Comparative Examination of the Olive Mill Wastewater Biodegradation Process by Various Wood-Rot Macrofungi

Olive mill wastewater (OMW) constitutes a major cause of environmental pollution in olive-oil producing regions. Sixty wood-rot macrofungi assigned in 43 species were evaluated for their efficacy to colonize solidified OMW media at initially established optimal growth temperatures. Subsequently eight strains of the following species were qualified: Abortiporus biennis, Ganoderma carnosum, Hapalopilus croceus, Hericium erinaceus, Irpex lacteus, Phanerochaete chrysosporium, Pleurotus djamor, and P. pulmonarius. Fungal growth in OMW (25%v/v in water) resulted in marked reduction of total phenolic content, which was significantly correlated with the effluent's decolorization. A. biennis was the best performing strain (it decreased phenolics by 92% and color by 64%) followed by P. djamor and I. lacteus. Increase of plant seeds germination was less pronounced evidencing that phenolics are only partly responsible for OMW's phytotoxicity. Laccase production was highly correlated with all three biodegradation parameters for H. croceus, Ph. chrysosporium, and Pleurotus spp., and so were manganese-independent and manganese dependent peroxidases for A. biennis and I. lacteus. Monitoring of enzymes with respect to biomass production indicated that Pleurotus spp., H. croceus, and Ph. chrysosporium shared common patterns for all three activities. Moreover, generation of enzymes at the early biodegradation stages enhanced the efficiency of OMW treatment.


Introduction
Three-phase centrifugal olive mills are widespread in most olive-oil producing countries and their operation results in the production of olive oil, solid pomace, and olive mill wastewater (OMW). The disposal of the latter creates major environmental problems since it is rich in organic compounds and contains large amounts of polyphenolics in high (and therefore) toxic concentrations [1,2]. In addition, the olive mills seasonal operation, small size, and scattered distribution hinder the effective management of the effluent [3].
In particular, wood-rot macrofungi constitute a group of organisms equipped with a potent biochemical arsenal including one or more nonspecific groups of enzymes, which permits them to effectively decompose various lignocellulosic compounds [19][20][21]. The chemical affinity of the latter to a wide range of agroindustrial wastes (including OMW) led to their treatment with filamentous basidiomycetes originally isolated from woody substrates. Therefore, such organisms were examined either at a OMW pretreatment stage for decreasing the effluent's high content of phenolics and hence toxicity prior to its processing with other microorganisms [22,23] or as the main/sole agents for OMW's remediation [24][25][26][27]. However, most pertinent investigations have focused on the exploitation of a few species only, for example, Phanerochaete chrysosporium, Ph. flavido-alba, Pleurotus ostreatus, and P. eryngii [28][29][30][31][32], and consequently a large number of other potentially useful macrofungi remain understudied or not examined at all.

BioMed Research International
The objective of the present work was the comparative evaluation of a wide variety of mushroom fungi for OMW biotreatment and the subsequent establishment of a battery of strains able to efficiently degrade OMW. The process was assessed by determining the activities of ligninolytic enzymes produced by qualified organisms at different time periods and through their association with concomitant OMW decolorization, reduction of phenolics, and decrease of plant toxicity.

Fungal Strains.
Sixty wood-rot macrofungi belonging to 43 species of basidiomycetes (phylum Basidiomycota) were evaluated for the purposes of this study. Details of the strains identity are presented in Table 1. All the biological material is maintained in the fungal culture collection of the authors institution (AUA-LGAM).

Nutrient Media: Fungal Growth Substrates and Conditions.
OMW was obtained from an olive-oil mill equipped with three-phase centrifugal decanters located in Kalamata (Peloponnese, S.W. Greece). Its composition and main physicochemical properties were previously assessed [2].
For the preparation of OMW-containing growth media, the effluent was adjusted at pH 6 by adding CaO and centrifuged for 20 min at 5000 g (5 ∘ C), and the supernatant was diluted or not with deionized water to prepare 25%, 50%, and 100% v/v OMW-based substrates, which were further heat-sterilized for 30 min (121 ∘ C, 1.1 atm). Fungal strains were subcultured on a 25% v/v OMW medium solidified with 1.7% w/v agar. Then, agar plugs (6 mm diam.) originating from the actively growing part of the fungal colonies were used to inoculate either solidified OMW media (for the purposes of mycelium growth rate determinations) or static liquid cultures (for OMW degradation experiments with selected strains).

Screening of Fungal Strains: Establishment of Temperature
Optima and Measurements of Linear Growth Rates. The temperature optima for mycelium linear growth were assessed on potato dextrose agar (PDA; Conda) medium and over the range of 12-47 ∘ C with a 5 ∘ C graduation. Experiments were conducted in Petri dishes as previously described [33], and measurements were taken every 24 h for all species examined with the exception of Phanerochaete chrysosporium and Hapalopilus croceus strains whose growth was measured every 12 h until seven consecutive values were obtained for all strains.
Growth rates were determined by measuring the distance of the colony's front from the centre of the inoculum at four different points along two perpendicular lines [34]. Mycelium growth rates ( ) were calculated by fitting the linear growth function = + (where is the distance covered by the hyphae and is the respective time) and were expressed in mm d −1 . All experiments were conducted in three replicates.
After establishing the optimal temperature for mycelium growth for each strain, all cultures were evaluated with respect to their growth rates on various OMW dilutions (25%, 50%, and 100% v/v OMW in water) at their optimal temperatures as previously determined, in the same way as described above. In addition, PDA was used as control. All mycelium growth rate experiments were conducted in three replicates. A destructive sampling process was adopted at five time points for dividing the entire cultivation at intervals of four, five, or six days (for growth periods of 20, 25, or 30 days, resp.), in order to determine pH, electric conductivity, biomass, total phenolics, decolorization, plant seed germination, and enzymes activity.

Treatment of OMW by Qualified
Measurements of pH and electric conductivity in OMW were performed by using a Corning EEL 12 pH meter and a Jenway 4010 conductivity meter, respectively. Biomass was harvested by filtration and its weight (mycelium dry weight) was determined by drying at 60 ∘ C until constant weight.
Total phenolics were analyzed by the Folin-Ciocalteu method [35], by measuring spectrophotometrically (at 760 nm) the formation of a blue complex as a result of the reduction of a phosphomolybdic-phosphotungstic reagent from the presence of phenolics. The concentration of total phenolics was determined against a syringic acid calibration curve (1 mg mL −1 syringic acid-initial concentrationresulted in an optical density of 0.377 at 760 nm).
Decolorization was estimated by measuring the absorbance of OMW samples at 525 nm using a U-2001 spectrophotometer (Hitachi Instruments Inc., USA).
The plant seed germination indexes were estimated on the basis of Zucconi et al. [36] protocol. Twenty-five cress seeds (Lepidium sativum L.) were placed on filter papers moistened with OMW (25%) and incubated in Petri dishes for three days at 25 ∘ C. The effect of OMW on germination was evaluated, and germination indexes (G.I.) were calculated as follows: G.I. = (% root elongation × % germination)/100 (control used: cress seeds moistened with water).
2.6. Statistical Analysis. Analysis of variance (ANOVA) followed by Gabriel's multiple comparison tests ( < 0.05) was used to estimate statistical differences between treatment means through the use of SPSS (version 18) software. Standard deviations were calculated for all mean values, and regression analysis was carried out to evaluate relationships between variables at significance levels of 1% and 5%.

Results and Discussion
Data about the optimal conditions for the production of fungal biomass are available for a rather few species of macrofungi (i.e., mainly those exploited for edible mushroom production). The effect of temperature, in particular, is of fundamental importance in evaluating the biotechnological potential of such organisms [33,34], and hence it was the first parameter which was assessed in the present study. The results of the comparative mycelium linear growth rate examination covering a wide temperature range (12-47 ∘ C) showed that most strains presented their optima at a narrow value area of 27-32 ∘ C ( PHC571 was the only one presenting optimal growth at 37 ∘ C. It is noteworthy that almost all Pleurotus strains showed a growth (albeit limited) at 12 ∘ C; Pleurotus was the only genus among the 28 different genera evaluated demonstrating such a property. In contrast, Hapalopilus croceus and Phanerochaete chrysosporium were the only species showing hyphal development at the other temperature extreme (42 ∘ C). Furthermore, mycelium growth rates of almost all strains presented a gradual increase from lower temperatures to the optimal one and then a sharp decrease to the temperature of no growth; in fact, this effect was more pronounced for individuals demonstrating high growth values ( Figure 1).
In the next comparative evaluation test, the mycelium growth rates of all strains were measured at their optimal temperatures and on three different OMW concentrations (25%, 50%, and 100% OMW v/v, in water) plus on PDA (Table 3).   [32], which advocates for a strain (rather than a species) dependent behavior. Consequently, eight fungi (i.e., Abortiporus biennis ABL436, Ganoderma carnosum GCL448, Hapalopilus croceus HCC522, Hericium erinaceus HEL801, Irpex lacteus ILC238, Phanerochaete chrysosporium PHL322, Pleurotus djamor PDC855, and P. pulmonarius PPL111) were qualified for further study on the basis of the outcome of the previous experiment, on the effect their growth had on the decolorization of the OMW-based substrate (assessed by visual inspection, data not shown), and on the prospects that individual species present in their subsequent/future exploitation for the generation of value-added products (e.g., edible biomass from Pleurotus spp.).
In all cases, fungal growth resulted in pH decrease from an initial value of 5.65 to values of as low as 4.48 for I. lacteus, while for the rest of the strains it ranged from 5.00 to 5.50 (Table 4). However, differences in pH were less pronounced for Pleurotus spp., which is in accordance with previous pertinent results obtained after the use of P. ostreatus [39]. On the other hand, electric conductivity measurements demonstrated a gradual increase from 1.80 mS cm −1 in the initial material to 3.85-5.97 mS cm −1 in the biodegraded effluent (Table 4); this seems to be the general trend for conductivity values after microbial treatment of raw OMW [40].
All selected strains produced abundant mycelium in static batch cultures; yet four out of the eight strains (A. biennis, H. croceus, P. djamor, and P. pulmonarius) provided significantly higher yields (200-230 mg) than the rest (Table 4). Nevertheless, biomass production at initial growth stages was rather poor since substrate's content in readily assimilated nutrients was low [2]. In general, fungal growth was accompanied with varying degrees of enzymes production among the various strains examined. A. biennis was the only fungus to demonstrate a particularly high laccase activity within the initial cultivation period; that is, a peak value of 205 U L −1 was obtained after 10 days only, and then it subsided and remained rather constant at values of 115-122 U L −1 until the end of the cultivation (Figure 2(a)). The rest of the strains demonstrated either significantly lower laccase activities or no activities at all. Within the former group, laccase production for both Pleurotus spp. and G. carnosum increased in parallel with biomass growth and reached the highest value by the end of cultivation (63-79 U L −1 ), whereas H. croceus and Ph. chrysosporium presented much lower values (3.6-5.1 U L −1 ). In contrast, in the cases of H. erinaceus and I. lacteus no laccase activity was detected.
As regards Mn peroxidase (MnP), a large variability was detected among strains in terms of both the peak activity values and the time periods where these were produced (Figure 2(b)). Hence, the highest MnP activities were observed for Table 3: Mycelium linear growth rates ( , mm day −1 ) for 60 strains of macrofungi growing on PDA and on OMW-based substrates prepared in various dilutions in water (v/v); ngd: no growth detected. Values are expressed as means ± standard deviation of means, = 3. Lack of common superscripts indicates statistically significant differences (Gabriel's -test, < 0.05) for comparisons of treatment means between different strains (numbers) and different substrates (lowercase letters).  12.83 ± 0.14 3,a 5.81 ± 0.07 8,9,b 4.71 ± 0.03 6,7,c 3.61 ± 0.08 2,d THC610 9.08 ± 0.06 6,7,a 5.78 ± 0.05 8,9,b 3.74 ± 0. 19  chrysosporium and the two Pleurotus spp. were early MnP producers since their maximum production (0.4-3.0 U L −1 ) was noted within the first 12 days of growth, and then it gradually subsided to levels of no detection at the end of cultivation. Production of MnP was previously associated with secondary metabolism activated by lack of adequate nitrogen and/or carbon [41]. This was apparently not the case during the early stages of incubation of Ph. chrysosporium and Pleurotus spp. used in this study, but it might well be the case for the other five strains which presented their peak activities with a time delay. On the other hand, no Mn independent peroxidase (MnIP) activities were detected by the two Pleurotus spp. tested (Figure 2(c)). All the other strains produced their peak MnIP values towards the second half of their growth period starting from day 12 (H. croceus, 2.3 U L −1 ) and ending at day 25 (Irpex lacteus and A. biennis, 32 U L −1 and 7.1 U L −1 , resp.). In general, monitoring of enzymes production with respect to fungal growth indicated that P. djamor and P. pulmonarius shared identical patterns for all three activities, and this was also the case for H. croceus and Ph. chrysosporium. On the other hand, A. biennis presented common production patterns with I. lacteus as regards MnP and MnIP and so did G. carnosum with H. erinaceus.
Although in some cases the findings of the present study are in line with previous literature reports on enzyme activities in relation to fungal growth, for example, MnP activity for Pleurotus and Ganoderma spp. [32,39], different types of data were also obtained. For example, MnIP activities in Ganoderma spp. showed their peak production earlier during the growth period [32] in comparison to the findings of the present study. Furthermore, previous pertinent results on P. eryngii and P. ostreatus revealed late production of MnP [32], which is in contrast with the results of this work (albeit with different Pleurotus spp.). While differences in enzyme activity values among strains used in various investigations could be attributed to both experimental conditions (including initial substrate composition and dilution rate) and properties of the particular fungus used, variations in enzyme production with respect to time could be primarily explained by the latter factor (i.e., variability of the biological material employed).
Total phenolic content was markedly reduced by all strains; however, particularly impressive was the reduction achieved by A. biennis, which after only six days of incubation decreased the initial phenols by more than 50%, while by the end of the experiment the respective figure reached 92% (Table 4). Very good performers (although significantly inferior than the first) were the two Pleurotus species (71%-76% of decrease), followed by I. lacteus and G. carnosum. The latter was the second fastest (after A. biennis) in reducing total phenols since a 49% decrease was measured after just 12 days of incubation. It is noteworthy that other Ganoderma species/strains were reported to present similar behavior as regards the time needed to decrease to a similar extent OMW's phenolics, that is, by 41-44% within a 10-day incubation period and by reaching a total reduction exceeding 64% [26,32]. On the other hand, selected P. ostreatus and P. eryngii strains demonstrated a rather delayed phenols degradation which exceeded the 50% level only after the completion of the first half of growth [32] much like as it was observed for P. djamor and P. pulmonarius in the present study. In certain cases (e.g., for A. biennis, G. carnosum, and I. lacteus), a notable deceleration in the reduction rate of total phenols reduction was observed during the second half of the growth period, which could be attributed to the recalcitrance of certain phenolics remaining in culture (e.g., trans-cinnamic acids which are not degraded by laccase) [42].
Of special interest was that total phenol reduction was significantly correlated with laccase and MnP activities ( 2 = 0.43 and 2 = 0.71, resp.) when data from all eight fungi Table 4: Biomass production by eight selected strains of macrofungi in OMW (25% v/v) substrates and pH, electric conductivity (E.C.), plant-seed germination index (G.I.), and biodegradation-related aspects as measured in five different time periods during fungal growth.   were collectively evaluated (  [32,39] but also when effluents from the debittering of green olives were treated by several white-rot fungi [43]. In this category of microorganisms, production of laccase is often induced by the presence of the appropriate substrate in the growth environment [44]. Especially as regards OMW media, such induction for Pleurotus spp. was associated with the concentration of the effluent in the culture substrate; high laccase activities were detected when phenolics exceeded 1.5 g L −1 [39,45,46], which is also the case with the initial phenolics content in the samples of this study (>2.0 g L −1 ). Most of the fungi examined here (with the only exception of A. biennis) demonstrated a lag phase for laccase production indicative of the time they needed to adapt their growth requirements to OMW [24,47]. As previously noted with phenols content, A. biennis was the most efficient fungus in decolorizing OMW by reaching a 64% value by the end of the cultivation period (Table 4). During the process, color increased (for this and some other strains) during the first measurement(s) as a consequence of the oxidation and/or polycondensation of phenolic compounds in other darker-colored forms [44,48]. Still, A. biennis achieved fast a very high decolorization rate (48% after 12 days) which is significantly higher than the respective values obtained from the other fungi during that same period (2%-18%). I. lacteus and Pleurotus spp. provided the next best total decolorization (56% to 39%), whereas G. carnosum performed rather poorly (14%). On the other hand, H. erinaceus was the only fungus that performed significantly better during the first half of the growth period, and then its decolorization subsided considerably.
As regards the phytotoxicity evaluation, A. biennis was again the best performer by demonstrating 30% increase of the germination index with respect to untreated OMW, while G. carnosum, I. lacteus, and H. erinaceus provided significantly lower values (15%-18%). In these cases, increase of plant seed germination was particularly slow during the first 1 to 2 weeks of incubation, which is in line with previous reports on P. ostreatus and P. eryngii achieving 19%-27% increase of germination index towards the end of the growth period on OMW [32]. On the other hand, the other four strains produced a very low increase in seed germinability (<10%) that could be possibly attributed to the formation of phenoxy radicals and/or quinonoids that are more toxic than the initial phenolic compounds [53,54]. In general, toxicity of OMW was associated with the presence of aromatic compounds of low molecular weight and with synergistic inhibition caused by phenolics [55][56][57]. Increase in seed germination was significantly correlated with reduction in phenolics and decolorization ( 2 = 0.84 and 0.71, resp., in cumulative calculations made for all strains tested); this high correlation with the former factor is in accordance with past reports on the germination of Triticum and Lepidium plant seeds in OMW treated by L. edodes and Pleurotus spp., respectively [24,32].
When the results presented above were reassessed by taking into consideration the time needed for obtaining the final values for total phenols reduction, decolorization, and increase of plant seed germination, classification of selected strains did not change considerably. For example, A. biennis was again significantly more efficient for all parameters evaluated; that is, it demonstrated a daily average of 3.70 percentage units of total phenols reduction, 2.56 percentage units of decolorization, and 1.20 percentage units of seed germination increase; Pleurotus spp., G. carnosum, H. croceus, and I. lacteus followed as regards total phenols reduction (with values of 2.35-2.53), I. lacteus for decolorization (2.24), and G. carnosum, I. lacteus, and H. erinaceus for seed germination increase (0.60-0.70).
It is particularly noteworthy that when the experimental data from all eight fungi were assembled and evaluated vis-à-vis their total phenol reduction, decolorization, plant seed germination increase, and enzymes activity (Table 5), then biomass and germination index showed statistically significant correlations in all comparisons ( 2 = 0.55-0.88 and 2 = 0.51-0.88, resp.). Laccase production was significantly   ( 2 ) values for OMW decolorization and total phenolics reduction versus the laccase activities presented by the eight selected macrofungi as calculated from the respective data obtained at five time points ( 1 to 5 ) during their entire growth period. correlated with total phenolics reduction and decolorization (Figure 3), and so was decolorization with MnIP ( 2 = 0.46). Last, MnP, and MnIP activities were also correlated ( 2 = 0.55).

Conclusions
All wood-rot macrofungi selected for the purposes of this study were particularly effective in colonizing and degrading OMW through an efficient enzyme-producing mechanism. Especially A. biennis and secondly I. lacteus, G. carnosum, and Pleurotus spp. achieved total phenol reduction and decolorization values, which were significantly correlated with laccase and/or peroxidases activities. Enhancement of plant seed germination was less pronounced but it demonstrated high correlation with the other two biodegradation parameters. This approach seems to be suitable (alone or in combination with other techniques) in the development of a sustainable methodology leading to the efficient treatment of OMW.