Mycorrhizal fungi form extensive mycelia in soil and play significant roles in most soil ecosystems. The estimation of their biomasses is thus of importance in order to understand their possible role in soil nutrient processes. For arbuscular mycorrhizal (AM) fungi the signature fatty acid 16:1ω5 provides a new and promising tool for the estimation of AM fungal biomass in soil and roots. For ectomycorrhizal fungi 18:2ω6,9 dominates among the fatty acids and can be used as an indicator of mycelial biomass of these fungi in soil in experimental systems. In biomass estimation primarily the phospholipid fatty acids (PLFAs) are suitable. Through the use of specific PLFAs it is possible to study interactions between mycorrhizal mycelia and bacteria in soil as well as between AM fungal mycelia and mycelia of saprophytic and parasitic fungi in soil and in roots. AM fungi, in particular, store a large proportion of their energy as lipids and by using the signature fatty acids it is possible to determine the relation between membrane and storage lipids, which could be an indication of energy storage levels. Various aspects of how the fatty acid signatures can be used for studies related to questions of biomass distribution and nutritional status of mycorrhizal fungi are discussed.
The mycorrhizal mycelium
Mycorrhizal symbiosis differs from other forms of microorganism/plant symbiosis in that it involves a soil dwelling phase (the external mycelium), essential for symbiosis function through the uptake of nutrients [1,2]. Since the mycorrhizal fungi obtain carbon energy from their associated plants, they do not compete with soil saprophytes for organic carbon and can be considered a separate functional group. Therefore it is of importance to differentiate between and compare the biomass of mycorrhizal fungi and the biomass of saprophytic microorganisms. The vegetative stages of the fungi of both arbuscular mycorrhizae (AM) and ectomycorrhizae (the two dominant types of mycorrhiza) are difficult to identify microscopically and differ from those of other fungi in soil.
The biomass of AM fungi in soil and roots is usually quantified microscopically. In pot culture studies, AM fungal hyphae are often counted and differentiated visually and subjectively from the hyphae of other fungi (e.g. [3,4]), but AM fungal hyphae have only rarely been counted in field soil [5,6]. The hyphae of ectomycorrhizal fungi (mainly basidiomycetes) cannot be differentiated from other basidiomycete hyphae.
Quantification of fungal hyphae by microscopy includes both living and dead hyphae, unless specialized vital staining methods are applied [7–9]. Chitin is a fungal cell wall compound that has been used to estimate the biomass of both ectomycorrhizal  and AM (, H. Rouhier, personal communication) fungi. Chitin persists after the death of the fungus and its suitability in estimating living or recently dead hyphae must be questioned. Furthermore, it is also produced by both insects and arthropods sometimes leading to high background values in field soil. The membrane compound ergosterol can be used instead for estimation of the biomass of ectomycorrhizal fungi. Ergosterol has also been identified in AM fungi and has been proposed as a biomass indicator for AM fungi . However, the content is much lower than in other fungi, making it unsuitable for most systems (see [13,14]).
Biochemical marker compounds
Two important types of lipids found in organisms are phospholipids (membrane constituents) and neutral lipids (energy storage in eukaryotes). Both these types of lipids contain fatty acids connected to a glycerol backbone. Phospholipid fatty acids (PLFAs) have been used to describe the microbial community structure in soil [15,16] and have become valuable tools in the study of changes in the microbial community in soils caused by pollution and environmental management practices (see ). Many of the fatty acids can thus be used to analyze microbial community dynamics and the PLFAs are useful biomass indicators . Phospholipids are, however, correlated to membrane area and the surface to volume ratio may vary widely, e.g. between soil bacteria and fungi (see ). This means that the amount of phospholipids per unit biomass is not the same in these different organisms and this should be considered when calculating the biomass. The phosphate group of the phospholipids is easily and rapidly released through enzymatic actions in soil , and thus PLFAs reflect the occurrence of mainly living or recently dead organisms. Neutral lipid fatty acids (NLFAs) provide an indication of the energy store and the neutral lipid to phospholipid ratio in eukaryotes may reflect nutritional status .
Objectives of this review
To elucidate the possible impact of different functional groups in the soil, it is important to estimate their relative biomasses. The objectives of this review are to describe how specific fatty acids can be used to estimate the biomass of mycorrhizal fungi in soil and roots as well as to provide information on their interactions with other microorganisms. Furthermore, the use of fatty acids to estimate carbon allocation to energy stores within the mycelium will be discussed. Special attention will be paid to the sensitivity and specificity of different signature compounds.
Estimation of the biomass of the external AM fungal mycelium using PLFA 16:1ω5
Fatty acids in AM fungi and the level of specificity
The fatty acid composition of lipids in AM fungi differentiates them from other organisms. In most AM fungi, a large proportion of the total fatty acids is found as 16:1ω5 and 18:1ω7 [22–24], which is also indicated by the example shown in Table 1. These fatty acids are normally not found in other fungi , but are constituents in some bacterial genera (see ). Furthermore, AM fungi have a rather high content of polyunsaturated 20-carbon fatty acids [27,28]. Twenty-carbon fatty acids are also present in algae and protozoa but are rare in non-AM fungi and not present in significant amounts in bacteria [29,30]. Thus, there is no completely specific fatty acid for AM fungi, but the fact that these fungi contain some rather unusual fatty acids can be used as a detection criterion. There are, however, considerable differences between AM fungal species and genera in their fatty acid composition , which may cause problems for biomass estimation in mixed AM fungal communities.
|Fatty acid||Lipid fraction|
|PL (nmol mg−1)||NL (nmol mg−1)||PL (%)||NL (%)|
|Fatty acid||Lipid fraction|
|PL (nmol mg−1)||NL (nmol mg−1)||PL (%)||NL (%)|
Note the differences in the proportions of 16:1ω5, 18:1ω7, 20:4 and 20:5 between the two lipid classes.
Estimation of biomass in soil and roots using PLFA 16:1ω5
It has been demonstrated that it is possible to use PLFA 16:1ω5 to estimate the biomass of external AM mycelium in different soil systems [14,26]. Through the analysis of PLFA 16:1ω5 in soil and root samples it is possible to estimate the AM fungal biomass in soil and roots in a comparable way and thus to compare the biomass of internal and external mycelium (, P.A. Olsson and A. Johansen, unpublished). However, there will always be background amounts of PLFA 16:1ω5 which are not derived from this target group of organisms. The background level of PLFA 16:1ω5 in soil has been found to be between 30 and 60% in different studies [14,26,32]. The background of PLFA 16:1ω5 in soil originates mainly from bacteria (but it may also occur in some other fungi) and is a problem in the accurate biomass estimation for AM fungi. Bacterial biomass in soil is usually correlated with organic matter content and estimation may thus be particularly difficult in soils high in organic matter. Results so far indicate that with an organic matter content of 2% it is possible to detect AM fungal mycelium, using PLFA 16:1ω5, if 2–5 m hyphae per gram soil are present [26,31]. In experimental systems often over 10 m AM fungal hyphae per gram soil is obtained in soils with this amount of organic matter (see [4,26]). The background of PLFA 20:5 in non-mycorrhizal control soil (probably derived from saprophytic fungi) can be lower than for PLFA 16:1ω5 , but since 16:1ω5 is present in significant amounts in more AM fungal species and usually at higher amounts than 20:5 [27,28], it is considered to be a better biomass indicator of AM fungi in soil.
NLFA 16:1ω5 is more sensitive than PLFA 16:1ω5 as an indicator of AM mycelia in soil systems since the ratio between NLFA and PLFA 16:1ω5 is high in AM fungi (between 1 and 200), while it is low in bacteria (<1) (P.A. Olsson and E. Bååth, unpublished). Using NLFA 16:1ω5 it is possible to detect low densities of AM fungal hyphae and it was thus possible for Nadian et al.  to estimate hyphal growth using NLFA 16:1ω5, while the levels of PLFA 16:1ω5 varied and were hardly detectable in their system. However, PLFA 16:1ω5 is a better biomass indicator, when it can be used, since its concentration in the mycelium depends less on growth conditions and is more related to hyphal length than to sporulation .
Ensuring the correct biomass estimation of AM fungal mycelium
To ensure that the amount of PLFA 16:1ω5 gives a reasonable measure of AM fungal biomass the following recommendations can be adopted.
Non-mycorrhizal controls should be used. These may be obtained by growing plants under non-mycorrhizal conditions or by separating compartments from the mycorrhizal hyphae with a membrane. In sand dunes in the field it is possible to use mesh bags containing sand originally free of hyphae. These can be buried in sand near vegetation and the AM fungal growth can be measured by extracting hyphae from the bags (P. Wilhelmsson and P.A. Olsson, unpublished).
Data on bacterial biomass should be collected since changes in PLFA 16:1ω5 may be due to changes in the bacterial community. Different techniques (including PLFA analysis) can be used to ensure that no major changes in bacterial biomass have occurred in a particular experiment, see .
NLFA 16:1ω5 should be used to monitor neutral lipids of AM fungi. If an increase in PLFA 16:1ω5 is detected, with no accompanying increase in NLFA 16:1ω5, the change is probably not due to a change in the biomass of AM fungi. NLFA 16:1ω5 can thus be used as support for the biomass estimation using PLFA 16:1ω5, such support can also be gained from the 20-C fatty acids of AM fungi [26,28].
Estimation of energy storage levels in AM fungi
Neutral lipids of AM fungi detected using NLFA 16:1ω5
In AM fungi, energy is mainly stored in neutral lipids  (P.A. Olsson and A. Johansen, unpublished), triglycerides being the dominant type. Neutral lipids are found in large amounts in AM fungal storage structures: spores  and vesicles . The amount of neutral lipids correlates very well with the number of spores formed by a mycelium , but accumulation of neutral lipids may also occur in the hyphae before sporulation .
For the efficient extraction of lipids from mature AM fungal spores the spore walls must be broken (P.A. Olsson and A. Johansen, unpublished). A mortar and pestle can be used to grind mycelium, but for soil samples, special ball milling equipment, made of material harder than quartz, must be used. Freeze drying of samples prior to extraction may also increase the yield.
The neutral lipid to phospholipid ratio as an indication of carbon allocation to storage structures
The ratio between the concentrations of neutral lipids and phospholipids (NL/PL) in AM fungi can be estimated using PLFA 16:1ω5 and NLFA 16:1ω5. However, here it must be considered that the proportion of 16:1ω5 is not the same in neutral lipids and phospholipids (see Table 1). The ratio indicates the proportion of carbon allocated to storage structures in the AM fungal mycelium since the amount of neutral lipids is closely correlated to the number of spores formed (see ). The NL/PL ratio for the plant pathogenic fungus Aphanomyces has similarly been used to study the nutritional status of the fungus during infection (J. Larsen, personal communication).
There are indications that a high NL/PL ratio indicates a good carbon supply to the AM fungus from the plant. Phosphorus application to soil normally reduces mycorrhiza formation in plant (see e.g. ) and reduces the carbon allocation to root and mycorrhizal components [37,38]. In one experiment where the addition of phosphorus to the soil reduced root colonization and the amount of external hyphae equally, the lipid composition of the external mycelium was affected; the NLFA 16:1ω5/PLFA 16:1ω5 ratio decreased from around 60 to 20 following high phosphorus addition . Another indication of the NL/PL ratio in AM fungi as an indication of carbon flow from the plants was that the external mycelia growing from the dune grass Ammophila arenaria L. in southern Sweden showed a lower NL/PL ratio in October than in September (P. Wilhelmsson and P.A. Olsson, unpublished). October is characterized by less light and lower soil temperature compared with September, factors that may reduce the carbon flow from the plant.
Other examples which show that information can be gained from the NL/PL ratio in AM fungi are that the NL/PL ratio for the internal mycelium of Glomus intraradices differs in different plant species (K. H. Söderberg et al., unpublished) and that the external mycelium exhibits a reduced NL/PL ratio during rapid growth in soil patches rich in organic material .
Biochemical signature compounds for the estimation of biomass of ectomycorrhizal fungi
PLFA 18:2ω6,9 makes up a large proportion of the PLFAs in most fungi  and is suitable for detecting the growth of external mycelium of ectomycorrhizal fungi in soil . The amount of PLFA 18:2ω6,9 has been shown to be correlated with the concentration of ergosterol, a fungal-specific sterol, in soil samples . Ergosterol is widely used for estimating the biomass of ectomycorrhizal fungi in roots and soil , although it often lacks sensitivity and specificity for soil samples . PLFA 18:2ω6,9 may be the most sensitive compound for fungal biomass determination in many cases, especially since PLFAs in dead organisms are rapidly degraded . Ergosterol is, however, the compound of choice for the estimation of fungal biomass in ectomycorrhizal roots since PLFA 18:2ω6,9, but not ergosterol, is also a constituent of plant membranes.
Fatty acids and ergosterol in ectomycorrhizal fungi
High proportions of PLFA 18:2ω6,9 are typical in basidiomycetes, ascomycetes and deuteromycetes, but 18:2ω6,9 is also present in varying amounts in most other eukaryotes [29,30]. Species of the ectomycorrhizal genus Suillus showed about 10 times higher concentrations of PLFA 18:2ω6,9, as well as total PLFAs, than another ectomycorrhizal fungus (Paxillus involutus). This was evident both when grown in pure culture on agar and in soil associated with pine seedlings (P.A. Olsson and H. Wallander, unpublished). In contrast to this high variation between species, sterols seem to represent a relatively constant part of the fungal biomass, constituting somewhere between 0.5 and 1.5% in most fungal groups . This means that the sensitivity of PLFA 18:2ω6,9 as a biomass indicator differs widely between different fungal species.
The total fatty acid content in ectomycorrhizal fungi has been shown to be fairly constant with different carbon/nitrogen ratios of the growth substrate  (P.A. Olsson and H. Wallander, unpublished) and the signature PLFA 18:2ω6,9 should thus be suitable as a biomass indicator.
The use of PLFA 18:2ω6,9 to estimate the biomass of ectomycorrhizal fungi
Since PLFA 18:2ω6,9 seems to vary considerably between different fungal species it may not always be a relevant indicator of general fungal biomass in soil communities. Instead, this variation means that it is a sensitive biomass indicator for certain fungal species in controlled experiments.
The fungal biomass was estimated by using both the PLFA 18:2ω6,9 and ergosterol contents in experiments with ectomycorrhizal fungi growing with pine seedlings in soils amended with various primary minerals [40,46]. The biomass was calculated using conversion factors from pure culture studies, and values representing saprophytic fungi in non-mycorrhizal controls were subtracted. The mycorrhizal mycelium of Suillus variegatus and P. involutus grew most extensively in soils amended with biotite. Both methods indicated that the mycorrhizal mycelia comprised more than 50% of the total fungal biomass. Both fungi grew less in soil without additional minerals and S. variegatus could only be detected here by the use of PLFA 18:2ω6,9. This could be expected due to the high concentration of this compound in this fungus (see Section 4.1).
Estimation of the biomass of organisms interacting with mycorrhizal fungi
It is possible to use PLFAs to simultaneously estimate the biomass of many combinations of species and thus to study the interactions between a range of different microorganisms. A few such examples are discussed below.
Interactions between AM fungi and other fungi
The ability of AM fungi to protect plants from pathogenic fungi has been extensively studied (see ). The parameters studied are usually plant performance and disease symptoms. To elucidate the mechanism behind such interactions it is, however, necessary to study the two fungal groups in soil since several pathogenic fungi proliferate in soil and disperse in this way between hosts. Larsen et al.  found considerable differences in PLFA patterns between AM fungi and some saprophytic and pathogenic fungi. The biomass of the pathogenic fungus Fusarium culmorum (estimated using PLFA 18:2ω6,9) and the arbuscular mycorrhizal fungus G. intraradices (estimated using PLFA 16:1ω5) was estimated in the same soil samples. In a similar system interactions between the mycoparasitic biocontrol agent Trichoderma harzianum and G. intraradices were studied . The T. harzianum used in that study was GUS-transformed and could thus also be detected by the specific GUS activity. Both methods showed, for example, a decrease in the density of the mycelium of T. harzianum after the first 10 days of the experimental period.
Interactions between AM fungi and saprophytic fungal communities can be studied in soil in systems similar to those described above , for example, in systems involving a soil interface between mycorrhizal roots and organic material colonized by saprophytic fungi (see ).
Interactions between mycorrhizal fungi and bacteria
Bacteria contain several fatty acids that are usually not found in eukaryotic organisms, and these can be used to estimate the bacterial biomass or change in the bacterial community [41,49]. By using fatty acid signatures it is thus possible to estimate bacterial and fungal biomass in the same sample . Changes in bacterial community structure may be indicated by changes in the pattern of bacterial PLFAs (see ).
Fatty acid signatures provide suitable tools for estimating the amount of external AM fungal mycelium in soil, and can be used to determine relative carbon allocation to storage structures in AM mycelium.
PLFA 18:2ω6,9 is, in many cases, a sensitive signature compound for ectomycorrhizal mycelia in soil and can be determined simultaneously with the signature fatty acids for other groups of organisms, such as bacteria and AM fungi.
Proper controls without mycorrhizal fungi must be used in most cases for the estimation of the biomass of the mycorrhizal fungi. Fatty acid data should be supplemented with other measurements to confirm possible changes in biomass of different organisms. In the field it is difficult to obtain mycorrhiza-free controls, but these can be obtained by using fungicides or by the use of bags impermeable to fungal hyphae. These methods may, however, to a large extent influence the whole microbial community.
Simultaneous estimation of the biomass of mycorrhizal fungi and bacteria is possible using fungi-specific fatty acids in combination with bacteria-specific fatty acids. Published data may indicate which fatty acid signatures can be used for a particular interaction study, but when using model organisms in experiments, it is recommendable to estimate the fatty acid composition of these when grown in pure culture to obtain reliable conversion factors and avoid mistakes.
The use of signature fatty acids should be further evaluated in field situations to determine whether they can be used to indicate the dynamics of certain communities of mycorrhizal fungi.
The financial support provided by the Swedish Council for Forestry and Agricultural Research is gratefully acknowledged. Thanks are also due to Erland Bååth, Iver Jakobsen, Sally Smith, Bengt Söderström and Håkan Wallander for their valuable suggestions and comments on the manuscript.