p!

q n 5&9”...

.rt

 

 

. V rl,..n.4:.susm. n. .a..MWm.1 .M
. 7.x. 47...; an... ..¥..«.< .1 , . . _ ..
. . 0‘ . . A ' A . ‘
, .5. . A

 

 

 

 

r,

 

IHIIHHIHHIIIllUHHNHIHWilli“!INIIHIIIUIHI

i’HESIs 293 01565 0173

 

l==Tlalmtww
Mlchlgan State

Un‘livenolty

 

 

 

This is to certify that the

dissertation entitled

LIPIDS: THEIR VALUE AS MOLECULAR MARKERS AND THEIR
ROLE IN THE CARBON CYCLE OF ARBUSCULAR FUNGI

presented by

Francisco Jose Calderon

has been accepted towards fulfillment
of the requirements for

/’{0 degreein Crdg 45/ 50:. /

S Q;(hC05

(fi‘fl/fl/

Major professor

 

Date ////7/f7

MSUL: an Affirmative Action/Equal Opportunity Institution 0- 12771

 

 

 

PLACE IN RETURN BOX to MOI. this checkout from your record.
TO AVOID FINES return on or before date duo.

DATE DUE DATE DUE DATE DUE

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1
* WW?

MSU to An Afflmottvo Action/Equal Opportunity Instituton

 

 

 

 

 

LIPIDS: THEIR VALUE AS MOLECULAR MARKERS AND THEIR ROLE IN THE
CARBON CYCLE OF ARBUSCULAR FUNGI

By

Francisco Jose Calderon

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Crop and Soil Science

1997

ABSTRACT

LIPIDS: THEIR VALUE AS MOLECULAR MARKERS AND THEIR ROLE IN THE
CARBON CYCLE OF ARBUSCULAR FUNGI

By

Francisco J. Calderon

Arbuscular mycorrhizae are associations between fungi and the
roots of vascular plants. Part of this dissertation is devoted to analyzing
the fatty acid and sterols of mycorrhizal and non-mycorrhizal Sorghum
with the aim of identifying universal molecular markers of mycorrhizal
infection. The mycorrhizal fungi contain high amounts of unusual
lipids that may be used to mark their presence in infected roots. My
results show that phospholipid fatty acid 16:1, as well as campesterol are
molecules that can be used to consistently identify mycorrhizal
infection. In addition, lipid profiles may provide insight as to which
fungal species is present in the roots.

In a second experiment, the fatty acids and sterols of several
isolates of root pathogenic fungi were surveyed to assess the taxonomic

value of lipid profiles. My results show that the genera Rhizoctonia and

Pythium can be reliably identified because of their characteristic lipid
composition.

Another question that Iaddress in this work is: What is the C
turnover time of mycorrhizal lipids? For this purpose, Icarried out a
pulse-chase experiment in which I followed the incorporation and
subsequent turnover of C in mycorrhizal lipids. Mycorrhizal and non-
mycorrhizal plants were subjected to a pulse exposure to 14C02, followed
by sequential harvesting. Infected plants assimilated more ”C than non-
mycorrhizal plants, and had a higher absolute and percentage allocation
of 1“C to root tissue, below ground respiration, and soil. Despite the
increased fixation of C by mycorrhizal plants, mycorrhizal shoots had
reduced biomass. This indicates that the C drain imposed by the fungus
results in a reduced shoot growth, suggesting that the mycorrhizal fungus
was acting as a parasite. The pulse-chase experiment demonstrated that
the lipids of mycorrhizal roots are a dynamic pool of C with measurable
turnover of 1‘C. The C turnover time of the mycorrhizal fatty acid 16:1
(05 was calculated at 210 h". The lipids of non-mycorrhizal roots
incorporated less radiolabel, underscoring the difference in the lipid C
cycle between the arbuscular mycorrhizae and non-mycorrhizal roots.
To my knowledge, this is the first measurement of the C turnover of a

biomass component of the mycorrhizal fungus.

To my wife and family.

iv

ACKNOWLEDGMENTS

I am grateful to Dave Harris from the Department of Crop and Soil
Sciences for all his help with the design of the radiotracer experiment
and for his valuable advice regarding the data and calculations. Thanks
to M. J. Klug at the W. K. Kellogg Biological Station for the use of the
FAME facility, and to S. Crespo, H. Corlew-Newman, and A. O’Neill for
their assistance in the FAME analysis. David Ringelberg, Julia Stair, and
David C. White at the University of Tennessee provided the equipment
and expertise for the sterol and PLFA analysis, and contributed with a
thorough discussion of the results. Iam grateful to Dave Schultz of the
Department of Botany and Plant Pathology for the help with the TLC
analysis and for his assistance in the interpretation of the data.
Professor J. Ohlrogge provided the equipment and materials necessary for
the gas chromatography and radiography of the fatty acids. Iwant to
acknowledge Robert Blackburn and the Phoenix Memorial Laboratory at
the University of Michigan, for his help with the sterilization of the soils.

Finally, Iwant to show my appreciation to my thesis committee
for their advise and their suggestions for improvement of the

manuscript. My advisor, E. A. Paul, has been both a friend and a

V

mentor, and Isincerely appreciate all the motivation, support, and
guidance I received.

The funds for this work were provided by The NSF Center for
Microbial Ecology (NSF BIR 912006) and the KBS Long-term Ecological

Research Project (NSF DEB 8702332).

vi

TABLE OF CONTENTS

LIST OF TABLES .................................................................................... x
LIST OF FIGURES ................................................................................. xii
INTRODUCTION .................................................................................. 1
Lipids as Molecular Markers ....................................................... 1
Structure and Function of Lipids ................................................ 5
The Lipid Composition of Arbuscular Fungi ................................. 6
The Carbon Cycle and Arbuscular Fungi ...................................... 8
About the Dissertation ............................................................... 9
LISTOFREFERENCES ................................................................... 12
CHAPTER 1
THE CHEMICAL TAXONOMY OF THREE LIPID
FRACTIONS FROM ROOT-ASSOCIATED FUNGI OF SORGH UM .................... 2 3
Abstract ................................................................................... 23
Introduction ............................................................................ 25
Materials and Methods ............................................................. 27
Broth Cultures ................................................................. 28
Harvests .......................................................................... 28
Analysis of Whole—Cell Fatty Acids (FAME) ......................... 29
Analysis of Phospholipid Fatty Acids (PLFA)
and Sterols ...................................................................... 30
Statistical Analysis ........................................................... 33
Results and Discussion .............................................................. 33
Sterols ............................................................................. 33
Phospholipid Fatty Acids (PLFA) ........................................ 35
Whole-Cell Fatty Acids (FAME) .......................................... 36
Specialized Structures ...................................................... 38
Lipids and Environmental Samples ................................... 39
Conclusions ..................................................................... 40
LIST OF REFERENCES ................................................................... 4 2

vii

CHAPTER 2
IDENTIFICATION OF POSSIBLE MOLECULAR

MARKERS IN THE LIPIDS OF ARBUSCULAR MYCORRHIZAE ........................ 6 5
Abstract ................................................................................... 65
Introduction ............................................................................ 67
Materials and Methods ............................................................. 69

Soil, Inoculum, and Growth Conditions ............................. 69
Root Harvesting and Preservation ...................................... 71
Whole-Cell Fatty Acids (FAME) .......................................... 71
Phospholipid Fatty Acids (PLFA) ........................................ 72
Sterols ............................................................................. 73
Microscopic Determination of Percent Mycorrhizal
Infection ......................................................................... 74
Statistical Analysis ........................................................... 74
Results and Discussion .............................................................. 75
Whole-Cell Fatty Acids (FAME) .......................................... 75
Phospholipid Fatty Acids (PLFA) ........................................ 78
A Comparison of FAME and PLFA ....................................... 79
Sterols ............................................................................. 80
Multivariate Analysis ....................................................... 81
Mycorrhizal Infection as Determined by
Microscopy ...................................................................... 82
Conclusions ..................................................................... 83
LIST OF REFERENCES ................................................................... 8 5

CHAPTER 3

PATTERNS OF CARBON ALLOCATION AND

FATTY ACID TURNOVER IN MYCORRHIZAL SORGH UM ............................ 9 7
Abstract ................................................................................... 97
Introduction ............................................................................ 98
Materials and Methods ........................................................... 101

Environmental Conditions and Inoculation ..................... 102
1‘C02 Pulse Labelling ....................................................... 103
Harvests ........................................................................ 103
Plant Biomass ................................................................ 103
Shoot Phosphorus and 1‘C Content .................................. 104
Root l‘C Content and Lipid Analyses ............................... 105
Below-Ground Respiration .............................................. 109
Sofl ............................................................................... 110
Statistical Analysis ......................................................... 112
Results and Discussion ............................................................ 112
Carbon Allocation .......................................................... 112

viii

Allocation to Plant Tissues ............................................. 115

Allocation to Below-Ground Respiration .......................... 117

Allocation to Soil ........................................................... 119
Concentration of Lipids in Mycorrhizal and Non-

Mycorrhizal Roots .......................................................... 120

Radiolabel and Fine Root Lipids ...................................... 122

Turnover of Root Fatty Acid 16:1 (05 ............................... 125

Conclusions ................................................................... 129

LIST OF REFERENCES ................................................................. 131

OVERALL CONCLUSIONS .................................................................... 1 4 7

About the Lipid Composition of Mycorrhizae ............................ 147

About the C Cycle of the Lipids of Arbuscular Fungi .................. 150

LIST OF REFERENCES ................................................................. 15 2

APPENDD(A .................................................................................... 153

APPENDIXB .................................................................................... 154

ix

LIST OF TABLES

INTRODUCTION

Table 1. Examples of several fatty acids found in fungi ....................... 17
Table 2. Marker fatty acids and sterols in the phospholipid

fraction of several groups of organisms ................................................ 18
Table 3. Examples of studies showing the effect of

mycorrhizae in the C allocation pattern of the host ............................. 2O
CHAPTERl

THE CHEMICAL TAXONOMY OF THREE LIPID
FRACTIONS FROM ROOT-ASSOCIATED FUNGI OF SORGH UM
Table 1. Root-pathogenic fungi analyzed for FAME,

PLFA and sterols ................................................................................. 46
Table 2. Mean percent of total sterols recovered from each

of the root pathogenic fungi ............................................................... 47
Table 3. Mass spectral characteristics of sterols recovered

from root pathogenic fungi ................................................................ 51
Table 4. Molecules giving the highest loadings for the

components ...................................................................................... 52
Table 5. Most abundant PLFA in the fungal root pathogens

of Sorghum ...................................................................................... 5 3
Table 6. Most abundant FAME in the fungal root pathogens

of Sorghum ...................................................................................... 57
CHAPTER2

IDENTIFICATION OF POSSIBLE MOLECULAR MARKERS IN THE LIPIDS OF
ARBUSCULAR MYCORRHIZAE
Table 1. Mass spectral characteristics of sterols recovered

from mycorrhizal and non-mycorrhizal roots ...................................... 89
Table 2. Microscopical estimates of infection ...................................... 90
Table 3. FAME of mycorrhizal and non-inoculated Sorghum

roots .............................................................................................. 91
Table 4. PLFA of mycorrhizal and non-inoculated Sorghum

roots .............................................................................................. 93
Table 5. Sterols of mycorrhizal and non-inoculated Sorghum

roots .............................................................................................. 95

X

CHAPTER 3

PATTERNS OF CARBON ALLOCATION AND FATTY ACID
TURNOVER IN MYCORRHIZAL SORGH UM

Table 1. Average distribution of labeled C per individual for

the 24 day harvest period ................................................................ 136
Table 2. Plant biomass (g) after the pulse-label ................................. 137
Table 3. Total radiolabel content (MBq per plant) of the

plant tissues and soil ....................................................................... 138
Table 4. Results of the ANOVA for the plant biomass, soil, and

respiration variables. ....................................................................... 1 39
Table 5. Specific activity (MBq/gC) of the plant tissues and

soil ............................................................................................ 140
Table 6. Lipid concentrations in fine roots (mg/g) ............................ 141
Table 7. Total radiolabel in the fine root biomass accounted

for by each lipid type (KBq/plant) .................................................... 142
Table 8. Results of the ANOVA for the lipid concentration

and radiolabel content of fine roots .................................................. 143

xi

LIST OF FIGURES

INTRODUCTION

Figure 1. General structure of a phospholipid and a

triacylglycerol ................................................................................... 21
Figure 2. Examples of different sterols ................................................. 22
CHAPTER 1

THE CHEMICAL TAXONOMY OF THREE LIPID FRACTIONS
FROM ROOT-ASSOCIATED FUNGI OF SORGH UM

Figure 1. Principal components anlyses of the lipid profiles ................. 61
Figure 2. Tree diagram of the sterol profiles from the

mycelium, conidia, or sclerotia ........................................................... 62
Figure 3. Tree diagram of the phospholipid fatty acid profiles

from the mycelium, conidia, or sclerotia ............................................. 63
Figure 4. Tree diagram of the whole-cell fatty acid profiles

from the mycelium, conidia, or sclerotia ............................................. 64
CHAPTER2

IDENTIFICATION OF POSSIBLE MOLECULAR MARKERS IN
THE LIPIDS OF ARBUSCULAR MYCORRHIZAE
Figure 1. Hierarchical cluster analysis of the FAME profiles .................. 9 6

CHAPTER3

PATTERNS OF CARBON ALLOCATION AND FATTY ACID

TURNOVER IN MYCORRHIZAL SORGHUM

Figure 1. Argentation chromatography of the fatty acid

extracts from M and NM Sorghum roots ............................................ 144
Figure 2. 1“C trapped from the below-ground atmosphere ................. 145
Figure 3. Ln(A.) versus time for the 1‘C turnover in the

16:1 (05 fatty acids from the roots of Sorghum infected with

G. clarum .................................................................................... 146
APPENDD(A
FAME, PLFA, and sterol analysis of roots and fungal material .............. 153

xii

APPENDIXB
Root lipid extraction and radiation measurements ............................ 154

xiii

INTRODUCTION

Lipids as Molecular Markers

Scientists are confronted with the problem of quantitatively and
qualitatively isolating and describing the components of microbial
communities. This process is often limited by the fact that microbes
exist as part of a consortium of mutually dependent metabolic types, or
as with arbuscular mycorrhizal fungi, rely on a host for their nutrition.
In addition, bacteria and fungi may exist attached to surfaces, embedded
in biofilms, or live inside the host’s tissues. The analysis of nucleic acids
has shown that soil communities contain a high diversity of
microorganisms that are not yet culturable in the laboratory (Torsvik,
1990). This is an important limitation since studies of microorganisms
in pure culture are likely to miss important members of the microbiota.
In addition, laboratory conditions may not be able to replicate the
physical and biological conditions that the microbes find in their
natural habitat. It is for this reasons that microbial ecologists rely on
the direct analysis of unusual molecules or nucleic acid sequences to

detect different groups of microbes in situ.

.1

The extraction of specific signature molecules such as DNA and
lipids has helped to identify a microbial diversity that could not be
observed by subculturing methods (Holben et al., 1988; White, 1994,
Cavigelli et al., 1995). Lipids are a set of molecules that has been used to
describe the microbial communities of soils (White, 1994). Lipid analysis
involves the extraction, derivatization of the lipids present in the
sample, followed by identification and quantification of the molecules
using chromatography,. Fatty acids are a group of lipids that may be
particularly valuable for differentiating microbial taxa. Fatty acids vary
in the number of carbons and number of unsaturations. Moreover, fatty
acids with the same number of double bonds and carbon atoms may
exist in a variety of positional and geometric isomers. Hundreds of
different fatty acids exist in microorganisms (Harwood and Russell,
1984). Table 1 gives some examples of fatty acids found in fungi.

Fatty acids may exist free in the cytoplasm or bound to molecules
such as phospholipids, triacylglycerols, and diacylglycerols. Many
different kinds of fatty acids may be present in the cells of a given
species. The number of different lipids and their relative amounts is
called the fatty acid profile of a sample. In some instances, fungal
isolates may produce a similar diversity of fatty acid molecules but
differ significantly in the relative amounts of the fatty acids. However,

in other occasions fungi can be differentiated by the type of fatty acids

that they produce as well as the relative content of each molecule (Stahl
and Klug, 1996). Because of these reasons, a commercially available
method for analyzing fatty acid profiles has proven useful for
characterizing large and diverse groups of fungal isolates (Stahl and Klug,
1996).

There are several microbial taxa that contain unusual molecules in
their lipid profiles (Table 2). These peculiar lipids may serve as
molecular markers for detecting specific groups of microorganisms in the
environment. For instance, fatty acids such as 18:2 m6 are relatively
abundant in fungal lipids (Vestal and White, 1989; Stahl and Klug,
1996). The polyenoic fatty acids are those that have more than one
carbon to carbon double bond. The concentration of polyenoic fatty
acids in natural samples has been used to detect shifts in fungal
abundance. It has been observed that polyenoic fatty acids increase in
habitats manipulated to make fungi better competitors than other kinds
of microorganisms (White et al., 1980).

The analysis of lipid profiles provides quantitative and
physiological information that may not be obtained by nucleic acid
analysis alone (White, 1994). Phospholipids are ionic molecules
composed of 1,2 diacylglycerol with a phosphodiester link that connects
the glycerol to a polar head group that is usually a nitrogenous base

(Figure l). The phospholipids are cell and membrane components and

are rapidly degraded after death (White et a1. 1979). The fatty acids
that are ester-linked to phospholipids are referred to as the phospholipid
fatty acids (PLFA). Since the PLFA can be extracted, characterized, and
quantified, the presence of PLFA markers in environmental samples may
serve as an index of the bacterial and fungal biomass in soil (Tunlid and
White, 1979; White et al., 1980; Petersen et al. 1991). An advantage 'of
analyzing PLFA rather than the total fatty acid content of the sample is
that the fatty acids bound to storage molecules are excluded. Storage
fatty acids such as those bound to triacylglycerides (Figure 1) may vary
in abundance depending on the nutritional status of the microflora
(White, 1988). Possible problems for the study of microbial communities
using PLFA analysis is that changes in physical and nutritional
conditions may alter the fatty acid composition of microbes (White,
1988), and a minimal amount of 107 cells is required for adequate
sensitivity (White, 1994).

Other lipids besides fatty acids have value as molecular markers.
Sterols are characterized by a common four-ringed structure of 1,2-
cyclopentano-perhydrophenanthrene (Fig.2). Different substituents to
the rings, methyl groups, and position and amount of unsaturations
characterize individual sterols (Figure 2). A wide variety of sterols have
been identified in fungi, while sterols are notably absent from most

bacteria, insects, and nematodes (Weete, 1980). Unusual sterols have the

potential to serve as molecular markers of eukaryotes in the same way as
fatty acids are used for other microbial groups. For example, ergosterol is
abundant in Ascomycetes and Deuteromycetes and has been used as an
indicator of fungal biomass as well as an indicator of fungal growth rate
in field samples (White et al., 1980; Newell et al., 1988; Newell and

Fallon, 1991).

Structure and Function of Lipids

Why do fungal species have a variety of fatty acids and sterols?
The diversity in the structure of lipids suggests a corresponding
functional diversity. Membrane components such as sterols and
phospholipids are thought to modulate the transport into and out of
the cell by their effect on the viscosity of the lipid bilayer (Weete, 1980).
It is generally known that increasing the amount of unsaturation of the
fatty acids causes changes in the melting point of the lipids to which the
fatty acids are bound. Unsaturated fatty acids remain fluid at lower
temperatures than saturated fatty acids. In fungi, this pattern has
implications for the adaptation to cold environments. For example, the
relative content of unsaturated fatty acids in some fungi increase when
they are cultured under lower temperatures (Weete, 1980). With some
exceptions, psychrophylic microorganisms have a higher proportion of

shorter and/or monounsaturated fatty acids than mesophylic

microorganisms (Harwood and Russell, 1984). Sterols are another
component of fungal membranes. It has been hypothesized that sterols
have a functional role on membrane permeability because of their
impact on the viscosity of lipids (Weete, 1980). Sterols have also been
associated with other functions such as acting as precursors of steroid
hormones that modulate fungal reproduction (Weete, 1980), and as
regulatory agents of fungal enzymes (Weete, 1989).

A different group of lipids with important functional implications
on fungi are the triacylglycerides and their bound fatty acids (Figure 1).
Triacylglycerides are the main carbon reserves in fungi and may be
oxidized for energy or used as a carbon source for the synthesis of sugars
(Weete, 1980). As a general rule, bacteria are different from fungi in the
sense that bacteria do not accumulate C in the form of triacylglyceride,

and use poly-hydroxy butyrate instead (Weete, 1980).

The Lipid Composition of Arbuscular Fungi

Arbuscular mycorrhizal fungi (AMF) are obligate root symbionts
with low host specificity and a wide geographic distribution. Six genera
from three families have been recognized based on diagnostic spore
morphology (Morton and Benny, 1990) and nucleic acid sequences
(Simon et al., 1993). The Acaulosporaceae includes the genera

Acaulospora and Entrophospora, the Glomaceae contains the genera

Glomus and Sclerocystis, and the Gigasporaceae includes the genera
Gigaspora and Scutellospora. Traditionally, estimating mycorrhizal _
formation involves root clearing and staining followed by microscopical
examination. This process is prone to observer differences, cannot
distinguish infection from the various AMF genera, and often lacks
correlation to the cost/benefit balance of the AMF-root symbiosis. Other
approaches for quantifying AMF have shown limitations. The
measurement of chitin and AMF associated pigments may not show a
close correspondence with microscopical evaluation of infection (Schmitz
et al., 1991). Fluorescein diacetate staining is not reliable for measuring
infection or arbuscularization because of the natural autofluorescence of
the AMF and interference of root pigmentation (Ames et al., 1982).
Specific fatty acids can be used as molecular markers of AMF at the
genus level (Graham et al., 1995). Roots infected with Glomus contain
several lipids in higher concentrations relative to non-mycorrhizal roots,
and this is true across different plant species (Graham et al., 1995; Nagy
et al., 1980; Nordby et al., 1981; Pacovsky, 1987; Pacovsky and Fuller,
1988; Pacovsky, 1989). Most of the previous work done on AMF fatty
acids has dealt with whole-cell fatty acid (FAME) extracts. These studies
show that a significant fraction of whole-cell fatty acids of several AMF
genera may be comprised of the unusual fatty acid 16:1c05 (Graham et

al., 1995). However, there is limited information about specific lipid

fractions such as the phospholipid fatty acids and sterols of mycorrhizal
roots. One of my objectives is to analyze the PLFA and sterols of AMF

genera that have not been studied to date.

The Carbon Cycle and Arbuscular Fungi

Plants allocate a large fraction of their fixed C to roots and the
associated mycorrhizal fungi. Mycorrhizal roots may contribute
amounts of C to soil that are comparable to the C inputs from the leaf
litter, and a significant fraction of the C turnover in a forest could be
attributed to fine roots and their associated fungi (Paul and Clark, 1996).
Arbuscular fungi and root tissue are equivalent in their ecological
function because both are dependent on soil resources such as P and N,
while relying on direct transfers of photoassimilated C from
aboveground. However, little is known about how the productivity and
mineralization of C in root tissue differs from that in the fungus. Several
studies using different plant-fungus combinations show that the AMF
affects the allocation pattern of C to the host’s tissues and soil. Table 3
shows that mycorrhizal infection has a particularly strong effect on the
distribution of C below—ground by increasing the allocation to root
respiration and root exudation.

The dominant cytoplasmic constituent of the arbuscular fungus is

lipid (Cox and Sanders, 1974). The amount of C stored in the root lipids

of different plant species increases when they are infected with Glomus
(Cox and Sanders, 1974; Cooper and Lose], 1978; Nagy et al., 1980; Peng et
al., 1993). Because of this, the analysis of fatty acids in infected roots
has shown value for estimating fungal carbon cost and the development
of colonization (Graham et al., 1995). However, there has been no
attempt of measuring the residence time of C in the mycorrhizal lipids.
This is relevant since elucidating the C cycle of the mycorrhizal fungus
will help to understand an important part of the nutrient cycling in

root-microbe symbioses.

About the Dissertation

The unifying theme of this work is the lipids of arbuscular fungi.
The overall goal of this thesis is to evaluate particular lipids as
molecular markers of arbuscular fungi in roots, and also provide insight
into the role of lipid molecular markers in the carbon cycle of the
arbuscular fungus-host association. Sorghum has been chosen as a
common host species for all the experiments. Sorghum is an important
crop that is known to form arbuscular mycorrhizae, and it’s lipid
composition is affected by mycorrhizal infection (Pacovsky, 1989).

The dissertation is divided into three experiments. In the first
chapter, I studied the lipids of known fungal root pathogens of Sorghum.

The purpose of this work is to determine if the molecules proposed as

10

markers of arbuscular fungi are present in other rhizosphere microbes.
This is a necessary step for validating a molecular marker, since the
presence of the marker in other microbes limits the value of the marker
analysis. In addition to the survey, Iproduced a chemical taxonomy of
the fungal root colonizers of Sorghum by subjecting the lipid profile data
to multivariate statistical analysis.

The second chapter of this dissertation is a study of the fatty acids
and sterols of the arbuscular fungus in situ. Ianalyzed the sterols and
PLFA of fungal genera for which this information is lacking with the
object of identifying possible universal markers of arbuscular fungi. In
addition to the marker aspect, Ianalyzed the complete set of molecules
using multivariate analysis. The multivariate analysis illustrates the
value of the lipid profiles to distinguish the mycorrhizae of different
fungal taxa from non-mycorrhizal roots.

The aim of the final chapter was to measure the turnover rate of
C in the lipids of the mycorrhizal fungus. The cost of the mycorrhizal
association to the plant has been investigated in only a few cases, and
nobody has been able to estimate turnover rates of the fungal partner.
In order to achieve this, I carried out a pulse-chase experiment in which I
exposed mycorrhizal plants to 1"C02, and then followed l‘C incorporation
and subsequent disappearance in the fungal fatty acids. Overall, fatty

acids were used to monitor the populations of arbuscular fungi in the

ll

roots of Sorghum, as well as to measure the turnover time of carbon in

this important soil organism.

LIST OF REFERENCES

LIST OF REFERENCES

Ames, R. N., Ingham, E. R., and Reid, C. P. P. 1982. Ultraviolet
induced autofluorescence of arbuscular mycorrhizal root
infections: an alternative to clearing and staining methods for
assessing infections. Can. J. Microbiol. 28, 351-355.

Cavigelli, M. A., Robertson, G. P., and Klug, M. J. 1995. Fatty acid
methyl esters (FAME) profiles as measures of soil microbial
community structure. Plant and Soil. 170, 99-113.

Cooper, K., and Losel, D. 1978. Composition of lipids in roots of onion,
clover, and ryegrass infected with Glomus mosseae. New
Phytol. 80, 143-151.

Cox, G., and Sanders, F. 1974. Ultrastructure of the host-fungus
interface in a vesicular-arbuscular mycorrhiza. New Phytol. 73,
901-912.

Fedrele, T. W. 1986. Microbial distribution in soil. pp. 493-498. In
Perspectives in Microbial Ecology. F. Megusar and Gantar, M (eds).
Slovene Society for Microbiology, Ljubljana.

Graham, J. H., Hodge, N. C., Morton, J. B. 1995. Fatty acid methyl ester
profiles for characterization of glomalean fungi and their
endomycorrhizae. Applied. Environ. Microbiol. 61, 58-64.

Harris, D., Pacovsky, R., and Paul, E. 1985. Carbon economy of
soybean-Rhizobium-Glomus associations. New Phytologist. 101,
427—440.

12

13

Harwood, J. L., and Russell, N. J. 1984. Major lipid types in plants and
microorganisms. pp. 7-32. In Lipids in Plants and Microbes.
Harwood, J. L. and Russell, N. J. (eds). Allen & Unwin,
Winchester Massachusetts.

Holben, W. E., Jansson, J. K., Chelm, B. K., and Tiedje, J. M. 1988. DNA
probe method for the detection of specific microorganisms in
the soil bacterial community. Applied and Environmental
Microbiology. 54. 703-711.

Jabaji-Hare, S. 1988. Lipid and fatty acid profiles of some vesicular-
arbuscular mycorrhizal fungi: contribution to taxonomy.
Mycologia 80, 622-629.

Jakobsen, I., and Rosendahl, L. 1990. Carbon flow into soil and
external hyphae from roots of mycorrhizal cucumber plants.
New Phytologist. 115, 77-83.

Kroppenstedt, R. M. 1985. Fatty acids and menaquinone analysis of
actinomycetes and related organisms. pp. 173-199. In
Chemical Methods in Bacterial Systematics. M. Goodfellow and
D. E. Minnikin (eds). Academic Press, London.

Morton, J. B., and Benny, G. L. 1990. Revised classification of
Arbuscular mycorrhizal fungi (Zygomycetes). A new order
Glomales, two new suborders Glomineae and Gigasporineae, and
two new families, Acaulosporaceae and Gigasporaceae, with an
emendation to the Glomaceae. Mycotaxon. 37, 471-492.

Nagy, S., Nordby, H. E., and Nemec, S. 1980. Composition of lipids in
roots of six citrus cultivars infected with the vesicular-arbuscular
mycorrhizal fungus Glomus mosseae. New Phytol. 85, 377-384.

Newell, S., Y., Arsuffi, T. L., and Fallon, R. D. 1988. Fundamental
procedures for determining ergosterol content of decaying plant

14

material by liquid chromatography. Applied and Environmental
Microbiology. 54, 1876-1879.

Newell, S., Y., Fallon, R. D. 1991. Toward a method for measuring

instantaneous fungal growth rates in field samples. Ecology.
72, 1547-1559.

O'Leary, W., M., and Wilkinson, S. G. 1988. Gram-positive bacteria.
pp. 117-201. In Microbial Lipids. C. Ratledge and S. G.
Wilkinson (eds). Academic Press, London.

Pacovsky, R. S. 1989. Influence of inoculation with Azospirillum
brasiliense and Glomus fasciculatum on Sorghum nutrition. pp.
35-239. In nitrogen fixation with non-legumes. F. A. Skinner
et al. (Eds). Kluwer Academic Publishers.

Pacovsky, R. S., and Fuller, G. 1989. Mineral and lipid composition of
Glycine-Glomus-Bradyrhizobium symbioses. Physiologia
Plantarum. 72, 733-746.

Paul, E., and Clark, F. E. 1996. In Soil Biology and Biochemistry.
Academic Press, New York.

Paul, E., and Kucey R. 1981. Carbon Flow in plant-microbial
associations. Science. 213, 473-474

Peng, S., Eissenstat, D. M., Graham, K. W., and Hodge, N. C. 1993.
Growth depression in mycorrhizal citrus at high phosphorus
supply. Analysis of carbon costs. Plant Physiol. 101, 1063- 1071.

Petersen, S. O., Henriksen, K., Blackburn, T., H., and King, G. M. 1991.
A comparison of phospholipid and chloroform fumigation
analyses for microbial biomass in soil: potentials and limitations.
FEMS Microbiology Ecology. 85, 257-268.

15

Schmitz, O., Danneberg, B., Hundeshagen, A., Klingner, A., and Bothe,
H. 1991. Quantification of vesicular-arbuscular mycorrhiza by
biochemical parameters. J. Plant Physiol. 139, 106-114.

Simon, L., Bousquet, J., Levesque, R. C., and Lalonde, M. 1993. Origin
and diversification of endomycorrhizal fungi and coincidence
with vascular land plants. Nature. 363, 67-69.

Stahl, P. D., and Klug, M. J. 1996. Characterization and differentiation
of filamentous fungi based on fatty acid composition. Applied
and Environmental Microbiology. 62, 4136-4146.

Torsvik, V. L., Gosk¢yr, J., and F. I. Daae. 1990. High diversity in DNA
of soil bacteria. Applied and Environmental Microbiology. 56,
782-7 87.

Tunlid, A., and White, D. C. 1988. Validation of quantitative analysis
for microbial biomass, community structure, and metabolic
activity. Adv. Limnol. 31: 1-18.

Vestal, R. and White, D. 1989. Lipid analysis in microbial ecology.
Bioscience. 39, 535-541.

Weete, J. D. 1980. Fungal lipids. In Lipid Biochemistry of Fungi and
Other Organisms. (Weete, J. D. Ed). Plenum Press, New York.

Weete, J. D. 1989. Structure and function of sterols in fungi.
Advances in Lipid Research. 23, 115-167.

White, D. C., Davis, W. M., Nickels, J. S., King, J. D., Bobbie, R. J. 1979.
Determination of sedimentary microbial biomass by extractable
lipid phosphate. Oecologia. 40: 51-62.

16

White, D. C., Bobbie, R. J, Nickels, J. S., Fazio, S. D., Davis, W. M. 1980.
Non-selective chemical methods for the determination of fungal
mass and community structure in estuarine detrital microflora.
Botanica Marina. 23, 239-250.

White, D. C. 1988. Validation of quantitative analysis for microbial
biomass, community structure, and metabolic activity. Arch.

Hydrobiol. Beih. 31, 1-18.

White, D. C. 1994. Is there anything else that you need to understand
about the microbiota that cannot be derived from analysis of
nucleic acids? Microbial Ecology. 28, 163-166.

Wilkinson, S. G. 1988. Gram-negative bacteria. pp. 299-489. In
Microbial Lipids. C. Ratledge and S. G. Wilkinson (eds).
Academic Press, London.

17

Table 1. Examples of several fatty acids found in fungi.

 

 

Molecule Nomenclature *
CH3(CH1),0COOH 12:0
CH3(CH1)12COOH 14:0
CH3—CH2CH(CH2)9COOH a 1 4:0
|
CH,
CH3-CH(CH2),0COOH i14:0
I
CH3
CH3(CH2)3CH=CH(CH2)9COOH 16: 1 a) 5
CH3(CH2)7CH=CH(CH2)5COOH 16: 1 co 9
CH3(CH1)4CH£HCH2CH=CH (CH2)7COOH 18:2 0) 6
CH3CH2CH=CHCH2CH=CHCH2CH=CH(CHZ),COOH 18 :3 co 3
CH3(CH2)7CH=CH(CH2)10COOH 20: 1 C0 9
CH3(CH,_),9COOH 2 1 :0

 

*- Nomenclature is in the form “A:B coC” where “A” indicates the total number

of carbon atoms, “B” the number of unsaturations, and “co ” antecedes the
number of carbon atoms between the nearest unsaturation and the methyl end
of the molecule. The suffixes “c” and “t” indicate cis and trans geometric

6“ 7’ u H

isomers. The prefixes l and a refer to iso and anteiso methyl branching.

Table 2.

18

Marker fatty acids and sterols in the phospholipid fraction of

several groups of organisms.

 

 

Molecule: Orggism:

12:0 Eukaryotic.

30H 14:0 Eubacteria.

15:0 Eubacteria in general, cyanobacteria, actinomycetes.

115:0 Eubacteria in general, gram-positive bacteriaa,
cyanobacteria, actinomycetes.

a15:0 Eubacteria in general, cyanobacteria, actinomycetes.

cy15:l Clostridia.

16:0 Fungi.

116:0 Gram-positive bacteria‘.

10Me16:0 Actinomycetes, gram-positive bacteria“.

16:1 m3t Diatoms.

16:1m5 Cyanobacteria, Glomus (AMF fungus)b.

16:1037 Eubacterial acrobes.

16:1co7t Eubacterial acrobes.

16:1m9 Eubacteria in general, cyanobacteria, actinomycetes.

l6:1col3t Green algae.

16:3 (06 Micro algae.

17:0 Eubacteria in general, cyanobacteria, actinomycetes.

il7:0 Eubacteria in general, cyanobacteria, actinomycetes.

al7:0 Eubacteria in general, cyanobacteria, actinomycetes.

cy17:0 Eubacterial anaerobes, gram-negative bacteriac.

17:1 (06 Sulfate-reducing eubacteria,, actinomycetes.

i17:1co7 Sulfate-reducing eubacteria, actinomycetes.

lOMe 18:0 Actinomycetesd.

18:1 (05 Eubacteria in general, cyanobacteria, actinomycetes.

18:1m7 Eubacterial aerobes, gram-negative bacteriac.

18:1co7t Eubacteria in general, cyanobacteria, actinomycetes.

18:1co9 Fungi, green algae, higher plants, gram-positive bacteriaa.

18:1coll Higher plants.

18:2036 Eukaryotes, cyanobacteria, fungie

18:3 (03 Fungi, green algae, higher plants.

19

Table 2 (cont’d).

 

 

Molecule: Organism:

18:3 (06 Fungi.

cy19:O Eubacterial anaerobes, gram-negative bacteria.
i19:O Eubacteria in general, cyanobacteria, actinomycetes.
al9:0 Eubacteria in general, cyanobacteria, actinomycetes.
20:3 m6 Protozoa.

20:4 (06 Protozoa.

20:5 Barophyllic, psychrophyllic eubacteria.

20:5 (03 Diatoms, higher plants.

20:5 (05 Diatoms.

22:6 Barophyllic, psychrophyllic eubacteria.

26:0 Higher plants.

Polyenoic fatty Eukaryotes.

acids.

Branched Gram-positive bacteria:-

chain fatty

acids.

Monoenoic and Gram-negative bacteria:-
cyclopropane.

30H fatty acids.

mosterol Fuggi'.

From Vestal and White (1989). Other references are indicated by a
superscript number and are listed below.

a- O'Leary, W., M., and Wilkinson, S. G. (1988).

b- Jabaji-Hare, S. (1988).

c- Wilkinson, S. G. (1988).

d- Kroppenstedt, R. M. (1985).

e- Fedrele, T. W. (1986).

f- Newell et al., (1988)

20

.365 w. $338 a». 353m msozgm 9m ammo" em Bknoqsfimm 3 9m a 35980: 2:83 0m 9m so?

 

meanw—
mnmnmmm

908 cm
39.9.-
nEmEm
908 5
888mm

963 cm
Ema-
2:35

903 on
«man...
can Em

EMS"
mnoamm

0:88?
8955

Sam mug

o smmw OE
Evian
Smx

0 Snow 05
Quinn
Ema

 

zn music:

38. 35 OS
wrooQE.
K E
rm an
m.~ No
wk No
~.~ Np

 

.x. £5858
3.28.859

K E
Awb an
mmb mNb
ANA. Nu.»
warm 3:».

 

5% 3.08 308..

K
wub

murm

woL

-.w

mm_Q£m8c:Q n 3530346 ex. £58503

0

23
an

wmh

Hob

H 06

 

8.4.2430?

K E
mob 5Q

NN Nmb
m.~ ab
9m 5%

 

woo"

9.523.

K
PM

Ah

Fm

Em

23

Pm

9m

H.m

Fm.

 

<>Z 33mm.—

489350:

3.0

Po

5h

Ab

 

<>Z 35mg

295%

Po

Hb

Nu

Nm

 

>= <Ecmm m8 mxcwmmmma 9m 9m 62.83 on 85— mxma 0 98m? 9m 3830: 38. Zn thnoQENmr ZZu :0:-B<83.E~mr sun so" canon—:58.
H- >m mxcawnmm + noon 8.4.3390: + 998.2“:an M50830: 8 Evan—EB. arm—68:89 069.5% mmnémAzoB cm mmmnmnEmEmu hmxocmms man
”8395. 389. 9035 Beach? SQ.» whom" any: man ~9an Com; 9035 «39.8526- Sam Rom" $2.1m on a. 30mm?

21

0
ll
CHz-CH3-O-C-R
0
ll
CH-CH3-O-C-R

O

 

H
CH Z-O-P-O-X

O'
Phospholipid

0
ll
CH2-CH3-O-C-R
0

||
CH-CH3-O-C-R

 

0

ll
CHz—CH3-O-C-R

Triacylglyceride

 

 

X= polar head group.

R= aliphatic chains of the fatty acids

Figure 1. General structure of a phospholipid and a

triacylglycerol.

22

.91..

Ergosterol ..

24-methylcholesterol

CH

CH

(Campesterol)

 

Sitosterol

 

Cholesterol

Figure 2. Examples of different sterols.

Chapter 1

THE CHEMICAL TAXONOMY OF THREE LIPID FRACTIONS FROM

ROOT-ASSOCIATED FUNGI OF SORGHUM

Abstract

The whole-cell fatty acid (FAME), phospholipid fatty acid (PLFA),
and sterol profiles of 16 species of root inhabiting fungi were analyzed by
gas chromatography. The object of this study was to evaluate the
usefulness of lipid profiles for distinguishing among the fungal taxa.
Hierarchical cluster analysis was used to illustrate relationships between
the lipid profiles; principal component analysis was used to determine
which lipids defined the grouping patterns.

Twenty different sterol molecules were recovered from the set of
fungi. Ergosterol, lanosterol, and a c28:2 sterol were the most abundant
sterols. The relatively high amounts of ergost-7-enol and low percentage
of ergosterol distinguish the genus Rhizoctonia from the rest of the taxa.
The lack of sterols is characteristic of the genus Pythium.

A total of 25 different phospholipid fatty acids were recovered
from the set of fungal isolates. 18:1 (09, 18:2 036, and 16:0 were the most

common PLFA and were present in all of the fungi. The PLFA profiles of

23

24

the genera Rhizoctonia and Pythium are unique and serve to reliably
identify these taxa. The genus Pythium had the highest diversity of PLFA
and can be distinguished from the rest of the taxa because of the
relatively high amounts of 18:1 039 and 16:0, combined with the
relatively low content of 18:2 (06. In addition, the Basidiomycete
anamorphs Rhizoctonia and Sclerotium are recognizable because of the
high percentage of 18:2 (06, and a relatively low amount of 18:1 m9.

Thirty four different FAME were recovered from all the fungi. In a
similar fashion to the PLFA,18:1 (09, 18:2 (06, and 16:0 were the most
abundant fatty acids in the FAME. In general, the taxonomic
information in the PLFA parallels that of the whole cell fatty acids
(FAME). As with the PLFA, the FAME profiles of Rhizoctonia and Pythium
are useful for the identification these genera. The Basidiomycete genera
Rhizoctonia and Sclerotium can similarly be identified because of their
high content of FAME 18:2 m6, and low amount of FAME 18:1 039.
Pythium is different from the rest of the fungal pathogens in the
relatively high amounts of FAME 14:0 and 16:0.

The Ascomycete anamorphs included in this study showed a
relatively high diversity in their lipid composition and for this reason,
lipid profiling showed little value in identifying the ascomycetous
genera. For example, the genera Fusarium, Bipolaris, and Phoma are not

recognizable by their lipid profiles.

25

This is the first report of the lipid profiles of Bipolaris, Nigrospora,
and Periconia. This research shows the value of lipids for classifying

fungal isolates that cause disease in the roots of Sorghum.

Introduction

Fungal identification has traditionally been based on
morphological and reproductive characters. Some fungal taxa do not
sporulate in laboratory conditions and as a result important
morphological features for their classification are not available.
Chemical profiling of cellular constituents using chromatography is a
widely used tool for determining the chemotaxonomy of microbes
(Lechevalier and Lechevalier, 1988). Fungal mycelium averages 17% by
weight of lipid material, ranging from less than 1% to 55% depending on
the species, development stage, and growth conditions (Murray, 1953;
Weete, 1980). Specific lipids such as sterols, whole-cell fatty acids (FAME)
and ester-linked phospholipid fatty acids (PLFA) have been used to
characterize and identify eukaryotic and prokaryotic microorganisms.
More than 500 fatty acids have been found in plants and microbes
(Harwood and Russell, 1984) and some are unique to specific taxa
(Kroppenstedt, 1985; Fedrele, 1986; O'Leary and Wilkinson, 1988;
Wilkinson, 1988).

For some fungal groups such as the Ascomycetes and

Basidiomycetes, specific lipids lack value as molecular markers because

26

most taxa share the same diversity of lipid molecules (Lechevalier and
Lechevalier, 1988). However, multivariate statistical treatment of fungal
lipid profiles may be used to estimate diagnostic differences between
isolates by small variations in the relative abundance of the molecules
(Blomquist et a1. 1992). Such FAME profiles are obtained readily using a
commercially available protocol (Microbial Identification System (MIS),
Microbial ID, Newark, DE, USA.) The MIS system may be used to
identify isolates by comparison of their fatty acid content to a database.
The FAMEs are produced by the saponification, derivatization, and
extraction of the whole cellular material of the isolates. Because of this,
the fatty acids present in phospholipids, triacylglycerides and
diacylglycerides, and free fatty acids are part of the FAME profile. This
method has proven useful for characterizing fungal isolates at the species
and more inclusive taxonomic levels (Stahl and Klug, 1996).

Recent lipid methods for the chemical taxonomy of fungi focus on
the fatty acids and sterols obtained from chromatographic analysis of
single extracts (Miiller and Hallaksela, 1994; Miiller et al., 1994; Miiller
et al., 1995). The results of these experiments suggest that sterols, in
addition to fatty acids, may offer taxonomic information that can be
used to characterize fungal taxa. The fatty acid content of the
phospholipids of fungi may differ from the fatty acid composition of the
rest of the cell’s lipids. For example, Jabaji-Hare et al., (1984) showed

that in mycorrhizal fungi, the relative amount of 18:2 fatty acid is

27

higher in the phospholipids when compared to other lipid fractions such
as the glycolipids and neutral lipids. It is for this reason that studies are
needed to investigate if the PLFA and the FAME vary in their value as
taxonomic characters

The objectives of this work were: i) To survey the sterol, FAME, and
PLFA content of fungi that infect the roots of Sorghum. ii) Compare the
diagnostic value provided by each lipid fraction at the genus level. iii)
Test if the lipid profiles are stable across different developmental stages

such as mycelia, conidia, and sclerotia.

Materials and Methods

The fungal species included in the experiment are potential
parasites associated with the roots of Sorghum (Farr et al., 1989).
Fourteen species belonging to nine genera were analyzed (Table 1).
Cephalosporium gramineum was supplied by D. Fulbright at the
Department of Botany and Plant Pathology, Michigan State University.
Periconia macrospinosa was isolated from soil at the Kellogg Biological
Station, Michigan, by G. Thorn, Department of Crop and Soil Sciences,
Michigan State University. The rest of the cultures were supplied by G.
Adams, Department of Botany and Plant Pathology, Michigan State
University. All the cultures were maintained on Potato Dextrose Agar

(DIFCO).

28

Broth Cultures

Environmental conditions such as the C:N ratio, pH, temperature,
oxygen. concentration, and time of incubation may affect the lipid
profiles of fungi (Weete, 1980; Miiller et al. 1994). For this reason, the
growth conditions must be carefully controlled to achieve
reproducibility. Iused Malt Extract Broth (DIFCO) as a uniform growth

medium. For each culture, 100 ml were placed in 250 ml glass flasks,

capped with aluminum foil, and autoclaved at 117 0C for 15 min. To
inoculate the media, mycelium (and spores or conidia when available)
were obtained from actively growing PDA cultures and mixed with the
broth. The growth period was 7 days. Slower growing fungi such as C.

gramineum, N. sphaerica, and P. macrospinosa were grown for 30, 10,

and 10 days respectively. The fungi were grown at 25°C under dark,

stationary conditions.

Harvests

Fungal mycelium was separated from the broth by filtration
through a Buchner funnel with filter paper (25 um pore size, Whatman #
114). The mycelium was rinsed with distilled water and blot-dried in
gel-blotting paper (Schleicher & Schuell, Keene, NH, USA.) The biomass
was then placed on a glass slide, finely cut, mixed, and freeze-dried.
Cultures from three different flasks were sampled for FAME and PLFA

analysis. Two different flasks were analyzed for sterols. Whenever

29

possible, specialized structures were analyzed separately from the
mycelium. The sclerotia of S. rolfsii were removed from 15 day old
cultures using a spatula. Conidia of F. moniliforme and Phoma sp. were
obtained from 7 and 10 day old cultures respectively. These formed a
sediment layer at the bottom of the flask. The mycelium and growth
medium were decanted from the flasks, then the conidial sediments were
resuspended with 25 ml of distilled water and centrifuged at 2400 rpm
for 2 min. The supernatants were removed, and the conidia] pellets
saved.

Appendix A is a flowchart that represents the sampling and

analysis scheme for the fungal lipids.

Analysis of Whole-Cell Fatty Acids (FAME)

The FAME content of the samples (0.1 g blot-dry weight) was
determined using the protocol developed for bacterial fatty acid analysis
using the MIDI MIS (MIDI. Newark, Delaware, USA). The samples were
put in l3x100 mm test tubes with Teflon-lined screw caps. ' One ml of

150 g NaOH/lL 50% MeOH in H20 was added to each tube, vortexed

briefly and saponified at 100 0C for 30 min. After the 30 minutes the
tubes were cooled to room temperature in a water bath. Two milliliters

of 54% 6N HCL in MeOH were added at this point. The material was

then vortexed briefly, methylated at 80 0C for 10 min. in a water bath

and immediately cooled to room temperature. The fatty acid methyl

30

esters were extracted by adding 1.25 ml of 1:1 Hexane:Methyl-tert butyl
ether and rotating the tubes for 10 min. The tubes were left to stand for
a clear separation of the phases and the aqueous phase (bottom) was
removed with a Pasteur pipette and discarded. The organic extract
remaining in the tube was washed by adding 3.0 ml of 1.2% NaOH in

H20 to each tube, and turning end over end for five minutes. The
organic phase was removed, transferred to a crimp-top GC vial and

stored at -20°C until analysis. The derivatized extracts were analyzed by
gas-liquid chromatography using a non-polar fused silica capillary
column (30 m x 0.25 mm) and a flame ionization detector. The time of

the run was 38 minutes, which allowed detecting fatty acids up to 28
carbons long (170 0C to 300 0C oven temperature; 250 OC injector
temperature ; 300 0C detector temperature). The flow rate of the carrier

gasses N, H, and air were 30, 30, and 400 ml min.'1 respectively. The GC
setup was developed by Microbial I. D. (Newark, DE). The system was
coupled to a computer programmed to identify fatty acids based on
retention times relative to a known standard mix (that was run every
ten samples) and to quantify them relative to other fatty acids in each

sample according to the peak width and area data.

Analysis of Phospholipid Fatty Acids (PLFA) and Sterols
Total lipids were extracted from 0.02 g of dry mycelium by use of a

Bligh & Dyer (1959) chloroform-methanol extraction modified to

31

incorporate a phosphate buffer. Fungal material was extracted at room
temperature in chloroform/methanol/potassium phosphate buffer
(1:2:0.8 by volume; 50 um , pH 7.4) for at least 2 h. Then, 0.29 volumes
of distilled water and chloroform were added resulting in a separation of

the aqueous and organic phases. The organic phase was dried by rotary

evaporation at 37°C and subsequently fractionated into neutral-, glyco-,
and polar lipid using silicic acid (100 mesh) column chromatography as
detailed by Guckert et al. (1985). This process involves the loading of
the lipid extract to a capillary column, followed by the sequential
elution of the lipids with solvents of increasing polarity (chloroform,
acetone, methanol). The neutral lipids are found in the chloroform,
while the polar lipids are obtained by the elution with methanol.

The phospholipids were obtained from the polar lipid fraction.
The polar lipids were subjected to a mild alkaline methanolic
transesterification (Guckert et al., 1985), producing the phospholipid
fatty acid methyl esters (PLFA). The PLFA were then separated,
quantified, and identified by capillary column gas chromatography (FID
detector, Hewlett-Packard HP-l 50m by 0.2 mm inner diameter non-polar
methyl silicone column) using the conditions described by Ringelberg et
al. (1989). Quantification was based on comparison of peak areas to an
internal injection standard (19:0) assuming equi-molar responses within
the range of chain lengths between 12-24 C. The structures were verified

using a mass selective detector (Hewlett Packard 5791). The extraction,

32

fractionation, derivatization, and analysis of PLFA and sterols was carried
out at the Center for Environmental Biotechnology of the University of
Tennessee, Knoxville.

The sterols were recovered from the neutral lipid fraction after
saponification in 5% KOH in methanol:water (80:20, vzv). The sterols
were derivatized using N, O bis(Trimethylsilyl)trifluoroacetamide (BTSFA)
that formed trimethyl silyl ethers. The trimethyl silyl ethers of the
sterols were then separated, quantified and identified using gas
chromatography (as for the PLFA). Sterol and PLFA identification was
based on co-injection with standards and relative retention time
measurements (Nes and Parish, 1989). All structures were verified using
GC-mass spectrometry at an electron energy of 70 eV with the same
columns and conditions described above.

Three ml of the growth medium were freeze-dried and analyzed
using the same procedure as for the fungal biomass. The amount of
FAME, PLFA, and Sterols in the growth medium was lower than the
detection limit of the instruments. For this reason, the results from this
experiment are assumed to represent the biosynthesis by the isolates.

Fatty acid terminology utilizes “A:B 03C” where “A” indicates the
total number of carbon atoms, “B” the number of unsaturations, and “co”
antecedes the number of carbon atoms between the closest unsaturation

“ 7’
1

and the methyl end of the molecule. The suffixes “c” and indicate

33

‘6",

cis and trans geometric isomers. The prefixes 1 and “a” refer to iso and

anteiso methyl branching.

Statistical Analysis

The lipid profiles were evaluated using Systat version 5.2.1 (SPSS
Inc., Chicago, IL, USA). Principal components analysis (PCA) with the
covariance matrix was used to compare the profiles of the isolates. All
variables are represented on each principal component. The loading
values give insight into which molecules are responsible for the
distribution of the taxa in the PCA diagrams. Isubjected the data to
hierarchical cluster analysis (single linkage method) and constructed
dendrograms for the same sets of data used for the PCA. The cluster
analysis was used to indicate which isolates have the most similar lipid
profiles. Molecules that made up less than 0.25% of the total profile

were near the detection limit and were excluded from the analyses.

Results and Discussion
Sterols

Atotal of 20 sterols were detected and used for the multivariate
analysis. Table 2 shows the number and relative amounts of the sterol
molecules recovered from each isolate, while Table 3 lists the results of
the mass spectral analysis used to identify each molecule. The separate

extraction and chromatography allowed the recovery of a higher

34

diversity of molecules than did previous studies of fungal sterols (Miiller
and Hallaksela, 1994; Miiller et al., 1994; Miiller et al., 1995). The most
common sterols in the fungal root pathogens of Sorghum were ergosterol,
lanosterol and a C28:2 sterol (Table 2). However, these three molecules
were not universally distributed since some fungal taxa lacked one or
more of them.

The absence of sterols may be used to distinguish Pythium from
other filamentous fungi. Other studies have also shown the absence of
sterols in Pythiales (Mc Corkindale, 1969; Weete, 1980). Furthermore, it
has been shown that members of the family Pythiaceae may require
sterol amendments in order for sporulation to occur in pure cultures
(Weete, 1980). The sterol profiles were also useful to differentiate the
genus Rhizoctonia from the rest of the genera included in the
experiment. Component 1, which separates the Rhizoctonia species from
the rest of the extracts (Figure l), is mainly accounted for by ergost-7-
enol and ergosterol (Table 4). Table 2 shows that ergost-7-enol is
enriched in the extracts of the Rhizoctonia species, and may partially
explain the diagnostic value of the sterols (Figs. 1 and 2). In addition, I
found low relative amounts of ergosterol in the extracts of Rhizoctonia
(Table 2). This is relevant because ergosterol is usually a major sterol of
the Basidiomycetes, and has been used to mark the presence of fungi in
field samples (Weete, 1980; Harwood and Russell, 1984; Newell and

Arsuffi, 1988).

35

Phospholipid Fatty Acids (PLFA)

Atotal of 25 PLFA were used for the multivariate analysis (Table
5). The major PLFAs were 18:1 (99, 18:2 (116, and 16:0. Combined, these
three molecules made up more than 90 % of the PLFA of most of the
isolates analyzed (Table 5). In this study, 18:1 (09, 18:2 036, and 16:0
were found in each one of the isolates (Table 5), and others have shown
that these three molecules are common fatty acids of fungi and other
eukaryotes (Weete, 1980; Miiller et al. 1994; Stahl and Klug, 1996).

The Pythium isolates have the highest diversity of PLFA of the
material analyzed. Atotal of 20 fatty acids in proportions higher than
0.25% were produced by P. ultimum alone (Table 5). Other studies have
found that dikaryotic fungi such as Ascomycetes and Basidiomycetes
produce a reduced variety of fatty acids relative to Oomycetes such as
Pythium (Stahl and Klug, 1996). Among the root pathogens of Sorghum,
Pythium is exceptional because of the production of several 20 C
polyunsaturates and 18:3. This set of molecules have also been found in
other members of the Pythiales (Weete, 1974; Stahl and Klug, 1996). The
two species of Pythium formed a distinct cluster, which suggests that this
genus has a diagnostic PLFA profile (Figure 1; Figure 3). Components one
and three provide the resolution to separate the Pythium isolates from
the rest of the extracts. The values for component one are primarily
determined by the relative amounts of 18:2 66, while the distribution of

component three is largely determined by the values of 18:1 m9, and

36

16:0 (Table 4). This suggests that it is the relatively high percent of 18:1
(99 and 16:0 and the low relative concentration of 18:2 (06 that
distinguishes Pythium from the rest of the isolates. Sterols as well as the
amount of unsaturated fatty acids in the phospholipids are two features
that have been associated with the permeability of cellular membranes.

It is remarkable that Pythium lacks sterols and has the highest
diversity of polyunsaturated PLFA. Ihypothesize that this distinctive
composition of cellular membrane components has implications for the
functioning of this taxon. Besides Pythium, Rhizoctonia has a
characteristic PLFA profile at the genus level (Figure 3). The mycelial
and sclerotial extracts of S. rolfsii formed a distinct cluster, suggesting
that there is a species-specific PLFA signature for this isolate (Figure 3).
The principal components diagram and the loading values of the
principal components analysis show that it is the relatively high
amounts of 18:2 (06 and the relatively low amounts of 18:1 (09 that helps
distinguish S. rolfsii and the Rhizoctonia isolates from the rest of the

taxa (Table 4; Figure l).

Whole-Cell Fatty Acids (FAME)

A total of 34 different FAME were identified and included in the
analyses (Table 6). As with the phospholipid fatty acids, the major
FAME were 18:1 (09, 18:2 (06, and 16:0. These three molecules have been

found to be common FAMEs in fungi (Stahl and Klug, 1996), and our

37

data shows that they make up more than 90% of the FAME content of
each of the root pathogens of Sorghum. Our results suggest that the
fatty acid composition of membrane-associated lipids (PLFA) is similar to
that of the total lipid content of the fungal cells (FAME). This is also
evident from the remarkably similar distribution of the fungal taxa in
the principal component diagrams of the PLFA profiles and the FAME
profiles (Figure 1).

The cluster diagram shows that FAMEs are useful for the
identification of the genera Pythium and Rhizoctonia (Fig 4). The FAME
profile of the genus Pythium is recognizable from the rest of the fungi
(Figure 1; Figure 4). The Pythium species have the highest molar
percentage of 16:0 and 14:0 (Table 6). These two molecules are
important in determining the distribution of component three (Table 4),
and distinguish the Pythium isolates from the rest of the genera (Figure
1). Arachidonic acid (20:4) and 18:3 FAMEs are exclusively found in the
Pythium isolates (Table 6), and this supports the results found by others
(Stahl and Klug, 1996). The genus Rhizoctonia has a diagnostic FAME
signature (Figure 4). This is a similar pattern as that obtained with the
sterol and PLFA profiles (Figs. 1 and 2). The mycelial and sclerotial
extracts of S. rolfsii formed a distinct cluster (Figure 4), which suggests
that the FAME fingerprint for this species is recognizable regardless of the
development stage of the organism. As with the PLFA profiles, the FAMEs

of the Basidiomycete anamorphs (Rhizoctonia and Sclerotium) are

38

distinguishable from the rest of the isolates because of the relatively low
amounts of 18:1 (09, combined with the relatively high amounts of 18:2

(06 (Figure 1; Tables 4 and 6).

Specialized Structures

Lanosterol is the precursor of ergosterol in fungi (Weete, 1973). In
the mycelia and sclerotia of Fusarium moniliforme and Phoma sp. the
production of ergosterol predominates, while the conidia of these two
genera are enriched in lanosterol (Table 2). This results suggest less
conversion of precursor (lanosterol) to product (ergosterol) in the
conidia, relative to the mycelium of each species. Other studies have
shown that sterol accumulation may be interrupted once sporulation
begins (Stevens and Jones, 1992; Weete, 1981). Because of the relatively
high amounts of lanosterol, the sterol profiles of the conidial extracts of
Fusarium and Phoma are more similar to each other than to the mycelia
of their respective species (Fig 2). As with the sterols, the relative
proportions of fatty acids vary among mycelia of different ages (Weete,
1980; Stevens and Jones, 1992). In this study, 18:1 fatty acids decreased
while 18:2 m6 from PLFA and FAME were enriched in the conidia and
sclerotia relative to the mycelia (Table 5; Table 6). This resulted in the
lack of similarity of the conidia to the mycelia of the same species

(Figure 3; Figure 4).

39

Lipids and Environmental Samples

Polyenoic fatty acids such as 18:2 036 have been shown to increase
in microbial communities that favor the proliferation of fungi (White et
al., 1980). Our results validate the use of this molecule as a general
molecular marker of fungi. Fatty acid 18:2 (06 was a dominant molecule
in the fungal root colonizers of Sorghum, accounting for up to 75.9 % of
the PLFA, and up to 75.2 % of the FAME (Tables 5 and 6). One potential
problem with the use of this molecule as a marker of fungi in soils, is the
possible background from plant roots, which may contain significant
amounts of this molecule (see chapter 2).

Ergosterol has been used as an indicator of fungi in soil (Grant et
al., 1986) and in plant material (Newell et al., 1988). The concentration
of this molecule varies between species and is absent from Pythium and
Bipolaris (Table 2). Because of this, any conclusions drawn from the
concentration of ergosterol in environmental samples should consider
that it is not a universal marker of fungi.

Campesterol and 16:1 m5 have been proposed as molecular
markers of arbuscular mycorrhizae (Pacovsky, 1989; Schmitz et al.,
1991). Our data shows that these molecules are absent from non-
mycorrhizal fungi that colonize roots and the rhizosphere. This
underscores the unique nature of these lipids and supports their use as

markers of arbuscular mycorrhizae (see chapter 2).

40

Conclusions

Stahl and Klug (1996) observed that in general, the FAME profiles
of filamentous fungi agree with their taxonomic relationships. They
found that in general, Ascomycetes and Basidiomycetes can be
distinguished by their FAME composition. The results of our study
partially agree with those of Stahl and Klug (1996). For instance, the
sterol and FAME profiles of the Basidiomycete anamorphs Rhizoctonia
and S. rolfsii show a close chemical taxonomy when compared to the rest
of the taxa. However, the Ascomycete anamorphs included in this
experiment showed a relatively high diversity in their lipid profiles and
lacked cohesiveness for this reason. For example, the isolates of
Fusarium, Bipolaris, and Phoma have relatively wide differences in their
FAME, PLFA, and sterol profiles and lack distinctive lipid signatures at the
genus level and higher taxonomic level (Figs. 3 and 4).

I conclude that lipid profiles provide sensible taxonomic
information for the classification of fungi at the level of species and
genus. To my knowledge, this study provides the first account of the
separate use of sterol profiles for the production of the chemical
taxonomy of microbes. The taxonomies of the sterol, FAME, and PLFA
profiles show some similar trends. For example, the sterol, FAME and
PLFA profiles all have specific fingerprints for the genera Pythium and
Rhizoctonia (Figs. 3 and 4). Future work should be aimed at enlarging

the available data base of fungal lipids so that broader comparisons can

41

be carried out. In addition, future studies should test if the presence of
fungal parasites in roots can be detected by changes in the lipid

composition of infected plant material.

LIST OF REFERENCES

LIST OF REFERENCES

Bligh, E. G., and Dyer, W. M. 1959. A rapid method of lipid extraction
and purification. Canadian Journal of Biochemistry and
Physiology 35, 911-917.

Blomquist, G., Andersson, B., Andersson, K., and Brondz, I. 1992.
Analysis of fatty acids. A new method for the characterization
of moulds. J. Microbiol. Meth. 16, 59-68.

Federle, T. W. 1986. Microbial distribution in soil. pp. 493-498. In
Perspectives in Microbial Ecology. F. Megusar and Gantar, M (eds).
Slovene Society for Microbiology, Ljubljana.

Grant, W. D., and West, A. W. 1986. Measurement of ergosterol,
diaminopimelic acid and glucosamine in soil: evaluation as
indicators of microbial biomass. J. Microbiol. Meth. 6, 47-53.

Guckert, J. B., Aintworth, C. P., Nichols, P. D., and White, D. C. 1985.
Phospholipid ester-linked fatty acids as reproducible assays for
changes in prokaryotic community structure of estuarine
sediments. FEMS Microbiology Ecology. 31, 147-158.

Harwood, J. L., and Russell, N. J. 1984. Major lipid types in plants and
microorganisms. pp. 7-32. In Lipids in Plants and Microbes.
Harwood, J. L. and Russell, N. J. (eds). Allen & Unwin,

Winchester Massachusetts.

Jabaji-Hare, S., Deschene, A., and Kendrick, B. 1984. Lipid content
and composition of vesicles of a vesicular-arbuscular
mycorrhizal fungus. Mycologia. 76(6), 1024-1030.

42

43

Kroppenstedt, R. M. 1985. Fatty acids and menaquinone analysis of
actinomycetes and related organisms. pp. 173-199. In
Chemical Methods in Bacterial Systematics. M. Goodfellow and
D. E. Minnikin (eds). Academic Press, London.

Lechevalier, H. and Lechevalier, M. P. 1988. Chemotaxonomic use of
lipids-an overview. pp. 869-902. In Microbial Lipids. C.
Ratledge, & S. G. Wilkinson (eds). Academic Press London and
New York.

McCorkindale, N. J., Hutchinson, S. A., Pursey, B. A., Scott, W. T.,
Wheeler, R. 1969. A comparison of the types of sterol found in
species of the Saprolegniales and Leptomitales with those found in
other phycomycetes. Phytochemistry. 8, 861-867.

Miiller, M. M., and Hallaksela, A. M. 1995. Variation in combined
fatty acid and sterol profiles of Ascoryne,Nectria, and

Neoglobularia-strains isolated from Norway spruce. Eur. J. For.
Path. 24, 11-19.

Miiller, M. M., Kantola, R., and Kitunen, V. 1994. Combining sterol
and fatty acid profiles for the characterization of fungi. Mycol
Res. 98, 593-603.

Miiller, M. M., Kantola, R., and Korhonen, K, and Uotila, J. 1994.
Combined fatty acids and sterol profiles of Heterobasidion
annosum intersterility groups S, P, and F. Mycol. Res. 99, 1025-
1033.

Murray, 8., Woodbine, M., and Walker, T. M. 1953. Microbiological
Synthesis of Fat. Journal of Experimental Botany. 4: 251-256.

Nes, W. D., and Parish, E. J. 1989. Analysis of sterols and other
biologically significant steroids. Academic Press, Inc. San Diego,
California.

Newell, S. Y., Arsuffi, T. L., and Fallon, R. D. 1988. Fundamental
procedures for determining ergosterol content in decaying plant

44

material by liquid chromatography. Appl. Env. Microbiol. 54,
1876-1879.

O'Leary, W., M., and Wilkinson, S. G. 1988. Gram-positive bacteria.
pp. 117-201. In Microbial Lipids. C. Ratledge and S. G.
Wilkinson (eds). Academic Press, London.

Pacovsky, R. S. 1989. Influence of inoculation with Azospirillum
brasiliense and Glomus fasciculatum on Sorghum nutrition. pp.
235-239. In Nitrogen Fixation with Non-legumes. F. A. Skinner
et al., (eds). Kluwer Academic Publishers, London.

Ringelberg, D. B., Davis, J. D., Smith, G. A., Pfiffner, S. M., Nichols, P. D.,
Nickels, J. S., Henson, J. M., Wilson, J. T., Yates, M., Kampbell, D.
H., Read, H. W., Stocksdale, T. T. and White, D. C. 1989.
Validation of signature polar lipid fatty acid biomarkers for

alkane-utilizing bacteria in soils and subsurface aquifer
materials. FEMS Microbiology Ecology. 62, 39-50.

Schmitz, O., Danneberg, B., Hundeshagen, A., Klingner, A., and Bothe,
H. 1991. Quantification of vesicular-arbuscular mycorrhiza by
biochemical parameters. J. Plant Physiol. 139, 106-114.

Stahl, P. D., and Klug, M. J. 1996. Characterization and differentiation
of filamentous fungi based on fatty acid composition. Applied
and Environmental Microbiology. 62, 4136-4146.

Stevens, J., and Jones, R. 1992. Determination of whole-cell fatty acids
in isolates of Rhizoctonia solani. Phytopathology. 82, 68-72.

Weete, J. D. 1973. Sterols of the fungi: distribution and biosynthesis.
Phytochemistry. 12, 1843-1864.

Weete, J. D. 1974. In Fungal lipid biochemistry. Distribution and
metabolism. Plenum Press, New York.

45

Weete, J. D. 1980. Fungal lipids. pp. 9-43; 224-260. In Lipid
Biochemistry of Fungi and Other Organisms. Weete, J. D. (ed).
Plenum Press, New York

Weete, J. D. 1981. Lipids in fungal growth and reproduction. pp.
463-485. In The Fungal Spore: Morphogenetic controls.
Turian, G. and Hohl, H. R., (eds) Acad. Press, New York,.

White, D. C., Bobbie, R. J, Nickels, J. S., Fazio, S. D., Davis, W. M. 1980.
Non-selective chemical methods for the determination of fungal
mass and community structure in estuarine detrital microflora.
Botanica Marina. 23, 239-250.

Wilkinson, S. G. 1988. Gram-negative bacteria. pp. 299-489. In
Microbial Lipids. Eds. C. Ratledge and S. G. Wilkinson. Academic
Press, London.

46

Table 1. Root—pathogenic fungi analyzed for FAME, PLFA, and sterols.

 

Bipolaris sorokiniana (Sacc.) Shoem.
Bipolaris victoriae (Meehan & Murphy) Shoem.
Cephalosporium gramineum Nisikado & Ikata
Fusarium equiseti (Corda) Sacc.

Fusarium graminearum Schwabe

Fusarium moniliforme Sheld.

Nigrospora sphaerica Lefebvre & Johnson
Periconia macrospinosa Lefebvre & Johnson
Phoma exocarpina Peck

Phoma sp.

Pythium graminicola Subramanian

Pythium ultimum Trow

Rhizoctonia oryzae Ryker & Gooch
Rhizoctonia solani Kiihn

Rhizoctonia zeae Voorhees

Sclerotium rolfsi S acc.

 

47

Haw—n N. 33: @083" cm 8:: mane—m 80398 $25 one: em :5 32 cwaromoio 35mm.

 

 

9037 056 mg”? 0333» ONmHuw 033 name. 05$» 90:»- nmmnmz
£23 .fiano— flame. nmmnhn

53°32.“ o fa o No 5o Sb o fix o

Exotica“.

5.333; o o o o o o o o o

SnSlnm

Omnriouhelzs Eb Nb o NPA mb 3b o o c

«35.52::

$5335 PM PM Nb Nu oh mwb o o O

«22.32.

3.32.5: o o o o o 3.0 o o o

wait—«.35:

3.31:3 o ob 5.0 We EH 3b o o o

Soazmxelzm

$362.23 C o o o o o o o o

5032233:

83%».

ZmeoauoS Pu m; o o o .30 o o o

“vagina

 

48

.520 N 303.8.

 

 

manor 036 Gum”? Sumac» nwmuum 03$ Owwua name- ONmHNm 92:7 03qu
«83— «82: 303— ONmneo

$32636 o Pa O Pm a; o amb o o o

Senweuhmaaa

383a o o o o o o ~90 o o o

33.3328

38:8 uh oh oh o 3b ~._ fo 3. m o o N. A

$593 E o o o Pm o 0 mm o o o

835:»

zineaoio o o o o o o obs o mo o

o~v~am

zinenaoie o o o o o o ob o; Pm o

ae~a=m

5:86.833 o o o o o 0 ob o 56 o

Nana

mlmwozzs o fig 0 o fo o 2L flu 3.0 o

352;.

M23335 o uh NQ o H5 o 3% o 3b o

35;... 333:»

 

49

Hugo N 839.8.

 

 

manner Omaha um; nwmnwo NmUN n nwmnwa 2mg? 08% :30- ONcHN Our» nuonw
2-2:: «83.

53032.9. MNQ o c o Ape o o Pm u; —.e :9

33.22.3333

93235 c o o o 30 o o o o o 0

$632.23

032.33.33.33 Nu o o 0 Ab o o c EA o o

N33323:

3.32.5: HQ o o o 3.; 0.x 0 Ab 9a 3..» fm

35.32.

3232.5: o o o o Pu o o 3.0 c 3.0 0

@3323335

3232.53 —.m o o o _o.u o 0 Wm uh ~92 o

333236253

3.32:5 o o uh o o o 0 mad 0 mb o

Scat—.333

333».

233.333 o 9o o 9 New o _. a u. 3 o ab o

3:32.63

 

50

Hug—a N 903.5.

 

 

manor ONmHNc Nm; meuwo gnu n Ommuwa «£37 .30- finch Owru Ououw
2-30. mano—

.§2.3:..: a; o o o Sb 0 o o u... o

3333333

$333 0 o o o Eh o Eu o umh o

33:32::

38:8 3 ah o _.u o APm o o... Eu .5 9m

3.9:: .3 fu o o o Ho... o mhb o 3.0 _.u

00:33

5:23:33: PM ob c on. S: 3... oh oh Nd o

cva:m

523205.: o o o o on: 3.”. on o; No o

u:.::.

52.3233: o; o._ o ob PM «oh ob om ob o

mmam

3332...: o o o o .9: 3.... 3b 0 ob C

256.2...

@23335 Pm ob o ob o Nmb fl. u o o o

33.... 353:»

aum. .26 $222.5: aorzam Enron was; :3 :8 so: 3338.

51

Table 3. Mass spectral characteristics of sterols recovered from root
pathogenic fungi.

 

 

Name* RRt1' Systematic name M+1' b.p.T other ions
C2726 0.47 448 251 433, 376, 361, 350. 333
C2824a 1.12 466 251 376, 361, 325, 69
brassicasterol 1.14 (22E)-Ergosta- 470 69 380, 365, 341, 255, 129
5,7—dienol
C2823a 1.16 468 363 453, 378, 337, 253
C2724 1.20 452 362 437, 325, 237, 195, 69
ergosterol 1.26 (22E)-Ergosta- 468 69 453, 378, 363, 337, 253
5,7,22-trienol
C2822a 1.30 470 73 455, 441, 365, 343, 227
stellasterol 1.32 470 69 455, 343, 255, 229
C2823b/C2824c 1.33 468 73 363, 337
I466
C2822b 1.37 470 69 455, 365, 343, 227, 213
C2821 1.38 472 472 457, 367, 341, 255, 229
C2823c 1.43 468 363 337, 253, 73
C2822c 1.46 470 73 455, 380, 365, 355, 343
C28:2d 1.50 470 75 455, 386, 372, 365, 343
fungisterol 1.53 ergost-7-enol 472 75 457, 367, 345, 255, 229
C29:3 1.53 472 69 446, 392, 377, 351, 253
lanosterol 1.61 Lanosta-8,24- 498 69 483, 393
dienol
C2922 1.70 484 69 469, 394, 379, 227
C31:2 1.80 512 69 497, 407, 241
C3022 1.92 498 73 483, 408, 393, 365, 355

 

*- The sterols are represented in the form szy, where x: number of carbons
in the molecule and y= the number of double bonds.

’r- RRt= relative retention time; M+= molecular weight of the protonated
compound; b.p.= most abundant fragment on the mass spectrum.

52

Table 4. Molecules giving the highest loadings for the components.

 

 

 

 

 

 

 

 

 

 

FAME
Moment—1* W’“ W’“
Mic 1 in my: 1 in my: Lea—ding
1822(06 -20.4 1821039 -7.0 16:0 3.5
18:1m9 8.9 1420 3.3 1420 2.5
1620 4.1 1620 2.3 1820 -1.6
PIFA
m 11 nt 1* m 11 nt 2* Component 3*
W logging molooolo 1 in molooplo loaoing
18:2co6 -16.5 1823003 -3.3 1821(09 —3.0
1821(09 6.4 1820 -3.2 1620 2.7
1620 2.8 1821m9 2.0 1820 -1.2
Sterol
m n n 1* Component 2* Component 3*
ergost-7-enol -29.6 C2822d 20.7 ergosterol 18.3
ergosterol 12.7 lanosterol -15.3 lanosterol -l4.5
C28:2d 7.0 C31:2 -4.4 C2822d -8.9

 

 

 

*- The component axes from figure 1 can be used to determine which

component (1, 2 or 3) is valuable for identifying each fungal genera. In this
table, the higher absolute values for the loadings indicate which molecules
have a greater influence on the distribution of the respective components in
figure 1. The percent of the total variance accounted by each component is:
FAME- component 1: 83.6%; component 2: 11.4%; component 3: 3.7%. PLFA-
component 1: 84.7%; component 2: 7.2%; component 3: 4.8%. sterol-
component 1: 40.9%; component 2: 24.2%; component 3: 22.4%.

.520 m. Eofi 3:38: 21?» 3 90 95mm: 82 3850803 o». @3832?

 

 

53

5:3. :3“ :Ho 38 38 3; ea 3; so 38 38 :5 ea 3; so 583a 888 so
9.8332,“. c ch 38 ob o n 88 oh 8.8 a; 8
33:33:88 8.3 8.3 8.3 8.3 8.3 8.3 8.3
5.33:2.“ o Pu 888 ch o H 8.8 8.8 88.8 mare 8
5682.8 8.3 8.3 8.8 8.3 8.8V 8.8V 8.3
0885836915: 8 oh 898 H o oh. :9 88 3.0 3.8 8
8223.322: 8.3 8.3 8.3 8.: 8.8V 8.8V 8.5
38.3433 . a 3.8 8.0 o 8 86 oh Eu 3.8 o
35.82. 8.3 8.3 8.5 8.3 8.3 8.3
83.32.23 o o 36 Pm o a 3.8 o 88.8 in o
83383335 8.3 8.88 8.3 8.3 8.3
3.32.23 o o 88.8 ob o . Ab H 3.8 mmb 8
532.282: 8.3 8.3 8.3 8.: 8.3
38.32.25 0.8 8 8828 8.3 o 8 mb 8.8 888 8.8.8 o
53.288248 8.3 8.3 8.3 8.3 8.3 8.3 8..:
833»

mewoauowa o o 8.0 86 o o 8.8 8.9 won 808 o
“humming 8.8V 8.3 8.8V 8.8V 8.3 8.3
wmlneamu o o 888 a o p 8 g o .8 35 8.8 o

3:233:23 8.3 8.3 8.: 8.3 8.3

 

54

8.888880 m 8288.3.

 

 

5883. 83888" 8888 38 898 898 8.8 898 80 8.88 886 8m; 88 8888 80 88888.; 8888 so
8688588 8%. o o 88.8 8.8 o 8 8. o 8.0 88.8 88.8 8
8.3 8.3 8. 83 8.83 8.3 8.3
$8558 .888. 8 8 890 8. .8 o o. u u. a 8. 8 88.88 3.8 8
888888888 8.88 8. 83 8. 3 8. 3 8. 3 8.3 8.3
28:28 8 8 886 o. .8 o 8 88.8 8. 8 88.8 888.8 C
382882.828 8.83 8. 3 8.3 8. 3 8.3 8.3
83.88885: 88 o 88.8 8 o o 8 o. m 8. 8 8.8.88 88; o. 8
83383.33 8.3 8.83 8. 3 8. 3 8. 83 8.3 8.3 8. 3
898.882.8288 8 8 8 88a 8. m o. u 8 8 m 8. m 88.8 88.8 o. m
88888.35: 8. 3 8.3 8. 3 8. 3 8. 3 8. 3 88.3 8.3 8. 83
”888.38.88.88 8 8 80.x o. u o o o. .8 o 8.8 gob o
38:88 8.3 8. 3 8. 83 8.83 8.3
8288.88.88.88 o o 8.8.8 8 o o 8 o o 88.8 88.8 o
8685:. 8.3 8. 3 8.3 8.3
822.868.8888.: 8 o 88 o o o 8. m o 88.8 3.8 8
~28 8.3 8. 83 8.3 8.3
9.2368823 o c. .8 8.8.8 8 o o. u 8 c 8 8.0 .890 8
38.3.8... 8. 83 8.3 8. 3 8. 3 8.3 8.3
88.830885: 8 8.8 88.8 8 o o. c 8. m 8. m 8.8 888 o
3885.8. 8. 83 8.3 8. 3 8. 3 8. 3 8.3 8.3

38038888

55

8.888880 m 88888.3.

 

 

8383. 3888" 8888 88 886 898 aw 898 898 898 8988.3 8988.3 89m 98 800... 8880
988 8808

88888888828. 8.8 o o o c o o o o o 0

3388833888 8.3

88.882828. 8h 8 8 o 8 8 o o o o o

8886.82.88 8.3

62888888088838.8883 8.88 8 8 o o 8 8 o o o 8

8338.88.22: 8.83

88.88.32.883 o 8 o 8 o 8 8 o o o 8

2882.828.

88.88.32.883 8.8 8 o 8 o o o o o o 8

82858882822: 8.83

88.88.32.883 0 o o o o o o o o o o

532.88%353

38.38.8885 888 o o o o o o o o o

30:888xow3m 8.83

8588888888

28.83.8338 8.8 o o o o o o o o o o

8888:3128 8.3

88.38.838.88 8.8 8 o 8 8 8 o o o o 8

588383883: 8.83

 

56

8.888880 m 88888.3.

 

 

8.6888. 88888888 8888 88 888 898 so 8on8 8on8 898 898 8a 898 ea 88% 88 800... 888
888 88088

888388 .888. o o o 8 8 8 o o o o o

2838 .888. 8.8 8 8 8 o 8 o o o o e

888888888 8.3

8.883: 8.8 ob o 8 8 8 o o o o 8.8

883.68.888.38 8.3 8.83 8.3

8888888882: 8 o 8 8 8 8 8.8 9m 8.8 o o

8338:8863 8.83 8.3 8.83

888.888.8288 8.8 98 88 oh 8.o oh oh 8.8 8.8 o8 oh

888885838 8.3 8.83 8.3 8.83 8.83 8.83 8.83 8.83 8.3 8.83 8.3

8388.88.83.88 o 0.8 8.8 8 8 8 8 ob o o C

883.28 8.83 8.3 8.3

8888.88.83.88 o o o o o o o o o o 0

“©8838

8288.38.83.88 o o o o o o o o o o o

Nana

822088.82: o o o o o o o o o o 8

8688.88...

.8. 8888:... 288088888 o o o o o o o o o o o

 

...- 858868580 885888880. 8.85 88888828 8884 8880 880888880 628% 5 88888885928. 588888 888 888 8.080888 08 88888 388888888828
889888888288. 8" 838 AAo.8m $3. gnu. 08888. 8.85 885809888 8808888888888 mon 808 8888888 c.8u 8° 888 £55..

57

8.888880 9.

7888 8888858888 8n>8<8m 888 8880 388mm: 82 8888885808; 08. 8.338838.

 

 

8388a. 838888 8880 88880 8880 888880 8888 890 88m; 890 8890 898 898 898 8880 8888
am 880 8m 88 80 am

3.838881... . 0 0.8 0.8 0 0 8 06 898 8.8 0 0.0 8 0.8 8

88388888888888 8.3 8.3 8.83 8.3 8.3 8.3 8.3

3.888882.“ 0 0.8 0.88 0.8 8 8 8 88.8 8.8 0.88 0.0 0 8 8

888.8881: 8.3 8.83 8.3 8.3 8.3 8.83 8.3

68888988838883 0 0 0.8 0 0 0.8 0 88.8 0 0 0.8 0 0.88 0.8

88.838.88.283 8.3 8.3 8.3 8.3 8.3 8.3

88.88.32.588 0 0 0.8 0 0 0.8 0 88.8 0 0 8.8 0 8 0

«8:88.38. 8.3 8.3 8.3 8.3

38.32.23 0 0 0.8 0 0 0 0 88.8 0 0 8.8 0 0 0

88.883888888828888 8.3 8.3 8.83

88.28.882.883 0 0 0.0 0 0 8 0 8PM 0 0 0.8 0 0 0

5388.88.83: 8.3 8.3 8.3

8.188.888.8883 0 0 8.88 0 0 0 0 88.8 0 0 0.8 0 0 0

38838888633 8.3 88.88 8.3

888888888

28.83.8883 0 8 8.8 0 0 8 0 88.0 0 0 8b 0 0 8

8888882886 8.3 8.3 8.3

883.88.638.88 0 0 8 0 0 0 0 80.8 0 0 0.8 0 0 0

35.38.888.888: 8.3 8.83

8885388 0 0.8 8.8 0 0 8 0 88.8 0 0 0.8 0 0 0

«83.889885 8.3 8.88 8.88 8.3

 

58

8.888880 8 8288.3.

 

 

8.38888 .8888“ 8880 8880 8888... 8888 888880 8888 8880 8080 8088 80888 8088 80% 8888
80: 80 98> 8a 8.8. 080 80 80 new 88

5888888818. 8.8 0 0 80.8 0 80.0 0 0 0 0 0 0 0

86388888853 8.3 8.3 8.3

3.0888882.“ 90 0 0 88.8 0 88.8 0 8 0 0 0 0 0

Englam 8.03 8.83 8.3

68808888888888.8883 8.8 0 0 88.8 0 88.0 0 0 8.0 0 0 0 0

88.838.882.888 8.83 8.3 8.3 8.03

88.88.32.885 m8 0 0 88.8 8.8 89m 0 8 0 0 0 0 0

«8888.388. 8.3 8.3 8.83 8.3

88.88.862.883 8.8 0 0 no.0 0 80.88 0 0.8 0 0 0 0 0

88.838.38.522: 8.83 88.3 8.03 8.83

88.82.882.883 8.8 0 0 80.0 0 88.8 0 0 0 0 0 0 0

3388888638888 8.03 8.83 8.3

88.88.32.885 m8 0 0 80.8 0 88.8 0 0 8.8 0 0 0 0

53:56:88 8.3 8.83 88.3 8.3

88888888.

28.83.863.88 98 0 0.80 88.8 0 88.8 0 0.8 0 0 0 0 0

€833.83 8.3 8.3 8.3 8.03 8.83

8828.85.88 8.0 0 0 00.8 0 890 0 0 0 0 0 0 0

58238888888 8.3 8.03 8.3

08858 88.8 0 0 88.0 0 88.0 0 0 0 0 0 0 0

833388.388 8.3 8.3 8.3

 

59

8.888880 0 88888.83.

 

 

88888888 888888 88 88880 8880 888880 8888 8080 88088 38 88080 8088 8088 8088 8880 8888
80 880 am 88 80 em

0883: .888. 0 8 8 0 0 8 0 80.8 0 0 8.8 0 8 8

8.3 8.3

2538 .80. 0 0 8 0 0 8 0 80.8 0 0 8.8 0 0.8 8

888888888 88.83 8.3 8.3

088888.838 0.8 0 80.8 0 0.8 0 0 80.0 0 0.0 8.0 0.8 0 8

8338:8888 8.3 8.3 8.3 8.3 8.3 8.3 8.3

08.888882: 0.0 0 80.0 0.0 0.0 8 0 88.8 0 0 8.8 0.0 0 8

88888.55: 8.3 8.03 8.3 8.83 8.83 8.03 8.3

8288.38.83.88 8 8 0.0 8 0 0.8 0 80.0 0 0 0. 0 8 0 0

382$ 8.3 8.3 8.3 8083

88883388388 8 0 8 8 0 8 0 88.0 0 0 8 8 0 0

.3883. 8.3

8288.38.83.88 8 0 8 0.0 0 8 0 88.0 0 0 8 0.8 0 8

~28 8.3 8.3 8.3

382088.885 0 0 0 0 0 0.0 0 88.0 0 0 0 0 8 8

288:... 8.3 8.3

0.88888388885 0 0 0 0 0 8.8 0 88.0 0 0 0 0 0 0

3858.8... 8.3 8.3

380388»

60

0.88080 0 88888.00.

 

 

0.8883. 928—" 000 0080 00;... 00; 008080 0080 0080 ~00 00"— 00:8 ~03. 0080 MN;
09.— 80 g 80 80 new 80 80 080 80

28388 at. 0.0 0 0 00.0 0 AA.— 0 0 0 0 0 0 0
80.3 80.00 3.3

38:8 .800. P0 0 0 00.0 0 00.0 0 0 0 0 0 0 0

0038:» 80.00 80.00 80.5

008088.23 0.0 0.0 0.0 00.0 0 0.0 8 8 8 8.0 0.0 0.0 8

0223.363: 80.00 80.3 80.00 80.0 80.00 80.00 80.00 80.80

008088.82: 0.0 0.0 0 00.0 0 3.0 0.0 0.0 :8 0 0.0 0.0 0

8:88.35: 80.00 80.00 2.00 80.00 80.00 80.00 80.00 80.00 2.00

022.38.403.88 8.0 0 0 5.0 0 00.0 0 8 8 8 0 0 8

303m 80.3 88.: 2.00

00832833 0.0 0 0 00.0 0 00.0 0 8 8 0 0 0 8

.3338. 80.3 80.00 3.3

”02.32383 0.0 0 0 00.8 8 00.0 0 0 8 0 0 0 0

Nana 80.3 3.80 2.3

023088.838 0.0 0 0 0.0 0 00.0 0 0 0 0 0 0 0.0

3008.2... 80.00 80.00 80.: 80.3

022088.53 0.08 8.0 0 08 0 00.0 0 0 0 0 0 0 0

no? 020388» 3.80 8.00 8.00 80.00

 

0.00 $8.883 86 883 8350:88— 888308 8888803 0888800 090250.
8:88 20% 88588 0.00 .000.

88030888280. 8"

88n0.

*

588:3 883 088 0080888 00 88888 088880~ 88888 3888880883

8880 090880088 om Ea 88858888838888 on 85m 885888880 5 858 Homo—<00.

 

 

 

 

 

61

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 1. Principal components anlyses of the lipid profiles.

 

 

 

5
4O sterol 9 2:; 3
6
(3 Q 9 1| F
10|PLFA 77 6
8 7 -6O
. 1
3 80\ 3
(3 9 8 1 3 \/30
6 (:2
-6O -70 C1
~10 3
10 |
40 7
(2 FAME 7
-15 -30 C1 10
8 3
C3
9 9 {13 6 3
2
E»
’10 m ||
30
/45
C2
_20 _5 5 C1
C1=component1
C2=component2
C3= component 3
1) Bipolaris 6) Phoma
2) Cephalosporium 7) Pythium
3) Fusarium 8) Rhizoctonia
4) ngmspora 9) Sclem ti um
S) Pericom'a

62

magnum»: 858888
0.0 IIIIIIIIIIIIIIIII

EBEQm 5.201%

 

No.0

 

§Won8o=0m 035mm I7

 

03800905.» 838: I—
gfiofioam N80. l, _

023088.858 358.8.

 

 

 

 

 

$058.85. 033858.800 ._

 

3058.25 885.5858 ._

 

08.38.088.885 8.003... 099.085

 

 

358m @8835»

 

 

 

mafiaca Boa 8.0.3380.

 

 

0.585.858 035588.885
28.025008.» gunman.»

3883.5: 38:.me

 

Flooim anamanmm

 

 

Ombafiombolca mfiaamca

 

 

353» mu.

 

 

 

08600QO 836.2.»3

 

 

ESE» mp. 83me L.

 

mummlca 508880000843 88:9»

30:8 N. 18.8.28 883808.888 00 9m «8868 030808 0.08 9m 8508:8885. 88808:». 08. £58.00».

63

0. 0 IIIIII m cmammblmmhmsnh IIIIII 0.0

$050.53 08338.80» _

$088.5: 885.5885 _
meaombowm @0318 TI
0:82.25 50:08.00an 88889» .
2. 05» 98898.3» _ _

3883.5: 083588.885

 

 

 

 

 

 

 

 

3.05» wt. 83%»

 

0:82.858 38.880.

005.8888.» anwombSOmm
mcgES Banmowam U

nmhbfiombo: E: 588850.885 IIL

 

 

 

 

 

EBEQm 5.22.8.3

 

085058.? 88.008533

 

 

xbwonsam Nfim

000N883» oQNmm J
xbfionaoafi 80%: IL

2805» up.

 

 

 

 

02262.5: 358.8. .8532»

02902.5: 358.8.

30888 0. Ham 888038888 em 888m 0880008808805 @8800 mag 08.0008 08.0888 8.88m 8508:8885. 839m. 08. wagon»

64

man—Emma Ema—888
0.0 IIIIIIIIIIIIIIIII 0.0

3058.883 0383.380»

 

 

 

3058.85. 52.85:

 

 

 

0:82.885 358.880.888.885
0:82.883 88803006286

0:83.885 35.88.. .

3.05m Mb. Ul—
wmadoim anwofiaofi

0:62.888 508.08.849.36 88858»

 

 

 

 

 

 

0.660QO 885.385»

mmnafiomboacs 0358.88.88:

 

3.802% Snaoamm _

 

28.08.8083 «3.88.8.8

3.03m @8885» _

283» up. 839»

 

 

 

 

 

 

muonoEm 80888:.

098N885.» Nmmm II..—

 

02329:“. oQNmm IL

022.688.8883 358.8. _ _
022088.883 358.8. 85388»

 

 

 

205m “8. Hana 9880me Om 85m 2:05.88— mm88v0 mna 08.038 03888 88.5 8508:8885. 83%». 08. .8530»

Chapter 2

IDENTIFICATION OF POSSIBLE MOLECULAR MARKERS IN
THE LIPIDS OF ARBUSCULAR MYCORRHIZAE

Abstract

We analyzed the total fatty acid methyl esters (FAME),
phospholipid fatty acids (PLFA) and sterols of the mycorrhizal and non-
mycorrhizal roots of Sorghum by gas chromatography. The purpose of
this work was to identify lipid markers of mycorrhizal infection. To
achieve this, the roots of Sorghum infected with several species of
glomalean fungi were compared to non-mycorrhizal roots. Fungi
belonging to the genera Glomus, Gigaspora, Acaulospora, and
Scutellospora were analyzed for their lipid content.

Increases in hexadecenoic acid (16:1) in the root phospholipids
indicates the presence of arbuscular mycorrhizae. Mycorrhizal infection
is associated with increased amounts of either PLFA16:1 035 or PLFA16:1
(07 in the roots of Sorghum, depending on which fungal symbiont is
present. The 16:1 (07 isomer was highest in the G. occultum symbiosis,
accounting to more than twice the molar percentage of this fatty acid in

the control roots. Phospholipid fatty acid 16:1 (05 increased in the rest

65

66

of the fungal treatments, and it was present in at least 50 % higher
molar percentage when compared to non-mycorrhizal roots.

The mycorrhizal roots of all the fungal treatments had at least
21.1% more total FAME concentration than uninfected roots. The
mycorrhizae of Glomus etunicatum, Glomus intraradices, Acaulospora
mellea, and Scutellospora erytropa have relative amounts of 16:1 (05
FAME which range from 4.9% to 13.3 % of the total fatty acids. In
contrast, the concentration of 16:1 m5 in non-mycorrhizal roots, as well
as in the mycorrhizae of Glomus occultum and Gigaspora rosea was
beyond the detection limit of the instruments. Hierarchical cluster
analysis shows that the FAME profiles are valuable in the diagnosis of
non-mycorrhizal roots and provide insight as to which fungus is present
in the mycorrhizal association.

Besides the FAME and PLFA, the sterols are valuable for marking the
presence of arbuscular fungi in roots. Infected roots can be distinguished
from non-mycorrhizal roots because the molar percentage of campesterol
is at least 23% higher when the roots are colonized.

This is the first report of the sterol content of the mycorrhizae of
Acaulospora and Scutellospora, and the first account of the
phospholipid fatty acids of roots infected by Acaulospora, Gigaspora,

and Scutellospora.

67

Introduction

The dominant cytoplasmic constituent of the glomalean fungus is
lipid (Cox and Sanders, 1974), comprising more than half of the dry
weight of the fungus in some cases (Beilby and Kidby, 1980). The
difference between mycorrhizal and non-mycorrhizal lipids is not only
quantitative, but it is also qualitative. For example, the roots of several
plant species infected with Glomus contain various unsaturated fatty
acids which are not detectable in non-mycorrhizal roots (Graham et al.,
1995; Nagy et al., 1980; Nordby et al., 1981; Pacovsky, 1987; Pacovsky
and Fuller, 1988; Pacovsky, 1989). The high lipid content combined with
presence of unusual molecules provides the opportunity for the use of
lipids as molecular markers of arbuscular mycorrhizal fungi. For
instance, fatty acid 16:1 «)5 has been used to estimate the presence of
Glomus vesicles in the roots of monocotyledons and dicotyledons
(Graham et al., 1995; Pacovsky, 1987; Pacovsky and Fuller, 1988;
Pacovsky, 1989).

Recent studies of the fatty acids of glomalean mycorrhizae have
concentrated on the analysis of the whole-cell fatty acids (FAME)
(Graham et al., 1995). The fatty acids that are bound to other cell
components are separated by methylation and analyzed by gas
chromatography to produce the FAME profile. Because of this, the FAME
analysis includes the fatty acids bound to neutral lipids (such as

triacylglycerides and diacylglycerides), those bound to polar lipids (such

68

as phospholipids), as well as free fatty acids. The neutral and polar
lipid fractions of glomalean fungi may differ in the abundance of 16:1
m5, indicating the importance of separating the lipids prior to the fatty
acid analysis (Jabaji-Hare, 1984; Olsson et al., 1995). However,
information on the fatty acid composition of specific lipid fractions such
as the phospholipids is lacking for several glomalean genera. This is
relevant because phospholipids fatty acids have been shown to be good
measures of the extra-radical phase of Glomus (Olsson et al., 1995), and
may be used as indicators of the biomass of microbes in the environment
(Vestal and White, 1989). Sterols are another group of lipids which have
proven useful for quantifying arbuscular mycorrhizae (Nagy et al.,1980;
Nordby et al., 1981; Schmitz et al., 1991; Dalpé et al. 1994). However,
information about the sterols of genera such as Acaulospora and
Scutellospora is lacking.

The objectives of this experiment are: 1) To test the value of
FAME, PLFA and sterols as markers of mycorrhizae, and 2) To gather
information on the PLFA and sterols of glomalean taxa for which this
information is not available. To do this, Iinoculated Sorghum with
different species of glomalean fungi which belong to different genera.
Following a growth period, the roots were harvested and analyzed for the
FAME, PLFA, and sterol composition. This information will be used in
chapter 3 as the basis of our studies investigating the C transformation

rates in the lipids of mycorrhizal roots.

69

Materials and Methods

Fatty acid terminology utilizes “A:B mC” where “A” indicates the
total number of carbon atoms, “B” the number of unsaturations, and “co”
precedes “C”, the number of carbon atoms between the closest

“ 9’

unsaturation and the methyl end of the molecule. The suffixes c and

“t” indicate cis and trans geometric isomers. The prefixes 1 and a

refer to iso and anteiso methyl branching.

Soil, Inoculum, and Growth Conditions

A sandy clay loam (60.0% sand, 20.0% silt, and 20% clay) pH 6.9,
and cation exchange capacity 10.2 me/lOOg was used as a growth
medium. The soil was supplied by the Plant Biology greenhouses at

Michigan State University. It contained 4.5% by weight organic matter,

24 mg kg'1 total P, 80 mg kg“ K, 1642 mg kg] Ca, and 220 mg kg! Mg.
The soil was passed through a 2 mm sieve and 0.8 liters (740 g) of air-
dry soil were added to each pot. The pots were closed with a plastic lid
and then irradiated with cobalt-6O at the Phoenix Memorial Laboratory,
University of Michigan. Irradiation was carried out for 13 hours with a
gamma dose rate of 3826.1 Gy/hr. Each pot received 1.20 ml of nutrient
solution which contained 1.5 mM CaClz, 0.5 mM K2804, 2.5 mM NH4N03,
0.25 mM MgSO,, 25 uM H3303, 20 uM FeDDHA, 20 uM ZnSO,, 0.5 uM CaSO,,,

0.4 uM H2MoO,, and 0.6 uM CoCLz. The pH was adjusted to 6.8 with a

70

KOH solution. The nutrient solution was filtered (0.2 pm) before being
dispensed to the pots. The pots were watered with 140 m1 of distilled
water every two days throughout the duration of the experiment.

Surface-sterilized seeds (70% ethanol for 30 sec., then 20%
commercial bleach for 20 min., then washed with distilled water) of
Sorghum bicolor L. Moench were germinated in a petri dish for two days
over a sterile moist filter paper before planting. Two seeds were planted
per pot and the plants were thinned to one per pot after the first week of
the experiment.

The pots were placed randomly in a growth chamber and they
were randomized once more four weeks after the start of the experiment

to minimize positional effects. The photosynthetically active radiation
-1
ranged from 450 to 500 uM m'2 s and the photoperiods lasted 14 hours.

The average temperature was 25°C, and the average relative humidity
was 75%. Plants were grown for 8 weeks, left for four days without
watering, and harvested.

The inocula were obtained from the International Culture
Collection of Arbuscular and Vesicular-Arbuscular Fungi at West Virginia
University. These consisted of Acaulospora mellea Spain & Schenck
FL266-2, Gigaspora rosea Nicolson & Schenck BR151A-3, Glomus
intraradices Schenck & Smith UT143-2, Glomus etunicatum Becker &
Gerdemann SC306A, Glomus occultum Walker IA702, Scutellospora

erythropa (Koske & Walker) Walker & Sanders MA453B-2*. The inocula

71

consisted of a mixture of soil, roots, and spores which contained at least
12 spores g". A minimum of 25 g of inoculum were used per pot. Each
pot in the mycorrhizal treatments received inoculum from a single
species. Six replicates of each species and six uninoculated controls were
included in the experiment. Achlamydospore-free filtrate of each kind
of inoculum was added to all pots to achieve a homogeneous diversity of

saprophytic microbes throughout the treatments.

Root Harvesting and Preservation

The roots were separated from the soil by wet sieving (2 mm
mesh). Root material retained in the sieve was finely cut and mixed in
distilled water. The roots and supernatant were then washed with water
through a 200 um sieve, and root fragments with diameters of > 1.0 mm
were lyophilized before lipid analysis, or stored in 5% formalin until
clearing and staining.

Appendix Ais a diagram of the methodology used to obtain the

lipid profiles.

Whole-Cell Fatty Acids (FAME)

The Microbial Identification System (Microbial ID, inc., Newark DE)
was used to determine the total FAME profile of the roots. Samples of
0.02 g dry weight were analyzed using the Microbial ID protocol. The

method was modified to include an internal standard (50 pl of 15:0,

72

0.001 g/ml in hexane) added after the saponification step. This made it
possible to obtain quantitative data in addition to the relative molar
percentage of the molecules. The fatty acids were quantified by

comparison of the peak areas with those of the standard peak.

Phospholipid Fatty Acids (PLFA)

The total lipids were extracted from 0.2 g of freeze-dried roots by a
chloroform-methanol extraction (Bligh & Dyer, 1959) modified to
incorporate a phosphate buffer. The root material was extracted at
room temperature with the chloroform/methanol/potassium phosphate
buffer (12:08 by volume; 50 um , pH 7.4) for at least 2 h, at which time
0.29 volumes of organic-free water and chloroform were added to

separate the aqueous and organic phases. After rotary evaporation at

37°C, the total lipid was fractionated into Neutral-, Glyco-, and Polar
lipid fractions using silicic acid (100-200 mesh) column chromatography
(Guckert et al., 1985). This way, the sterols and phospholipids were
recovered from the same total lipid extract.

The phospholipids were obtained from the polar lipid fraction.
The polar lipid was subjected to a mild alkaline methanolic
transesterification (Guckert et al., 1985) from which the PLFA-methyl
esters were recovered. The PLFA were then separated, quantified, and
identified by capillary column gas chromatography (FID detector,

Hewlett-Packard HP-l 50 m by 0.2 mm inner diameter non-polar methyl

73

silicone column) using the conditions described by Ringelberg et al.
(1989). Quantification was based on comparison of peak areas to an
internal injection standard (19:0) assuming equi-molar responses within
the range of chain lengths between 12-24 C. The structures were verified

using a mass selective detector (Hewlett Packard 5791).

Sterols

The sterols were recovered from the neutral lipid fraction after
saponification in 5% KOH in methanol:water (80:20, vzv). The extract
was derivatized using N, O bis(Trimethylsilyl)trifluoroacetamide (BTSFA).
This formed trimethyl silyl ethers, which were then separated and
identified using gas chromatography (as for the PLFA). Sterol
identification was based on co-injection with the standard cholestane,
relative retention time, and by comparison of the mass spectra with
previously reported spectra. The structures were verified using GC-mass
spectrometry at an electron energy of 70eV with the same columns and
conditions described above (Table 1).

The concentration and mol % of each molecule were determined
with the following formulas:
mol g": ((area of sample peak)(area of standard peak"))(mol ul" of
standard)(dilution factor of sample in ul)(grams of sample)".

mol %= (mol of individual peak)(total mol in sample)"(100).

74

Microscopic Determination of the Percent Mycorrhizal Infection.

The roots were cleared and stained using the technique of
Kormanik and McGraw (1982). The percentage root length occupied by
vesicles, arbuscles, and hyphae was determined by the line intercept
method (Giovanetti and Mosse, 1980). The values are shown in Table 2.
This was modified to use the lines of an eyepiece graticule to determine
the observation points. The specimens were observed at 40 x and the
magnification was increased when necessary to diagnose specific
structures. Three subsamples of 100 intercepts were observed for each
root system. Idid not observe any mycorrhizal infection in the control

treatment.

Statistical Analysis

The lipid profiles of the six pot cultures in each fungal treatment
were evaluated by hierarchical cluster analysis using the average linkage
method of Systat 5.2.1 (SPSS Inc., Chicago, IL, USA). Iused a one-way
ANOVA followed by the Tukey’s mean comparisons to test for differences
in the relative. amount of each molecule between the treatments.
Molecules in concentrations of <1500 pmol or making up <0.25 % of the
total profile were excluded from the statistical and multivariate

analyses.

75

Results and Discussion
Whole-Cell Fatty Acids (FAME)

Mycorrhizal infection is associated with an increase in the total
fatty acid content of roots. The mycorrhizal treatments had at least
21.1 % more total fatty acid content than the controls (Table 3), and the
difference was statistically significant (p<0.05) for all treatments except
G. rosea. Likewise, Pacovsky (1989) and Peng et al., (1993) found
increases in fatty acid concentration in the roots of different host-fungus
combinations. Arbuscular mycorrhizae may improve the phosphorus
nutrition of the host plant. One question that we may ask is whether
the increased lipid concentration in infected roots is due to nutritional
effects, or to the presence of the fungus. Pacovsky (1989) carried out
experiments in which the non-mycorrhizal Sorghum plants were
amended with P fertilizer to achieve a comparable Pnutrition to that of
the mycorrhizal plants. He showed that the increase in fatty acid
content of mycorrhizal roots is not related to the phosphorus nutrition
of the host. Because of this, the increase in the fatty acids is attributed
to the presence of the fungal biomass in the infected roots.

A total of 20 different fatty acids with molar percentages higher
than 5 % were recovered from mycorrhizal and non-mycorrhizal roots
(Table 3). Besides the effect on the concentration of total lipids in roots,
arbuscular mycorrhizae are also associated with a qualitative change in

the lipid composition. For example, the molar percentage of fatty acid

76

16:1 035 was significantly higher (p< 0.05) for all mycorrhizal treatments,
except for G. rosea and G. occultum. Glomus occultum is characterized
because it produces 16:1co7 instead of the 16:1 (05 isomer (Table 3). This
opens the possibility of reliably discerning between the infection by G.
occultum and that of other glomalean species in field samples and
controlled experiments. Molecular markers for detecting G. occultum
should be particularly useful since this fungus is difficult to stain and
quantify microscopically (Table 2). The roots infected by G. rosea are
distinguishable because of the increased percentage of 18:1 009 relative to
the non-mycorrhizal roots (p< 0.01). Others have shown that this fatty
acid is also valuable for differentiating the spores of Gigaspora from
those of other genera (Bentivenga and Morton, 1994).

Several lipids such as 18 C and 20 C unsaturated fatty acids have
been proposed as mycorrhizal markers (Pacovsky, 1987), while 18 C
polyenoic fatty acids has been used as molecular markers of fungi in
general (Vestal and White, 1989). Our results show that the usefulness of
18 and 20C fatty acids as molecular markers depends on the fungal
species present in the roots. For example, Idid not find these molecules
to be generally higher in the mycorrhizae of the different fungal
treatments, except for the increase in fatty acid 18:1 (09 in G. rosea
(Table 3).

The results from this study show that the occurrence of FAME 16:1

coS can not be predicted using genetic similarities or morphologic

77

differences. The occurrence of 16:1 65 as a dominant FAME is shared by
members of the Glomineae (Acaulospora and Glomus) and
Gigasporineae (Scutellospora). In spite of the common occurrence of
16:1 (05, the Glomineae and Gigasporineae have different development
patterns and spore morphologies which separate them taxonomically
(Morton and Benny, 1990). Despite the fact that these two different
glomalean families are similar in their content of 16:1 (05, members
within another glomalean family, the Gigasporaceae, show different
patterns of 16:1 coS abundance.

Studies using nuclear gene sequences and spore morphology place
the genus Scutellospora together with Gigaspora in the family
Gigasporaceae (Morton and Benny, 1990; Simon et al., 1993). Nucleic
acid sequences indicate that Gigaspora diverged from Scutellospora and
may be one of the most recent evolutionary lines of the arbuscular fungi
(Simon et al., 1993). It is remarkable that Gigaspora is distinguished
from the rest of the genera by having the 18:1 (69 as a dominant FAME
instead of 16:1 035 (Table 3). For this reason, Ihypothesize that the loss
of 16:1 (05 FAMEas a dominant fatty acid is a derived character within
the Gigasporaceae. The presence of 16:1 (07 in G. occultum may also
have important evolutionary implications. Glomus occultum, together
with G. leptotichum produce 16:1 (07 as the most abundant 16 C
monoenoic fatty acid (Bentivenga and Morton, 1996) (Table 3). Glomus

leptotichum is thought to be one of the more ancestral members of the

78

arbuscular fungi (Bentivenga and Morton, 1996) and it is for this reason
that Ipropose 16:1 67 is an ancestral character within the Glomales.
The unusual nature of the G. occultum fatty acids is also demonstrated
in Figure 1, which shows that the FAME composition of roots infected
with this fungus are distinct from uninfected roots and to mycorrhizal

roots of other fungal taxa.

Phospholipid Fatty Acids (PLFA)

Sixteen different PLFA in molar percentages of at least 0.25 % were
recovered form the set of mycorrhizal and non-mycorrhizal roots of
Sorghum (Table 4). All mycorrhizal treatments except G. occultum have
a higher mean concentration of total root PLFA than the controls (Table
4). The difference is significant (p< 0.01) for the A. mellea and G.
etunicatum treatments. This agrees with the results of Nagy et al.
(1980), which found that the total amount of phospholipid in the roots
of Citrus increased with the formation of Glomus mycorrhizae.

All fungal treatments except G. occultum have at least 2.6 times
the molar percentage of 16:1 m5 than uninfected roots (p< 0.05; Table 4).
As with the FAME, roots infected G. occultum have 16:1 (07 as the major
unsaturated 16 C PLFA (Table 4). This suggests that PLFA1621 may be a
general indicator of arbuscular mycorrhizal infection. Other studies
have shown that polyunsaturated 20 C PLFA can be used as a measure of

Glomus mycelium in soil (Olsson et al, 1995). In our study of root

79

extracts, Ifound that the concentration of root PLFA 20:1 and 20:4 did
not always increase with inoculation (Table 4). While 20:1 and 20:4
may be good indicators of the extraradical phase, our results suggest that
they are not consistent markers of glomalean fungi in roots (Table 4).
Phospholipid 16:1 (05 was detected in the control roots (Table 4).
The control treatment of this experiment is not designed to produce
sterile roots, and the non mycorrhizal roots may have supported a
rhizoplane microbial community. Ihypothesize that the PLFA1621m5
present in the control roots may be accounted for by rhizoplane
microorganisms. For instance, 16:1co5 may be found in low amounts
(<0.6 % of total FAME content) in the fungal root pathogens Pythium
and Bipolaris (see chapter 1 of this thesis; Gandhi and Weete, 1991).
However, a significant background of 16:1 is unlikely because heavy
colonization by root pathogenic microorganisms is likely to produce

recognizable symptoms of disease and cause root death.

A Comparison of the FAME and PLFA

The mycorrhizae of G. rosea are remarkable because of the
distribution of 16:1 035 in FAME and PLFA. Gigaspora rosea follows the
pattern of the rest of the fungal treatments in that it is associated with
an increase of 16:1 (95 in mycorrhizal phospholipids (Table 4). However,
the concentration of 16:1 m5 in whole-cell extracts of roots infected with

G. rosea is beyond the detection limit of the instruments (Table 3). I

80

hypothesize that this is explained by a lack of accumulation of 16:1 005
in the neutral lipids of G. rosea. This difference in the fatty acid
composition of phospholipids and total lipid underscores the value of

separating the PLFA from the total lipid before the fatty acid analysis.

Sterols

A total of six different sterols were recovered from the mycorrhizal
and non-mycorrhizal roots of Sorghum (Table 5). The total
concentration of sterols is statistically indistinguishable in mycorrhizal
and control roots (Table 5). Similarly, Schmitz et al., (1991) found that
mycorrhizal infection is not associated to increases in the total
concentration of sterols in roots. However, the concentration of
campesterol in the roots of various plant species increases with the
infection of Glomus and Gigaspora (Ho, 1977; Nagy et al, 1980; Nordby et
al., 1981; Schmitz et al., 1991; Dalpé et al. 1994). The fungal symbiont
may be a contributor to this increase, since Glomus spores have
campesterol as a major sterol (Beilby and Kidby, 1980; Dalpé et al.
1994). In this study, all the fungal treatments had at least 23 % more
campesterol than non-mycorrhizal roots (Table 5). The difference was
significant for all fungal treatments (p< 0.05; Table 5) except for the
roots inoculated with A. mellea. This suggests that the use of
campesterol to measure mycorrhizal formation may be restricted to

specific fungal taxa. An advantage of campesterol as a molecular marker

81

of glomalean fungi is the fact that other root infecting microorganisms
should not contribute a significant background signal to sterol
measurements of mycorrhizae, since campesterol is absent from non-
mycorrhizal fungi (see chapter 1 of this thesis), bacteria, and nematodes.
Previous work has shown that ergosterol increases with mycorrhiza
formation in the roots of Trifolium and Zea (Frey et al., 1992). However,
in this experiment the concentration of this molecule was beyond the

detection limit of our instruments.

Multivariate Analysis

Cluster analysis is a procedure by which multiple variables such as
fatty acids or sterols are compared. As a result, a tree diagram is
produced which illustrates the similarity between the different fungal
treatments. Closely related cases form distinct clusters in the tree
diagram. The FAME, PLFA, and sterol profiles extracted from mycorrhizal
and control roots were examined using cluster analysis. Our results show
that the whole set of FAME molecules provides a signature that identifies
control roots (Figure 1). Additionally, the FAME profiles provide insight
about which fungus is present. Figure 1 shows that the roots infected
with G. rosea and A. mellea are recognizable by their FAME profiles. The
PLFA and sterol profiles were similarly analyzed by hierarchical cluster
analysis. However, for both sterols and PLFA, the control roots and the

different fungal treatments did not show distinct clustering patterns

82

(data not shown). This suggests that infection has a more pronounced

effect on the FAME than on the PLFA and sterol profiles.

Mycorrhizal Infection as Determined by Microscopy

The different AMF species varied in the amount of infection as well
as the formation of storage (vesicles) and exchange structures (arbuscles)
(Table 2). The lowest score for the percent infection was observed in G.
occultum with 8.8 %, while the highest values were observed in G
etunicatum with 44.3 %. The different AMF species varied in their
amount of arbuscle and vesicle formation. For example, no vesicles were
observed in G. rosea or S. erytropa. This agrees with previous studies
which have shown that Gigaspora and Scutellospora form few or no
vesicles inside the host’s roots (Gerdemann and Trappe, 1974). The
percent arbuscularization ranged from 3.9 in S. erytropa to 20.2 in G.
etunicatum. Icompared the infection data to the lipid data from the
six replicates in each fungal treatment to test if the concentration of the
specific lipids increase with the microscopical estimates of infection. No
correlation of the amount of specific lipids and the infection estimates
was found (data not shown). Ihypothesize that the lipid content will
follow microscopical measurements when wide differences in the
intensity of infection are compared. However, this experiment was not
designed to obtain a wide range of percent infection within each fungal

treatment. Others have also found that tests for determining the

83

relationship of total fatty acids to percent infection have shown
contrasting evidence. Graham et al. (1995) found that 16:1 fatty acid
correlates with the percent infection, while others did not find a
significant relationship (Pacovsky, 1987). Similarly, measures of intra-
radical biomass of glomalean fungi such as chitin analysis fail to
correspond to percent infection (Paul and Clark, 1989). Ihypothesize
that this may be due to several reasons: 1) The inability of the staining
procedure to discern between active and non-living mycelium; 2) The
line intercept method may not be a good measure of the intensity of
infection since heavily colonized root intercepts score the same as
intercepts with few hyphae, and 3) Possible fluctuations in the lipid

composition of mycorrhizae during the life cycle of the symbionts.

Conclusions

The fatty acid composition of the fungal symbiont is not affected
significantly by the species of host (Bentivenga and Morton, 1994). This
suggests that the use of fatty acid markers to indicate the presence of
arbuscular mycorrhizae may be useful across different host species and
field collected samples. This study shows that the lipids of glomalean
fungi provide the possibility of marking the presence of these organisms
In Situ and may provide insight as to which fungal symbiont is present

in the host’s roots. Future studies will be able to use this methodology to

84

carry out experiments where the populations of glomalean fungi are

monitored in situ.

LIST OF REFERENCES

LIST OF REFERENCES

Beilby, J. 1980. Fatty acid and sterol composition of ungerminated
spores of the vesicular-arbuscular mycorrhizal fungus Acaulospora
laevis. Lipids. 15(11), 949-952.

Beilby, J., and Kidby, D. K. 1980. Biochemistry of ungerminated and
germinated spores of the vesicular-arbuscular mycorrhizal
fungus, Glomus caledonius: changes in neutral and polar lipids.
Journal of Lipid Research. 21, 739-750

Bentivenga, S., and Morton, J. B. 1994. Stability and heritability of
fatty acid methyl ester profiles of glomalean endomycorrhizal
fungi. Mycol. Res. 98, 1419-1426.

Bentivenga, S., and Morton, J. B. 1996. Congruence of fatty acid
methyl ester profiles and morphological characters of arbuscular
mycorrhizal fungi in the Gigasporaceae. Proc. Nat. Acad. Sci. 93,
5659-5662.

Bentivenga, S. 1996. Congruence of fatty acid methyl ester profiles
and morphological characters of arbuscular mycorrhizal fungi
in Gigasporaceae. Proc. Nat. Acad. Sci.

Bligh, E. G., and Dyer, W. M. 1959. A rapid method of lipid
extraction and purification. Canadian Journal of Biochemistry
and Physiology. 35, 911-917.

Cooper, K., and Losel, D. 1978. Composition of lipids in roots of onion,
clover, and ryegrass infected with Glomus mosseae. New
Phytol. 80, 143-151.

85

86

Cox, G., and Sanders, F. 1974. Ultrastructure of the host-fungus
interface in a vesicular-arbuscular mycorrhiza. New Phytol. 73,
901-912.

Dalpé, Y., Grandmougin, A., Veignies, B., Rafin, C., and Sancholle, M.
1994. Effect of arbuscular mycorrhizal infection by Glomus
mosseae on the lipid content of Leek plants (Allium porrum L.).
Fifth International Mycological Congress. Vancouver, Canada.

Farr, D. F., Bills, G. F., Chamuris, G. P., and Rossman, A. Y. 1989. Fungi
on plants and plant products in the U. S. pp. 419-420.
American Phytopathological Society. St. Paul.

Frey, B., Buser, H. R., and Schiiepp, H. 1992. Identification of
ergosterol in vesicular-arbuscular mycorrhizae. Biol. Fertil. Soils.
13: 229-234.

Frey, B., Villariiio, A., Schiiepp, H., and Arines, J. 1994. Chitin and
ergosterol content of extraradical and intraradical mycelium of

the vesicular-arbuscular mycorrhizal fungus Glomus
intraradices. Soil Biol. Biochem. 26, 711-717.

Gandhi, S. R., and Weete, J. D. 1991. Production of the
polyunsaturated fatty acids arachidonic acid and

eicosapentanoic acid by the fungus Pythium ultimum. Journal
of General Microbiology. 137, 1825-1830.

Gerdemann, J. W. and Trappe, J. M. 1974. The endogonaceae in the
Pacific northwest. Mycologia Memoir. 5, 76.

Graham, J. H., Hodge, N. C., Morton, J. B. 1995. Fatty acid methyl ester
profiles for characterization of glomalean fungi and their
endomycorrhizae. Applied. Environ. Microbiol. 61, 58-64.

Guckert, J. B., Aintworth, C. P., Nichols, P. D., and White, D. C. 1985.
Phospholipid ester-linked fatty acids as reproducible assays for

changes in prokaryotic community structure of estuarine
sediments. FEMS Microbiology Ecology 31, 147-158.

87

Ho, 1. 1977. Phytosterols in root systems of mycorrhizal and non-
mycorrhizal Zea mays L. Lloydia. 40, 476-478.

Jabaji-Hare, S., Deschene, A., and Kendrick, B. 1984. Lipid content
and composition of vesicles of a vesicular-arbuscular
mycorrhizal fungus. Mycologia. 76(6), 1024-1030.

Kormanik, P. and McGraw, A. 1982. Quantification of VA mycorrhizas
in plant roots. pp. 37-45. In Methods and Principles of
Mycorrhizal Research. Schenck, N (ed). American
Phytopathological Society, Minnesota.

Nagy, S., Nordby, H. E., and Nemec, S. 1980. Composition of lipids in
roots of six citrus cultivars infected with the vesicular-

arbuscular mycorrhizal fungus Glomus mosseae. New Phytol.
85, 377-384.

Nordby, H. E., Nemec, S., and Nagy, S. 1981. Fatty acids and sterols
associated with citrus root mycorrhizae. J. Agric. Food Chem. 9,
(2), 396-401.

Olsson, P.A., Baath, E., Jakobsen, I., and S6derstr6m, B. 1995. The use
of phospholipid and neutral fatty acids to estimate biomass of
arbuscular mycorrhizal fungi in soil. Mycol. Res. 99, 623-629.

Pacovsky, R. S. 1987. Nutrition of maize inoculated with Azospirillum
brasiliense and Glomus fasciculatum. 7th North American
Congress on Mycorrhizae. University of Florida, Gainesville
Florida.

Pacovsky, R. S. 1989. Influence of inoculation with Azospirillum
brasiliense and Glomus fasciculatum on Sorghum nutrition. pp.
235-239. In Nitrogen fixation with non-legumes. F. A. Skinner
et al. (eds). Kluwer Academic Publishers.

88

Pacovsky, R. S., and Fuller, G. 1989. Mineral and lipid composition of
Glycine-Glomus-Bradyrhizobium symbioses. Physiologia
Plantarum. 72, 733-746.

Paul, E. A. and Clark, F. E. 1989. Mycorrhizal Relationships. pp. 218.
In Soil microbiology and biochemistry. E. A. Paul and F. E. Clark
(eds). Academic Press, San Diego.

Peng, S., Eissenstat, D. M., Graham, K. W., and Hodge, N. C. 1993.
Growth depression in mycorrhizal citrus at high phosphorus
supply. Analysis of carbon costs. Plant Physiol. 101, 1063- 1071.

Ringelberg, D. B., Davis, J. D., Smith, G. A., Pfiffner, S. M., Nichols, P. D.,
Nickels, J. S., Henson, J. M., Wilson, J. T., Yates, M., Kampbell, D.
H., Read, H. W., Stocksdale, T. T. and White, D. C. 1989.
Validation of signature polar lipid fatty acid biomarkers for

alkane-utilizing bacteria in soils and subsurface aquifer
materials. FEMS Microbiology Ecology. 62, 39-50.

Sancholle, M., and Dalpé, Y. 1993. Taxonomic relevance of fatty acids
of arbuscular mycorrhizal fungi and related species. Mycotaxon.
49, 187-193.

Schmitz, O., Danneberg, B., Hundeshagen, A., Klingner, A., and Bothe,
H. 1991. Quantification of vesicular-arbuscular mycorrhiza by
biochemical parameters. J. Plant Physiol. 139, 106-114.

Simon, L., Bousquet, J., Levesque, R. C., and Lalonde, M. 1993. Origin
and diversification of endomycorrhizal fungi and coincidence
with vascular land plants. Nature. 363, 67-69.

Vestal, J. R., and White, D. C. 1989. Lipid analysis in microbial
ecology. Bio Science, 39, 535-541.

Whipps, J. M., and Lewis, D. H. (1980). Methodology of a chitin assay.
Transcripts British Mycological Society. 74, 416-418.

89

0.88000 8. Emma 008838 28888888610803 3. 8.86380 888.080 8.33 83.88.1808— »:0 85878880823088— 8.880.

 

 

2888886 5... 00380888800 8888885 7.8+... 0.0. ... 8888861280

3803838 00.00 008088-028— 800 800 800. 000.000.080.000. 800
080083838 80.00 80090800888888.888008800008858 808 00 008.000.000.000. 800
800838888: 80.00 800 00 008.000.080.000
3838088 80.00 80850880888080-0883 800 000 808.000. 000.000. 000
08:30:28-8 0.88 8888580888 800 00 800.000.000.00800
888802802880 00.00 8:13.02: 8888... 00 000.080

 

.8- an 880888888 80888 888 88:88:88“ 2:." 80809888: 2080888 08. 885 0888888800 88098889 0.0." 88888

8808888088888 08.880338 0: 8886 888882. 0060828888“ :0. :08 0086288088088.

90

Table 2. Microscopical estimates of infection.

 

 

Percent Percent Percent
Inoculum: arbuscles vesicles infection
A. mellea 11.2(4.8) 0.7(0.5) 36.6(8.6)
G. rosea 12.6(9.4) 0.0(0.0) 20.4(10.2)
G. etunicatum 20.2(8.7) 0.6(0.3) 44.3(5.1)
G intraradices 12.8(3.7) 3.6(0.9) 44.0(8.7)
G occultum* 5.1(4.4) 0.2(0.3) 8.8(4.4)
S. erytropa 3.9(1.5) 0.0 (0.0) 20.4(4.5)

 

The values are the mean with the standard error in parenthesis. n=6
pots.

*- The mycorrhizae of G. occultum stained noticeably weaker than the
rest of the treatments, the microscopical estimates may be biased for
this reason.

91

0.8820 0.

0.97.20 00 BEoQEaE 88880 85875098088860 08:00:38 Boa.

 

 

8.888880. moan 08:883— >3=~9€38 0:88:85 Q0335. Q0335. 9.05.0028 020880008053
3103 «8838.225: 8.3833808.» 02208838 3.2.88 «3.8308
888 Eco—0008.03.00 88.82.00 08.00.00 00.20.00 00.08.00 80.00.00 00.08000
3032
8880 0.0 0.20.: 0.0 0.0 0.880.: 0.0 0.8 80.5
808880 0.0 0.0 0.0 0.0 8.8 80.00 0.0 0.0 80.80
80:80 0.0 8.08080 80.82.00 80.08000 0.0 0.0 0.08000
00 88080.00 00.03.00 00080.00 80080.00 80.20.00 00080.00 80.20.00
0088 0.0 0.0 0.0 0.20.3 8.20.00 0.080.00 0.03.00
00 0.880.: 0.0 0.880.: 0.080.: 0.080.: 0.0 0080.00
8080 00.080.00 00,080.00 80.08.00 00.08000 00.02.: 08.03.00 80.03.00
8088 80 00.03.00 0.08080 0.88080 0.880.: 0080.00 3.08080 0080.00
808 ... 0.0 0.0 0.0 0.880.: 0.08000 0.0 0.03.00
00 8080.00 0.08000 0.08000 8.20.: 0.080.: 8880.00 5080.:
80; 80.20.00 0.080.00 0.08080 08080.00 80.03.00 88.20.00 0.03.:
8080 003 8.08000 0.08080 0.080.: 0.20.00 0080.00 8.880.: 0.20.00

 

92

0.88080 0 88888.00.

 

 

0.9880. 8889 008888.0— >8=09€03 00058;. 0:058; Q0035. 9.098.003 09328063
52?: «8.38.2885: 8.3828332». ennirtz 3.8.28 «3.83088
8080 003. 0.0 0.0 0.0 0.080.: 0.0 0.20.: 0.080.:
00; 80 0.0 0.0 0.0 0.0 0.0 0.03.80 0.0
0080 000 8.03.00 0.03.00 0.03.00 8.08.00 0.20.00 0.080.00 0080.00
0080 0.03.00 0.080.00 0.83.80 8880.00 8.03.00 0880.00 8.83.00
0880 800 0.20.80 8.03.00 0.03.00 0.080.80 0.83.00 0.080.00 8.83.00
0880 8.08080 0.08080 80.08.00 80.83.00 00.08000 80.03.00 80.83.:
0080 0.03.00 0080.00 0.03.00 0.0 0.20.: 0.83.80 0.08080
0880 8.080.: 0.83.00 0.03.00 0.0 0.080.: 0.83.80 0.08000

 

58:88 883 8885300 808850883 9.33.

988858.

88.0 320 0888:2886 8880 0636888 om 885 80888— m>§m 888288883 00. 800 00850 80888886.

*

- Him 00 8888 8888800780 88588888 00 @958— 00; 8.8880. 888%.

0

08880. 8880 088880. woman 8880888880880 804 888 86858 0.00 888 085888. 88"..
.280 $88.88 :8 8:0 80 8.2 88808288 806 8888888888088 2. 888: 025m 5 885 $8080. $8800 8880 808 08.

93

0.8805 8.

31?? 00 30.88.8380 8:80 8805-3000888860 03.00285 3080.

 

 

8.88880. moan 008888.08 082833.053 Q0055. Q3383. Q3583. 9.098.003 020303003
5130 «8.38.2822: 8.3832333 2.2808883 8.8.3 «0.83088
8888— 8.80:0 0.03.00 0.080.: 0.080.: 0.080.: 0.08080 0.08080 0.080.:
830.3
:00 0.080.00 0.080.: 0.03.00 0.03.00 0.080.: 0.080.: 8.03.00
800 0.080.: 0.080.: 0.080.: 0.03.5 0.080.: 0.080.: 0.03.3
80880 8.080.: 0.08080 0.08080 0.03.00 0.03.00 0.080.: 8.03.00
00880 0.08080 8.080.: 0.08.00 0.080.: 0.080.: 0.880.: 0.03.00
00 00.03.00 00.08080 00.03.00 00.02.00 00.08080 00.02.80 00.03.80
3.800 8.080.: 0.080.: 0.080.: 8.20.: 8.03.00 0.080.: 8.080.:
80080 00.03.00 00.08000 00.83.00 00.03.80 08.03.80 00080.00 00.82.00
0030 0.080.: 0.080.: 0.03.00 0.080.80 8.20.00 0.03.80 8.080.:
0890 08880.00 0.08.00 0.080.: 80.20.00 0.83.80 80.83.00 0080.00
0890 0.08000 8.20.00 8.03.00 0.080.00 0.83.80 8.08080 8.20.00
00 0080.00 8.080.: 0.080.: 0.08080 0.080.: 0.080.: 0.080.:
3.800 0080.00 0.080.: 8.080.: 0.03.00 0,080.00 0.080.: 0.080.:
00880 0.20.00 8.080.: 0.08000 0.0 0.0 0.080.: 0.0

 

94

.320 0 903.3.

 

 

3.8% moan nonna— >8=~9€o§ Q35: @355. 03:23. @536an 0.3308003
szma «222.235: 53035.29. 62:22: 3.30 «Jawown
00:80 0.20.00 0.20.00 0.03.00 0.03.8 0.20.00 0.03.00 0.20.00
NNH0 0.03.5 0.20.5 0.20.: 0.20.3 N.03.uv 0.03.: 320.3
0.03 0.20.3 903.8 0.03.5 003.00 0.02.3 0.03.3 0.03.3

 

50:3 ”8 w<oBma 835030 2.83.

On; 90 33. momma monocaasm an ”H 5me 0.8.0 30 $595. an o. .25

«£52. 3 :5 80 8;. 5938 9a 8:838:05 cm 880 3.?» 0: So mwBEo. 95:0 :5 now” on :8 84;
5938 :6 382: cm :5 86— wCu> 88:33 3. :5 muonwmo Bang—o.

95

.505 9

083; 00 Swooiawm: M50 sonLaonEwfia 00‘0025 808.

 

 

«83: 2:22: >. Satan Q. 3:38:35 Q. 033353,. Q. 22:33 Q. ~33 0. «35009
88— 38020 0902903 04.03.03 00.03003 00.03003 00030.3 0.3.20.3 09200.03
803?

0850382: 00.03.03 0002.3 0903.3 00.293 0903.3 0903.3 0903.3
0308382: 00.20.03 00.03.03 0.3.20.3 00.0.3.3 530.03.03 00.02.03 h3.03.03
0308325390303 920.03 903.03 0.03.3 0.0 N903.03 0.20.03
0:383— N9i~.3 0903.3 0903.03 09.20.53 0.03.03 00.293 3.03.3
5:395.-— 0.~C.03 N903.03 N903.3 0.20.03 0.03.03 $120.3 0.03.03
5:39:50 0.03.03 0.03.3 0.03.03 0.03.03 0.0 0.03.03 0.03.03

 

598m 30 55300 385030 383.

$6050 8202:?

a. .25 $38 0: :5 80 3s 5938 :5 8:83:50: cm 83—
.4.88; w: :5 3820. 83—0 :5 8m" cm :6 32m 0:938 90 0082: cm :5 88— mnoaow ”89580 03 3. 9n

96

0.0 Euclidean distance

 

G. occul tum 3
S. eryn'opa 3
5' mp3 1 ——j—
S. erytropa 6
G. occul tum 6
G. occul tum 4
G. occultum I:
G. occul tum S
G. occultum 2
con rm] 2
con tro14
con (to) 1
con U013
con U016
con 17015

G. rosea 6 :—
G. roses 4

G. r050! 2 —— ‘
G. rosa 1

6' mm 5:—

G. r0501 3

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

A. mellea 4
A. mellw 2
A. mella 3
A. mellea 6
A. melIea 5
A. mellm 1
G. in traradices 5
G. in mdices 2
G. mmdices4
G. in tramdices6
G. ina'aradices 1 :—
G. etunica tum S
G. etum'ca tum
G. etunica tum 3
G. etum'ca tum 1
G. etunica tum 4
G. etumca tum 2 H

SW“:—
Serytropa4

S. erytmpa 5

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

G. in mdices 3

Figure 1. Hierarchical cluster analysis of the FAME profiles.

5.0

Chapter 3

PATTERNS OF CARBON ALLOCATION AND FATTY ACID TURNOVER

IN MYCORRHIZAL SORGHUM

Abstract

Radioactive tracers were used to study the carbon allocations to
plant tissues, to below-ground respiration, and to soil in the Sorghum
bicolor-Glomus clarum mycorrhizal association. Sorghum bicolor L.
Moench was grown in sterile soil or inoculated with the arbuscular-
mycorrhizal fungus (AMF) Glomus clarum Nicolson & Schenck.
Mycorrhizal and non-mycorrhizal plants 51 day old, were subjected to a
3 hour l‘CO, pulse exposure followed by a 24 day chase period with
sequential harvesting. Infected plants assimilated 21 % more 1“C than
non-mycorrhizal plants, and had a higher percentage and absolute
allocation of 1‘C to root tissue, below-ground respiration, and soil.
Mycorrhizal plants had a higher content of total lipids and total fatty
acids. The mycorrhizal fatty acid 16:1 m5 comprised up to 29.5 % of the
total fatty acid content of mycorrhizal roots. Non mycorrhizal roots had
only trace levels of this molecule. The pulse-chase experiment
demonstrated that the lipids of mycorrhizal roots are a dynamic pool of
C with measurable incorporation and turnover of 1“C. Thin layer

chromatography was used to physically separate the fatty acids

97

98

extracted from the roots. The radiolabel content of particular fatty acids
was determined by radiography. The ”C turnover time of the
mycorrhizal fatty acid 16:1 m5 was calculated at 210 h". The lipids of
non-mycorrhizal roots incorporated less radiolabel and there was not a
distinct turnover of the assimilated 1‘C, underscoring the difference in
the lipid C cycle between the arbuscular mycorrhizae and non-

mycorrhizal roots.

Introduction

A majority of the land plants are associated with mycorrhizal
fungi, and the allocation of C from host to fungus is a fundamental part
of the economy of the mycorrhizal symbiosis. It has been shown the
infection with mycorrhizal fungi increases the flow of fixed C to the root
system and soil (Pang and Paul, 1980; Snellgrove et al. 1982; Jakobsen
and Rosendahl, 1990). For this reason, roots and their associated
mycorrhizal fungi are an important supply of reduced C to soil
ecosystems and affect the amount of C compounds available to the soil
biota (Finlay and deerstrom, 1992). This increase in the quantity of C
available to soil organisms influences root interactions with pathogens,
decomposers, the grazing of roots and mycelium by fauna, and soil
aggregation (Finlay and Stiderstrom, 1992).

The arbuscular-mycorrhizal fungus (AMF) appears to be exclusively

dependent on the host plant for it’s C nutrition (Ho and Trappe, 1973),

99

and it has been observed that the fungus will rely on it’s own C reserves
when separated from the host plant. For example, Olsson et al. (1993)
observed a reduction in the storage lipids of the fungus after harvesting
the host and incubating the mycorrhizal biomass in the soil. The AMF
has an extraradical phase that increases the absorptive area of the root
system, and an intraradical phase that acts both as the exchange surface
between the symbionts and as the carbon storage compartment of the
fungus. The amount of C that is allocated to the intraradical phase of
the AMF has never been measured directly because the close association
of the root and the fungus limits the feasibility of a physical separation
(Tinker et al. 1994). Because of this, little is known regarding the C cycle
of arbuscular-mycorrhizal roots (AM) and the associated fungus.

Plants contribute up to 50% of their photoassimilated C to their
roots and ultimately to the below-ground ecosystem (Warembourg and
Paul, 1973). The mycorrhizal fungus is an important factor both in the
retention and loss of C in terrestrial ecosystems since the AMF can act as
a net C immobilzer, while mineralizing C at a high rate. For example,
Horwath (1993) hypothesized that slow root turnover values in AM are
explained by immobilization of C by the fungal symbiont. However, the
molar growth yield of the AMP has been proposed to be low because of
the C cost of active transport and the associated high respiratory loss
that results in a relatively short permanence of C in the AMF biomass

(Harris and Paul, 1987).

100

Lipids play an important role in the C economy of the arbuscular
mycorrhizal fungus. Ericoid mycorrhizal fungi accumulate C in
carbohydrates such as glycogen and mannose (Stribley and Read, 1974).
Fungal-specific sugars are not abundant in AMF, and lipid synthesis may
have an analogous role to the sugars of the ericoid fungi. For example,
Cox et al. (1975) showed the incorporation of photoassimilated 1‘C in
lipid droplets of AMF cells. Mycorrhizal roots contain high amounts of
several fatty acids that are absent in non-mycorrhizal roots (Nagy et al.,
1980; Nordby et al., 1981). The higher lipid content of mycorrhizae has
an effect on the C dynamics of the roots. Infected roots may have higher
construction costs than non-mycorrhizal roots because of the increased
lipid content (Eissenstat et al., 1993). Specifically, fatty acid 16:1 (95 has
been proposed as an indicator of the C cost of the mycorrhizal symbiosis
(Graham and Hodge, 1993), since this molecule is a predominant
component of the cell mass of the AMF (Beilby, 1980; see chapter 2 of
this thesis).

Carbon turnover is a result of the exchange of C into and out of a
biological compartment (Robertson, 1957). The C turnover time is the
amount of time required to replace the C present in a C pool. The C
turnover time is an important variable because it may indicate if the C
pool of interest is acting as a long-term storage pool. Determining the
turnover of AMF in roots is necessary for understanding the role of these

microbes in the retention and cycling of C. The AMF lipids offer the

101

opportunity to determine the turnover of an important biomass
component of the AMP by a pulse-incorporation of radiolabeled C,
followed by sequential measurements of the radioactive signal in specific
lipids.

The objectives of this experiment are: 1) Measure the patterns of
C incorporation and retention in the plant tissues, root lipids, soil, and
soil atmosphere of mycorrhizal and non-mycorrhizal Sorghum, 2)
Measure the turnover of C in the fungal-specific lipids of the mycorrhizal

I'OOIS.

Materials and Methods
Environmental Conditions and Inoculation

Plants were grown in 2 Kg of sieved (5 mm) sandy loam of pH 6.2,
cation exchange capacity 4.5 me/IOO g, 1.2 % by weight organic matter,
12.5 mg kg" total P, 88.0 mg kg“ K, 647.5 mg kg-1 Ca, 121.5 mg kg‘1 Mg.
The soil was collected from the Ap horizon of a Kalamazoo loam from
Kellogg Biological Station, Michigan. The soil was irradiated with cobalt-
60 (13 hours with a gamma dose rate of 3826.1 Gy/hr) and stored at -
20°C until the start of the experiment. Each pot received 300 ml of
nutrient solution (1.5 mM CaClz, 0.5 mM K2804, 2.5 mM NH4N03, 0.25
mM MgSO4, 25 uM H3BO3, 20 uM FeDDHA, 20 uM ZnSO‘, 0.5 uM CaSO4, 0.4
uM H2MoO‘, and 0.6 uM CoCLZ, pH to 6,8 with KOH). Each pot was

watered to field capacity with distilled water every two days.

102

Surface-sterilized seeds of Sorghum bicolor L. Moench (70% ethanol
for 30 sec., then 20% commercial bleach for 20 minutes, then washed
with distilled water) were germinated in a petri dish for two days over a
sterile moist filter paper before planting. Two seeds were planted per pot
and thinned to one after the first week of the experiment.

The pots in the mycorrhizal treatment (M) were inoculated with
50 g of Glomus clarum Nicolson & Schenck (INVAM BR147B-4) by adding
a mixture of infected roots and soil-borne spores. The non-mycorrhizal
treatment (NM) received a chlamydospore-free and root-free filtrate of
the mycorrhizal inoculum. All M and NM plants were inoculated with
Azospirillum braziliense (ATCC # 29729). The bacteria were grown in PSS
(Hylemon et al., 1973) at 30 °C for 72 h and a 1.0 ml cell suspension
(0.015 g ml“) was added to the soil 7 days after planting.

A total of 40 plants (20 M and 20 NM) were placed inside a 5.43
m3 Plexiglas chamber with a sealed wood frame and base. The plants
were grown with natural light supplemented with high pressure sodium

lamps placed outside the chamber. The photosynthetically active
radiation ranged from 250 to 1200 11M m”2 5'1 during the 16 hour

photoperiod. The temperature ranged from 23 to 29 0C.

l‘COzPulse Labeling
Plants were subjected to a single exposure of 1‘CO, 51 days after the

start of the experiment. A total amount of 92 11M of labeled C in the

103

form of Na,”co,, with a specific activity of 154.2 MBq/mg C was used.
The labeled Naf‘CO, was mixed with unlabeled Na2C03 to form a solution
of 0.77 M. The concentration of the CO2 in the chamber was maintained
at ambient concentration manually by monitoring the internal CO2
concentration with an infra-red gas analyzer and producing CO2 as
needed. The 1“CO; was produced by reacting the Na2C03 solution with an
excess 85% lactic acid. A total of 170.2 MBq were added to the chamber
during a 3 hour period. The canopy was opened at the end of the pulse-
1abeling and the atmosphere within the canopy was purged to the

outside of the greenhouse at a rate of 17.7 m3 hr“.

Harvests
Five harvests were carried out 1, 3, 6, 12, and 24 days after the
1“C02 exposure. At each harvest time, the shoots were separated from the

roots by clipping and the whole cylinders and shoots were placed at -20°.

Plant Biomass

The shoots were dried at 65 °C for 24 h and their mass was
determined gravimetrically. After the biomass determination, the shoots
were ground and stored at 5°C until the nutrient and biomass analyses
were carried out. The roots were separated from the soil by wet sieving.
After washing, the material was freeze dried and the biomass of the fine

(< 1 mm diameter), and coarse (> 1 mm diameter) roots was recorded.

104

The freeze-dried roots were stored at 4°C until the lipid and radioactivity

analyses.

Shoot Phosphorus and 1‘C Content

The shoots of the first harvest were analyzed for P content.
Samples of 0.5 g were ashed (500 °C for 5 hours). The ashes were
digested for 1 hr in 25 ml 3N HNO3 in 1000 mg x kg'l LiCl. The digests
were filtered and mixed with a 0.3 N NaOH solution (1:9 volzvol). These
were then analyzed colorimetrically with a Flow injection Analyzer
(Lachat Chemicals Inc., Wisconsin, USA) (Murphy and Riley, 1962). The
mean total phosphorus in the shoot of the M and NM plants was
statistically indistinguishable as planned. M plants had a mean P
content of 12.2 mg, while that of the NM was 10.3 mg.

Ground shoot subsamples of known weight were analyzed for
specific activity and C content. The samples were combusted with a
biological sample converter (Europa Scientific Roboprep-CN) in series
with a mass spectrometer (Europa Scientific 20-20 Stable Isotope
Analyzer). The mass spectrometer provided the area under the curve for
the CO2 peak. The C content of the samples was determined by
comparison of the sample areas to those generated by sucrose standards
of known mass. The CO2 evolved by combusting the samples was
trapped in Carbon 14 Cocktail (R. J. Harvey Instrument Co., Hillsdale, N.

J.) and the radioactivity was measured by liquid scintillation. The

105

effectiveness of the traps was estimated by combusting and trapping the

CO2 from samples of 1‘C leucine of known mass and specific activity.

Root l‘C Content and Lipid Analyses

Coarse and fine root subsamples were cut, mixed, and analyzed for
the radiolabel content in the same way as the shoot material. The
specific activity of the coarse and fine roots was statistically equal (data
not shown). The total radiolabel content of the root system was
determined by the adding the values of the coarse roots and fine roots.

Appendix B is a flowchart that represents the sampling and
analysis scheme for the root lipids.

The presence of mycorrhizal infection was determined by the
measurement of molecular marker fatty acid 16:1 (:35. Samples of 0.02 g
dry weight were analyzed using the methodology and software of the
Microbial Identification System (Microbial ID, inc., Newark DE). Samples
of freeze-dried fine roots (0.02 g) were placed in a glass centrifuge tube
with a Teflon-lined screw cap. Two ml of saponification reagent (150 g

NaOH in IL 50% MeOH) were added to each tube. The tubes were

vortexed briefly, and heated at 100°C. After the 30 minutes, the tubes
were cooled to room temperature in a water bath. At this point, 50 ul of
an internal standard (15:0, 0.001 g/ml in hexane) was added to the
saponified mixture. The methylation was carried out by adding 4.0 ml

of 54% 6N HCL in MeOH to each tube. The tubes were then vortexed

106

briefly, heated to 80°C for 10 min. in a water bath and immediately
cooled to room temperature.

The fatty acid methyl esters were extracted by adding 1.25 ml of
1:1 hexane: methyl-tert butyl ether and rotating the tubes end over end
for 10 min. The organic phase (top) was removed with a Pasteur pipette
into a 13x100 screw capped glass culture tube. A second extraction of
the remaining root debris was carried out and the organic phases were
combined. The pooled extracts were washed by adding 3.0 ml of 1.2%

NaOH in H20 to each tube, and turning for five minutes. If necessary,

the tubes were centrifuged at low rpm for a clear separation of the
phases. The organic phase was removed, transferred to a glass 13x100

culture tube, and dried under a stream of N2. The dry samples were

stored at -20°C. Before analysis, the samples were resuspended in 1.25
ml of 1:1 hexane: methyl-tert butyl ether and placed in crimp-top GC
vials. The fatty acids were quantified based on the relative mass
response of the detector to the standard peak.

Fatty acid terminology utilizes “A:B coC” where “A” indicates the
total number of carbon atoms, “B” the number of unsaturations, and “co”
antecedes the number of carbon atoms between the nearest unsaturation
and the methyl end of the molecule. The suffixes “c” and “t” indicate

cis and trans isomers. The prefixes 1 and a refer to iso and anteiso

methyl branching.

107

The total lipids of root subsamples (30 mg) were obtained using
the method of Bligh and Dyer (1959). The extract was dried under N2
and re-suspended in 1 ml of chloroform. A subsample of known volume
was placed in a glass fiber filter of known mass (type A/E, Gelman
Sciences, Ann Arbor, USA.) and dried at 75 °C for one hour. The weight
of the lipid subsample was measured gravimetrically and the specific
activity of the lipid material in the filter was determined using the same
procedure as for the plant biomass. With these data, the lipid mass and
radioactivity per unit of root weight were calculated. A second
subsample of the total lipids was taken for the measurement of the
radiolabel content of the root fatty acids.

The fatty acid methyl esters (FAME) were obtained by methylating
the lipid extract using the procedure of Morrison and Smith (1964). The
methylation product was dried and re-suspended in hexane.

Subsamples of the methylation product were taken for liquid
scintillation analysis and argentation TLC. The argentation TLC allows
for a physical separation of the FAMEs by the number of double bonds
and C chain length (Morris, 1966). The plate preparation,
developments, and visualization were carried out following the methods
detailed by Cahoon and Ohlrogge (1994). The FAMEs bands were
identified by comparison with authentic standards (16:1, 18:1, 20:1 and
22:1 Sigma Chemical, St. Louis USA.) A commercial standard for fatty

acid 16:1 015 was not available. For this purpose the FAME extract from a

108

E. coli clone (pDES 16) that produces 16:1 (05, 16:1 037, and 18:1 m9 as
the dominant monoenoic fatty acids was included (Schultz et al., 1996).
The argentation TLC achieved a full resolution of saturated, dienoic, 16:1
035, 16:1 co7, 18:1 (97, 18:1 (09, and 20:1 FAME. The plates were analyzed
by radiography using a Packard Instant Imager (Packard Instrument
Company, Illinois, USA.) at a scan time of 1.5 h to measure the
radiation contributed by each fatty acid band. Up to three main bands
corresponding to 16:1 (95, saturated fatty acids, and polyenoic fatty
acids were observable in the radiographs (Figure 1). Some bands which
were not visualized produced a detectable radioactive band in the
radiogram and were quantified. The readings of the liquid scintillation
analysis of the same extracts as those used for the TLC plates were used
to calculate the radioactivity per unit of root mass contributed by the
different FAME bands.

Transport of a substance (such as C) into and out of a
compartment (such as 16:1 o5 fatty acid) results in the incorporation
and turnover of C in 16:1 (05. The turnover time is the interval of time
necessary for the amount of C that goes in and out of 16:1 m5 to be equal
to the total amount of C present in 16:1 (1)5 (Robertson, 1957). The
turnover of the labeled C in the mycorrhizal fatty acid 16:1 015 was
calculated using four assumptions: 1) Aconstant amount of 16:1 (05 in
the roots (referred to as the steady state), 2) A single pool of 16:1 015

with a decomposition curve of one slope, 3) No recycling of the label

109

between 16:1 (05 and other pools, and 4) The maximum specific activity
of the lipid occurs 24 hours after labeling and corresponds to the time of
the first harvest. Given these assumptions, the turnover of 1‘C in the
16:1 (05 fatty acid should conform to the exponential first order
function:

A1=At(e-u)

Where A,- is the net increase in labeled C in 16:1 015 after incorporation in
terms of mass of 1‘C per root; A, is the mass of the tracer in 16:1 (115 at
time t measured in mass of 14C per root; k is the turnover rate constant
in units of inverse time, t is the time between the end of the pulse
labeling and the measurement. The turnover rate constant k is the slope
of the line defined by the line of Ln A,. against time. The turnover time
is 1/k. The turnover rate (4n) is calculated using the relationship:
¢1=kMi

Where M, represents the total mass of 16:1 o5 in the root system. For
example, using the average mass of 16:1 (05 for the first harvest (8.95

mg), I calculated a C turnover rate of -42.5 pg plant’l h“.

Below-Ground Respiration
The soil atmosphere was sampled from a port located at the
bottom of the cylinder. The below-ground respiration of the 8 plants of

the last harvest was measured starting at 8 hours after the pulse label.

110

The soil atmosphere was flushed for 10 minutes before the first sampling
period, then each pot was sampled by continuously extracting air from
the below-ground airspace at a rate of 0.02 m3/hour. The CO2 was
trapped with 350 ml of 3 M NaOH, and the Bq/hour were estimated using
a Packcard 1500 Tri-carb Liquid Scintillation Analyzer. Six sampling
periods of 8 hours were carried out sequentially, with one hour intervals
between samplings. The below-ground atmosphere was not sealed from
the above ground atmosphere. This system assumes that the
atmosphere sampled from the below-ground sampling has negligible
exchange with the atmosphere from the above ground canopy. This was
achieved by flushing the atmosphere of the above ground canopy with
l‘COz-free air at a rate of 17.7 m3 hr'1 concurrently with the sampling of

the below- ground respiration.

Soil

Soil subsamples were observed under a dissecting microscope and
all fine roots were removed. Root-free soil samples were dried at 65 °C
and the specific activity of the soil C was determined in the same
manner as the plant biomass. To test for the presence of radiolabel in
soil carbonates, randomly selected soil samples were acidified (0.5 ml 1N
HCL per gram of soil) and their specific activity was compared to non-

acidified samples. No detectable amount of soil 1‘C was present in the

111

form of carbonate. The measured radiolabel in the soil is therefore
accounted for by the cellular and extracellular organic matter.

All the soils from the M treatment were sampled for the
concentration and radioactivity of mycorrhizal spores. The spores were
obtained from 15 g soil samples by wet sieving (38 um mesh), followed by
the method of sucrose-gradient centrifugation modified from Daniels
and Skipper, (1982). The soil was sieved by washing with tap water
through a 420 um sieve into a 38 um sieve that collected the spores. The
spores were then transferred with tap water from the 38 um sieve into a
50 ml centrifuge tube, and the spore suspension was centrifuged (5 min.
at 2000 rpm) using a swinging bucket type rotor. The water and root
debris in the supernatant were decanted, and the spore pellet was re-
suspended in 20 ml of 45 % sucrose in H20. The spore suspension was
then balanced and centrifuged for 1.5 minutes at 2000 rpm. The sucrose
solution containing the spores and hyphae was decanted into a clean
centrifuge tube and centrifuged for 1.5 min. at 2000 rpm. The
supernatant containing the spores was transferred into a 38 um sieve
and the spores were rinsed with distilled water. The spores were then
transferred into a petri dish for counting under a dissecting microscope.
After this, the number of mycorrhizal spores per gram of soil was
calculated. The spores were separated from the solution by filtration

(Gelman Sciences type A/E glass fiber filter, Ann Arbor, MI), dried (65°C,

112

24h). The spores with the glass filter were then analyzed for the specific

activity in the same way as the soil and plant material.

Statistical Analysis

The pots were placed in four blocks as a Split-Plot design with the
main plots determined by harvest time and the sub-plots determined by
the mycorrhizal treatments. The main plots and sub-plots were
completely randomized. The harvest times and mycorrhizal addition
were analyzed as treatments using the Split-Plot Analysis of Variance of
SAS version 6.11 (SAS Institute Inc., Cary, N. C., U. S. A.). n: 4 plants per

mycorrhizae and harvest treatment combination.

Results and Discussion
Carbon Allocation

The infection by G. clarum of the Sorghum plant affects the
amount of C assimilation and the distribution of the fixed 1"C to the
Sorghum tissues and soil. On average, M individuals assimilated a total
of 21 % more labeled C than the NM (Table 1). The 1“C allocation to the
below-ground portion of the individuals is the sum of the amount of MC
recovered from the root biomass, the soil, and the below-ground
respiration. In this experiment, mycorrhizal infection is associated with
a higher C incorporation plus an increased below-ground C allocation.

The M plants allocated an average of 2.2 MBq below-ground; the average

113

for the NM plants was 1.5 MBq (Table 1). Mycorrhizal individuals
allocated 52.1 % of their fixed 1‘C to below-ground, while the NM plants
distributed 43.5 % of their label below-ground (Table 1). Other pulse-
chase experiments to study the transfer of C from host to fungus have
shown that mycorrhizal plants allocate more of their C fixation to the
mycorrhizal roots when compared to non-mycorrhizal plants (Pang and
Paul, 1980; Snellgrove et al. 1982; Jakobsen and Rosendahl, 1990). These
experiments involved different growth media, pulse and chase periods,
host-fungus combinations, and plant ages. Despite the differences in
experimental design, all of these studies observed distinct effects of the
mycorrhizal fungus on the C assimilation and allocation (Tinker et al.,
1994).

The difference in the 1“C allocation below-ground between M and
NM individuals can be used to quantify the additional diversion of C to
the root/soil system caused by the fungal symbiont. This should be a
legitimate comparison for this experiment, since the total shoot
phosphorus content of the M and NM plants is statistically equal. This
suggests that the differences in allocation pattern between the M and NM
should be due to the C sink imposed by the fungus and not by
differences in the nutritional status of the M and NM. My results
demonstrate that infection with Glomus clarum is associated with an
additional 0.7 MBq of labeled C distribution to below-ground (Table 1).

This represents 15 % of the total fixed 1‘C of the M plants and is the C

114

demand imposed by the fungal symbiont. The 15 % increased allocation
of C to mycorrhizae may be accounted for by several factors: a) Fungal
respiration, b) Root respiration, c) Allocation to mycelial biomass, d)
Allocation to root biomass, e) Mycelial respiration (internal and
external), and f) Exudation from roots or hyphae.

In this experiment, Idid not measure the photosynthetic rate of
the individuals directly. However, this can be estimated from the
radioactivity data. Table 1 shows that the plants in the M treatment
assimilated a larger amount of total label. Nevertheless, the M shoots
have less biomass than the NM shoots in the first harvest (Table 2). On
average, the photoassimilation rate of radiolabel by the M plants was 30
% higher than that of the NM plants (0.26 MBq g shoot" h" for the M
treatment, vs. 0.20 MBq g shoot" h" for the NM treatment). The ability
of plants to compensate for the needs of the mycorrhizal partner was
first suggested by Paul and Kucey (Paul and Kucey 1981; Kucey and
Paul; 1982). My data with a different plant type than that of Paul and
Kucey further shows the ability of the plant to adapt it’s photosynthetic
rate to the C demand of the fungus. In this study, there was not a
complete compensation, for the M shoots have less biomass than the NM
(Table 2). It has been hypothesized that the effect on the
photosynthetic rate is explained by a the improvement of the water

balance, increased leaf tissue P, higher specific leaf area, or

115

phytohormones associated with mycorrhizal infection (Harris et al.,

1985).

Allocation to Plant Tissues

The growth of the shoots was higher than root growth during the
chase period in both M and NM individuals (Table 2). This difference in
shoot and root growth rate is reflected in the radioactivity
measurements, which show that the shoots incorporated more radiolabel
than the roots in both NM and M (Table 3). The total amount of label
in the shoots and roots was relatively stable throughout the harvest
periods and this was true for both the M and the NM treatments (Table
3). This is illustrated by the lack of a harvest effect for the total ”C
content of both shoots and roots in the analysis of variance (Table 4).
This pattern indicates that the majority of the C remaining in the plants
24 h after the 1‘C pulse was already incorporated into long-term storage
or structural components and is not turned over to a significant degree.
This should be expected for a rapidly growing plant that is allocating a
great deal of it’s photosynthate to the production of structural and long-
term storage C pools.

Our results imply that the C allocation from Sorghum to Glomus is
a C drain that may be regarded as a potential loss of shoot growth unless
full photosynthetic compensation occurs. Non-mycorrhizal shoots had a

higher biomass than the M shoots (Tables 2 and 4). However, the total

116

amount of radiolabel allocated to the M and NM shoot tissues is
statistically indistinguishable because the specific activity of the M
shoots is higher than that of the NM shoots (Table 5). Other studies
show contrasting results in regards to the beneficial, or parasitic nature
of mycorrhizal fungi. In some instances, previous work found that the
cost associated with mycorrhizae represents a C drain that is not

completely compensated for by the host’s C fixation. For example, Harris
et al. (1985) used 1"C02 labeling to study the tripartite symbiosis of

Rhizobium-Glomus-Glycine. Six week old mycorrhizal-rhizobial plants
had a higher fixation rate than the controls. However, the growth rate of
the symbiotic plants was less than that that of the non-inoculated
controls because of the allocation of photosynthate to the symbionts.
Snellgrove et al. (1982) found a similar kind of pattern in Allium
infected with Glomus. They carried out a pulse-chase experiment and
observed that mycorrhizal plants transported 7% more C to the roots,
while having a reduced leaf mass relative to non-mycorrhizal plants. In
contrast to the two previous references, other studies have found that
mycorrhizal plants had similar or higher growth rates relative to non-
mycorrhizal plants, implying that the host compensated for the C
demand from the symbionts with an increased photosynthetic rate (Pang
and Paul 1980; Paul and Kucey, 1981). The degree of compensation is
dependent on the efficiency of the symbiosis relative to the soil-plant

nutrient status. Higher fertility may result in parasitism, whereas the

117

AMFmay act as a mutualist when it can help obtain limiting resources
such as soil P. It should be possible to modify the rates of fertilizer
amendment to achieve a more similar growth patterns of M and NM
plants.

Mycorrhizal infection was associated to increased allocation and
long-term incorporation of C to the root system. Mycorrhizal roots had
significantly higher specific activity and total radiolabel content than
the NMroots (Tables 3, 4, and 5). This contrasts with previous studies
that did not measure an increase in the incorporation of labeled C to the
root biomass of mycorrhizal plants (Paul and Kucey, 1981; Harris et al.,

1985).

Allocation to Below-Ground Respiration

The amount of 1‘C recovered from the below-ground atmosphere
was consistently higher in the M pots relative to the NM (Figure 2; Table
4). Each M plant evolved a total of 0.55:0.08 MBq, while NM individuals
respired a total of 0.40:0.05 MBq. This agrees with previous studies that
show increased CO2 evolution of mycorrhizal plants relative to non-
mycorrhizal controls using pulse-chase experiments with different plant-
fungus combinations. (Pang and Paul, 1980; Kucey and Paul, 1982;
Snellgrove et al. 1982; Harris et al., 1985). For both treatments, 72% of
the total below-ground respired 1“C was recovered within 34 hours after

the pulse label. The root biomass of the M plants was statistically equal

118

to that of the NM (Tables 2 and 4), which suggests a higher specific
respiration rate of the M plants. The observed increase in below-ground
l4CO2 recovery from the M soil may be accounted for by any one of
several factors: a) An increase in the specific respiration rate of the
host, b) The respiration of the AMF, c) Increased rhizosphere respiration
in the M treatment, and/or (1) As shown on chapter 2, mycorrhizal roots
have increased amounts of fatty acids relative to uninfected roots. The
biosynthesis of fatty acids is an energetically expensive process, requiring
the expense of relatively high amounts of ATP and NADPH. Because of
this, the production of additional fatty acids in mycorrhizal roots may
be partly responsible for the increased below—ground respiration in M
individuals. As stated above, soil may be a contributor to the M and NM
below-ground respiration because of the observed (although non-
significant) reduction of soil 1“C content during the 24 day chase period
periods. Other studies show that the microbial respiration of root

exudates, can be an important source of below-ground CO2 production.

For example, Martin and Kemp (1986) studied the translocation ofC to

the roots of wheat plants by pulse labeling and measurement the 14CO2
content of the shoot, soil + roots, and rhizosphere of fumigated and
untreated soil. Their results show that the rhizosphere of non-fumigated
soil had increased respiration (twice) compared to the soil + root of the

same treatment.

119

Allocation to Soil

The average specific activity of the M soil was at least twice that of
the NM soil, and this difference was measured across all of the harvest
periods (Tables 4 and 5). This result, combined with the observed higher
accumulation of radiolabel in the M soil atmosphere, demonstrate a
higher allocation of fixed 1‘C to the M soil relative to the NM. Other
studies show contrasting evidence regarding increases in soil C allocation
associated with infection. Snellgrove (1982) found a marginal difference,
while Jakobsen and Rosendahl (1990) found that the allocation of C to
the extraradical phase in mycorrhizal cucumber was twice that of non-
mycorrhizal plants and represented 3.1% of the total C fixation by the
host. Paul and Kucey (1981) observed that extraradical mycorrhizal
hyphae accounted for one percent of the fixed C in the Vicia-Glomus
symbiosis, suggesting that the C allocation to fungal structures may
explain in part the increased C allocation to soil. In this study, the
specific activity of the M soil varies throughout the harvest period,
although there is a net decrease since the soil specific activity at day 76
is lower than that measured at the first harvest (Table 5). However, this
decrease is not statistically significant (Table 4), suggesting an overall C
mineralization in the M and NM soils for the time frame of the chase
period. The M soil shows a net decrease in total label content during the
first 12 days of the chase period (from 0.4 to 0.2 MBq), followed by a net

increase in labeled C during the last 12 days of the chase period (from

120

0.2 to 0.3 MBq) (Table 3). Ihypothesize that these fluctuations may be
explained by the mineralization of root exudates during the first half of
the chase period followed by the turnover of mycorrhizal tissues from

day 12 to 24 after the pulse.

Concentration of Lipids in Mycorrhizal and Non-Mycorrhizal Roots

The roots of the M plants had significantly higher concentrations
of total lipid, total fatty acids, and fatty acid 16: 1015 when compared
to the controls (Table 6). The increased accumulation of lipids in
mycorrhizae has been documented by previous studies (see chapter 2 of
this thesis). For example, the roots of Allium, clover, ryegrass, and Citrus
have a higher trygliceride concentration when infected by Glomus than
when the roots are non-mycorrhizal (Cox and Sanders, 1974; Cooper and
Lose], 1978; Nagy et al., 1980), and the concentration of fatty acids in
mycorrhizal roots may be twice that of uninfected roots (Peng et al.,
1993). Several studies show that a significant fraction of the lipids of the
AMF genus Glomus may be comprised of the unusual fatty acid 16:1co5
(Graham et al., 1995), and this molecule has been proposed as the
principal AMF storage molecule in intra-radical vesicles (Pacovsky and
Fuller, 1988). This effect in the concentration of root lipids indicates that
the arbuscular fungus may be favoring the accumulation of C in the root
system. As a result, the infection with an arbuscular fungus has an

associated increase in the cost of root production (Eissenstat et al.,

121

1993). The effect of arbuscular mycorrhizae in the lipid composition of
roots has been demonstrated for several fungus-plant combinations and
different growth conditions, suggesting that the increased storage of C in
roots is a widespread pattern in mycorrhizal symbioses. In spite of this,
the growth efficiency of the AMF has been shown to be relatively low
(Harris and Paul, 1987). My results suggests that the net effect of the
AMF is to increase the C storage in root lipids, despite the high losses of C
associated with the metabolism of the mycorrhizal fungus.

The average concentration of the total lipid, total fatty acids, and
the molecular marker 16: 1 (05 remained in a relatively steady state
throughout the 24 day chase period (Table 6). Other studies have shown
that the concentration of 16:1 (05 fatty acid in roots increases with the
formation of the storage structures of the fungus. For instance, Graham
and Hodge (1993) observed a rapid accumulation of 16:1 (05 in roots
after 65 days in the Glomus -Citrus association, when the development
of storage vesicles takes place. My results suggest that during the chase
period, G. clarum was not concentrating stored C in fatty acids to a
significant extent. However, the total amount of lipids in the roots
increased because of the growth in root biomass (Table 2).

Throughout the harvest period, fatty acid 16:1 (05 accounts for up
to 29.5 % of the total fatty acid content of the M roots (Table 6).
Previous studies of the Sorghum-Glomus symbiosis show that molar

percentages of 13.3 % correspond to microscopical estimates of root

122

colonization of 44 % (see chapter 2 of this thesis). Assuming an overall
correlation of fungal infection with the relative amounts of 16:1 015 in
roots, Iestimate that the percent infection of the M roots in this
experiment exceeds 44 % root colonization. This is a reasonably high
amount of mycorrhizal infection in the M roots. Arbuscular mycorrhizal
fungi cannot be grown in pure culture, therefore, it is impossible to get
an accurate ratio of 16:1 035 to fungal biomass that could be applied to

the lipid results to calculate fungal weight.

Radiolabel and Fine Root Lipids

The radiolabel content of the total lipid, total fatty acids,
monoenoic fatty acid, saturated fatty acids, and 16:1 015 fatty acid is
higher in the M relative to the NM fine roots (Tables 7 and 8). The
difference is pronounced during the first 12 days of the harvest period as
indicated by the significant interaction between harvests and
mycorrhizal treatment (Table 8). In general, the total extractable lipid,
total fatty acid, and 16:1 w5 fatty acids of the M roots followed a similar
pattern of high radiolabel incorporation before the first harvest, followed
by a decrease in the label content throughout the chase period (Table 8).
Ishowed in chapter 2 that mycorrhizae formation in Sorghum has an
effects the qualitative and quantitative composition of total fatty acids,
phospholipid fatty acids, and sterols. This experiment further

demonstrates that the amount of C incorporation and loss from the

123

lipids of the roots of Sorghum are affected by mycorrhizal infection. The
incorporation pattern of M roots is different from the 1"C incorporation
and turnover pattern of NM roots. The results for the first harvest show
that the total lipids and the different fatty acid classes of NM roots
incorporated 30 % or less of the label incorporated by the corresponding
fractions in the M treatment (Table 6). Furthermore, the turnover of
radiolabel in NM lipids was slight or absent during the chase period,
indicating that NM root lipids were acting as relatively long term C
storage or structural compounds with limited net oxidation or
inineralization.

The saturated fatty acids are characterized by a lack double bonds
in the C chain. The most abundant saturated fatty acid in the M and
NM roots in palmitic acid 16:0. Table 6 shows that the average
concentration of 16:0 in M roots is higher than that of NM throughout
all the harvest periods, suggesting that part of the increase in 16:0 is due
to the presence of this fatty acid in the fungal biomass. The
incorporation and turnover of radiolabel in the saturated fatty acids of
M roots is different from that of NM roots and follows a similar pattern
to the incorporation and turnover of 1‘C in the mycorrhizal fatty acid
16:1 015 (Table 7). Ihypothesize that this pattern is explained by the
fact that a portion of the 16:0 in M roots is of fungal origin, which
explains the similarity in the C cycle between 16:1 and saturated fatty

acids.

124

Polyenoic fatty acids are characterized by having 2 or more
unsaturations in the carbon chain. In the M and NMroots of Sorghum.
the predominant polyunsaturated fatty acid is linoleic acid 18:2. The
incorporation and turnover of “C in polyenoic fatty acids follows a
relatively similar pattern in M and NM roots (Table 7). This contrasts
with the data from the saturated fatty acids, total fatty acids, and 16:1
(05 where M lipids incorporated more label before the first day of the
harvest period and had a marked turnover of radiolabel that was not
observed in the NM treatment. One major difference between polyenoic
fatty acids and the rest of the fatty acid classes is that the relatively high
incorporation of radio tracer between days 0 and 1 of the chase period is
not observed in the M polyenoic fatty acids. Table 6 shows that the
mass of 18:2 fatty acid per gram of root is mostly lower or equal in M
relative to the NM roots, suggesting that the mycorrhizal fungus is not
contributing significantly to the concentration of this molecule. It is for
this reason that Ibelieve 18:2 fatty acid to be mostly a component of
plant biomass rather than of fungal biomass. If we accept the
hypothesis that polyenoic fatty acids present in the roots of Sorghum are
mostly of plant origin regardless of mycorrhizal infection, this may
explain the similar C cycle of 18:2 fatty acid of M and NM roots.

The total lipids of M toots incorporated 97.1 KBq per plant before
day 1 of the chase period, whereas total fatty acids incorporated or total

of 38.2 KBq (Table 7). With these values Iestimated that the total fatty

125

acids of M roots account for 39% of the radiolabel incorporated by total
lipids between days 0 and 1 of the chase period. This implies that other
non-fatty acid, lipid-soluble molecules are also responsible for the
relatively high incorporation of radiolabel of M root lipids. The results
from Chapter 2 support this hypothesis, since we showed that the
concentration non-fatty acid lipids in the roots of Sorghum, such as
sterols may be affected by mycorrhizal infection. In addition, other lipid
soluble molecules such as waxes and carotenoids can potentially be

involved in the increased l‘C incorporation of M roots.

Turnover of the Root Fatty Acid 16:1 (05

The ten fold drop in the 1“C content of 16:1 (05 shows that the C
pool in the fatty acid 16:1 (05 is dynamic and suggests that the
mycorrhizal lipid C is either oxidized or converted to non-lipid
compounds within the time frame of the chase period of 24 days.
Despite the observed transience of C in the mycorrhizal lipids, the
increased mass of root fatty acids in the M treatment indicates that fatty
acids serve as compartments for net C immobilization in mycorrhizal
roots.

The equation of the solid line in Figure 3 is: y=4.3-0.00475*(x).
The slope of the solid line represents the rate constant k. Using these
values, the rate constant for the turnover of C in 16:1 «15 was determined

to be -0.00475 h" (Figure 3), which gives a turnover time of 210 hours.

126

The turnover time of fatty acid 16:1 (05 is within the turnover time
frame of the fungal arbuscles reported by Cox and Tinker, (1976). They
calculated that fungal arbuscles have a life span of 4 to 15 days within
the root cells, after which time they are re-absorbed by the plant cell.
My results verify the relatively fast turnover of mycorrhizal structures
relative to uninfected roots. Such short turnover is probably a
prerequisite to an efficient symbiotic system, since a long-term
permanence of fungal material in the roots may indicate that the fungus
is acting as a parasite.

We observed a relatively fast incorporation of labeled C and a
subsequent slower turnover rate during the chase period (Figure 3).
Pulse-chase studies of detrital microflora have shown a similar pattern of
C incorporation and turnover of microbial lipids. King et al., (1977)
observed that labeled C is rapidly incorporated into lipids and attain a
maximum specific activity within 20 hours, while the turnover of the
label occurs over a much longer span of time that may last several days.
This pattern of high incorporation rate followed by a slow turnover rate
in the mycorrhizal fatty acid 16:1 035 provides insight into the C
physiology of the mycorrhizal fatty acids. In systems that are in the
steady state, the mass of the substance of interest remains constant, and
the turnover rate must equal the incorporation rate (Robertson, 1957).

The incorporation and turnover of C tracer in fatty acid 16:1 015 does

127

not follow that of a single C pool in the steady state. This may be
explained several ways:

a) Increase in the C pool in root fatty acid 16:1 c05- Table 6 shows
that the concentration of fatty acid 16:1 (115 in the roots remains
constant throughout the chase period. However, the total pool of C in
root fatty acid 16:1 o5 increases throughout the experiment because of
the increase in root biomass.

b) Recycling of the tracer— Recycling of tracer may occur between
lipid and non-lipid pools. In fungi, carbohydrates may be converted to
fat (Weete, 1980). Actetyl-CoA is the main precursor for fatty acid
synthesis, and sugars are converted to Acetyl-CoA when subject to
glycolysis followed by decarboxylation. The reverse process of conversion
of fat to carbohydrate is also true (Weete, 1980). Triacylglycerides are an
important source of energy and C skeletons in fungi. The
triacylglycerides can be degraded to it’s component fatty acids by
enzymatic action. The C in fatty acids such as 16:1 (05 can then be
transferred to sugar molecules by the process of B oxidation followed by
the glyoxylate pathway and reverse glycolysis. Besides the possibility of
C recycling between fungal lipids and non-lipid molecules, there is also a
likelihood that some recycling also occurs between the fungus and the
host plant. There are contrasting observations regarding the role of
lipids in the C economy of the host-fungus relationship. It has been

hypothesized that the storage of C in fungal-specific lipids creates a

128

metabolic sink that favors the net movement of C to the fungal
symbiont (Finlay and Soderstrom, 1992). However, others have reported
the movement of fat globules from the fungus to the host’s cell,
suggesting that there may be a limited movement of C from fungus to
host (Mosse, 1973).

c) Multiple pools of fatty acid 16:1 oS- Fatty acid 16:1 (115 may
exist free in the cytoplasm or be bound to different lipid classes such as
triacylglycerides, diacylglycerides, or phospholipids. Triacylglyceride
serves as long-term storage of C, while phospholipids serve a structural
function by being part of cellular membranes. Ihypothesize that 16:1
(05 obtained from whole-cell extracts is in effect obtained from a mixture
of different C pools, with the possibility that each pool has a C cycle of
different time span.

In this study, Iisolated the extraradical spores of the arbuscular
fungus to measure their concentration in the M soil, as well as their 1“C
turnover. The spores of arbuscular fungi are rich in C storage
compounds (Beilby, 1980), and should be a good index of the C storage
in the fungal biomass outside of the roots. The possibility of physically
separating the spore biomass from the soil should allow for the direct
measurement of the incorporation and turnover of C in the fungal
biomass without having to recur to the analysis of fungal-specific
molecules. However, in this experiment the sporulation of the external

phase of the fungus was low, with concentrations of less than 1 spore per

129

gram of soil recorded for all the harvest periods. Because of this, the
mycorrhizal spores accounted for less than 0.8 % of all the soil radiolabel
content, and I was not able to detect a distinct pattern of 1“C
incorporation and turnover in the external phase of the mycorrhizal
fungus. This information is valuable, however, because it indicates that
the increased 1‘C content of the M soil relative to the NM soil cannot be
explained by fungal sporulation. Other factors such as fungal hyphae,
root exudation, and C immobilization in soil saprophytes may explain

the response in radiolabel content of the M soil.

Conclusions

This study shows that the infection of Glomus clarum affects the C
assimilation and allocation pattern of Sorghum bicolor. The higher root
respiration and large movement of 1"C underground is a consequence of
mycorrhizal interactions. The allocation of 6.3 % of the photosynthate
to soil is higher than other results obtained in our laboratory that
tended to show values of l-3%. The value of 6.3 % of soil allocation
represents 18 % of the below-ground production. This is a large enough
amount of C to have an effect in the activity of rhizosphere microbiota.

To this date, there are no studies that have measured the C cycle
of the fungal symbiont separately from the C cycle of the host because
the fungus is an integral part of the mycorrhizal root. The separation of

the mycorrhizal fatty acids is useful from both a quantitative and

130

qualitative standpoint, since it allows for the first time the calculation of
the turnover of an important cytoplasmic component of the arbuscular
fungus. 16:1 (05 is not a long term structural molecule and does not
necessarily give a total turnover of the AMF biomass. However, fatty
acids account for a high percentage of the AMF cytoplasm, so the
turnover rate of the mycorrhizal fatty acids represents the C turnover an
important C pool of the fungus. In this experiment I measured the
incorporation followed by a sustained decrease in the content of 1"C of
the 16:1 (1)5 fatty acid during the 24 day chase period, which results in a
1"C turnover time in 16:1 (05 of 210 hours. This pattern is absent in the
non-mycorrhizal roots and represents an important difference in the C
cycle of the mycorrhizal fungus and the host. This technique opens the
possibility of further experiments to study environmental impacts on the

turnover of mycorrhizal components.

LIST OF REFERENCES

LIST OF REFERENCES

Beilby, J. P. 1980. Fatty acid and sterol composition of ungerminated
spores of the vesicular-arbuscular mycorrhizal fungus Acaulospora
laevis. Lipids. 15, No. 11, 949-952.

Bligh, E. G., and Dyer, W. M. 1959. A rapid method of lipid
extraction and purification. Canadian Journal of Biochemistry
and Physiology. 35, 911-917.

Cahoon, E. B., and Ohlrogge, J. B. 1994. Apparent role of
phosphatidylcholine in the metabolism of petroselinic acid in
developing Umbelliferae endosperm. Plant Physiol. 104, 845-
855.

Cox, G., Sanders, F. B., Tinker, P. B., and Wild, J. A. 1975.
Ultrastructural evidence relating to host-endophyte transfer in
vesicular-arbuscular mycorrhizae. In Endomycorrhizas (F. E.
Sanders, B. Mosse and P. B. Tinker, Eds) pp. 149-174. Academic
Press, London.

Cox, G., and Tinker, P. B.. 1976. Translocation and transfer of
nutrients in vesicular-arbuscular mycorrhizae. 1. The arbuscle
and phosphorus transfer: a quantitative ultrastructural study.
New Phytol. 77, 371-378.

Cooper, K., and Losel, D. 1978. Composition of lipids in roots of onion,
clover, and ryegrass infected with Glomus mosseae. New
Phytol. 80, 143-151.

Daniels, B. A., and Skipper, H. D. 1982. Methods for the recovery and
quantitative estimation of propagules from soil. In “Methods and
Principles of Mycorrhizal Research”. N. C. Schenck, ed. pp. 29-35.
American Phytopathological Society Press, St. Paul, MN.

131

132

Eissenstat, D. M., Graham, J. H., Williams, K., and Peng, S. 1993.
Carbon cost of mycorrhizal citrus roots. Ninth North American
Conference on Mycorrhizae. University of Guelph, Guelph,
Ontario, Canada.

Finlay, R. and deerstrom, B. 1992. Mycorrhizae and carbon flow to
soil. In Mycorrhizal Functioning. Ed. Allen, M. pp. 134-170.
Chapman & Hall, New York.

Giovanetti, B., and Mosse, B. 1980. An evaluation of the techniques
for measuring Vesicular-Arbuscular mycorrhizal infection in
roots. New Phytol. 84: 489-500.

Graham, J. H., and Hodge, N. C. 1993. Evaluation of carbon cost of
citrus mycorrhizae by fatty acid analysis. Ninth North
American Conference on Mycorrhizae. University of Guelph,
Guelph, Ontario, Canada.

Hanis, D., Pacovsky, R., and Paul, E. 1985. Carbon economy of
soybean-Rhizobium-Glomus associations. New Phytol.‘ 101, 427-
440.

Harris, D., and Paul, E. A. 1987. Carbon requirements of vesicular—
arbuscular mycorrhizae. pp. 93—106. In Ecophysiology of VA-
mycorrhizal plants. G. R. Safir, (ed). CRC Press, Boca Raton,
Florida.

Ho, 1., and Trappe, J. 1973. Translocation of 14C from Festuca plants
to their endomycorrhizal fungi. Nature, The New Biology. 244,
30-31.

Horwath, W. 1993. The dynamics of carbon, nitrogen, and soil
organic matter in Populus plantations. Ph. D. thesis, Michigan
State University.

133

Hylemon, P. B., Wells, J. S., Krieg, P. B., and Hannasch, H. W. 1973.
The genus Spirillum: A taxonomic study. International Journal
of Systematic Bacteriology. 23(4) pp. 340-380.

Jakobsen, I., and Rosendahl, L. 1990. Carbon flow into soil and
external hyphae from roots of mycorrhizal cucumber plants.
New Phytol. 115, 77-83.

King, J. D., White, D. C., and Taylor, C. T. 1977. Use of lipid
composition and metabolism to examine structure and activity
of estuarine detrital microflora. App. Environ. Microbiol. 33,
1177-1183.

Kucey, R., and Paul, E. 1982. Carbon flow, photosynthesis, and N2

fixation in mycorrhizal and nodulated faba beans (Vicia faba
L.). Soil Biol. Biochem. 14, 407-412.

Martin, J. and Kemp, J. 1986. The measurement of C transfers within
the rhizosphere of wheat grown in field plots. Soil Biology and
Biochemistry. 18:1, 103-107.

Morris, L. J. 1966. Separation of lipids by silver ion chromatography. J.
Lipid Res. 7, 717-732.

Morrison, W. R., and Smith L. M. 1964. Preparation of fatty acid
methyl esters and dimethyl acetals from lipids with boron
fluoride-methanol. J. Lipid Res. 5, 600-607.

Mosse, B. 1973. Advances in the study of vesicular-arbuscular
mycorrhizae. Annual Review of Plant Pathology. 11, 171-196.

Murphy, J., and Riley, J. P. 1962. A modified single solution method
for the determination of phosphate in natural waters. Anal.
Chim. Acta. 27, 31-36.

134

Nagy, S., Nordby, H. E., and Nemec, S. 1980. Composition of lipids in
roots of six citrus cultivars infected with the vesicular-arbuscular
mycorrhizal fungus Glomus mosseae. New Phytol. 85, 377-384.

Nordby, H. B., Nemec, S., and Nagy, S. 1981. Fatty acids and sterols
associated with citrus root mycorrhizae. J. Agric. Food Chem. 9,
(2), 396-401.

Olsson, P.A., Baath, E., Jakobsen, 1., and deerstrom, B. 1995. The use
of phospholipid and neutral fatty acids to estimate biomass of
arbuscular mycorrhizal fungi in soil. Mycol. Res. 99, 623-629.

Pang, P., and Paul, E. 1980. Effects of vesicular-arbuscular mycorrhiza on

14C and 15N distribution in nodulated faba beans. Canadian
Journal of Soil Science. 60, 241-250.

Paul, B., and Clark, F. E. 1989. Mycorrhizal Relationships. In Soil
Biology and Biochemistry. Academic Press, New York.

Paul, B., and Kucey R. 1981. Carbon Flow in plant-microbial
associations. Science. 213, 473-474

Pacovsky, R. S., and Fuller, G. 1989. Mineral and lipid composition of
Glycine-Glomus-Bradyrhizobium symbioses. Physiologia
Plantarum. 72, 733-746.

Peng, S., Eissenstat, D. M., Graham, K. W., and Hodge, N. C. 1993.
Growth depression in mycorrhizal citrus at high phosphorus
supply. Analysis of carbon costs. Plant Physiol. 101, 1063-1071.

Schultz, D. J., Cahoon, E. B., Shanklin, J., Craig, R., Cox-Foster, D. L.,
Mumma, R., and Medford, J. Expression of a A 9 14:0-acyl
carrier protein fatty acid desaturase gene is necessary for the
production of (05 anacardic acids found in pest-resistant

geranium (Pealrgonium xhortum). Proc. Natl. Academy Sci. USA.
93, 18771-8775.

135

Robertson, J. S. 1957. Theory and use of tracers in determining
transfer rates in biological systems. Physiological Reviews. 37,
133-154.

Snellgrove, R., Splittstoesser, W., Stribley, D., and Tinker, P. 1982. The
distribution of carbon and the demand of the fungal symbiont in
leek plants infected with vesicular-arbuscular mycorrhizas. New
Phytol. 92, 75-87.

Stribley, D. P., and Read, D. J. 1974. The biology of mycorrhizae in
the Ericaceae. III. Movement of 14C from host to fungus. New
Phytol. 73, 731-741.

Tinker, P., Durall, D. M., and Jones, M. D. Carbon use efficiency in
mycorrhizas: theory and sample calculations. New Phytol. 128,
115-122.

Warembourg, F. R., and Paul, E. A. 1973. The use of C1402 canopy

techniques for measuring carbon transfer through the plant-soil
system. Plant and Soil. 38, 331-345.

Weete, J. D. 1980. Fungal lipids. In Lipid Biochemistry of Fungi and
Other Organisms. Weete, J. D. (ed). Plenum Press, New York.

136

.0030 3.

0200000 00050000: :0 0000000 0 00:. 0:930:00 30 50 N0 000. 002000 00300.

 

 

 

 

 

34000803” 70 270

3003 «a .0303 “Va

730 0:000:00: 7000 03000030:
0:03 00 3.8 0.90 3.8 0.3 3.: 00.0 3.8
Woo” 3.0 3.8 00.0 3.8 3.0 3.: ~90 3.8
00: 0.0 3.00 0.0 3.8 0.3 3.00 0.0 3.8
000900330 000000000: ... 0.0 0.8 3.0 3.00 0.0 0.8 30.0 3.8
0002 030000303... 0.0 0.1.0 0.0 3.8 0.0 01.0 0.3 3.8
.0900 000 0:830:0— 00 3.8 0.0 3.8

 

.000 208008 000 00 00.00000 30 90 020 003.008.

1.- 0039:0000 00:0 0.0 000903. em 00000000 :0 002200: 0 0:0 00 05:3 0200 90 00300 30003.
00:8 00 90 0000 0020.2 2000 0080000 mom 008,203.50 400000000?
0052-08.50 000000000: 0». 0.00 2:00. 2:00 23 00:8 :00 0: 0<0n000 on 0.00 7000. .000 038 000

.000 000:000 03:. cm 0.0 800: 00 0: 0000:9005.
0:; 90
30.82300 030:8 r00 0: 0<0~000

0000800 00:00 000 90 30700005 on 0: 90 0020.3 001000.

.3..- .200 0:02 000000000: 200 :2 000008500 0000000000. 000 000: 50700000. .0: 0000:3000 90 .052
000000000? 3 0.30800 0.00 90 «000000000 £300 00 :9 0009580 03 3. 30 .052. noon. mom. 0:0 00302.
03.50 000000000: 00 0.0 0002 000000000? 09:00:00 0308000: 00 900 0:0 0502 000000000: €00 0:033:
000 0: 3 0:0 23 0007000000. .35.. £00.. 3 0008000 0 301808 <0E0 cm 0.0 300 em 0009 :0 000000000:
0.00 00 0:032: 09. 0: 00:8.

137

.0020 N. 000:" 32:03 Amv 030a 30 050000000.

 

U00: 0m 00 00 00 00

0:00"
308000 0.0 3.8 0.0 3.: 0.0 3.3 0.0 3.8 0.0 3.8 0.0 3.3 0.0 3.8 0.0 3.00 3.0 3003.0 3.8

00:0 82
308000 0.0 8.8 0.0 8.3 0.0 8.00 0.0 3.00 0.0 68 Pm 8.3 0.0 8.00 0.0 3.8 .00 8.: 0.0 3.3

.0300
~00"
33:03 0.0 8.00 0.0 3.8 0.0 8.00 0.0 3.00 0.0 8.8 «N 3.3 0.0 3.3 Mg 3.8 00.0 3.88.0 8.3

.000 <0E0m 0~0 0.0 0<0~0m0 $25000 033 o». 0.0 800:0 cm 0000 000:". :u A 000:" 00a 20050:"
8800:0000? Zn 500000000“ ZZH :o:-0:00:_0fi00. .000 000:8 £08 00 00v~ 000 0" 90 080 a». 0.0 00000
0000050 8 :00“.

00000 0. .0000— 0000000000 00:00:” 9000 000 000:00 00 H00 000:" 000000 0:0 00:.

 

 

138

 

 

 

 

 

 

 

U003 0N 00 00 00 00
0000"
2— PM 3.8 0.0 3.8 PM 3.8 0.0 3.8 N; 3.8
23 P0 3.3 PM 3.: N0 3.: 0.0 3.8 NA 3.8
€020 0000
K 0.0 3.00 0.0 3.8 0.0 3.3 09 3.8 0b 3.8
23 0.0 3.8 0.0 3.8 0.0 3.8 0.0 3.3 0.0 3.8
00:0 ~00:
3 0.0 3.00 0.0 3.8 0.0 3.8 0.0 3.8 0.0 3.8
200 00 3.8 0.0 3.8 0.0 3.8 0.0 3.8 0.0 3.:
00:”
E 0.0 3.8 0.0 3.: 0b 3.: 0.0 3.: 0.0 3.8
22— 0.~ 3.8 0.0 3.00 0.0 3.00 0.0 3.8 0.0 3.00

 

.000 <0E0m 00.0 000 0<0~0m0 030300 0:00 00 ~00 800:0 00 0000 30700000. :n 0 000:» 000 3009803
00803000? Zn 30000080“ 20$" :0:-0:00:_0H00. .000 000:8 £000 00 000~ 000 0, H00 00:0 00 "00 0050.
00000000 8 :00». :0" :0" 0083030.

139

Table 4. Results of the ANOVA for the plant biomass, soil, and
respiration variables.

 

 

Harvest x
Harvest Mycorrhizae Mycorrhizae

Variable:

Shoot biomass (g) + + -
Shoot specific activity + + _
(MBq/gC)

Shoot total radiolabel (MBq) - - -
Whole root biomass (g) + - _
Fine root biomass (g) + - -
Root specific activity (MBq/gC) - + _
Root total radiolabel (MBq) - + -
Soil specific activity (KBq/gC) - + -
Soil total radiolabel (MBq) - + -
Below-ground respiration * + *

radiolabel (KBq)

“-” indicates p > 0.05; “+” indicates p s 0.05.
*- The below-ground respiration was only mesured for the individuals of the
last harvest.

00000 0. 000050 000300 9000300 0». 000 000:. 000000 0:0 00:.

 

 

140

 

 

 

 

 

 

 

00$ 00 00 00 00 00
0000:
E 0.0 3.8 0.0 3.8 0.0 3.8 0.0 3.00 0.0 3.00
22— 0.m 3.: 0.0 3.8 0.0 3.8 ., i 0.0 3.: 0.0 3.8
$0000 0002
E 0.0 3.8 0.0 3.8 0.0 3.8 0.0 3.8 0.0 3.8
20(— 0.0 3.8 0.0 3.00 0.0 3.: 0.0 3.8 0.0 3.:
0.80 0003
Z 0.0 3.8 0.0 3.00 0.0 3.8 0.0 3.8 0.0 3.:
270 0.0 3.00 0.0 3.00 0.0 3.8 0.0 3.8 0.0 3.00
00:”
Z 0.05 3.0000 0.000 3.003 0.000 3.008 0.0: 3.003 0.0003003
22 0.000 3.003 0.000 3.0000 0.000 3.008 0.0003003 0.000 3.008

 

.000 50:00 00.0 000 0<0~0m0 0000:0000 02.00 00 ~00 800:0 00 0000 80700000. :n 0 000:" 000 20080:"
00808000? Zn 8000—0000“ ZZH :0:-800:_0000. 000 000:3 £000 00 000~ 000 00 H00 080 00 ~00
08000000050 8 :00”. :0" :00 000008800.

141

 

.0030 0. :05 0000030000000 00. 200 00000 000.038.
U00 00 00 0.0 00 .00

K. E K 2.3 K., BK. K E K E
0.0:: :03 00.20.08 00.03.08 00 00 00 00 00.02.08 0003.8 00.20.08 0003.08
0.0:: 0.0000
003 0.02.8 0.03.3 0.03.8 0.03.08 0.02.8 0.03.08 0.02.08 0.03.08 0.02.3 0.03.8
0005. 00000 0008 2000000: 0000008”
00:80 0.03.08 0 0.03.08 0 0.03.8 0.03.08 0.03.08 0 0.03.8 0
000 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.3 0.03.8 0.03.8 0.03.8
3; 0 0.03.08 0.03.8 0 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8
000 90 0.03.08 0.03.8 0.03.08 0.03.8 0.03.08 0.03.8 0.03.8 0.03.8 0.03.3 0.03.8
00; e0 0.03.08 0.03.08 0.03.00 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.08
00; _. 0.03.8 0 0.03.08 0 0.03.00 0 0.03.8 0 0.03.8 0
00; 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8
0000 000 0.03.8 0.03.8 0.03.08 0.03.8 0.03.8 0.03.08 0.03.8 0.03.8 0.03.8 0.03.8
070 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8 0.03.8

 

 

 

 

 

 

 

H00 <0_=00 000 50 0<000m0 90000000 00000 00 90 50008 00. 0000 E03.

00000—0000“ 23" 000-0000E0000. 0000. 50 :33 £50: 8000 00 00 8000 0.00 503 000 00005.
0" 00000 AA 0.00 8038.

0.0 00 E0 050 00. E0 0050 00000000 00 1000.

00H :00 0000000000.

.7 .050 00 00 000000200 8000000 00 m0<000_ 00; 0005000.

an 0 0.000 000 000008000 008300000. Zn
.30 E0000 $000 0— 00v~

142

Table 7. Total radiolabel in the fine root biomass accounted for by
each lipid type (KBq/plant).

 

Day: 52 64 76

Total lipid-extractable material
M 97.1 (20.0) 60.2 (9.0) 34.4 (2.3)

NM 28.1 (9.5) 32.2 (1.7) 28.4 (5.8)

Total fatty acids
M 38.2 (12.2) 14.9 (4.3) 10.9 (4.6)

NM 11.2 (4.3) 12.3 (2.1) 6.8 (4.2)

Polyenoic fatty acids
M 1.8 (0.7) 4.1 (0.5) 2.6 (1.2)

NM 2.1 (1.0) 2.7 (0.5) 2.9 (0.6)

16:1 035 fatty acid
M 13.9 (4.0) 2.5 (0.5) 1.7 (0.7)

NM 0.9 (0.4) 0.3 (0.2) 0.4 (0.2)

Saturated fatty acids
M 18.9 (5.5) 6.3 (1.4) 4.4 (2.3)

NM 4.4 (2.0) 4.6 (1.0) 3.4 (1.4)

The values are the average (standard error of the mean) of each plant.
11: 4 plant per treatment combination. M: inoculated; NM: non-
inoculated. The plants were 51 day old at the time of the exposure to
1“C02.

143

Table 8. Results of the ANOVA for the lipid concentration and radiolabel
content of fine roots.

 

 

Harvest x
Harvest Mycorrhizae Mycorrhizae

Variable:
Total lipid (mg/g) - + _
Total fatty acids (mg/g) - + -
16:1 (05 (mg/g) - + -
Total lipid (KBq/plant) - + +
Total fatty acid + + -
(KBq/plant)
16:1 fatty acid + + +
(KBq/plant)

 

“-” indicates p > 0.05; “+” indicates p s 0.05.

144

 

 

 

 

saturated
fatty acids

3
04'0'“,

2v

it

18:1 (07
16:1 05

18:1 09 h: '

16:1 «:7

fatty acids up '

 

 

 

 

Argentation TLC plate after staining

with dichlorofluorescein and
visualization under UV light,

 

Radiograph of the same plate as
the image on the left

followed by marking of the fatty

acid bands with pencil.

Figure 1. Argentation chromatography of the fatty acid extracts from M

and NM Sorghum roots.

145

 

 

 

 

5.5-
M M

4.4” [i] EINM
i

0* €72

m -

2.2 Ii] 0 315 Q
1.1- '1] m [I]

 

 

10 2 3O 4O 50 6O
time(h)

Figure 2. 14C trapped from the below-ground atmosphere.

146

 

 

A 5—
E
8' 4‘
fr?
8
Ed 3"
.S
s.) 2"
"a
0 1‘
5
O 1 12 24
Days after pulse labelling

Figure 3. Ln(At) versus time for the 1“C turnover in the 16:1 05
fatty acids from the roots of Sorghum infected with G. Clamm.

OVERALL CONCLUSIONS

About the Lipid Composition of Mycorrhizae

The principal aim of this work was to characterize the species of
fungus present in mycorrhizal roots by lipid analysis and to measure the
C turnover of mycorrhizal lipids. Ihave shown that mycorrhizal
infection has an impact in the kinds and amounts of lipids in the roots
of Sorghum. Chapter 2, identifies several lipids such as campesterol and
16:1 fatty acids which are useful for discerning between mycorrhizal and
non-mycorrhizal roots. Multivariate analysis of the lipid profiles was
shown to be a valuable approach for the identification of the fungal
species engaged in the symbiosis. In chapter one, Ifurther validated the
lipid markers of mycorrhizal fungi by demonstrating that campesterol
and 16:1 fatty acids are present only in small amounts in the biomass of
the non-mycorrhizal fungi that can potentially share the habitat with
the AMF.

There are limitations to the use of lipid analysis for the
identification of mycorrhizal fungi in roots. Ihave shown that 16:1 05
FAME is not detectable in uninfected roots and for this reason its

presence reliably indicates mycorrhizal roots. However, other fungal

147

148

lipid markers like 16:1 05 PLFA,16:1 m7 PLFA, and campesterol are
present in detectable amounts in non-mycorrhizal roots. For this reason,
it is necessary to obtain information about the lipid composition of
infected and uninfected roots to make a correct assessment of infection
when using this set of lipid markers. This technology should be useful in
ecological studies in greenhouse or field conditions where we know the
host species, plant age, as well as the dominant species of arbuscular
fungi in the soil community. This way, the lipid markers and lipid
profiles of samples can be compared to a known data base of infected
and uninfected roots, so that a putative identification of the fungal
species present in the roots can be carried out. An additional advantage
of the use of lipids for the analysis of arbuscular fungi is that we not
only have the potential to identify the species of fungus present in“ the
root, but there is the potential to estimate the amount of biomass of the
fungus per unit mass of root. This possible because if we make the
assumption of a constant ratio of lipid to fungal biomass, the mass of
the fungus in the root can be directly calculated from the lipid data.
For instance, Jabaji Hare (1988) has measured the ratio of phospholipid
to biomass in the vesicles of arbuscular fungi. Using this information,
Olsson (1995) was able to calculate the biomass of Glomus in the growth
medium of mycorrhizal plants by measuring the relative amount of 16:1

05 in the phospholipid extracts of the samples. Using the same

149

approach, we can calculate the biomass of Glomus clarum in the roots of
Sorghum. The assumptions are that 13.1 % of the dry weight of the
fungus is FAME fatty acid, and that approximately 50 % of the FAME is
16:1 05 (Jabaji Hare, 1988). Using the data for the fatty acid
concentration from chapter 3, we can estimate that the mass of fungus
per unit mass of root ranges from 23.0 to 27.5 mg/g. Nonetheless, it is
important to remark that this type of estimate is speculative, since
assumptions have to be adopted regarding a constant lipid to biomass
ratio in different fungal cell types, fungal species, fungal age, and
environmental conditions.

The results of my work show some interesting patterns with
possible implications in the functionality of the lipids AMF which are
worthy of further analysis. For example, it is widely known that
alterations in the kinds of fatty acids and sterols present in lipids may
have an impact on the melting point of lipids and the fluidity of
membranes. Because of this, we can hypothesize that organisms
enriched in unsaturated fatty acids (such as 16:1 05) are able to
maintain membrane function at lower temperatures than other
organisms richer in saturated fatty acids. Chapter 2 showed that the
genus Gigaspora forms mycorrhizae which have a reduced amount of
16:1 05 fatty acid when compared to the rest of the mycorrhizal species

included in the experiment. Relevant questions are: Does Gigaspora has

150

an adaptation to a different temperature regime than the rest of the
arbuscular fungi? Does the qualitative differences in the sterol
composition between mycorrhizal and non-mycorrhizal roots have
implications in the viscosity of cellular lipids, or do they serve as
precursors for the production of steroid hormones which affect plant and
fungal growth? These are important questions which should be

addressed in future work.

About the C Cycle of the Lipids of Arbuscular Fungi

In the third chapter of this thesis, I provide the turnover time of C
in the mycorrhizal fatty acid 16:1 05. As shown in Chapter 2, lipids can
be an important component of the biomass of the arbuscular fungus.
Because of this, the C turnover in the mycorrhizal fatty acids represents
the length of the C cycle of the most abundant C pool of the arbuscular
fungus. This experiment sets the precedent for future work to expand
our understanding of the C cycle of AMF. A variety of factors are known
to affect the C budget and the C cycling of plants and their associated
fungi. For example, physical conditions such as available oxygen,
temperature, and water status are known. to affect the respiration rate of
organisms, while biological factors such as growth habit, fruit set, and
plant age are associated with changes in the respiration rate and C

cycling in plant tissues. Further, we do not know if the different

151

arbuscular fungal species vary in their inherent C turnover times,
respiration rate and patterns of C storage and utilization. These are
important variables, since the C demand, accumulation and loss by the
fungus may be determinants of the mutualistic or parasitic nature of the
symbioses. Future studies should address how the C turnover of the
fungus is affected by various biological and physical conditions
mentioned above. This will be an important step to improve our

understanding of the complex economy of the mycorrhizal symbiosis.

LIST OF REFERENCES

LIST OF REFERENCES

Jabaji-Hare, S., Deschene, A., and Kendrick, B. 1984. Lipid content
and composition of vesicles of a vesicular-arbuscular
mycorrhizal fungus. Mycologia. 76(6), 1024-1030.

Olsson, P.A., Baath, B., Jakobsen, I., and deerstrém, B. 1995. The use
of phospholipid and neutral fatty acids to estimate biomass of
arbuscular mycorrhizal fungi in soil. Mycol. Res. 99, 623-629.

152

APPENDIX A

APPENDIX A

FAME, PLFA, AND STEROL ANALYSIS OF ROOTS AND FUNGAL

MATERIAL

Sample

/\

Bligh and Dyer extraction Analysis Of whole-cell
of lipids fatty acids (MIDI system)

i l

Fractionation by silicic acid FAME profile
column chromatography

/\

Neutrfi lipid Polar lipid
Saponification TransesteJification
Gas Chromatography Gas Chromatography

Mass Spectrometry Mass Spectrometry

Sterol profile PLFA profile

Flowchart representing the procedure used to analyze the FAME, sterol,
and PLFA profiles of the root and fungal material in chapters 1 and 2.

153

APPENDIX B

APPENDIX B

ROOT LIPID EXTRACTION AND RADIATION MEASUREMENTS

Root Sample
Bligh Dyer . Identification of fatty acids
Total hpid extraction by gas chromatography (MIDI).
Total lipid: ,
Specific activity Methylation
and mass 1
Whole-cell FAMEs
Physical separation of
FAMEs by
silver ion TLC
Radiography of total
and specific fatty acid bands

Flowchart representing the sampling and analysis scheme for the root
lipids in chapter 3.

154

"Illllllllllllllllllllf