Influence of Carbon Sources on the Phenolic Compound Production by Euglena gracilis Using an Untargeted Metabolomic Approach

Industrial development and urbanization has led to the diverse presence of metals in wastewater that are often improperly treated. The microalgae Euglena gracilis can tolerate high concentrations of metal via the excretion of organic metabolites, including phenolics. This study aims to evaluate how carbon amendment stimulates phenolic compound production by E. gracilis. The number, relative intensity and molecular composition of the phenolic compounds were significantly different between each of four carbon amended cultures (i.e., glutamic acid, malic acid, glucose, reduced glutathione) during the log phase. Phenolic compounds were mainly produced during the minimum growth rate, likely a response to stressful conditions. A better understanding of phenolic compounds production by E. gracilis and the impact of growth conditions will help identify conditions that favor certain phenolic compounds for dietary and metal chelation applications.


Introduction
Metal presence in aquatic ecosystems is an increasing global problem largely due to rapid urbanization and industrialization [1]. An estimated 40% of lakes and rivers has already been contaminated [2]. Metals can cause adverse and hazardous effects on living organisms, vegetation, biodiversity and human health [3,4]. Therefore, it is important to find a way to properly treat industrial and municipal effluents. Several physical, chemical, and biological methods have been developed to this end. One of them is phycoremediation, a form of bioremediation, using algal biomass producing chelating ligands able to recover significant amounts of metals [5].
Euglena gracilis is a unicellular aquatic organism part of the euglenoids family. This freefloating protist is able to produce energy under autotrophic, heterotrophic and mixotrophic conditions [6] by utilizing carbon from several sources [7,8]. Its metabolic flexibility suggests the existence of several physiological pathways affecting the biosynthesis and proportion of different metabolites [9]. Euglena gracilis also shows high adaptation capacities for a broad range of environmental conditions [10] through the plasticity of its atypical and unique metabolism [6,11], including its ability to tolerate high concentrations of metals (ppm range) [12] via the production of chelating ligands [13]. These studies focused mainly on primary metabolites production, but little is known in terms of secondary metabolites [14]. Secondary metabolites are specialized compounds produced in response to environmental changes [15]. They play a key role in the defense mechanism against abiotic and biotic stress via their anti-allergenic, anti-microbial, anti-inflammatory and antioxidant properties [16,17]. For example, oxidative stress led to the formation of oxidative compounds and free radicals, and the breakdown of photosynthetic and metabolic enzymes [14,18]. Phenolic compounds are a major class of secondary metabolites and act as reducing agents and hydrogen donors to minimize oxidative stress [19,20]. In addition, phenolic compounds show an excellent metal chelating potential due to the abundance of oxygen functional groups in their structure [21,22]. These specialized metabolites are synthesized because of interactions with the environment [23] and can reflect evolutive environmental conditions, such as nutrient availability or abiotic stress presence. Altered growth and medium conditions can promote the production of specific metabolites [18,24,25]. For example, metabolites with carboxyl functional groups were more abundant in Bacillus subtilis cells harvested during the exponential phase than during the stationary phase [26]. Other studies exploring the effects of changing growth light regimes found a significant effect on metabolite composition and the production of metal binding ligands [27,28]. Few reports have focused on phenolic compounds and their bioactive functions [29] but their molecular composition remains largely unknown.
The aim of this study was to assess how the production of phenolic compounds by Euglena gracilis cultures is influenced by the carbon source (i.e., glutamic acid, malic acid, glucose or reduced glutathione). Earlier studies [30][31][32][33] have shown the direct and important impact of growth conditions, as variable organic carbon sources have an impact on secondary metabolite production. The most common analytical method for the quantification and characterization of phenolic compounds is UV-visible spectrophotometry using the Folin-Ciocalteu (FC) reagent method [34]. The FC assay is a rapid, repeatable method to measure total phenolic content and oxidative capacity of algal cells, but phenolics, proteins and thiols are also reactive to the FC reagent [35,36]. High performance liquid chromatography (HPLC) combined with mass spectrometry (MS) has been shown to be suitable for the detection of phenolic compounds [37,38]. Here we describe an untargeted metabolomic approach to unravel, for the first time, the molecular composition of the cellular phenolic compounds produced by Euglena gracilis as a response to different carbon sources.

Algal Growth and Biomass Harvesting
Euglena gracilis culture was obtained from the Canadian Phycological Culture Centre (CPCC; University of Waterloo, Waterloo, ON, Canada) and grown in modified acid medium (MAM) [39] at pH 3 and supplemented with one simple carbon source, i.e., glutamic acid, malic acid, glucose or reduced glutathione (GSH) at a concentration of 5 g L −1 . Cultures were grown in a pre-combusted 1L Erlenmeyer flask and maintained under photoautotrophic conditions with an alternating light:dark cycle of 16:8 h. The light intensity was fixed at 150 µmol photons m −2 s −1 and the temperature kept between 20-25 • C. Previous studies [40][41][42] have shown that the addition of simple carbon sources in photoautotrophic conditions allowed faster growth. Having a greater biomass was crucial to carry out an in-depth study of the molecular characterization of phenolic compounds produced by E. gracilis. The cell concentration was counted using a Leica DM500 light microscope and a hemacytometer. The initial cell density of each culture was 1.0 × 10 6 cell mL −1 , and the biomass was harvested on day 1, 3, 5, 6, 8 and 10 of growth. The cell pellet was washed 3 times with milliQ water and stored freeze-dried until further analysis.

Antioxidant Activity Assay
The antioxidant activity was measured with the FC assay [43]. Briefly, 10 mg of dry biomass was ground and sonicated for 1 h in 85% methanol then centrifuged for 30 min at 4900 rpm. The FC reagent (1:10) was added to the sample extract or standard. After a 5 min incubation at room temperature, 1.6 mL 1 M Na 2 CO 3 was added, and the reaction mixture incubated for 2 h in the dark at room temperature. The absorbance was measured at 760 nm with a spectrophotometer (Shimadzu UV-1800). The antioxidant activity was expressed in gallic acid equivalent (mg g −1 dry weight) using a calibration curve ranging from 0 to 62.5 µg mL −1 .

Mass Spectrometry Analysis
The metabolites associated with each of the five LC fractions were analyzed using a MAXIS time-of-flight mass spectrometer (qTOF; Bruker Daltonics, Bremen, Germany) equipped with an electrospray ionization (ESI) source in negative ionization mode. The samples were directly infused at a flow rate of 300 µL h −1 (Lewis et al., 2021). The signal was acquired for 2 min across a m/z range of 100-1000. The source capillary voltage was set to 5000 V with a nitrogen gas flow rate of 4 L min −1 and a capillary temperature of 180 • C. An external standard (sodium formate) was used to ensure a good calibration from day to day. MS grade methanol was infused between each sample to minimized sample carry over and contamination. A blank sample was also acquired every 10 samples and all m/z found in the blank were removed from sample spectra to remove background. Spectra processing was performed using Compass DataAnalysis (v4.4, Bruker Daltonics, Bremen, Germany) where a mass list was generated at S/N > 4 and a relative intensity cut-off of 0.1%. The elemental composition was then attributed using SmartFormula (DataAnalysis v4.4) with the following criteria: 12 C(1-50), 1 H(1-100), 16 O(1-30), 14 N(0-2), 32 S(0-2) within 10 ppm mass error. We confirmed the molecular formulas for the most abundant peaks with the help of the signal intensities of 13 C 1 12 C n−1 compounds. Only the m/z present in replicate samples were kept for phenolic analysis. Here, phenolic compounds were defined as 0.6 < H/C < 1.5 and 0.3 < O/C < 0.85 [46,47] and a modified aromaticity index (AImod) inferior to 0.67 [48].

Statistical Analysis
The different growth phases were identified for each carbon source using a curve fitting model (GraphPad Prism 9). Each carbon source had 4 days corresponding to the log phase except for glucose, which had only 3 days. During those days, we were able to identify the maximum slope and the minimum slope using the growth rate equation for a population as follows: A Shapiro-Wilk normality test was performed followed by the Wilcoxon nonparametric test (R studio). A significant difference was considered at the level of p < 0.05.

Growth Curve and Carbon Sources
The MAM medium was supplemented with simple carbon sources to achieve faster growth and a higher cell density relative to inorganic media ( Figure S1). Each growth medium presented a different profile because each carbon source is metabolized differently by E. gracilis favoring specific and variable metabolic pathways that impact on growth. The length of the log phase was also dependent on the carbon source ( Figure 1). Glutamic acid and its amine group supports the growth of cells via its ability to biosynthesize amino acids and nucleic acids. Glucose is the main carbon source in cells and the elemental metabolite for glycolysis, which helps produce energy and many molecules of biological interest. Malic acid is involved in the tricarboxylic acid cycle (TAC) [49], which is responsible for the majority of energy production in cells. These three carbon sources have a direct impact on energy production and growth as depicted by a significantly higher biomass compared to the inorganic amended culture. In comparison, the GSH amendment showed a smaller increase in cell density compared to the other C sources but it was higher than MAM alone. GSH is a major endogenous antioxidant responsible for the redox balance management and is involved in the detoxification of xenobiotic and endogenous compounds in cells [50].

Growth Curve and Carbon Sources
The MAM medium was supplemented with simple carbon sources to ach growth and a higher cell density relative to inorganic media ( Figure S1). Each g dium presented a different profile because each carbon source is metabolized by E. gracilis favoring specific and variable metabolic pathways that impact o The length of the log phase was also dependent on the carbon source ( Figure 1 acid and its amine group supports the growth of cells via its ability to biosynthe acids and nucleic acids. Glucose is the main carbon source in cells and the elem tabolite for glycolysis, which helps produce energy and many molecules of bi terest. Malic acid is involved in the tricarboxylic acid cycle (TAC) [49], which i ble for the majority of energy production in cells. These three carbon sources ha impact on energy production and growth as depicted by a significantly high compared to the inorganic amended culture. In comparison, the GSH amendme a smaller increase in cell density compared to the other C sources but it was h MAM alone. GSH is a major endogenous antioxidant responsible for the red management and is involved in the detoxification of xenobiotic and endoge pounds in cells [50].

Phenolic Concentration Based on the Folin-Ciocalteu Assay
The concentration in gallic acid equivalent (GAE) of the extracted phe pounds varied from 2.15 to 2.46 mg g −1 and from 1.44 to 2.11 mg g −1 in the me ethanol extracts, respectively ( Figure S2). More phenolic compounds were fo methanol extracts than in the ethanol extracts (p < 0.05), suggesting that the phe pounds in E. gracilis were highly hydrophilic. Previous studies reported bette extraction with organic aqueous mixtures [51] and a better extraction yield us nol [50]. Overall, both the 50% and 80% methanol extracts showed the highest tions with values of 2.39 ± 0.04 and 2.46 ± 0.02 mg g −1 , respectively. Further ana performed using the 80% methanol extracts.

Phenolic Concentration Based on the Folin-Ciocalteu Assay
The concentration in gallic acid equivalent (GAE) of the extracted phenolic compounds varied from 2.15 to 2.46 mg g −1 and from 1.44 to 2.11 mg g −1 in the methanol and ethanol extracts, respectively ( Figure S2). More phenolic compounds were found in the methanol extracts than in the ethanol extracts (p < 0.05), suggesting that the phenolic compounds in E. gracilis were highly hydrophilic. Previous studies reported better phenolic extraction with organic aqueous mixtures [51] and a better extraction yield using methanol [50]. Overall, both the 50% and 80% methanol extracts showed the highest concentrations with values of 2.39 ± 0.04 and 2.46 ± 0.02 mg g −1 , respectively. Further analyses were performed using the 80% methanol extracts. The mean antioxidant capacity using the FC assay ranged from 4.04 ± 0.52 mg g −1 with malic acid amendment to 6.08 ± 0.73 mg g −1 with GSH amendment (Figure 2). Comparable values have been previously reported for Euglena gracilis and other microalgae [52,53]. The mean antioxidant capacity was 1.2-1.5 times higher with the sulfur-rich amendment (GSH amendment) than with any other carbon amendments. This result was expected as sulfur has a good antioxidant capacity [54].
lecules 2022, 12, x FOR PEER REVIEW The mean antioxidant capacity using the FC assay ranged fro with malic acid amendment to 6.08 ± 0.73 mg g −1 with GSH amendm parable values have been previously reported for Euglena gracilis a [52,53]. The mean antioxidant capacity was 1.2-1.5 times higher amendment (GSH amendment) than with any other carbon amendm expected as sulfur has a good antioxidant capacity [54].

Phenolic Compound Molecular Characterization
The reverse-phase HPLC method presented good resolution an the 9 different standards tested (Table S1; Figure S3) and good linea tration range (12.5-200 µM; Table S1). None of the pure standards w samples isolated in this study ( Figure S4). The elemental ratios (O/C fied aromaticity index (AImod), double bond equivalent (DBE), an tion state of carbon (NOSC) were calculated (Table S2) for the five fr S4) obtained from chromatographic separation of the cellular ext trends were observed between fractions, highlighting the heteroge compounds isolated in the chromatographic fractions. The molecu performed on the combined fraction only (F1 + F2 + F3 + F4 + F5).
The number of phenolic compounds varied greatly from day t phase between C sources ( Figure 3A-D). A range of 6-31 different m nolics were found with no significant differences between carbon so relative intensities varied between C sources (0.057-0.296; Figure 3Eacid amendment showing the greatest variability.

Phenolic Compound Molecular Characterization
The reverse-phase HPLC method presented good resolution and reproducibility for the 9 different standards tested (Table S1; Figure S3) and good linearity over the concentration range (12.5-200 µM; Table S1). None of the pure standards was found in the algal samples isolated in this study ( Figure S4). The elemental ratios (O/C and H/C), the modified aromaticity index (AImod), double bond equivalent (DBE), and the nominal oxidation state of carbon (NOSC) were calculated (Table S2) for the five fractions F1-F5 ( Figure S4) obtained from chromatographic separation of the cellular extracts. No significant trends were observed between fractions, highlighting the heterogeneity of the phenolic compounds isolated in the chromatographic fractions. The molecular composition was performed on the combined fraction only (F1 + F2 + F3 + F4 + F5).
The number of phenolic compounds varied greatly from day to day during the log phase between C sources ( Figure 3A-D). A range of 6-31 different m/z attributable to phenolics were found with no significant differences between carbon sources (p > 0.05). Their relative intensities varied between C sources (0.057-0.296; Figure 3E

Impacts on Phenolic Production
Significant differences in phenolic production were found, depending on the cellular growth rate. The lowest number and relative intensity of phenolic compounds in the glutamic acid amendment ( Figure 3A,E) were found on day 3 of growth, which represents the middle of the log phase and the start of the maximum growth rate ( Figure 1A). Glutamic acid supports cell growth by producing nitrogen-containing compounds. The variation in phenolics and the minimum production of phenolic compounds when cell division is the most important may imply that glutamic acid is used preferentially for other essential biological functions when the cell division rate is maximal and is more favorable for phenolic production when the cell is less metabolically active. The glucose amendment ( Figure 3B,F) showed a relatively similar pattern, with a decrease in the number of phenolic compounds as the growth curve slope steepened. Glucose is used directly by the cells to produce energy and different biological compounds. Although glucose is involved in the production of phenolic compounds via the synthesis of phosphoenolpyruvate [55], it has many additional functions. The malic acid amendment ( Figure 3C,G) had the lowest overall relative intensity relative to the other carbon sources, likely due to its involvement in the TAC and the production of adenosine triphosphate (ATP). The cellular energy demand throughout the log phase is relatively steady, which means that malic acid was con-

Impacts on Phenolic Production
Significant differences in phenolic production were found, depending on the cellular growth rate. The lowest number and relative intensity of phenolic compounds in the glutamic acid amendment ( Figure 3A,E) were found on day 3 of growth, which represents the middle of the log phase and the start of the maximum growth rate ( Figure 1A). Glutamic acid supports cell growth by producing nitrogen-containing compounds. The variation in phenolics and the minimum production of phenolic compounds when cell division is the most important may imply that glutamic acid is used preferentially for other essential biological functions when the cell division rate is maximal and is more favorable for phenolic production when the cell is less metabolically active. The glucose amendment ( Figure 3B,F) showed a relatively similar pattern, with a decrease in the number of phenolic compounds as the growth curve slope steepened. Glucose is used directly by the cells to produce energy and different biological compounds. Although glucose is involved in the production of phenolic compounds via the synthesis of phosphoenolpyruvate [55], it has many additional functions. The malic acid amendment ( Figure 3C,G) had the lowest overall relative intensity relative to the other carbon sources, likely due to its involvement in the TAC and the production of adenosine triphosphate (ATP). The cellular energy demand throughout the log phase is relatively steady, which means that malic acid was constantly used to provide energy to support growth rather than to produce phenolic compounds. The small overall increase in phenolic compounds production during the log phase may be explained by ATP production leading to the activation of different secondary metabolic pathways, including those for the synthesis of phenolic compounds. The phenolic compounds in the GSH amendment ( Figure 3D,H) were maximal at midexponential phase. Metabolic waste and toxins accumulate as the log phase progresses, and since the GSH is used for detoxification purposes [50], this can explain the diminution at the end of the growth phase. The relative phenolic intensity decreased as the growth slope slightly increases with the glutamic acid, malic acid, and glucose amendments.
The phenolic compounds produced during the log phase changed rapidly from day to day. Up to 11 common m/z (corresponding to 9-24% of assigned m/z) were found between two consecutive days in each carbon source ( Figure S5), highlighting the rapid transformation of phenolic compounds produced by E. gracilis. Most of the assigned formulas were unique to a carbon source, indicating a significant impact of the carbon source on the phenolic compounds profile.
Phenolic compounds with significantly higher m/z were found at the end of the log phase in comparison to the beginning for glucose and GSH ( Figure S6). This is consistent with the fact that the cell is the most metabolically active during this growth stage and has enough time and resources to biosynthesize more complex molecules, which often requires more energy and more sophisticated biochemical pathways.

Impacts on Phenolic Molecular Composition
The growth rates showed significant differences in the phenolic profiles ( Figure S7). The maximum growth rate was characterized by a lower number and relative intensity of phenolic compounds for each carbon amendment except for the number of phenolic compounds in the glucose amendment. Phenolic compounds have antioxidant properties, and as such, are more prone to be produced when the cells are stressed. During the maximum growth rate, the cells are under optimal conditions for metabolic activity and growth, in agreement with the observed reduction in the production of phenolic compounds.
The elemental composition of the phenolic compounds showed significant differences between minimum and maximum growth rates (Figure 4). The elemental composition for each carbon source was more balanced during the minimum growth rate with a particularly comparable proportion of CHO (26-36%), CHON (26-29%) and CHOS (12-26%) ( Figure 4A). These results contrast with the maximum growth rates where the elemental composition was more variable ( Figure 4B). A certain level of stress was present during the minimum growth rate that can influence the production of phenolic compounds which can cope with the stress. In contrast, during the days corresponding to the maximum growth rate, the stress was minimal, and the carbon source was used to support cellular growth and metabolic activity, which greatly affects the phenolic elemental profiles.
Biomolecules 2022, 12, x FOR PEER REVIEW 7 of 11 stantly used to provide energy to support growth rather than to produce phenolic compounds. The small overall increase in phenolic compounds production during the log phase may be explained by ATP production leading to the activation of different secondary metabolic pathways, including those for the synthesis of phenolic compounds. The phenolic compounds in the GSH amendment ( Figure 3D,H) were maximal at mid-exponential phase. Metabolic waste and toxins accumulate as the log phase progresses, and since the GSH is used for detoxification purposes [50], this can explain the diminution at the end of the growth phase. The relative phenolic intensity decreased as the growth slope slightly increases with the glutamic acid, malic acid, and glucose amendments. The phenolic compounds produced during the log phase changed rapidly from day to day. Up to 11 common m/z (corresponding to 9-24% of assigned m/z) were found between two consecutive days in each carbon source ( Figure S5), highlighting the rapid transformation of phenolic compounds produced by E. gracilis. Most of the assigned formulas were unique to a carbon source, indicating a significant impact of the carbon source on the phenolic compounds profile.
Phenolic compounds with significantly higher m/z were found at the end of the log phase in comparison to the beginning for glucose and GSH ( Figure S6). This is consistent with the fact that the cell is the most metabolically active during this growth stage and has enough time and resources to biosynthesize more complex molecules, which often requires more energy and more sophisticated biochemical pathways.

Impacts on Phenolic Molecular Composition
The growth rates showed significant differences in the phenolic profiles ( Figure S7). The maximum growth rate was characterized by a lower number and relative intensity of phenolic compounds for each carbon amendment except for the number of phenolic compounds in the glucose amendment. Phenolic compounds have antioxidant properties, and as such, are more prone to be produced when the cells are stressed. During the maximum growth rate, the cells are under optimal conditions for metabolic activity and growth, in agreement with the observed reduction in the production of phenolic compounds.
The elemental composition of the phenolic compounds showed significant differences between minimum and maximum growth rates (Figure 4). The elemental composition for each carbon source was more balanced during the minimum growth rate with a particularly comparable proportion of CHO (26-36%), CHON (26-29%) and CHOS (12-26%) ( Figure 4A). These results contrast with the maximum growth rates where the elemental composition was more variable ( Figure 4B). A certain level of stress was present during the minimum growth rate that can influence the production of phenolic compounds which can cope with the stress. In contrast, during the days corresponding to the maximum growth rate, the stress was minimal, and the carbon source was used to support cellular growth and metabolic activity, which greatly affects the phenolic elemental profiles.  Malic acid contains no nitrogen or sulfur, consistent with the very low CHON (22%), CHOS (6%) and CHONS (19%) abundances in the produced phenolics with the malic acid amendment. The abundance of CHO phenolic compounds accounted for 53% of the phenolic compounds. The presence of sulfur in GSH was translated into S-rich phenolic compounds. Up to 71% of the phenolic compounds identified contained sulfur ( Figure 4B). The elemental composition of the phenolic compounds produced with the glutamic acid and glucose amendment were comparable, with CHO dominating the phenolic formula at both growth rates (p > 0.05; 26-36%; Figure 4B). No distinct heteroatom rich phenolic compounds were found in the glucose amendment, likely due to the fact that glucose is the primary energy source for cellular activity and is used to produce a great variety of biological compounds [56]. A similar explanation can justify the absence of a distinct profile with glutamic acid treatment, since it is used for a great diversity of N-rich molecules, including nucleic acid and amino acid biosynthesis [57]. The glutamic acid treatment did not show a higher abundance of CHON and CHONS since nitrogen is mainly used in the production of biomass [58]. Together these results showed that the number, the intensity, and the elemental composition of cellular phenolic compounds were dependent on the growth rate of E. gracilis and the carbon source present.

Conclusions
This study focuses on the influence of four simple carbon additions on the production of phenolic compounds by Euglena gracilis cells using an untargeted metabolomic approach. The antioxidant capacity of E. gracilis cellular extracts determined by Folin-Ciocalteu assay was highly variable between carbon sources during the log growth phase. The cellular phenolic compounds were highly diversified in number, relative intensity and molecular composition between the four carbon treatments and during the log phase of E. gracilis growth. The minimum growth rate period showed up to a 3-fold increase in cellular phenolic production compared to the optimal growth rate, suggesting that cellular phenolic compounds were mostly produced to mitigate stressful conditions. The predominance of sulfur-and nitrogen-rich phenolic compounds was 7-14% higher during the minimum growth rate. Favorable conditions for the production of phenolic compounds could be particularly advantageous in dietary supplement and metal chelation applications.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/biom12060795/s1, Figure S1: Growth curves in inorganic medium supplemented with carbon sources; Figure S2: Concentrations of gallic acid equivalent (GAE) found in methanol and ethanol extracts; Figure S3: Chromatogram of the phenolic standards; Figure S4: Chromatogram of a cellular extract on day 3 of growth in de MAM medium supplemented with GSH; Figure S5: Venn diagrams of the molecular formulas attributed for each day of the lag phase in the (A) glutamic acid, (B) malic acid, (C) glucose, and (D) GSH amendment; Figure S6: Box plot of all the m/z found at the start and at the end of the log phase in (A) glutamic acid, (B) glucose, (C) malic acid, and (D) GSH amendment; Figure S7: (A) Number of polyphenols and (B) their relative intensities associated with the minimum and maximum growth rates; Table S1: Retention time and linearity of the phenolic standards; Table S2

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.