**Exploring the Antibacterial Activity of** *Pestalotiopsis* **spp. under Di**ff**erent Culture Conditions and Their Chemical Diversity Using LC–ESI–Q–TOF–MS**

**Madelaine M. Aguilar-Pérez 1,**† **, Daniel Torres-Mendoza 1,2,**† **, Roger Vásquez <sup>1</sup> , Nivia Rios <sup>3</sup> and Luis Cubilla-Rios 1,\***


Received: 16 July 2020; Accepted: 17 August 2020; Published: 19 August 2020

**Abstract:** As a result of the capability of fungi to respond to culture conditions, we aimed to explore and compare the antibacterial activity and chemical diversity of two endophytic fungi isolated from *Hyptis dilatata* and cultured under different conditions by the addition of chemical elicitors, changes in the pH, and different incubation temperatures. Seventeen extracts were obtained from both *Pestalotiopsis mangiferae* (*man-1* to *man-17*) and *Pestalotiopsis microspora* (*mic-1* to *mic-17*) and were tested against a panel of pathogenic bacteria. Seven extracts from *P. mangiferae* and four extracts from *P. microspora* showed antibacterial activity; while some of these extracts displayed a high-level of selectivity and a broad-spectrum of activity, *Pseudomonas aeruginosa* was the most inhibited microorganism and was selected to determine the minimal inhibitory concentration (MIC). The MIC was determined for extracts *man-6* (0.11 µg/mL) and *mic-9* (0.56 µg/mL). Three active extracts obtained from *P. mangiferae* were analyzed by Liquid Chromatography-Electrospray Ionization-Quadrupole-Time of Flight-Mass Spectrometry (LC–ESI–Q–TOF–MS) to explore the chemical diversity and the variations in the composition. This allows us to propose structures for some of the determined molecular formulas, including the previously reported mangiferaelactone (**1**), an antibacterial compound.

**Keywords:** Endophytic fungi; *Hyptis dilatata*; *Pestalotiopsis mangiferae*; *Pestalotiopsis microspora*; chemical elicitors; antibacterial activity; LC–ESI–Q–TOF–MS

#### **1. Introduction**

The World Health Organization (WHO; Geneva, Switzerland) has established an urgent pathogen list of antibiotic-resistant bacteria to guide the research, discovery, and development of new antibiotics. This list includes carbapenem-resistant *Pseudomonas aeruginosa* and *Enterobacteriaceae* and third generation cephalosporin-resistant bacteria as critical priorities as a result of the continuous and indiscriminate use of antibiotics, not only in the treatment of human diseases, but also in animals [1,2]. This list includes antifungal compounds [3].

Pharmaceutical conglomerates have abandoned this field of research due to the high costs. Despite the efforts made in recent years, e.g., investment in research and development (R&D) as well as in scientific and technological research, the strategies for the search of new antibiotics and antifungals

remain uncertain [4,5]. In this context, natural products produced by endophytic fungi provide an alternative to supply new molecules with antimicrobial activities [6–9].

Endophytic fungi spend a large part of their life cycle inside the tissue of the host organism without causing apparent damage [10]. In the last 15 years, interest in endophytic fungi has grown exponentially because of their ability to produce a wide range of secondary metabolites with diverse and important biological activities. Plant endophytes are considered one of the least studied groups of microorganisms and have proven to be a source of natural products and therefore provide a way to discover novel compounds with biological activities [11,12]. Two species of the genus *Pestalotiopsis*, isolated from *Hyptis dilatata* (Labiatae), a plant that is distributed in the north and east of the Republic of Panama and that is known for producing abietane and pimarane diterpenes [13] were selected for these studies.

The genus *Pestalotiopsis* is considered a vast source of natural products from which more than 300 compounds have been isolated and characterized, including terpenoids, polyketides, chromones, quinones, coumarins, lactones, and nitrogen-containing molecules with a wide range of biological activities such as antifungal, antibacterial, anticancer, antioxidant, antiparasitic, antihypertensive, anti-inflammatory, and neuroprotective activities [14–19]. Previously, our group reported the isolation of a set of eleven compounds (see Figure S1) from the crude extract of *P. mangiferae*, including a polyhydroxylated macrolide named mangiferaelactone [20]; the crude extract showed growth inhibition against *Listeria monocytogenes* (29 mm diameter inhibition zone), and showed a minimal inhibitory concentration (MIC) of 1.69 mg/mL and 0.55 mg/mL against *L. monocytogenes* and *Bacillus cereus*, respectively. This compound belongs to the nonalide class that is associated with important biological activities such anticancer, antifungal, antibacterial, and antiviral activities. Its synthesis has been developed by different research groups [21–23]. Compounds such as taxol, isopestacin, and some polyketides have been isolated from *P. microspora* [24–27].

The production of secondary metabolites by microorganisms could be impacted by environmental dynamics, such as growing conditions, which include biotic and abiotic factors [28]. Therefore, the selective variation of these parameters during the cultivation of fungi [29,30] and/or the induction of stress through competition with other microorganisms in a co-culture represent interesting ways to generate greater activity, chemical diversity, and novel active molecules [31–33]. Hence, the opportunity to modify culture conditions allows for the optimization of secondary metabolite production [34]. Therefore, given the growing interest in enhancing the production of secondary metabolites by endophytic fungi, the study of the methods and strategies to stimulate the gene clusters responsible for the biosynthesis of new molecules has been intensified and could include chemical or physical factors [35,36]. For example, the use of metallic ions, organic and inorganic compounds, pH, and incubation temperature to optimize the production of enzymes or secondary metabolites have been described [37–47].

For this study, we focused on modifying the conditions of the culture medium by varying abiotic parameters and through this, activate fungal silent gene clusters [48–50] in *P. mangiferae* Hd08 and *P. microspora* Hd18 in order to increase the chemical diversity and to detect new antibacterial activities.

#### **2. Materials and Methods**

#### *2.1. Chemicals and Reagents*

All of the following chemicals were acquired from Sigma–Aldrich® (Sigma–Aldrich, St. Louis, MO, USA): arginine, glutamic acid, CuSO4, CaCl2, FeSO4, tri-sodium citrate dihydrate, dimethyl sulfoxide (DMSO), and formic acid (FA). Ethyl acetate, acetone, and methanol used for extraction were American Chemical Society grade (Tedia®, Tedia Company Inc., Fairfield, OH, USA). The methanol for the liquid chromatography-mass spectrometry (LC–MS) analysis was LC–MS grade (J. T. Baker®, Avantor Performance Materials, Inc., Center Valley, PA, USA).

#### *2.2. Isolation and Identification of Fungal Isolates*

A healthy specimen of *Hyptis dilatata* (Labiatae) was collected in La Mesa, Veraguas Province in the Republic of Panama, in November 2010. An exsiccate from the plant material was deposited in the Herbarium of the University of Panama (PMA 084861). Mature leaves were surface sterilized as we previously reported in [51], and small fragments were cultured on 2% malt-extract agar (MEA; DifcoTM, Becton, Dickinson and Co., Sparks, MD, USA) under sterile conditions. Strains Hd08 and Hd18 were further isolated from the collection plate and successively re-plated until pure strains were obtained. Pure fungal strains were stored at −80 ◦C in a cryoprotectant solution of 10% glycerol and were preserved in the collection of the International Cooperative Biodiversity Group (ICBG) at the University of Panama. The identification of endophytic fungi was carried out as described previously [20]. Briefly, the total genomic DNA of each strain was isolated from fresh mycelium following U'Ren et al. [52]. Polymerase chain reaction (PCR) was used to amplify the nuclear ribosomal internal transcribed spacers and 5.8s gene (ITS rDNA; ca. 600 bp), and the first ca. 600 bp was sequenced bidirectionally [52]. The entire sequences for each strain were compared to the nucleotide database of the National Center for Biotechnology Information (NCBI) using the Basic Local Alignment Search Tools (BLAST®) Website.

#### *2.3. Media Preparation and Cultivation of Fungal Strains*

Strains Hd08 and Hd18 were reactivated aseptically on Petri dishes containing potato dextrose agar (PDA; DifcoTM, Becton, Dickinson and Co., Sparks, MD, USA) and incubated at 26 ◦C for seven days. Then, the mycelium was removed using a sterile spatula and was placed in sterile water to obtain a homogeneous solution. This solution was poured on the surface of Petri dishes (145 × 90 mm) containing MEA for all of the experimental conditions. After 15 days of incubation, the material was extracted, and the amount of crude organic extract was measured. Sterile controls were established for all of the experiments.

Chemicals as elicitors. MEA was prepared as indicated on the label; the resulting pH was measured using a pH meter (Thomas Scientific, Swedesboro, NJ, USA). The medium was then sterilized at 121 ◦C. When it had cooled to 45 ◦C, each of the elicitors was added, mixed well, and poured on Petri dishes (145 × 90 mm), and the strains then were inoculated and incubated at 26 ◦C for 15 days.

pH as an elicitor. MEA was prepared as described above and buffered with a 50 mmol/L tri-sodium citrate dihydrate solution to set values of 4.0, 4.6, and 5.6. The chemical elicitors were CaCl<sup>2</sup> and CuSO4. The strains then were inoculated, and the plates were incubated at 26 ◦C for 15 days.

Incubation temperature as an elicitor. MEA was prepared and buffered at pH = 4.0 by adding CaCl<sup>2</sup> or CuSO4, as described above. The incubation temperatures were set at 24, 28, and 30 ± 2 ◦C in the incubation chamber (Sheldon Manufacturing, Inc., Cornelius, OR, USA) for 15 days.

#### *2.4. Extraction and Sample Preparation*

In all of the experiments, after the incubation time, the mycelium was cut into small pieces and placed into a 1 L beaker and 500 mL of ethyl acetate was added. After 30 min, the mixture was triturated and homogenized using a Polytron® (Brinkmann Instruments, Westbury, NJ, USA) and subsequently filtered through filter paper (Whatman No. 7) using a vacuum. The organic solvent was evaporated under a vacuum at 30 ◦C using a rotary evaporator. The resulting crude extract was re-dissolved in acetone and transferred to scintillation vials, which were previously labeled and weighed, and then evaporated on a Speed Vac® Plus (Thermo SavantTM, Thermo Fisher Scientific, Waltham, MA, USA) for 24 h. Thereafter, the amount of extract was determined.

#### *2.5. Antibacterial Activity*

Tested microorganisms: Among the microorganisms used for the antimicrobial test, eight strains (*Bacillus cereus* CECT 5050, *Escherichia coli* CECT 433, *Kocuria rhizophila* CECT 241, *Legionella pneumophila* CECT 7109, *Listeria monocytogenes* CECT 935, *Pasteurella multocida* CECT 962, *Salmonella enterica* CECT 7161, *Salmonela enterica* CECT 7160, and *Shigella flexneri* CECT 4804) were acquired from the Spanish Types Culture Collection of the University of Valencia, Spain, and eight strains (*Enterobacter cloacae* ATCC 13047, *Enterococcus faecalis* ATCC 19433, *Klebsiella pneumonia* ATCC 13883, *Klebsiella pneumonia ozaenae* ATCC 11296, *Proteus vulgaris* ATCC 9484, *Pseudomonas aeruginosa* ATCC 10145, *Staphylococcus aureus* ATCC 25923, and *Streptococcus oralis* ATCC 35037) were acquired from the American Type Culture Collection.

In vitro bacterial growth inhibition: The antibacterial activity of each extract was determined through the susceptibility test of the British Society for Antimicrobial Chemotherapy (BSAC) [53]. The turbidity standard (0.5 McFarland Turbidity Standard) was a BaCl<sup>2</sup> solution which absorbance (0.08–0.10 at 625 nm) was verified in a spectrophotometer (Spectronic 21, Bausch & Lomb, Rochester, NY, USA). The solution was stored in the dark at 24 ± 2 ◦C. The bacterial inoculum of the seventeen pathogenic strains were prepared in Trypticase-Soy Agar (TSA; BactoTM, Becton, Dickinson and Co., Sparks, MD, USA) for 24 h. Thereafter, five colonies were picked up and transferred into a tube containing a saline and isotonic solution, then visually compared to the turbidity standardas previously reported [54].

Evaluation of the minimal inhibitory concentration (MIC): A broth dilution susceptibility testing method was applied for the determination of the (MIC) [55], using a stock solution prepared by adding 15 mg of the organic extract in 3 mL of Trypticase-Soy Broth (TSB). Serial dilutions of the organic extract (3.33 µg/mL, 1.67 µg/mL, 0.56 µg/mL, 0.18 µg/mL, 0.061 µg/mL, 0.0021 µg/mL, and 0.007 µg/mL) and positive control (gentamycin sulfate: 104.5 µg/mL, 35.0 µg/mL, 11.6 µg/mL, and 3.87 µg/mL) were performed. Each solution was inoculated with 50 µL (0.5 McFarland) of a culture of *Pseudomonas aeruginosa* and incubated at 37 ◦C for 18 h. The negative control was DMSO. A sterile culture media control was also used. Each assay was performed in duplicate.

#### *2.6. Analysis of Organic Extracts by LC–MS*

LC–MS analysis was carried out in an Agilent 1290 Infinity LC System (Agilent Technologies, Santa Clara, CA, USA) using a Zorbax® Eclipse Plus (1.8 µm) C<sup>18</sup> reverse phase LC column, 100 × 3 mm (Agilent Technologies, Santa Clara, CA, USA). The mass spectrometer was a micrOTOF-QIII (Bruker Daltonics, Billerica, MA, USA) supplied with an electrospray ionization (ESI) source. For the positive mode Electrospray Ionization-Quadrupole-Time of Flight-Mass Spectrometry (ESI+–Q–TOF–MS) analysis, extracts were re-dissolved in methanol and filtered through a 0.45 µm cellulose acetate membrane filter. Solutions of 0.5 µg/mL were prepared, and aliquots of 10 µL were injected. The chromatographic analysis was carried out using a 37 min step gradient (UHPLC) run using mixtures of methanol and acidified water (99.9% H2O-0.1% FA) as mobile phase, starting from: (a) 5–95% MeOH-H2O for 2 min; (b) a 25 min gradient from 5:95 methanol:H2O to 100% methanol; (c) 100% methanol for 8 min. The column was returned to the initial condition for 2 min. Prior to collecting the data, two level of calibration were employed; before the analysis, an external calibration was performed using an Agilent ESI-L Low-Calibration Tuning Mix, and during the evaluation of each sample, we used hexakis (1H, 1H, 2H-difluoroethoxy)-phosphazene (*m*/*z* 622.0290 [M + H]+; Synquest Laboratories, Alachua, FL, USA) as an internal reference, for the lock mass calibration.

#### **3. Results**

#### *3.1. Culture Conditions*

The results obtained from evaluating the changes in culture conditions are listed in Table S1. The addition of chemical elicitors impacted the amount of crude extract obtained. In *P. mangiferae*, the best result was achieved in presence of Fe2<sup>+</sup> and Ca2<sup>+</sup> ion (*man-3*, 256.0 mg and *man-4*, 232.0 mg); in *P. microspora*, under the presence of Cu2<sup>+</sup> and Ca2<sup>+</sup> ions (*mic-4*, 297.0 mg and *mic-5*, 329.0 mg).

Low pH values increased the amount of crude extract obtained. For both species, the best results were obtained at pH = 4.0 and Cu2<sup>+</sup> as elicitor: in *P. mangiferae* (*man-9*, 264.3 mg); in *P. microspora* (*mic-9*, 260.0 mg).

The third factor to consider was the incubation temperature. Maintaining constant the pH at 4.0, we found that the highest amount of extract in *P. mangiferae* was at 30 ◦C for both elicitors (*man-14*, 528.0 mg; *man-17* 448.0 mg); the production of extract using Ca2<sup>+</sup> was 34-fold higher and 1.7-fold higher using Cu2<sup>+</sup> compared to the amounts obtained in phase II. For *P. microspora*, the highest amount of extract was obtained at 24 ◦C using Cu2<sup>+</sup> as elicitor (*mic-15*, 1194.0 mg) that was 4.6-fold higher that the obtained amount in phase II.

#### *3.2. Antibacterial Activity*

A total of 34 extracts were assayed against a panel of pathogenic bacteria in a preliminary antibacterial test (disc diffusion method, mm), and only 11 extracts showed growth inhibition: seven from *P. mangiferae* and four from *P. microspora* (Table 1). These extracts were capable of inhibiting the growth of 13 of 17 pathogenic bacterial strains. Larger inhibition zones were observed against *P. aeruginosa* (12.5 mm) and *L. monocytogenes* (11.0 mm). Extract *man-6* displayed the highest broad-spectrum of antimicrobial activity (inhibited seven pathogenic strains), followed by extracts *man-9* and *mic-9* (both inhibited six pathogenic strains). Nevertheless, the culture condition for *man-6* (CaCl2, pH = 4.0 and T = 26 ◦C) induced one of the lowest amounts of organic extract (15.4 mg). The culture condition for extracts *man-9* and *mic-9* (CuSO4, pH = 4.0, T = 26 ◦C) were more favorable.

Extracts *man-15* and *mic-11* exhibited selectivity against *P. aeruginosa*, the most sensitive strain. To determine the MIC using *P. aeruginosa*, five of the eleven extracts were selected. In this experiment, only extracts *man-6* (0.11 µg/mL) and *mic-9* (0.56 µg/mL) demonstrated growth inhibition.

Hence, to correlate the chemical profile with the antibacterial activity against *P. aeruginosa*, three samples were selected from *P. mangiferae* to be analyzed by LC–MS: (1) one active, *man-7*; (2) one with broad-spectrum activity, *man-9* (12.5 mm inhibition zone); and (3) one with selective activity, *man-15* (9.5 mm inhibition zone).

To our knowledge, there are only two reports of secondary metabolites from *P. mangiferae* [20,56].

#### *3.3. Evaluation of the Chemical Diversity*

Table 2 lists the principal molecular ions present in the analyzed extracts and their determined molecular formulas. Extracts *man-7* and *man-15* showed a similar chemical composition; nevertheless, at least four compounds were only present in extract *man-15*. The molecular ions are linked, mainly to polyoxygenate compounds, but some nitrogenous are present too. In all three extracts (*man-7*, *man-9*, *man-15)*, the presence of mangiferaelactone was determined (retention time *t*<sup>R</sup> 21.55–21.59 min; *m*/*z* 401.2017 [M + H]+); none of the other previously isolated compounds from *P. mangiferae* was detected as principal components of the analyzed samples (see Figure S1). The pseudo molecular ion *m*/*z* 338.341 [M + H]+, which appeared at *t<sup>R</sup>* 28.3 min, has been established as a possible molecular formula (calculated for C22H44NO). This compound appeared in the chromatograms of extracts *man-9* and *man-15*, but it was absent in the *man-7* extract (Figure 1, Table 2), suggesting that it could be responsible for the antibacterial activity against *P. aeruginosa*.


**Table 1.** Antibacterial activity in the disk diffusion test <sup>1</sup> of organic extracts produced by *Pestalotiopsis* spp.

<sup>1</sup> Results are given in mm of inhibition.


**Table 2.** Molecular ions of secondary metabolites present in extracts obtained from *P. mangiferae.*

<sup>1</sup> Time is given in minutes.

 *‐ ‐ ‐* **Figure 1.** Total ion chromatograms (TICs) of extracts *man-9*, *man-7,* and *man-15*.

 *‐ ‐ ‐ ‐ ‐ ‐ ‐* Chemical diversity of the three extracts from *P. mangiferae* (*man-7*, *man-9,* and *man-15*) was analyzed by LC–ESI–Q–TOF–MS. The total ion chromatograms for each sample are presented in Figure 1. The peak at *t*<sup>R</sup> 21.5 min is common to all three analyzed samples. The MS spectrum of this peak showed ions [M + H]<sup>+</sup> and [M + Na]<sup>+</sup> at *m*/*z* 401.217 and 423.199, respectively. Additionally, two ion clusters were detected: M2H<sup>+</sup> and M2Na<sup>+</sup> at *m*/*z* 801.424 and 823.407, respectively (Figure 2A), which matched the MS data for mangiferaelactone, a previously characterized compound. This compound had a relatively lower concentration in sample *man-15*. Based on its selectivity against *P. aeruginosa* (9.5 mm inhibition zone), this compound could be proposed as the major component of the extract, for example, the peak at *t*<sup>R</sup> 27.8 min with [M + H]+, [M + Na]+, and [2M + Na]<sup>+</sup> ions at *m*/*z* 391.283, 413.261, and 803.536, respectively (Figure 2D). The polar section of *man-15* was the most complex, indicating a higher level of chemical diversity then *man-7* and *man-9*.

*‐ ‐ ‐* Extract *man-7* showed a higher relative concentration among the components of the polar section (peaks at *t*<sup>R</sup> 19.4, 20.2, 20.7, and 21.5 min). Its moderate polarity section included a peak at *t*<sup>R</sup> 28.4 min with [M + H]+,[M + Na]+, and [2M + Na]<sup>+</sup> ions at *m*/*z* 419.308, 441.276, and 859.579, respectively (Figure 2B), compared with extracts *man-9* and *man-15* that had a peak at *t*<sup>R</sup> 28.3 min with [M + H]+, [M + Na]+, and [2M + H]<sup>+</sup> ions at *m*/*z* 338.337, 360.319, and 675.670, respectively (Figure 2C).

 *‐ ‐ ‐ ‐ ‐ ‐ ‐ ‐* **Figure 2.** Mass spectra of selected peaks. (**A**) Mangiferaelactone at *t*<sup>R</sup> 21.5 min in extracts *man-7*, *man-9*, and *man-15*. (**B**) Peak in extract *man-7* at *t*<sup>R</sup> 28.4 min. (**C**) Peak in extracts *man-9* and *man-15* at *t*<sup>R</sup> 28.3 min. (**D**) Peak in extracts *man-7* and *man-15* at *t*<sup>R</sup> 27.8 min.

#### **4. Discussion**

 ‐ ‐ ‐ ‐ ‐ As we mentioned above, there are only two reports related to the isolation and characterization of secondary metabolites from *P. mangiferae*; early culturing under two different conditions produced antibacterial compounds mangiferaelactone (**1**) and 4-(2,4,7-trioxa-bicyclo[4.1.0]heptan-3-yl) phenol (see Figure S1) [20,56]. Neither one was among the main components present in the extracts analyzed here by high resolution LC–MS, nor the other compounds with a low molecular weight.

 Our results showed that in this initial study, *Pestalotiopsis* showed prolific antibacterial activity. The results of Table 2 indicate that each extract had a different biological activity profile. This means that the chemical composition changes in diversity and concentration.

  ‐ The changes in the culture conditions played a role in differences in the chemical diversity of *P. mangiferae* and the genus *Pestalotiopsis*. This was confirmed by the wide range of the preliminary antibacterial activities determined for each of the extracts and through the LC–HRMS analysis of the three extracts. According to the revised reviews and recent publications (Table S2) on the secondary metabolites isolated from the genus *Pestalotiopsis* and the antibacterial activity previously determined, most of the molecular formulas for the metabolites reported here did not match with those reported earlier for the genus. Nonetheless, taking into account the previously isolated compounds from the genus *Pestalotiopsis* or from *P. mangiferae* and the molecular formula obtained through high-resolution mass spectrometry, we proposed some molecular structures; also, in most of the cases they are related with a previously isolated compound with antibacterial or antifungal activity.

 Five major reviews on the chemistry and bioactivity of the genus *Pestalotiopsis* were published until 2017 [14–18]. An exhaustive exploration of the available compounds' structures allowed us to establish that the majority of the compounds present in the analyzed extracts of *P. mangiferae* have not been isolated from a species of the genus *Pestalotiopsis*. Nevertheless, it could be proposed that they belong to three of the main classes of compounds isolated from the genus, namely: (a) polyketides/polyols derivatives; (b) terpenoids/triterpenoids; and (c) nitrogen-containing compounds.

 ‐ The major group of compounds present in the analyzed extracts could belong to polyketide/polyol derivatives. For example, we proposed the hydrolysis of mangiferaelactone (**1**) (C20H32O8) that will result in the hypothetical trihydroxylactone (**1a**) (not-yet-detected by MS-analysis), its successive

dehydration and methylation could lead to two lactones (**1b**) (C16H26O4) and (**1d**) (C17H26O3), respectively (Figure 3). Compound (**1b**) has the same molecular formula as koninginins B (**1e**; C16H26O4) and E (**1f**; C16H26O4) isolated from the genus *Trichoderma* [57–59], and they probably have the same precursor as compound (**1a**) (C16H28O5). 

 *‐ ‐ ‐* **Figure 3.** Proposed polyketide/polyol derivatives **1b**,**d** that could be present in extracts *man-7*, *man-9*, and *man-15,* having as a precursor compound **1** and the intermediates **1a**,**c**. Polyketides **1e**,**f** previously isolated from the genus *Trichoderma* with same molecular formula as **1b**.  *‐ ‐ ‐* 

 ‐ ‐ ‐ ‐ α ‐ ‐ *β α* ‐ ‐ *β* ‐ ‐ *‐ ‐* Two molecular formulas could correspond to triterpenoids (C30H46O4/ C30H48O5). From the genus *Pestalotiopsis,* only oleanane- and ursane-type triterpenes have been isolated [14,17,60]. Ursane-type triterpenes have been reported when to the culture medium was added ursolic acid [61]. Nevertheless, oleanane-type were isolated (15α)-15- hydroxysoyasapogenol B (**2**), (7β, 15α)-7, 15-dihydroxysoyasapogenol B (**3**) and (7β)-7, 29-dihydroxysoyasapogenol B (**4**) from *Pestalotiopsis clavispora* [60], together with ursolic acid. These three compounds (**2**–**4**) could be synthesized by biological oxidation mechanism derived in one of the triterpenoids **2a**, **3a** or **4a**, respectively; they have the molecular formula C30H46O<sup>4</sup> or C30H48O<sup>5</sup> established by high resolution MS in extracts *man-7* and *man-15* (Figure 4). ‐ ‐ ‐ ‐ α ‐ ‐ *β α* ‐ ‐ *β* ‐ ‐ *‐ ‐*

**Figure 4.** *Cont.*

 *‐ ‐*  **Figure 4.** Proposed structure for compounds **2a**, **3a**, and **4a** based on the molecular formula determined by HRMS in extracts *man-7* and *man-15* and their proposed biosynthetic precursor.  *‐ ‐* 

 *‐ ‐* In our results two molecular ions have the same molecular formula C34H52O8, but they eluted at different time and are both present in extracts *man-7* and *man-15*. These isomers could be related to fusapirone (**5**; C34H54O9) a compound with antifungal activity, previously, isolated from *Fusarium semitectum* [62], it possess multiple chiral centers and can derived into compound (**5a**; C34H52O8) by dehydration (Figure 5). *‐ ‐* 

 *‐ ‐*  *‐ ‐* **Figure 5.** Dehydration of the polyketide derivative fusapirone **5** could produce compound **5a** with a molecular formula C34H57O<sup>8</sup> , determined in extracts *man-7* and *man-15.*

 *‐ ‐ ‐ ‐ ‐ ‐* The dehydrogenation of asperacine (**6**; C40H36N6O4) results in compound **6a** (C40H36N6O4) with an imine function (Figure 6), this molecular formula was determined for extracts *man-7*, *man-9*, and *man-15*.

 *‐ ‐ ‐* **Figure 6.** Nitrogen derivative **6a**, that could be present in extracts *man-7*, *man-9*, and *man-15.*

#### **5. Conclusions**

 ‐ For this study were selected two strains of *Pestalotiopsis* endophytic fungi (*P. microspora* Hd18 and *P. mangiferae*), that are capable of producing secondary metabolites with relevant biological activities. The strategy developed in this work included variations of the culture conditions, the determination of the antibacterial activity of the obtained extracts, together with the effective analysis of the chemical profile using LC-HRMS. This strategy could improve the discovery of new molecules with a pharmaceutical potential, in this case antibacterial. Hence, this work confirmed changes in the chemical diversity and biological activity of *P. microspora* Hd18 and, principally, *P. mangiferae* Hd08 under varying the culture conditions.

 Taking into account the chemical diversity and the preliminary antibacterial activity displayed by *P. mangiferae,* further work will need to establish and confirm the chemical composition of each extract as well as the antibacterial activity of a single compound.

 **Supplementary Materials:** The following are available online at http://www.mdpi.com/2309-608X/6/3/140/s1, Figure S1: Structure of compounds previously isolated from *P. mangiferae*. Table S1: Culture parameters and amount of organic extract produced by *Pestalotiopsis* spp., Table S2: Molecular ions and formulas of compounds isolated from the genus *Pestalotiopsis*

 ‐ ‐ ‐ **Author Contributions:** M.M.A.-P., R.V., and N.R. designed and performed the experiments and analyzed the data. D.T.-M. analyzed the data and reviewed, edited, and wrote the paper. L.C.-R. designed the experiments, analyzed the data, and edited and wrote the paper. All of the authors have read and agreed to the published version of the manuscript.

 **Funding:** This work was partially supported by the National Secretariat for Science, Technology and Innovation of Panama (SENACYT, grants COL10-060 and FID11-051), the Projects of Nagoya Protocols´ Application in Panama, and the National Research System of Panama (SNI).

 ‐ ‐ **Acknowledgments:** We gratefully acknowledge the Government of Panama through the Ministerio de Ambiente (MiAMBIENTE) for granting the corresponding permits to collect the samples used in this study.

 **Conflicts of Interest:** The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

#### **References**



© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

#### *Article*

#### **Talaat H. Habeeb <sup>1</sup> , Mohamed Abdel-Mawgoud <sup>2</sup> , Ramy S. Yehia 3,4 , Ahmed Mohamed Ali Khalil <sup>5</sup> , Ahmed M. Saleh 4,\* and Hamada AbdElgawad 6,\***


Received: 25 July 2020; Accepted: 4 September 2020; Published: 9 September 2020

**Abstract:** Arbuscular mycorrhizal fungi (AMF) and elevated CO<sup>2</sup> (eCO2) have been effectively integrated to the agricultural procedures as an ecofriendly approach to support the production and quality of plants. However, less attention has been given to the synchronous application of AMF and eCO<sup>2</sup> and how that could affect the global plant metabolism. This study was conducted to investigate the effects of AMF and eCO2, individually or in combination, on growth, photosynthesis, metabolism and the functional food value of *Thymus vulgare*. Results revealed that both AMF and eCO<sup>2</sup> treatments improved the photosynthesis and biomass production, however much more positive impact was obtained by their synchronous application. Moreover, the levels of the majority of the detected sugars, organic acids, amino acids, unsaturated fatty acids, volatile compounds, phenolic acids and flavonoids were further improved as a result of the synergistic action of AMF and eCO2, as compared to the individual treatments. Overall, this study clearly shows that co-application of AMF and eCO<sup>2</sup> induces a synergistic biofertilization impact and enhances the functional food value of *T. vulgare* by affecting its global metabolism.

**Keywords:** mycorrhizae; elevated CO2; *Thymus vulgare*; growth; photosynthesis; metabolites; biological activity

#### **1. Introduction**

Herbal plants have been widely used in traditional and folk medicine as an effectual solution to cure many diseases, being a big store for bioactive compounds, especially secondary metabolites [1]. They are known to have various biological activities such as antioxidant, antimicrobial, anti-inflammatory and anticancer properties [2]. Recently, a priority was given to herbal plants in terms of enhancing the production of the economically important phytochemicals through the application of cultivation procedures under stimulated growth conditions [3]. In this aspect, arbuscular mycorrhizal fungi (AMF) have been regarded as one of the most important beneficial microorganisms that are able to associate with almost two thirds of terrestrial plants improving their growth and stress tolerance [4]. In some cases, mycorrhizal symbiosis is essential as the host plant cannot grow normally and/or survive without it [5]. The beneficial effects of AMF symbiotic association with plants include enhanced levels of mineral nutrients and accumulation of primary and secondary metabolites [6]. From environmental point of view, AMF can keep the balance of soil aggregates, hence, able to fight erosion [7]. Accordingly, AMF represent a promising trend that has found its way in the sustainable agricultural productivity [8]. For instance, utilization of AMF for enhancing the production and quality of aromatic plants have been reported [8]. In this regard, several medicinal aromatic plants, such as pennyroyal and parsley, showed enhanced levels of bioactive metabolites when associated with AMF [6].

In another aspect, the exposure of plants to elevated CO<sup>2</sup> (eCO2) has been regarded as an effective approach to improve the nutritional and medicinal values of herbal plants [9]. eCO<sup>2</sup> can increase plant growth and productivity either directly by enhancing photosynthesis [10] or indirectly by stimulating plant water use efficiency [11]. On the other hand, the effect of eCO<sup>2</sup> on belowground communities, including AMF, are still not fully understood [12]. What is well known is that the higher the photosynthetic activity, under eCO2, the more the photosynthate transfer to plant roots and the higher release to the associated microbial communities [13]. Furthermore, as being dependent on their host plant for carbon, AMF may be sensitive to global climatic changes that influence their host [14]. Therefore, such triple effect resulting from interactions among plant, AMF and eCO<sup>2</sup> is expected to have beneficial roles in increasing the productivity and quality of crops and medicinal plants.

One of the well-known plants for both culinary and medicinal purposes is *Thymus vulgaris* L., a member of the family Lamiaceae, being widely used in folk medicine for treatment of several diseases like bedwetting, diarrhea, stomach ache, arthritis, sore throat, cough, bronchitis and chest congestion [15]. The biological activities of *T. vulgaris* are mainly ascribed to its content of secondary metabolites, particularly essential oils that have been extensively studied for antioxidant, antimicrobial and antitumor activities [16]. Thus, improving the accumulation of these phytochemicals in *T. vulgaris* could support its nutritional, medicinal and pharmacological properties. In this regard, previous studies have reported the positive impacts of both AMF and eCO<sup>2</sup> on the growth and quality of herbal plant [17,18], however, the complete picture on how AMF-eCO<sup>2</sup> combination affect primary and secondary metabolomes is not fully drawn [19,20]. In addition, metabolic profiling of the host plant is essential to understand the mechanisms behind the changes occurring in response to the individual and/or combined effect of AMF and eCO2. So far, the detailed metabolic implications induced by the synchronous application of eCO<sup>2</sup> and AMF on plants are not investigated. Thus, the current study was conducted to explore, in details, the individual and combined impacts of eCO<sup>2</sup> and AMF on *T. vulgaris*, as a model herbal plant. We have assessed the changes in mycorrhizal colonization, plant biomass production, photosynthesis, respiration and levels of individual primary (sugars, amino acids, fatty acids and organic acids), secondary (phenolic acids and flavonoids) metabolites and volatile oils. Further, the associated changes in nutritional and medicinal values of *T. vulgaris* were investigated.

#### **2. Material and Method**

#### *2.1. Experimental Setup, Plant Materials and Growth Conditions*

Soil potting was mixed with sterilized sand (1:3) and inculcated with a pure commercial inoculum of *Rhizophagus irregularis* (MUCL 41,833 obtained from Glomeromycota in vitro collection (GINCO)) at a concentration of 50 spores per soil in a pot (25 × 15 cm). The control treatments were represented by non-inoculated soil. The seeds of *T. vulgaris* were disinfected then sown in both treated and non-treated soils. Plants were grown in a controlled greenhouse at 21/18 ◦C, 16/8 h day/night, and 60% humidity, they were regularly watered. The pots inn each of the control and AMF-inoculated groups were equally subdivided into two sub-groups, one subjected to 410 ppm CO<sup>2</sup> (ambient CO2; aCO2) and the other subjected to 620 ppm CO<sup>2</sup> (elevated CO2; eCO2,) through the time course of the experiment. The plants were harvested after 6 weeks, then the aerial parts were immediately frozen in liquid nitrogen and

stored at −20 ◦C to be used in different plant analyses. For determination of dry matter and mineral elements, plant shoots were washed with distilled water and dried at 75 ◦C for 72 h.

#### *2.2. Mycorrhizal Parameters*

Mycorrhizal colonization was demonstrated following Phillips and Hayman [21]. About 0.5 g of fresh roots were clarified with potassium hydroxide (10% *w*/*v*) and potassium hydroxide (10%) + hydrogen peroxide (10% *v*/*v*) in a ratio of 1:1 (*v*/*v*), then stained with 0.05% trypan blue in lactoglycerol. A stereomicroscope (40×) was used to show the stained roots, while the colonization rate was calculated by using gridline intersect method [22].

#### *2.3. Photosynthesis Parameters*

The light-saturated photosynthetic rates (µmol CO<sup>2</sup> m−<sup>2</sup> s −1 ) of mature leaves were measured (LI-COR LI- 6400, LI-COR Inc., Lincoln, NE, USA), according to AbdElgawad et al. [23]. Dark respiration was determined as the absolute CO<sup>2</sup> exchange rate determined at photosynthetic photon flux density (µmolm−<sup>2</sup> s −1 ).

#### *2.4. Metabolic Profiling*

For extraction of sugars, plant tissues were homogenized in 50% (*v*/*v*) acetonitrile. The method described by Hamad et al. [24] was applied to identify the individual sugars in the plant extract by using high-performance liquid chromatography (HPLC), then comparing their retention time with those of a standard mixture. Quantification of the sugar samples was achieved based on peak area comparison with a calibration curve of the corresponding standards. Organic acids were extracted in phosphoric acid (0.1% *v*/*v*) supplemented with butylated hydroxyanisole (3 g/L) and then analyzed using HPLC with a SUPELCOGEL C-610H column coupled to a UV detection system set at 210 nm (LaChromL-7455 diode array, LaChrom, Tokyo, Japan). The concentration of each organic acid was calculated by using a calibration curve [24]. For extraction of amino acids, a known weight of plant tissues was vigorously homogenized in 80% aqueous ethanol. Amino acids were measured using a Waters Acquity UPLC-tqd system (Milford, Worcester County, MA, USA) equipped with a BEH amide 2.1 × 50 column. The lipophilic fraction of plant samples was obtained by extraction in chloroform/methanol (2:1, *v*/*v*). Thereafter, fatty acids were detected, according to Hassan et al. [25], by using GC/MS analysis (Hewlett Packard, Palo Alto, CA, USA) with an HP-5 MS column (30 m × 0.25 mm × 0.25 mm). Fatty acids were quantified using NIST 05 database and Golm Metabolome Database (http://gmd.mpimp-golm.mpg.de). Phenolic acids and flavonoids were extracted in acetone-water solution (4:1 *v*/*v*) for 24 h. The method outlined in Hamad et al. [24] was followed up for determination of Phenolic acids and flavonoids using an HPLC system (SCL-10A vp, Shimadzu Corporation, Kyoto, Japan). The concentration of each compound was calculated with a calibration curve of the corresponding standard. For extraction of volatile oils, two hundred gm of fresh plant material were subjected to steam distillation with about 500 mL of water, where the volatiles were collected [26]. The levels of volatiles were determined using gas chromatography–mass spectrometry (GC–MS) according to the method outlined by El Hattab et al. [27].

#### *2.5. Determination of Biological Activities*

Several methods were used to determine the total antioxidant capacities of the plant extract, including the ferric reducing antioxidant power (FRAP), oxygen radical absorbance capacity (ORAC), inhibition of LDL (low density lipoprotein) oxidation (TBARS and conjugated dienes) and inhibition of hemolysis assays [23,24]. For LDL oxidation, dialyzed LDL (100 µg protein/mL) was diluted in 10 mM PBS (phosphate buffered saline containing 0.01 Mphosphate-buffer and 0.15 M NaCl, pH 7.4) and incubated at 37 ◦C in presence or absence of 10 µM CuSO4. Oxidation was performed with or without the sample solution of colostrum proteins. After incubation, lipid peroxidation of the LDL was measured. Thiobarbituric acid reactive substances (TBARS) was determined at 532 nm/600 nm,

using 1,1,3,3-Tetramethoxypropane as standard for calibration curve, while conjugated diene formation was measured at 232 nm of LDL solution (100 µg protein/mL) in PBS incubated with CuSO<sup>4</sup> (10 µM) in the absence or presence of various concentrations of bovine colostrums protein [28].

#### *2.6. Statistical Analysis*

Experiments were carried out following a randomized complete block design. Data normality and the homogeneity of variances were checked using the Kolmogorove–Smirnov test and Levene's test, respectively. Each experiment was done in five replicates (n = 5). All the data was subjected to one-way analysis of variance (ANOVA). Student's *t*-test at probability levels of 0.05, 0.01 or 0.001 was used to test the difference between the treatment and control or between AMF alone and the combined AMF+eCO<sup>2</sup> treatment. All statistical tests were performed using the computer program PASW statistics 18.0 (SPSS Inc., Chicago, IL, USA).

#### **3. Results and Discussion**

#### *3.1. AMF Colonization and Hyphal Growth*

It is known that AMF are largely dependent on their host plant for carbon, so they are sensitive to climatic changes that affect their host plant [14]. In this sense, eCO<sup>2</sup> could have an indirect effect on mycorrhizal colonization by promoting carbon assimilation and allocation to roots [29]. Since AMF are attached to plant roots, they are lucky to receive higher amount of photosynthates under eCO<sup>2</sup> before other soil microbes [30]. Herein, the mycorrhizal growth was significantly enhanced in *T. vulgare* by AMF treatments (Table 1). Such mycorrhizal proliferation was much more stimulated under eCO<sup>2</sup> conditions. Several studies have demonstrated some positive effects for eCO<sup>2</sup> on AMF-plant association such as increased mycorrhizal root length [14] and increased extra-radical hyphae [31]. However, other studies did not show any beneficial effects for eCO<sup>2</sup> on AMF growth in host plants [32,33]. Therefore, the impact of eCO<sup>2</sup> levels on mycorrhizal growth seems to be dependent on plant species, AMF species and soil type [34]. In fact, as being a member of *Glomeraceae*, the ratio of *R. irregularis* has been reported to be more positively influenced by eCO<sup>2</sup> than others, e.g., *Gigasporaceae* [35].

**Table 1.** Mycorrhizal colonization and growth parameters in roots of *Thymus vulgare* grown under normal conditions (control) or the effect of eCO<sup>2</sup> (620 ppm), arbuscular mycorrhizal fungi (AMF) or their combination (AMF-eCO<sup>2</sup> ). Values are mean ± standard error of five independent replicates. Asterisks indicate significant changes (\*\*\* *p* < 0.001) between AMF alone and the combined AMF+eCO<sup>2</sup> treatment.


nd = not detected.

#### *3.2. AMF and eCO<sup>2</sup> Acts Synergistically to Improve Photosynthetic Capacity and Biomass Production*

It has been known that the photosynthetic rate, and consequently biomass production, could be improved under the effect of AMF inoculation, as a result of the expected increased nutrients uptake [36], and also under eCO<sup>2</sup> atmosphere due to the enhancement of the carboxylation reaction of rubisco [37]. Supporting this hypothesis, the current results revealed that eCO<sup>2</sup> and AMF independently, and to more extent in combination, promoted photosynthetic rate and biomass production in *T. vulgare* (Figure 1). Such increments were much more induced by the interaction between both treatments.

**Figure 1.** Fresh mass (**A**), dry mass (**B**), and rates of photosynthesis (**C**) and respiration (**D**) in *Thymus vulgare* grown under normal conditions (control) or the effect of eCO<sup>2</sup> (620 ppm), AMF or their combination (AMF-eCO<sup>2</sup> ). Values are mean ± standard error of five independent replicates. Asterisks indicate significant changes (\* *p* < 0.05; \*\* *p* < 0.01) compared to control, as revealed by the Student's *t*-test. Lowercase letters indicate significant differences (<sup>a</sup> *p* < 0.05; <sup>b</sup> *p* < 0.01) between AMF alone and the combined AMF+eCO<sup>2</sup> treatment.

Similar to our results, the positive effects of eCO<sup>2</sup> on biomass of *T. vulgaris* and some other medicinal plants, *Ocimum basilicum*, *Origanum vulgare*, *Mentha piperita* and *Mentha spicata*, have been previously investigated [38]. Moreover, the increased biomass production in plants inoculated with AMF was reported [39]. Regarding the interaction between eCO<sup>2</sup> and AMF, It is well known that eCO<sup>2</sup> stimulates the photosynthetic rate and plant growth [40], which in turn, affects the allocation of photosynthates to AMF, consequently makes more C available to AMF colonizing the roots [13], thus, increasing sink strength in mycorrhizal plants, and eventually this leads to increased C storage in soils [41]. Such effect is hypothesized to create a balance between carbon cost and nutrient benefits, besides reducing the negative effects of down regulation of photosynthesis caused by acclimation of plants to long-term exposure to eCO<sup>2</sup> [42]. In this regard, it has been found that both eCO<sup>2</sup> and AMF, when applied individually or in combination, improved biomass production of *Pisum sativum* and lettuce [32,43]. In addition, it has been indicated that mycorrhizal plants have higher photosynthetic rate [33] when grown under high CO<sup>2</sup> levels. However, such an effect might differ among variable cultivars [44]. In contrast, it was supposed that eCO<sup>2</sup> may impair the beneficial effects of AMF on plant biomass, especially when the fungal community is dominated by *Glomus* species [45]. This might be due to the difference among AMF taxa in their exchange of carbon and nutrients [46].

### *3.3. Application of AMF and eCO<sup>2</sup> Improves the Nutritional Value of T. vulgare*

It was assumed that the nutritive value of plants is highly related to its content of primary metabolites, e.g., sugars, proteins and lipids [9]. In this regard, sugars and organic acids are related to taste and flavor [47], essential amino acids are involved in some biological processes, such as protein synthesis [48] and a lower saturated/unsaturated fatty acids (SFA/USFA) ratio is linked to cardio-protective effects [49]. It has been reported that the higher the CO2 levels, the higher the rate of photosynthetic activity, which is linked to the enhancement of the carboxylation reaction of rubisco, the enzyme responsible for CO<sup>2</sup> fixation [37]. As a consequence of photosynthesis improvement, sugars could be accumulated and also broken down via dark respiration, resulting in production of the precursors necessary for synthesis of different classes of primary and secondary metabolites [50]. Supporting such a concept, the individual AMF and eCO<sup>2</sup> treatments induced significant increases in the content of total soluble sugar of *T. vulgare,* about 1.6 folds, however, starch was significantly accumulated under eCO<sup>2</sup> only (Table 2, Figure 2). Further, the synchronous application of AMF and eCO<sup>2</sup> caused a significant accumulation in the levels of the majority of the measured sugars relative to AMF alone treatment. Similarly, CO<sup>2</sup> enrichment enhanced the accumulation of sucrose and starch in oil palm [51], and increased the accumulation of total soluble carbohydrates and starch in ginger varieties [52]. AMF treatments were found to induce the accumulation of total soluble sugars in lettuce [53]. Further, the interaction between AMF and CO<sup>2</sup> improved forage quality of alfalfa plants by increasing the levels of glucose, fructose and hemicellulose and decreasing that of lignin [54].

**Table 2.** Levels of primary metabolites (mg g−<sup>1</sup> dry weight) in *Thymus vulgare* grown under normal conditions (control) or the effect of eCO<sup>2</sup> (620 ppm), AMF or their combination (AMF-eCO<sup>2</sup> ). Values are mean ± standard error of five independent replicates. Asterisks indicate significant changes (\* *p* < 0.05; \*\* *p* < 0.01) compared to control, as revealed by the Student's t-test. Lowercase letters indicate significant differences (<sup>a</sup> *p* < 0.05; <sup>b</sup> *p* < 0.01) between AMF alone and the combined AMF+eCO<sup>2</sup> treatment.



**Table 2.** *Cont*.

Besides, the current results revealed that the combined AMF and eCO<sup>2</sup> treatment induced a significant increase in the majority of the detected organic acids, amino acids (including both essential and non-essential amino acids) and fatty acids in *T. vulgare*, relative to AMF alone (Table 2, Figure 2). Regarding the individual treatments, AMF was more efficient in inducing the accumulation of these primary metabolites than eCO2. All AMF and/or eCO<sup>2</sup> treatments did not affect the SFA/USFA ratio. Similarly, it was reported that AMF-inoculated maize plants, under low temperature, had higher amino acid concentrations than non-mycorrizal ones, especially for Thr, Lys, Gly, Ala and His contents [55]. In contrast, proline content was reduced in mycorhizal *Capsicum annuum* grown under saline conditions [56]. Moreover, different effects of eCO<sup>2</sup> on amino acids were reported, which were reduced in barley [57], increased in spring wheat [58], or were not affected in maize [59]. It was also shown that organic acid levels were increased in mycorrhizal *Pinus sylvestris* grown under

heavy metal concentrations [60], while they were not increased in *Portulacaria afra* under eCO<sup>2</sup> [61]. The concentration of most fatty acids of soybean was unchanged under higher levels of CO<sup>2</sup> [62]. On the other hand, a significant increase in the levels of individual fatty acids was reported in parsley and dill grown under eCO2, which is more evident on UFA than SFA [9]. Therefore, the synchronous application of AMF and eCO<sup>2</sup> could be beneficial to avoid the negative impact of the individual treatment and/or to support their positive effects.

**Figure 2.** Heatmap of fold change in the contents of primary and secondary metabolites of *Thymus vulgare* grown under the effect of eCO<sup>2</sup> (620 ppm), AMF or their combination (AMF-eCO<sup>2</sup> ). Asterisks indicate significant (\* *p* < 0.05; \*\* *p* < 0.01) increased fold changes compared to control (untreated plants), as revealed by Student's t-test. Lowercase letters indicate significant differences (<sup>a</sup> *p* < 0.05; <sup>b</sup> *p* < 0.01) between AMF alone and the combined AMF+eCO<sup>2</sup> treatment.

### *3.4. AMF and eCO<sup>2</sup> Promote the Accumulation of Phenolic Compounds and Volatile Oils in T. vulgare*

Mycorrhizal symbiosis with medicinal plants has been recognized to induce the accumulation of secondary metabolites, especially phenolic compounds which play an important role in curing several ailments [63]. The present results showed that protocatechuic, p-coumaric and rosmarinic acids are the most abundant phenolic acids; while apigenin, kaempferol, quercetin and luteolin are the predominant flavonoid in *T. vulgare* (Table 3). Similarly, previous studies revealed the presence of some phenolic acids such as cinnamic, carnosic and rosmarinic acids, and also flavonoids such as luteolin and apigenin derivatives in *T. vulgare* [64]. There was a significant increment in the levels of the majority of the detected phenolic acids and flavonoids in *T. vulgare* under AMF and/or eCO<sup>2</sup> treatments, however the combined treatment was much more efficient than the individual ones (Figure 2). On the other hand, in consistence with the previous studies [65], the present results revealed the presence of 16 volatile oils in *T. vulgare*, whereas 1,8-cineol, carvacrol and p-cymene are the most dominant followed by less amounts of linalol, α- and β-pineno, α-Phellandrene, myrecene and thymol (Table 3). There is also a significant increase in the volatile oils of *T. vulgare*, under the individual and combined treatments.

**Table 3.** Levels of phenolic compounds and volatile oils (mg g−<sup>1</sup> dry weight) and biological activities in *Thymus vulgare* grown under normal conditions (control) or the effect of eCO<sup>2</sup> (620 ppm), AMF or their combination (AMF-eCO<sup>2</sup> ). Values are mean ± standard error of five independent replicates. Asterisks indicate significant changes (\* *p* < 0.05; \*\* *p* < 0.01) compared to control, as revealed by Student's t-test. Lowercase letters indicate significant differences (<sup>a</sup> *p* < 0.05; <sup>b</sup> *p* < 0.01) between AMF alone and the combined AMF+eCO<sup>2</sup> treatment.


Supporting our results, several studies have investigated the potential effects of eCO<sup>2</sup> and AMF, separately and in combination, on the levels of phenolic compounds and antioxidant activity in a variety of plant species. For instance, AMF treatments caused an increase in phenolic compounds content of lettuce [53,66], and in the antioxidant capacity of sweet basil [67]. Moreover, the flavonoids of some wild plants, such as *Libidibia ferrea*, were found to be accumulated by mycorrhizal association [68]. Similarly, eCO<sup>2</sup> induced the accumulation of some phenolic compounds in birch [69], and *Zingiber o*ffi*cinale* [52]. However, a low phenolic content was reported for some plants such as rice [70] under eCO2. Regarding the interaction between eCO<sup>2</sup> and AMF, it was reported that the induction of secondary metabolites

in lettuce and alfalfa by AMF was negatively affected under eCO2, probably due to utilization of photoassimilates for increasing plant biomass and for AMF growth as well, at the expense of secondary metabolism [43,44]. Therefore, it could be suggested that climatic changes might have an impact on AMF, which in turn, affect the metabolic functions of their host plants.

Several scenarios have been proposed to explain the induction of secondary metabolites in response to AMF associations and eCO2. It was found that AMF could affect secondary metabolites through improved photosynthesis and mineral content of the host plants, activation of pathways involved in synthesis of secondary metabolites, or higher expression of some genes related to secondary metabolism [68]. On the other hand, the eCO2-induced changes in plant secondary metabolism have been attributed to either excess amount of non-structural carbohydrates, resulting in an increment in carbon-based secondary metabolites [9,37,51].

#### *3.5. AMF and eCO2-Induced Changes in Secondary Metabolites Support the Biological Activities of T. vulgare*

Reactive oxygen species and free radicals have been recognized to induce harmful effects on living organisms. In this aspect, antioxidants, such as phenolic compounds and volatile oils, could act as free radical scavengers [71]. The present results showed an increase in the total antioxidant capacities of *T. vulgare*, tested by different methods (FRAP, ORAC, inhibition of LDL oxidation (TBARS and conjugated dienes) and inhibition of hemolysis), under the effects of AMF and/or eCO<sup>2</sup> (Table 3). It was previously reported that the antioxidant properties exhibited by *T. vulgare* extracts have been attributed to their content of volatile oils, especially carvacrol and thymol [72], flavonoids such as apigenin and luteolin derivatives and phenolic acids such as cinnamic and rosmarinic acids [64]. Moreover, some phenolic compounds were previously isolated from *T. vulgare* and proved to inhibit oxidative hemolysis [73]. The decreased levels of lipid peroxidation products such as TBARS and conjugated dienes might be ascribed to some protective effects of thymol [74].

#### **4. Conclusions**

Based on the above results, it is clear that the tested plant, *T. vulgare*, has benefited from the independent and combined effects of both AMF and eCO2, however their synchronous application is much more beneficial. Such positive impacts are being reflected on improved biomass production and higher accumulation of primary (sugars, amino acids, fatty acids and organic acids), and secondary (phenolic acids, flavonoids and volatile oils) metabolites. *T. vulgare* plants grown under synchronous application of AMF and eCO<sup>2</sup> have taken much advantage over those grown under the individual effects of both factors in terms of improved growth and bioactive components. Thus, the current study clearly shows that co-application of AMF and eCO<sup>2</sup> is a promising approach to improve the growth and the nutritional and health promoting values of *T. vulgare*. Further, the robust monitoring of primary and secondary metabolites presented herein could support our understanding about the mechanisms behind the positive impacts of AMF and eCO<sup>2</sup> on plants.

**Author Contributions:** Conceived and designed the experiments: A.M.S., H.A., A.M.A.K. Conducted the main experiment: T.H.H., R.S.Y, A.M.A.K. Measured the fungal related parameters: R.S.Y., A.M.A.K. Performed the metabolite profiling: H.A., A.M.S., M.A.-M. Measured the biological activities: M.A.-M., Analyzed the data: A.M.S., M.A.-M., A.M.A.K., T.H.H. Wrote the original draft: M.A.-M., T.H.H., Reviewed, edited and prepared the MS for submission: A.M.S., H.A., A.M.A.K. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Conflicts of Interest:** The authors declare no conflict of interest.

### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

*Article*
