

samples of Glomus luteum SA101, Glomus versiforme

IT104, Gi. decipiens AU102, and Glomus mosseae WY110.

Table 2 lists the designation of the sequences recovered (as

labelled in Fig. 1) and their most closely related isolate(s)

determined by BLAST search of GenBank. BLAST

searches yielded a minimum of 98% sequence similarityAnalysis of reference spores by PCR-DGGE was

performed in triplicate. No discernable difference in

DGGE pattern was observed. The DGGE analysis of the

NS31-GC/AM1 primed products yielded banding patterns

within the range of 3940% denaturant under our conditions

(Fig. 1). Isolates, G. luteum SA101, G. versiforme IT104,

Gi. decipiens AU102, and G. mosseae WY110, wereFig. 1. DGGE profiles of 18S rDNA fragments for reference AMF spores.

Lane 1: G. luteum SA101; Lane 2: G. versiforme IT104; Lane 3: Gi.

decipiens AU102; Lane 4: G. mosseae WY110; Lane 5: G. etunicatum

UT316; Lane 6: mixed spores (Gi. decipiens AU102, G. luteum SA101, G.

mosseae WY110, and G. versiforme IT104); Lane 7: G. versiforme NT7;

Lane 8: G. luteum NT4. Each arrow locates a single band. Each band is

labelled with the INVAM accession number followed by a sequential

designation (e.g. SA101-1 denotes the PCR-DGGE band from INVAM

reference culture G. luteum SA101, and it was the first band excised from

the gel). Sequencing of bands for G. etunicatum UT316, G. luteum NT4,

and G. versiforme NT7 (Lanes 5, 7 and 8, respectively) was attempted but

no usable sequences were obtained.

distinguishable from each other based upon DGGE

mobility. Isolates SA101, Glomus etunicatum UT316, G.

versiforme NT7, and G. luteum NT4 were visually

indistinguishable. Isolates IT104 and AU102 produced a

distinctive double-band DGGE signature, and arguably, all

reference isolates produced this doublet feature in the

DGGE gel. BLAST results indicated bands IT104-2/IT104-

3 and AU102-4/AU102-5 were18S rDNA sequences of G.

versiforme and Gi. decipiens, respectively (Table 2).

We tested whether the procedure was able to delineate

members of a simple AMF community. Spores from

reference cultures Gi. decipiens AU102 and G. mosseae

WY110 were collected together (3 and 10 spores,

respectively) and subjected to the optimized extraction

and PCR-DGGE procedures. From Fig. 1, Lane 6

contained bands with similar DGGE mobility to isolate

Gi. decipiens AU102 (Lane 3) and isolate G. mosseae

WY110 (Lane 4).

3.3. Detection limit of the developed molecular procedure

when applied to reference soils

Griffiths et al.s (2000) DNA extraction method was able

to extract amplifiable template from reference root and soil

samples, but the desired AMF 18S rDNA fragments were

not produced consistently (Fig. 2, Lanes 2, 6 and 10). With

AU102, respectively). Bands not corresponding to AMF

signatures were observed (bands encompassed by dotted

box). The identity of two of these non-AMF bands was

determined (Table 2).

To overcome the inconsistent DNA extraction from soil,

a scaled-up version of the mortar and pestle method

adapted from the reference spore DNA extraction was used

to generate the soil PCR-DGGE profile of Fig. 2 (Lanes 3, 4,

7, 8, 11, and 12). With the exception of Lane 11, the

optimized methodology produced detectable AMF signa-

tures in all reference soil samples. However, the detection

limit of the procedure for Gi. decipiens AU102 must be

considered because of the absence of an AU102 band in

Lane 11. The spore density of AU102 reference soil was two

spores per gram of soil (sp gK1). Given the absence of an

AMF band in Lane 11, the detection limit must be greater

than two spores per gram of soil. From Fig. 3, an AU102

signature from soil extracts was produced when spiked with

four or greater sp gK1; therefore, the detection limit of the

method for AU102 was 26 sp gK1.

3.4. PCR-DGGE detection of AMF in organic farm soils

The success of the developed assay for detecting AMF

was judged by its ability to detect AMF in soil samples with

various physical, chemical, and biological properties.

W.K. Ma et al. / Soil Biology & BiocFig. 2. DGGE profiles of 18S rDNA fragments for reference soils. Lanes 1,

5, 9: reference AMF 18S rDNA DGGE signatures generated from reference

spore extracted DNA using the optimized procedure (G. luteum SA101, G.

versiforme IT104, and Gi. decipiens AU102 are represented, respectively);

Lanes 2, 6, 10: 18S rDNA DGGE profiles generated from reference soil

extracted DNA using Griffiths et al.s (2000) method; Lanes 3, 7, 11: 18S

rDNA DGGE profiles generated from reference soil extracted DNA using

the optimized method; Lanes 4, 8, 12: 18S rRNA gene DGGE profile

generated from reference soil extracted DNA using the optimized method

with extra spores (SA101Z30, IT104Z30, AU102Z11) added to the soilprior to DNA extraction. Unlabelled arrows locate AMF bands. Non-AMFthe exception of Lane 2, no corresponding AMF DGGE

signature was observed (reference AMF mobility signaturesbands (as determined by sequencing) are encompassed by dotted box.were represented by Lanes 1, 5, and 9 which corresponded

to G. luteum SA101, G. versiforme IT104, and Gi. decipiens

Fig. 3. DGGE profiles of 18S rDNA fragments from Gi. decipiens AU102

soils with different numbers of spores added to the soil sample prior to DNA

extraction. Lane 1: 18S rDNA DGGE signature generated from spore

extracted DNA from reference culture AU102; Lanes 26: 18S rDNA

DGGE profiles generated from reference soil extracted DNA from

reference culture AU102 using the optimized method with 0, 4, 8, 12,

and 16 spores per gram of soil (sp gK1), respectively, spiked into the soil

prior to DNA extraction. Arrow locates AU102 bands. Non-AU102 bands

are encompassed by dotted box. The decline in band intensity in Lanes 5

and 6 was an artefact, rather than inhibition of PCR by increasing template

concentration, because the samples in these two lanes were accidentally

flushed with the pipette during sample loading.hemistry 37 (2005) 15891597 1593Analysis of all farm samples was done in duplicate and

BiocFig. 4. Sample DGGE profile of 18S rDNA fragments from organic farm

soils. Lane L1 is a ladder constructed with reference spore PCR products of

G. luteum SA101 (top two bands), G. versiforme IT104 (second set of two

bands), Gi. decipiens AU102 (third set of two bands) and PCR products of

two non-AMF bands (last set of two bands). Lane L2 is a ladder constructed

with PCR product of DGGE gel eluted DNA from Fig. 1 (Band 1ZSA101-1, Band 2ZWY110-6, Band 3ZIT104-2, Band 4ZIT104-3, Band5ZAU102-4, Band 6ZAU102-5). Other lane designations denote thecorresponding soil sample number (e.g. Lane 1-1 denotes the soil sample

W.K. Ma et al. / Soil Biology &1594the resulting patterns per sample were similar (result not

shown). Twenty-three of the 38 samples processed had

prominent DGGE bands of AMF origin (i.e. bands within

mobility range of reference bands; Fig. 4). Fifteen of the 38

samples had no detectable AMF bands, but non-AMF

DGGE bands (i.e. those bands outside the DGGE mobility

range of the reference AMF species but similar to the two

non-AMF bands sequenced from Fig. 2) were observed in

these samples. The majority of recovered DGGE bands (38

of 50 bands) were identified as Glomus sp. either by DGGE

mobility or sequencing. Attempts were made to sequence all

bands within the mobility range of the reference bands but

some bands could not be recovered from DGGE gels or did

not produce usable sequences.

Bands that produced usable sequences are listed in

Table 3 along with the identity of their closest related

sequence from GenBank. The majority of sequenced bands

were identified as Glomus sp. In addition, Scutellospora

calospora, a previously undetected AM fungus in Saskatch-

ewan, Canada was found in this survey (Fig. 4, Lane 4-1,

band 4-1-1). Bands corresponding to all species represented

by the reference culturesexcept for Gi. decipiens

AU102were detected by band mobility in field samples.

nested PCR-DGGE technique identified AMF isolates in

reference spores and soils (Fig. 1 and 2) and in farm soil

from field #1 of producer #1 whereas Lane 1-2 denotes the soil sample from

field #2 of producer #1). Designations below the sample numbers refer to

the soil zones from which the samples came: BZBlack, BrZBrown,DBZDark Brown, DGZDark Grey, GZGrey. Bands with usablesequences are located by arrows and labelled with the respective lane

designation and a sequential number (e.g. Band 4-1-1 was the first

sequenced band and it is from Lane 4-1). Bands with mobility in the range

of non-AMF reference bands were considered non-AMF without sequen-

cing. Information concerning Lane 13-2 was lost during sample collection

and processing; hence, its result was not considered.samples (Fig. 4). However, the primers used have poor

specificity as they co-amplified non-AMF DNA fromHowever, G. mosseae was the only sequence-confirmed

reference species found in field soils.

4. Discussion

Our adapted DNA extraction procedure and optimized

Table 3

Sequences recovered from DGGE bands of organic farm field soil

Sequence

designationaMost related isolate from Gen-

Bank (% sequence similarity

by BLAST)b

GenBank accession no.

of the most related

sequences

4-1-1 Scutellospora calospora (99%) AJ306445, GI:15211856

6-1-2 Glomus sp. Glo18 isolate

(100%)

AY129625,

GI:23092378

12-1-3 Glomus mosseae isolate

EEZ21 (100%)

AJ506089, GI:22474490

13-2-4 Glomus sp. Glo4 isolate (99%) AF074353, GI:3342472

19-1-5 Glomus mosseae BEG122

(99%)

AJ505616, GI:22293517

21-1-6 Glomus sp. Glo4 isolate (98%) AF074353, GI:3342472

36-1-7 Glomus sp. Glo18 isolate

(99%)

AY129625,

GI:23092378

47-2-8 Glomus sp. 5014b25.Llao5

(96%)

AF480158,

GI:23451949

47-2-9 Glomus sp. Glo4 isolate (99%) AF074353, GI:3342472

a Sequence designations are as labelled in Fig. 4.b 97% sequence similarity is minimum requirement for identity.

hemistry 37 (2005) 15891597reference soil samples (Fig. 2). At the time of selection,

the primer AM1 appeared specific to AMF only (based on

database check). From our results and recent check of

GenBank, this primer sequence can amplify non-AMF

templates (e.g. Fig. 2). Primer specificity for AMF in soil

samples need to be stringent to exclude non-AMF templates

during PCR amplification (Anderson et al., 2003). This is

significant when PCR is used in conjunction with commu-

nity profile techniques such as DGGE where each band is

assumed to be of fungal origin. In retrospect, selection or

development of specific primers may improve detection of

AMF in those field samples with only non-AMF DGGE

bands (Fig. 4).

Extraction efficiency of AMF DNA, as a function of

soil inoculum level, relative to other non-AMF DNA

must be sufficiently high to overcome the detection limit

of PCR procedures (Clapp et al., 1995). This is

illustrated by the detection limit test performed on Gi.

decipiens AU102 (Fig. 3). Below the detection limit,

non-AMF DNA is amplified by PCR and its product

visible by DGGE (Lane 2). Above the detection limit,

Gi. decipiens AU102 DNA was amplified and visualized

by DGGE whereas non-AMF bands were excluded

Bioc(Lanes 35). Martin-Laurent et al. (2001) demonstrated

that different extraction protocols may provide conflicting

estimates of soil microbial diversity depending on their

efficiency.

The similar DGGE mobility of isolates G. luteum SA101

and G. etunicatum UT316 was unexpected under the

assumption that different species differ in 18S nucleotide

sequence and GC-content (Fig. 1, Lanes 1 and 5,

respectively); however, the high percentage of sequence

similarity (Table 2) of SA101 to UT316 may indicate

insufficient nucleotide sequence and GC-content differences

in the targeted 18S rDNA fragment to distinguish between

SA101 and UT316 by DGGE. Isolates G. versiforme NT7

(Lane 7) and G. luteum NT4 (Lane 8) had similar DGGE

mobility as G. luteum SA101 and G. etunicatum UT316.

Mobility similarity between G. luteum NT4 and G. luteum

SA101was justified because NT4 was the voucher specimen

for SA101. Contamination with G. luteum NT4 or

mislabelling of G. luteum NT4 cultures during successive

cultivation may cause the observed results with G.

versiforme NT7. Alternatively, isolate G. versiforme NT7

relates closer to G. luteum SA101 and G. etunicatum UT316

than to G. versiforme IT104 (Lane 2).

Operon heterogeneity appeared to play an important role

in DGGE banding patterns. We observed a double-band

pattern for all reference isolates. Operon heterogeneity has

been reported by others (Sanders et al., 1995; Clapp et al.,

1999; Kuhn et al., 2001). Whether the two variants of the

18S rRNA gene observed is from a single spore or different

spores are unknown because our reference spore PCR-

DGGE profile was generated with extractions from multiple

spores. This observation complicates the interpretation of

field DGGE profiles because any AMF isolate may be

represented by two or more bands. Therefore, the actual

number of AMF in a field DGGE profile may be less than

half of what is visually detected by DGGE.

Other phenomena may explain the doublet feature

observed. First, primer specificity could contribute to the

doublet pattern if related but non-AMF templates were

present in a sample. Poor specificity at the primer 3 0-endcould produce non-specific products (Anderson et al., 2003;

Innis and Gelfand, 1999). Second, chimeric DNA molecule

formation has been recognized as a source of sequence

infidelity (Wang and Wang, 1997). These hybrid molecules

from two organisms with sequence homology could produce

doublets with related mobility during DGGE. Third, the

error rate of the proprietary Taq used in the Master Mix is

unknown, but non-proofreading polymerases have reported

error rates ranging from 4.0!10K2 to 2.2!10K4 (Innis andGelfand, 1999). For the 30-cycle reaction used to amplify a

fragment of w550 bp, the number of potential mismatchinsertions range from 1 to 40 bases. The potential error

is doubled because of the nested PCR strategy used.

Finally, Kocherginskaya et al. (2001) demonstrated that

single-stranded and double-stranded molecules from the

W.K. Ma et al. / Soil Biology &same template have different mobility during DGGE.This cursory survey of 38 field soils from organic farms

by molecular techniques found 04 species per site

(assuming each band represented a different isolate). This

is similar to Talukdar and Germidas (1993) identification of

36 species by trap culture techniques for conventional farm

sites across the province. Both results are low when

compared to the number of species detected at other

temperate low-input sites (26 species, Oehl et al., 2003),

conventional cultivated sites (13 species, Hamel et al.,

1994) and native sites (37 species, Bever et al., 2001). The

combination of extraction inefficiency in conjunction with

low primer specificity, i.e. non-AMF is amplified instead of

AMF (e.g. Fig. 3), and similar migration behaviour of DNA

fragments with different origin but same GC-content

(Kowalchuk et al., 2002) may underestimate the number

of species in a sample. The sampling strategy used may not

be optimally suited to assess community composition of

AMF in agricultural soils. For example, the 015 cm sample

used for DNA extraction will miss species present at greater

depth (Douds et al., 1995), and compositing of samples may

dilute the number of spores per gram of soil because of

patchy occurrence of AMF spores (Smith and Read, 1997).

Or, simply, there are fewer species in Saskatchewan soils.

More work is needed to address these possibilities.

The dominant number of Glomus-like bands observed

was not surprising (Fig. 4 and Table 3). In particular, the

large number and wide distribution of G. mosseae-like

bands (those with similar mobility to G. mosseae WY110

bands) observed agrees with the literatures general

assessment of G. mosseae as a common AM fungus found

in a variety of cultivated field soils (Sylvia and Schenck,

1983). The near absence of members from the Diversispor-

ales (i.e. Gigaspora and Scutellospora) concurs with the

correlation of Gigaspora and Scutellospora population

decline with cultivation (Douds et al., 1993). From the

high proportion of bands observed from the Grey to Dark

Grey soil zones (33 of 50 bands) it is arguable cropping

history and management (data not shown) in relation to soil

type and climate promoted or inhibited AMF establishment

and maintenance. For example, rotation of poorly mycor-

rhizal crops (An et al., 1993) with fallowing and tillage

(Kabir et al., 1998) in the southwestern part of the province

(Brown and Dark Brown soil zones) will select for specific

AMF species and diminish soil inoculum levels. In contrast,

producers in the Grey Dark Grey soil zones generally grow

strongly mycorrhizal leguminous crops (e.g. pea and lentil)

with no-till or maintained forage cover such as alfalfa.

These conditions could promote AMF diversity and build up

soil inoculum levels (Douds and Millner, 1999). The

aforementioned problems with DNA extraction and PCR

are aggravated by the likely lower diversity and inoculum

levels as a function of the agriculture practiced in the areas

of the Dark Brown and Brown soil zones.

A diverse AMF population is a key factor to improve the

sustainability of low-input and organic cropping systems.

hemistry 37 (2005) 15891597 1595To increase our ability to optimize management of AMF in

community. New Phytologist 130, 259265.

Clapp, J.P., Fitter, A.H., Merryweather, J.W., 1996. Arbuscular mycor-

rhizas. In: Hall, J.S. (Ed.), Methods for the Examination of Organismal

Douds, D.D., Janke, R.R., Peters, S.E., 1993. VAM fungus spore

populations and colonization of roots of maize and soybean under

conventional and low-input sustainable agriculture. Agriculture,

W.K. Ma et al. / Soil Biology & Biochemistry 37 (2005) 158915971596Ecosystems and Environment 43, 325335.

Douds, D.D., Galvez, L., Janke, R.R., Wagoner, P., 1995. Effect of tillage

and farming system upon population and distribution of vesicular

arbuscular mycorrhizal fungi. Agriculture, Ecosystems and Environ-

ment 52, 111118.

Gange, A.C., West, H.M., 1994. Interactions between arbuscular mycor-

rhizal fungi and foliar-feeding insects in Plantago lanceolata L. New

Phytologist 128, 7987.

Griffiths, R.I., Whiteley, A.S., ODonnell, A.G., Bailey, M.J., 2000. Rapid

method for coextraction of DNA and RNA from natural environments

for analysis of ribosomal DNA- and rRNA-based microbial communityDiversity in Soils and Sediments. CAB International, New York,

pp. 145161.

Clapp, J.P., Fitter, A.H., Young, J.P.W., 1999. Ribosomal small subunit

sequence variation within spores of an arbuscular mycorrhizal fungus,

Scutellospora sp. Molecular Ecology 8, 915921.

Douds, D.D., Millner, P.D., 1999. Biodiversity of arbuscular mycorrhizal

fungi in agroecosystems. Agriculture, Ecosystems and Environment 74,

7793.field situations, there is a need for more information on how

agricultural practices influence the variation in AMF

development and function in different crop species (Smith

and Read, 1997). The reliance on spore morphology to

characterize AMF communities is subjective and provides

an incomplete interpretation of their in situ reality. The

molecular techniques described here are one tool to

objectively characterize complex fungal communities in

agro-ecosystems.

Acknowledgements

This study was supported by the Natural Sciences and

Engineering Research Council of Canada. We thank Rachel

Buhler for providing the organic farm field soils for this

work.

References

An, Z.Q., Guo, B.Z., Hendrix, J.W., 1993. Mycorrhizal pathogens of

tobacco: cropping history and current crop effects on the mycorrhizal

community. Crop Protection 12, 527531.

Anderson, I.C., Campbell, C.D., Prosser, J.I., 2003. Potential bias of fungal

18S rDNA and internal transcribed spacer polymerase chain reaction

primers for estimating fungal biodiversity in soil. Environmental

Microbiology 5, 3647.

Bever, J.D., Schultz, P.A., Pringle, A., Morton, J.B., 2001. Arbuscular

mycorrhizal fungi: more diverse than meets the eye, and the ecological

tale of why. BioScience 51, 923931.

Clapp, J.P., Young, J.P.W., Merryweather, J.W., Fitter, A.H., 1995.

Diversity of fungal symbionts in arbuscular mycorrhizas from a naturalcomposition. Applied and Environmental Microbiology 66, 54885491.Gupta, R., Kumar, P., 2000. Mycorrhizal plants in response to adverse

environmental conditions. In: Mukerji, K.G., Chamola, B.P., Singh, J.

(Eds.), Mycorrhizal Biology. Kluwer Academic/Plenum Publishing,

New York, pp. 6784.

Hamel, G., Dalpe, Y., Lapierre, C., Simard, R.R., Smith, D.L., 1994.

Composition of the vesiculararbuscular mycorrhizal fungus popu-

lation in an old meadow as affected by pH, phosphorus, and soil

disturbance. Agriculture, Ecosystems and Environment 49, 223231.

Helgason, T., Daniell, T.J., Husband, R., Fitter, A.H., Young, J.P.W., 1998.

Ploughing up the wood-wide-web?. Nature 384, 431.

Ibijbijen, J., Urquaiga, S., Ismali, M., Alve, J.R., Boddey, R.M., 1996.

Effect of arbuscular mycorrhizal fungi on growth, mineral nutrition, and

nitrogen fixation of three varieties of common bean (Phaseolus

vulgaris). New Phytologist 134, 353360.

Innis, M.A., Gelfand, D.H., 1999. Optimization of PCR: conversations

between Michael and David. In: Innis, M.A., Gelfand, D.H.,

Sninsky, J.J. (Eds.), PCR Applications: Protocols for Functional

Genomics. Academic Press, San Diego, CA, pp. 322.

Jakobsen, I., 1999. Transport of phosphorus and carbon in arbuscular

mycorrhizas. In: Varma, A., Hock, B. (Eds.), Mycorrhiza: Structure,

Function, Molecular Biology. Springer, Berlin, pp. 535542.

Jarstfer, A.G., Sylvia, D.M., 2002. Isolation, culture, and detection of

arbuscular mycorrhizas. In: Hurst, C.J., Crawford, R.L., Knudsen, G.R.,

McInerney, M.J., Stetzenbach, L.D. (Eds.), Manual of Environmental

Microbiology. American Society of Microbiologists Press, Washington,

DC, pp. 535542.

Kabir, Z., OHalloran, I.P., Fyles, J.W., Hamel, C., 1998. Dynamics of the

mycorrhizal symbiosis of corn (Zea mays L.): effects of host

physiology, tillage practice and fertilization on spatial distribution of

extra-radical mycorrhizal hyphae in the field. Agriculture, Ecosystems

and Environment 68, 151163.

Kennedy, L.J., Morton, J.B., 1999. Glomus eburneum and G. luteum, two

new species of arbuscular mycorrhizal fungi, with emendation of G.

spurcum. Mycologia 91, 10831093.

Kocherginskaya, S.A., Aminov, R.I., White, B.R., 2001. Analysis of the

rumen bacterial diversity under two different diet conditions using

denaturing gradient gel electrophoresis, random sequencing, and

statistical ecology approaches. Anaerobe 7, 119134.

Kowalchuk, G.A., Gerards, S., Woldendorp, J.W., 1997. Detection and

characterization of fungal infections of Ammophila arenaria (marram

grass) roots by denaturing gradient gel electrophoresis of specifically

amplified 18S rDNA. Applied and Environmental Microbiology 63,

38583865.

Kowalchuk, G.A., de Souza, F.A., van Veen, J.A., 2002. Community

analysis of arbuscular mycorrhizal fungi associated with Ammophila

arenaria in Dutch coastal sand dunes. Molecular Ecology 11, 571581.

Kuhn, G., Hijri, M., Sanders, I.R., 2001. Evidence for the evolution of

multiple genomes in arbuscular mycorrhizal fungi. Nature 414, 745

748.

Lu, X.H., Koide, R.T., 1994. The effect of mycorrhizal infection on

components of plant-growth and reproduction. New Phytologist 128,

211218.

Martin-Laurent, F., Philippot, L., Hallet, S., Chaussod, R., Germon, J.C.,

Soulas, G., Catroux, G., 2001. DNA extraction from soils: old bias for

new microbial diversity analysis methods. Applied and Environmental

Microbiology 67, 23542359.

Muyzer, G., de Waal, E.C., Uitterlinden, A.G., 1993. Profiling of

complex microbial populations by denaturing gradient gel electro-

phoresis analysis and polymerase chain reaction-amplified genes

coding for 16S rRNA. Applied and Environmental Microbiology 59,

695700.

Oehl, F., Sieverding, K., Ineichen, K., Mader, P., Boller, T., Wiemken, A.,

2003. Impact of land use intensity on the species diversity of arbuscular

mycorrhizal fungi in agroecosystems of Central Europe. Applied and

Environmental Microbiology 69, 28162824.

Sanders, I.R., Alt, M., Groppe, K., Boller, T., Wiemken, A., 1995.Identification of ribosomal DNA polymorphisms among and within

spores of Glomales: application to studies on the genetic diversity of

arbuscular mycorrhizal fungal communities. New Phytologist 130,

419427.

Schwarzott, D., Schussler, A., 2001. A simple and reliable method for SSU

rRNA gene DNA extraction, amplification, and cloning from single AM

fungal spores. Mycorrhiza 10, 203207.

Simon, L., Lalonde, M., Bruns, T.D., 1992. Specific amplification of

18S fungal ribosomal genes from vesiculararbuscular mycorrhizal

fungal communities. Applied and Environmental Microbiology 58,

291295.

Smith, S.E., Read, D.J., 1997. Mycorrhizal Symbiosis. Academic Press,

Cambridge, MA. 605 pp.

Smit, E., Leeflang, P., Glandorf, B., van Elsas, J.K., Wernars, K., 1999.

Analysis of fungal diversity in the wheat rhizosphere by sequencing of

cloned PCR-amplified genes encoding 18S rRNA and temperature

gradient gel electrophoresis. Applied and Environmental Microbiology

65, 26142621.

Stackbrandt, E., Goebel, B.M., 1994. A place for DNADNA reassociation

and 16S ribosomal RNA sequence analysis in the present species

definition in bacteriology. International Journal of Systematic Bacteri-

ology 44, 846849.

Sylvia, D.M., Schenck, N.C., 1983. Application of superphosphate

to mycorrhizal plants stimulates sporulation of phosphorus-

tolerant vesiculararbuscular mycorrhizal fungi. New Phytologist 95,

655661.

Talukdar, N.C., Germida, J.J., 1993. Occurrence and isolation of vesicular

arbuscular mycorrhizae in cropped fields of Saskatchewan, Canada.

Canadian Journal of Microbiology 39, 567575.

Thomas, A.G., 1985. Weed survey system used in Saskatchewan for cereal

and oilseed crops. Weed Science 33, 3443.

Vainio, E.J., Hantula, J., 2000. Direct analysis of wood-inhabiting fungi

using denaturing gradient gel electrophoresis of amplified ribosomal

DNA. Mycological Research 104, 927936.

Wang, G.C.Y., Wang, Y., 1997. The frequency of formation of chimeric

molecules as consequence of PCR coamplification of 16S rRNA genes

from mixed bacterial genomes. Applied and Environmental Micro-

biology 63, 46454650.

Xavier, L.J.C., Germida, J.J., 2002. Response of lentil under controlled

conditions to co-inoculation with arbuscular mycorrhizal fungi and

rhizobia varying in efficacy. Soil Biology & Biochemistry 34,

181188.

W.K. Ma et al. / Soil Biology & Biochemistry 37 (2005) 15891597 1597

A PCR-DGGE method for detecting arbuscular mycorrhizal fungi in cultivated soilsIntroductionMaterials and methodsReference AMF speciesSample collection from organic farmsDNA extraction from spore and soil samplesNested PCR strategy and conditionsDGGE analysisSequence analysis of DGGE bands and partial 18S rDNA sequences from spores and soilDetection limit of optimized procedures

ResultsDNA extraction from and PCR-DGGE results of reference spore samplesDGGE band analysis of reference spore samplesDetection limit of the developed molecular procedure when applied to reference soilsPCR-DGGE detection of AMF in organic farm soils

DiscussionAcknowledgementsReferences