New Insights on Protein Recovery from Olive Oil Mill Wastewater through Bioconversion with Edible Filamentous Fungi

Olive oil mills represent an important sector in the Mediterranean Sea Basin but also an environmental hazard due to untreated wastewater. Recovery of nutrients from olive oil mill wastewater (OMWW) as protein-rich microbial biomass can produce novel feed and reduce its chemical oxygen demand; however, low-protein containing products have been reported. New strategies leading to higher protein-containing fungal biomass could renew the research interest on bioconversion for pollution mitigation of OMWW. In this work, through cultivation of edible filamentous fungi (Aspergillus oryzae, Neurospora intermedia, and Rhizopus delemar), a link between the protein content in the originated fungal biomass, and the addition of nitrogen and medium dilution was established. Addition of nitrogen in the form of NaNO3 reduced the cultivation time from 96 h to 48 h while achieving a similar biomass mass concentration of 8.43 g/L and increased biomass protein content, from w = 15.9% to w = 29.5%. Nitrogen addition and dilution of OMWW, and consequent reduction of suspended solids, led to an increase in the protein content to up to w = 44.9%. To the best of our knowledge, the protein contents achieved are the highest reported to date and can open new research avenues towards bioconversion of OMWW using edible filamentous fungi.


Introduction
Olive oil production is an important economic sector in the Mediterranean Sea Basin. Olive oil is extracted through the so-called discontinuous (pressing) or continuous (centrifugation) processes. The pressing process is the oldest method and benefits from cheap equipment, technical simplicity, and low generation of wastewater, but it is labor-intensive. The centrifugation process is an automated process and more common within the sector; however, it produces a large amount of wastewater [1]. In this process, olives are washed and crushed, and then mixed in a tank for a certain period of time to obtain oil droplets. This is followed by a series of centrifugation steps to achieve olive oil separation. Concomitantly, large amounts of byproducts, namely, pomace and olive oil mill wastewater (OMWW), are generated ( Figure 1).

Figure 1.
Overview of the main steps taking place in an olive oil mill, adopting the continuous (centrifugation) process. The production of olive oil mill wastewater (OMWW), presently left untreated and thereby with high environmental pollution potential, is highlighted.
The use of microorganisms, especially edible filamentous fungi, for the valorization of industrial wastewaters and other low-value byproducts, has been among the most studied strategies. In addition to the wide range of enzymes, enabling growth in most (if not all) substrates, and valueadded products that can be produced, the macroscopic mycelium, obtained following assimilation of nutrients and easily recovered from the medium, provides superior performance of filamentous fungi towards reduction of COD levels, in comparison to unicellular microorganisms (e.g., bacteria, yeasts, and microalgae) [14]. The use of the latter normally entails costly biomass recovery processes. Filamentous fungal biomass normally contains high protein levels, essential amino acids, polyunsaturated fatty acids, and its cell wall is rich in compounds with immunostimulant activity [15]. Therefore, it has increasingly been studied as a replacement of fish meal and soybeans in the production of animal feed, a product in increasing demand due to population growth. The growth of filamentous fungi in other fat-rich substrates such as dairy products [16][17][18] and fish processingderived wastewaters [19,20], for protein recovery, has been reported in the literature. Research studies on protein recovery from OMWW using filamentous fungi, although scarce and more than eight years old, also exist [21][22][23]. However, these are characterized by low protein-containing products if the landscape of protein contents reported among fungal biomasses obtained from the growth in various low-value substrates is considered.
The aim of the present study was to develop a new cultivation strategy to enhance the protein level present in fungal biomass obtained from the cultivation of edible filamentous fungi in OMWW. OMWW represents an environmental hazard in view of its slightly acidic pH, high content of toxic compounds such as phenols, and high chemical oxygen demand (COD) [2,3]. Moreover, olive oil production is seasonal and connected to the harvest period, which influences the range of treatment strategies that can potentially be applied. If left untreated, OMWW can affect the color of natural waters and create odor and pollution due to emission of methane and hydrogen sulfide [4]. Consequently, OMWW is considered a potential contaminant source for flora and fauna associated with soil and water pollution [3]. To overcome its environmental hazard, several strategies, including aerobic and anaerobic biological processes, have been applied to reduce the polluting load of OMWW [1,5]. Alternatively, research approaches have also attempted to extract value from OMWW. For instance, extraction of phenolic compounds, such as hydroxytyrosol and tyrosol, with potential use as antimicrobial, antioxidant, and anti-inflammatory agents, has been reported [6,7]. Due to its composition, rich in, e.g., sugars, fat, and polyalcohols, OMWW has also been considered a potential substrate for production of hydrogen [8], biofuel through supercritical water gasification [9], enzymes [10][11][12], and protein-rich microbial biomass [13].
The use of microorganisms, especially edible filamentous fungi, for the valorization of industrial wastewaters and other low-value byproducts, has been among the most studied strategies. In addition to the wide range of enzymes, enabling growth in most (if not all) substrates, and value-added products that can be produced, the macroscopic mycelium, obtained following assimilation of nutrients and easily recovered from the medium, provides superior performance of filamentous fungi towards reduction of COD levels, in comparison to unicellular microorganisms (e.g., bacteria, yeasts, and microalgae) [14]. The use of the latter normally entails costly biomass recovery processes. Filamentous fungal biomass normally contains high protein levels, essential amino acids, polyunsaturated fatty acids, and its cell wall is rich in compounds with immunostimulant activity [15]. Therefore, it has increasingly been studied as a replacement of fish meal and soybeans in the production of animal feed, a product in increasing demand due to population growth. The growth of filamentous fungi in other fat-rich substrates such as dairy products [16][17][18] and fish processing-derived wastewaters [19,20], for protein recovery, has been reported in the literature. Research studies on protein recovery from OMWW using filamentous fungi, although scarce and more than eight years old, also exist [21][22][23]. However, these are characterized by low protein-containing products if the landscape of protein contents reported among fungal biomasses obtained from the growth in various low-value substrates is considered.
The aim of the present study was to develop a new cultivation strategy to enhance the protein level present in fungal biomass obtained from the cultivation of edible filamentous fungi in OMWW. Through nitrogen addition, and nitrogen addition and medium dilution, to lower the content of suspended solids, the protein content of Aspergillus oryzae biomass grown in OMWW, could be increased from w = 15.9% to w = 32.3% and w = 44.9%, respectively.

Filamentous Fungal Strains
Three edible filamentous fungi were used in this work, namely, Aspergillus oryzae var. oryzae CBS 819.72, isolated from tané koji used for sake making, Neurospora intermedia CBS 131.92, isolated from oncom, and Rhizopus delemar CBS 145940, isolated from leaves used in tempeh production (Centraalbureau voor Schimmelcultures, Utrecht, The Netherlands). The strains were maintained on PDA (Potato Dextrose Agar) medium plates containing 4 g/L potato infusion, 20 g/L glucose, and 15 g/L agar. New plates were prepared every 2-4 weeks via addition of 20 mL of sterile distilled water to pre-grown plates, where the spores were brought into solution using a L-shape sterile disposable plastic spreader. New plates were inoculated with 100 µL of spore solution, which was evenly distributed onto the agar surface with another plastic spreader, followed by incubation at 30 • C for three days and then stored at 4 • C until use.

Substrate
The olive oil mill wastewater (OMWW) was obtained from Dizem Olive Oil Factory (Çanakkale, Turkey). The steps leading to the production of OMWW are presented in Figure 1 and its composition provided in Table 1. The OMWW was heat-sterilized in an autoclave (VX-95, Systec, Linden, Germany) at 121 • C for 20 min, and then stored at 4 • C until use.

Cultivation in Shake-Flasks
Series of cultivations were carried out in 250 mL wide-neck Erlenmeyer flasks containing 50 mL of OMWW. The factors studied included fungal strain, initial pH (pH 5.2 vs. pH 6.5 adjusted with 2 mol/L NaOH, incubation time (varied within 24-120 h), OMWW dilution (diluted with ultrapure water to make up (20,40,60, and 80) % OMWW), nitrogen supplementation using NaNO 3 and NH 4 Cl, NaNO 3 addition mass concentration, and OMWW fraction. For the latter, OMWW was centrifuged (Heraeus Megafuge 8, Thermo Scientific, Waltham, MA, USA) at 4000× g for 10 min; the supernatant was used directly for fungal cultivation, while the same amount of ultrapure, as that of supernatant, was added to the wet cake, preceding pH adjustment to 5.2, sterilization, and fungal cultivation. Each flask was inoculated with 1 mL of spore solution, prepared as aforementioned, followed by incubation in water baths (Excella E24, New Brunswick Scientific, Nürtingen, Germany) at 35 • C with rotational frequency of 125 min −1 . No pH control and readjustment was carried out during cultivation. After the cultivation, the originated cultures were poured into a sieve (1 mm 2 pore area), and the recovered fungal biomass was thoroughly washed with distilled water for further oven drying. All cultivations were carried out in duplicate.

Analytical Methods
The harvested biomass samples were dried to constant weight in an oven at 70 • C. Total solids (TS), suspended solids (SS), dissolved solids (DS), and ash contents were determined according to Sluiter et al. [25] and Sluiter et al. [26]. Crude protein content of OMWW and fungal biomass were analyzed according to the Kjeldahl method [27]. To determine the crude protein content of the samples, a nitrogen-to-protein conversion factor of 6.25 was used [28]. The COD levels of both OMWW and the liquids remaining after fungal cultivation and fungal biomass separation were determined using a COD Kit (COD 1500, Nanocolor, Düren, Germany). Total fat contents were determined based on diethyl ether extraction method described in a previous work [19]. Alkali-insoluble material, representing the fungal cell wall fraction, was quantified according to a previously published method [29]. Sugars, organic acids, and ethanol levels were monitored using high-pressure liquid chromatography (HPLC) (Waters 2695, Waters Corporation, Milford, MA, USA) equipped with an analytical ion exchange column based on hydrogen ions (Aminex HPX-87H, BioRad, Hercules, CA, USA), operating at 60 • C and using 0.6 mL/min of 0.005 mol/L H 2 SO 4 as eluent, and a refractive index (RI) detector (Waters 2414, Waters Corporation, Milford, MA, USA).

Statistical Analysis
The software Minitab17 ® (Minitab Ltd., Coventry, UK) was used for the statistical analysis of the obtained results with ANOVA (analysis of variance) tables using general linear models. Pairwise comparisons among groups of data were also carried out according to the Tukey test. Significant differences were considered at p-value < 0.05 within a 95% confidence interval. All error bars and intervals presented represent two times the standard deviation.

Composition of Olive Oil Mill Wastewater (OMWW)
The chemical composition of OMWW is presented in Table 1. OMWW composition may vary according to olive type, season, and olive processing method [1]. The COD of the OMWW used in this work was found to be 102.5 g/L, being within the range, of 30-320 g/L, reported in the literature [1,5]. Fat, protein, sugars (mainly glucose), and ash were found to be the main compounds present in OMWW, where a C/N ratio of 22.21 was determined. Altogether, at a first glance, OMWW contained all needed nutrients to support fungal growth.

Production of Fungal Biomass
In the first approach, the three fungal strains were cultivated in undiluted OMWW for 96 h ( Figure 2). A. oryzae was found to lead to the highest biomass mass concentration, namely of 8.1 g/L, in comparison to that achieved with Neurospora intermedia (2.1 g/L) and Rhizopus delemar (0.49 g/L). Therefore, further studies, namely, at varying initial pH using undiluted OMWW (Figure 3a,b) and at varying initial pH and varying OMWW mass concentration (Figure 3c,d), were carried out using only A. oryzae. The cultivation of the ascomycete fungus in undiluted OMWW at varying initial pH (5.2 vs. 6.5) led to similar maximum biomass mass concentrations after 96 h of cultivation, where the initial pH generally did not influence the fungal growth rate (except when comparing the biomass mass concentration at 72 h of cultivation where statistically significant differences were found). However, the initial pH was found to play a role on final biomass mass concentration obtained after 96 h of cultivation at varying OMWW mass concentrations, where significantly higher biomass mass concentrations at 40% and 80% OMWW were obtained when the initial pH was 5.2. As it is presented in Figure 3c,d, gradually higher mass concentrations of OMWW did not impair the growth of A. oryzae. The biomass mass concentrations based on undiluted OMWW at 20%, 40%, 60%, 80%, and 100% were 8.55 ± 1.03 g/L, 9.33 ± 1.55 g/L, 7.95 ± 0.21 g/L, 9.88 ± 0.30 g/L, and 8.46 ± 0.51 g/L, respectively, which corroborates lack of impairment of fungal growth. Overall, the results show that neither chemicals, for pH adjustment, nor dilution, are needed while attaining the same cultivation output. Therefore, further studies were conducted using undiluted OMWW at its natural initial pH of 5.2. initial pH generally did not influence the fungal growth rate (except when comparing the biomass mass concentration at 72 h of cultivation where statistically significant differences were found). However, the initial pH was found to play a role on final biomass mass concentration obtained after 96 h of cultivation at varying OMWW mass concentrations, where significantly higher biomass mass concentrations at 40% and 80% OMWW were obtained when the initial pH was 5.2. As it is presented in Figure 3c,d, gradually higher mass concentrations of OMWW did not impair the growth of A. oryzae. The biomass mass concentrations based on undiluted OMWW at 20%, 40%, 60%, 80%, and 100% were 8.55 ± 1.03 g/L, 9.33 ± 1.55 g/L, 7.95 ± 0.21 g/L, 9.88 ± 0.30 g/L, and 8.46 ± 0.51 g/L, respectively, which corroborates lack of impairment of fungal growth. Overall, the results show that neither chemicals, for pH adjustment, nor dilution, are needed while attaining the same cultivation output. Therefore, further studies were conducted using undiluted OMWW at its natural initial pH of 5.2.  initial pH generally did not influence the fungal growth rate (except when comparing the biomass mass concentration at 72 h of cultivation where statistically significant differences were found). However, the initial pH was found to play a role on final biomass mass concentration obtained after 96 h of cultivation at varying OMWW mass concentrations, where significantly higher biomass mass concentrations at 40% and 80% OMWW were obtained when the initial pH was 5.2. As it is presented in Figure 3c,d, gradually higher mass concentrations of OMWW did not impair the growth of A. oryzae. The biomass mass concentrations based on undiluted OMWW at 20%, 40%, 60%, 80%, and 100% were 8.55 ± 1.03 g/L, 9.33 ± 1.55 g/L, 7.95 ± 0.21 g/L, 9.88 ± 0.30 g/L, and 8.46 ± 0.51 g/L, respectively, which corroborates lack of impairment of fungal growth. Overall, the results show that neither chemicals, for pH adjustment, nor dilution, are needed while attaining the same cultivation output. Therefore, further studies were conducted using undiluted OMWW at its natural initial pH of 5.2.   A previous study has shown that the addition of nitrogen, in the form of NH 4 Cl and NaNO 3 , to OMWW could increase the activity of lipase produced by A. oryzae strains [12]. Therefore, a similar strategy was used in this study (Figure 4), and its effect on biomass mass concentration and growth rate investigated. In the first approach, the initial nitrogen amount in OMWW of 1.32 g/L was doubled by addition of NaNO 3 and NH 4 Cl, representing the addition of 8.01 g/L and 5.04 g/L, respectively. Following a 48 h cultivation, the addition of NaNO 3 led to a significant increase in the mass concentration of biomass in comparison to both control cultivation and that where NH 4 Cl was added (Figure 4a). A further investigation was carried on the effect of NaNO 3 addition to reach a C/N ratio of 11.11 (8.01 g/L) and 5.55 (32.06 g/L), on the biomass mass concentration obtained following cultivation at 24, 48, and 96 h (Figure 4b). It was found that a C/N ratio of 11.11 was preferable for growth of A. oryzae in OMWW, which led to a reduction of the cultivation time needed from 96 h to 48 h, while achieving a similar maximum biomass mass concentration. The yield of biomass based on consumed COD was 0.19 g/g. A previous study has shown that the addition of nitrogen, in the form of NH4Cl and NaNO3, to OMWW could increase the activity of lipase produced by A. oryzae strains [12]. Therefore, a similar strategy was used in this study (Figure 4), and its effect on biomass mass concentration and growth rate investigated. In the first approach, the initial nitrogen amount in OMWW of 1.32 g/L was doubled by addition of NaNO3 and NH4Cl, representing the addition of 8.01 g/L and 5.04 g/L, respectively. Following a 48 h cultivation, the addition of NaNO3 led to a significant increase in the mass concentration of biomass in comparison to both control cultivation and that where NH4Cl was added (Figure 4a). A further investigation was carried on the effect of NaNO3 addition to reach a C/N ratio of 11.11 (8.01 g/L) and 5.55 (32.06 g/L), on the biomass mass concentration obtained following cultivation at 24, 48, and 96 h (Figure 4b). It was found that a C/N ratio of 11.11 was preferable for growth of A. oryzae in OMWW, which led to a reduction of the cultivation time needed from 96 h to 48 h, while achieving a similar maximum biomass mass concentration. The yield of biomass based on consumed COD was 0.19 g/g.

Composition of Fungal Biomass
In addition to the impact on fungal biomass production, the effect of different cultivation factors on fungal biomass composition was concomitantly monitored. While varying the initial pH ( Figure  5a,b), cultivation time (Figure 5a), and mass concentration of OMWW (Figure 5b), the protein content of the harvested biomass was w = 13.3-19.2%. However, when nitrogen was added to the medium (Figure 5c), fungal biomass containing up to w = 32.3 % protein was obtained. Therefore, the addition of nitrogen led to a decrease of the needed cultivation time and to an increase of 54% in protein content, in comparison to the condition without nitrogen addition, following a 48 h cultivation.
A more extensive compositional analysis of A. oryzae biomass, including fat, cell wall (alkaliinsoluble material), and minerals (ash), was carried out and the effect of nitrogen addition, OMWW mass concentration, and presence of suspended solids investigated (Figure 5c). A striking observation is that the amount of fungal biomass that could not be analyzed was the highest when undiluted OMWW or a derived solution based on only suspended solids diluted with ultrapure water, were used. Inversely, a more complete characterization of the biomass was acquired as the level of suspended solids, as a result of OMWW dilution, was carried out, reaching the highest level of

Composition of Fungal Biomass
In addition to the impact on fungal biomass production, the effect of different cultivation factors on fungal biomass composition was concomitantly monitored. While varying the initial pH (Figure 5a,b), cultivation time (Figure 5a), and mass concentration of OMWW (Figure 5b), the protein content of the harvested biomass was w = 13.3-19.2%. However, when nitrogen was added to the medium (Figure 5c), fungal biomass containing up to w = 32.3 % protein was obtained. Therefore, the addition of nitrogen led to a decrease of the needed cultivation time and to an increase of 54% in protein content, in comparison to the condition without nitrogen addition, following a 48 h cultivation. of 4.23 g/L biomass, while 7.58 g/L of fungal biomass were obtained following cultivation in the wet cake obtained after centrifugation and with added ultrapure water. Yields of fungal biomass per gram of dissolved solids (0.12-0.29 g/g), per gram of total solids (0.13-0.16 g/g), and per liter of undiluted OMWW (8.43-10.24 g/L) increased as OMWW mass concentration decreased. Altogether, the results of this study point out that optimization of C/N ratio and content of suspended solids can play a substantial role on production of fungal biomass as well as on protein recovery from OMWW.

Composition of Streams Remaining after Fungal Cultivation and Biomass Harvesting
Due to their growth in the form of a macroscopic filamentous structure easily recoverable from the medium, the use of filamentous fungi for bioconversion of low-value sidestreams can contribute to substantial COD reduction. The characterization of the streams left after 48 h or 96 h fungal A more extensive compositional analysis of A. oryzae biomass, including fat, cell wall (alkali-insoluble material), and minerals (ash), was carried out and the effect of nitrogen addition, OMWW mass concentration, and presence of suspended solids investigated (Figure 5c). A striking observation is that the amount of fungal biomass that could not be analyzed was the highest when undiluted OMWW or a derived solution based on only suspended solids diluted with ultrapure water, were used. Inversely, a more complete characterization of the biomass was acquired as the level of suspended solids, as a result of OMWW dilution, was carried out, reaching the highest level of characterization when only dissolved solids were used as cultivation medium. The latter cultivation condition also led to the highest protein content obtained in this work, namely, of w = 44.9%. The fungal cultivation in the supernatant obtained following OMWW centrifugation led to the production of 4.23 g/L biomass, while 7.58 g/L of fungal biomass were obtained following cultivation in the wet cake obtained after centrifugation and with added ultrapure water. Yields of fungal biomass per gram of dissolved solids (0.12-0.29 g/g), per gram of total solids (0.13-0.16 g/g), and per liter of undiluted OMWW (8.43-10.24 g/L) increased as OMWW mass concentration decreased. Altogether, the results of this study point out that optimization of C/N ratio and content of suspended solids can play a substantial role on production of fungal biomass as well as on protein recovery from OMWW.

Composition of Streams Remaining after Fungal Cultivation and Biomass Harvesting
Due to their growth in the form of a macroscopic filamentous structure easily recoverable from the medium, the use of filamentous fungi for bioconversion of low-value sidestreams can contribute to substantial COD reduction. The characterization of the streams left after 48 h or 96 h fungal cultivation and biomass separation at different conditions is presented in Table 2. The reduction range of total COD was 35-44%. Table 2. Characterization of the remaining streams following cultivation of A. oryzae and biomass separation. The conditions presented are the same as presented in Figure 5c.

Discussion
The world production of olives relies on a few concentrated areas, namely, the Mediterranean Sea Basin, Australia, China, and some countries in North (USA and Mexico) and South (Peru, Argentina, and Chile) America. For instance, it has been reported that 290,000 tons of olives were processed in 2017 in Jordan, producing 175,000 m 3 of OMWWs [30]. Therefore, considering the production of 3.12 million tons of olives in 2019/2020 as provided by Statista, 2.6 million m 3 of OMWWs are potentially being generated. The magnitude of the problem is exacerbated by the absence of (efficient) wastewater treatment processes in place [30]. Therefore, processes that can contribute to COD removal are urgently needed. Filamentous fungi are able to assimilate a wide range of compounds, both organic and inorganic, and after recovery of the mycelium, can highly contribute for COD removal with concomitant production of value-added products. Wastewater characteristics including mass concentration of inhibitory compounds, lack of nutrients, content of suspended solids, and pH can influence fungal growth and composition of the produced fungal biomass. Therefore, these aspects were investigated in this work.
In the first stage of the study, the ascomycete A. oryzae was found to be a more robust microorganism, in comparison to the ascomycete N. intermedia and the zygomycete R. delemar, for growth in OMWW, achieving 8.1 g/L of biomass within 96 h cultivation. In an another study investigating the production of lipase enzyme, it was reported that biomass weight of two different A. oryzae strains were 5.4-5.7 g/L after 168 h in OMWW containing a comparatively lower COD of 25.6-51.5 g/L [12]. The motivation behind the use of these three strains is related to the different characteristics they can provide to processes. For instance, N. intermedia is a superior ethanol producer than A. oryzae [31], while the cell wall composition of R. delemar, containing both chitin and chitosan, differs from that of the two ascomycetes, containing mostly chitin and no chitosan [32]. Although other cultivation parameters might have influenced the final output, several research studies have shown the superior performance of A. oryzae when its comparison with other strains, e.g., N. intermedia, was carried out. Examples include research works studying the growth on filamentous fungi in fish processing-derived wastewaters [19,20], vinasse [33], starch plant wastewater [34], and ethanol production-derived side streams [31].
Simple processing and extraction of value-added products can increase the feasibility of new bio-based processes with positive environmental impacts. Accordingly, the need of pH adjustment with addition of chemicals or of wastewater dilution increase the complexity of the process. The results from this study support the growth of A. oryzae directly in OMWW, where higher or comparable growth performance were obtained at initial pH of 5.2 in comparison to pH 6.5, and at varying mass concentrations of OMWW, respectively. A. oryzae lipase, produced during cultivation in fat-rich dairy products, was found to be significantly more active on fat degradation at pH >5.3 [16]; neutral pH conditions improved the enzymatic activity of Aspergillus species [35,36]; and a pH value of 6.5 was found to be suitable for growth of the ascomycete in vinasse [33]. However, adjusting the initial pH (6.1) of fish processing-derived wastewater to 5.0 led to a lower fungal growth performance [20]. Clearly, medium composition and physicochemical conditions play a role on fungal growth performance, and a tailored process development is needed while using wastewater of different origins [37]. OMWW can be toxic to microorganisms due to the presence of phenols and heavy metals. Inhibition of fungal growth has been reported in previous studies [37][38][39], and therefore OMWW or its phenols-rich extracts have been investigated as an antimicrobial agent. The growth of Aspergillus strains in undiluted OMWW, with or without nutrients addition has been reported for COD range of 25.6-123.7 g/L [12,40,41] pointing out the robustness of the genus for bioconversion of this wastewater.
Another factor during process development is productivity and, in this work, the cultivation time was reduced by 50% through addition of a nitrogen source. Similarly to the pH value for cultivation, the source of nitrogen should be tailored to the wastewater to go through bioconversion and, in the case of OMWW, it should also be tailored to the strain type and desired value-added product. For instance, NH 4 Cl was found to be a more suitable nitrogen source than (NH 4 ) 2 SO 4 , NaNO 3 , and urea for lipase production [41]. In addition to the need for finding the right nitrogen source, the results obtained point out the need for mixing different low-value substrates in order to attain satisfactory fungal growth and offset the need of addition of chemicals. Such strategy is commonly used in anaerobic digestion; however, it has not been so common in the research with filamentous fungi under submerged cultivation [14]. Co-cultivation strategies have, however, been applied in solid-state fermentation [23,42]. The first proposal for co-cultivation approaches is the addition of fish processing-derived wastewaters higher in protein and lower in carbohydrates [19,20]. The addition of yeast extract has also been reported to be needed to ensure fungal growth in OMWW [43], or leading to higher lipase production [41]. In order to enhance the feasibility of the process, it is proposed that yeast extract can be replaced by a filamentous fungal autolysate, obtained from fungal growth in low-value streams, as previously investigated [44]. Although in this study, the focus was on the effect of nitrogen addition, other types of nutrient unbalances might exist in OMWW and should be investigated in future studies.
Cultivation of A. oryzae in undiluted OMWW led to a protein content of w = 15.9%, which is in accordance with previous studies on growth of filamentous fungi in OWMM using both submerged and solid-state fermentation strategies. For instance, Hamdi et al. [21] have obtained a protein content range of w = 11.5-23% with Aspergillus niger, depending on OMWW composition and submerged cultivation time; Laconi et al. [22] have reported fungal biomass containing w = 13% protein following submerged growth of a fungal mixture in OWMM that had undergone an alkaline-oxidative treatment; and Giannoutsou et al. [23] have reported a product containing w = 19% protein following solid-state fermentation of Paecilomyces variotii on dried OMWW and molasses. Nonetheless, higher protein contents-within w = 42-58%-have been obtained following growth of, e.g., Aspergillus oryzae in various low-value substrates such as vinasse [33] and thin stillage [31,45]. Therefore, even after addition of nitrogen and significant improvement of the protein content of the fungal biomass, a plethora of research studies have reported fungal biomass containing higher protein contents. Several research works have pointed out that the macroscopic filamentous structure and consequent entanglement with suspended solids might be related to lower protein contents. Examples include the growth of filamentous fungi in fish processing-derived sidestreams [19,20], thin stillage [46], and potato starch wastewater [34]. This study was pioneer on scrutinizing the effect of the content of suspended solids on the final protein content of the fungal biomass, where the highest protein content of w = 44.9% was obtained following growth in the supernatant, being similar to that reported in the literature and the highest reported, to the best of our knowledge, for microbial biomass obtained from growth in OMWW.
The COD reduction, following fungal cultivation and biomass separation, was 35-44%. Such COD reduction range is comparable to those reported in the literature using OMWW. For instance, 50% COD reduction was obtained under repeated batch process of Pleurotus ostreatus in OMWW (94.8 g/L initial COD) [40]; 78% removal was obtained after acclimatization of filamentous fungi in OMWW (82 g/L initial COD) [47]; and 55% removal after 74 h of cultivation of Aspergillus niger in bubble column (123.7 g/L initial COD) [48]. These comparisons might point out that there is room for improvement on COD removal by A. oryzae. The wastewater remaining after fungal cultivation and biomass separation still contains high chemical oxygen demand entailing further processing such as through activated sludge process. Therefore, the effect of a fungal cultivation on downstream processing also needs to be investigated in future studies.
Altogether, based on the results obtained in this study, it is proposed that research interest on bioconversion of OMWW into feed protein and COD removal should be renewed. Through close control of cultivation medium composition, high protein-containing fungal biomass can be obtained applying short and simple bioprocessing. Further research works should, e.g., investigate alternative nitrogen sources, preferably those that remain as waste, and their impact on protein content and COD removal, or cultivate filamentous fungi in bioreactor to take advantage of continuous supply of air. The integration of cultivation of filamentous fungi with membrane bioreactors should also be investigated. Similarly to filamentous fungi, membrane bioreactors function as selective sieves and can attain separation of several value-added streams, such as phenols-rich stream, and high-cell mass concentrations conditions. Such a strategy has widely been investigated for detoxification of industrial wastewaters containing a wide range of micropollutants [14].

Conclusions
A. oryzae is a robust microorganism for conversion of OMWW into protein-rich fungal biomass with concomitant COD removal. The significant increase in the protein content of fungal biomass through simple nitrogen addition and OMWW phase-separation might represent a basis for renewed research interest in the use of filamentous fungi for management of OMWW. It is hypothesized that laying efforts in the development of bioprocesses for the management of OMWW can benefit from the clustered availability of OMWW in the world. Funding: This research was funded by the Swedish Agency for Economic and Regional Growth through a European Regional Development Fund.