Valorization of Tagetes erecta L. Leaves to Obtain Polyphenol-Rich Extracts: Impact of Fertilization Practice, Phenological Plant Stage, and Extraction Strategy

Abstract

Tagetes erecta L. is an ornamental crop known for its medicinal qualities. Large amounts of waste are produced in the commercial usage of T. erecta flowers, including leaves that could be used to develop new eco-friendly phenolic extracts with additional value for the food industry. To maximize the phenol content in the leaf extracts, this study used a Box–Behnken design with Response Surface Methodology, considering three extraction methods (Soxhlet distillation, heat, and vacuum-assisted extraction), three cropping practices (without fertilizer, chemical fertilizer, and vermicompost), and three phenological stages (plants without buds, with buds, and in flower). Extracts from plants fertilized with vermicompost (Eisenia foetida, 10 t ha⁻¹), collected during the blossoming stage and extracted via Soxhlet distillation, exhibited the highest phenol content (25.66 mg GAE/g). Further chemical characterization of the optimized extract (UV-Vis, UV-fluorescence, FTIR, GC-MS, HPLC) confirmed the occurrence of polyphenols in the extract, including quercetin, chlorogenic, gallic, p-coumaric, 3-hydroxycinnamic, and caffeic acids. This underscores the significance of T. erecta leaf residues as a valuable source of bioactive molecules, highlighting the importance of integrating agricultural practices and chemical extraction methods to enhance the phenolic content in leaf extracts from this species.

1. Introduction

Tagetes erecta, commonly referred to as “Cempoalxóchitl” or the “flower of dead”, is an ornamental species native to Mexico, belonging to the Asteraceae (Compositae) family [1]. This plant is renowned for its aesthetic appeal, the aromatic essential oil it produces, which is highly valued in premium perfumes, and its medicinal attributes. The flowers of Cempoalxóchitl hold significant cultural importance in Mexico, particularly during the “Death Day” celebration. Its cultivation and agricultural production significantly impact the rural economy in this country and beyond, as this plant is commonly available in markets for decorative, social, and religious purposes [2].

Conventional cultivation of T. erecta relies on field management techniques constrained by limited fertilizer application and plant maintenance knowledge. As a result, farms often struggle to achieve full market readiness due to inadequate or delayed flowering. The farming of T. erecta economically affects numerous families in Mexico who rely on the income generated from selling its flowers. When the plants fail to produce many buds, they are often discarded and burned, contributing to greenhouse gas emissions [3]. This leads to the disposal of substantial quantities of T. erecta plant material, including leaves and stems. Burning these residues in situ can severely affect health and the environment, releasing harmful particulate matter and greenhouse gases into the air [4]. Thus, there is an urgent need to utilize T. erecta residues more efficiently.

Crop residue management is a well-known approach to sustainable agriculture and a circular economy. The high availability of agro-waste and its richness in bioactive molecules make it a potentially interesting raw material for producing food antioxidants and active ingredients [5]. The food industry is presently investigating naturally safe derived antioxidants as substitutes for synthetic options [6]. Furthermore, there is an increasing trend towards integrating natural antioxidants into active food packaging [7].

Among antioxidants, phenolic compounds emerge as exciting molecules due to their notable abundance in plants, especially within agro-alimentary waste, which renders them especially appealing for developing innovative high-value-added products [8]. Moreover, the multiplicity of action of phenolic compounds as antioxidant, anti-inflammatory, anti-bacterial, and anti-cancer agents makes them attractive for developing novel functional foods [9].

Previous research has demonstrated that extracts from T. erecta are rich in phenolic compounds, especially digallic acid, quercetin, patuletin, quercetagetin glycosides, laricitrin, and myricetin derivatives [10,11]. However, the relative concentrations of phenolic compounds in natural extracts may vary depending on the plant tissue, the agronomic strategies, and the extraction method [12].

Among agronomic approaches, applying nitrogen fertilization or biofertilization has demonstrated a notable influence on plant development and the chemical composition of natural extracts [13]. In Mexico, more than 50% of agricultural operations use chemical fertilizers [14], often applied imprecisely and, in some cases, excessively [15]. A potential approach to mitigate these practices is the application of vermicompost. Research indicates that vermicompost not only enhances the availability of phosphorus, nitrogen, potassium, iron, and manganese in the soil [16] but also increases the phenolic content in natural extracts [17,18].

The biosynthesis of secondary metabolites is a temporally regulated process occurring at specific developmental stages, influenced by phenological plant stages and environmental conditions [19]. The metabolites’ content varies across different plant organs (roots, leaves, flowers, etc.) and is also determined by the plant’s developmental stage [20].

The most common extraction techniques to obtain high polyphenol yields involve conventional Soxhlet distillation, microwave, ultrasound-assisted extraction [21,22,23], and heat-assisted extraction [24,25]. More recently, vacuum-assisted extraction has emerged as a novel technology to extract polyphenols from foods [26,27]. It results in higher extraction efficiencies and shorter times at low temperatures than conventional methods [28].

While prior studies have explored the antioxidant properties of T. erecta flower extracts [11], limited scientific knowledge exists on how fertilization methods, extraction techniques, and the plant’s developmental stage affect the phenolic content in leaf extracts. This lack of understanding hinders the ability to develop strategies for the sustainable utilization of T. erecta agricultural leaf residues as a source of bioactive phenolic compounds.

This study aimed to investigate the influence of various factors, including fertilization method, plant phenological stage, and extraction at varying plant-to-solvent ratios considering three extraction methods, on the phenolic content of T. erecta leaf extracts using the Box–Behnken design (BBD) alongside Response Surface Methodology (RSM).

This study provides, for the first time, a holistic strategy integrating agricultural practices (fertilization and harvest time related to phenological plant stage) and chemical extraction aspects (type of extraction and plant/solvent ratio) to obtain a rich-phenol extract from this plant. The aim is to enhance the value of T. erecta leaf residues by promoting their sustainable utilization in developing new natural polyphenolic extracts to potentially be used in the food industry.

2. Materials and Methods

2.1. Agronomical Trials

The agronomic trials were conducted in Tlilapan, Veracruz, Mexico (18°48′25.9″ N; 97°5′51.66″ W), at an altitude ranging from 900 to 1800 m. The experimental period covered July to October 2019 and was replicated in 2020. T. erecta seeds were planted in July within a designated area of the study. After a period of fifteen days, the seedlings had emerged, reaching an approximate height of 10 cm. Then, they were transferred to three adjacent plots and were transplanted with a spacing of 30 × 30 cm. The experiment was conducted in a randomized complete design in a plot size of 12 × 2 m² with three replications.

The plants were cultivated in accordance with the experimental design described in Section 2.3 as follows: (a) plants grown without fertilization (plot #1); (b) cultivation with chemical fertilizers (CF) at a ratio of N:P₂O₅:K₂O of 100:60:60 kg ha⁻¹) [24] (plot #2); and (c) fertilization with organic fertilizer from Eisenia foetida (10 t ha⁻¹) (plot #3).

The species E. foetida was obtained from the Training Unit for Rural Development (UNCADER) located in Coatepec, Veracruz, and the vermicompost was produced at the Faculty of Biological and Agricultural Sciences at the Veracruzana University, Peñuela campus, Córdoba, Mexico. The vermicompost exhibited a dark brown color, characterized by a porous and odorless structure. Its physical properties at 25 °C included a pH of 7.4, an electrical conductivity of 1.53 Sdm⁻¹, and a C/N ratio of 36.86.

Fertilization was conducted on the day of transplanting in the study plots, followed by additional fertilization at 30, 45, and 60 days. On the day of transplantation, the plots received irrigation, which was repeated every 15 days.

Leaves were collected from plants at three distinct phenological stages: (a) without buttons (30 days after sowing), (b) with buttons (60 days after sowing), and (c) in the blooming stage (90 days after sowing). This selection was made to ensure a representative sample of vegetal material of healthy plants following Good Agricultural Practices (GAP). The collected plants were carefully transported in pots to the laboratory, where leaves were promptly separated from the plants.

2.1.1. Processing of T. erecta Leaves

Polyphenols are unstable and particularly susceptible to hydrolysis, thermal degradation, and/or reactions with oxygen and metal ions, which alter their structures during processing and storage, ultimately leading to a reduction in their biological activities [29]. Consequently, it was decided to process the T. erecta leaves immediately after harvesting to mitigate the degradation of their chemical content.

Leaves were pretreated with distilled water to remove dust and impurities. Excess moisture was eliminated by oven-drying at 40 °C until leaves reached a moisture content of less than 10%. Then, the leaves were cut into small pieces, frozen, and freeze-dried using a freeze-drying system (FreeZone 4.5 Liter-50C, Labconco, Kansas, MO, USA) under 0.06 mbar at −80 °C for two days. After freeze-drying, the leaves were crushed to a particle size of 0.595 mm and used for extraction according to the experimental design.

2.1.2. Determination of Soil Nitrogen Content

Previous studies have indicated that optimum application of nitrogen favors both the vegetative and floral characteristics of T. erecta Linn [30,31,32,33]. Nitrogen is the major macronutrient for plants, playing a significant role in determining the concentrations of total phenols and flavonoids in leaves of Mexican marigold (T. erecta L.) [34]. As a result, the nitrogen content in the soil was identified as a crucial factor influencing the production of polyphenols in T. erecta throughout the experiments.

The content of nitrogen in the soils without fertilization, with chemical fertilizer, and with vermicompost was determined using the micro Kjeldahl method [35]. Briefly, soil samples were collected from 20 points of the plots at 25 cm depth. Samples were stored at 4 °C for 24 h. Prior to analyses, soil samples were oven-dried at 55 °C for 12 h, and hand-cleaned to remove foreign particles. The dried and sieved soil samples (200 mg, <2 mm) were digested with H₂SO₄-H₂O₂ (2:1 v/v) using the Speeddigester K-439 (Büchi, Flawil, Switzerland) coupled with the Scrubber B-414 (Büchi, Flawil, Switzerland) at 360 °C for homogenization and digestion. Then, the samples were analyzed in the KjelFlex K-360 distillation unit (Büchi, Flawil, Switzerland) [35].

2.2. Extraction Strategies

Three extraction approaches were considered: (a) Soxhlet distillation, (b) heat-assisted extraction, and (c) vacuum-assisted extraction. In all cases, the extraction was performed in triplicate [36]. Once the extraction process was completed, all the extracts were freeze-dried and stored at 4 °C until use. All the extracts were prepared using 0.01, 0.05, and 0.1 g of freeze-dried (FD) leaf material per 10 mL of distilled water.

Water was chosen as the extraction solvent for all extraction methods. As a green solvent, water is non-toxic and appropriate for use in food products. Its cost-effectiveness and widespread availability make it essential for environmentally friendly formulations that are easy to manage, thereby minimizing the handling of hazardous waste. The use of water provides farmers with a scalable, cost-effective, and straightforward approach to extracting valuable chemicals from T. erecta feedstocks.

2.2.1. Soxhlet Distillation

The Soxhlet distillation method was performed as described [36] with minor modifications. Twenty grams of FD leaf powder were packed into a Grade 645 cellulose extraction thimble (Fisherbrand^TM, Toronto, ON, Canada) and placed inside the primary chamber of the Soxhlet extractor, positioned above the collecting flask and beneath a reflux condenser. The round bottom flask was filled with water as the extraction solvent. The solvent was heated using a heating mantle. As the water reached its boiling point, the resulting vapors ascended and encountered the condenser. These vapors were subsequently condensed and dripped into the thimble containing the T. erecta plant material. Once the thimble was adequately filled with solvent, it was siphoned off through a siphon tube, returning the solvent to the collecting flask. The ratio of plant material/solvent (distilled water) was 0.001, 0.005, and 0.01 g/mL, and the reflux was maintained for 6 h.

2.2.2. Heat-Assisted Extraction

A 20 g FD leaf powder sample was dispersed in distilled water using the same ratio plant/solvent as previously described. The mixture was refluxed for 45 min at 86 °C and filtered through filter paper Whatman No. 1.

2.2.3. Vacuum-Assisted Extraction

FD leaf powder (20 g) was dispersed in distilled water at a solvent ratio of 0.001, 0.005, and 0.01 (g/mL), and the mixture was placed in a Pado Ozen blender (HAF-HB300SV, Valencia, CA, USA) equipped with a vacuum system preventing oxidation and foam formation during blending. The extraction was performed for 20 s at 25 °C. Afterward, the mixture was filtered (Whatman No.1) and centrifuged at 2000 rpm for 30 s. The supernatant extract was freeze-dried and stored at 4 °C until use.

2.3. Optimization of Phenolic Extraction

The optimization of phenolic extraction in T. erecta leaf extracts, considering the three extraction methods outlined earlier, was performed through the application of a Box–Behnken design utilizing response surface methodology (RSM). RSM is an empirical optimization technique to evaluate the relationship between desired experimental responses (phenolic content in the extracts) and factors that could impact the studied process. It has been successfully combined with factorial methods such as the Box–Behnken design to optimize the polyphenolic extraction of plants and foods [37].

The experiment involved three factors considering each extraction method and contemplated 15 runs, encompassing three replicates at the central point (

Table 1. Coded variables, factor levels, and limits used in the experimental design.

Independent Variable	Coded Variable	Factor Levels
		Min	−1	0	1	Max
FT Soil nitrogen content (%)	X1	0	0.22	0.26	1.02	1.50
DAS (days)	X2	0	30	60	90	94
PS (g of FD leaves/10 mL)	X3	0	0.01	0.05	0.1	0.15

FT (Fertilization type); DAS (Days after Sowing); PS (Plant/solvent ratio); FD (freeze-dried).

). The independent variables were: (1) days after sowing (DAS), (2) plant/solvent (PS) ratio (0.001, 0.005, and 0.01 g/mL), and (3) fertilization type (FT). DAS were related to different phenological stages of the plant: (a) without buttons (30 days after sowing), (b) with buttons (60 days after sowing), and (c) blooming stage (90 days after sowing) [38].

Three conditions of fertilization impacting the nitrogen content of soils were studied: crops without fertilizer (0.221% of N), crops with chemical fertilizer (0.266% of N), and crops with vermicompost with Eisenia foetida (1.02% of N) at 95% (p < 0.05).

The ranges and levels of the independent variables were presented in their coded forms (

Table 2. The total phenol content of T. erecta leaf extracts using different extraction methods according to the Box–Behnken design.

Run	Independent Variables				Total Phenol Content (mg GAE g−1 FD Extract)
				Soxhlet		Heat-Assisted Extraction		Vacuum-Assisted Extraction
	X1	X2	X3	P	E	P	E	P	E
1	−1	−1	0	9.48 ± 0.04	9.80 ± 0.05	9.22 ± 0.16	9.70 ± 0.05	11.17 ± 0.04	11.30 ± 0.0
2	−1	1	0	20.16 ± 0.05	20.00 ± 0.05	16.4 ± 0.1	16.00 ± 0.10	14.37 ± 0.11	14.50 ± 0.05
3	1	−1	0	6.6 ± 0.26	6.70 ± 0.11	9.86 ± 0.05	7.50 ± 0.17	13.03 ± 0.05	13.00 ± 0.11
4	1	1	0	22.6 ± 0.1	25.49 ± 0.05	21.86 ± 0.05	21.90 ± 0.10	14.43 ± 0.1	17.20 ± 0.11
5	−1	0	−1	9.06 ± 0.05	9.90 ± 0.10	9.77 ± 0.15	9.70 ± 0.10	14.03 ± 0.05	14.50 ± 0.10
6	−1	0	1	21.17 ± 0.15	20.30 ± 0.05	20.57 ± 0.12	18.40 ± 0.10	11.15 ± 0.12	14.00 ± 0.10
7	1	0	−1	17.11 ± 0.02	14.40 ± 0.17	15.08 ± 0.07	14.20 ± 0.05	13.07 ± 0.05	11.20 ± 0.05
8	1	0	1	21.32 ± 0.11	19.20 ± 0.05	16.67 ± 0.12	15.30 ± 0.05	14.07 ± 0.1	13.10 ± 0.13
9	0	−1	−1	4.12 ± 0.11	2.80 ± 0.11	4.63 ± 0.03	3.90 ± 0.17	14.4 ± 0.1	14.00 ± 0.05
10	0	−1	1	6.54 ± 0.05	5.30 ± 0.05	6.6 ± 0.1	6.40 ± 0.00	13.17 ± 0.07	14.40 ± 0.11
11	0	1	−1	17.5 ± 0.01	18.60 ± 0.05	18.63 ± 0.21	18.80 ± 0.17	11.07 ± 0.04	11.20 ± 0.05
12	0	1	1	19.57 ± 0.15	14.30 ± 0.11	21.55 ± 0.05	20.30 ± 0.15	14.17 ± 0.06	13.00 ± 0.05
13	0	0	0	13.67 ± 0.06	9.30 ± 0.10	14.17 ± 0.21	9.90 ± 0.11	14.37 ± 0.11	17.20 ± 0.05
14	0	0	0	14.6 ± 0.2	16.40 ± 0.17	14.17 ± 0.29	9.80 ± 0.05	14.4 ± 0.1	19.60 ± 0.17
15	0	0	0	13.53 ± 0.05	20.20 ± 0.05	14.1 ± 0.1	9.91 ± 0.11	14.43 ± 0.1	16.40 ± 0.11
R2					93.48		96.00		94.00

The values represent the mean ± standard deviation of 3 replicates for each run. P (predicted value); E (experimental value); X₁ (fertilization type); X₂ (days after sowing); X₃ (plant/solvent ratio).

). The highest values were established based on previous management experiences with the crop. The nitrogen percentage of 1.5% aligns with the maximum level of nitrogen typically found in agrochemicals used by farmers in the region under study. The 94 DAS represents the longest time recorded for the plant to reach flowering under the experimental conditions. The leaf/solvent ratio of 0.15 g/10 mL corresponds to the maximum amount that could be accommodated in the Soxhlet extraction apparatus. The minimum possible value for these parameters was determined to be 0.

The experimental results were fitted using a general quadratic model represented in the response surface Equation (1):

$$Y={\beta }_0+{\sum }_{i=1}^k{\beta }_i{X}_i+{\sum }_{i=1}^k{\sum }_{j>i}^k{\beta }_{ij}{X}_i{X}_j+{\sum }_{i=1}^k{\beta }_{ii}{X}_i^2$$

(1)

where Y represents the response value; ${\beta }_0$ is the constant; ${\beta }_{\mathrm{i}}$ is the coefficient of the linear effect; ${\beta }_{ij}$ is the coefficient of the interaction effect; and ${\beta }_{ii}$ is the coefficient of the quadratic effect. The variables $X_i$ and $X_j$ denote the process factors, the coefficients of the response surface equation, and the fitting model (Equation (1)). The significance of the model was assessed through analysis of variance (ANOVA) for multiple responses (p < 0.05).

Total Phenol Content Determination in the Extracts

The total phenol content in the extracts was assessed as previously described with minor modifications [39]. Briefly, 250 μL of 50% Folin–Ciocalteu reagent, 250 μL of 7.5% sodium carbonate (Sigma-Aldrich, St. Louis, MO, USA), and 250 μL of extracts (100–700 µg/mL were incubated at 40 °C for 30 min. Subsequently, 2.0 mL of distilled water was added to the mixture. The absorbance was then measured at 750 nm using a UV/VIS Spectrophotometer, VE-5600UV (McAllen, TX, USA). The total phenol content (TPC) was calculated considering a standard curve of gallic acid (y = 0.049x − 0.0117, R² = 0.99). Results were expressed as mg of gallic acid equivalent (GAE) per gram of FD extract (mg GAE/g). All the extracts were evaluated in triplicate.

2.4. Chemical Characterization of the Optimized Extract

The optimized extract from T. erecta leaves, which demonstrated the highest total phenolic content as anticipated by the Box–Behnken design and RSM, was experimentally reproduced and further analyzed through a series of chemical assessments. These included the measurements of hydroxycinnamic acid and total flavonoid content, UV fluorescence, proximal and elemental analysis, as well as FTIR, GC-MS, and HPLC analysis.

2.4.1. Determination of the Hydroxycinnamic Acid Content

The determination of hydroxycinnamic acid content involved the addition of 500 μL of the extract (500 µg/mL) to a mixture comprising 1.0 mL of 0.5 M HCl, 1.0 mL of Arnow’s reagent, 1.0 mL of 2.125 M NaOH (Sigma-Aldrich, St. Louis, MO, USA), and 1.5 mL of distilled water. Subsequently, absorbance was measured at 525 nm (A525 nm). The hydroxycinnamic acid content was quantified considering a standard curve of chlorogenic acid (1–150 µg/mL; y = 0.00169x + 0.007628; R² = 0.99). Results are expressed as mg of chlorogenic acid equivalents (ChAE) per g of FD extract (mg ChAE/g) [39]. Standardized maritime pine bark extract (Oligopin^®) (D.T.R, Dax, France) was used as a positive control. This extract is recognized for its high phenol content and its wide application in the food industry as a dietary supplement [40].

2.4.2. Determination of the Total Flavonoid Content

To determine the optimized extract’s total flavonoid content (TFC), 2.0 mL FD extract (500 µg/mL) was mixed with 2.0 mL of 2% aluminum chloride (Sigma-Chemical, St. Louis, MO, USA). This mixture was incubated at 20 °C for 1 h, and the absorbance (A415 nm) was recorded. The total flavonoid content was quantified considering a standard curve of quercetin (1–40 µg/mL; y = 0.03191x + 0.0106; R² = 0.99). TFC was expressed as milligrams of quercetin equivalents (QE) per gram of FD extract (mg QE/g) [41]. Given its high flavonoid content, Oligopin^® was used as a positive control [40].

2.4.3. UV-Fluorescence Spectroscopy

T. erecta extract was diluted in HPLC grade methanol (Fisher Scientific, Fairlawn, NJ, USA) at 10 ppm and subjected to analysis using a Shimadzu RF 5301PC spectrofluorophotometer equipped with Panorama Fluorescence software (2.1 version; Duisburg, Renania, Germany). The analytical approach was based on the previous methodology [42]. In summary, the excitation wavelength remained constant and was scanned from 250 to 500 nm, while emission wavelengths were recorded with a 5 nm difference (ranging from 260 to 510 nm). Excitation and emission slit widths were set up at 5 nm to ensure precision in the fluorescence measurements. Data was collected every 1 nm.

2.4.4. Proximal and Elementary Analysis

This analysis was performed as previously reported [42], which involved the determination of fixed carbon, volatiles, and ash content using a Mettler Toledo thermogravimetric analyzer (SDTA851e) equipped with STARe data analysis software version 9 (Mettler Toledo, Columbus, OH, USA) for obtaining thermogravimetric/differential thermal data curves (TGA/DTG). Elements composition, including contents of carbon (C), hydrogen (H), nitrogen (N), and sulfur (S) in the most promising extract from T. erecta, was determined utilizing a Leco TruSpec Micro elemental analyzer 628 series instrument (St Joseph, MI, USA). The oxygen content was calculated by difference, considering the ash content on a dry basis.

2.4.5. Fourier Transform Infrared (FTIR) Spectroscopic Analysis

FTIR analysis was conducted to identify the main functional groups present in the optimized extract of T. erecta, as previously described [43]. The spectra were recorded using a PerkinElmer Spectrum 100 FTIR spectrometer (Beaconsfield, Bucks, UK). To assess the number, position, and area of the bands, samples were analyzed over a wavelength spectrum ranging from 4000 to 650 cm⁻¹. The resulting data were exported, plotted, and analyzed with OriginPro^® 2015 software. All measurements were taken in triplicate. The assignment of the functional groups was determined as previously described [44,45].

2.4.6. GC-MS Analysis

Volatile and semi-volatile molecules were determined in the optimized extract from T. erecta leaves via GC-MS analysis using an Agilent GC 7890B system, coupled with a 7000C GC/MS Triple Quadrupole mass spectrometer (Santa Clara, CA, USA), equipped with an HP-5MS column (30 m × 0.25 mm × 0.25 μm) and a G4513A injector. Samples (1.0 μL) were injected in splitless mode with helium as the carrier gas at a constant flow rate of 1.0 mL/min. The temperature gradient program was as follows: 70 °C (3 min), 6 °C/min to 100 °C (3 min), 6 °C/min to 150 °C (3 min), 6 °C/min to 200 °C (3 min), 6 °C/min to 250 °C (3 min), 6 °C/min to 320 °C (10 min).

To enhance the volatility of target molecules, such as phenols, and reduce their polarity, the extract underwent derivatization as previously described [46]. Briefly, 10 mg of lyophilized extract was subjected to derivatization by adding 200 µL of pyridine (containing 1.5 mg of DMAP per mL) and 200 µL of BSTFA (with 10% TMCS). The reaction mixture was gently agitated, heated at 70 °C for 4 h, and then cooled in an ice bath (0 °C) for 15 min. Subsequently, the sample was diluted with 400 µL of ethyl acetate, centrifuged at 5000 rpm for 5 min, and the supernatant was transferred to a 1.5 mL amber glass vial for GC-MS injection.

Mass spectral data for underivatized and derivatized extracts were acquired and processed using Agilent Mass Hunter Qualitative Analysis (B.07.00) and NIST MS Search 2.2 software. Compound identification was achieved by comparing mass spectral fragmentation patterns and retention times with reference spectra from the NIST Mass Spectral Library and the published literature data.

2.4.7. HPLC Analysis

Before HPLC analysis the samples were filtered with 0.45 µm Nylon acrodiscs. The separation of compounds was conducted via HPLC using a Varian ProStar HPLC Model 350 (East Lyme, CT, USA), which is equipped with a binary pump, a temperature-controlled autosampler, a temperature-controlled column compartment, and a multi-wavelength photodiode array (PDA) and UV detector. The separation of analytes was accomplished using a Phenomenex Gemini C18 column (250 mm × 4.6 mm, 5 µm, Torrance, CA, USA) at ambient temperature. Compounds were monitored at 260 nm at a flow rate of 0.5 mL/min, with the mobile phases comprising 0.1% formic acid in water and acetonitrile (70–30 v/v). The injection volume was 100 µL. The procedure was founded on the methodology established by Hung-Ju Chen et al. (2012) [47].

Identification of the compounds was achieved by comparing their retention times and UV spectra with those of pure standards, including gallic acid, malic acid, chlorogenic acid, quercetin, caffeic acid (all standards were sourced from Sigma Aldrich, St. Louis, MO, USA). Calibration curves were utilized to quantify the identified compounds, achieving a correlation coefficient greater than 0.99 using Clarity 8 DataApex software, version 8.4.0.47, Serial No.: 011-053604.

Linear regression data for each of the identified compounds were as follows: (1) malic acid, concentration range: 0.05–0.5 mg/mL, regression equation: y = 10061.24x, r² = 0.99; (2) caffeic acid, concentration range: 0.05–0.5 mg/mL, regression equation: y = 30474.27x, r² = 0.99; (3) chlorogenic acid, concentration range: 0.05–0.5 mg/mL, regression equation: y = 23346.59x, r² = 0.99; (4) quercetin, concentration range: 0.05–0.5 mg/mL, regression equation: y = 22140.74x; r² = 0.99; (5) gallic acid, concentration range: 0.004–0.3 mg/mL, regression equation: y = 101971.52x, r² = 0.99.

2.5. Statistical Analysis

The Box–Behnken design with Response Surface Methodology used to optimize the phenolic yields from T. erecta leaves was analyzed using Minitab 2019^® software (version 18.0; State College, PA, USA). A p-value of less than 0.05 indicates significant model terms. The student’s t-tests were used to analyze differences in the content of hydroxycinnamic acid and flavonoids at p < 0.05.

3. Results and Discussion

3.1. Extraction Optimization

Approximately 7000 hectares are estimated to be used globally to cultivate T. erecta [11], whose production reaches 600,000 tons annually [48]. While primarily grown for ornamental purposes, T. erecta is a source of bioactive compounds, such as lutein [49] and polyphenols [50]. Numerous strategies have been suggested for the valorization of T. erecta byproducts. Most research has concentrated on extracting residues from the inflorescence [51] to obtain antioxidant, anti-diabetic, anti-inflammatory, antilipemic, and chelating agents [11,52]. Various uses for inflorescence waste encompass the production of essential oils for perfumery [53], phytoremediation [54], and soil fertilization.

The leaves have received significantly less attention in valorization than the flowers; nonetheless, the essential oil derived from the foliage of T. erecta has been proposed for applications as an herbicide, antifungal agent, and insecticide [55]. Other important benefits of leaf extracts are their blood-coagulating, insecticidal, and antibacterial properties [56,57,58].

Despite numerous studies employing RSM to enhance the extraction of lutein and polyphenols from the residues of T. erecta flowers [21,59,60,61], optimization of leaf extraction has been limited. Prior research applied RSM to optimize the coagulating activity of T. erecta leaf extracts, examining various concentrations of leaf extract and blood [57]. Unlike previous research, the present study emphasizes optimizing leaf extraction through a comprehensive strategy that includes agricultural practices (fertilization and harvest time related to phenological plant stage) and chemical extraction, considering the type of extraction and plant/solvent ratio. As a result, it provides farmers with methods to increase the phenol content of plants. Simultaneously, it offers information about the chemical characterization of the optimized extract, enhancing the value of T. erecta leaf residues.

3.1.1. Soxhlet Distillation

Soxhlet distillation was chosen for this study as it is a widely used method to obtain polyphenolic-rich extracts from plants and foods [62]. It is a simple and economical approach in which plant materials are repeatedly extracted with new portions of the solvent, thereby shifting the mass transfer equilibrium. Additionally, the temperature during extraction is kept relatively high in the area where extraction occurs, and no post-leaching filtration step is needed [62].

As observed in Table 2, the TPC ranged from 2.8 to 25.49 mg GAE g⁻¹ of FD extract. Both linear and quadratic models were applied to model the relationship between the studied factors (FT, DAS, and PS) and the TPC, resulting in a regression in Equation (2). The analysis revealed that the quadratic model exhibited a significantly good fit, and the associated terms were statistically significant (p < 0.05), as indicated in Table S1. The fitted quadratic model in coded factors for TPC is provided below:

$$\begin{gathered} \mathrm{T}\mathrm{P}\mathrm{C}=15.30+0.88X_1+6.77X_2+1.77X_3+2.9X_1^2+2.6X_2^2-2.45X_3^2 \\ +2.25X_1{X}_2-1.6 2.25X_1{X}_3+1.70 2.25X_2{X}_3 \end{gathered}$$

(2)

Figure 1 shows the response surface plots, which account for the impact of nitrogen content (FT), days after sowing (DAS), and the ratio of plant/solvent (PS) on the total phenol content (TCP) of T. erecta leaf extracts. The ANOVA results presented in Table S1 indicate that the linear factor linked with DAS significantly impacted TPC in the extracts. However, according to the ANOVA (Table S1), no significant interactions were observed among these factors.

The plant’s developmental stage significantly impacted on TPC, particularly during the 9 days after sowing, when the plant undergoes essential adaptations to biotic and abiotic stress. A previous study performed with two varieties of Tagetes, Pusa Basanti Gainda (PBG) and Pusa Narangi Gainda (PNG), also showed that leaves’ developmental stage impacts their phenolic content. Young leaves in the PBG variety exhibit higher levels of phenols, tannins, and flavonoids. In contrast, PBG showed higher amounts of free sugars, starch, amino acids, and free proline, indicating a correlation with different phases of plant growth [50].

The model adequacy was confirmed through a coefficient of determination (R²) up to 83.48%. In line with the optimization model, it can be predicted that factors at high levels X₁ (1.02% N), X₂ (leaves from plants at 90 DAS), and X₃ (PS ratio of 0.1 g/10 mL) will achieve the highest TPC (25.74 mg GAE g⁻¹) in the FD T. erecta extract obtained via Soxhlet extraction (R² > 93.48%) (Figure 1I)

3.1.2. Heat-Assisted Extraction

Heat-assisted aqueous extraction is an environmentally friendly methodology successfully used to obtain phenolic compounds from the agri-food industry byproducts, avoiding harmful solvents [63,64]. High temperatures favor mass transfer, allowing high yields of phenols. Unlike Soxhlet distillation, where the plant biomass was refluxed for 6 h, in this method, the sample was refluxed for 45 min, which significantly reduces energy and time consumption, making the process more efficient. However, this method requires a filtration step as a work-up.

According to the BBD design, the TPC of FD T. erecta leaf extracts are outlined in Table 2. The observed TPC values exhibited variability within the range of 3.9 to 21.9 mg GAE g⁻¹ of FD extract. Linear and quadratic models were applied to establish the relationship between the factors and TPC, resulting in a regression equation. The analysis revealed that the quadratic model fit well (p < 0.05). The fitted quadratic model is represented in Equation (3):

$$\begin{gathered} \mathrm{T}\mathrm{P}\mathrm{C}=9.870+0.638 {\mathrm{X}}_1+6.188{\mathrm{X}}_2+1.725{\mathrm{X}}_3+{2.977\mathrm{X}}_1^2+0.928{\mathrm{X}}_2^2 \\ +1.553{\mathrm{X}}_3^2+2.025{\mathrm{X}}_1{\mathrm{X}}_2-1.900{\mathrm{X}}_1{\mathrm{X}}_3-0.250{\mathrm{X}}_2{\mathrm{X}}_3 \end{gathered}$$

(3)

The ANOVA results indicated that the linear terms related to DAS and the PS ratio were significant for TPC using this extraction method (Table S1). Moreover, the quadratic effect was substantial for the nitrogen content in the soil related to the fertilization type. The nitrogen content and DAS interaction showed statistical significance (Table S1). In this study, plants were collected without buttons (DAS = 30), with buttons (DAS = 60), and at the blooming stage (DAS = 90), thereby reflecting different phenological stages. Previous studies have shown that essential oils obtained from the aerial tissues of Tagetes patula at different phenological stages showed changes in the relative proportion of molecules contained in the oils, impacting their antioxidant and antibacterial properties [65]. The plant’s developmental stage significantly influences phenolic content, particularly during DAS, when the plant undergoes essential adaptations to biotic and abiotic stress [66].

The results demonstrated that the soil’s nitrogen content related to the fertilization strategy significantly impacts the extract’s phenol content obtained via heat-assisted extraction. Previous studies have shown the influence of fertilization on growth, flower yield, seed yield, flowering stimulation, and content of chemical constituents in T. erecta [62,67,68,69]. Nitrogen is one of the main macronutrients required for plant growth. Its great relevance is related to its presence in amino acids, chlorophyll, hormones, amino acids, proteins, and nucleic acids, which are critical mediators involved in the biosynthesis of secondary metabolites, including carotenoids, polyphenols, and glucosinolates [70]. Using nitrogen at 12.6 mg L⁻¹ for T. erecta cultivation significantly increased the content of total phenols and flavonoids in the leaves but not in the flowers, as previously demonstrated [34]. The relevance of the phenological stage and the type of fertilization to produce secondary metabolites in plants could explain why the interaction of both parameters significantly impacted the TPC of the FD T. erecta leaf extracts procured using this extraction approach.

According to the optimization strategy (Figure 1II), it can be predicted that the selection of leaves from plants 90 DAS, cultivated using organic fertilization with Eisenia foetida (1.02% nitrogen) and employing a PS ratio at a low level (0.01 g of FD extract per 10 mL) during the heat-assisted extraction, allows achieving an optimal TPC (24.35 mg GAE g⁻¹, R² > 96%).

3.1.3. Vacuum-Assisted Extraction

Vacuum treatment is used in the food industry for drying, cooking, and packaging [27]. Vacuum systems can preserve and/or improve both the sensory and nutritional properties of foods, reducing the impact of their degradation and improving their shelf life [71]. Eliminating oxygen in vacuum systems reduces oxidation-reduction processes, decreasing enzymatic activity. During vacuum treatment, the mechanical and structural properties of the vegetable tissue are altered, including texture and porosity [72], thereby provoking an increase in extraction yields [73]. While Soxhlet distillation and heat-assisted extraction take six hours and 45 min, respectively, this methodology completes extraction in just 20 s at 25 °C. This considerably reduces extraction time and energy consumption, making the process more efficient.

According to the BBD design, the TPC values exhibited variability ranging from 11.3 to 19.6 mg GAE g⁻¹ (Table 2). The fitted quadratic model for this extraction method is represented in Equation (4):

$$\begin{gathered} \mathrm{T}\mathrm{P}\mathrm{C}=17.73+0.025{\mathrm{X}}_1+0.400{\mathrm{X}}_2+0.450{\mathrm{X}}_3-1.84{\mathrm{X}}_1^2-1.89{\mathrm{X}}_2^2-2.69{\mathrm{X}}_3^2 \\ +0.25{\mathrm{X}}_1{\mathrm{X}}_2+0.6{\mathrm{X}}_1{\mathrm{X}}_3+0.35{\mathrm{X}}_2{\mathrm{X}}_3 \end{gathered}$$

(4)

The ANOVA for this extraction method indicates that the studied factors do not significantly influence TPC when using this extraction method (Table S1, p > 0.05). According to the optimization model, it can be predicted that the use of leaves from plants 90 DAS, cultivated with Eisenia foetida organic fertilizer (1.02% nitrogen) and employing a PS ratio at a low level (0.05 g of FD extract per 10 mL), achieves an optimal TPC (20.83 mg GAE g⁻¹, R² > 94%) via vacuum-assisted extraction (Figure 1III).

Overall, using different extraction approaches, RSM was applied to predict the optimum conditions for extracting phenolic compounds from T. erecta leaves. The highest phenolic content (25.74 mg GAE g⁻¹) in the T. erecta leaf extracts was predicted at the following conditions: plants grown in soil fertilized with Eisenia foetida vermicompost (1.02% nitrogen), harvested at the blooming stage (DAS = 90), and extracted in water using a PS ratio of 0.1 g per 10 mL using the Soxhlet method.

The optimal values predicted for each extraction method regarding the TPC of extracts were experimentally validated (

Table 3. Experimental validation of the predicted values at optimal extraction conditions.

Response Variable/ Extraction Method	Optimal Predicted Value	Experimental Value
TPC (mg GAE/g)/Soxhlet	25.74	25.66 ± 0.06
TPC (mg GAE/g)/HAE	24.35	23.10 ± 0.88
TPC (mg GAE/g)/VAE	20.83	19.60 ± 0.86

TPC (total phenol content); GAE (gallic acid equivalents); HAE (heat-assisted extraction); VAE (vacuum-assisted extraction). Experiments were performed in triplicate.

). The correlation between the prediction and the experimental results for all extraction methods confirmed that the model adequately reflected the optimization since the difference between the predicted and the experimental values was around 5% for all responses [74].

Compared to heat-assisted and vacuum-assisted extraction, the higher yields in extracting phenols via Soxhlet extraction may be attributed to the higher temperatures maintained for an extended period. Previous research has highlighted that the extraction temperature is related to natural extracts’ enhanced phenolic content and antioxidant activity [75,76,77]. At elevated temperatures, a considerable transfer of mass takes place. Concurrently, insoluble phenolic compounds may be released by disrupting their linkages with the structural cellulose, hemicellulose, and lignin, leading to enhanced phenolic content and antioxidant activity in natural extracts [78].

3.2. Chemical Characterization of the Optimized Extract

3.2.1. Determination of Hydroxycinnamic Acids and Flavonoid Content in the Optimized Extract

Figure 2 represents the content of hydroxycinnamic acids and flavonoids in the optimized extract obtained using the Soxhlet distillation. All results were compared, with the commercial extract Oligopin^® (D.T.R, Dax, France) used as a positive control. Oligopin^® is a standardized extract derived from French maritime bark which is recognized for its richness in phenolic compounds and used by the food industry as an antioxidant and nutraceutical agent [79,80].

Oligopin^® exhibited a higher content of hydroxycinnamic acids than T. erecta extract (Figure 2A), as previously reported [81]. Hydroxycinnamic acids such as chlorogenic, ferulic, p-coumaric, and caffeic acids have been identified in the flowers and leaves of T. erecta [11,50,82]. GC-MS analysis confirmed the occurrence of hydroxycinnamic acids in the optimized extract of T. erecta, specifically 3-hydroxycinnamic acid and p-coumaric acid. Additionally, HPLC analysis revealed the presence of chlorogenic and caffeic acids.

In contrast, the flavonoid content in the optimized T. erecta extract (16.68 mg QE/g) was higher than that shown by Oligopin^® (Figure 2B). Flavonoids represent a significant group of secondary metabolites identified in flower extracts from T. erecta, including patulitrin, quercetagetin, 6-hydroxy kaempferol, quercetin, isorhamnetin [60], and their derivatives [83]. HPLC analysis confirmed the occurrence of quercetin in the optimized extract. The versatility of action of flavonoids and their high content in the leaves of this species suggest excellent potential for leaf extract commercialization that warrants further investigations.

3.2.2. UV-Fluorescence of the Optimized T. erecta Extract

UV-Fluorescence has been successfully used to identify the presence of phenols in natural extracts [42,84]. Phenols are present in plants as simple structures or conjugated ring systems with different dihedral angles in their chemical structures, resulting in diverse fluorescence spectra. The UV-fluorescence spectra may include compounds with carbonyl groups and one-ring aromatic compounds as simple phenols, which can be detected at a wavelength range of 260 to 290 nm [85]. The wavelengths of fluorescence emission increase with the number of aromatic rings, allowing the detection of short conjugated systems initially and subsequently polycondensed structures between 290 and 400 nm. As shown in Figure 2C, the highest peak in the UV-fluorescence spectra was observed after 400 nm, which could be related to the presence of large conjugated phenolic polycondensed structures in the optimized T. erecta extract [85]. Further separation and chemical characterization by NRM are necessary to identify these compounds.

3.2.3. Proximate and Elemental Analysis of the Optimized Extract

Table 4. Proximate and elemental analysis of the optimized T. erecta extract.

T. erecta Optimized Extract
Proximate analysis (wt.%)
Moisture a	7.55 ± 0.09
Volatiles b	57.24 ± 0.21
Fixed carbon b	33.15 ± 0.11
Ash b	9.61 ± 0.32
Elemental analysis (wt.%)
C b	40.85 ± 0.24
H b	5.98 ± 0.01
N b	4.53 ± 0.07
O c	48.64 ± 0.15

^a As-received basis. ^b Dry basis. ^c By difference. Carbon (C), Hydrogen (H), Nitrogen (N), Sulfur (S), Oxygen (O). Data represents the mean ± SD of two experiments.

displays the proximate and elemental analysis of the optimized T. erecta extract. The study reveals a notable concentration of volatile compounds, fixed carbon, carbon, and oxygen in the optimized extract. Volatile compounds have been described in the essential oil from T. erecta leaves [86]. The GC-MS analysis verified the occurrence of volatile compounds previously reported in T. erecta oils such as piperitone and piperitenone [86]. The elemental carbon and oxygen content in the T. erecta extract could relate to the extract’s fixed carbon and volatile matter content [87].

3.2.4. FTIR Analysis

The FTIR technique has been widely employed to assess the quality and composition of natural extracts, as it provides valuable information regarding the specific functional groups that define their chemical composition. Figure 3 shows the FTIR spectra of the optimized extract from T. erecta leaves.

Stretching vibrations of the O-H bond (3600–3100 cm⁻¹) are typically assigned to phenols, alcohols, and carboxylic acids [37]. Consequently, the initial absorption band, which reaches a maximum at 3276.41 cm⁻¹, may be associated with the presence of phenolic compounds. This finding aligns with the outcomes derived from UV-fluorescence analysis and supports a previous investigation that analyzed the FTIR spectrum of an aqueous extract of T. erecta [88]. The increase of a band at 2983.18 cm⁻¹ was assigned to C-H stretching vibrations [89]. The significant peak observed at 1585.82 cm⁻¹ was associated with the C=O stretching vibration and amide linkages [90] which may indicate the presence of flavonoids in the T. erecta extract. Previous FTIR analyses have shown vibrations in the range of 1650–1560 cm⁻¹ for various types of flavonoids, including flavones, isoflavones, flavanones, flavonols, flavanols, and anthocyanins [91]. This result aligns with the flavonoid content in the optimized extract of T. erecta, as well as with the HPLC analysis that verified the quercetin occurrence in the extract. The presence of amide linkages is consistent with the proportion of amides found in the optimized extract (8% of total identified molecules) according to the GC-MS analysis.

The stretching signal detected at 1404.81 cm⁻¹ for the C=C bond is linked to the presence of conjugated aromatic systems in the extract and may suggest the existence of flavonols or anthocyanins, which typically display a distinctive band within the range of 1475–1400 cm⁻¹ [91]. This result aligns with the detection of quercetin, a flavonol identified in the HPLC analysis. Vibrations at 1057.28 cm⁻¹ are associated with C-N stretching [92] and C-O-C glycosidic bonds [93], consistent with the presence of carbohydrates in the extract as confirmed through GC-MS analysis. Vibrations in the range of 900–800 cm⁻¹ were attributed to C-H bending and can be regarded as complementary to various stretches, including C-O and C-N [94]. The functional groups identified in the FTIR spectrum align with those previously documented for T. erecta leaves, concerning the presence of phenols, amide linkages, and aromatic compounds [88].

3.2.5. GC-MS Chemical Identification

The proximate analysis revealed a notable concentration of volatile compounds in the optimized T. erecta extract, accounting for 57.24% of its composition. To further explore the volatile and semi-volatile compounds, the extract was analyzed using GC-MS (Figure 4a). Since chemical derivatization drives the transformation of less volatile compounds, including phenols, into volatile and thermally stable molecules, the extract underwent derivatization prior to GC-MS analysis (Figure 4b).

Table 5. Compounds identified in Tagetes erecta leaves using GC-MS.

No.	Identified Compound	RT (min)	Extract	Derivatized Extract	Groups *
1	Benzaldehyde	7.430	*		Ah
2	Benzyl alcohol	9.134	*		A
3	Piperitone	16.400	*		T
4	Octanoic acid	16.611	*		FA
5	Unknown	16.844		*	Uk
6	Unknown	17.203		*	Uk
7	Indole	17.468	*		Ø
8	Butanedioic acid	18.106		*	CA
9	Piperitenone	18.803	*		T
10	2-Butenedioic acid, (E)-	18.953		*	CA
11	Nonanoic acid	19.217	*		FA
12	2-Allyl-4-methylphenol	19.466	*		Ph
13	Decanoic acid	21.952	*		FA
14	Unknown	23.140	*		Uk
15	Malic acid	23.411		*	CA
16	2,4-Di-tert-butylphenol	23.656	*		Ph
17	2,6-Di-tert-butyl-4-methylphenol	23.785	*		Ph
18	Xylose	26.348		*	CH
19	2-Pentenedioic acid	26.726		*	CA
20	Methyl jasmonate	27.460	*		E
21	Unknown	27.883	*		Uk
22	Unknown	28.822	*		Uk
23	3-Hydroxycinnamic acid	30.696		*	Ph
24	Unknown	31.364	*		Uk
25	D-(-)-Fructofuranose	31.534		*	CH
26	Neophytadiene	31.558	*		T
27	Citric acid	31.718		*	CA
28	Myristic acid	31.816	*		FA
29	D-(-)-Fructopyranose	31.828		*	CH
30	Versalide	32.110	*		O
31	D-(+)-Talofuranose	32.464		*	CH
32	Quininic acid	32.860		*	CA
33	7,9-Di-tert-butyl-1-oxaspiro(4,5)deca-6,9-diene-2,8-dione	33.639	*		K
34	β-D-(+)-Mannopyranose	33.680		*	CH
35	p-Coumaric acid	34.214		*	Ph
36	Myo-2-Inosose	34.370		*	Cyc
37	Unknown	34.689	*		Uk
38	Gallic acid	35.034		*	Ph
39	Unknown	35.149	*		Uk
40	Scyllo-Inositol	35.338		*	Cyc
41	D-Allofuranose	35.559		*	CH
42	Talose	35.945		*	CH
43	Palmitic Acid	36.614	*	*	FA
44	10-Octadecenoic acid, methyl ester	37.756	*		E
45	Myo-Inositol	38.276		*	Cyc
46	Unknown	38.465	*		Uk
47	9-Octadecenamide	38.888	*		Ad
48	Tetradecanamide	39.266	*		Ad
49	9,12-Octadecadienoic acid (Z,Z)-	39.929	*		FA
50	α-Linolenic acid	40.058	*		FA
51	α-D-Glucopyranosiduronic acid	40.155		*	T
52	Stearic acid	40.482	*		FA
53	Heptacosane	41.384	*		Ak
54	2-O-(beta-D-glucosyl)glycerol	42.439		*	CH
55	Oleamide	42.462	*		Ad
56	cis-11-Eicosenamide	42.858	*		Ad
57	Octadecanoic acid, 4-hydroxy-, methyl ester	43.051	*		E
58	2-Myristynoyl-glycinamide	43.171	*		Ad
59	Unknown	44.073	*		Uk

RT: retention time. * Groups: Alcohol (A); Aldehyde (Ah); Phenol (Ph); Other (O); Cyclitol (Cyc); carbohydrate (CH); Amide (Ad); Polycyclic (Ø); Ester (E); Carboxylic Acid (CA); Ketone (K); Fatty acid (FA); Alkane (Ak); Terpene (T); and Unknown (Uk).

lists the numbered compounds shown in Figure 4. A total of 59 compounds were detected in chromatograms—36 in the underivatized extract and 24 in the derivatized extract. The major identified molecules were oleamide (55) and xylose (18). The largest proportion of the identified compounds belonged to the family of fatty acids (14%), carbohydrates (14%), phenols (10%), and carboxylic acids (10%). The occurrence of compounds like piperitone, piperitenone, indole, gallic acid, coumaric acid, and neophytadiene, as well as α-linolenic, myristic, palmitic, and and stearic acids has been previously recorded in T. erecta extracts [11,50,86,95,96,97].

Oleamide (55), identified as a major compound, is recognized for its vasodilatory [98], hypnotic [99], and antinflammatory [100] properties. Myo-inositol (45) has been used in addressing human conditions linked to insulin resistance, as well as in the management of diabetic complications, including neuropathy, nephropathy, and cataracts [101]. Neophytadiene (26), a compound related to carotene metabolism, is recognized by its antimicrobial [102] activity.

The presence of cyclitols is striking, accounting for 5% of the identified compounds. Cyclitols are cyclic polyols derived from glucose. Due to their low toxicity and positive impacts on managing insulin resistance, obesity, and polycystic ovary syndrome, these molecules are gaining attention in the dietary supplement industry [103].

GC-MS analysis confirmed that the extract was rich in phenolic antioxidant compounds, including 2,4-di-tert-butylphenol (16), 2,6-di-tert-butyl-4-methylphenol (17), 3-hydroxycinnamic acid (23), p-coumaric acid (35), gallic acid (38), and 2-allyl-4-methylphenol (12). The findings validated the importance of leaf residues from T. erecta as a valuable source of bioactive compounds.

3.2.6. HPLC Analysis

The occurrence of phenolic compounds in the optimized extract (caffeic acid, chlorogenic acid, quercetin and gallic acid), previously documented in T. erecta extracts [11,50], was confirmed through HPLC analysis (Figure 5). Furthermore, the detection of malic acid was validated through HPLC, aligning with its identification using GC-MS.

The optimized extract of T. erecta leaves is particularly rich in phenolic compounds such as chlorogenic acid, quercetin, and caffeic acid. Chlorogenic acid was found to have the highest concentration regarding the dry leaves (36.0 mg/g), followed by quercetin (35.0 mg/g), caffeic acid (6.7 mg/g), malic acid (6.7 mg/g), and gallic acid (2.4 mg/g). These results were consistent with a previous report that identified chlorogenic acid as the dominant compound in T. erecta leaf extracts [104].

The leaf extracts of two T. erecta cultivars, Pusa Narangi Gainda (PNG) and Pusa Basanti Gainda (PBG), exhibited lower concentrations of gallic acid (0.05 mg/g), caffeic acid (0.49–0.67 mg/g), and quercetin (1.35–1.91 mg/g) [50] compared to those found for the optimized extract according to HPLC quantification. This illustrates the significance of the outlined optimization strategy for the efficient extraction of phenols from T. erecta leaves.

Phenolic compounds here identified are significant to the food industry due to their antioxidant, anti-inflammatory, and anti-hyperglycemic effects [11,42], suggesting that the optimized extract of T. erecta could be valuable for developing new value-added alimentary products.

4. Conclusions

T. erecta flowers have been widely studied as a source of bioactive molecules, primarily phenols and carotenoids. However, the species’ leaves have received little attention and are currently considered a low-value waste.

This research integrates, for the first time, agronomic aspects like fertilization type and the timing of the harvest during various phenological stages with chemical extraction techniques to maximize the phenol content of the leaves of T. erecta. Soxhlet extraction (0.01 g FD leaf powder/mL) in plants harvested at the blooming stage leads to high yields of phenols. Major identified polyphenols in the optimized extract included 2,4-di-tert-butylphenol, 2,6-di-tert-butyl-4-methylphenol, 3-hydroxycinnamic acid, p-coumaric acid, gallic acid, 2-allyl-4-methylphenol, quercetin, chlorogenic acid, and caffeic acid, according to GC-MS and HPLC analyses.

This study demonstrated that fertilization with E. foetida (10 t ha⁻¹) represents a better option than chemical fertilization. The irrational use of agrochemicals represents an economic burden for farmers, contributing to soil degradation and greenhouse gas emissions. Transitioning to fertilization with Eisenia foetida is a more cost-effective option, thus ensuring the viability of the crops for farmers. The ideal time to harvest this species occurs 90 days after sowing, coinciding with the bloom of T. erecta. During this timeframe, farmers can reap dual benefits by selling bouquets while utilizing the leaves to produce extracts with potentially high phenolic content.

The optimization conducted in this study enabled the identification of optimal conditions at the laboratory scale, which could be a foundation for scaling up to a pilot level by considering additional factors like energy use, temperature fluctuations, equipment performance, etc. Further research is required to evaluate the antioxidant properties of the optimized extract produced at a pilot scale, compared to those derived from T. erecta flowers and food antioxidant additives, to determine its potential commercial value for the food industry.