OUP user menu

Effect of continuous olive mill wastewater applications, in the presence and absence of nitrogen fertilization, on the structure of rhizosphere–soil fungal communities

Dimitrios G. Karpouzas , Constantina Rousidou , Kalliope K. Papadopoulou , Fotios Bekris , Georgios I. Zervakis , Brajesh K. Singh , Constantinos Ehaliotis
DOI: http://dx.doi.org/10.1111/j.1574-6941.2009.00779.x 388-401 First published online: 5 November 2009


Olive mill wastewater (OMW) is rich in potentially toxic organics precluding its disposal into water receptors. However, land application of diluted OMW may result in safe disposal and fertilization. In order to investigate the effects of OMW on the structure of soil fungal groups, OMW was applied daily to pepper plants growing in a loamy sand and a sandy loam at two doses for a period of 3 months (total OMW equivalents 900 and 1800 m3 ha−1). Nitrogen (N) fertilization alleviated N scarcity and considerably enhanced plant biomass production; however, when applied in combination with the high OMW dose, it induced plant stress. OMW applications resulted in marked changes in the denaturing gradient gel electrophoresis patterns of soil basidiomycete communities, while concurrent N fertilization reduced these effects. In contrast, the ascomycete communities required N fertilization to respond to OMW addition. Cloning libraries for the basidiomycete communities showed that Cryptococcus yeasts and Ceratobasidium spp. dominated in the samples treated with OMW. In contrast, certain plant pathogenic basidiomycetes such as Thanatephorus cucumeris and Athelia rolfsii were suppressed. The observed changes may be reasonably explained by the capacity of OMW to enrich soils in organic substrates, to induce N immobilization and to directly introduce OMW-derived basidiomycetous yeasts.

  • olive mill wastewater
  • denaturing gradient gel electrophoresis
  • fungi
  • basidiomycetes
  • ascomycetes
  • soil microbial community


Olive oil production constitutes one of the most important agricultural–industrial activities for the Mediterranean region in which over 95% of the world's olive oil is produced, reaching 3 × 106 tons per year (International Olive Oil Council, 2009). However, the three-phase extraction process, still applied by the majority of olive oil extraction industries, produces 1–1.2 m3 of olive mill wastewater (OMW) per ton of olives, which is equivalent to about 6 m3 of OMW per ton of olive oil (Azbar et al., 2004). This leads to a total production of over 10 × 106 tons of OMW per year. The OMW show highly variable properties (Aktas et al., 2001), but they are generally characterized by acidic pH (4.5–5.5) and a high organic load, with average biological and chemical oxygen demand values of 60 and 100 g L−1, respectively. The organic matter of OMW mainly consists of sugars, tannins, lipids, aromatic acids and phenolic compounds, with the latter being suggested as the main cause of its reported antibacterial and phytotoxic effects (Capasso et al., 1995; Isidori et al., 2005). The high load of potentially toxic organics established the notion that OMW should be regarded and treated as a pollutant. Indeed, its direct discharge into aquifers, or even into civil wastewater treatment plants, may create significant environmental problems. On the other hand, OMW is rich in biodegradable organic matter and inorganic elements, notably K, P, Fe and Mg. Indeed, soil application of OMW results in an increase in organic matter content and in available inorganic nutrients, mainly K, P, Mg and Fe (Levi-Minzi et al., 1992; Piotrowska et al., 2006; Brunetti et al., 2007). Thus, several studies have stressed its potential fertilizer value when applied in soils poor in organic matter and nutrients, such as those of most Mediterranean agricultural regions (De Monpezat & Denis, 1999; Paredes et al., 1999; Ehaliotis et al., 2003; Rinaldi et al., 2003). However, OMW application in land receptors leads to increased soil electrical conductivity and temporal immobilization of available nitrogen (N) (Levi-Minzi et al., 1992; De Monpezat & Denis, 1999; Sierra et al., 2007). Furthermore, its addition increases soil bulk density and reduces hydraulic conductivity (Colluci et al., 2002).

The changes imposed by OMW application on soil chemical properties and nutrient status are expected to have a significant effect on soil biological properties and soil microorganisms. However, information on changes in soil microbial communities after application of OMW is generally lacking. Early studies by Paredes et al. (1986) showed a substantial increase in coryneform bacteria and a decrease in Bacillus spp. Mekki et al. (2006) showed that addition of OMW resulted in significant increases in soil actinomycetes, spore-forming bacteria and soil fungi and a major decrease in bacterial nitrifiers. Further studies by Kotsou et al. (2004) showed a general increase in soil respiration immediately after the addition of OMW, while Piotrowska et al. (2006) demonstrated an increase in soil respiration, dehydrogenase and urease activity, but a decrease in phosphatase, β-glucosidase, nitrate reductase and diphenol oxidase activity in OMW-amended soil. All the above studies have used cultivation–enumeration methods or enzymatic activity measurements to identify the potential effects of OMW on soil microorganisms. However, it is now well documented that only 1–5% of the soil microorganisms can be cultivated in vitro (Aman et al., 1995; Torsvik & Ovreas, 2002), limiting the applicability of cultivation-based techniques in studying the effects of stressors at the soil microbial community level (Roszak & Colwell, 1987; Prosser, 2002).

Recently, biochemical and culture-based experimental approaches have been complemented by the investigation of the community structure using cultivation-independent methods including phospholipid fatty acid (PLFA) analysis and molecular fingerprinting techniques such as denaturing gradient gel electrophoresis (DGGE) and terminal restriction fragment length polymorphism (Singh et al., 2006). PLFAs were used for identifying the potential effects of varying rates of OMW application on the soil microbial community. It was found that addition of a dose of OMW, exceeding 30 m3 ha−1, resulted in increases in soil fungi, gram-negative bacteria and actinomycetes, but also in a significant decrease in gram-positive bacteria 1 year later (Mechri et al., 2007). In a subsequent study, it was shown that OMW application led to reduced colonization of the roots of olive trees by arbuscular mycorrhizal fungi (Mechri et al., 2008).

Among soil microorganisms, fungi constitute the key primary decomposers of soil organic matter, while bacteria contribute to cycling of important nutrient elements, such as N, S and P. Microbial composition often appears to have a functional significance in soil ecosystems (Bell et al., 2005; Reed & Martiny, 2007), and to affect aboveground plant communities as well (Hooper et al., 2000). This study used a DGGE approach to identify the effects of OMW on the structure of two main groups of soil fungi, basidiomycetes and ascomycetes, in the presence of plants. The choice was based on the biological traits of these fungal groups. Basidiomycetes include lignolytic fungi and yeasts participating in different stages of organic matter decomposition in soil. Basidiomycetous yeasts proliferate in soils enriched in aromatic compounds (Bergauer et al., 2005), whereas filamentous white-rot fungi have been used for the treatment of OMW (Sayadi & Ellouz, 2007) and are expected to be favored by the phenolic-rich composition of OMW and by the N immobilization induced by OMW application in the soil environment. On the other hand, Ascomycetous yeasts proliferate in OMW and have even been used for its treatment (Lanciotti et al., 2005; Ben Sassi et al., 2008), whereas several plant pathogens belong to the phylum of Ascomycota and the application of OMW might influence, positively or negatively, their proliferation in agricultural soils. Fingerprinting analysis of these fungal communities was complemented by a cloning strategy in order to identify fungi particularly responsive to OMW addition. Sequential application of OMW was practiced, which allows the gradual application of high total loads of OMW, and could be particularly useful for soils of the Mediterranean region, which suffer from water deficiency and are characterized by a low organic matter content. As OMW application to soils leads to N-limited conditions, which could be prevented by N fertilization, the experiments were performed in the presence or absence of N fertilization.

Materials and methods

OMW irrigation experiment

The two soils used in this study were collected from Velika, a typical olive grove area, in the vicinity of Kalamata city (SW Peloponnese, Greece), and were characterized as loamy sand (LS) and sandy loam (SL), with their physicochemical characteristics as shown in Table 1.

View this table:

The physical and chemical properties of the soils used in the study

Soil propertiesSoils
Sandy (%)52.386.6
Loam (%)30.611.1
Clay (%)17.12.3
Cation exchange capacity (cmol(+) kg−1 soil)10.05.7
Organic matter (%)2.180.27
Total nitrogen – Kjeldahl (%)0.110.04
Electrical conductivity (dS m−1)0.7150.370
POlsen (μg g−1)22.310.6
Naexch+ (cmol(+) g−1 soil)0.200.22
Kexch+ (cmol(+) g−1 soil)0.680.15
Caexch2+ (cmol(+) g−1 soil)3.330.96
Mgexch2+ (cmol(+) g−1 soil)0.851.30

Each soil was sieved to pass through a 3-mm mesh sieve in order to be homogenized and placed into 30 pots (3 L per pot). A single green pepper plant (Capsicum annuum) at the three- to four-leaf stage was transplanted in each pot. Immediately after transplantation, pots were placed under a light shade net in the field to protect plants from excessive sunlight during the entire experimental period. All pots and plants received 4.5 g of the phosphorus fertilizer in the form of superphosphate (0-20-0) as a regular agronomic practice. In addition, half of the pots for each soil received extra fertilization with 9 g of a N fertilizer in the form of ammonium nitrate (35-0-0). Watering was performed on a daily basis in all pots with water solutions containing 0%, 2% and 4% OMW for a period of 3 months (July to September). The experimental setup for each soil comprised of six treatments: 0% OMW with (OMW 0%+F) and without N addition (OMW 0%), 2% OMW with (OMW 2%+F) and without N addition (OMW 2%), 4% OMW with (OMW 4%+F) and without N addition (OMW 4%). Five replicate pots were used for each treatment. At the end of the irrigation period, the soil in the pots had received a total amount of 0, 900 or 1800 m3 of OMW ha−1 for the 0%, 2% and 4% treatments, respectively. When the 3-month irrigation period was completed, plants were uprooted, shoots were separated from fruit, dried at 65 °C and weighed, whereas soils from all replicates for each treatment were pooled and homogenized. Subsequently, three 50 g soil subsamples were placed in plastic airtight bags and stored at −20 °C until DNA extraction.

The OMW used in the current experiment was collected by a three-phase olive oil mill in the area of Kalamata, Messinia, and its characteristics are shown in Table 2. Before use, it was mildly treated with CaO (1% w/v), resulting in a pH increase from 4.5 to 5.7, and a large part of the suspended solids was removed by natural sedimentation.

View this table:

The physical and chemical properties of the OMW used in the study

Electrical conductivity (dS m−1)11
Chemical oxygen demand (g L−1)48
Total organic carbon (g L−1)26
Total solids (g L−1)42
Concentration of phenolics (g L−1)8.8
Total nitrogen (g L−1)0.9
K (g L−1)6.1
Ca (g L−1)1.1
P (g L−1)0.21
Mg (g L−1)0.12
Na (g L−1)0.07
Fe (mg L−1)65
Cu (mg L−1)2.4
Zn (mg L−1)3.4
Mn (mg L−1)0.9
  • OMW was treated before analysis with CaO to reduce acidity. Suspended solids were partially removed by natural sedimentation.

Soil DNA extraction and PCR

DNA was separately extracted from 0.5 g of soil from each of three stored samples of each treatment using the Power Soil DNA Isolation Kit according to the manufacturer's instructions (Mo Bio Laboratories).

In order to increase sensitivity a semi-nested PCR technique was used. In the first round of PCR, group-specific primers were used targeting the amplification of the internal transcribed spacer (ITS) region of the fungal genome. The primer ITS1F (Gardes & Bruns, 1993) was used along with either the ascomycete-specific reverse primer ITS4A (Larena et al., 1999) or the basidiomycete-specific reverse primer ITS4B (Gardes & Bruns, 1993), both amplifying a product of about 700 bp. Thermocycling for the basidiomycetes-specific PCR consisted of an initial denaturation at 94 °C for 3 min, followed by 35 cycles of 94 °C for 60 s, 55 °C for 30 s and 72 °C for 60 s. Thermocycling conditions for the ascomycetes-specific PCR consisted of an initial denaturation at 94 °C for 2.5 min, followed by 40 cycles of 94 °C for 15 s, 55 °C for 30 s and 72 °C for 60 s. The products obtained from the first PCR round were semi-nested with primers ITS1F-GC and ITS2. The forward primer was exactly the same as in the first PCR round, but with addition of a GC-clamp at its 5′ end (Muyzer et al., 1993). Thermocycling conditions consisted of an initial denaturation at 94 °C for 5 min, followed by 30 cycles at 94 °C for 30 s, 55 °C for 30 s and 72 °C for 60 s. In all PCR rounds, a final extension at 72 °C for 10 min was applied.

The final concentrations of the different components in the PCR amplifications were: 1 × polymerase buffer, 1.5 mM MgCl2, 200 μM of each deoxynucleoside triphosphate, 1 U DNA polymerase (EXT-DyNAzyme DNA polymerase, Finnzymes, Finland), 0.2 and 0.4 μM of each of the basidiomycetes-specific and ascomycetes-specific primers, respectively, and sterile-distilled water. Bovine serum albumin at a concentration of 400 ng μL−1 was included in all first-round PCR amplifications to prevent polymerase inhibition due to the presence of contaminants such as humic acids in the soil DNA. During the first PCR round, 1 μL of soil DNA was added to 24 μL of PCR mastermix, while in the second PCR round 1 μL of the amplified product from the first round was added to 49 μL of PCR mixture.

DGGE analysis

DGGE analyses were carried out on an INGENYphorU-2 × 2 system (Ingeny International BV, the Netherlands). Polyacrylamide gels (8%) were prepared in 1 × TAE buffer (40 mM Tris base, 20 mM acetic acid and 1 mM disodium EDTA, pH 8.2). Approximately 500 ng of DNA were loaded in each lane. The polyacrylamide gels were made with a denaturing gradient ranging from 30% to 55% (where 100% denaturant contains 7 M urea and 40% formamide). The electrophoresis was run for 16 h at 60 °C and 75 V. After the electrophoresis was completed, gels were silver stained as described previously by McCaig et al. (2001). The image was captured using a digital camera, and subsequent analysis of all DGGE gel pictures was performed using the gelcompar ii software 3.0v (Applied Maths, Sint Martens-Laten, Belgium).

Data on the presence and absence of bands in the DGGE fingerprints obtained were used for statistical analysis (genstat 7.0v): the presence/absence data matrices obtained by each DGGE gel were used for multivariate statistical analysis to estimate the effect of OMW alone or in combination with N fertilization on the structure of the soil microbial community. Principal coordinate analysis (PCoA) with a Jaccard similarity matrix was applied to the presence/absence dataset generated from the DGGE banding patterns of the different treatments to decrease dimensionality. Subsequently, the first six components of the PCoA were subjected to canonical variate analysis (CVA). Sample groupings were specified before CVA: OMW 0%, OMW 2%, OMW 4%, OMW 0%+F, OMW 2%+F and OMW 4%+F. CVA is a subjective statistical method that finds linear combinations of variates and is designed to maximize between-group differences, thus identifying subtle differences that other statistical approaches fail to discriminate (McCaig et al., 2001; Campbell et al., 2003; Singh et al., 2006).

Cloning of PCR products

Clone libraries for basidiomycetes were constructed based on the ITS rDNA fragments generated with the primer pair ITS1F–ITS4B using the same PCR conditions as described above. As the DGGE results showed that replicate samples of the same treatment had low variability, the triplicate PCR products of the same treatment were pooled and concentrated to a final volume of 30 μL using the Macherey-Nagel Nucleospin II PCR Clean-Up Kit (Macherey-Nagel GmbH, Germany). An aliquot of the concentrated product (5 μL) was subjected to agarose gel electrophoresis for estimation of the approximate DNA concentration. Subsequently, the purified product was cloned into plasmid vector pGEM®-T easy and the ligation product was transformed into Escherichia coli (DH5a High Efficiency Competent Cells – Invitrogen) following the manufacturer's instructions. The transformed cells were plated onto a Luria–Bertani (LB) medium containing ampicillin (100 μg mL−1), X-Gal (5-bromo-4-chloro-3-indolyl-b-d-galacto-pyranoside: 2% solution) and isopropyl-β-d-thiogalactopyranoside (0.1 M) to identify white-colored recombinant colonies.

Screening of clone libraries by DGGE

Screening of the clone libraries by PCR and DGGE was carried out as described by Liang et al. (2008). To confirm the presence of inserts, approximately 50 white colonies were selected for each sample, and subjected to colony PCR using the primers ITS1F-GC and ITS2 as described before, with the only modification that 25 cycles were performed instead of 30. Depending on the PCR result, 0.8–1.2 μL of the amplification product from each positive clone was run on DGGE gel to determine its electrophoretic mobility. Inserts with different DGGE mobilities were compared with the band pattern of the environmental original sample. Representative inserts for each band type/mobility matching the mobility of bands in the profile of the original environmental samples were sequenced. In cases where several clones showed an identical migration pattern in the DGGE, three clones were sequenced to check for comigration of diverse sequences.

Sequencing and affiliation of clones

For sequencing, plasmid DNA was extracted and purified from selected colonies using the NucleoSpin Plasmid Kit (Macharey-Nagel GmbH, Germany). Sequencing reactions were performed according to the manufacturer's instructions using a PRISM BigDye Terminator Cycle Sequence Reaction Kit (Applied Biosystems, Warrington, UK). Sequences were deposited in the European Molecular Biology Laboratory database under accession numbers FM866334FM866395. Similarity comparisons with known ITS sequences in the database were performed using the online basic local alignment search tool program (blast; http://www.ncbi.nlm.nih.gov/BLAST).


DGGE analyses demonstrated that the two soils support diverse microbial communities. The number of bands identified in the DGGE profiles of the basidiomycetes community in the LS and SL soil were 25–29 and 20–33, respectively. Accordingly, the corresponding numbers for the ascomycetes community in the two soils were 20–24 and 18–29. The number of different bands included in the statistical analysis ranged from 71 for the ascomycetes community in the LS soil to 81 for the basidiomycetes community in the same soil.

Overall, the DGGE profiles obtained for basidiomycetes and ascomycetes in the SL and the LS soil showed good reproducibility between the PCR-amplified DNAs from the replicate soil extracts of each treatment (Fig. 1; see also Supporting Information).


DGGE analysis of the community of basidiomycetes in the LS (a) and SL (b) soils. Lanes 1–9: samples without N fertilization; lanes 10–18: samples with N fertilization. In terms of OMW treatment rates, lanes 1–3 and 10–12 received no OMW (OMW 0%), lanes 4–6 and 13–15 were treated on a daily basis with a solution of OMW 2%, lanes 7–9 and lanes 16–18 were treated on a daily basis with a solution of OMW 4%. Lanes designated with M correspond to a fungal marker that contained 20 ng μL−1 of the ITS-PCR products of the following fungi with the sequence in which they appear on the gel from top to bottom: Pleurotus djamor, Fusarium oxysporum f. sp. radici-lycopersici, Fusarium solani, Pleurotus eryngii, Pleurotus ostreatus and Pleurotus cystidiosus. Bands identified through screening with cloning libraries are designated by arrows accompanied by a code number as shown in Table 3.

Effects of OMW and N fertilization on plant growth

The pepper plants grew better in the SL compared with the LS soil; however, the lack of N fertilizer severely restricted plant biomass production in both soils (Fig. 2). In the absence of N, OMW applications further restricted plant growth, especially regarding fruit production, a result more clearly observed in the SL soil. In the presence of N fertilizer, the high OMW application led to a major reduction in plant growth apparently related to plant toxicity in the LS soil (Fig. 2a); in the SL soil, however, plant biomass was not affected by OMW application, but the ratio of fruit to shoot biomass increased significantly (Fig. 2b).


Above-ground biomass production by pepper plants grown in the LS soil (a) and the SL soil (b). The gray part of the stalked bars indicates shoot biomass; the white part of the bars indicates fruit biomass. Error bars indicate SEs of means (n=5).

Effects of OMW and N fertilization on the microbial community of the LS soil

The first two principal coordinates of the PCoA represented, in all cases, a low percentage of the overall variation of the dataset (<25%). However, PCoA efficiently reduced the number of variates for the subsequent CVA, which was used on the first six components of the PCoA. CVA on the basidiomycetes community (Fig. 3a) clearly distinguished the high OMW dose treatment (OMW 4%) from the control and the low OMW dose treatment (OMW 0% and OMW 2%) along canonical variate (CV) 1, which explained most of the variance of the dataset. In the presence of N fertilization, however, the effect of OMW appeared to be comparatively smaller and the sample scores for the OMW treatments (OMW 2%+F, OMW 4%+F) did not differ. However, the OMW-treated samples were separated from the nontreated samples (OMW 0%+F) along CV1.


Ordination of CV produced from multivariate analysis of DGGE banding data from triplicate samples of the LS soil that were treated daily with a solution of 0% (○, ●), 2% (◻, ▪) or 4% (△, ▲) OMW with N fertilization (open symbols) or without (closed symbols). Graphs represent the differences between treatments in the soil community of basidiomycetes (a) and ascomycetes (b).

Regarding the community of ascomycetes in the LS soil, in contrast to basidiomycetes, a clear difference was observed between the nontreated samples and the corresponding OMW-treated samples in the presence of N fertilization along CV1, which explained most of the variance of the dataset (Fig. 3b). A less clear difference was evident for the nonfertilized samples that were separated along CV2. No effect of OMW dose was, however, evident, and samples receiving different doses of OMW (2% and 4%) were not distinguished either in the presence or in the absence of N fertilization.

Effects of OMW and N fertilization on the microbial community of the SL soil

Similarly, the first two PCoA scores accounted for <30% of the variance in the community data. Subsequent CVA on the first six principal components for the community of basidiomycetes in the SL soil distinguished clearly the different treatments according to the quantity of OMW received (Fig. 4a). The clearest separation between treatments was obtained for the samples that received OMW without N fertilization (OMW 0%, 2% and 4%), which were separated along both CV1, which explained most of the variance, and CV2. On the other hand, a relatively weaker, but still clear, separation along CV1 was evident between N-fertilized samples that were not treated (OMW 0%+F) and N-fertilized samples treated with 2% and 4% OMW.


Ordination of CV produced from multivariate analysis of DGGE banding data from triplicate samples of the SL soil that were treated daily with a solution of 0% (○, ●), 2% (◻, ▪) or 4% (△, ▲) OMW with N fertilization (open symbols) or without (closed symbols). Graphs represent the differences between treatments in the soil community of basidiomycetes (a) and ascomycetes (b).

In contrast to what was observed for basidiomycetes, the ascomycetes community again appeared to be more responsive to OMW effects in the presence of N fertilization: CVA of the ascomycetes community in the SL soil demonstrated a very clear separation along CV1 in the presence of N fertilization between the untreated (OMW 0%+F) and the OMW-treated samples (OMW 2%+F, OMW 4%+F) (Fig. 4b). In contrast, the treatments that did not receive N fertilization, but only OMW (OMW 2%, OMW 4%), grouped together with the corresponding nontreated samples (OMW 0%), but also with the nontreated samples that were N-fertilized (OMW 0%+F).

Cloning of the basidiomycetes ITS sequences

The DGGE fingerprints obtained from the different soil treatments suggested that the community of basidiomycetes was the most affected by the application of OMW, while the effect on ascomycetes could not be clearly attributed to OMW, but rather to N fertilization. Thus, we focused on the development and screening of cloning libraries for the basidiomycetes community in the different soils and treatments in order to identify members of the community that are responsive to OMW application.

The dominant DGGE bands and several of the less dominant ones were represented in the cloning libraries prepared. Overall, more than 400 clones were screened; 95 clones that showed the same migration patterns with bands in the DGGE profiles were sequenced and blast analysis showed that 63 out of the 95 sequences were unique and their highest homology is reported in Table 3. The bands in the DGGE profile of the soil samples that showed the same migration pattern as the clones sequenced are indicated in Fig. 1.

View this table:

Identity of selected DGGE bands from clones obtained for soil basidiomycetes in the different treatments

Band numberSoil originClosest match from GenBank (% sequence similarity by blast)*GenBank accession no.
B3-1LSUncultured Cryptococcus clone (99%)AY254865
B3-2LSUncultured basidiomycetes clone (83%)EU292299
B3-3LSUncultured soil fungus clone (96%)DQ421094
B3-4LSThanatephorus cucumeris (99%)AF354076
B3-6LSUncultured soil fungus clone (99%)DQ421268
B3-7LSCeratobasidium sp. (93%)DQ858827
B3-8LSUncultured Ceratobasidium sp. (98%)EF154354
B3-9LSUncultured ectomycorrhiza Entolomataceae (93%)FJ210729
B3-10LSCryptococcus terreus (99%)AB032682
B4-1LSCryptococcus podzolicus clone (99%)AJ581036
B4-2LSUncultured fungal clone (97%)DQ054552
B4-4LSCryptococcus elinovii (100%)AF444367
B4-5LSLycoperdon pratense (99%)AJ237625
B4-6LSPisolithus tinctorius (99%)AF374712
B4-7LSCryptococcus cellulolyticus (99%)AF444442
B4-9LSUncultured ectomycorrhiza fungus (96%)AJ633588
B5-1LSCryptococcus elinovii (100%)AF444367
B5-2LSCeratobasidium sp. (99%)AF354082
B5-3LSRhizoctonia sp. (97%)AJ242887
B5-4LSUncultured Tremellales isolate (99%)EU046042
B5-6LSUncultured Tremellales isolate (99%)EU046042
B5-7LSCeratobasidium sp. (99%)AF354082
B5-8LSUncultured soil fungus (99%)DQ421268
B5-9LSUncultured Filobasidiaceae clone (99%)EU046038
B7-1SLThanatephorus cucumeris (100%)DQ102446
B7-2SLCeratobasidium sp. (97%)AF354081
B7-3SLUncultured soil fungal clone (99%)DQ421268
B7-4SLCeratobasidium sp. (97%)AF354081
B7-5SLConocybe crispa specimen voucher (92%)AY194556
B7-6SLLaetisaria arvalis (99%)EU622841
B7-7SLUncultured ectomycorrhiza clone (93%)EF411074
B7-8SLHannaella sinensis (99%)DQ297411
B7-9SLEntoloma sp. (93%)DQ974695
B7-10SLMarasmius oreades (98%)EF187911
B7-11SLUncultured soil basidiomycete clone (94%)EU090897
B7-12SLCeratobasidium cornigerum (92%)EU273525
B8-1SLThanatephorus cucumeris (99%)DQ102449
B8-2SLThanatephorus cucumeris (100%)DQ102446
B8-3SLUncultured ectomycorrhiza isolate (Tomentella) (94%)EF218834
B8-4SLCeratobasidium cornigerum isolate (92%)EU273525
B8-5SLUncultured Filobasidiaceae isolate (98%)EU046047
B8-6SLUncultured soil basidiomycete clone (94%)EU090897
B8-7SLCeratobasidium isolate (95%)DQ028790
B8-8SLCryptococcus aerius (99%)AF444376
B8-9SLCryptococcus aerius (99%)AB032666
B9-1SLCeratobasidium sp. (97%)AF354081
B9-2SLUncultured ectomycorrhiza isolate (Tomentella) (94%)EF218834
B9-3SLUncultured basidiomycete clone (92%)AM901940
B9-4SLCeratobasidium sp. (99%)EU591764
B9-5SLCeratobasidium sp. (97%)AF354092
B9-6SLFungal endophyte isolate (92%)EU686118
B9-7SLCeratobasidium sp. (99%)AB219145
B9-9SLCryptococcus phenolicus (99%)AF444351
B10-1SLAthelia rolfsii strain (99%)AY684917
B10-2SLUncultured soil fungus clone (94%)EF433959
B10-3SLLeucocoprinus cretaceus isolate (98%)AY176447
B10-4SLBovista capensis (99%)DQ112621
B10-5SLUncultured ectomycorrhiza isolate (Tomentella) (93%)EF218826
B10-6SLUncultured soil basidiomycete clone (99%)DQ672332
B10-7SLThanatephorus cucumeris (99%)DQ102445
B10-10SLUncultured Filobasidiaceae isolate (99%)EU046047
B11-1SLUncultured soil fungus clone (99%)DQ421268
B12-1SLUncultured Filobasidiaceae clone (100%)EU046047
  • * Matches are based on c. 700-bp-long sequences of the ITS region of basidiomycetes.

Generally, all clone sequences showed the highest homology to basidiomycetous sequences verifying the specificity of the primer set used for basidiomycetes. In the LS soil, clones that showed the highest sequence homology to Cryptococcus sp. (i.e. B4-4 and B5-1: 100% to Cryptococcus elinovii; B5-9: 99% to an uncultured Filobasidiaceae ribotype; B3-1, B4-1, B5-6 and B3-10: 98–99% to different Cryptococcus ribotypes) showed a matching migration with dominant bands in the DGGE profiles of the OMW-amended samples, regardless of N addition (Fig. 1a). Similarly, clones with high sequence homology to Ceratobasidium ribotypes (Table 3), such as B3-8 (98%), B5-2 (99%) and B5-7 (99%), were associated with DGGE bands that were more prominent in the profiles of OMW-amended samples (Fig. 1a).

On the other hand, a number of clones only appeared in soil samples that were not amended with OMW or became less intense at the highest OMW rates (Fig. 1a). These clones included a clone that has a 97% sequence homology to a Rhizoctonia ribotype (B5-3, Table 3) and clone B4-9 (96%), with the highest sequence homology to an ectomycorrhizal Tomentella ribotype (Fig. 1a). Clones showing the highest sequence homology (99%) to Thanatephorus cucumeris ribotypes were associated with DGGE bands appearing in the OMW 0% and reappearing less intensely in OMW 4% (Fig. 1a, B3-4).

In the SL soil clone, B10-1 showed the highest sequence homology to Athelia rolfsii (99%) (Table 3). This clone showed matching migration, with a DGGE band appearing only in the OMW 0%+F samples (Fig. 1b). In line with the LS soil, clones showing the highest sequence homology (99–100%) to T. cucumeris (B7-1, B8-1, B8-2 and B10-7) were associated with bands seen only in the OMW 0% or OMW 2% samples (Table 3, Fig. 1b).

On the other hand, several clones showed identical migration patterns with DGGE bands that were generally more prominent in the samples treated with OMW (Fig. 1b). Thus, clones showing the highest sequence homology to Cryptococcus sp. (B8-8 and B8-9: 99% to Cryptococcus aerius; B9-9: 99% to Cryptococcus phenolicus) were associated with dominant DGGE bands from the samples that were treated with OMW (2% and 4%) (Table 3, Fig. 1b). Similarly, clone B12-1 (100%) showed the highest sequence homology to an uncultured clone belonging to the family of Filobasidiaceae, which comprises the genera of Cryptococcus and Fillobasidium (Table 3), and showed matching migration with a band visible only in the OMW 4%+N samples (Fig. 1b). Several clones (B8-7, B9-1, B9-7, B9-4 and B9-5) in the SL soil showed the highest sequence homology to Ceratobasidium ribotypes (Table 3). These clones appeared mainly in samples treated with OMW.

In general, clones derived from different cloning libraries (samples), but having the same migration pattern in DGGE, showed 99–100% sequence homology between them. The only exception was for clones B7-3, B11-1, B8-5 and B10-10 (Table 3).


The continuous irrigation of the soils with diluted OMW results in high availability of variable easily decomposable C sources and nutrients, increased amounts of recalcitrant phenolics and concurrent immobilization of N (Levi-Minzi et al., 1992; Gioacchini et al., 2007; Sierra et al., 2007). N availability in both soils studied was limited, as clearly indicated by the dramatic response of plants to N fertilizer (Fig. 2). N availability became more restricted following OMW additions, confirming the expected N immobilization effects by OMW. N fertilizer boosted plant growth in both soils, and mostly alleviated OMW effects at the lower OMW dose, indicating the importance of N supply to plants in combination with OMW applications. However, plants were sensitive to the higher OMW dose; cumulative effects of both N fertilizer and OMW on soil salinity, lower oxygen availability in the rhizosphere and increased availability of phenolics are probably partly responsible for toxicity to plants in the LS soil. In the SL soil, although plant biomass was not reduced, the increased fruit-to-shoot biomass ratio also indicates a stress effect. The SL soil has greater clay and organic matter content (Table 1) and the improved plant tolerance in this soil is probably related to buffering effects by the respective colloidal surfaces. The role of clays and other soil colloids as absorbents and oxidative agents for phenolics (Huang et al., 1977; Cecchi et al., 2004) is in line with the above interpretation. It should be noted that both OMW doses used in this study were rather extreme, as in practice, doses for field applications do not typically exceed 200 m3 of OMW ha−1 (Ehaliotis et al., 2003).

The addition of OMW alone (no N fertilizer applied) at two dose levels, which correspond to total volumes of 900 and 1800 m3 ha−1 for a 3-month period, resulted in significant changes in the community of basidiomycetes. The changes were consistent regardless of the type of soil studied. These results are in line with the physiological and ecological attributes of basidiomycetes, which constitute one of the most important and ecologically diverse group of soil fungi encompassing plant pathogens, ectomycorrhizal symbionts, general saprotrophs and specialized recalcitrant-C decomposers. Apart from the above, yeast populations in mineral soils are mostly basidiomycetous (Wuczkowski & Prillinger, 2004; Vishniac, 2006). White-rot fungi are known saprotrophic basidiomycetes that, under N deficiency, activate extracellular enzymatic mechanisms and metabolize a wide array of aromatic compounds including organic pollutants, such as phenolics (Bumpus et al., 1985; Rodriguez et al., 2004), which are present at high concentrations in OMW (Kissi et al., 2001; Aggelis et al., 2002). Yeasts and saprotrophic white-rot fungi are important decomposers of organic matter in soil, dominating different stages of the process (Deacon, 1997; Thormann, 2006). Chernov (1985) observed yeast dominance during the early stages of decomposition, but as readily soluble carbohydrates became limited, white-rot fungi succeeded the initial yeast community to continue the decomposition process.

A different picture regarding the structure of the basidiomycete community in the two soils was obtained when application of OMW was supplemented with N fertilization. Indeed, the supplementation of OMW with N fertilizer markedly reduced the changes observed in the structure of the community of basidiomycetes compared with the corresponding nonamended samples, and this was evident in both soils. Possibly, N fertilization reversed the N deficiency conditions that favor certain basidiomycetous groups such as the white-rot fungi and the ectomycorrhizal fungi, thus creating a soil microenvironment that is less conducive for them. The larger rooting systems developed in the N-fertilized treatments may have also affected C-source quality and availability in the rhizosphere soils, contributing to the development of well-established, rich and robust zymogenous antagonistic microbial communities, less prone to changes by OMW additions.

The sequencing of clone libraries for the communities of basidiomycetes provided information on the phylogeny of the dominant DGGE bands that is in agreement with the above regarding the differences brought about by OMW. In both soils, ribotypes with high homology to Cryptococcus yeasts, Ceratobasidium and Rhizoctonia were identified, but demonstrated different responses to OMW addition.

Cryptococcus spp. were generally favored by OMW application regardless of N supplementation. Previous studies have demonstrated that yeasts in mineral soils are largely members of the polyphyletic genus Cryptococcus (Vishniac, 2006). Application of OMW resulted in an enrichment of soil with easily decomposable carbon substrates, which could be metabolized by Cryptococcus. Members of this genus are known to metabolize various simple and complex sugars, organic acids, sugar alcohols, glycosides and amino acids (Thormann et al., 2007), which are present in large amounts in the organic fraction of OMW. Moreover, C. elinovii, C. aerius and C. phenolicus, which appear to become dominant in the OMW-amended samples of the two soils, belong to the Aerious clade of Filobasidiales (Scorzetti et al., 2002) and are closely related to ribotypes of the other two members of this clade (Cryptococcus terreus and Cryptococcus terricola) that have recently been shown to degrade phenolic compounds efficiently (Bergauer et al., 2005). It appears therefore that degradation of phenolics is a common characteristic shared by Cryptococcus species of the Aerious clade. Indeed, C. phenolicus degrades phenolics (Fonseca et al., 2000), and C. elinovii may remove phenol from industrial wastewaters (Morsen & Rehm, 1987) and grow at the expense of several benzene compounds that utilize them as sole sources of carbon (Middelhoven, 1993). Basidiomycetous yeasts were also shown to dominate rhizosphere DGGE profiles at late plant growth stages dominated by root senescence and high carbon availability, in contrast to ascomycetes that dominated the rhizosphere soil at earlier plant growth stages (Gomes et al., 2003). Increased plant root senescence probably occurred in our experiments, especially at the high OMW dose, further contributing to substrate availability favoring Cryptococcus yeast proliferation.

Regarding the continuous OMW applications, the possibility of direct enrichment of the soils with basidiomycetous yeasts should not be disregarded. However, Cryptococcus yeasts are rarely reported in OMW. Indeed, yeasts proliferate in OMW (Ben Sassi et al., 2006; Amaral et al., 2008), but they belong mainly to the genera of Candida and Pichia (Ben Sassi et al., 2008); Candida and Geotrichum species have also been isolated from olive-mill wastes (Giannoutsou et al., 2004), and Yarrowia lipolytica has been tested extensively for the treatment of OMW (Lanciotti et al., 2005).

Several bands in the DGGE profile of samples that were either nontreated with OMW or treated with 2% OMW showed high sequence homology (>99%) to T. cucumeris and its anamorph Rhizoctonia solani; the R. solani complex includes several plant pathogens (Otero et al., 2002). The absence of ribotypes highly homologous to Rhizoctonia from the samples that were treated with the higher OMW dose could be attributed to a suppressive effect of OMW on this plant pathogen. In previous reports, it was suggested that a nutrient-rich environment that is dominated by r-strategists (Kotsou et al., 2004) and/or direct toxicity exerted by the OMW phenolic constituents (Yangui et al., 2007) could be responsible for this suppressive activity. A similar OMW-induced suppressive effect was observed for A. rolfsii, which is the teleomorph of Sclerotium rolfsii, a soilborne fungal pathogen that causes southern blight disease on a wide range of agricultural and horticultural crops (Dodd et al., 2000). Its presence in this soil is not unexpected because the soils used in this study were previously cultivated with lupinus (Lupinus albus), a host plant of A. rolfsii. Suppressiveness against S. rolfsii has been related to high doses of organic amendments (Blum & Rodríguez-Kábana, 2004) and increased concentrations of biodegradable carbon (Danon et al., 2007), conditions that predominated in the high OMW dose treatment.

Ceratobasidium ribotypes were also identified as dominant members of the basidiomycete community in both soils. Many Ceratobasidium spp. have a Rhizoctonia anamorph (Hietala et al., 1994; Otero et al., 2002), and they are known as pathogens of turfgrass and cereals (Currah, 1991). On the other hand, Ceratorhiza, the anamorphic genus of Ceratobasidium, is one of the most common endophytes isolated from temperate orchids (Zelmer & Currah, 1995). Our cloning results suggest that Ceratobasidium strains are promoted in OMW-amended soils in contrast to other members of the R. solani complex such as Thanatephorus strains, which are suppressed. Other fungal clones that appear to be inhibited by OMW addition in the SL soil showed the highest sequence homology to the saprotrophs Leucocoprinus cretaceus and Entoloma sp.

Another important point in our findings is the presence of clones in the library of the OMW 0% samples of the SL soil showing the highest sequence homology to Laetisaria arvalis (99%) and Hannaella sinensis (99%). The former is a soil-dwelling basidiomycete that secretes an allelopathic agent which induces rapid hyphal lysis in several phytopathogenic fungi including R. solani (Conway et al., 1997). On the other hand, strains of H. sinensis were found to secrete a fungicidal thermolabile and protease-sensitive toxin; this mycocin acted only against tremellaceous yeasts (Cryptococcus) (Golubev & Nakase, 1997). The presence of H. sinensis and L. arvalis could be associated with the extensive presence of their respective antagonists, Cryptococcus and Rhizoctonia spp., in the studied soils. Overall, addition of OMW in agricultural soils appears to inhibit the proliferation of certain soil basidiomycetes, including certain plant pathogens, as demonstrated by our findings.

Regarding the ascomycetes, the OMW application had less intense effects on their community by hardly altering its structure in the SL soil and by producing a slight effect in the LS soil. However, the addition of OMW had an impact on the ascomycete community when N fertilization was also applied. We suggest that, in contrast to the basidiomycetes, the availability of N in soil constitutes a prerequisite for the growth and proliferation of soil ascomycetes and probably enables them to utilize the surplus of decomposable organic carbon introduced into the soil by OMW addition. However, further work is needed to confirm this hypothesis.

A limited number of studies have so far addressed the effects of OMW addition on the structure of the soil microbial community. Furthermore, the majority of these studies have used cultivation-dependent techniques to enumerate soil microbial populations in response to soil application of OMW (Paredes et al., 1986; Tardioli et al., 1997; Ehaliotis et al., 1999; Mekki et al., 2006). Only recently, Mechri et al. (2007),(2008) studied the effect of a single application of OMW on the structure of the soil microbial community using PLFAs. They reported that addition of OMW at dose rates lower than those used in our study (30–150 m3 ha−1) resulted in a significant increase in the fungal/bacterial PLFAs ratio. In addition, they found that addition of OMW produced a significant reduction of the ratio of PLFAs cy17:0/16:1ω7 or cy19/18:1ω7, which indicated that the soil microbial population was not under stress (Kaur et al., 2006). These results further support our suggestion that the observed effects of OMW on the different microbial groups reflect their capacity to drastically alter the nutritional status of the soil via enrichment of the soil in organic and inorganic nutrients.

Major ecosystem processes, such as decomposition and nutrient mineralization, are generally carried out by a large number of microbial taxa and often appear to be insensitive to microbial diversity changes, suggesting high functional redundancy (Wardle & Giller, 1996; Wardle, 2006). However, processes based on a restricted number of specialized bacterial taxa such as nitrification, N fixation and methane oxidation may be sensitive to microbial diversity changes (Wardle et al., 2002; Hawkes et al., 2005; Singh et al., 2007). Changes observed in the structure of the community of basidiomycetes could also be related to functional changes affecting the decomposition of recalcitrant xenobiotics such as polyphenols and humification processes. Further studies investigating the effects of OMW on long-term soil humification processes and on different bacterial groups including actinomycetes and ammonia oxidizers will provide answers to these questions.


OMW, used alone or in combination with N fertilization, had significant and rather consistent effects on the structure of the rhizosphere–soil fungal communities in two different soils, which may be summarized as follows:

  • 1. Addition of OMW significantly influenced the structure of the community of basidiomycetes, which, as efficient degraders of organic carbon and N scavengers, were the most responsive group to OMW additions especially in the lack of N fertilization. Ascomycetes, however, generally appear in need of N fertilization for responding to OMW addition.

  • 2. OMW appeared conducive for certain basidiomycetous yeasts, in both soils including phenolic-degrading Cryptococcus yeasts; however, it is not clear whether this is a result of changes in carbon quality and availability by OMW or whether continuous direct enrichment with OMW-derived microbial communities contributed to this result.

  • 3. Application of OMW appears to have a suppressive effect on various soil basidiomycetes at the microbial community level, and certain plant pathogens such as R. solani and A. rolfsii appear to be among them.

Supporting Information

Fig. S1. DGGE analysis of partial ITS sequences of the community of ascomycetes in the LS soil. In lanes 1-9 all samples have received no nitrogen fertilization unlike lanes 10-18 which correspond to samples which received nitrogen fertilization. As far as OMW treatment rates are concerned lanes 1-3 and 10-12 received no OMW (OMW 0%), lanes 4-6 and 13-15 were treated on a daily basis with a solution of OMW 2%, 7-9 and 16-18 were treated on a daily basis with a solution of OMW 4%. Lanes designated with M correspond to fungal marker lanes as described in Fig.1.

Supporting Information

Fig. S2. DGGE analysis of partial ITS sequences of the community of ascomycetes in the SL soil. In lanes 1-9 all samples have received no nitrogen fertilization unlike lanes 10-18 which correspond to samples which received nitrogen fertilization. As far as OMW treatment rates are concerned lanes 1-3 and 10-12 received no OMW (OMW 0%), lanes 4-6 and 13-15 were treated on a daily basis with a solution of OMW 2%, 7-9 and 16-18 were treated on a daily basis with a solution of OMW 4%. Lanes designated with M correspond to fungal marker as described in Fig. 1.

Please note: Wiley-Blackwell is not responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.


This research was funded by the Greek General Secretariat of Research and Technology (EPAN 4.5.1 –‘FP66’). B.K.S.'s laboratory is supported by grants from the Scottish Government.


  • Present address: Georgios I. Zervakis, Laboratory of General and Agricultural Microbiology, Department of Agricultural Biotechnology, Agricultural University of Athens, Athens, Greece

  • Editor: Philippe Lemanceau


View Abstract