Rapid, simple and accessible phytonutritional assessment methods in agricultural research

Abstract

BACKGROUND

Agricultural research with respect to studying the effects of genetic background and environmental conditions on plant and commercial parameters is ample. Nevertheless, data on the effects on nutritional value and phytonutritional composition are scarce, despite its significance to the consumer, as well as its importance in providing nutritional security. The present study aimed to simplify, unify and make accessible a set of spectrophotometric nutritional assessment assays, comprising a valuable tool in agricultural research. Efforts were made to provide simple, rapid, small‐scale and cost‐effective protocols, aiming to minimize resources and time, learning‐curve and erring.

RESULTS

We included assays alongside literature‐compliant values for antioxidant capacity (ABTS, DPPH, FRAP), polyphenols and flavonoids (free, conjugated, bound, total), lignans, anthocyanins, tocopherols, phytosterols, glucosinolates, carotenoids, betalains, saponins, phlorotannin, phycobilin, vitamin C, total oil and soluble protein content, in addition to antinutrients‐ oxalic acid and phytic acid. Calibration curves of recommended standards are included and workflows are demonstrated. Guidelines are also provided for protocol adjustments.

CONCLUSION

This work thus compiles a broad arsenal of rigorous methods, rendering a good spectrum of nutritional assessment protocols, including nutriphytochemical composition alongside a basic bioactivity assessment (antioxidant measurements). These methods are recommended in evaluating the effects of agricultural management practices, such as water and mineral nutrition availability or postharvest storage, or as part of breeding programs, such as in research or commercial R&D laboratories that do not have expertise in nutritional analysis, and require an accessible high‐throughput nutritional assessment tool. © 2025 The Author(s). Journal of the Science of Food and Agriculture published by John Wiley & Sons Ltd on behalf of Society of Chemical Industry.

INTRODUCTION

Over the past decade, there has been a significant increase in consumer awareness of food's health‐promoting effects, consequently leading to an increased demand for nutritionally beneficial functional foods, rich in plant‐derived bioactive phytochemicals. This trend is accompanied by concomitant rising consumer expense on fruit and vegetables, with US household expenditure (being the largest globally in total) increasing from USD $876 in 2019 to $1099 in 2022, ¹ as well as an increasing demand for functional foods, projected to reach USD 280 billion by 2025. ² Inevitably, growers comply with demands for high‐quality, nutritionally beneficial products, leveraging the health benefits as a marketing driving force, resulting in much emphasis on crop health‐related and nutritional values, at the same time as investing more resources in research related to crop's health properties.

In face of current and forecasted major climatic and environmental changes, modern agriculture is undergoing constant evolution, re‐assessing existing practices and developing new strategies to withstand the emerging challenges. Extreme weather conditions, diminishing arable land, declining water quality and quantity, shifts in traditional growth regions, and heightened plant protection issues, all pose major difficulties. These factors necessitate the breeding of new varieties to adapt to the changing environment and stress conditions. Maintaining efficient agriculture is essential in light of food and nutritional security challenges with respect to feeding an increasing population with diminishing resources. This requires the production of food that is both abundant and nutritious, encompassing primary metabolites (carbohydrates, lipids, protein, vitamins and minerals) alongside specialized metabolites essential for maintaining proper health, as recommended by the World Health Organization. ³

The effects of both genetic and environmental factors on phytochemical composition are well established. Screenings of genetic collections have demonstrated the vast inherent genetic variability in phytonutritional composition, underscoring the critical role of genetic background in determining nutritional quality. ⁴ , ⁵ As for the effects of environmental conditions, previous works including our own have indicated the impacts of mineral nutrition, as well as micronutrient availability on phytonutritional composition. ⁶ , ⁷ , ⁸ Additionally, irrigation practices and water availability, including the use of marginal, saline, desalinated and wastewater, have also been observed to possess major effects on nutritional quality. ⁹ , ¹⁰ Further studies have revealed the influence of other practices on phytochemical composition, including pruning, harvest time, sowing time and spacing, cultivation conditions (e.g. region and temperature), as well as the use of rootstocks, to encourage vigor, vegetation, fruit quality and pest tolerance. ¹¹ , ¹²

Despite the crucial role of each of these individual factors, the most prominent impacts are still attributed to the interaction between genetics and environment, such as the differential effect of environmental conditions on the nutritional profile being genotype‐dependent. These interactions are expected to become critical when introducing existing and new varieties from breeding programs into new growth areas as a result of changing climatic conditions. ¹³ , ¹⁴

In this context, it is important to clarify and unify the terminology of ‘fat’ versus ‘oil’ as used in the present study. The primary distinction lies in their physical state at room temperature: fats are typically solid or semi‐solid, whereas oils are liquid. ¹⁵ The term ‘oil content’ is commonly used in scientific literature; therefore, we adopted it in this study to maintain consistency with standard terminology. Moreover, given the agricultural focus of this study (i.e. primarily concerning plant‐derived materials), the analyzed substances are predominantly oils. Although exceptions exist (e.g. cocoa, shea and mango butters), our methods and terminology remain applicable. Thus, even when referring to ‘oil,’ the methodology may also be relevant to fats in a broader context.

Collectively, these trends present a paramount need for the evaluation of crop phytonutritional quality, including the levels of antinutrients, from both nutritional security and consumer demand standpoints. To face this need, a simple tool is thus required, for assessing the influence of genetic and environmental factors on food quality; for example, for evaluating the effect of new varieties from breeding programs, new cultivation environments, novel agricultural practices and their interactions. However, to fulfill its purpose, such a tool should be accessible, easy to learn and perform, quick, require minimal specialized manpower or expensive equipment, and be cost‐effective, thereby enabling a preliminary screening of a large number of samples, or effectively comparing management or postharvest treatments.

The present study aimed to provide a comprehensive tool to meet the growing demand for phytonutritional evaluation. The methods compiled here are designed to be user‐friendly, efficient and accessible, thus facilitating broader a application of phytonutritional assessments in various agricultural contexts. Although not offering high‐resolution, they are simple and economical, yet yield valuable information, only requiring basic equipment, such as a spectrophotometer/plate reader and a benchtop centrifuge, alongside readily available reagents and solvent. We have also addressed other related concepts, including sampling, sample volume scaling and the analysis of more complex matrices such as oilseeds.

MATERIALS AND METHODS

Plant material and sampling

Recommended sampling is ideally based on six biological replicates, each of at least six to 10 fruits, and no less than four replicates. Fruits from a few plants can be combined into one biological replicate; however, each replicate should be from a different group of plants. Additionally, a small part can be sampled from each fruit, nevertheless, it should be taken from across the fruit (not only the top, bottom, middle, etc.), and similarly from all fruits.

Our presented data comprise measurements performed in foodstuff purchased in local markets including argan (oil and seed), blackberry fruit, pitaya fruit, tomato (fruit and leaf), pollen, goji berry fruit, cocoa beans, pomegranate fruit, date (fruit and seed) avocado fruit, honey, local herbs (Malva pauliflora, Rumex cypus, Silybum marianum, Gundelia tournefortii, Portulaca oleracea) and algae (Chlorella spp, Hematococcus spp. Gracilaria conferta), or grown locally as reported previously, including almond (kernel and oil), ¹⁶ tef seed, ⁷ sesame (paste, oil and seed), ¹⁷ sweetpotato (tuber and leaf), ¹⁷ olive oil ⁹ and nigella (oil and seed). Analyses were carried out on plant material at optimum freshness.

Sample preparation

To stop all enzymatic and chemical reactions, as well as to normalize for water content, once samples arrived at the laboratory, they were immediately flash frozen in liquid nitrogen, followed by two‐to‐three‐day lyophilization. Once dried, all samples were kept at −20°C until analysis. Next, dried samples were ground to a fine uniform powder, using an IKA grinder (IKA, Staufen, Germany), or a coffee grinder, pre‐cooled with liquid nitrogen.

Powder should be fine and uniform to ensure optimized extraction and to avoid large variance within results as a result of differences in particle size among samples.

All reagents and solvents used in the present study were purchased from Merck KGaA (Rehovot, Israel).

Calculations

Calculations are common to all analyses, unless otherwise specified.

m: sample mass taken for analysis (g).

V: final extract volume (mL).

D: extract dilution factor ($=\frac{\text{Final volume after dilution}}{\text{Initial volume taken for dilution}}$).

A: sample absorbance.

Intercept, slope, as taken from the standard's calibration curve

$$\text{concentration}\left(\frac{\mathrm{mg}}{\mathrm{g}\text{sample}}\right)=\frac{\left(A-\mathit{\text{intercept}}\right)*D*V}{\mathit{\text{Slope}}*m}$$

This calculation is derived from the principles of utilizing a calibration curve to determine the analyte's concentration in an unknown sample. In general, a calibration curve is based on at least five data points consisting of known concentration of the analyte or an external standard which was chosen based on chemical resemblance. Readings of the known concentrations are then used to draw a linear function, Y = aX + b, consisting of absorbance (A; Y axis) versus concentrations (X axis), where a is the slope and b is the intercept. Then, using the linear equation, the concentration of an unknown sample can be calculated by measuring its absorbance, using the same slope and intercept data. Understandably, this calculation also accounts for the sample weight taken into analysis (m), total sample volume (V) and any dilution steps (D). The reader is further referred to previous works describing the principles of calibration curves. ¹⁸ , ¹⁹

Antioxidant activity (ABTS, DPPH and FRAP)

Sample extraction

For this, 100 mg of powdered sample was weighed and 1 mL of 80% aqueous ethanol solution in distilled water (DW) (V/V) was added. The sample was vortexed, thermoshaken (AccuTerm; Rocket Software, Waltham, MA, USA) for 10 min at 25°C and centrifuged (17 000 × g for 5 min at room temperature) and supernatant (extract, or standard) was used for analysis as described in the Supporting information (Fig. S1).

ABTS antioxidant analysis

2,2′‐Azino‐bis (3‐ethylbenzothiazoline‐6‐sulfonic acid) (ABTS) antioxidant activity was determined as reported previously. ⁷ , ¹⁶ In a 96‐well plate, 50 μL of extract (or standard) was combined with 150 μL of ABTS solution. After 30 min, absorbance was measured in a plate reader (Multiskan; Thermo Fisher Scientific, Waltham, MA, USA) at 745 nm against a blank sample. Trolox was used as a standard for a 5‐point calibration curve (1 mg mL⁻¹ stock solution diluted to 0–0.1 mg mL⁻¹ in 80% ethanol). The results were expressed as mg of Trolox equivalent (TE) per g sample. In many cases, the results for this analysis are presented as micromol TE g⁻¹ sample, and, to convert mg TE g⁻¹ into micromole, the TE g⁻¹ numeric results can be multiplied by 3.9954 (or 4) (with the molecular weight for ABTS being 250.29 g mol⁻¹).

ABTS solution was prepared as follows: In a 100‐mL glass volumetric flask, 6.6 mg of potassium persulfate was weighed, and 25.1 g of ABTS radical was added, followed by 15 mL of DW. The bottle was put on a stirring plate, and heat was set to 45 °C. After 1 min, 65 μL of H₂SO₄ was slowly pipetted when stirring. After 30 s, 35 mL of ethanol was added [analytical research (AR) grade]. The solution was stirred for another 60 min at 45 °C, with the cap untightened to prevent an increase in pressure, and then maintained at 4 °C overnight before use. ²⁰ The solution can be kept at 4 °C for 1 month and the absorbance, when checked before use, should be in the range 0.8–1.5.

DPPH antioxidant activity

2,2‐diphenyl‐1‐picrylhydrazyl (DPPH) radical scavenging activity was determined as reported previously. ⁷ , ¹⁶ In a 96‐well plate, 100 μL of extract (or standard) was combined with 150 μL of DPPH solution. After 30 min, absorbance was measured at 520 nm against a blank sample. Trolox was used as a standard for a 5‐point calibration curve (1 mg mL⁻¹ stock solution diluted to 0–0.1 mg mL⁻¹ in 80% ethanol). The results were expressed as mg of TE per g sample.

DPPH solution was prepared as 0.182 mg mL⁻¹ in 80% aqueous ethanol solution. The DPPH solution can be kept at 4 °C for 1 month and the absorbance, when checked before use, should be in the range 0.8–1.5 (although the reagent would work well in this entire absorbance range, values above 1 are preferable).

FRAP antioxidant activity

Ferric reducing ability of plasma (FRAP) antioxidant activity was determined as previously reported previously. ⁷ , ¹⁶ In a 96‐well plate, 50 μL of extract (or standard) was combined with 200 μL of FRAP solution. After 30 min, the absorbance was measured at 593 nm, against a blank sample. Trolox was used as a standard for a 5‐point calibration curve (1 mg mL⁻¹ stock solution diluted to 0–0.1 mg mL⁻¹ in 80% ethanol). The results were expressed as mg g⁻¹ TE per sample.

FRAP solution was prepared as follows. In a 15‐mL tube, 10 mL of acetate buffer (0.3 m) was added, followed by 1 mL of TPTZ solution (10 mm) and 1 mL of ferric chloride solution (20 mm). The solution was made fresh on the day of the analysis and was heated to 37 °C for 10 min before analysis.

Acetate buffer (pH 3.6) was prepared as follows. In a 250‐mL glass volumetric flask, 455.3 mg of sodium acetate was weighed and 50 mL of double distilled water (DDW) was added, followed by 3.97 mL of glacial acetic acid. DDW was then added to a final volume of 250 mL and the pH was adjusted to 3.6.

10 mm TPTZ solution in 40 mM HCl was prepared. In a 100‐mL glass volumetric flask, 156.2 mg of TPTZ was weighed, followed by 50 mL of DW. Next, 170 μL of HCl 12 n was added and DW to a final volume of 100 mL.

10 mm ferric chloride solution was prepared. In a 50‐mL glass volumetric flask, 270.03 mg of FeCl₃ was weighed, and DW was added to a final volume of 50 mL.

Total polyphenol content

Total free phenolic content is usually referred to in the literature as total phenolic content, or TPC. At the same time, the total phenolic content should also include the conjugated as well as the bound forms of phenolic compounds, and thus the term is inaccurate. In the present study, we refer to total phenolics as the calculated sum of free, conjugated and bound compounds (Fig. 1A).

Figure 1.

Workflow for polyphenol and flavonoid analysis. (A) Total polyphenols. (B) Total flavonoids. (C) Combined total polyphenol+ total flavonoid analysis.

For extraction of phenolic compounds, 50 mg of powder was weighed, and 1 mL of 80% aqueous ethanol solution in DW (V/V) was added. Tubes were vortexed, thermoshaken for 10 min at 25 °C and centrifuged at 17 000 × g for 5 min. Supernatant was analyzed for free and conjugated phenolic compounds, whereas the pellet was analyzed for bound phenolic compounds.

Free phenolic content analysis was performed using Folin–Ciocalteu colorimetric method as reported previously. ⁷ , ¹⁶ In a 96‐well plate, 30 μL of extract (or standard) was mixed with 100 μL of 0.2 n Folin solution in 80% ethanol and 100 μL of sodium carbonate 2% (W/V) solution in DDW. Plate was covered and incubated for 90 min. The absorbance was measured at 765 nm, against a blank sample. Gallic acid (GA) was used as a standard for a 5‐point calibration curve (3 mg mL⁻¹ stock solution diluted to 0.018–0.3 mg mL⁻¹ and a 0 mg mL⁻¹ blank in 80% ethanol solution) and the results were expressed as mg GA equivalent (GAE) per g sample.

For conjugated phenolic content, 100 μL extract was combined with 100 μL of 4 m HCl solution, and incubated at 80 °C for 30 min. ²¹ Then, 100 μL of NaOH 4 m solution in DW was then added (to achieve pH 4.0–4.5) and the sample was vortexed. Readings were performed in a 96‐well plate, as described for free phenolic content, also using GA as a standard. The calculation accounted for extract dilution by multiplying in dilution factor (D = 300/100).

For bound phenolic content, the pellet after free phenolic extraction was dried in a SpeedVac (Thermo Fisher Scientific) for 15 min. Next, 750 μL of NaOH 2 m in DW was added and the sample was vortexed and then thermoshaken for 70 min at 70°C and 1000 r.p.m. Next, 250 μL HCl of 6 m solution in DW was then added (to achieve pH 4.0–4.5), and the sample was vortexed and then centrifuged at 17 000 × g for 5 min. Bound phenolic content analysis was performed using the Folin–Ciocalteu colorimetric method as reported previously. ⁷ , ¹⁶ Readings were performed in a 96‐well plate, as described for free phenolic content, also using GA as a standard.

Total flavonoid content

Because flavonoids are also phenolic compounds, their extraction steps are essentially similar to those of polyphenols, whereas analysis is carried out separately to quantify each group using a different colorimetric method (Fig. 1B). A combined protocol is also suggested (Fig. 1C).

For extraction of flavonoids, a 50‐mg powdered sample was weighed, and 1 mL 80% aqueous ethanol solution in DW (V/V) was added. Tubes were vortexed, thermoshaken for 10 min at 25 °C and centrifuged at 17 000 × g for 5 min. Supernatant was analyzed for free and conjugated phenolic compounds, whereas the pellet was analyzed for bound phenolic compounds.

Free flavonoid content was analyzed as reported previously. ⁷ , ²² in an Eppendorf tube, 600 μL of DW, 45 μL of NaNO₂ (0.05 g mL⁻¹ in DW) and 150 μL of extract (or standard) were added, then the sample was vortexed, and 90 μL of AlCl₃ (0.1 g mL⁻¹ in DW) was added, followed by vortexing and thermoshaken for 5 min at 25 °C and 1000 r.p.m. Then, 300 μL of DW and 300 μL of NaOH 2 m in DW were added and samples vortexed and centrifuged for 5 min at 17 000 × g. Next, 200 μL of each sample was transferred into a 96‐well plate, and read at 415 nm if using quercetin as standard, orat 506 nm for rutin. Quantification was performed based on 8‐point quercetin or rutin calibration curve (1 mg mL⁻¹ stock solution diluted to 0.05–0.9 mg mL⁻¹ and a 0 mg mL⁻¹ blank), and the results are expressed as mg quercetin equivalent (QE) or rutin equivalent (RE) g⁻¹ per sample.

For conjugated flavonoid content, 100 μL of extract was combined with 100 μL of 4 m HCl solution and incubated at 80 °C for 30 min. ²¹ Then, 100 μL of NaOH 4 m solution in DW was added (to achieve pH 4.0–4.5) and the sample was vortexed. Readings were performed in a 96‐well plate, as described for free phenolic content, also using GA as a standard. The calculation accounted for extract dilution by multiplying in dilution factor (D = 300/100).

For bound flavonoid content, the pellet after free phenolic extraction was dried in a SpeedVac for 15 min. Next, 750 μL of NaOH 2 m in DW was added, the sample was vortexed and thermoshaken for 70 min at 70 °C and 1000 r.p.m. Then, 250 μL of HCl 6 m solution in DW was then added (to achieve pH 4.0–4.5), the sample was vortexed and centrifuged for 5 min at 17 000 × g. Bound phenolic content analysis was performed using the Folin–Ciocalteu colorimetric method as reported previously. ⁷ , ¹⁶ Readings were performed in a 96‐well plate, as described for free phenolic content, also using GA as a standard.

Total lignan content

Free and bound lignans were analyzed as reported previously. ²³ , ²⁴ For free lignans, 50 mg of powder was weighed, and 1 mL of 80% ethanol solution was added. The sample was then vortexed, thermoshaken for 10 min at 25 °C, and centrifuged for 5 min at 17 000 × g. Then, 200 μL of supernatant was transferred to a 96‐well plate and read at 280 nm.

For bound lignans, the pellet after free lignan extraction was dried in a SpeedVac for 15 min. Next, 300 μL of HCl 4 m in DW was added the sample was vortexed and thermoshaken for 60 min at 80 °C and 700 r.p.m. Then, 300 μL of NaOH 4 m solution in DW was added and the sample was vortexed and centrifuged for 5 min at 17 000 × g. Next, 200 μL of each sample was transferred into a 96‐well plate and read at 280 nm. Quantification was performed based on 14‐point sesamol calibration curve (0.00051–0.222 mg mL⁻¹ and a 0 mg mL⁻¹ blank) and the results expressed as mg sesamol equivalent (SE) g⁻¹ per sample.

Because the extract is similar to polyphenol analysis, these analyses can be combined, using the same extract (Fig. 2).

Figure 2.

Analyses that can be performed using the same 80% ethanol extract.

Total anthocyanin content

Analysis of total anthocyanin content was carried out as reported previously. ²⁵ For this, 50 mg of powder was weighed, and 1 mL of 80% ethanol solution in DW (V/V) was added. Tubes were vortexed, thermoshaken for 10 min at 25 °C, sonicated for 5 min at 25 °C and centrifuged at 17 000 × g for 5 min. In one well of a 96‐well plate, 20 μL of extract (supernatant) was mixed with 180 μL of HCl (2 n in DW), and, in another well, a 20‐μL aliquot of the same extract was mixed with 180 μL of DW). The absorbance of both samples was read at 520 and 700 nm. Cyanidin‐3‐glucoside (C3G) was used as a standard for a 12‐point calibration curve (1 mg mL⁻¹ stock solution diluted to 0.017–1 mg mL⁻¹ and a 0 mg mL⁻¹ blank in 80% ethanol solution). For C3G, 20 μL of standard solution were added to 180 μL of HCl (2 n in DW (colored), whereas another 20 μL of the same solution were added to 180 μL of NaOH 2 n in DW (colorless; the standard is acidic and the addition of NaOH neutralizes it, yielding a pH level at which the standard is colorless). Results were expressed as mg C3G equivalent (C3GE) g⁻¹ per sample.

For this protocol, A was calculated as:

$$A=\left({\mathrm{A}}_{520\mathrm{HCl}}-{\mathrm{A}}_{700\mathrm{HCl}}\right)-\left({\mathrm{A}}_{520\mathrm{DW}}-{\mathrm{A}}_{700\mathrm{DW}}\right)$$

Notably, the extraction for this analysis is similar to that of polyphenol analysis, allowing both these analyses to be carried out simultaneously, using the same extract (Fig. 2).

Total tocopherol content

Analysis of total tocopherol content was conducted as reported previously. ⁹ , ¹⁶ For this, 50 mg of powder was weighed and 1 mL of isopropanol (IPA) was added. Tubes were vortexed, thermoshaken for 10 min at 25 °C and 700 r.p.m. and then centrifuged at 17 000 × g for 5 min. Next, 750 μL of supernatant (or standard) was transferred into a new Eppendorf tube into which 200 μL of FeCl₃ solution (0.2% W/V in IPA) was added and vortexed, and 200 μL of 2,2′‐dipiridyl (0.2% W/V in 3:1 IPA: ethanol solution) was added and vortexed. Samples were thermoshaken for 20 min at 25 °C and 700 r.p.m. Then, 200 μL was transferred into a 96‐well plate and read at 515 nm. ⍺‐tocopherol was used as a standard for a 5‐point calibration curve (1 mg mL⁻¹ stock solution diluted to 0.00625–0.1 mg mL⁻¹ and a 0 mg mL⁻¹ blank in IPA) and the results were expressed as mg ⍺‐tocopherol equivalent (TOE) g⁻¹ per sample.

Total phytosterol content

For the analysis of total phytosterol content, ¹⁶ 50 mg of powder was weighed, and 1 mL of ethyl acetate was added. Tubes were vortexed, thermoshaken for 10 min at 25 °C and 700 r.p.m. and centrifuged at 17 000 × g for 5 min. In a 96‐well plate, 100 μL of extract (or standard) was mixed with 100 μL of Liebermann‐Burchard (LB) reagent, plates were covered with a sticker and then read at 675 nm after 90 min. ꞵ‐sitosterol was used as a standard for a 5‐point calibration curve (10 mg mL⁻¹ stock solution diluted to 0.125–2 mg mL⁻¹ and a 0 mg mL⁻¹ blank in ethyl acetate) and the results were expressed as mg b‐sitosterol equivalent (SE) g⁻¹ per sample.

LB reagent was prepared as follows. In an ice‐water bath mounted on a stirring plate, 10 mL of pre‐cooled acetic anhydride was added into a 20 mL glass vial, followed by slow addition of 1 mL of H₂SO₄ with constant stirring, and this was maintained at −20 °C pending analysis.

Total glucosinolate content

Total glucosinolate content was analyzed as repored previously. ²⁶ For this, 50 mg of powder was weighed, and 1 mL of 80% ethanol solution in DW (V/V) was added. Tubes were vortexed, thermoshaken for 10 min at 25 °C and centrifuged at 17 000 × g for 5 min. In a 96‐well plate, 50 μL of extract (or standard) was mixed with 150 μL of 2 mm NaPdCl₄ solution in DW with 0.17% HCl (prepared by mixing 5.9 mg of NaPdCl₄ with 10 mL of DW and adding 17 μL of concentrated HCl). The plate was covered and incubated for 60 min. Absorbance was measured at 425 nm, against a blank sample. Sinigrin was used as a standard for an 8‐point calibration curve (3 mg mL⁻¹ stock solution diluted to 3–0.0625 mg mL⁻¹ and a 0 mg mL⁻¹ blank in 80% ethanol solution) and the results were expressed as mg sinigrin equivalent g⁻¹ per sample.

Because the extraction protocol is similar to that of polyphenol analysis, both these can be performed at once, using the same extract (Fig. 2).

Oxalic acid

Oxalic acid analysis was performed as reported previously. ²⁷ , ²⁸ For tbis, 50 mg of powdered sample was weighed, and 1 mL of HCl (2 n in DW) was added. Tubes were vortexed, thermoshaken for 60 min at 25 °C and 700 r.p.m. and centrifuged at 17 000 × g for 30 min. In a new Eppendorf tube 70 μL of the middle phase (or standard) was mixed with 700 μL of potassium permanganate solution (0.2 mg mL⁻¹ in DW) and the samples were vortexed, followed by a 5 min centrifugation at 17 000 × g. Then, a 200‐μL sample was transferred into a 96‐well plate and read at 524 nm. Oxalic acid was used as a standard for a 7‐point calibration curve (5 mg mL⁻¹ stock solution diluted to 0.1563–5 mg mL⁻¹ and a 0 mg mL⁻¹ blank in HCl 2 n) and the results were expressed as mg oxalic acid per g sample.

Phytic acid

Phytic acid analysiss was preformed as reported previously. ²⁹ , ³⁰ For this, 50 mg of powder was weighed, and 1 mL of HCl 2 n was added. Tubes were vortexed, thermoshaken for 60 min at 25 °C and 700 r.p.m. and centrifuged at 17 000 × g for 30 min. In an Eppendorf tube, 30 μL of the middle phase (or standard) was mixed with 800 μL of Wade reagent and samples were vortexed, followed by a 5 min centrifugation at 17 000 × g. Then, 200‐μL samples were transferred into a 96‐well plate and read at 500 nm. Phytic acid was used as a standard for a 5‐point calibration curve (20 mg mL⁻¹ stock solution (made by adding 40 ul of phytic acid standard 50% to 960 μL of 2 n HCl) diluted to 7–20 mg mL⁻¹ and a 0 mg mL⁻¹ blank in HCl 2 n) and the results were expressed as mg phytic acid g⁻¹ per sample.

Wade reagent was prepared as follows. In a 50‐mL tube, 150 mg of sulfosalicylic acid was weighed, together with 15 mg of ferric chloride hexahydrate and 50 mL of DW was added.

Because the extract for phytic acid is similar to that for oxalic acid analysis, they can both be carried out at the same time, using the same extract (see Supporting information, Fig. S2).

Total carotenoid contents

For analyzing total carotenoid content, ³¹ 100 mg of powdered sample was weighed, and 0.5 mL of hexane:acetone solution (3:2, v/v) was added. Tubes were vortexed, and then centrifuged at 17 000 × g for 10 min. Next, 200 μL of extract (or standard) was transferred into a different tube, and 0.8 mL of cold DW (4 °C) was added, followed by 0.7 mL of hexane. Tubes were centrifuged for 2 min, and 200 μL of supernatant (upper phase) was read in a 96‐well plate at 446 nm. β‐carotene was used as a standard for a 5‐point calibration curve (1 mg mL⁻¹ stock solution diluted to 0.00312–0.1 mg mL⁻¹ and a 0 mg mL⁻¹ blank in hexane:acetone solution [1:1, v/v](and the results were expressed as mg β‐carotene g⁻¹ per sample). Note that, for calculation, the extract volume is 1 mL.

Total betalain content

For analyzing total betalains (betacyanins + betaxanthins) content, ³² , ³³ 50 mg powder was weighed, and 1 mL of DW was added. Tubes were vortexed, thermoshaken for 10 min at 25 °C and 700 r.p.m. and centrifuged at 17 000 × g for 5 min.Then, 200 μL supernatant (or standard) were transferred into a 96‐well plate and read at 535 (for betacyanins) and 483 nm (for betaxanthins). Betanin was used as a standard for a 5‐point calibration curve (30 mg/mL stock solution diluted to 0.9375–15 mg/mL and a 0 mg/mL blank in ethanol 80%), and the results were expressed as mg betanin equivalent (BE) per g sample.

Total saponin content

Total saponin content analysis was carried out as reported previously. ³⁴ For this, 50 mg of powder was weighed, and 1 mL of 80% ethanol solution in DW (V/V) was added. Tubes were vortexed, thermoshaken for 10 min at 25 °C and 700 r.p.m. and centrifuged at 17 000 × g for 5 min. Then, 15 μL of the supernatant (or standard) were transferred into a new Eppendorf tube kept on ice, to which 100 μL of vanillin solution (8% (W/V) in ethanol) and 800 μL of H₂SO₄ were added. The sample was thermoshaken at 4 °C for 5 min, and then temperature was set to 60 °C for another 20 min. Tubes were let cool to room temperature, and 200 μL of sample was then transferred into a 96‐well plate and read at 560 nm. Saponin or aescin was used as a standard for a 9‐point calibration curve (30 and 15 mg mL⁻¹ stock solution, respectively, diluted to 0.234–30 mg mL⁻¹ and a 0 mg mL⁻¹ blank in ethanol 80% solution), and the results were expressed as mg saponin aescin equivalent g⁻¹ per sample. Because the extract procedure resembles that of polyphenol analysis, the analyses can be carried out together using the same extract (see Supporting information, Fig. S2).

Total phlorotannin content

Total phlorotannin content analysis was carried out as reported previously. ³⁵ For this, 50 mg of algae powder (prepared by grinding the algal sample followed by 10 min sonication at room temperature) was weighed, and 1 mL of 80% ethanol solution in DW (v/v) was added. Tubes were vortexed, and centrifuged at 17 000 × g for 5 min. In a 96‐well plate 30 μL of extract (or standard) was mixed with 100 μL of 0.2 n Folin solution in 80% ethanol and 100 μL of sodium carbonate 2% (W/V) solution in DDW. Plate was covered and incubated for 90 min. The absorbance was measured at 765 nm, against a blank sample. Phloroglucinol was used as a standard for a 5‐point calibration curve (3 mg mL⁻¹ stock solution diluted to 0.018–0.3 mg mL⁻¹ and a 0 mg mL⁻¹ blank in 80% ethanol solution) and the results were expressed as mg phloroglucinol equivalent (PGE) g⁻¹ per sample.

Total phycobilin content

Total phycobilin (phycoerythrin and phycocyanin) content was analyzed. ³⁶ For this, 50 mg of algae powder was weighed, and 1 mL of phosphate‐buffered saline (PBS) was added. Tubes were vortexed, termoshaken at 25 °C for 20 min, and centrifuged at 17 000 × g for 5 min. In a 96‐well plate, 20 μL of extract was pipetted, and absorbance was measured at 455, 564, 592, 618 and 645 nm, against a blank sample. The calculation was performd as follows:

$$\text{Phycoerythrin}\left(\mathrm{mg}{\mathrm{mL}}^{–1}\right)=\left[\left(A_{564}-A_{592}\right)-\left(A_{455}-A_{592}\right)\times 0.20\right]\times 0.12$$

$$\text{Phycocyanin}\left(\mathrm{mg}{\mathrm{mL}}^{–1}\right)=\left[\left(A_{618}-A_{645}\right)-\left(A_{592}-A_{645}\right)\times 0.15\right]\times 0.15$$

Vitamin C

Vitamin C contents was quantified as reported previously. ³⁷ , ³⁸ For this, 100 mg of powder was weighed (m) and 1 mL (V1) of 2% HCl in DW solution (v/v) was added. Tubes were vortexed and centrifuged at 17 000 × g for 5 min. Next, 0.3 mL of the supernatant (V2) were transferred into a glass container, and tittered with 25 mg mL⁻¹ 2,6‐dichlorophenolindophenol (DCPIP) solution in DW to a steady light pink color, and volume used for titration was recorded (X). Ascorbic acid standard [1 mg mL⁻¹ (C1) in 2% HCl in DW (v/v)] was used as a standard, and 0.5 mL (V3) was titrated, and the titration volume was recorded (Y), with the results expressed as mg ascorbic acid g⁻¹ per sample.

The calculation was performed as follows:

$$\begin{array}{l}\text{Ascorbic acid content}\left(\frac{\mathrm{mg}}{\mathrm{g}\text{sample}}\right)\\ =\frac{V3\left(\mathrm{mL}\right)\times C1\left(\frac{\mathrm{mg}}{\mathrm{mL}}\right)\times X\left(\mathrm{mL}\right)\times V1\left(\mathrm{mL}\right)}{Y\left(\mathrm{mL}\right)\times V2\left(\mathrm{mL}\right)\times m\left(\mathrm{g}\right)}\end{array}$$

This analysis can be combined with oxalic and phytic acid analysis, by using the same extract and diluting it 1:3, from a concentration of HCl 2N to HCl 2% (see Supporting information, Fig. S2).

Another reliable assay for ascorbic acid analysis has been suggested, by blocking the ascorbic acid content with metaphosphoric acid. ³⁹

Oil content

For oil content, 200 mg of powder was weighed (M1), 0.5 mL of hexane was added and the sample was vortexed followed by centrifugation at 17 000 × g for 10 min. A new Eppendorf tub was weighed (M2), to which supernatant was transferred. Hexane extraction was repeated two more times, and supernatants pooled. Hexane was evaporated either (in the hood or in a SpeedVac), until the weight was constant, and the tube containing the resulting oil was weighed (M3).

The calculation was performed as follows:

$$\text{Oil}\left(\%\right)=\frac{\left(M3-M2\right)\times 100}{M1}$$

Soluble protein content

For soluble protein analysis, ⁴⁰ 10 mg of powder was weighed, and 1 mL of PBS buffer was added. Sample was vortexed for 30s followed sonication for 10 min at 35 °C, thermoshaken for 10 min at 35 °C, and centrifuged at 17 000 × g for 10 min. In a new Eppendorf tube, 10 μL of extract (or standard) and 200 μL of Bradford reagent were added (in that order) and the tube was allowed tiosit in the dark. After 20 min, samples were shaken for 20 s and read at 595 nm. Bovine serum albumin was used as a standard for a 7‐point calibration curve (0.077–1 mg mL⁻¹ and a 0 mg mL⁻¹ blank in PBS) the and results were expressed as mg bovine serum albumin (BSA) equivalent per g sample. For oilseed analysis, Triton‐100 at 0.008% was added to the PBS buffer. ⁴¹

Statistical analysis

Statistical analysis was performed using JMP, version 16 (https://www.jmp.com) using analysis of variance (ANOVA) model for the analysis and Tukey's test for further pairwise comparisons. Prior to applying the ANOVA model, all data were first assessed for suitability by conducting normality and homogeneity tests.

RESULTS AND DISCUSSION

The present study introduces a comprehensive set of phytonutritional assays designed to support the profiling of phytochemical composition and nutritional assessment of food products within agricultural research. These protocols provide a straightforward, accessible, rapid and cost‐effective tool for nutritional evaluation and screening, especially useful when advanced analyses are unavailable, unnecessary or impractical because of high costs, specialized equipment and skilled labor requirements. This collection includes both our own and others' published methods for analyzing nutriphytochemical composition, along with guidelines for customization as needed. Although these methods are not intended for in‐depth profiling, they nonetheless provide a broad spectrum of chemical analysis that might be challenging to achieve through other means, yielding valuable information.

The chemical and clinical properties of the discussed compounds are out of the scope of this work. For these, the reader is referred to some recent reviews which provide a good coverage of the topic, including for antioxidants, ⁴² polyphenols, ⁴³ flavonoids, ⁴⁴ lignans, ⁴⁵ anthocyanins, ⁴⁶ tocopherols, ⁴⁷ phytosterols, ⁴⁷ glucosinolates, ⁴⁸ saponins ⁴⁹ and betalains, ⁵⁰ as well as antinutrients, including oxalic and phytic acid. ⁵¹

To provide the reader with exemplary results obtained by the assays brought here, quantitative data of analyzed samples is given in Table 1, with works illustrating the importance of phytonutritional assessment, and the significance of tracking the nutritional value to improving our food and providing nutritional security. To compare the methods presented here to other available phytochemical screening assays, ensuring consistency with the methods and expressions previously noted, a comparison with results obtained using other methodologies is also provided.

Table 1.

Quantitative data obtained using the current assays, compared to data from other methodologies

	Quantitative data‐ current assay	Comparable data	Crop
ABTS	0.19–0.26 (mg TE g−1) 16	0.11 (mg TE g−1) 52	Almond (kernel)
DPPH	0.19 (mg TE g−1) 16	0.26 (mg TE g−1) 52	Almond (kernel)
FRAP	0.22 (mg TE g−1) 16	0.2 (mg TE g−1) 52	Almond (kernel)
Polyphenols	0.37 (mg GAE g−1) 16	0.11–0.42 (mg g−1) (UPLC‐MS/MS) 53	Almond (kernel)
Flavonoids	0.39 (mg g−1) 54	3.96–10.7 (HPLC‐MS) 54	Almond (kernel)
Lignans	10 740 (mg kg−1) 23	10 170 (mg kg−1) (HPLC‐PDA) 23	Sesame (seed and oil)
Anthocyanins	0.81 (mg C3G g−1) 55	1.42 (mg C3G g−1) (HPLC‐PDA) 55	Black currant (fruit)
Tocopherols	0.37 (mg g−1) 16	0.28–0.52 (mg g−1) (HPLC‐UV) 56	Almond (kernel)
Phytosterols	2.05 (mg g−1) 16	2.1 (mg g−1) (HPLC) 57	Almond (kernel)
Glucosinolates	24.25 (μmol g−1) 58	25.14 (μmol g−1) (HPLC‐DAD) 58	Camelina (seed)
Oxalic acid	9.83 (mg g−1) 59 , 60	12.72 (mg g−1) (HPLC‐PDA) 61	Spinach (leaf)
Phytic acid	322 (mg 100 g−1) 29	366 (mg 100 g−1) (Ion exchange+ spectrophotometric) 29	Infant food
Carotenoids	955 (μg g−1) 62	797 (μg g−1) (HPLC‐DAD) 62	Potato (tuber)
Betalains	6 (mg g−1) 32	6 (mg g−1) (HPLC‐PDA) 33	Pitaya (fruit)
Saponins	2.01 (mg g−1) 34	1.76 (mg g−1) (HPLC‐DAD); 1.79 (mg g−1) (HPLC‐MS) 34	Sweetpotato (tuber)

Notably, most measurements are satisfactory, except those for flavonoids, presenting a large gap between methods. Indeed, several concerns have been previously raised regarding this analysis, ⁶³ attributing the gap to the large structural variation alongside differences in sample composition. Additionally, flavonoid profiling is often performed as a targeted analysis, only measuring specific compounds; thus, their total might be lower. At the same time, it is still being widely used, also yielding consistent values, ⁵⁴ and we thus leave It to the reader to decide whether to use it or not. Another group of compounds showing result discrepancies between methods is anthocyanins, although, in this case, the HPLC assay detects higher level compared to those found using the spectrophotometric method. This trend also depends on the specific crop, and was thus stipulated to result from structural and compositional variations. ⁶⁴

In this section, we chose to bring some possible limitation and pitfalls, which the users should account for. These will be mentioned according to the workflow of the analytical work, starting with sampling and sample preparation, through extraction and various sample‐ type related aspects, and concluding with reporting.

Sampling

Some aspects are to be considered for proper sampling. Evidently, sample size required for analysis is very small because usually approximately 3–5 g is sufficient to perform the entire set of analyses described here. As the same time, it is crucial that sampling is representative, and large enough to cover as much variance, Hence, at least four to six replicates of at least three to six different plants/fruit/vegetables each (biological replicates) are required to ensure a sufficient sampling.

Another concept that is often brough up in this respect is the identity of the specific tissues within the fruit or vegetable that should be sampled; for example, skin in potato or sweetpotato, seeds such as in dragon fruit, and flavedo in citrus fruit. As a rule of thumb, conforming to Food Science perspective, we chose for analysis the edible parts, addressing the sample as if it was food, deciding which parts would then be consumed and taking these tissues into analysis. Another question that often rises is related to maturity and, here again, the sample is regarded as food, and sampled at the same maturity level it would have been consumed. In this respect, it is also important to sample all fruit/vegetable parts, avoiding sampling only the top or middle part, but rather slicing it so to sample across. Lastly, sampling a field or net house experiment should also follow the original experimental design (i.e. consider original experiment's treatments, plots, random blocks, etc.).

Sample preparation

In this work, much emphasis was put on minimizing sample weight and extraction volume, aiming to make the analyses as small‐scale as possible, increasing their high‐throughput at the same time as minimizing labor, consumables and reagents. Thus, all extracts are adjusted to 1 mL, and all reaction volumes are restricted to 2 mL, so that Eppendorf tubes can be used.

In some cases, the pellet is dried after the first stage of extraction, and is then sequentially re‐extracted (e.g. for bound flavonoids). We have evaluated the reminiscent in these cases to be insignificant, between 1% and 3% (data not shown).

Regarding centrifugation, the rule of thumb is to use the highest speed available, and adjust time so that the plant material is well precipitated on the bottom and a stable pellet is formed. This step in crucial because absorbance cannot be read in unclear, cloudy extracts.

When performing the analyses, it is important to verify that the hood is darkened and no light is turned on, because most assays are light sensitive, resulting in inaccurate data if carried out in the presence of light. Moreover, to avoid a batch effect, sample randomization is highly recommended and can be easily achieved by randomly selecting the order of the sample sequence.

Unless otherwise specified, all solutions are prepared fresh, and can be stored for a week in 4 °C. This is even more crucial in the case of standards.

Understandably, not all food/plants have all phytochemical groups, and an initial literature survey is thus recommended to decide on which assays are to be performed.

Calibration curve

In this work, we refrained from using specific extinction coefficients because these necessitate the light pathlength for the Beer–lambert law for calculating the concentration. ⁶⁵ Thus, these coefficients are usually in use for 1‐cm reading cells, whereas the presented analyses are intended for a plate reader. In a plate reader, the pathlength is shorter than 1 cm, is usually not known, and often varies between solvents ⁶⁶ and assays. As a more viable alternative, we highly recommend the use of a calibration curve as a means of qualification, yielding more accurate and consistent data, as long as performed following the same assay. In some instruments pathlength correction to 1 cm exists; however, it is not available for many of the solvents in use here, ⁶⁶ (Thermo Fisher Scientific, personal communication).

Moreover, we recommend running a calibration curve together with every set of samples, which provides accuracy by avoiding variation in conditions, solvent, solutions and reagents, etc., between samples and standard.

Because solvents are highly volatile and easily evaporated, it is generally advisable to work with up to 16 tubes at a time, which, in three technical replicates, yields 48 wells, comprising half a 96‐well plate. In addition, rapid work is also recommended to avoid variance in pathlength among samples and between sample extract and standards as a result of evaporation, compromising data quality.

Another important technical aspect that is worth mentioning is that standard volumes taken into the reaction must be similar to those of sample extract (although sample and calibration curve total volumes may differ). The calibration curve data for all methods are provided in Table 2.

Table 2.

Equations and regression coefficients for the calibration curves of various phytochemicals evaluated in this work

	Standard	Equation	R 2	Linear range (mg mL−1)
ABTS	Trolox	Y = −6.2453x + 0.9467	0.9918	0–0.1
DPPH	Trolox	Y = −8.8117x + 1.1133	0.977	0–0.1
FRAP	Trolox	Y = 10.431x + 0.099	0.9994	0–0.1
Polyphenols	Gallic acid	Y = 5.2517x + 0.0274	0.9985	0–0.3
Flavonoids	Rutin/quercetin a	Y = 3.3929x + 0.0242	0.9993	0–0.9
Lignans	Sesamol	Y = 7.0035x + 0.0582	0.9981	0–0.22
Anthocyanins	C3G	Y = 0.4716x + 0.0128	0.9921	0–1.0
Tocopherols	⍺‐tocopherol	Y = 0.629x + 0.5847	0.9891	0–0.1
Phytosterols	ꞵ‐sistosterol	Y = 0.6613x + 0.063	0.999	0–2
Glucosinolates	Sinigrin	Y = −0.1656x + 0.2738	0.9991	0–0.0625
Oxalic acid	Oxalic acid	Y = −0.3608x + 1.7459	0.9762	0–5
Phytic acid	Phytic acid	Y = −0.1094x + 2.2254	0.9992	0–20
Carotenoids	ꞵ‐carotene	Y = 0.456x‐0.39237	0.9977	0–0.1
Betalains	betacyanin	Y = −0.1019x + 0.0536	1	0–15
	betaxanthin	Y = −0.0719x + 0.0486	1	0–15
Saponins	Saponin/aescin b	Y = 0.0284x + 0.0752	0.9969	0–30
Soluble protein	BSA	Y = 0.3327x + 0.4206	0.9933	0–1

Abbreviations: C3G, cyanidin‐3‐glucoside; BSA, bovine serum albumin.

^a Presented for quercetin.

^b Presented for aescin.

When performing spectrophotometric analysis using a calibration curve, two calculation options are commonly used; the first is to subtract from each sample reading the blank reading, whereas the other is to read the blank as part of the standard calibration curve as zero standard concentration. Upon careful evaluation of these options, we decided on the latter because it is simpler and usually renders lower variance.

Standards

Most standards are readily availablel however, in the case of very expensive compounds, these can be purchased once and used for a wide‐range calibration curve, which can be further used for all future analyses.

In addition, because these analyses are often meant to read the total levels of a chemical group comprising various chemical structures, in some cases, a specific standard is better for a specific crop, better reflecting the present compounds. Thus, if a calibration curve of a specific standard results in considerably too low or too high levels, it is recommended to try a different standard.

Assay adjustments

The presented assays are potentially compatible with all plant samples. At the same time, compound levels greatly vary among crops, and thus might require empirical adjustments. Such adjustment can be made in two points along the protocol: sample mass taken into extraction, and volume/ concentration of extract taken into the reaction. Thus, prior to analysis, a preliminary screening of two to three samples is advisable to initially assess whether adjustments are required to the protocol for the specific crop or variety, and how they should be made. In case extracts are too concentrated (i.e. absorbance reading in a plate reader exceeds 1.5–2) or too diluted (readings <0.1–0.2) then extracted plant material can be reduced or increased to 10–200 mg. It should be noted, however, that smaller samples are less representative of the sampled population, and are more prone to outliers and high variance.

Alternatively, extract can be diluted (usually 1:5 or 1:10), using the same volume of diluted extract for analysis, and accounting for the dilution factor in the calculation. Another less favorable option is to take a smaller volume into the analysis, as using lower extract volumes increases data variance. Importantly, in both cases, calibration curve range or volumes must be adjusted to comply with that of sample extracts. In addition, assays can be scaled up or down according to the amount of plant material available for analysis.

Solvents

Another key determinant in these analyses is the choice of solvent. In the present study, the principles of green chemistry were applied, preferring less toxic, more environmentally‐friendly, degradable solvents. ⁶⁷ Traditional solvents were thus substituted according to the available recommendations and guidelines. ⁶⁸ , ⁶⁹ , ⁷⁰ , ⁷¹ , ⁷²

The ethanol, isopropanol or ethyl acetate used in this work are compatible with most plant and food samples. However, when samples are more apolar, the use of other, more apolar solvents is recommended (e.g. methyl‐tert‐butyl ether, 1‐pentanol, or a combination of several of these). Solvents can thus be further adjusted; however, it should be stressed that the wavelength of the assay should be adjusted accordingly. This should be done by screening the modified solution for optimal absorbance wavelength between 200 and 900 nm.

Solvent for these assays should be AR grade, and water should be DW. Higher grades [e.g. HPLC‐grade or DDW (Milli‐Q; Merck Millipore, Burlingotn, MA, USA)] are more expensive and are not required for these assays.

For many analyses, 80% ethanol solution was chosen, despite its high price, because of its relative low toxicity, and the ability to play a role as a universal solvent compatible with most specialized metabolites. Some analyses can be performed in water (e.g. anthocyanin and glucosinolates); however, using ethanol also enables enzymatic inactivation.

96‐Well plates

96‐well plates are being used for the majority of these analyses; Plate material needs to be chosen according to two main considerations. The first is the wavelength at which an assay is carried out, whereas the other is assay solvents. Measures taken in the UV range (200–400 nm) necessitate plate polymer to be UV‐compatible and have no absorbance in this range (e.g. cycloolefine plates (Greiner 655 801; Greiner Bio‐One are located in Kremsmünster, Austria)). The second consideration is plate compatibility to solvents in use because some solvents would dissolve plate polymer (e.g. acetone, ethyl acetate or acids). A good indication for plate–solvent incompatibility is the appearance of cloudiness in samples, although polymer–solvent interactions are also available from vendors. Cycloolefine also comply with these solvents, as long as the measurement is performed within 15–20 min. In all other instances, however, polystyrene plates are sufficient (e.g. Greiner 655 101).

Oilseed analysis

For oilseed samples, defatting is sometimes recommended. Nevertheless, assays brought here have been evaluated in oilseeds without a defatting pre‐treatment and found to be consistent, accurate and reproducible. All assays can thus be adjusted for oil and oilseed analysis. It should be noted that the presence of cloudiness might imply sample–solvent incompatibility, usually because of polarity issues, possibly as a result of the content of specific metabolites, and solvent substitution should be considered (e.g. isopropanol, ethyl acetate or vice versa, or a combination of these or others). It should also be kept in mind that only cold‐press oils can be analyzed for quality because, in heated, chemically extracted oils, specialized metabolites are diminished.

Statistical analysis

With an appropriate experimental design, one should have a minimum of 12 readings per treatment (four biological replicates with three technical replicates for each). To account for natural variance inherent to biological systems, prior to analysis, careful evaluation of the data to thoughtfully consider removing outliers is recommended. Moreover, when analyzing the data, it is important to be aware of possible batch effects, and account for them, or repeat the analysis if necessary.

Reporting

More uniform analytical methods and assays, alongside more uniform reporting, allow better comparison between studies, broadening the available information. For example, when different solvents are used for the same protocol, some discrepancies in the data are expected. In addition, when a different standard is used for the calibration curve reported, the results are incomparable (e.g. ascorbic acid instead of trolox, quercetin/rutin for total flavonoids, saponin/aescin for saponin, etc.). At the same time, it should be noted that, in some cases, when samples contain only one form of a group of compounds, using a comparable specific standard is advisable. Reporting the results per sample weight unit is helpful, rather than extract weight unit. In this respect, reporting the results for sample dry weight is also important because, in plant and food samples, water content may vary considerably. Furthermore, antioxidant activity is sometimes reported in IC₅₀ values, rather than as trolox equivalent, which prevents a viable comparison of data sources and, in the context of the present study, this is considered highly unadvisable.

Future prospects

Despite the significant advantages of the assays considered in the present study, these still comprise destructive methods, which require specific chemicals, solvent labware, consumables and specialized equipment, as well as trained personnel. The future of truly cost‐effective, rapid and simple food chemical analysis lies in non‐destructive techniques [e.g. near Infra Red spectroscopy (NIRS), Fourier transform infrared spectroscopy (FTIR) and Raman spectroscopy]. At the same time, these non‐destructive methods still require specialized equipment and constant calibration with wet chemical methods to ensure robustness and reliability. Additionally, data analysis is complex, requiring expertise in statistical modelling. Nevertheless, recent advances in machine learning and artificial intelligence hold great promise for both analytical approaches, highlighting the need for further studies in these areas, as well as emphasizing the importance and novelty of these techniques, which are expected to become standard methods, facilitating analysis and enhancing our understanding of food.

Although much data are available, and constantly added, regarding the effects of agricultural management practices and genetics on plant parameters, nutritional aspects are rarely regarded in this framework. Hence, research in this area is limited, although the nutriphytochemical composition has been reported to be majorly affected by cultivation conditions (e.g. growth region, temperature and management, such as irrigation and fertilization or postharvest storage). The present study provides tools that enable the study of nutritional effects more routinely, incorporating it as an integral part of agricultural research, such as for breeding programs, or comparing management treatments, also emphasizing the importance of standardized protocols. Importantly, when an accurate quantification is required, more advanced methods are recommended, such as gas and liquid chromatography, preferably coupled to a mass spectrometer, with the use of commercial standards.

In conclusion, the simple spectrophotometric assays compiled here are proposed as an integral part of agricultural research, aiming to assess nutritional value through high‐throughput screening. The methods suggested here have been optimized for efficiency, simplicity and scalability, facilitating the steps and streamlining the stages for a rapid performance. Future works should include more spectrophotometric assays to measure the total contents of additional groups of compounds, not included in the present study (e.g. alkaloids, which are challenging due to their large structural variability, quinones, phytoestrogens, chlorophyll and tannins). In addition, given the advances in NIRS and Raman technologies, we consider that efforts need to be made to improve these techniques, as well as the accompanying data processing and modeling, using machine leaning in designing new strategies to support high‐throughput chemical analysis. Moreover, simple accessible bioassays, including anti‐inflammatory, anti‐metabolic syndrome or anti‐cancer assays, may be valuable to support the analyses presented here.

Short videos of some assays filmed in our laboratory are available under the @TietelLab channel on YouTube.

Supporting information

Figure S1. Workflow for various antioxidant capacity analysis.

Figure S2. Analyses that can be performed using the same 2n HCl extract.

DATA AVAILABILITY STATEMENT

The data that support the findings of this study are available from the corresponding author upon reasonable request.