Subscribe Now Subscribe Today
Research Article
 

Assessing Microbial Community in Andisol Differing in Management Practices by Biochemical and Molecular Fingerprinting Methods



M. Hasinur Rahman, Atsushi Okubo, Shigenao Kawai and Shu-ichi Sugiyama
 
Facebook Twitter Digg Reddit Linkedin StumbleUpon E-mail
ABSTRACT

Several biochemical and molecular methods are used to investigate the microbial community structure and microbial diversity in soils. Biochemical (Phospholipid fatty acid: PLFA profiles) and molecular (Denaturing gradient gel electrophoresis of polymerase chain reaction amplified DNA: PCR-DGGE) fingerprinting methods were employed to quantify soil microbial community in Andisol differing in management practices. No-tillage (NT) soil had significantly higher microbial biomass carbon and microbial biomass nitrogen than conventional tillage (CT) soil. PLFA for gram-positive bacteria, gram-negative bacteria, aerobes, cyanobacteria and fungi were significantly higher in NT than CT. On the other hand, PLFA for sulfate-reducing bacteria, methane-oxidizing bacteria and mycorrhizae were significantly higher in CT than PD and NT. The total DNA extracted from Andisol with differing management practices ranged from 21.0 to 33.0 μg g-1 soil. Soil DNA yielded from puddling (PD) showed highest amount and NTg showed lowest amount. There were no significant variations in DNA yield obtained from Andisol of CT and NTg. Highest bacterial diversity evaluated by DNA band number in DGGE analysis based on PCR amplification of 16S rDNA fragments was observed in PD and can be arranged as: PD>CT>NT. On the other hand, highest fungal diversity evaluated by DNA band number in DGGE analysis based on PCR amplification of 18S rDNA fragments was observed in NT and can be arranged as: NT>CT>PD. Results indicated that microbial community was responsive to management practices demonstrating their usefulness as indicators of soil quality in temperate Andisol.

Services
Related Articles in ASCI
Search in Google Scholar
View Citation
Report Citation

 
  How to cite this article:

M. Hasinur Rahman, Atsushi Okubo, Shigenao Kawai and Shu-ichi Sugiyama, 2008. Assessing Microbial Community in Andisol Differing in Management Practices by Biochemical and Molecular Fingerprinting Methods. International Journal of Soil Science, 3: 1-10.

DOI: 10.3923/ijss.2008.1.10

URL: https://scialert.net/abstract/?doi=ijss.2008.1.10

INTRODUCTION

Management practices have a strong influence on the soil physical, chemical and microbiological properties. Microbial-based indicators of soil quality are believed to be more dynamic than those of physical and chemical properties. Therefore, only physical and chemical properties do not draw detailed conclusions on soil fertility. Microbial community is an important determinant of soil organic matter decomposition rates and nutrient turnover and availability in agricultural soils. Microbial characteristics of soils are being evaluated increasingly as sensitive indicators of soil health because of the clear relationships between microbial diversity, soil and plant quality and ecosystem sustainability (Doran et al., 1994). While the understanding of microbial properties such as biomass, activity and diversity are important to scientists in furthering knowledge of the factors contributing to soil health, results of such analyses may also be useful to extension personnel and farmers in devising practical measures of soil quality (Hill et al., 2000).

Soil ecosystems are highly complex, containing a tremendous amount of microbial species. Indigenous microbial populations in soil are of fundamental importance for ecosystem functioning, through determining nutrient cycling, organic matter decomposition and energy flow (Doran and Zeiss, 2000). Despite all attempts to measure fluxes and gross microbial pools, the soil and its microbiota still remain a black box. Most soil micro-organisms are still unknown, while very few have been isolated, cultured and identified and directly related to their function in agroecosystems. The development of effective methods for studying the diversity, distribution and behavior of micro-organisms in soil habitats is essential for a broader understanding of soil health. Traditionally, the analysis of soil microbial communities has relied on culturing techniques using a variety of culture media designed to maximize the recovery of diverse microbial populations. Although new culture media have been recently developed to maximize the recovery of diverse microbial groups from soils (Balestra and Misaghi, 1997; Mitsui et al., 1997), only a small fraction (<0.1%) of the soil microbial community has been accessible with this technique (Atlas and Bartha, 1998; Hill et al., 2000). Widmer et al. (2001) reported that standard microbiological culture techniques leave over 90% of the micro-organisms unaccounted for in the environment. Prosser (2002) reported that less than 5% of soil micro-organisms are cultivable. To overcome these problems, several methods have been utilized in an attempt to assess a greater proportion of the soil microbial community. On the other hand, recent techniques like Phospholipid Fatty Acid (PLFA) and Denaturing Gradient Gel Electrophoresis (DGGE) analyses have provided a new understanding of microbial communities in soil. The analysis of PLFA is a useful assay for soil microbial community because (i) the concentration of total PLFA can be used as an index of viable microbial biomass since phospholipids are rapidly degraded after cell death, (ii) certain fatty acids may be used as molecular markers for specific taxa and as indicators of microbial stress and (iii) multivariate analysis of the PLFA profiles can be used to detect changes in community composition (White et al., 1996). More recently, molecular techniques for identification that do not require cultivation have been developed to study of different groups of micro-organisms. One of those techniques is DGGE of PCR-amplified genes, which can be used to evaluate the diversity of complex microbial systems (Muyzer et al., 1993). We used (i) PLFA analysis with gas chromatography and mass spectrometry and (ii) DGGE analysis of polymerase chain reaction-amplified genes coding for 16S rDNA and 18S rDNA for bacteria and fungi, respectively to investigate the microbial communities in Andisol. This is the first attempt done to check whether the two methods give identical results in detecting the soil microbial communities especially in Andisol. Such information will improve our understanding in investigating microbial communities in Andisols.

MATERIALS AND METHODS

Collection of Soil Samples
Soil samples were collected from the experimental field of Iwate University, Japan established in 1980 and located in about the latitude of 39° 42´ N and longitude 141° 10´ E with an altitude of about 155.2 m above sea level. The mean monthly maximum and minimum temperature, precipitation and humidity of the study area for last 20 years are shown in Table 1.

Three sites under NT, CT and PD systems were selected in this study. The samples were collected from A horizons at a depth of 0-10 cm of sandy clay pumice Andisol (Melanudands) which was partially modified by siltation of Kitakami river from apple orchard in which apple was grown with NT, grassland in which fodder crops (Orchardgrass: Dactylis glomerata L., Kentucky bluegrass: Poa prantensis L., White clover: Trifolium repens L., Tall fescue: Lolium arundinaceum [Schreb.] SJ Darbyshire, Meadow fescue: Lolium pratense [Huds.] SJ Darbyshire, Rumex: Rumex obtusifolius L. Sere etc.) were grown with no-tillage (NTg), dryland soil cultivated with sweet potato, corn, wheat, barley, rye and colza in rotation with conventional tillage (CT) and puddling with continuous rice (PD) systems for more than 20 years. It can be noted that two or three grass species (Orchardgrass: Dactylis glomerata L., White clover: Trifolium repens L., Tall fescue: Lolium arundinaceum [Schreb.] SJ Darbyshire, Rumex: Rumex obtusifolius L. Sere etc.) existed or were present in NT-apple orchard as interplants. Tillage methods, cropping patterns and changes in soil properties were measured as per the methods followed by Rahman et al. (2003).


Table 1: Meteorological data of field area during 1980-2006
Image for - Assessing Microbial Community in Andisol Differing in Management Practices by Biochemical and Molecular Fingerprinting Methods
Morioka meteorological station, Iwate, Japan

The soil was sampled in September 2006 (after ploughing) from surface horizon because this is where the maximum difference in soil properties occurs among treatments (Rahman et al., 2003). Ten samples were collected from each plot and made into three composite samples. Each soil sample was placed in a plastic bag in the field and kept cool until it was processed in the laboratory. Soil samples were sieved (<4 mm) in the laboratory. After removing visible plant litters, rocks and roots, a portion of soil was immediately frozen for microbial analysis.

Chemical Analysis
The total C and N contents of soils were determined using a Sumigraph-90A automatic analyzer (Sumitomo Chemical Co. Ltd., Osaka, Japan) and expressed as g kg-1 soil. Since there was no indication of the presence of carbonates, total C was equivalent to organic C. The flush of C was converted to total microbial biomass carbon (MBC) by multiplying by 2.64 (Vance et al., 1987) and the flush of N was converted to total microbial biomass nitrogen by multiplying by 1.86 (Brookes et al., 1985).

Microbial Community Structure
PLFA analysis was done using a composite mixture of three replicates in 20 samples for each site. The method described by Frostegård et al. (1993) was followed to obtain PLFA profiles for soil samples. Briefly, lipids were extracted from fresh soils (equivalent to 10.0 g of dry soil) following 4 h shaking using a mixture of chloroform: methanol: citrate buffer (1:2:0.8 v/v/v). The chloroform phase of the extraction was collected and then separated into functional classes using silicic acid column chromatography (Sep-Pak Silica, Waters, Milford, USA). The polar lipid was eluted with 5 mL methanol and then dried under N2 gas. Then, fatty acid methyl esters (FAMEs) were extracted. The dried samples were saponified with 1 mL of saponification reagent (NaOH 15 g, methanol 50 mL and distilled water 50 mL) and heated in a 100 °C water bath for 30 min. The samples were then cooled with water stream and then methaylated by adding methylation reagent (6.0N HCl 130, 110 mL methanol) and heated in a 80 °C water bath for 10 min. FAMEs were extracted by adding a 1.25 mL mixture of 1:1 methyl-t-butyl ether and n-hexane (v:v) and shaked for 10 min. The organic phase was then transferred to another test tube, dried under a stream of N2 gas, resuspended in 10 μL of n-hexane and 1 μL was analyzed on a gas chromatography (GC-14B, Shimadzu, Kyoto, Japan) equipped with a capillary column of 30 m with 0.25 m inner diameter (SPB-1, Supelco, Bellefonte, USA) in a condition of oven temperature of 150 °C (4 min) to 250 at 4 °C increment min-1. Helium was used as carrier gas, 20 cm sec-1, at 150 °C and FID at 280 °C. In addition, PLFA mixtures extracted from representative soils were analyzed by GC-MS (PQ2010, Shimadzu, Kyoto, Japan) and chemical structure of each PLFA peak was identified. We also used a commercial bacterial FAME mixture for peak identification (FAME, Supelco, Bellefonte, USA). PLFAs of each soil sample were identified from the chromatographic retention time by comparison with nonadecanoic acid (19:0, Supelco, Bellefonte, USA) as an internal standard. In this study PLFA peaks with more than 0.071 μg g-1 soil equivalent proportion to the amount almost higher than 0.5 % of total PLFA were used (Kourtev et al., 2003). A total of 58 PLFA peaks were found and 45 of them were identified and 34 were considered for evaluation or further studies. For each sample, the abundance of individual fatty acid methyl-ester was calculated by comparison with internal standard and was expressed as nmol g-1 soil.

According to Hill et al. (2000), common fatty acid signatures were used for groups of organisms viz., common bacteria, aerobes, anaerobes, sulfate-reducing bacteria, methane-oxidizing bacteria, cyanobacteria and fungi. The fatty acid signatures i15:0, a15:0, 15:0, 16:0, 16:1ω5, 16:1ω9, i17:0, a17:0, 17:0, 18:1ω7t, 18:1ω5, i19:0 and a19:0 were chosen to represent bacterial PLFA. The fatty acid signatures 16:1ω7, 16:1ω7t and 18:1ω7t were chosen to represent aerobes. The fatty acid signatures cy17:0 and cy190 were used for anaerobes. For sulfate-reducing bacteria, fatty acid signatures 10Me16:0, i17:1ω7 and 17:1ω6 were used. Fatty acid signatures 16:1ω8c, 16:1ω8t, 16:1ω5c, 18:1ω8c, 18:1ω8t and 18:1ω6c were considered for methane-oxidizing bacteria. The polyenoic, unsaturated PLFA 18:2ω6 was used as an indicator for cyanobacteria. The PLFAs 18:1ω9, 18:2ω6, 18:3ω6 and 18:3ω3 were used for fungi. On the other hand, the branched, saturated PLFAs 10Me16:0, i15:0, a15:0, i16:0, i17:0 and a17:0 were chosen to represent Gram-positive bacteria (O`Leary and Wilkinson, 1988; Zogg et al., 1997). The monoenoic and cyclopropane unsaturated PLFAs 16:1ω7t, 16:1ω9c/a16:0, 18:1ω5c, 16:1ω7c, cy17:0, 18:1ω7c and cy19:0 were chosen to represent Gram negative bacteria (Zogg et al., 1997; Porazinska et al., 2003). The PLFAs 16:1ω5c and 10Me18:0 were used to represent mycorrhizae and actinomycetes, respectively (Bossio and Scow, 1998; Zak et al., 2003).

Bacterial and Fungal Diversity
DNA extraction was performed with the Ultra Clean Soil DNA kit (MoBio Laboratories, Solana Beach, USA). One gram soil was processed according to the protocol. Concentration of DNA was measured by BIO-RAD SmartSpecTM Plus Spectrophotometer, USA and expressed as μg g-1 soil.

Bacterial 16S rDNA were amplified using the primers 341f-GC and 534r (Muyzer et al., 1993). Amplification of fungal 18S rDNA required nested PCR. First round PCR used the primers NS5 and NS8 (White et al., 1990) and the nested-PCR used FR1-GC and FF390 (Vainio and Hantula, 2000). The sequence, band-size obtained and PCR thermal cycles of each primer are shown in Table 2. The reaction mixture (25 μL) consisted of 5 ng template DNA, 2.5 μL of 10xPCR buffer (100 mM Tris-HCl, 500 mM KCl, 15 mM MgCl2), 2 μL of dNTP Mixture (2.5 mM of eath dNTP), 0.5 μM of each primer and 0.3U of Smart Taq DNA Polymerase (Abgene®, UK).

Bacterial DGGE was performed with 8% polyacrylamide gels (Acrylamide/bis-acrylamide: 37.5/1) containing 45% to 65% vertical gradient of denaturant (100% was defined as containing 7 M Urea and 40% Formamide). On the other hand, 7% polyacrylamide gels containing 25 to 55% gradient were used for fungal DGGE. Approximately 24 μL of each PCR product was loaded onto the gels. Electrophoresis was performed in 1xTAE buffer (40 mM Tris/Acetate, pH 8;1 mM EDTA) at 80 V and 60 °C for 16 h for bacterial DNA, while 65 V and 60 °C for 13.5 h for fungal DNA. After electrophoresis, gels were stained in 1xTAE containing 0.01% SYBR Green I Nucleic Acid Gel Stains (Cambrex Bio Science Rockland, Inc. Rockland, ME USA) for 30 min. The gels were visualized by UV illumination and the gel images were digitally captured by a CAMEDIA digital camera C-750 Ultra Zoom (OLYMPUS CO., LTD., Japan) with a Sybr Green filter. DGGE fingerprints were processed to waveform with Image J 3.14n (National Institutes of Health, USA) to normalize background. After background subtraction, the clustering of patterns was calculated with NTSYSpc Ver.2.11 (Biostatistics Inc.) using the unweighted pair group method using average linkages (UPGMA).


Table 2: Properties of each primer and PCR conditions
Image for - Assessing Microbial Community in Andisol Differing in Management Practices by Biochemical and Molecular Fingerprinting Methods
*: 1The size described is estimated from this experiment and/or references

Experimental Design and Statistical Analysis
Soil samples were collected according to a systematic sampling design across the S-shaped transects. The least significant differences (LSD) test was also used to determine whether differences between means were statistically significant (p<0.05). Statistical analysis was conducted by JMP 4.0 (SAS Institute, Cary, NC, USA). Unless otherwise noted, all the results were calculated using oven-dried soil (105 °C, 24 h).

RESULTS AND DISCUSSION

Microbial biomass-C was influenced by management practice (Fig. 1a) being highest in the NTg and lowest in the CT. Soil under CT had lowest microbial biomass-N (Fig. 1b). The lowest microbial biomass-C in CT associated with the lowest microbial biomass-N.

NT systems generally show higher amount of soil organic carbon (Rahman et al., 2003) as well as microbial biomass-C (Paustian et al., 1997). Franchini et al. (2007) found 80 and 104% higher microbial biomass-C and microbial biomass-N in NT than CT, respectively.

Many of the individual fatty acids were affected by management practice (Table 3). In CT soils individual fatty acids of i14:0, a14:0, 16:1ω7t, i17:0, a17:0, 18:2ω6, 18:1ω9 and 18:0 were least abundant while 3-OH 14:0, 16:1ω5c, cy17:0, 2-OH 16:0 and i18:0 had more abundance than PD and NT. On the other hand, individual fatty acids of a15:0, 15:0, 16:1ω9c, i17:1ω5, 17:0, 20:2 and 20:1ω9c did not show any observable difference. Total PLFA, aerobes, gram-positive bacteria, gram-negative bacteria, cyanobacteria and fungi were significantly higher in NT than CT (Table 4). Bardgett et al. (1996) observed that, the ratio of fungal to bacterial PLFA was higher in unfertilized upland grassland soil than in adjacent fertilized treatments. Correlation between microbial biomass-C and N and microbial community indicated that MBC and MBN were highly and positively correlated with total PLFA (r = 0.777 and 0.878, respectively), fungal PLFA (r = 0.917 and 0.868, respectively), gram-positive bacteria (r = 0.665 and 0.911, respectively), gram-negative bacteria (r = 0.911 and 0.871, respectively) and cyanobacteria (r = 0.617 and 0.873, respectively) while highly and negatively correlated with methane-oxidizing bacteria (r = -0.969 and-0.827, respectively) and mycorrhizae (r = -0.968 and -0.825, respectively). Sulfate-reducing bacteria showed nonsignificant correlation with either MBC or MBN. PLFA for methane-oxidizing bacteria and mycorrhizae were higher in CT than PD and NT while for sulfate-reducing bacteria it was higher in NTa than CT (Table 4). Methane-oxidizing microbial community is very important because of their importance for global methane cycling. Methane is considered as the most potential greenhouse gas after carbon dioxide (IPCC, 2001). Due to the fact that methane concentration in the atmosphere has more than doubled in the post-industrial era, much research effort has been expended to identify sources and sinks of methane and the organisms involved. The balance between the production of methane by methanogenic bacteria under anoxic conditions and the consumption of methane by methanotrophic bacteria (MOB) under oxic conditions determines whether a particular environment acts as a source or a sink for atmospheric methane. Submerged wetland soils like. rice are regarded as the most important source of atmospheric methane while non-flooded upland soilslike grasslands are regarded to be the only biological sink of atmospheric methane (LeMer and Roger, 2001). In both wetland and upland soils, obligate aerobic methane-oxidizing bacteria use molecular oxygen to oxidize methane to carbon dioxide and cell carbon (Hanson and Hanson, 1996). In wetland soils these bacteria are active in the surface soil layers and in the rhizosphere of oxygen releasing plants (Frenzel, 2000). Rahman et al. (2007) conducted a research on Andisol, observed that under dry condition, soils has higher methane-oxidizing bacteria although those soils have been under puddling for long time. In this study, it was observed that Andisol with puddling contained substantially less methane-oxidizing bacteria than conventional tillage under submerged condition. It is necessary to carry out intensive research on methane-oxidizing microbial communities with the special reference to Andisols.


Image for - Assessing Microbial Community in Andisol Differing in Management Practices by Biochemical and Molecular Fingerprinting Methods
Fig. 1: Microbial biomass carbon (a) and microbial biomass nitrogen (b) of temperate Andisol Bars showed the standard errors

Table 3: Phospholipid fatty acids (PLFAs: nmol g-1 soil) profiles of temperate Andisol
Image for - Assessing Microbial Community in Andisol Differing in Management Practices by Biochemical and Molecular Fingerprinting Methods
CT1: Conventional Tillage, PD2: Puddling, NTa3: No-tillage (apple), NTg4: No-tillage (grass), LSD5: (Least Significant Different) at p<0.05 between treatments (rows)

Table 4: Microbial community (nmol g-1 soil) of temperate Andilsol
Image for - Assessing Microbial Community in Andisol Differing in Management Practices by Biochemical and Molecular Fingerprinting Methods
CT1: Conventional Tillage, PD2: Puddling, NTa3: No-tillage (apple), NTg4: No-tillage (grass), LSD5: (Least Significant Different) at p<0.05 between treatments (rows)

DNA extracted from Andisols were colorless indicating no humic acid contamination. Soil DNA of sufficient purity for reproducible PCR amplification yielded 21.0 to 33.0 μg g-1 soil (Fig. 2). Highest bacterial diversity evaluated by DNA band number in DGGE analysis was observed in PD and can be arranged as: PD>CT>NTg>NTa (Fig. 3a and 4a) while, highest fungal diversity evaluated by DNA band number in DGGE analysis was observed in NTa. In this study inconsistent relationship was observed between bacterial as well as fungal abundance and diversity measured by number of PLFA peaks and DNA bands, respectively. Rahman et al. (2007) observed inconsistent relationship between bacterial abundance and diversity measured by number of PLFA peaks and DNA bands, respectively. They also observed consistent relationship between fungal abundance and diversity evaluated by the number of PLFA peaks and DNA bands. We inferred that microbial abundance could be high although the diversity might be low. In addition methods like PLFA and DGGE used for measuring microbial community in mineral soils may not give parallel results for microbial community composition in organic soils (Andisols) hence, more studies are necessary to understand the reason behind.


Image for - Assessing Microbial Community in Andisol Differing in Management Practices by Biochemical and Molecular Fingerprinting Methods
Fig. 2: Concentration of DNA reproducible PCR amplification of temperate Andisol (Bars showed the standard errors)

Image for - Assessing Microbial Community in Andisol Differing in Management Practices by Biochemical and Molecular Fingerprinting Methods
Fig. 3: DGGE fingerprint (a) and dendrogram analysis (b) of bacterial community (1, 2, 3: Conventional tillage; 4, 5, 6: Puddling; 7, 8, 9: No-tillage-apple; 10, 11, 12: No-tillage-grass)

Image for - Assessing Microbial Community in Andisol Differing in Management Practices by Biochemical and Molecular Fingerprinting Methods
Fig. 4: DGGE fingerprint (a) and dendrogram analysis (b) of fungal community (1, 2, 3: Conventional tillage; 4, 5, 6: Puddling; 7, 8, 9: No-tillage-apple; 10, 11, 12: No-tillage-grass)

The dendrogram was generated by UPGMA and percentage of similarity among the lanes was calculated taking into account the band migration distance and the relative intensity of all bands (Fig. 3b and 4b). Cluster analysis of the data from PCR-DGGE methods grouped the replicates of the different soils. For bacterial community, cluster analysis revealed closer association of the characteristics of NTa and NTg with a relative distance value of 80%. On the other hand, for fungal community, cluster analysis revealed closer association of the characteristics of NTa and PD with a relative distance value of 57%. The result suggested a linkage between fungal diversity and tillage practices in temperate Andisol. This study justified the need of measuring soil quality based on soil microbial communities. These findings indicated that management inputs, such as fertilizer, herbicide and irrigation were associated with the distinctive microbial community composition of the different management practices.

REFERENCES

1:  Atlas, R.M. and R. Bartha, 1998. Microbial Ecology: Fundamentals and Applications. 4th Edn., Benjamin Cummings, USA., ISBN-13: 9780805306552

2:  Balestra, G.M. and I.J. Misaghi, 1997. Increasing the efficiency of the plate counting method for estimating bacterial diversity. J. Microbiol. Methods, 30: 111-117.
CrossRef  |  Direct Link  |  

3:  Bardgett, R.D., P.J. Hobbs and A. Frostegard, 1996. Changes in soil fungal: Bacterial biomass ratios following reductions in the intensity of management of an upland grassland. Biol. Fert. Soils, 22: 261-264.
CrossRef  |  Direct Link  |  

4:  Bossio, D.A. and K.M. Scow, 1998. Impacts of carbon and flooding on soil microbial communities: Phospholipid fatty acid profiles and substrate utilization patterns. Microb. Ecol., 35: 265-278.
CrossRef  |  Direct Link  |  

5:  Brookes, P.C., A. Landman, G. Pruden and D.S. Jenkinson, 1985. Chloroform fumigation and the release of soil nitrogen: A rapid direct extraction method to measure microbial biomass nitrogen in soil. Soil Biol. Biochem., 17: 837-842.
CrossRef  |  Direct Link  |  

6:  Doran, J.W., D.C. Coleman, D.F. Bezdicek and B.A. Stewart, 1994. Defining soil quality for a sustainable environment. American Society of Agronomy, SSSA Special Publication No. 35, Madison, WI.

7:  Doran, J.W. and M.R. Zeiss, 2000. Soil health and sustainability: Managing the biotic component of soil quality. Applied Soil Ecol., 15: 3-11.
CrossRef  |  Direct Link  |  

8:  Franchini, J.C., C.C. Crispino, R.A. Souza, E. Torres and M. Hungria, 2007. Microbiological parameters as indicators of soil quality under various soil management and crop rotation systems in Southern Brazil. Soil Tillage Res., 92: 18-29.
CrossRef  |  Direct Link  |  

9:  Frenzel, P., 2000. Plant Associated Methane Oxidation in Rice Fields and Wetlands. In: Advances in Microbial Ecology, Schink, B. (Ed.). Vol. 16, Springer, New York, pp: 85-114

10:  Frostegard, A., A. Tunlid and E. Baath, 1993. Phospholipid fatty acid composition, biomass and activity of microbial communities from two soil types experimentally exposed to different heavy metals. Applied Environ. Microbiol., 59: 3605-3617.
Direct Link  |  

11:  Hanson, R.S. and T.E. Hanson, 1996. Methanotrophic bacteria. Microbiol. Rev., 60: 439-471.
Direct Link  |  

12:  Hill, G.T., N.A. Mitkowski, L. Aldrich-Wolfe, L.R. Emele and D.D. Jurkonie et al., 2000. Methods for assessing the composition and diversity of soil microbial communities. Applied Soil Ecol., 15: 25-36.
CrossRef  |  Direct Link  |  

13:  IPCC, 2001. Climate Change, 2001. The Scientific Basis. In: Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change, Houghton, J.T., Y. Ding, D.J. Griggs, M. Noguer and P.J. van der Linden et al. (Eds.). Cambridge University Press, New York, USA., pp: 881

14:  Kourtev, P.S., J.G. Ehrenfeld and M. Häggblom, 2003. Experimental analysis of the effect of exotic and native plant species on the structure and function of soil microbial communities. Soil Biol. Biochem., 35: 895-905.
Direct Link  |  

15:  Le Mer, J. and P. Roger, 2001. Production, oxidation, emission and consumption of methane by soils: A review. Eur. J. Soil Biol., 37: 25-50.
Direct Link  |  

16:  Mitsui, H., K. Gorlach, H.J. Lee, R. Hattori and T. Hattori, 1997. Incubation time and media requirements of culturable bacteria from different phylogenetic groups. J. Microbiol. Methods, 30: 103-110.
CrossRef  |  Direct Link  |  

17:  Muyzer, G., E.C. de Waal and A.G. Uitterlinden, 1993. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Applied Environ. Microbiol., 59: 695-700.
Direct Link  |  

18:  O'Leary, W.M. and S.G., Wilkinson, 1988. Gram-Positive Bacteria. In: Microbial Lipids, Ratledge, C. and S.G. Wilkinson (Eds.). Vol. 1, Academic Press, New York, pp: 117-201

19:  Paustian, K., H.P. Collins and E.A. Paul, 1997. Management Controls on Soil Carbon. In: Soil Organic Matter in Temperate Agroecosystems: Long-Term Experiment in North America, Paul, E.A., E.T. Elliot, K. Paustian and C.V. Cole (Eds.). CRC Press, Boca Raron, FL., pp: 227-234

20:  Porazinska, D.L., R.D. Bardgett, M.B. Blaauw, W. Hunt, A.N. Parsons, T.R. Seastedt and D.H. Wall, 2003. Relationships at the aboveground-belowground interface: Plants, soil biota and soil processes. Ecol. Monographs, 73: 377-395.
Direct Link  |  

21:  Prosser, J.I., 2002. Molecular and functional diversity in soil micro-organisms. Plant Soil, 244: 9-17.
Direct Link  |  

22:  Rahman, M.H., A. Tanaka and S. Hoque, 2003. Long-term effects of tillage on physicochemical properties of modified Andisol of northeast Honshu Island. Commun. Soil Sci. Plant Anal., 34: 1743-1757.
Direct Link  |  

23:  Rahman, M.H., A. Okobu and S. Sugiyama, 2007. Microbial community in Andisol as related to land use and tillage. Jap. J. Grassland Sci., 53: 398-399.

24:  Vainio, E.J. and J. Hantula, 2000. Direct analysis of wood-inhabiting fungi using denaturing gradient gel electrophoresis of amplified ribosomal DNA. Mycol. Res., 104: 927-936.
Direct Link  |  

25:  Vance, E.D., P.C. Brookes and D.S. Jenkinson, 1987. An extraction method for measuring soil microbial biomass C. Soil Biol. Biochem., 19: 703-707.
CrossRef  |  Direct Link  |  

26:  White, D.C., H.C., Pinkart and D.B. Ringelberg, 1996. Biomass Measurements: Biochemical Approaches. In: Manual of Environmental Microbiology, Hurst, C.J. (Ed.). ASM Press, Washington, DC., pp: 91-101

27:  White, T.J., T.D. Bruns, S.B. Lee and J.W. Taylor, 1990. Amplification and Direct Sequencing of Fungal Ribosomal RNA Genes for Phylogenetics. In: PCR Protocols: A Guide to Methods and Applications, Innis, M.A., D.H. Gelfand, J.J. Sninsky and T.J. White (Eds.). Academic Press, San Diego, CA., USA., ISBN-13: 9780123721808, pp: 315-322
CrossRef  |  Direct Link  |  

28:  Widmer, F., A. Flieβbach, E. Laczko, J. Schulze-Aurich and J. Zeyer, 2001. Assessing soil biological characteristics: A comparison of bulk soil community DNA-, PLFA- and BiologTM-analyses. Soil Biol. Biochem., 33: 1029-1036.
CrossRef  |  Direct Link  |  

29:  Zak, D.R., W.E. Holmes, D.C. White, A.D. Peacock and D. Tilman, 2003. Plant diversity, soil microbial communities and ecosystem function: Are there any links? Ecology, 84: 2042-2050.
CrossRef  |  Direct Link  |  

30:  Zogg, G.P., D.R. Zak, D.B. Ringelberg, D.C. White, N.W. MacDonald and K.S. Pregitzer, 1997. Compositional and functional shifts in microbial communities due to soil warming. Soil Sci. Soc. Am. J., 61: 475-481.
CrossRef  |  Direct Link  |  

©  2021 Science Alert. All Rights Reserved