Tải bản đầy đủ - 0 (trang)
IV. Effect on Microbial Activities and Composition

IV. Effect on Microbial Activities and Composition

Tải bản đầy đủ - 0trang

FUMIGANTS ON NON-TARGET ORGANISMS IN SOILS



9



similar when compared to the other three fumigants, suggesting a lesser effect of

1,3-D on heterotrophic. In another study, effects of SS on soil microbial

properties, including metabolic diversity of the microbial communities, were

examined in a greenhouse study and were compared to MeBr (Tanaka et al.,

2003). The authors found that the richness and average well color development

(AWCD) values in the microbial communities decreased markedly immediately

after treatment with MeBr but showed a rapid recovery, while those treated

with CP continued to decrease until the transplanting of tomato seedlings.

This was in agreement with Ibekwe et al. (2001a) who showed a sharp decrease in

microbial diversity with MeBr, followed by a quick recovery of the community

after 8 weeks when compared to a more sustained effect for a longer period with

CP. The shifts in microbial communities observed in the Biolog assays were due

to the toxic effects of fumigants on rapid growing microorganisms of high

population in the soils. Analysis of microbial communities from the Biolog GN

assay by DGGE confirmed that carbon source utilization profiles obtained with

Biolog GN plates do not necessarily discriminate the numerically dominant

members of the microbial community used as the inoculum (Engelen et al., 1998;

Smalla et al., 1998).

Under field conditions, natural fluctuations in carbon substrate utilizing

activity and community-level physiological profiles of microorganisms in low

input and conventional rice paddy soils were monitored using Biolog GN plates

for 2 years. The purpose was to establish criteria for assessing side effects of

pesticides on soil microbial ecosystems (Itoh et al., 2002). The activity changed

seasonally showing a regular pattern with more activity observed during late

summer. The level of microbial activities seemed to be directly influenced by soil

temperature and/or redox potential. Soil microbial communities grouped into

three clusters, August – December, January – April, and May – August, based on

the sampling season. Many studies have shown the effects of fumigants on

microbial activities and community structure. The effect of metam sodium

fumigation on community structure after a 5- and 18-week incubation showed

a separation of the two communities along the first principal component (PCA)

based on treatment dose (Macalady et al., 1998). There was a signicant

p ẳ 0:001ị dose treatment effect at week 5, whereas at week 18 there was no

dose significant effect ðp ¼ 0:05Þ; but there was a binary variable effect

ðp ¼ 0:001Þ between the treated and untreated samples. Toyota et al. (1999)

compared the AWCD values and richness (number of positive wells) of different

categories of the 95 substrates 105 days after fumigation with metam sodium.

They found that both AWCD and richness in all the substrate groups were

signicantly p ẳ 0:05ị lower in the fumigated soils than in the control soils. In

the radish rhizosphere and non-rhizosphere soils fumigated with CP, there was a

significant suppression in AWCD and richness in fumigated soils compared to the

control (Itoh et al., 2002). It was assumed that the bacterial populations with a

high substrate assimilation activity were damaged by CP fumigation and a



10



A. M. IBEKWE



different microbial community was developed in the radish rhizosphere. Cluster

analysis of these communities after 24 h of incubation separated between

rhizosphere and non-rhizosphere samples, and then between fumigated and nonfumigated samples, suggesting the effect of the rhizosphere by CP fumigation.

After 72 h, the Biolog samples showed a clear separation between the fumigated

and the non-fumigated samples in both the rhizosphere and the non-rhizosphere

samples. The authors concluded that fumigants affected mostly the slowsubstrate utilizing rhizosphere microbial communities or that fast growers

seemed to utilize most of the substrates during the early stages of incubation. The

problem with the Biolog system is due first to the respiratory activities of fast

growing heterotrophic bacteria resulting in the stimulation or reduction of the

catabolism of 95 carbon substrates (Engelen et al., 1998; Ibekwe et al., 2001a).

The shifts in microbial communities observed in the Biolog assays were due to

the rapid growth of organisms of their high population in the soils. For example,

Pseudomonas species are found in most soil samples and they respond well in

Biolog assays (Haack et al., 1994; Garland, 1996, 1997).



B. CHANGES IN MICROBIAL BIOMASS

SIR (Anderson and Domsch, 1978) has been the standard method for soil

microbial biomass measurements. It is based on the maximal initial response of the

soil microbial biomass to a substrate amendment. Microbial biomass is assessed

quantitatively by the C, N, ninhydrin-reactive compounds, ATP, quinones, and

phospholipid fatty acid (PLFA) composition of the cells. Total organic carbon in

the microbial biomass (biomass C) is considered as the general indicator of the

amount of microorganisms in the soil, and total nitrogen is considered to be the

indicator of potential available nitrogen in the soil. Ninhydrin-reactive

compounds represent the labile fraction of biomass N and are metabolized to

ammonia by heterotrophic microorganisms in soil. This is one of the major

fractions of available N to plants and microorganisms. The concentration of ATP

in soil is an indicator of the amount of microorganisms that can be readily and

rapidly measured. The total amount of respiratory quinones has been shown to be

an indicator of the microbial biomass since many microorganisms have only one

quinone species. Bacteria contain relatively constant amounts of viable biomass as

phospholipid, so this can also be used as a good biomass indicator.

Many studies have shown the impact of fumigation on microbial biomass

(Zelles et al., 1997; Toyota et al., 1999; Ibekwe et al., 2001a; Suyama et al.,

2001; Itoh et al., 2002). Zelles et al. (1997) reported a decrease in the microbial

biomass-C and -N of about 20% after chloroform fumigation. Griffiths et al.

(2000) examined a technique based on progressive chloroform fumigation of soil

to reduce soil microbial biodiversity, and measured the effects of the reductions



FUMIGANTS ON NON-TARGET ORGANISMS IN SOILS



11



upon the stability of key soil processes. The diversity of cultivable and noncultivable bacteria, protozoa, and nematodes was progressively reduced by

increasing fumigation times, with the total microbial biomass less in fumigated

soils than the unfumigated. Specific parameters like nitrification, denitrification

and methane oxidation decreased as biodiversity decreased. Suyama et al. (2001)

looked at the effects of fumigation on paddy rice soil to establish a criteria to

assess the short-term effects of pesticides on soil microorganisms. They

concluded that the degree of fluctuation of microbial biomass and population

in the paddies can be used as references to assess the degree of pesticide effects in

other Japanese paddy soils. In another study, the effects of the pesticides

fenitrothion, chlorothalonil, chloropicrin, linuron, and simazine on microbial

biomass were monitored for 28 days for changes in respiratory quinone profiles

(Katayama et al., 2001). Pesticides were applied to the soil at 10 times the

recommended rates. Application of CP decreased the amount (an indicator of

microbial biomass) and diversity (an indicator of taxonomic diversity of the

microbial community) of the different quinones species during the 28-day

treatment. Continuous change in the structure of the microbial community in the

CP-treated soil was documented by the changes in the dominant quinone species,

and there was no change in the control soil. The authors concluded that quinone

profile analysis is a potential method to detect the effect of pesticide on a soil

microbial community and biomass.

PLFA profiles are often used to study microbial diversity and biomass in

complex communities (Zelles, 1999). PLFAs are components of phospholipids

that are essential parts of membranes found in all living cells. Certain signature

fatty acids in the overall PLFA profile are specific for groups of bacteria, fungi,

and actinomycetes (Tunlid and White, 1992). The biomass of these groups can be

studied once fumigants are applied to any soil because this will represent the

living component of the population. Analysis of PLFA profiles of soils fumigated

with MeBr, MITC, 1,3-D, and CP was carried out over a 12-week period after

application. Biomarker peaks were analyzed and were determined to range from a

minimum of 1.3 nmol g21 dry wt for the four fumigants (week 1) to a maximum

of 55 nmol g21 dry wt for the 1,3-D- and CP-treated samples in week 12 (Ibekwe

et al., 2001a). The biomass contents, as indicated by the total PLFA, were

significantly different at different time points in some treatments ðp , 0:05Þ: At

week 1, the biomass contents in MeBr-amended microcosms were significantly

lower than those in weeks 8 and 12 (Fig. 3), and of the three other fumigants

(Ibekwe et al., 2001a). There was also a decrease in biomass of some of the

Gram-negative (cy17:0, 15:0, and 18:1v7c) and the fungal (18:2v6c) biomarkers, with increases in MeBr concentration during the first week (Fig. 3).

There was a significant increase in biomass for Gram-positive bacteria (a17:0,

i17:0), fungi (18:2v6c), and actinomycetes (10me 16:0) in weeks 8 and 12. 1,3-D

and CP had the strongest effects on actinomycetes, resulting in a significant

decrease in biomass for most treatments. The effects of MITC followed the same



12



A. M. IBEKWE



Figure 3 Biomass contents (nmol of PLFA g21 dry wt of soil) of samples collected after weeks 1,

8, and 12 from MeBr fumigated soils n ẳ 3ị: MeBr1, -8, and -12 indicates samples were taken 1, 8,

and 12 weeks after the start of the experiment. 1, 2, and 3 after the “-” sign indicates 50% below field

application rate, field application, and 1000% above field application rates, respectively. Error bars

represent standard deviation.



trend as MeBr, except that the recovery of Gram-negative bacteria biomass did

not occur during week 8. The effects of fumigants on microbial biomass may be

short term with biomass recovery after a few weeks, as was seen with MeBr,

MITC, and 1,3-D, or it may be long term, as was observed with CP.



C. CHANGES IN SOIL MICROBIAL COMMUNITY STRUCTURE

AND COMPOSITION

Several methods to characterize microbial communities in soils that do not

depend on culturing have been recently developed. These methods were based on

the analysis of biomarkers, such as 16S rRNA genes, PLFA, and respiratory

quinones (Morgan and Winstanley, 1997). Analysis of 16S rRNA genes in soil

was used to detect the long-term effects of phenylurea herbicides on soil

microbial communities (El Fantroussi et al., 1999). The analysis of PLFA

was applied to detect the long-term microbial effects of heavy metals



FUMIGANTS ON NON-TARGET ORGANISMS IN SOILS



13



(Frostegard et al., 1996; Pennanen et al., 1996). However, soil microbial DNA

does not reflect changes in soil microbial biomass, and the profile of PLFA does

not represent individual taxonomic groups (Katayama and Fujie, 2000).

Katayama et al. (2001) showed that a significant difference p ẳ 0:05ị between

the quinone proles of pesticide-treated soils and the control soil at a given period

of incubation was observed for CP-treated soil after 28 days of incubation. The

use of quinones for microbial community structure analysis is based on the

classification of quinones into different functional groups. The group consisted of

MK-8 [Gram-positive bacteria with a low guanine plus cytosine (G ỵ C) content

and Actinobacteria ], followed by MK-7 (Cytophaga-Flavobacterium cluster and

Gram-positive bacteria with low G ỵ C contents) and next, MK-9 (Micromonosporineae, Streptmycineae, and Streptosporangineae). The authors reported

that 3 days after treatment with CP, MK-9 comprised the largest proportion of

quinones, indicating a change in the dominant microorganisms. After 7 days of

incubation, the largest proportion of quinones changed to MK-7, after 14 days to

MK-9, and after 28 days the MK-8 represented the second largest proportion of

quinones in the soils suggesting a continuous change in the structure of the

microbial community in the CP-treated soils. Sigler and Turco (2002) examined

the impact of the fungicide chlorothalonil on dominant bacterial and fungal

populations following application to turfgrass, forest, and agricultural soils.

Increased rates of chlorothalonil impacted eight bacterial populations (two

inhibitions, four enhancements, and two non-specific responses) and four fungal

populations (all inhibitions). Denaturing gradient gel electrophoresis (DGGE)

band numbers of 16S rRNA and the Shannon –Weaver index of diversity (H0 )

indicated an altered but not significantly different ðp , 0:05Þ bacterial and fungal

community structure following chlorothalonil fumigation. Sequencing of the

dominant DGGE bands indicated an impact on several groups of bacteria

(Cytophaga-Flavobacterium-Bacteroides, alpha-, beta-, gamma-, and deltaproteobacteria) and two fungal taxa (zygomycota and ascomycota). The authors

concluded that changes to the overall community structure of dominant species

were less significant, but a single chlorothalonil application and a short

incubation period resulted in community changes including both enhancement

and inhibition of a variety of dominant organisms.

In another study, chloroform fumigation was used to manipulate the

composition of microbial communities as a means of investigating relationships

between community structure and the functioning of soil processes (Dickens and

Anderson, 1999). The results showed that chloroform fumigation after 24 h

caused a large reduction in total PLFAs and poor regrowth of the residual

community. The effect of metam sodium on soil microbial community structure

and function in two Japanese soils showed that the number and pattern of

amplified 16S rRNA restriction patterns (ARDRA) fragments were changed by

fumigation. Shifts were observed in the percent G ỵ C prole toward a greater

proportion of lower percentage G ỵ C classes in treated soils as compared to the



14



A. M. IBEKWE



untreated soils (Toyota et al., 1999). They showed that the effects of fumigation

on the soil microbial community structure and function were pronounced and for

some parameters were very persistent.

There are several studies on the effects of fumigants on bacterial community

structure and function. Some of these studies relied on cultural techniques and,

more recently, on biochemical or molecular approaches. Depending upon the

approach, different conclusions were reached. Macalady et al. (1998) concluded

that, although community fatty acid analysis showed promise as a screening tool

in soil microbial toxicity studies, more detailed information could be obtained

from PLFA and other specific lipid fractions. The abundance of indicator fatty

acids for bacteria after 5 weeks of incubation was correlated to MITC doses, but

after 18 weeks, very few were related to the MITC dose. MITC was also observed

to reduce populations of culturable organisms dramatically in the Biolog assay.

Ibekwe et al. (2001a), using the same concentration of MITC and other fumigants

as Macalady et al. (1998), did a detailed study of the impact of four fumigants on

soil microbial communities using the Biolog assay, PLFA, and DGGE. They

found that the community structures of fumigant-treated microcosms measured

by PLFA analysis shifted away from the first communities after 8 and 12 weeks.

The shift was greatest with MeBr, which doubled the amount of variation in

component 1 when compared to component 2. The major difference in the PLFA

profiles between the MeBr-treated and the control microcosms was that the MeBr

microbial communities contained significantly more branched chain PLFAs

[specifically, a17:0, i17:0, a15:0, and i15:0 ðp , 0:05Þ], indicative of Grampositive bacteria (White and Findlay, 1988; Heipieper et al., 1992; Tunlid and

White, 1992). In the MeBr-treated microcosms, the relative proportion of PLFAs

indicative of fungal biomass (Guckert et al., 1985), specifically 18:2v6c and

18:3v6c, increased over time. Comparison of PLFA profiles of MITC, 1,3-D, and

CP-treated microcosms to the control samples showed that microbial community

from week 1 was furthest away from the control, and after weeks 8 and 12, PLFA

profiles of the three fumigants and the control were not significantly different

from each other. The major advantage of PLFA analysis over other techniques is

that it has been regarded as an indicator of the total microbial biomass and certain

PLFAs can be used as biomarkers for specific groups of microorganisms (Tunlid

and White, 1992; Zelles et al., 1992; Vallaeys et al., 1997). The presence of large

proportions of branched fatty acids (a15:0, i15:0, a17:0, and i17:0), which are

markers for Gram-positive bacteria, showed that Gram-positive bacteria were

less affected by the impact of these fumigants when compared to Gram-negative

bacteria. Zelles et al. (1997) found that Gram-positive bacteria were less injured

by chloroform fumigation and attributed it to protection by their cell wall

structure, formation of spores, and ability to adapt to fumigant vapor more

quickly. The total amount of PLFAs decreased by about 50% after 10 days

of chloroform fumigation and monosaturated fatty acids, indicative of

Gram-negative bacteria, were more heavily affected (60 – 70%).



FUMIGANTS ON NON-TARGET ORGANISMS IN SOILS



15



D. ANALYSIS OF SOIL MICROBIAL COMMUNITY STRUCTURE

BY MOLECULAR TECHNIQUES

Analysis of 16S rRNA sequences retrieved from environmental samples is the

standard genetic technique for determining microbial community structure

without bias introduced by cultivation (Head et al., 1998). The technique uses

genetic fingerprinting to describe microbial diversity on a community scale. This

has given significant information on environmental changes that could not have

been observed by traditional techniques. However, some of the techniques still

measure a fraction of the soil community or the dominant population. Marked

changes in the bacterial community structure within the dominant population can

be measured by a fingerprinting pattern to illustrate the effects of xenobiotics.

Very few studies of fumigant effects of soil bacteria have used this technique.

Engelen et al. (1998) studied the effects of commercial formulations of the

herbicides dinoterb (field dose) and metamitron (10 times field dose) on the

bacterial community in a laboratory experiment on a previously unexposed soil.

The impact of the herbicide treatment was monitored by 16S rRNA-TGGE of soil

community DNA and other classical tests, such as SIR, dehydrogenase, carbon

and nitrogen mineralization, and community fingerprinting using Biolog. TGGE

gels of the 16S rRNA showed that dinoterb had a marked effect on bacterial

diversity. The community structure in the dinoterb soil was dominated by

sequences associated with nitrite-oxidizing bacteria (Nitrospina, Nitrospira) not

found in the control. In another study, El Fantroussi et al. (1999) showed the

long-term effects of three phenylurea herbicides on bacterial diversity in an

orchard soil. The number of bacterial colony forming units (CFUs) on R2A agar

medium was affected by all herbicide treatments, emphasizing the impact of

these pesticides on the microbial community. DGGE analysis of 16S rRNA

revealed a clear effect of the four pesticides.

DGGE was introduced into microbial ecology (Muyzer et al., 1993) as an

attempt to obtain an overview of the structural diversity of microbial

communities. It is based on the separation of ribosomal gene sequences directly

amplified from community DNA by using conserved primers on a denaturing gel

according to their melting point. DGGE analysis of 16S rRNA fragments was

used to examine the effects of four fumigants on soil microbial communities

(Ibekwe et al., 2001a). Figure 4a –c shows the DGGE patterns of the 16S rRNA

fragments (primers P338f and P518r) amplified from the four soils 1, 8, and 12

weeks after fumigation. During the first week, when the most drastic effects

occurred, MeBr treatments did not produce any dominant bands compared to the

other fumigants and the control (Fig. 4a). At week 8 there was a significant shift

in the microbial community structure of the MeBr-treated soils. As shown in

Fig. 4b, more bands begin to appear in MeBr treatments. There was also a

decrease in the number of bands as the concentration of MeBr increased (MeBr

8-1 to 8-3 from 8 bands to 2, with increasing concentration). At 12 weeks,



16



A. M. IBEKWE



Figure 4 DGGE analysis of 16S rRNA fragments of pooled soil samples collected from triple

microcosm treated with different fumigants. Amplified products were separated on a gradient gel of

30– 70% denaturant. (a) Community structures 7 days after the initiation of the experiment. (b) Community structures after 8 weeks of treatment. (c) Community structures after 12 weeks of treatment.



the microbial communities for all concentrations of MITC, 1,3-D, CP, and the

lowest concentrations of MeBr were similar to the control (Fig. 4c).

A second approach was used to determine community structure of different

bacterial groups based on the 16S rRNA peak height. The peak height values



FUMIGANTS ON NON-TARGET ORGANISMS IN SOILS



17



Figure 4 (continued )



generated from the sampling points (Fig. 4a – c) were integrated and analyzed

using Excel (Microsoft). Data obtained were used to integrate the area under each

peak for each gel lane of every treatment. Each band was presumed to represent

the ability of that bacterial species to be amplified. The Shannon – Weaver index

of diversity (H0 ) (Shannon and Weaver, 1963) was used to compare changes in

diversity of microbial communities within the four treatments at each time period

by using the following function:

H 0 ¼ 2{Pi log Pi }

Pi ¼ ni =N; where ni is the height of peak and N is the sum of all peak heights in

the curve.

This resulted in a direct comparison of the effect of each compound at one time

point in one gel on the structural diversity of the four microbial communities. The

Shannon –Weaver index of diversity H0 was calculated on the basis of the number

and relative intensity of bands on a gel strip. The four treated samples showed

different levels of diversity ranging between 0.11 and 1.26 at different sampling

times. It also showed that MeBr exerted the most significant effects on the

structural diversity of the soil compared to the other three fumigants and the

control. One week after MeBr application, the number of bands decreased from

16 in the control to almost undetectable numbers in MeBr-treated soil. This

indicated the collapse of the microbial community due to the acute toxicity of

MeBr. H0 decreased from 1.26 in the control to 0.11 in the treatment with the

highest concentration of MeBr, and continued to increase slightly during week 8



18



A. M. IBEKWE



and subsequently, in week 12 to about 0.75 but remained clearly below the

average control value ðH 0 ¼ 1:26Þ:

To determine a shift in the microbial community structure, DGGE bands from

Ibekwe et al. (2001a) were selected for reanalysis as shown in Fig. 4a – c. Figure 5

shows the phylogenetic analysis of prominent bands recovered from the DGGE

gel. Band sequence was analyzed using the BLAST database (National Center for

Biotechnology Information, www.ncbi.nlm.nih.gov). Sequence alignments were

performed using the PILEUP program from the University of CaliforniaRiverside Genetics Computer Group (GCG programs). Matrices of evolutionary

distances were computed using the Phylip program with the Jukes – Cantor model

(Jukes and Cantor, 1969). All clones extracted from fumigated soils and their

accession numbers are shown in bold. Phylogenetic tree was constructed and

checked by bootstrap analysis (100 data sets) using the program SEQBOOT.

Bootstrap values represent the frequency of resampling that supports a specific



Figure 5 Phylogenetic tree constructed for 16S rRNA gene sequences and aligned by the GCG

program from the University of California, Riverside genetic group. The tree was produced by using a

neighbor-joining algorithm. Bands were cut from the DGGE gel and cloned for analysis.



FUMIGANTS ON NON-TARGET ORGANISMS IN SOILS



19



branching pattern. Two prominent bands (F1 and F2) were observed during the

first week of analysis. The derived sequences from these bands confirmed F1 to

be 100% similar to Pseudomonas reactans and F2 to be 99% similar to

Pseudomonas putida. Mazzola et al. (2002) showed that microbial community

residents in a wheat soil fumigated with MeBr-suppressed components of the

microbial complex that incited replant apple disease. P. putida was the primary

fluorescent pseudomonad recovered from the suppressive soil, whereas

Pseudomonas fluorescence bv. III was dominant in the conducive soil. At

week 8, four dominant bands (F3 –F7) and three other bands, F16, F17, and F31

appeared in the MeBr treatments. Communities from the other three fumigants

were different from that of MeBr. Bands F3 and F6 showed relationships to the

Gram-positive species Heliothrix oregonensis and Bacillus subtilis, respectively

(Fig. 5). There were no significant new bands at week 12, as the community

profiles from the four fumigants shifted close to the control.

In another closely related study, the effect of different concentrations of PBr

and 1,3-D in unamended and manure-amended soil on the microbial community

was evaluated (Dungan et al., 2003a). DGGE analysis of the PCR fragments

(primers 63f and 518r) from weeks 1, 4, 8, and 12 showed a shift in the structural

composition of the communities during the 12-week study. The effects of the

fumigants on the microbial community structure were most dramatic 1 week after

application, as unamended and amended soils treated with PBr or 1,3-D clustered

away from the controls. Dominant 16S rRNA bands were not detected in the

unamended and amended soils treated with 100 mg kg21 of PBr and

500 mg kg21 of 1,3-D. Four weeks after treatment, PBr and 1,3-D treatments

began to cluster closer to the control in amended soils and in unamended soil

treated with 1,3-D. The impact of both fumigants on the microbial community

was less dramatic in the manure-amended soils than the unamended soil. PBr

treatment was more damaging to the microbial community structure than 1,3-D,

as significantly fewer bands were found in the PBr treatments and a longer time

was required for recovery. In general, the structural diversity of the dominant

microbial community decreased with increasing fumigant concentration,

regardless of the treatment. The Shannon – Weaver index of diversity H0 was

also affected by the fumigants. H0 ranged from zero in some of the highconcentration treatments to about 1.4 in some of the controls and 10 mg kg21

treatments. In amended soil treated with 10 mg PBr kg21 the H0 differed little

from the control. This was not the case in unamended soils. After week 4, the H0

values of 1,3-D concentrations were similar to the control-amended soil, but

decreased to zero at 500 mg kg21. By weeks 8 and 12, the H0 values at all

concentrations of 1,3-D were similar to that of the PBr treatments. They

concluded that PBr is most damaging to the microbial community in unamended

soil, but not in soil containing organic amendment. It should be noted that the

banding patterns obtained from these two studies reflected the most abundant

rRNA types in the community, but not the total members of the community.



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

IV. Effect on Microbial Activities and Composition

Tải bản đầy đủ ngay(0 tr)

×