jurnal ilmu tanah

MICROBIAL ECOLOGY

Microb Ecol (2002) 44:49-58 DOh 10.1007/s00248-001-0042-8

2002 Springer-Verlag New York Inc.

The Diversity and Function of Soil Microbial Communities Exposed to Different Disturbances

A.K, Miiller, 1 K. Westergaard, 1 S. Christensen, 2 S.J. Sorensen 1

1 Department of General Microbiology, University of Copenhagen, Solvgade 83 H, DK-1307 Copenhagen K, Denmark 2 Department of Terrestrial Ecology, University of Copenhagen, Universitetsparken 5, DK-2100 Copenhagen O, Denmark

Received: 2 May 2001; Accepted: 15 October 2001; Online publication: 30 April 2002

IA B ST R A CT

To improve understanding of the relationship between the diversity and function of the soil

ecosystem, we investigated the effect of two different disturbances on soil bacterial communi-

t ies- long-term exposure to the heavy metal mercury and transient exposure to the antibiotic

tylosin. In the mercury-contaminated soil the diversity (Shannon index) was reduced as assessed

from denaturing gradient gel electrophoresis (DGGE) of amplified 16S rDNA sequences from the

soil community DNA and from colony morphology typing of the culturable bacterial population.

However, analysis of the substrate utilization profiles did not reveal any differences in diversity.

In the tylosin-treated soil, DGGE revealed a small difference in the diversity of 16S rDNA

compared to the control soil, whereas analysis of the colony morphology typing or substrate

utilization results did not reveal any differences in diversity. Soil function was also affected by

mercury contamination. The lag time before soil respiration increased following addition of

glucose or alfalfa substrate was longer in the mercury-contaminated soil than in the control soil.

Moreover, it was markedly prolonged in mercury-contaminated soil subjected to heat treatment

prior to substrate addition, thus indicating reduced resistance to a new disturbance in the

mercury-contaminated soil as compared to the control soil. Tylosin treatment did not have any

significant effect on any of the respiration parameters measured, either with or without prior

heat treatment of the soil.

Introduction

There is growing interest in the relat ionships among ec-

osystem diversity, structure, and function, and a number

Correspondence to: S.J. Sorensen; Phone: +45 35 32 20 53; Fax: +45 35 32 20 40; E-maii: [email protected]

of theories have been formulated concern ing how species

diversity is related to ecosystem funct ion [31]. Thus, Na-

eem et al. [39] and T i lman et al. [53] have suggested that

enhanced species d ivers i ty is beneficial to ecosystem

function. In contrast, other authors suggest that the

propert ies of an ecosystem depend more upon the func-

50 A.K. M/iller et al.

tional abilities of particular species than on the total number of species [28, 54, 58].

One aspect of ecosystem function is stability, defined as the system's ability to avoid displacement following a perturbation (resistance) and to return to its former state following a perturbation (resilience) [5]. Stability has formerly and more recently been suggested to correlate positively with system diversity [12, 20, 33, 52], although the validity of this hypothesis is the subject of controversy [22].

Despite the suggestion of Wardle and Giller [57] that the diversity and functional importance of soil organisms present an excellent opportunity to test currently topical aspects of ecological theory, the hypotheses relating ecosystem diversity and function have mainly been developed by plant ecologists, with only a few studies having con- cerned the soil community [21, 35]. Hypotheses from the aboveground communities may not easily be applied to belowground, since there are differences. The microbial diversity in soil is enormous [55], and there may be sub- stantial overlap in function between microbial species [9]. Furthermore, it is likely that microorganisms within a functional group differ in their response to the environment. As microorganisms are fast growing they can quickly fill out empty niches occurring when the environment is changing [19]. These circumstances could create a high degree of stability, but it is unknown what level of diversity is necessary to maintain stability [57].

Investigation of the relationships among diversity, structure, and function poses the problem of how to manipulate communities so as to establish differences in diversity for comparative purposes. One possibility is to construct communities differing in level of diversity by introducing different numbers of species into micro- cosms [34, 35]. However, such experiments almost inevitably involve unrealistically few species within the communities. Moreover, on the microbial level a created community will be unrepresentative of a natural community since only known and culturable organisms can be included. They are unlikely to cover important func- tions in the soil system. Another approach is destructive reduction of species diversity [21, 47]. This inevitably entails selective reduction in diversity, and if species vulnerable to the treatment are removed the results may more reflect the absence of certain key species than a reduction in diversity. However, this is also the case in natural systems, where any given stress always selects for adapted species. The results of studies made using the

destructive approach will gradually become more inter- esting when different means of reducing diversity with different selective properties are compared.

The effect of a disturbance on microbial community function depends on its duration and specificity. After a transient disturbance, system function may eventually return to its former state, whereas a permanent disturbance will result in a new, altered state [46]. Disturbances with a specific mode of action only alter a few groups of organisms, whereas those that act non-specifically affect a wide range of organisms. Investigation of how different kinds of disturbances affect system function will enhance our knowledge of the relationships among diversity, structure, and function.

Heavy metals are highly persistent in the environment and are known to alter soil ecosystem diversity, structure, and function [7, 18]. Antibiotics dispersed into the soil in connection with the application of livestock manure represent transient disturbances that mainly affect specific groups of bacteria. Although the effects of antibiotics in soil have been investigated [6, 16, 29], their effect on the structure and function of the microbial community has not previously been examined.

The diversity of complex soil microbial communities has been studied using new techniques for describing different microbial populations [8, 17, 37, 51, 56]. As no one single method is currently available for exploring the whole bacterial community, a combination of methods is necessary to obtain a detailed view of its structure and diversity.

When characterizing system function a myriad of different processes can be examined. The key soil function is mineral cycling, and one of the most common measures of microbial mineralization is respiration [44].

The aim of the present paper is to investigate how two different disturbances--mercury contamination and tylosin treatment--affect the diversity and function of the soil microbial community. Community diversity was assessed by colony morphology typing, DGGE profiling, and substrate utilization studies. For both soils detailed results from these community analyses have already been published [38, 60], and here we only present diversity as the Shannon Index. Community function was assessed from CO2 production following substrate addition (glucose or alfalfa). In order to investigate whether the resistance of the soil microbial community was compromised by the initial disturbance, the mercury-contaminated and tylosin- treated soils were also subjected to an additional distur-

Disturbed Microbial Communities 51

bance (heat treatment) prior to the measurement of

respiration.

Materials and Methods

Tylosin Treatment

Soil samples were collected from the upper 20 cm of an agricultural sandy soil in Jyndevad, Southern Jutland, Denmark. The mineral fraction of the soil from this location consisted of 4.1% clay (200 ~m). The soil contained 1.2% organic carbon and had a water-holding capacity of 15 g water 100 g-1 of soil. The pH was 6.8 [27]. After sampling the soil was brought to the laboratory and stored at 10C in the dark until required. The soil was mixed, sieved (2 ram), air-dried overnight at room temperature, and transferred to 8 plastic boxes (750 g soil in each). Tylosin-treated soils (four replicate soil boxes) were pre- pared by dissolving tylosin (tylosin tartrate, Sigma) in water and thoroughly mixing the solution into the soil to a final concentration of 2000 ~tg tylosin g-Z dry soil and a water content of 15% (corresponding to the water-holding capacity of the soil). The remaining four soil boxes served as controls, with only water being added. The soil was incubated aerobically at 25C in the dark. On day 60, when the number of CFU in the tylosin-treated soils was no longer significantly different from that in the controls (t-test, p < 0.05), the soil boxes were transferred to 4C. It is unknown whether the proportion of culturable bacteria in the treated and control soil may have differed. The community analyses were made on each of the four tylosin-treated and control boxes of soil.

Mercury-Contaminated Soil

Soil samples were collected from the upper 5-15 cm at a mercury- contaminated site in Assens, Denmark. The contamination with elemental mercury had taken place 14 years prior to sampling. The samples were collected in a distance of 1 m and 19 m from the center of contamination in order to obtain one sample subjected to a high concentration of mercury and one sample free of mercury contamination to serve as a control. The mercury- contaminated soil (pH 7.2) contained 4.3-6.3% organic matter [38] and was sampled at a site devoid of vegetation. The control soil (pH 7.0) contained 6.6-6.9% organic matter [38] and was sampled in a garden. The samples were stored in glass jars at 4C in the dark up to 6 months. The storage at 4C has been shown to decrease the number of viable organisms [48]. The community analyses were made on three replicate soil samples from each collection point.

Quantification of Tylosin by HPLC

One g soil was mixed with 1 mL distilled H20 in an Eppendorf tube. The solution was shaken vigorously for 1 min and cen-

trifuged for 2 min at 10,000 g, and the supernatant was collected in a new tube. The procedure was then repeated twice on the pellet. Finally, the 3 mL supernatant was mixed thoroughly and filtered through a 0.2 ~tm cellulose acetate filter.

Samples were analyzed by HPLC (LKB Bromma, 2248, Phar- macia, Uppsala, Sweden) equipped with an RP C-18 column (5 ~tm particle size) and a variable-wavelength monitor (LKB Bromma, 2248, Pharmacia, Uppsala, Sweden) using the mobile phase as described by Ose and Tonkinson [42]. Peaks were detected at 280 nm. The standard curve was made from analyzing standard solutions of different concentrations of tylosin tartrate in distilled H20.

Quantification of Total and Bioavailable Mercury

The total mercury content of the soil was determined using static headspace analysis as previously described by Kriger and Turner [30] (soil method 2) whereby the Hg present is con- verted to the elemental form and measured on a Hg vapor an- alyzer (Jerome Model 431-X, Arizona Instruments Inc., Phoenix, AZ).

The bioavailable inorganic mercury was measured using a mer-lux biosensor as previously described [45] combined with calculation of the mer-lux expression factors [4]. A standard curve was established using the regression equation for the relationship between the amount of bioavailable mercury and the expression factors obtained from a standard assay (Hg(NO3)2 diluted in distilled H20). A control experiment was performed using a mutant strain constitutively expressing the lux genes [3] to investigate whether inhibition of light development occurred in soil samples.

Bacterial Colony Morphology

Soil was diluted 10-fold in phosphate-buffered saline (PBS; autoclaved solution of 1.44 g L -1 Na2HPO4-2H20, 0.24 g L -~ KH2PO4, 8.00 g L -I NaC1, and 0.20 g L -I KC1 in distilled water) and vortexed at maximum velocity for 1 rain. From this sus- pension, soil dilutions appropriate to ensure 30 to 100 colonies per plate were spread on TSA (Tryptic Soy Agar) plates supple- mented with fungicide (25 ~tg natamycin mL-1). The plates were inspected after 4 days of incubation at 25C. A nutrient-rich medium was used to enhance expression of the differences between colonies [43]. The colonies were grouped into morphotypes on the basis of visual differences according to Smibert and Krieg [50] using characteristics such as colony color, diameter, edge, surface (roughness and shininess), and other special characteristics, e.g., tendency to swarm. Two hundred random colonies were typed on each replicate of the tylosin-treated and control soils while 150 colonies were typed on the mercury- contaminated soil. The number of different colony morphotypes and the number of colonies belonging to each morphotype were used as parameters for further analysis.

52 A.K. Mfiller et al.

DNA Fingerprinting

Microbial community DNA from the soils used in the tylosin experiment was extracted using a bead-beating method described by van Elsas and Smalla [ 11]. Extracts were initially purified with CsC1 and thereafter using the Wizard DNA cleanup system (Promega, Madison, WI) in accordance with the manufacturer's instructions and then stored at -20C. Microbial community DNA from the mercury-contaminated soil was extracted using the bead-beating method (FastDNA SPIN Kit (for soil), Bio 101 Inc., USA) in accordance with the manufacturer's instructions. The extract was further purified by Wizard DNA cleanup system and then stored at -20C.

The community DNA extracted from the soil was amplified using the primer sequences described by Muyzer et al. [37]. These anneal to conserved regions of the 16S rDNA of eubacteria and contain a GC clamp. Amplification was undertaken using the Expand High Fidelity DNA polymerase (Boehringer Mannheim) in accordance with the manufacturer's instructions using 250 nM of each primer. The polymerase was added after a hotstart procedure (94C for 5 min). The PCR was then performed with a Perkin-Elmer 9600 thermocycler using the following cycles: 1 min at 94C, 1 min at 65C, 3 rain at 72C, with a touchdown of 0.5C per cycle for the first 20 cycles. Thereafter followed 10 cycles at the annealing temperature of 55C. The tenth cycle was followed by 7 min at 72C.

After gel electrophoresis (2% w/v agarose gel) of 5 ~L sub- samples of the PCR product the amount of amplified DNA was calculated by comparing band intensities to a standard curve based on intensities from a Low DNA Mass Ladder (Gibco-BRL). Comparisons were performed by image analysis using Quantity One 4.0.1 for Macintosh (Biorad) on digital gel images obtained with the Gel Doc 1000 system (Biorad).

DGGE of the amplified 16S rDNA sequences was performed as described by Muyzer et al. [37] with minor modifications. Equal amounts of DNA were loaded on the gel. Digital image of the gel was obtained and analyzed using image analysis (Quantity One 4.0.1, Bio-Rad). By carefully inspecting lane intensity curves in combination with enlarged images of the lanes bands were detected and quantified (average peak intensity). Background intensity was subtracted (option: Rolling disc, size 9). The number of bands and the intensity of each band were used as parameters for further analysis.

Sole Carbon Source Utilization Profile

Bacterial cells for inoculation of Ecoplates (Biolog Inc., Hayward, CA) were extracted from soil by homogenizing 10 g of soil in 100 mL sterile water for 1 min in a Waring blender followed by I min cooling on ice and further blending for 1 min. The slurry was then centrifuged for 10 min at 1000 g, and the supernatant transferred to a new tube. The pellet was resuspended in 100 mL sterile water, and the blending procedure repeated. The two supernatants were pooled and diluted to obtain a cell density (estimated by direct count after staining with acridine orange) of

approximately 4 x 105 cells mL -1. Extract (125 pl) was inoculated into each well and the plates incubated in the dark on a shaker (150 rpm) at 20C. The OD59s was measured with a microtiter plate reader (EL 340 Biokinetics Reader, Biotek Instruments, Winooski, Vermont, USA). In order to minimize the effects of different inoculation densities the plates were read when the average well color density (AWCD) on a plate was 0.5 _+ 0.1. The data were standardized by subtracting the reading of the blank well from the reading of each substrate-containing well and di- viding this value by the AWCD [18]. Wells with a mean OD < 0 were excluded from further analysis as they do not contribute to the biological information. The number of substrates utilized (well color > 0.1) and the standardized color development of each substrate were used as parameters for further analysis.

Diversity Index and PCA

The results of the three different community analyses were evaluated by determining the Shannon diversity index. The index based on colony morphology, DGGE, or substrate utilization was calculated as

$

Ht = -- ~ Pi In Pi i=1

where Pi is the percentage of (a) the total (S) intensity accounted for by the ith band, (b) the total coloration accounted for by the ith substrate, or (c) the total number of colonies accounted for by the ith morphotype, respectively.

Statistical analysis of the results was performed by analysis of variance. The principal component analysis was based on the correlation matrix using SPSS 6.1 for Macintosh.

Soil Respiration Parameters

We also examined the effect of tylosin-treatment and mercury- contamination on soil respiration following the addition of substrate. Moreover, to investigate the ability of the treated and contaminated soil to resist a further disturbance, we measured respiration on soil samples that had been heated in watertight plastic tubes at 50C for 12 h prior to the addition of the substrates.

The two substrates used were (a) glucose plus ammonium nitrate (40 mg glucose g-1 wet soil and 11.4 mg NH4NO3 g-~ wet soil) and (b) ground alfalfa (5 nag alfalfa g-1 wet soil). The substrate, 2.5 g soil and 10 mL sterile water were added to flasks, which were then sealed and incubated at 15C on a shaker.

The headspace CO2 from the glucose-amended soil samples was measured every second hour over a 48-hour period, while that from the alfalfa-amended soil samples was measured every 4 to 8 hours over a 72-hour period. The gas samples were analyzed on a gas chromatograph (Mikrolaboratoriet, Aarhus, DK) with a TC detector after separation on a 1.8 m 3 mm Porapak Q column.


The various respiration parameters determined were analyzed statistically by analysis of variance (SAS 6.12 for Windows).

Results

Quantification of Tylosin

The added tylosin was recovered completely on day 1-3 of

incubation (data not shown). On day 5, the first of 4 un-

identified peaks appeared on the chromatogram, probably

representing tylosin degradation products. After day 13,

tylosin could not be detected, and by day 17, all degra-

dation products had disappeared. None of the above-

mentioned compounds were detected in the control soils.

Quantification of Mercury

The total mercury concentration in the soil collected 1 m

from the centre of contamination was 511 ~tg Hg g-1 dry

soil, of which only 0.043% was bioavailable.

The total mercury concentration in the control soil was

6.8 ~tg Hg g-1 dry soil, with no bioavailable mercury being

present (detection limit: 35 ~g Hg g-1 soil). Even though

the total mercury concentration was higher than expected,

it was nevertheless considerably lower than in the soil

close to the center of contamination.

Diversity of the Bacterial Community

The bacterial diversity in the various soils based on

colony morphotypes, DGGE, and substrate utilization is

presented in Table 1. More results from the community

analysis other than the diversity index has previously

been published [38, 60]. Among a variety of different

diversity indices [32], the Shannon index is one of the

most widely used. However, this index requires clearly

Table 1. Bacterial diversity (mean + SE) in the different soils estimated as the Shannon index based on colony morphotypes, DGGE bands and substrate utilization (Ecoplates)

The Shannon diversity index based on:

Colony Substrate Treatment morphotypes DGGE bands utilization

Control 2.48 + 0.04 3.75 + 0.01 2.62 + 0.05 Tylosin 2.54 + 0.05 3.65 + 0.03* 2.54 _+ 0.07 Control 2.66 + 0.10 3.83 + 0.02 2.66 _+ 0.07 Mercury 2.10 + 0.15" 3.48 + 0.02*** 2.64 + 0.08

Statistically significant (ANOVA) effects of tylosin treatment and mercury contamination are indicated as *p < 0.05 and ***p < 0.001.

defined species and a distinct identification of individuals

[59], requirements that are not met when dealing with

bacteria. Estimation of the Shannon index as performed

in this investigation thus provides composite values for

the number and distribution of morphotypes, DGGE

bands, and substrates utilized that represent different

aspects of bacterial diversity, but not necessarily at a

species level.

The Shannon index values given in Table 1 are the

average values for the replicates of each soil. Although it

has been questioned whether it is statistically correct to

apply parametric statistics to the Shannon index [14, 32],

the issue is considered to be of minor importance in the

present study since the differences between the soils were

the same irrespective of whether the index was calculated

for all the separate replicates (data not shown), for the

pooled replicates (data not shown), or as the average of the

replicates (Table 1).

The Shannon indices calculated for the three methods

of describing the microbial community are not directly

comparable since one of the parameters determining the

index is the number of variables (in this case either the

number of bands, substrates or morphotypes), which of

course differs between the three methods.

There was no statistically significant difference in di-

versity between the tylosin-treated and control soils after 2

months of incubation as estimated on the basis of colony

morphotypes (iV = 0.36) and substrate utilization (p =

0.37), although there was a difference in diversity as esti-

mated on the basis of the DGGE bands (p = 0.02), mainly

due to the lower number of DGGE bands in the treated soil

(data not shown).

With the mercury-contaminated soil, diversity differed

significantly between the contaminated and control soils

as estimated on the basis of both colony morphotypes (p =

0.04) and DGGE bands (p = 0.0003). In both cases, bac-

terial diversity was lower in the mercury-contaminated

soil than in the control soil, mainly because of differences

in the number of morphotypes and the number of DGGE

bands (data not shown). Diversity as estimated on the

basis of substrate utilization did not differ between the two

soils (p = 0.85).

Soil Respiration Parameters

The respiration parameters of soil amended with glucose

plus ammonium nitrate was the substrate-induced respi-

ration (SIR) rate, the lag time before the exponential in-

54 A.K. M/flier et al.

crease in respi rat ion rate and the specific respirat ion in-

crement dur ing the exponent ia l phase [40]. As no expo-

nential increase in resp i rat ion occurred after the addit ion

of alfalfa, only two respi rat ion parameters were deter-

mined: the t ime interval before CO2 product ion started to

increase (respirat ion delay) and the respirat ion rate. The

latter was higher than the basal respirat ion rate.

Table 2 summar izes the results of the respirat ion

measurements on ty los in- t reated and mercury-contami-

nated soil amended with glucose plus ammonium nitrate

or amended with alfalfa. The SIR rate tended to be higher

(p = 0.09) and specific respi rat ion increment lower (p =

0.09) in the ty los in- t reated soil than in the contro l soil,

thus indicat ing a h igher microbia l b iomass with a slower

growth in the ty los in- t reated soil. The higher SIR rate in

the ty los in-treated soil cor responds to est imates of the

microbia l b iomass made using the fumigat ion-extract ion

method (data not shown). The lag t ime before the expo-

nential increase in resp i rat ion rate was not signif icantly

affected by the presence of tylosin (p = 0.45). Heat treat-

ment pr ior to the resp i rat ion measurements tended to

increase the lag t ime (p = 0.08) and the specific respirat ion

increment (p = 0.09), but with no differences between the

ty los in-treated soil and contro l soil. The SIR rate was not

affected by the heat t reatment (p = 0.23). Tylosin treat-

ment had no effect on the respi rat ion delay and respira-

t ion rate in soil amended with alfalfa. Heat t reatment

considerably increased the delay (p = 0.0001), but to the

same extent in both ty los in- t reated and control soils. The

response of the two soils to heat treatment did differ in

regard to respirat ion rate, however (p = 0.04). Thus res-

pirat ion rate increased fol lowing heat treatment in the

control soil but remained unchanged in the tylosin-treated

soil.

SIR rate in the mercury -contaminated soil did not

differ from that in the contro l soil (p = 0.76), whereas

the lag time before the exponent ia l increase in respira-

t ion rate was signif icantly longer (p = 0.004) and the

specific respirat ion increment lower (p = 0.05). The re-

sponse of the two soils to heat exposure differed. Thus

the lag t ime increased in the contaminated soil but de-

creased in the control soil (p = 0.05). With both soils,

the specific respirat ion increment was significantly re-

duced by the heat t reatment (p = 0.01). The respirat ion

delay following the addi t ion of alfalfa was significantly

longer in the mercury -contaminated soil than in the

control soil (p = 0.0001), a difference that was further

enhanced by heat treatment. Thus whereas there was no

respirat ion delay in either the unheated or heated con-

trol soil, the delay in the mercury-contaminated soil

doubled from an average of 22.7 h without heat treat-

ment to 44 h fol lowing heat treatment. When decom-

posit ion started in the mercury -contaminated soil, the

respirat ion rate was signif icantly higher than in the

Table 2. Soil respiration parameters (mean + SE) calculated after addition of glucose (SIR rate, lag time and specific respiration increment [Spec. resp. inc.]) or alfalfa (respiration delay and respiration rate) for tylosin-treated soil and mercury-contaminated soil with (+) and without ( - ) heat treatments

Addition of glucose and ammonium nitrate Addition of alfalfa

SIR Spec. resp. inc. Respiration rate (~tg CO2 g-t (p-g CO2 g~l Respiration (~tg CO2 g~

Soil Heat dw soil h -1) Lag time (h) dw soil h- ) delay (h) dw soil h- )

Control - 3.36 + 0.65 16.5 + 4.0 0.051 Control + 2.07 + 1.47 23.3 + 4.6 0.062 Tylosin - 5.74 + 1.08 18.5 + 3.9 0.039 Tylosin + 4.02 + 1.37 28.7 _+ 4.2 0.051 Tylosin treatment + Heat treatment + Interaction Control - 6.94 +0.55 24 + 4.2 0.063 Control + 5.76 +0.96 16 + 0 0.053 Mercury - 6.03 + 1.66 32.7 + 7.5 0.057 Mercury + 5.73 + 2.11 47.3 +_ 0.6 0.041 Mercury contamination ** Heat treatment Interaction +

__+ 0.006 0 __+ 0 6.60 __+ 0.40 + 0.007 22.5 -+ 2.5 7.87 __+ 0.15 __+ 0.002 0 + 0 7.26 __+ 0.48 __+ 0.007 24 __+ 1.6 7.00 + 0.21 + + ***

+ 0.004 0 + 0 6.66 + 0.27 + 0.001 0 + 0 6.11 + 0.13 + 0.001 22.7 + 2.7 6.91 + 0.11 + 0.005 44 + 0 7.45 + 0.27 + : ,g** *

Statistically significant (ANOVA) effects of tylosin treatment or mercury contamination, heat treatment and the interaction between treatments are indicated as +, p > 0.1; *, p > 0.05; **, p < 0.01; and * * *, p < 0.001.


control soil (p = 0.01). Moreover, the soils responded

differently to heat treatment (p -- 0.04), the respiration rate increasing in the mercury-contaminated soil but

decreasing in the control soil.

Discussion

The colony morphotype data enabled us to detect, in the case of mercury contamination, disturbance-induced reduction in diversity. Even though these findings are based upon a subjective determination of morphotypes, investigation of the diversity within morphotypes indicates that morphotypes are homogeneous [25, 60]. The culturable part of the microbial community represents only a minor fraction of the total community [2, 15] and only of the dominant bacteria. The diversity of colony morphotypes thus under- estimates the total diversity [1]. Moreover, it is uncertain whether it is representative of the whole community or even of the active part of the community [26, 51, 63].

Analysis of the DGGE profiles revealed reduced diversity in both the tylosin-treated and mercury-contaminated soils. This is in agreement with the findings based on colony morphology typing in the case of the mercury- contaminated soil, but not for tylosin treatment, where analysis of the colony morphotype data did not reveal any differences in diversity. Compared to the colony morphology approach, DGGE may include nonculturable bacteria. However, as the profiles contain a maximum of about 50 bands, they clearly do not include all species. It is important to bear in mind that some bacteria produce more than one band on the DGGE [41], and that DNA sequences from different bacteria can have identical melting properties and hence do not separate on DGGE. Furthermore, it is doubtful whether extracted and amplified DNA reflects the quantitative abundance of the species [13, 62]. Whether the bands represent the most abundant species, the most easily extractable species, the most active species, or a combination of all these groups is uncertain. Nevertheless, DGGE seems to be the most sensitive

method for detecting differences in community diversity. While differences in diversity could be detected from

the colony morphotype and DGGE data, no differences in diversity could be detected from the substrate utilization data. The latter reflects a different level of community structure since the utilization of a specific substrate is not related to specific bacterial types, as is the case for colonies and to some extent for DGGE bands. Substrate utilization

is attributable to more than one bacterial type [49, 61] and the degree of substrate utilization is not necessarily quantitatively related to the number of utilizing bacteria [23]. Because coloration in the analysis depends on bac-

terial growth, it only describes the potential activity of that fraction of the microbial community able to grow in the wells [61[, and not necessarily the potential of the nu-

merically dominant bacteria in the innoculum [49]. We wanted to relate the changes in the diversity of the

microbial community to functional parameters in the soils,

including the ability of the soil system to cope with an additional disturbance (transient heating) that is very different from the initial disturbance (tylosin treatment or mercury contamination). The heat treatment disturbed the soil system by changing one or more respiration parameters in all soils. Furthermore, it must have selected for different

bacteria than tylosin treatment and mercury contamination since the response of these soils was equivalent to or greater than that of the respective control soils.

Mercury contamination more strongly affected soil function than did tylosin treatment. The respiration parameters did not differ significantly between the tylosin- treated soil and control soil in the present study, although inhibition of soil respiration 7 weeks following addition of tylosin has previously been reported [6]. The effect of heat treatment seemed to be more or less the same for the tylosin-treated and control soils, i.e., prolongation of the lag time and respiration delay, and an increase in the

specific respiration increment. The most striking differences in the respiration meas-

urements were the enhanced lag time and respiration delay in the mercury-contaminated soil. Furthermore, the mercury-contaminated soil was more sensitive to the additional disturbance than the control soil since the decomposition of both glucose and alfalfa was inhibited longer in the contaminated soil than in the control soil following heat treatment. The lag time [40] or the corresponding decomposition time [24] have been shown to be sensitive indicators of heavy metal contamination, increasing with increasing contamination. Lag time has been interpreted as reflecting physiological state, as shown for E. coil in pure culture studies [36] and for soil microbial communities [40]. The lag time could also be attributable

to the direct toxic effect of the heavy metals on the microorganisms or in some cases to interactions between the heavy metal and the involved substrate or enzyme [10]. Also, the size of the microbial biomass could influence the delay of respiration [40].

56 A.K. M/iller et al.

In the present study, exposure to mercury affected the respiration delay before commencement of alfalfa decomposition in the same manner but to a greater extent than the lag time before growth on glucose. This is in agreement with the finding of Doelman and Haanstra [10] that lag time is particularly prolonged in the case of ma- terial that does not readily decompose. It is possible that the respiration delay reflects the same thing as the lag time. However, since the degradation of more complex substrates is likely to depend on a consortium of microorganisms, this could lead to the interpretation that the diversity of the microorganisms is important for the du-

ration of the respiration delay. Prolongation of the respiration delay could therefore be due to the changes in diversity, changes in the physiological state of the bacteria, a direct toxic effect on the bacteria, or a combination of all

factors. It is obvious that other factors than diversity influence the function of the soil community since the

mercury-control soil and the tylosin-control soil differed in heat tolerance but not in diversity. Instead the difference in stability could be due to, for example, the composition of the community or a lower microbial biomass in the tylosin-control soil compared to the mercury-control soil as indicated by the SIR values.

The soils differing most in diversity with respect to both the culturable and nonculturable part of the microbial population also exhibited the greatest differences in sensitivity to the additional disturbance. This could be interpreted as supporting the hypothesis of a link between diversity and resistance of the ecosystem. The same has been demonstrated in another soil community, where re-

duced diversity caused by CHC13 fumigation resulted in decreased resistance and resilience to additional disturbances [21]. Such an interpretation should be made with

caution, however, as the two disturbances (tylosin treatment and mercury contamination) differ considerably in

terms of duration and target specificity. Thus whereas tylosin disappears, the organisms are permanently exposed to the mercury. Even though the microbial population adapts to the mercury contamination, the organisms

may expend considerable energy to maintain their tolerance, thereby potentially diminishing their resistance to an additional disturbance. Since the tylosin disappeared 6 weeks before the measurements of diversity and function were made, tylosin does not act as a stress factor on the organisms in the present study. Another explanation for the reduced resistance of the mercury-contaminated soil is that organisms selected by the mercury exposure could be

more heat-sensitive than the general population or the

tylosin-resistant population. Tylosin was chosen to manipulate the soil system in

this experiment because it is specifically toxic to bacteria and presumably only has secondary effects on other groups of soil organisms. On the other hand, the effects of mercury are broader, also affecting the fungi, flora, and fauna--organisms with important functional roles in

the soil. For instance, mercury had very easily detectable effects on the vegetation at the site where the soil samples were collected, and soil structure clearly differed between the soils. The absence of vegetation and re-

sultant changes in substrate levels and microniches in the soil probably alter the diversity, structure, and

function of the soil community. The present study thus shows that functional per-

formance was generally unaltered by the transient disturbance, whereas it was slower and more sensitive to an additional disturbance in the presence of a permanent disturbance. As regards the relationship between system diversity and function, we thus found that if diversity was reduced, system performance was also altered. If there was no diversity reduction, in contrast, system function was unaffected. Whether this correlation is attributable to a relationship between diversity and function or to the use

of different disturbances remains to be elucidated.

Acknowledgments

This research was supported by the "Centre for biological processes in contaminated soil and sediment" (BIOPRO) (www.biopro.dk) under The Danish Environmental Re- search Programme. Special thanks to Dr. Jaap Bloem and

An Vos for their hospitality and assistance with the DGGE, and to Susanne Eriksen for help with quantification of the

tylosin.

References

1. Atlas RM (1984) Diversity of microbial communities. In: KC Marshal (ed) Advances in Microbial Ecolog. vol 7, Plenum, New York, pp 1-47

2. Bakken LR (1985) Separation and purification of bacteria from soil. Appl Environ Microbiol 49:1482-1487

3. Barkay T, Gillman M, Turner RR (1997) Effects of dissolved organic carbon and salinity on bioavailability of mercury. App1 Environ Microbiol 63:4267-4271


4. Barkay T, Turner RR, Rasmussen LD, Kelly CA, Rudd JWM (1998) Luminescence facilitated detection of bioavailable mercury in natural waters. In: RA LaRossa (ed) Methods in Molecular Biology, vol 102: Bioluminescence Methods and Protocols. Humana Press Inc, Totowa; NJ, pp 231-246

5. Begon M, Harper JL, Townsend CR (1996) Ecology, 3rd ed. Blackwell Science Ltd

6. Bewick MWM (1978) Effect of tylosin and tylosin fermen- tation waste on microbial activity of soil. Soil Biol Biochem 10:403-407

7. Bfifith E (1989) Effects of heavy metals in soil on microbial processes and population. Water Air Soil Pollut 47:335- 379

8. Bfifith E, Diaz-Ravina M, Frostegfird ,~, Campbell CD (1998) Effect of metal-rich sludge amendments on the soil microbial community. Appl Environ Microbiol 64:238-245

9. Chapin FS III, Walker BH, Hobbs RJ, Hooper DU, Lawton JH, Sala OE, Tilman D (1997) Biotic control over the func- tions of ecosystems. Science 277:500-504

10. Doelman P, Haanstra L (1979) Effects of lead on the decomposition of organic matter. Soil Biol Biochem 11:481- 485

11. van Elsas JD, Smalla K (1995) Extraction of microbial community DNA from soils. In: A Akkermans, JD van Elsas, FJ de Bruin (eds), Molecular Microbial Ecology Manual. Kluwer Academic Publishers, pp 1-11

12. Elton CS (1958) The Reasons for Conservation. The Ecology of Invasions by Animals and Plants. Chapman & Hall, London, pp 143-153

13. FarreUy V, Rainey FA, Stackebrandt E (1995) Effect on ge- nome size and rrn gene copy number on PCR amplification of 16S rRNA genes from a mixture of bacterial species. App1 Environ Microbiol 61:2798-2801

14. Finlay LB, Maberty SC, Cooper JI (1997) Microbial diversity and ecosystem function. Olkos 80:209-213

15. Fowler J, Cohen L (1990) Practical Statistics for Field Biol- ogy. John Wiley & Sons Ltd, Chicester, UK

16. Fa~gri A, Torsvik VL, Goks6yr J (1977) Bacterial and fungal activities in soil: Separation of bacteria and fungi by a rapid fractionated centrifugation technique. Soil Biol Biochem 29: 75-89

17. Gavalchin ], Katz SE (1994) The persistence of fecal-borne antibiotics in soil. J AOAC Int 77:481-485

18. Garland JL, Mills AL (1991) Classification and characteri- zation of heterotrophic microbial communities on the basis of patterns of community-level sole-carbon-source utilization. App1 Environ Microbiol 57:2351-2359

19. Giller KE, Witter E, McGrath SP (1998) Toxicity of heavy metals to microorganisms and microbial processes in agricultural soils: A review. Soil Biol Biochem 30:1389-1414

20. Griffiths BS, Ritz K, Wheafley RE (1997) Relationship between functional diversity and genetic diversity in complex microbial communities. In: H Insam, A Rangger (eds) Microbial Communities, Functional versus Structural Ap- proaches. Springer-Verlag, Berlin, pp 1-9

21. Griffiths BS, Ritz K, Bardgett RD, Cook, R, Christensen S, Ekelund F, Sorensen SJ, B~fith E, Bloem J, De Ruiter P, Dolfin J, Nicolardot B (2000) Stability of soil ecosystem processes following the experimental manipulation of soil microbial community diversity. Oikos 90:279-294

22. Grime JP (1997) Biodiversity and ecosystem function: The debate deepens. Science 277:1260-1261

23. Haack SK, Garchow H, Klug MJ, Forney LJ (1995) Analysis of factors affecting the accuracy, reproducibility, and interpretation of microbial community carbon source utilization patterns. Appl Environ Microbiol 61:1458-1468

24. Haanstra L, Doelman P (1984) Glutamic acid decomposition as a sensitive measure of heavy metal pollution in soil. Soil Biol Biochem 16:595-600

25. Haldeman DL, Amy PS (1993) Diversity within a colony morphotype: Implications for ecological research. Appl En- viron Microbiol 59:933-935

26. Hattori T, Mitsui H, Hattori R, Shlkano S, Gorlach K, Ka- sahara Y, E1-Beltagy A (1997) Analysis of the bacterial community according to colony development on solid medium. In: H Insam, A Rangger (eds) Microbial Communi- ties, Functional versus Structural Approaches. Springer- Verlag, Berlin, pp 229-236

27. Heidmann T (1989) Startkarakterisering af arealer til sys- temforskning. IV. Resultater Ira arealet ved Jyndevad. Re- port No. S 2021, Danish Institute of Plant and Soil Science.

28. Hooper DU and Vitousek PM (1997) The effects of plant composition on ecosystem processes. Science 277:1302- 1305

29. Ingham ER, Coleman DC (1984) Effects of streptomycin, cycloheximide, fungizone, captan, carbofuran, cygon, and PCNB on soil microorganisms. Microb Ecol 10:345-358

30. Kriger AA, Turner RR (1995) Field analysis of mercury in water, sediment and soil using static headspace analysis. Water Air Soil Pollut 80:1295-1304

31. Lawton JH (1994) What do species do in ecosystems? Olkos 71:367-374

32. Magurran AE (1988) Ecological Diversity and its Measure- ment. University Press, Cambridge, UK

33. MacArthur R (1955) Fluctuations of animal populations and a measure of community stability. Ecology 36:533-536

34. McGrady-Steed J, Harris PM, Morin PJ (1997) Biodiversity regulates ecosystem predictability. Nature 390:162-165

35. Mikola J, Setfil~i H (1998) Relating species diversity to ecosystem functioning: Mechanistic backgrounds and experimental approach with a decomposer food web. Olkos 83: 180-194

36. Mochizuki M, Hattori T (1988) Kinetic study of growth throughout the lag phase and the exponential phase of Escherichia coli, FEMS Microbiol Ecol 45:291-296

37. Muyzer, G, de Waal EC, Uitterlinden AG (1993) Profiling of complex microbial populations by denaturant gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Appl Environ Microbiol 59:695-700

58 A.K. Mtiller et al.

38. Miiller AK, Westergaard K, Christensen S, Sorensen S] (2001) The effect of long-term mercury pollution on the soil microbial community. FEMS Microbiol Ecol 36:11-19

39. Naeem S, Thompson LI, Lawler SP, Lawton ]H, Woodfin RM (1995) Empirical evidence that declining species diversity may alter the performance of terrestrial ecosystems. Phil Trans R Soc Lond 347:249-262

40. Nordgren A, B~th E, S6derstr6m B (1988) Evaluation of soil respiration characteristics to assess heavy metal effects on soil microorganisms using glutamic acid as a substrate. Soil Biol Biochem 20:949-954

41. Nfibel U, Engelen B, Felske A, Snaidr 1, Weishuber A, Amann RI, Ludwig W, Backhaus H (1996) Sequence het- erogeneities of genes encoding 16S rRNAs in Paenibacillus polymyxa detected by temperature gradient gel electrophoresis. ] Bact eriol 178:5636-5643

42. Ose EE, Tonkinson V (1985) Comparison of the antimyco- plasma activity of two commercially available tylosin pre- mixes. Poultry Sci 64:287-293

43. Palumbo AV, Zhang C, Liu S, Scarborough SP, Pfiffner SM, Phelps TJ (1996) Influence of media on measurement of bacterial populations in the subsurface. Appl Biochem Bio- tech 57/58:905-914

44. Prosser ]I (1997) Microbial processes within the soil. In: Dirk van Elsas ], Trevors IT, Wellington EMH (eds) Modern Soil Microbiology. Marcel Dekker, New York, pp 183-213

45. Rasmussen LD, Sorensen S], Turner RR, Barkay T (2000) Application of a Mer-lux biosensor for estimating bioavailable mercury in soil and its utility in relating the response of soil microbial communities to bioavailable mercury. Soil Biol Biochem 32:639-646

46. Rykiel El Jr (1985) Towards a definition of ecological disturbance. Aus ] Ecot 10:361-365

47. Salonius PO (1981) Metabolic capabilities of forest soil microbial populations with reduced species diversity. Soil Biol Biochem 13:1-10

48. Shishido M, Chanway CP (1998) Storage effects on indi- genious soil microbial communities and PGPR efficacy. Soil Biol Biochem 30:939-947

49. Smalla K, Wachtendorf U, Heuer H, Liu W, Forney L (1998) Analysis of biolog GN substrate utilization patterns by microbial communities. Appl Environ Microbiol 64:1220-1225

50. Smibert RM, Krieg NR (1994) Phenotypic characteriza- tion. In: P Gerhardt, RGE Murry, WA Wood, NR Krieg

(eds) Methods for General and Molecular Bacteriology. American Society for Microbiology, Washington DC, pp 607-615

51. Smit E, Leeflang P, Wernars K (1997) Detection of shifts in microbial community structure and diversity in soil caused by copper contamination using amplified ribosomal DNA restriction analysis. FEMS Microb Ecol 13:249-261

52. Tilman D (1996) Biodiversity: Population versus ecosystem stability. Ecology 77:350-263

53. Tilman D, Wedin D, Knops ] (1996) Productivity and su- stainability influenced by biodiversity in grassland ecosystems. Nature 379:718-720

54. Tilman D, Knops ], Wedin D, Reich P, Ritche M, Siemann E (1997) The influence of functional diversity and composition on ecosystem processes. Science 277:1300-1302

55. Torsvik V,Goksyr ], Daee FL (1990) High diversity in DNA of soil Bacteria. Appl Environ Microbiol 56:782-787

56. Torsvik V, Daae FL, Sandaa R, Ovre~s L (1998) Novel techniques for analysing microbial diversity in natural and perturbed environments. ] Biotech 64:53-62

57. Wardle DA, Giller KE (1996) The quest for a contemporary ecological dimension to soil biology. Soil Biol Biochem 28: 1549-1554

58. Wardle DA, Zackrisson O, H6rnberg G, Gallet C (1997) The influence of island area on ecosystem properties. Science 277:1296-1299

59. Watve MG, Gangal RM (1996) Problems in measuring bacterial diversity and a possible solution. Appl Environ Microbiol 62:4299-4301

60. Westergaard K, Mfiller AK, Christensen S, Bloem ], Sorensen S]. (2001) Effects of tylosin as a disturbance on the soil microbial community. Soil Biol Biochem, 33:2061-2071

61. Winding A, Hendriksen NB (1997) Biolog substrate utilization assay for metabolic fingerprints of soil bacteria: Incu- bation effects. In: H Insam, A Rangger (eds) Microbial Communities, Functional versus Structural Approaches. Springer-Verlag, Berlin, pp 195-205

62. von Wintzingerode F, G6bel UB, Stackebrandt E (1997) Determination of microbial diversity in environmental samples: Pitfalls of PCR-based rRNA analysis. FEMS Microb Rev 21:213-229

63. Ovre~s L, Torsvik V (1998) Microbial diversity and community structure in two different agricultural soil communities. Microb Ecol 36:303-315

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.

Documents

jurnal ilmu tanah