A Method for the Rapid Measurement of Alkylresorcinols in Flour, Bread and Related Products Based on 1H qNMR

The main objectives of the current work were to investigate differences among flours from traditionally preserved Greek varieties of cereals, and especially those of wheat, and in parallel, to correlate those potential differences with the presence of bioactive natural ingredients. In this context, we developed a new, fast, and simple method for the measurement of total 5-alkylresorcinols in cereals and related foods by qNMR. Several types of flour (white or whole-grain) coming from wheat, i.e., Triticum dicoccum, T. monococcum, T. aestivum, T. durum and T. turgidum, corn, barley, rye and oat from a certified producer in Greece were used either as raw materials or for the production of bread, pasta and flakes. A small portion of the flour or the corresponding product was extracted with DMSO-d6. The liquid part was directly analyzed by NMR (400 MHz). The simplicity of the NMR spectrum of the total extract and the lack of overlapping peaks permitted the development of a high throughput quantitative method for the measurement of total bioactive alkylresorcinols in less than 15 min. Grains, whole grain flours and breads from old varieties of T. dicoccum and T.monococcum showed high contents of alkylresorcinols (455–1148 mg/Kg), while the same compounds were completely absent from white flour and the corresponding bread. The term high-phenolic flour is proposed to distinguish among flour types.


Introduction
Wheat (Triticum spp.), olive trees and grapevines are considered to be the three basic plants of the Mediterranean diet; in particular, wheat is the most important plant in the history of human nutrition in Europe and the Middle East. Although famous bioactive small molecules like resveratrol or oleocanthal have been discovered in grape and olive oil, wheat products are not similarly known, from the consumer's perspective, as sources of health-protective ingredients. However, alkylresorcinols (ARS), an important class of phytochemicals found in wheat, have attracted the interest of scientists in the last thirty years [1][2][3].
Alkylresorcinols, 1,3-dihydroxy-5-n-alkylbenzenes, are phenolic lipids found in some plant families, as well as in some algae, sponges, fungi and bacteria [2]. The main nutritional sources of alkylresorcinols are whole grain, bran of wheat and rye, with contents that may reach 1-4 g/kg of dry material [4][5][6][7][8][9][10]. ARS are adequately absorbed [11] and have exhibited interesting bioactivities in vitro, including direct antioxidative properties, lipoxygenase inhibition, inhibition of copper-induced LDL oxidation, DNA-strand scission, colon and prostate cancer cell growth inhibition and inhibition of lipase activity in the adipose tissue cells [3,12,13]. Despite being present in high amounts in specific cereals like wheat and rye, and despite their reported bioactivities and potential role in health protection, their concentration is not widely used for the determination of grain quality or the quality of the final food products. For example, the term high-phenolic olive oil has been used as a marketing description in recent years based on the EU health claim [14] and the big variability in the phenolic content of olive oil; however, nothing similar has happened for cereals. One possible reason for this is the limited data from human nutritional intervention studies, and another is the rather complex analytical methodologies required for the selective measurement of alkylresorcinols in cereal products [4][5][6][7][8][9][10][15][16][17][18][19][20].
To overcome the challenges associated with analytical methodologies, and based on previous successful applications of high-throughput analyses of bioactive small molecules in several foods (olive oil [21], beer [22], wine [23]) or botanic extracts [24][25][26], our target is to develop a fast and simple method for the measurement of total 5-alkylresorcinols in cereals and related foods (bread, pasta, wafers etc.) by qNMR. The developed method requires a total time of less than 15 min per sample, without the use of standards, and makes it possible, for the first time, to study several Greek traditional (autochthonous) wheat species and varieties and compare them with several commercial cereals, as well as other common beans and seeds. As previously reported, huge differences were observed among whole-grain and refined flours and respective final food products; interestingly, the grains of some local wheat varieties showed some of the highest reported concentrations.

General
All solvents were of analytical grade (Merck). Syringaldehyde (98% purity, Sigma-Aldrich, Steinem, Germany) was used as internal standard (IS). IS solution was prepared in DMSO-d 6 at a concentration of 0.5 mg/mL and kept at 4 • C. The IS solution was left to reach room temperature prior to use. The quantitative determination of alkylresorcinols was performed using NMR spectroscopy with a Bruker Avance DRX 400 MHz. The 1 H-NMR spectra were processed using either the MNova (Mestrelab Research) or the TOPSPIN software (Bruker, Billerica, MA, USA).

Plant Material and Processed Products
Several types of flour (white or whole-grain) originating from different wheat species, i.e., Triticum dicoccum, T. monococcum, T. aestivum, T. durum and T. turgidum, and other cereals, i.e., Zea mays (corn), Hordeum vulgare (barley), Secale cereale (rye), Avena sativa (oat) as well as the grains of the above plants and processed products (bread, flakes, wafers, pasta) and seeds of Lens culinaris, Lathyrus clymenum, Phaseolus vulgaris, Cicer arietinum, Linum usitatissimum and Fagopyrum esculentum were obtained from a certified producer in Greece (Antonopoulos farm, Dilofo). All plants were grown in the same location and the same year to reduce the impact of pedoclimatic factors.

Extraction and Isolation
T. dicoccum whole-grain flour (100 g) was extracted in an ultrasonic bath for 20 min with CH 2 Cl 2 (300 mL). The solvent was evaporated and the residue was submitted to column chromatography using silica gel with cyclohexane and mixtures of cyclohexane with ethyl acetate (95:5, 90:10 and 85:15). Fractions obtained with cyclohexane-ethyl acetate 85:15 were pooled and submitted to preparative TLC. The alkylresorcinol spot was identified by the intense red color appearing after spraying with vanillin in methanol/sulfuric acid and heating. The spot with Rf = 0.32 (cyclohexane-ethyl acetate-acetic acid 80/20/1) was extracted with dichloromethane and analyzed by GCMS and 1 H-NMR in CDCl 3 , DMSO, pyridine and acetone.

GC-MS
GCMS was performed using a Hewlett Packard 6890-5973 GC-MS system operating in EI mode (equipped with a HP 5MS 30 m × 0.25 mm, 0.25 µm film thickness capillary column). Helium (1 mL/min) was used as carrier gas. The initial temperature of the column was 60 • C; it was then heated to 280 • C at a rate of 3 • C/min.

High-Throughput Extraction and NMR Analysis
A portion of the flour or the comminuted corresponding product (330 ± 0.1 mg) was extracted in an ultrasonic bath for 5 min with deuterated DMSO (1 mL) containing syringaldehyde (0.5 mg/mL) as an internal standard. The supernatant liquid part was obtained by centrifugation (5 min at 3000 rpm) and was directly transferred to a 5 mm NMR tube. Each sample was analyzed in triplicate using a standard 90 degree excitation pulse, with a pulse width of 10 µsec and a prescan delay of 6.5 µsec. All measurements were performed at 298 K. Typically, 16 scans were collected into 32 K data points over a spectral width of 0-13 ppm (5263.18 Hz) with a relaxation delay of 10 s, an acquisition time of 3.11 sec and a FID resolution of 0.32 Hz. The appropriate relaxation delay was determined by gradual increases (1,2,5,8,10,15,20 s until the ratio between the integration of the peak of internal standard and the peak of the target compounds remained unchanged. The matching, tuning, shimming, receiver gain adjustment as well as phasing and baseline correction were always first performed automatically and then manually to achieve the best result. Prior to Fourier transformation (FT), an exponential weighting factor corresponding to a line broadening of 0.3 Hz was applied. For the peaks of interest, accurate integration was performed manually. The concentration of ARS was measured by comparing the area of the selected signal at 6.00 ppm with that of the internal standard (IS) at 9.79 ppm, which was set as 1. The calculation of the concentration of total ARS in mmol/mg of dry material was performed using the following formula: C = [(I ARS /3) × mmol I.S. ]/M sample , where M sample = 330 mg and mmol I.S. = 0.5 mg/MW syringaldehyde = 0.00270. To express the concentration in mg ARS/mg of dry material, we used the average molecular weight between C19 (m/z 376) and C21 (m/z 404) alkylresorcinols.

Precision-Recovery-LOD-LOQ
The recovery was calculated by the comparison of successive extractions of samples for the studied ARS. The samples contained different levels of extractives in order to obtain unbiased results. The recovery achieved after one extraction was >95%. The intraday and interday precision was determined by analyzing three replicates of three random samples from the same day and from three different days, and the %RSD was found to be <10%. LOQ = 15 µg or 45 mg/Kg (S/N > 10) and LOD = 10 µg or 30 mg/Kg (S/N > 3).

Comparison among Flour Extracts
The obtained extracts of flours using DMSO-d 6 were directly recorded by 1 H-NMR without any other treatment. The spectra were compared and a singlet was observed at 6.0 ppm only in wheat and rye samples (Figures 1 and A1, Figure A2, Figure A3). Interestingly, it was not observed in refined (white) flours.
The same flour samples when extracted in other deuterated solvents showed more complicated spectra which precluded accurate quantitation ( Figures A4 and A5).

Structure Elucidation Results
Targeted fractionation and isolation of the molecules related with the peak at 6.0 ppm showed that it corresponded to 5-alkylresorcinols. The NMR spectrum in CDCl 3 ( Figure 2) was similar with that described in literature [12]. Interestingly, the spectrum of the same compound in DMSO ( Figure 2) showed only a single peak in the aromatic region, consistent with that observed in the flour extract. The same flour samples when extracted in other deuterated solvents showed more complicated spectra which precluded accurate quantitation ( Figures A4 and A5).

Structure Elucidation Results
Targeted fractionation and isolation of the molecules related with the peak at 6.0 ppm showed that it corresponded to 5-alkylresorcinols. The NMR spectrum in CDCl3 ( Figure 2) was similar with that described in literature [12]. Interestingly, the spectrum of the same compound in DMSO ( Figure  2) showed only a single peak in the aromatic region, consistent with that observed in the flour extract. (a)

Structure Elucidation Results
Targeted fractionation and isolation of the molecules related with the peak at 6.0 ppm showed that it corresponded to 5-alkylresorcinols. The NMR spectrum in CDCl3 ( Figure 2) was similar with that described in literature [12]. Interestingly, the spectrum of the same compound in DMSO ( Figure  2) showed only a single peak in the aromatic region, consistent with that observed in the flour extract.
(a) A GCMS analysis of the isolated compound showed that it was a mixture of 5-n-alkylresorcinols with a side chain ranging from C15 to C25, but predominantly from C19 and C21 ( Figure 3).  A GCMS analysis of the isolated compound showed that it was a mixture of 5-n-alkylresorcinols with a side chain ranging from C15 to C25, but predominantly from C19 and C21 ( Figure 3). A GCMS analysis of the isolated compound showed that it was a mixture of 5-n-alkylresorcinols with a side chain ranging from C15 to C25, but predominantly from C19 and C21 ( Figure 3).

Quantitation Results
Grains, whole grain flours and breads, especially from old varieties of Triticum monococcum and T. dicoccum, showed high alkylresorcinols (ARS) contents (>450 mg/Kg), while the compounds were completely absent from white flour and all the commonly consumed corresponding types of bread.
We also found a high total alkylresorcinol content only in samples of Secale cereale flour, while no 5-alkylresorcinols could be detected in Hordeum vulgare (barley) flour, Avena sativa (oat) whole grain flour and flakes, Zea mays whole grain flour, nor in the seeds of Lens culinaris, Lathyrus clymenum, Phaseolus vulgaris, Cicer arietinum, Linum usitatissimum, Fagopyrum esculentum and Zea mays. The quantitation results are presented in Table 1.

Discussion
The main objectives of the current work were to investigate the possible differences among different types of flours from traditionally preserved Greek indigenous varieties of cereals, and especially those of wheat, and in parallel, to correlate those differences with the presence of bioactive natural ingredients. Based on the previous successful use of 1 H-NMR for the qualitative and quantitative characterization of bioactive ingredients in olive oil, wine and beer [21][22][23], we first studied the total profile of selected flour extracts using microextraction with deuterated solvents and directly recorded the NMR spectra.
After several trials of extraction and simultaneous spectra recording with CDCl 3 , CD 3 OD, Pyridine-d 5 , DMSO-d 6 and Acetone-d 6 , we found that DMSO was the most efficient solvent, presenting in parallel the clearest spectra concerning the aromatic area which showed the most significant differences. A comparison among the different spectra (Figure 1) of the flours extracted with DMSO-d 6 revealed the presence of a major peak in the aromatic area (6.0 ppm) only in rye and the wheat species. Targeted isolation of the molecules corresponding to this peak using liquid chromatography and subsequent characterization by NMR and GCMS led us to attribute it to a series of 5-alkylresorcinols. Interestingly, when the NMR spectrum of the isolated compound was recorded in CDCl 3 , Acetone-d 6 or Pyridine-d 5 , it presented a doublet and triplet, as described in the literature for 5-alkylresorcinols [12], corresponding to the three aromatic protons ( Figure 2A); however, when it was recorded in DMSO, the spectrum was simplified, presenting only one singlet in the aromatic region (H-2,4,6) ( Figure 2B). In contrast to our results, the three protons H-2,4,6 of 5-alkylresorcinols in CDCl 3 [27] were reported as two singlets, and in DMSO [28] as a doublet and a triplet. An analysis of the isolated compound by GCMS revealed that it was a mixture of several alkylresorcinols with side chains ranging from C15 to C25, but predominantly from C19 and C21 (Figure 3).
The simplicity of the NMR spectrum of the total extract in DMSO and the lack of overlapping peaks from other compounds allowed us to develop a high throughput quantitative analytical method for the measurement of total alkylresorcinols. The method was based on the integration of the peak at 6.0 ppm (I ARS ) and comparisons with the integration of the internal standard at 9.79 ppm, which was set as 1. The calculation of the concentration of total ARS in mmol/mg of dry material was performed using the formula presented in Materials and Methods, which was based on the molar ratio between the target analyte and the internal standard, following the general guidelines presented by Pauli et al. [29] and Bharti and Roy [30]. The integration value of the singlet at 6.0 ppm was divided by 3, since it corresponded to three aromatic protons.
It should be noted that one extraction step in an ultrasonic bath for 5 min is sufficient for the quantitative recovery (>95%) of ARS. Longer extraction times or more than one cycle of extraction led to a negligible increase of the recovery, i.e., less than 5% ( Figure A6).
Advantages of the qNMR method in comparison to chromatographic methods Alkylresorcinols (ARS) have been previously investigated and quantified in several wheat products, mainly using time-consuming chromatographic methods and long extraction or pretreatment procedures [4][5][6][7][8][9][10][11][12][13][15][16][17][18][19][20]. Depending on the analyzed material (flour, grain, bread or pasta) and the analytical method (liquid or gas chromatography or colorimetry), several methods of extraction have been used, most commonly with ethyl acetate, acetone, hot propanol or propanol/water, with extraction times ranging from 3 to 48 h, followed by several filtration, evaporation and dilution or derivatization steps. In all cases, external standards of one or more ARS are necessary. Existing chromatographic methods (GCMS, HPLC-UV/DAD or LCMS) give very good results but the time of analysis ranges from 15 min to 90 min. Colorimetric methods require times ranging from 15 min to 4 h.
Our new method is the first to make possible the selective and high throughput quantitative determination of ARS in grains, flours and foods in less than 15 min per sample, including the extraction and analysis time. In addition, the most important advantages are: (i) no need for standards for target compounds; (ii) no need for derivatization; (iii) short time of extraction in ultrasonic bath (5 min); (iv) very simple procedure for the separation of the solution for analysis by centrifugation; and (v) adequate limit of detection and precision. The only limitation is that the current method cannot determine the length of the side chain, which, in some cases, is useful to differentiate rye from wheat [4,5].
Comparison among the studied cereals and other common seeds and beans As previously reported, ARS were found only in wheat and rye, and not in other cereals. Indeed, analyses of the grains of the studied cereals, as well as several of other commonly used beans and seeds, confirmed that the main nutritional source of ARS is wheat and rye. Wheat should be considered as a primary nutritional source, especially for Mediterranean populations, since rye is consumed to a lesser extent. Until the mid-20th century, the main flour used for bread preparation was whole-grain wheat flour, usually coming from local varieties, and consequently, the traditional Mediterranean diet should be considered as a diet that is very rich in ARS. In modern days, the massive use of white (refined) flour from T. aestivum has significantly reduced the nutritional intake of ARS in Mediterranean populations.
Comparison among wheat species and previously reported ARS content in wheat grains and flour In our study, the local, traditional variety of T. monococcum (known as "Kaploutzas") showed the highest content of ARS in whole grain flour, i.e., even higher than the corresponding rye flour. It showed a small difference in comparison with T. dicoccum and a big difference with T. durum, T. aestivum and T. turgidum. Interestingly, the analysis of the grains showed that T. dicoccum presented the highest content of ARS (>1000 mg/Kg), close with that of T. monococcum, but again, very different from those of T. aestivum and T. durum. The results obtained herein are in agreement with previous analyses [4,6,15] concerning whole grain and refined white flours. Previous studies have also shown that the grains of T. monococcum and T. dicoccum are richer than T. aestivum and T. durum, but herein, the concentrations in local varieties of T. monococcum (925 mg/Kg) and T. dicoccum (1148 mg/Kg) were found to be much higher than what was previously reported from Germany and Hungary (391-819 mg/Kg) [4,8,17] and also from all other wheat grains reported until now, with the exception of T. timophevi ( Figure A7). A comparative list of all previous studies measuring the levels of ARS in wheat grains and products is presented in Table A1.
ARS content of bread, pasta, wafers The results obtained for T. dicoccum grains and the fact that this species is becoming popular again for baking prompted us to study the fate of ARS during the production of bread, flakes, wafers and pasta. Interestingly, all products contained significant amounts of ARS (Table 1, Figure A8). The flakes which has undergone the least processing, as expected, showed the highest content. Bread retained the majority of ARS during baking, in contrast to previous studies showing that ARS disappeared during baking. In fact, DMSO extraction and the applied analysis methods probably aided in overcoming the problems of previous methods, where interactions with starch led to poor results.

Conclusions
Alkylresorcinols are a class of major bioactive ingredients present in wheat flour and wheat-based products, and found in high amounts in old wheat varieties while being almost absent from the common white flours that are used for the mass production of bread nowadays. Their concentration can easily be measured by qNMR, permitting high-throughput analyses of flour and related products.

Acknowledgments:
We would like to thank Antonopoulos Farm, Dilofo, Greece for offering the raw plant material and the processed products and Dan Flynn for its assistance in preparation of experimental breads.

Conflicts of Interest:
The authors declare no conflict of interest. Table A1 contains a comparative list of all previous studies measuring the levels of ARS in wheat grains and products. NMR spectra of analyzed seeds and beans are provided as Figures A1-A3. Examples of NMR spectra of extracts recorded in solvents other than DMSO are provided as Figures A4 and A5. NMR spectra showing the recovery of each extraction step are provided as Figure A6. NMR spectra of wheat grains and products are provided as Figures A7 and A8.       Please revise them.

Commented [m2R1]:
has arbitrary units that Please revise them.