Research Report 1.11

Measuring Changes In Soil Microbial Populations By Analysis Of Their Phospholipid Signatures


Dr. Ralph A. Chapman, J. Kohlmaier, S. Millar, and K. Henning,
Pest Management Research Centre,
Agriculture and Agri-Food Canada, 1391 Sandford St,
London, ONT, Canada, N5V 4T3

COESA Report No.:  RES/MAN-011/96

Objectives & Expected Outputs
Interpretative Summary
Technical Abstract
Table of Contents

View / Download Report  [219 KB pdf]

 

Report List | Green Plan Research


Objectives and Expected Outputs
Objectives:
  1. To establish sensitivity limits of the 31P NMR technique using a set of pure phospholipid reagents;
  2. To establish extraction protocols, minimum sample sizes, and extraction efficiencies as well as a set of phospholipid profiles for identifying the microfloral components in the soil samples using soil samples obtained from one set of the Green Plan Bioindicator sites (conventional vs. no-till vs. woodland).
Expected Outputs:
  1. To make comparisons and to establish correlations between this data set and other data sets from conventional microfloral characterizations of the soil, towards the objective of associating phospholipid profiles with specific groups of microflora. If these correlations prove successful, these associations can be used in the development of bioindicators;
  2. To develop a (bio)indicator of land use practice, based on either fatty acid or phospholipid analysis, that could be related to the underlying soil biological community, and other soil characterizations being made that relate to soil quality.
Type: Fed. Government, In-House
Spending Profile: 94-95: $8.0 K,    95-96: $8.0K;     Total: $16.0 K
Status: Available March 1998

 

INTERPRETATIVE SUMMARY

Microorganisms are the most abundant life form in soil. This soil microbial community consists of a large number of types of organisms - the actual number is so large that it will probably never be known. The biomass of each type present, the microbial community structure, is affected by many factors such as nutrient level and source, temperature, moisture, available oxygen and the presence of other types.

These organisms affect many aspects of agriculture-related activities from being essential for plant growth as providers of necessary nutrients, through such processes as organic matter decomposition and nitrogen fixation, to being unwanted pests as the agents of plant diseases. Their universal presence and the sensitivity of their community structure to changes in their environment also makes them good candidates for indicators of changes in soil quality.

To use them as indicators, simple, reliable and sensitive methods of measuring changes in the soil microbial community structure must be available. One method under development is based on the chemical composition of the phospholipids present in the cell walls of the microorganisms. Phospholipids are a major cell wall component and their chemical composition differs for different types of organisms thereby providing a link between chemical composition and community structure.

In this study we examined the merits of two different measures of phospholipid composition, fatty acid composition and class composition, for detecting changes induced by differences in tillage practice at four paired field sites. Fatty acid composition was determined by gas chromatograph and differences were observed between the paired sites, between the till and no-till treatments and between high and low plot elevations.

Class composition was determined by 31P NMR spectroscopy and consistent differences between tillage treatments were also found. The complexity of the differences in the fatty acid composition and the limited state of our knowledge of the composition of most soil organisms restricted the conclusions the could be drawn about the changes in community structure involved.

TECHNICAL ABSTRACT

Well documented, gas chromatography (gc)-based procedures were used to separate the fatty acids derived from the mixture of phospholipids extracted from cultured microorganisms (4 bacteria, 2 fungi and 1 actinomyces) and from 28 soil samples from four paired field sites (8 plots) under conventional tillage and no tillage treatments respectively. Twenty samples were taken from high elevations and 8 samples were taken from low elevations of a single pair. The fatty acid mixtures from the cultured microbes were relatively simple consisting to 3-10 major components which together accounted for 90+% of the total material.

Differences between the organisms tested were readily observable. In contrast, the fatty acid mixtures from the soil samples consisted of numerous components at similar low concentrations such that the 20-30 largest components accounted for only 75% of the total material. Differences in these samples were not easily detected by visual comparison. The percent composition of the fatty acid mixture was calculated for each of the soil samples based on the area of the gc response for each component and the total area for all components. A subset of the data with components >0.5% in at least one sample was selected (64 components).

The relative concentration of each of these components in the 28 samples was calculated and components which exhibited similar relative concentration among the samples, ie. a high correlation of changes in concentration with treatment, were grouped by hierarchical cluster analysis (6 clusters). Representative components were selected from each cluster and the percent composition data for these representative components in the 28 samples was tested by MANOVA to assess the significance of differences and by discriminant analysis to identify the components (clusters) most responsible and the samples which could be correctly classified based on the differences observed.

Significant differences were found between the four pairs of plots (P<0.001), the two tillage treatments (P<0.001) and the two elevations within the plots (P<0.002). Discriminant functions based on the representative components correctly classified all 28 samples among the four paired sites and between the two tillage treatments. The 16 samples taken at 2 elevations at 1 site also were all correctly classified. Components from all six clusters had large coefficients in the site discriminant function, but only four clusters were strongly represented in the tillage function. The representative components involved were: antiso-15:0, 16:1(9c), 16:0 and cyclo-19:0(9,10) usually associated with bacteria, 18:2(9,12) usually associated with fungi and an as yet unidentified fatty acid.

A 31P NMR-based procedure was adapted to provide high quality spectra of the phospholipids extracted from microbes and soils. Line widths of 3 Hz or less could be obtained even for crude total lipid extracts. The chemical shifts of some phospholipid classes present were found to be both concentration and matrix dependent which made assignment of the signals somewhat difficult when total lipid extracts were analysed. Some of this uncertainty could be removed by separating the phospholipids present by column chromatography and diluting to similar concentrations before recording the spectra. A total phospholipid concentration of ca. 8.3 mg/mL was selected as providing a good compromise between an acceptable matrix effect and a reasonable data acquisition time (4 - 5 hr).

The 25 mg quantity of phospholipids required could typically be obtained from 250 - 500 g of soil. The spectra of the phospholipids from the fungi and actinomyces examined (4 genera) differed both qualitatively and quantitatively suggesting that phospholipid class composition had potential as a measure of changes in microbial community structure. Lipids were extracted from 8 soil samples from the four paired field sites. 31P NMR spectra of the phospholipids in the total lipid extracts were recorded. Differences were observed in the phospholipid class composition estimated from the spectra. Statistical analysis of the data was made difficult by the fact that two of the four major components were found to be highly correlated in the discriminant analysis by site.

Only 6 of the 8 samples were classified correctly when three components were used in the analysis. Analysis based on the three major components and a minor component classified all 8 samples correctly with high probability. The major components were not correlated in the analysis by tillage and the resulting discriminant function classified all 8 samples correctly between the two tillage types with high probability. The treatment related differences observed in phospholipid class composition suggest that it has potential as a tool for detecting changes in soil microbial community structure.

 

Table of Contents

INTERPRETATIVE SUMMARY 1
TECHNICAL ABSTRACT 2
OBJECTIVES 4
METHODOLOGY 5
Microbial Culture 5
Soil Sources 5
Lipid Extraction 5
Phospholipid Separation 6
Transesterification of the Phospholipids 6
Gas Chromatographic Analysis of
Fatty Acid Methyl Esters
6
31P NMR Analysis 6
Data Analysis 7
RESULTS AND DISCUSSION 8
Introduction 8
GC-Based Phospholipid-Derived
Fatty Acid Analysis
9
31P NMR-Based Phospholipid Class Analysis 12
SUMMARY 15
REFERENCES 16

 


 

Report List  
Green Plan Research Home | Canada-Ontario GREEN PLAN

Last Updated: May 16, 2011 02:33:51 PM