Analysis of Broad-scale Differences in Microbial Community Composition of Two Pristine Forest Soils

System. App!. Microbio!. 21, 579-587 (1998) _©_G_us_tav_F_is_ch_er_V_erl_ag_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

SYSTEMATIC AND APPLIED MICROBIOLOGY

Analysis of Broad-scale Differences in Microbial Community Composition of Two Pristine Forest Soils ANTONIS CHATZINOTAS\ RUTH-ANNE SANDAA2, WILHELM SCHONHUBER3, RUDOLF AMANN3, FRIDA LISE DAAE2, VIGDIS TORSVIK2, JOSEF ZEYER\ and DITTMAR HAHN! lSwiss Federal Institute of Technology (ETH), Institute of Terrestrial Ecology, Soil Biology, Schlieren, Switzerland 2University of Bergen, Department of Microbiology, Bergen, Norway 3M ax-Planck-Institute for Marine Microbiology, Molecular Ecology, Bremen, Germany Received August 18, 1998

Summary Broad-scale differences in soil microbial community composition were analyzed in two contrasting soils using DNA reassociation and % G+C profiles for analysis on the community-level, and filter- and whole cell hybridization techniques for a coarse-level characterization of larger phylogenetic groups of bacteria. Reassociation analysis of DNA from bacterial fractions extracted from the organic soil Seim and the mineral soil Hau revealed similar complexity of the communities with 5700 and 4900 different bacterial genomes (g soil [dry wt])-l, respectively. Thermal denaturation studies showed wide % G+C distributions in DNA from bacteria of both soils. Differences in the median % G+C with 55 to 61 % for the bacterial community in soil Seim and 61 to 66% for that in soil Hau indicated a higher proportion of bacteria with a high DNA G+C content in soil Hau. In situ hybridization with fluorescent (Cy3-labeled) probes targeting larger phylogenetic groups showed minor differences between both soils, and between direct detection of bacteria in dispersed soil slurries and in bacterial fractions extracted from soils though about 90% of the total bacteria were lost during extraction. In dispersed slurries of both soils, only probes ALFI b, SRB385, and PLA46 hybridized to cells accounting for more than 1 % of the DAPIstained cells, while numbers obtained after hybridization with probes ARCH915, BET42a, GAM42a, HGC69a, and CF319a were below the detection limit set at <1 %. These results were confirmed by in situ hybridization with horseradish peroxidase (HRP)-labeled probes and subsequent Cy3-tyramide signal amplification. In contrast, dot blot hybridization with probe HGC69a indicated significant amounts of Gram-positive bacteria with a high DNA G+C content in both soils. These could subsequently be visualized in non-dispersed soil slurries by in situ hybridization with HRP-Iabeled probe HGC69a and Cy3-tyramide signal amplification. Filamentous Gram-positive bacteria with a high DNA G+C content, likely actinomycetes, which are present in soil Hau in significant numbers are obviously destroyed by procedures used for soil dispersion. Key words: DNA reassociation - in situ hybridization - % G+C-content - probes - rRNA

Introduction Soil is probably the most complex of all microbial habitats. It differs from most other habitats in that it is dominated by a solid phase consisting of particles of different sizes and composition which are surrounded by aqueous and gaseous phases that fluctuate markedly in time and space (NEDWELL and GRAY, 1987; STOTZKY, 1986). The heterogeneous and dynamic nature of soils together with the small and discontinuous size distribution of microhabitats often complicates studies on the inhabiting microbial communities (COLEMAN et al., 1992; PARKIN, 1993). In addition to the effects of environmental factors, studies on microbial communities are also

often impeded by the fact that most microorganisms resist cultivation, which is an essential prelude to characterization by many traditional methods. Detection methods relying on the isolation of microorganisms can therefore he extremly selective and usually underestimate numbers of microorganisms and diversity of microbial communities in soils (BOTH et al., 1990; FISCHER et al., 1995; RICHAUME et al., 1993; SKINNER et al., 1952; S0RHEIM et al., 1989). During the last years, growth-dependent methods were increasingly supplemented with molecular methods in studies on microbial communities in soils (MYROLD et

580

A. CHATZINOTAS et al.

al., 1994) which allow to study microbial communities unbiased by the limitations of culturability at different levels of resolution. For the analysis of bacteria on the community-level, methods such as DNA reassociation (RITZ et al., 1997; TORSVIK et al., 1990; 1996), cross-hybridization of DNAs (GRIFFITHS et al., 1996; 1997; XIA et al., 1995) or the analysis of % guanine + cytosine (% G+C) base profiles (GRIFFITHS et al., 1997; HOLBEN and HARRIS, 1995; RITZ et al., 1997) were used. Methods for a coarse-level of characterization included total fatty acid analysis (FAME) (BUYER and DRINKWATER, 1997; CAVIGELLI et al., 1995), substrate utilization analysis (BIOLOG) (BUYER and DRINKWATER, 1997; WDNSCHE et al., 1995), amplified ribosomal DNA restriction analysis (ARDRA) (MASSOL-DEYA et al., 1997; SMIT et al., 1997), denaturing or thermal gradient gel electrophoresis (DGGE or TGGE) (FELSKE et al., 1997; HEUER et al., 1997; KOWALCHUK et al., 1997), or hybridization techniques (e.g. filter- or whole cell hybridization) (HAHN et al., 1992; JACOBSEN and RASMUSSEN, 1992; STREIT et al., 1993; ZARDA et al., 1997). For a more specific characterization of microbial communities, e.g. a characterization on the species-level, nucleic acid sequence analysis is usually required (BORNEMAN et al., 1996; LIESACK and STACKEBRANDT, 1992; NDsSLEIN and TIEDJE, 1998; REIMS et al., 1996). The aim of our study was to determine broad-scale differences in soil microbial community structure in two contrasting soils, the organic soil Seim and the mineral soil Hau, respectively, focusing on differences at the community-level and a coarse-level of resolution rather than on specific differences in species composition. For the analysis on the community-level, DNA reassociation and the determination of % DNA G+C profiles were used, while filter- and whole cell hybridization techniques were used for the coarse-level characterization of larger phylogenetic groups of bacteria.

Material and Methods Characteristics of soil samples: Surface samples down to 10 cm depth from the organic soil Seim (Bergen, Norway) and the mineral soil Hau (Birmensdorf, Switzerland) (RICHARD et aI., 1978) were collected at the end of August 1995 and stored at 4 DC. Both soils differ significantly with respect to soil physico-chemical as well as to biological characteristics (Table 1). Extraction of bacteria from soils: Bacteria were extracted from 20-g-samples of the soils (n = 2 for hybridization studies,

and n = 3 for reassociation studies) after homogenization in 100 ml of ice-cold distilled water in a Waring blender in three oneminute-bursts at low speed with intermediate cooling on ice for 5 min. Cells were separated from soil particles by centrifugation at 5 DC and 1,000 x g for 15 min. Homogenization of the pellet in 100 ml of ice-cold distilled water and subsequent centrifugation were repeated twice. Bacteria in the combined supernatants were then concentrated by centrifugation at 5 DC and 10,000 x g for 40 min (TORSVIK, 1995). For reassociation studies, bacteria were further washed with 200 ml of 2% sodium hexametaphosphate (pH 8.5), and subsequently twice with 100 ml of Crombach buffer (CROMBACH, 1972). Finally, bacterial pellets were resuspended in isopropanol with a Ystral homogenizer, speed setting at 2, and stored in the refrigerator (TORSVIK et aI., 1990). Extraction of nucleic acids from soils: For reassociation and thermal denaturation studies, bacteria were centrifuged, resuspended in Crombach buffer and homogenized with the Ystral homogenizer. Bacteria of soil Seim were then lyzed and nucleic acids purified according to TORSVIK et al. (1990), except that potassium acetate instead of ammonium acetate was added to the lysate to remove proteins and humic material. DNA extraction from bacteria of soil Hau required some modifications to the original protocol in order to obtain sufficient amounts of clean DNA. Bacteria from soil Hau were lyzed with lysozyme (50,000 U mg-t, Sigma, St. Louis, USA,S mg ml- 1) in the presence of RNase rIa (Sigma, 50 )1g ml- 1) and RNase T1 (Sigma, 8 U ml- 1) at 37 DC for 1 h, followed by treatment with proteinase K (Sigma, 0.2 mg ml- 1 ) at 37 DC for another 30 min. Complete lysis of bacteria in soil Hau was achieved by subsequent incubation with SDS at a final concentration of 1 % at 65 DC for 15 min and finally by mixing with NaCl0 4 to a final concentration of 1 M. DNA of bacteria from soil Hau was then purified by extraction with chloroformlisoamylalkohol (24:1 v/v) (SAMBROOK et aI., 1989). Nucleic acids of bacteria from both soils were further purified on a hydroxyapatite column (Bio-Gel HT, BioRad) and subsequently concentrated by cetylpyridinium bromide (Sigma) precipitation (TORSVIK et aI., 1990). For hybridization studies, nucleic acid extraction from soil samples was based on a modified bead beating protocol (CRESSWELL et aI., 1991). This protocol was adapted to a microscale using 1.5 ml eppendorf tubes as homogenization vessels. The homogenization mixture contained 0.4 g of soil (wet weight) and 0.4 g of glass beads (0.1-0.11 mm, Braun, Melsungen, Germany) to which extraction buffer (100 mM Tris/HCl, pH 7.5; 1.5% SDS; 10 mM EDTA; 1 % deoxycholat; 1 % NP-40 (v/v); 5 mM thiourea; 10 mM DTT) (HUGHES and GALAU, 1988) was added to a final volume of 1.5 ml. Cells were disrupted in a bead beater (MSK Cell Homogenizer, Braun) in a one-minuteburst at maximum setting (4,000 rpm) with subsequent cooling on ice (HONERLAGE et aI., 1995). Released nucleic acids in the soil homogenates were further purified after centrifugation in an Eppendorf centrifuge at 6,000 rpm for 3 min. The combined

Table 1. Soil characteristics. Soil

Texture (% clay/silt/sand)

Org. material (%)

Dry weight (%)

Water-holding capacity (ml [g soil dry wt]-I)

pH (in water)

CFua (xl0 6 [g soil dry wt]-l)

Seim

nd b 4115217

76.0 5.6

20.0 68.8

0.96 0.56

2.9 6.8

9.5 1.4

Hau

acolony forming units (CFU) determined by incubation of serial dilutions on R2A agar plates (Difco, Detroit, Mich.) at 30 DC for 2 days. bnot determined.

Analysis of Broad-scale Differences supernatant and upper sediment layer (approx. 600 Ill) were extracted with an equal amount of phenol/chloroform, and the nucleic acids afterwards precipitated with ethanol, dried and resuspended in 100 III of distilled water (SAMBROOK et aI., 1989). After subsequent treatment with RNase A (Fluka, Buchs, Switzerland; 95.5 U mg-t, final concentration 0.1 Ilg Ill-I) at 37°C for 30 min, the nucleic acids were further purified by centrifugation through Sephadex G-200 columns (1 ml syringes filled with Sephadex G-200 to 1 ml) (TSAI and OLSON, 1992), again extracted with phenol/chloroform, precipitated, dried, and resuspended in 100 III of distilled water. The presence of DNA was controlled by gel electrophoresis and the total amount estimated by the OD 26o reading. Reassociation and thermal denaturation: For reassociation studies, DNA was sheared in a French press at 20,000 Psi followed by a second purification on a hydroxyapatite column and concentration by cetylpyridinium bromide precipitation. Reassociation of sheared and heat-denatured DNA from soils Seim and Hau (400 Ilg ml- I in 6 x SSC/30% DMSO) (TORSVIK et aI., 1990) was measured as the decrease in absorbance of the DNA kept at constant temperature of 54.5 °C which approximates 25°C below Tm until 50% of the DNA was reassociated (Cary 4 spectrophotometer, Varian, Walton-on-Thames, UK) . The complexity of the DNA in base pairs was calculated relative to the size of the genome of E. coli B (Sigma) assuming 4.1 x 10 6 base pairs (TORSVIK et aI., 1990). Thermal denaturation of DNA from soils Seim and Hau (approx. 25 Ilg ml- I in 0.1 x SSC) (TORSVIK et aI., 1990) was analyzed by the absorbance at 260 nm in a Cary 4 spectrophotometer using DNA of E. coli B (Sigma) as reference. Thermal denaturation of DNA was achieved by increasing the temperature at a rate of 0.1 °C min-I for 14 h. The denaturation profile was calculated in percent hyperchromicity (TORSVIK et aI., 1990). The moles percent guanine plus cytosine (G+C) was calculated using the formula by MANDEL et al. (1970). The first derivative was calculated and used for the analysis of the fine structure of the denaturation profile.

581

Probes: Oligonucleotide probes (Table 2) were synthesized either with or without a primary amino group at the 5' end (MWG Biotech, Ebersberg, Germany). The fluorescent dye Cy3 Reactive Dye (Cy3, Amersham, Zurich, Switzerland) was covalently bound to the amino group of the oligonucleotide probes. Horseradish peroxidase (HRP) (Boehringer, Mannheim, Germany) was conjugated to the amino group of the oligonucleotides via the cross-linking agent p-phenylene-diisothiocyanate. The oligonucleotide-peroxidase conjugates were purified from unreacted components and stored at 4 °C or frozen in small aliquots at -20°C (AMANN et aI., 1992). Digoxigenin-ddUTP (Boehringer) was coupled to the 3' end of the oligonucleotides with terminal transferase (Promega, Zurich, Switzerland) according to the manufacturer's instructions. Dot blot hybridization: For dot blot hybridization, serial dilutions of nucleic acid samples were denatured with NaOH (final cone. 0.2 M) for 20 min at room temperature. After neutralization with ammoniumacetate (final conc. 0.2 M) nucleic acids (final volume 60 Ill) were applied to nylon filters (Boehringer) with a dot blot manifold (BioRad) and eventually bound to the membrane at 120°C for 30 min. Membranes were hybridized with digoxigenin-labeled probes in 0.36 M Na2HP04' pH 7.2, 5% SDS, 1 mM Na2EDTA, and 1 % bovine serum albumin (BSA) (CHURCH and GILBERT, 1984) in the presence of 45% formamide for probe EUB338, and 30% for probe HGC69a, at 49°C for 16 h. The formation of stable hybrids was shown by binding of an antibody-alkaline phosphatase conjugate (Boehringer) to the digoxigenin reporter molecule. Alkaline phosphatase activity was visualized by light emission using CSPD® chemiluminescent substrate (Tropix, Catalys, Wallisellen, Switzerland) and exposure to Kodak X-OMAT AR film according to the manufacturer's instructions (Tropix). Signal intensities were determined by image analysis of scanned filmes (OneScanner®, Apple) using the OmniPage Professional, Prism View Rev. 3.6.2. program. Intensities obtained with group-specific probe HGC69a were compared to those

Table 2. Oligonucleotide probes. Probe

Target

Sequence

Reference

EUB338

Bacteria 16S rRNA, pos. 338-355 Eukarya 16S-like rRNA, pos. 502-516 Archaea 16S rRNA, pos. 915-934 a-subdivision of Proteobacteria 16S rRNA, pos. 19-35 ~-subdivision of Proteobacteria 23S rRNA, pos. 1027-1043 y-subdivision of Proteobacteria 23S rRNA, pos. 1027-1043 ii-subdivision of Proteobacteria 16S rRNA, pos. 385-402 Gram-positive bacteria with high DNA G+C content 23S rRNA, pos. 1901-1918 Cytophaga-Flavobacterium cluster of the CFB phylum 16S rRNA, pos. 319-336 Planctomycetes 16S rRNA, pos. 46-63

j'GCTGCCTCCCGTAGGAGT

(AMANN et aI., 1990b)

5' ACCAGACTTGCCCTCC

(AMANN et aI., 1990b)

EUK516 ARCH915 ALFlb BET42aa GAM42a h SRB385 HGC69a' CF319a PLA46

j'GTGCTCCCCCGCCAATTCCT (STAHL and AMANN, 1991) 5'CGTTCGYTCTGAGCCAG (MANZ et aI., 1992) 5'GCCTTCCCACTTCGTTT

(MANZ et aI., 1992)

5'GCCTTCCCACATCGTTT

(MANZ et aI., 1992)

5'CGGCGTCGCTGCGTCAGG

(AMANN et aI., 1990a)

5'TATAGTTACCACCGCCGT

(ROLLER et aI., 1994)

5TGGTCCGTGTCTCAGTAC

(MANZ et aI., 1996)

5'GACTTGCATGCCTAATCC

(NEEF et aI., in press)

aprobe BET42a was used with an equimolar addition of unlabeled GAM42a as competitor. bprobe GAM42a was used with an equimolar addition of unlabeled BET42a as competitor. 'probe HGC69a was used with an equimolar addition of unlabeled competitor probe.

582

A. CHATZINOTAS et al.

obtained with the bacterial probe EUB338 on DNA of reference bacteria (Azospirillum brasiliensis, Burkholderia cepacia, Pseudomonas aeruginosa, Streptomyces azureus, Bacillus subtilis, and Cytophaga johnsonii). Whole cell- and in situ hybridization: Extracted bacteria from soils Seim and Hau as well as samples from both soils were fixed in 4% paraformaldehydelphosphate buffered saline (PBS, composed of 0.13 M NaCl, 7 mM Na zHP0 4 and 3 mM NaH 2P0 4, pH 7.2 in water) at 0 °C for 16 h (HAHN et aI., 1992), subsequently washed in PBS and stored in 96% ethanol at -20°C at a concentration of 50 mg of soil (wet wt) per ml. Before application to slides, 20 )11 of the soil samples were dispersed in 980 )11 of 0.1 % sodium pyrophosphate in distilled water by mild sonication at a setting of 20 for 1 min (Branson Sonifier B-12, Danbury, Connecticut) (ZARDA et aI., 1997). Ten )11 from each fixed and dispersed sample were spotted onto gelatin-coated slides (0.1 % gelatin, 0.01 % KCr(S04h), dried at room temperature for at least 4 h, and subsequently dehydrated in 50, 80 and 96% ethanol for 3 min each. Hybridizations with fluorescent oligonucleotide probes and DAPI staining were exactly as recently described (ZARDA et aI., 1997). In this protocol, due to a requirement for maximum signal intensity some of the probes are hybridized at stringencies that were slightly lower than those recommended in the original publications. In control experiments with reference strains it was shown that this did not change the probe specificity. Hybridization with enzyme-labeled probes had to be performed at 37°C on samples permeabilized with SDS/DTT and lysozyme (FISCHER et aI., 1995). In order to maintain a sufficient stringency formamide concentrations were raised to 30% (ALF1b, HGC69a, and SRB385) or 40% (EUB338, PLA46, BET42a, GAM42a, and CF319a) with the general assumption that an addition of 1 % formamide was equivalent to a temperature rise of 0.5 °C (AMANN et aI., 1992). After hybridization the slides were washed in the same low salt buffer (64 mM NaCl, 5 mM EDTA and 10 mM TrislHCI, pH 7.2) at 37°C which was sufficient to maintain specificity. Peroxidase activity was visualized by using tyramide signal amplification (Cyanine 3 FISH, NEN, Boston, MA) for 5 min instead of 10 min (SCHONHUBER et aI., 1997). Slides were finally mounted with Citifluor solution and the preparations were examined with a Zeiss Axiophot microscope fitted for epifluorescence with a high-pressure mercury bulb (50 W) and filter sets 02 (Zeiss; G 365, FT 395, LP 420) and HQ-Cy3 (AHF Analysentechnik, Tubingen Germany; G 535150, FT 565, BP 610175).

Bacteria were counted at 1,000x magnification. Fourty fields, selected at random, covering an area of 0.01 mm 2 each were examined from a sample distributed over four circular areas of 53 mm 2 each.

Results and Discussion Analysis of bacterial communities by reassociation and thermal denaturation Reassociation of DNA extracted from bacteria separated from soils Seim, and Hau revealed largely identical Cot curves (Fig. 1). While a log Cot1/2 value of 3.61 was obtained for bacterial DNA from soil Seim, bacterial DNA from soil Hau showed a log Cotll2 value of 3.55. Based on the log Cot1l2 value of 0.72 and a genome size of 4.1 x 10 6 bp of the reference organism E. coli (TORSVIK et aI., 1990), 5,700 different bacterial genomes (g soil [dry wt])-l were calculated for soil Seim, whereas soil Hau harbored 4,900 different bacterial genomes (g soil [dry wt])-l. These values which were in the same range as those for other soils (TORSVIK et aI., 1990; 1996) indicated a similar complexity of the bacterial communities in both soils, which, however, might be due to two communities with quite different structures. The % G+C profiles showed wide % G+C distributions in both soils and indicated differences in bacterial community structure between both soils Seim and Hau (Fig. 2). The bacterial community in soil Seim had a lower median % G+C between 55 to 61 % than that in soil Hau with a median % G+C between 61 to 66%, indicating a relatively higher proportion of bacteria with a high DNA G+C content in soil Hau. A closer assignment to different bacterial populations might be based on the comparative analysis to the % DNA G+C contents of commonly enumerated bacterial genera (HOLBEN and HARRIS, 1995). However, realizing that our current knowledge of bacterial diversity is limited such an assignment remains speculative and can just be a standing point for further molecular studies on the bacterial community structure.

0

c 10 0

:;;;

CO

·u

20

0

(J) (J)

CO

Q) .....

-;J2 0

30 40 50

Seim -1

0

Hau 2

3

-1

o

Log Cot (moles 1-1 sec)

2

3

4

Fig. 1. Cot curves for DNA isolated from soils Seim and Hau. The abscissa gives the logarithm of the initial concentration of single-stranded DNA (in mole-nucleotides per liter) multiplied by time in seconds. The ordinate gives the percent reassociated DNA.

Analysis of Broad-scale Differences

Reassociation analysis and the determination of % G+C distributions of DNA from bacteria in soils depends on the availability of sufficient amounts of highly purified bacterial DNA (TORSVIK et a!., 1990). Though a direct extraction of nucleic acids might yield sufficiently pure DNA for reassociation (RITZ et a!., 1997) and thermal denaturation studies (HOLBEN and HARRIS, 1995; RITZ et a!., 1997), a reliable analysis requires the separation of the bacterial fraction from other organisms prior to lysis (RITZ et a!., 1997; TORSVIK, 1995). In our study, the original extraction protocol worked very well on soil Seim because it was developed and optimized for soils with a high content of organic material. The DNA extraction from the bacterial reaction from soil Hau characterized by high contents of silt and clay, however, required some modifications to the original protocol in order to obtain sufficient amounts of clean DNA. Large differences in the extraction protocols for nucleic acids, e.g. when direct extraction protocols for nucleic acids were compared to nucleic acid extraction from bacterial fractions, resulted in significant shifts in reassociation profiles and % G+C distributions (RITZ et a!', 1997). Although the differences in the protocols used in our study were quite small, we cannot exclude that reassociation analysis and the determination of % G+C distributions of DNA from bacteria in soils Seim and Hau were influenced by the extraction protocol.

Analysis of larger phylogenetic groups of bacteria by hybridization Affiliation with larger phylogenetic groups was analyzed in extracted bacterial fractions and in slurries of the original soils by in situ hybridization with fluorescent oligonucleotide probes and, in addition, in soil slurries with HRP-Iabeled probes and subsequent Cy3-tyramide signal amplification. Total numbers of bacteria in bacterial fractions determined after DAPI-staining were reduced in both soils by approximately 2 orders of magnitude when compared to total numbers of bacteria in homogenized slurries of the original soils (Table 3). Hybridization with fluorescent (Cy3-labeled) probe EUB338 enabled to detect approx. 40% of the DAPI-stained bacteria in both bacterial fractions and in soil slurries of

583

soils Seim and Hau (Table 3). While fluorescent probes could be used without initial permeabilization treatments (ZARDA et a!., 1997), the detection protocol with HRPlabeled probes including Cy3-tyramide signal amplification required pretreatments to increase the permeability of the cells for probes due to the fairly large size of the HRP-labeled probes (molecular weight of approx. 50,000) (AMANN et a!., 1992). Without pretreatments, detectability values for bacteria in soils Seim and Hau were only 32:!: 3% and 10 :!: 2%, respectively. After SDS/ DTT and subsequent lysozyme pretreatment (ZARDA et a!., 1997), however, the HRP-based assay enabled us to detect bacteria in soils Seim and Hau in comparable percentages (40 :!: 5% and 34 :!: 11 %, respectively) as with fluorescent probes (45 :!: 7% and 40:!: 10%, respectively) (Table 3). Detectability values could not further be increased by other pretreatments such as those with lysozyme alone, or with HCl and with peracetic acid, without or with subsequent lysozyme pretreatment (ZARDA et a!', 1997). In situ hybridization with group-specific probes indicated only minor differences between both soils, and only small differences between bacterial fractions and bacteria in soil slurries (Table 3). In slurries of both soils, only probes ALFlb, SRB385, and PLA46 hybridized to cells accounting for more than 1 % of the DAPI-stained cells, while numbers obtained after hybridization with probes ARCH915, BET42a, GAM42a, HGC69a, and CF319a were below the detection limit set at d %. These results are in accordance with earlier studies (ZARDA et a!., 1997). In slurry of soil Seim the percentage of cells detected with probe ALFI b (5 :!: 3 %) was slightly lower than that in soil Hau (10 :!: 3 %). In bacterial fractions extracted from soils, detection of cells hybridizing to probes ALFI band SRB385 was reduced to below the detection limit, whereas percentages of cells hybridizing to probe PLA46 were comparable to that in soil slurries (7 :!: 3 %) (Table 3). Similar results were obtained after hybridization with HRP-labeled probes including Cy3tyramide signal amplification (Table 3) although the quantification of bacteria was occasionally influenced by large differences in signal intensities of stained cells (Fig. 3), by the occasionally very intense staining of some cells which altered their microscopic image beyond recogni-

0.04

J

(J)

<

Fig. 2. % G+C distribution in DNA isolated from soils Seim and Hau, respectively, derived by fitting the formula by MANDEL et al. (1970) and differentiating the resultant curves.

0.02

-»

0.00

+==:::::;::;::..............,................- ...............~..--.......,.............,..............,....,........~~

30

35

40

45

50

55

60

% G+C content

65

70

75

80

584

A.

CHATZINOTAS

et al.

Fig. 3. Bacteria detected by HRP-labeled oligonucleotide probes EUB338 (left panel) and HGC69a (right panel) and subsequent Cy3-tyramide signal amplification in non-dispersed slurries of soils Hau (a, b) and Seim (c, d). Bar represents 5 pm.

Analysis of Broad-scale Differences

585

Table 3. Percentage of specific bacterial populations relative to DAPI-stained cells ([g soil dry wtj-l) in soils Seim and Hau after hybridization with probes targeting larger phylogenetic groups (n =40; X ± SD). Probes a

DAPI (xl08) DAPI EUB338 ALF1b BET42a SRB385 PLA46

Seim

Hau

extracted

in situ

in situ with signal amplification b

extracted

in situ

in situ with signal amplification b

9±2 100 40 ± 7 <1 <1 <1 7±3

404 ± 15 100 45 ± 7 5±3 <1 5±5 7±3

394 ± 38 100 40 ± 5 2±1 5±3 8±4 3±1

2±0 100 44 + 8 <1 <1 <1 6±2

450 ± 11 100 40 ± 10 10 ± 3 <1 4±2 7±3

440 ± 68 100 34 ± 11 h 7±3 4±2 7±2 3±2

"detection values with probes ARCH915, GAM42a, HGC69a, and CF319a were always below the detection limit set at <1 %. bafter pretreatment with SDSIDTT and lysozyme (FISCHER et aI., 1995).

tion (Fig. 3), and by problems of visualizing DAPI in cells strongly stained with Cy3-tyramide. These results do not indicate major systematic biases in the extraction efficacies for the larger phylogenetic groups from both soils. They also did not reveal significant differences in bacterial community composition between soils Hau and Seim which was in contrast to the differing % G+C profiles which indicated a higher proportion of bacteria with a high DNA G+C content in soil Hau. This might have various reasons including the broadness of the probes and the lack of full coverage of the microbial community be in situ hybridization. The latter might be due to the lack of suitable sets of probes targeting all organisms within the domain Bacteria, to a restricted permeability of vegetative and dormant cells for probes or to low levels of rRNA per cell (FISCHER et aI., 1995; ZEPP et aI., 1997). Detection of actinomycetes by in situ hybridization The assumption that methodological drawbacks might be responsible for the contrasting findings with extraction and in situ methods was further evaluated for a bacterial group which might have accounted for the different % G+C distributions in soils Seim and Hau, the Gram-positive bacteria with a high DNA G+C content. In contrast to in situ hybridization where counts with probe HGC69a always remained below the detection limit, dot blot hybridization with probe HGC69a on extracted DNA yielded 12 ± 3% and 15 ± 4% of the signals obtained with probe EUB338 for soils Seim and Hau, respectively. The failure of in situ hybridization could have been due to the fact that spores of Gram-positive bacteria with a high DNA G+C content are often impermeable for fluorescent probes (HAHN and ZEYER, 1994), whereas vegetative cells often exhibit only low probe conferred signals which may be quenched by autofluorescent signals of the soil matrix (HAHN et aI., 1992). A further problem in the analysis of Gram-positive bacteria with a high DNA G+C content might be their filamentous morphology

which could render them vulnerable for the ultrasonic treatment applied prior to in situ detection. In situ hybridization with fluorescent probes on non-dispersed soil slurries of soil Hau indeed revealed the presence of many filaments attached to soil aggregates (data not shown). Filaments were not detected in soil Seim. Detectability, however, was hampered by the low intensities of probeconferred signals which were often quenched by autofluorescence of the soil matrix. Hybridization with HRP-Iabeled probes including Cy3-tyramide signal amplification resulted in an increase of signal that enabled clear visualization of bacteria attached to organic material or to soil aggregates due to the increase in probe-conferred signal intensity (Fig. 3). Using the Cy 3-tyramide signal amplification with probe HGC69a different morphotypes were detected in non-dispersed slurries of both soils, coccoid cells in soil Seim and coccoid and filamentous cells in soil Hau (Fig. 3). The probe HGC69a-positive filamentous bacteria in the soil Hau could be actinomycetes which are well known for their high DNA G+C content. The destruction of this population by the ultrasonication prior to the in situ hybridization could therefore explain why we did not see differences in the microbial community composition as assessed by FISH whereas these were detected in the % G+C profiles.

Conclusions The results of our study demonstrate broad-scale differences in soil microbial community structure in soils Seim and Hau. Only the combination of different techniques enabled us to detect the methodologically introduced artifact of lack of in situ detection of filamentous bacteria of the Gram-positive bacteria with a high DNA G+C content in soil Hau. Future studies on microbial community structure in soils should therefore be based on multi-level approaches which must also include a characterization of populations with higher resolution, e.g., on the genus or species level. These studies might include DGGE or TGGE analysis (FELSKE et aI., 1997;

586

A. CHATZINOTAS et al.

HEUER et al., 1997; KOWALCHUK et al., 1997) or the application of genus- and species-specific hybridization probes which construction could be based on the data retrieved from soil16S rRNA gene libraries (BORNEMAN et al., 1996; FELSKE et al., 1997; NOSSLEIN and TIEDJE, 1998; ZHOU et al., 1997).

Acknowledgements This work was supported by grants from the Swiss National Science Foundation (Priority Program Biotechnology), the Swiss Federal Office of Environment, Forests and Landscape (BUWAL) and the EU (contracts BI02-CT94-3098 and EV5VCT94-0415). The authors are indebted to Svein Nordland and 0ivind Enger for their support in the analysis of thermal denaturation data.

References AMANN, R. 1., BINDER, B. J., OLSEN, R. J., CHISHOLM, S. W., DEVEREUX, R., STAHL, D. A.: Combination of 16S rRNA-targeted oligonucleotide probes with flow cytometry for analyzing mixed microbial populations. Appl. Environ. Microbiol. 56, 1919-1925 (1990a). AMANN, R. 1., KRUMHOLZ, L., STAHL, D. A.: Fluorescentoligonucleotide probing of whole cells for determinative, phylogenetic, and environmental studies in microbiology. ]. Bacteriol. 172,762-770 (1990b). AMANN, R. 1., ZARDA, B., STAHL, D. A., SCHLEIFER, K.-H.: Identification of individual prokaryotic cells by using enzyme-labeled, rRNA-targeted oligonucleotide probes. Appl. Environ. Microbiol. 58, 3007-3011 (1992). BORNEMAN, J., SKROCH, P. W., O'SULGIVAN, K. M., PALUS, J. A., RUMjANEK, N. G., JANSEN, ]. L., NIENHUIS, J., TRIPLETT, E. W.: Molecular microbial diversity of an agricultural soil in Wisconsin. Appl. Environ. Microbiol. 62, 1935-1943 (1996). BOTH, G.]., GERARDS, S., LAANBOEK, H.].: Most probable numbers of chemolithoautotrophic nitrite-oxidizing bacteria in well in well drained grassland soils: Stimulation by high nitrite concentrations. FEMS Microbiol. Ecol. 74, 287-294 (1990). BUYER, ]. S., DRINKWATER, L. E.: Comparison of substrate utilization assay and fatty acid analysis of soil microbial communities.]. Microbiol. Meth. 30, 3-11 (1997). CAVIGELLI, M. A., ROBERTSON, G. P., KLUG, M. J.: Fatty acid methyl ester (FAME) profiles as measures of soil microbial community structure. Plant Soil 170, 99-113 (1995). CHURCH, G. M., GILBERT, W.: Genomic sequencing. Proc. Nat!. Acad. Sci. USA 81, 1991-1995 (1984). COLEMAN, D. c., ODUM, E. P., CROSSLEY Jr., D. A.: Soil biology, soil ecology, and global change. BioI. Ferti!. Soils 14, 104-111 (1992). CRESSWELL, N., SAUNDERS, V. A., WELLINGTON, E. M. H.: Detection and quantification of Streptomyces violacelatus plasmid DNA in soil. Lett. Appl. Microbiol. 13, 193-197 (1991). CROMBACH, W. H.].: DNA base composition of soil arthrobacters and other coryneforms from cheese and sea fish. Antonie van Leeuwenhoek 38,105-120 (1972). FELSKE, A., RHEIMS, H., WOLTERINK, A., STACKEBRANDT, E., AKKERMANS, A. D. L.: Ribosome analysis reveals prominent activity of an uncultured member of the class actinobacteria in grassland soils. Microbiology 143, 2983-2989 (1997). FISCHER, K., HAHN, D., AMANN, R. 1., DANIEL, 0., ZEYER,].: In situ analysis of the bacterial community in the gut of the

earthworm Lumbricus terrestris L. by whole-cell hybridization. Can. J. Microbial. 41, 666-673 (1995). FISCHER, K., HAHN, D., HONERLAGE, W., SCHON HOLZER, E, ZEYER, ].: In situ detection of spores and vegetative cells of Bacillus megaterium in soil by whole cell hybridization. System. Appl. Microbiol. 18,265-273 (1995). GRIFFITHS, B. S., DIAZ-RAVINA, M., RITZ, K., McNICOL, ]. W., EBBLEWHITE, N., BAATH, E.: Community DNA hybridisation and % G+C profiles of microbial communities from heavy metal polluted soils. FEMS Microbial. Ecol. 24, 103-112 (1997). GRIFFITHS, B. S., RITZ, K., GLOVER, L. A.: Broad-scale approaches to the determination of soil microbial community structure: Application of the community DNA hybridization technique. Microb. Ecol. 31,269-280 (1996). HAHN, D., AMANN, R. 1. LUDWIG, W., AKKERMANS, A. D. L., SCHLEIFER, K.-H.: Detection of microorganisms in soil after in situ hybridization with rRNA-targeted, fluorescently labelled oligonucleotides. ]. Gen. Microbiol. 138, 879-887 (1992). HAHN, D., ZEYER, ].: In situ detection of bacteria in the environment. pp. 137-148. In: Beyond the Biomass (RITZ, K., DIGHTON, J., GILLER, K. E., eds.), John Wiley & Sons, Chichester, UK (1994). HEUER, H., KRSEK, M., BAKER, P., SMALLA, K., WELLINGTON, E. M. H.: Analysis of actinomycete communities by specific amplification of genes encoding 16S rRNA and gel-electrophoretic separation in denaturing gradients. Appl. Environ. Microbiol. 63, 3233-3241 (1997). HOLBEN, W. E., HARRIS, D.: DNA-based monitoring of total bacterial community structure in environmental samples. Mol. Ecol. 4, 627-631 (1995). HONERLAGE, W., HAHN, D., ZEYER, ].: Detection of mRNA of nprM in Bacillus megaterium ATCC 14581 grown in soil by whole cell hybridisation. Arch. Microbiol. 163, 235-241 (1995). HUGHES, D. W., GALAU, G. A.: Preparation of RNA from cotton leaves and pollens. Mol. BioI. Reporter 6, 253-257 (1988). JACOBSEN, C. S., RASMUSSEN, O. E: Development and application of a new method to extract bacterial DNA from soil based on separation of bacteria from soil with cation-exchange resin. Appl. Environ. Microbiol. 58, 2458-2462 (1992). KOWALCHUK, G. A., STEPHEN, J. R., DE BOER, W., PROSSER, ]. 1., EMBLEY, T. M., WOLDENDORP, ]. W.: Analysis of ammoniaoxidizing bacteria of the beta subdivision of the class proteo bacteria in coastal sand dunes by denaturing gradient gel electrophoresis and sequencing of PCR-amplified 16S ribosomal DNA fragments. Appl. Environ. Microbiol. 63, 1489-1497 (1997). LIESACK, W., STACKEBRANDT, E.: Occurrence of novel groups of the domain bacteria as revealed by analysis of genetic material isolated from an Australian terrestrial environment. ]. Bacterial. 174,5072-5078 (1992). MANDEL, M., IGAMBI, L., BERGENDAHL, J., DODSON Jr., M. L., SCHELTGEN, E.: Correlation of melting temperature and caesium chloride buoyant density of bacterial deoxyribonucleic acid. J. Bacteriol. 101,333-338 (1970). MANZ, W., AMANN, R., LUDWIG, W., VANCANNEYT, M., SCHLEIFER, K.-H.: Application of a suite of 16S rRNA-specific oligonucleotide probes designed to investigate bacteria of the phylum Cytophaga-Flavobacter-Bacteroides in the natural environment. Microbiology 142, 1097-1106 (1996). MANZ, W., AMANN, R., LUDWIG, W., WAGNER, M., SCHLEIFER, K.-H.: Phylogenetic oligodeoxynucleotide probes for the major subclasses of proteobacteria: Problems and solutions. System. Appl. Microbiol. 15,593-600 (1992).

Analysis of Broad-scale Differences MASSOL-DEYA, A., WELLER, R., RIOS-HERNANDEZ, L., ZHOU, ]. Z., HICKEY, R. E, TIEDJE, ]. M.: Succession and convergence of biofilm communities in fixed-film reactors treating aromatic hydrocarbons in groundwater. App\. Environ. MicrobioI. 63,270-276 (1997). MYROLD, D. D., HILGER, A. B., Huss-DANELL, K., MARTIN, K. J.: Use of molecular methods to enumerate Frankia in soil. pp. 127-136. In: Beyond the Biomass (RITZ, K., DIGHTON, J., GILLER, K. E., eds.), John Wiley & Sons, Chichester, UK (1994). NEDWELL, D. B., GRAY, T R. c.: Soils and sediments as matrices for microbial growth. Symp. Soc. Gen. Microbiol. 41,18-54 (1987). NEEF, A., AMANN, R., SCHLESNER, H., SCHLEIFER, K.-H.: Monitoring a widespread bacterial group: in situ detection of planctomycetes with 16S rRNA-targeted probes. Microbiology (in press). NUSSLEIN, K., TIEDJE, J. M.: Characterization of the dominant and rare members of a young Hawaiian soil bacterial community with small-subunit ribosomal DNA amplified from DNA fractionated on the basis of its guanine and cytosine composition. Appl. Environ. Microbiol. 64, 1283-1289 (1998). PARKIN, T. B.: Spatial variability of microbial processes in soila review. J. Environ. Qual. 22, 409-417 (1993). RHEIMS, H ., SPROER, c., RAINEY, EA., STACKEBRANDT, E.: Molecular biological evidence for the occurrence of uncultured members of the actinomycete line of descent in different environments and geographical locations. Microbiology 142,2863-2870 (1996). RICHARD, E, LUSCHER, P. , STROBEL, T.: Physikalische Eigenschaften von Boden der Schweiz. Eidgenossische Anstalt fur das forstliche Versuchswesen, Birmensdorf, Switzerland (1978). RICHAUME, A., STEINBERG, c., JOCTEUR-MoNROZIER, L., FAURIE, G.: Differences between direct and indirect enumeration of soil bacteria: The influence of soil structure and cell location. Soil BioI. Biochem. 25, 641-643 (1993). RITZ, K., GRIFFITHS, B. S., TORSVIK, V. L., HENDRIKSEN, N . B.: Analysis of soil and bacterioplankton community DNA by melting profiles and reassociation kinetics. FEMS Microbiol. Lett. 149, 151-156 (1997). ROLLER, c., WAGNER, M., AMANN, R., LUDWIG, W., SCHLEIFER, K.-H.: In situ probing of Gram-positive bacteria with a high DNA G+C content using 23S rRNA-targeted oligonucleotides. Microbiology 140, 2849-2858 (1994). SAMBROOK, J., FRITSCH, E. E, MANIATIS, T: Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY (1989). SCHONHUBER, W., FUCHS, B., JURETSCHKO, S., AMANN, R.: Improved sensitivity of whole-cell hybridization by the combination of horseradish peroxidase-labeled oligonucleotides and tyramide signal amplification. Appl. Environ. Microbial. 63,3268-3273 (1997). SKINNER, E E., JONES, P. C. T, MOLLISON,]. E. A.: Comparison of a direct- and a plate-counting technique for the quantitative estimation of soil micro-organisms. ]. Gen. Microbiol. 32,261-271 (1952). SMIT, E., LEEFLANG, P., W ERNARS, K.: Detection of shifts in microbial community structure and diversity in soil caused by

587

copper contamination using amplified ribosomal DNA restriction analysis. FEMS Microbiol. Ecol. 23, 249-261 (1997). S0RHEIM, R., TORSVIK, V. L., GOKS0YR, J.: Phenotypical divergences between populations of soil bacteria isolated on different media. Microb. Ecol. 17,181-192 (1989). STAHL, D. A., AMANN, R. 1.: Development and application of nucleic acid probes. pp. 205-248. In: Nucleic acid techniques in bacterial systematics (STACKEBRANDT, E., GOODFELLOW, M., eds.), John Wiley & Sons, Chichester, UK (1991). STOTZKY, G.: Influence of soil mineral colloids on metabolic processes, growth, adhesion, and ecology of microbes and viruses. pp. 305-428. In: Interactions of soils minerals with natural organics and microbes. (HUANG, P. M., SCHNITZER, M., eds.), Soil Science Society of America, Inc., Madison, USA (1986). STREIT, W., BJOURSON, A. J., COOPER, J. E., WERNER, D.: Application of subtraction hybridization for the development of a Rhizobium leguminosarum biovar phaseoli and Rhizobium tropici group-specific DNA probe. FEMS Microbiol. Ecol. 13,59-67 (1993) . TORSVIK, V.: Cell extraction method. pp. 1-15. In: Molecular microbial ecology manual (AKKERMANS, A. D. L., VAN ELSAS, J. D., DE BRUIN, E, eds.), Kluwer Academic Publishers. Dordrecht, The Netherlands (1995). TORSVIK, V., GOKS0YR, J., DAAE, E L.: High diversity in DNA of soil bacteria. Appl. Environ . Microbiol. 56, 782-787 (1990). TORSVIK, v., S0RHEIM, R., GOKS0YR, J.: Total bacterial diversity in soil and sediment communities: A review. J. Industr. Microbiol. 17, 170-178 (1996). TSAI, Y.-L., OLSEN, B.: Rapid method for separation of bacterial DNA from humic substances in sediments for polymerase chain reaction. Appl. Environ. Microbiol. 58, 2292-2295 (1992). WONSCHE, L., BRUGGEMANN, L., BABEL, W.: Determination of substrate utilization patterns of soil microbial communities: An approach to assess population changes after hydrocarbon pollution. FEMS Microbiol. Ecol. 17,295-305 (1995). XIA, X., BOLLINGER, J., OGRAM, A.: Molecular genetic analysis of the response of three soil microbial communities to the application of 2,4-D. Mol. Ecol. 4, 17-28 (1995). ZARDA, B., HAHN, D., CHATZINOTAS, A., SCHONHUBER, W., NEEt~ A., AMANN, R. 1., ZEYER, J.: Analysis of bacterial community structure in bulk soil by in situ hybridization. Arch. Microbiol. 168, 185-192 (1997). ZEPP, K., HAHN, D., ZEYER, J.: Evaluation of a 23S rRNA insertion as target for the analysis of uncultured Frankia populations in root nodules of alders by whole cell hybridization. System. Appl. Microbial. 20, 124-132 (1997). ZHOU, J., DAVEY, M. E., FIGURAS, J. B., RIVKIN A, E., GILICHINSKY, D., TIEDJE, J. M.: Phylogenetic diversity of a bacterial community determined from Siberian tundra soil DNA. Microbiology 143,3913-3919 (1997).

Corresponding author: DITTMAR HAHN, ETH Zurich, Institute of Terrestrial Ecology, Soil Biology, GrabenstraRe 3, CH - 8952 Schlieren, Switzerland. Tel.: 41-1-6 33 60 39; Fax: 41-1-6 331122; e-mail: [email protected]