Materials and methods

Plant materials. Leaves of O. capitatus and O. echinops species were recollected on January 25 and 26, 2021 in the Clavijero Botanic Garden of the Instituto de Ecología, A. C. in Xalapa municipality, Veracruz; at Latitude 19° 40′ 39″ N; Longitude 97° 01′ 07″ W, elevation 1,400 m asl. Leaves of O. flaccidus were recollected in the locality of Atapalchico, Tlacolulan municipality, Veracruz, 3 km from Tlacolulan town; Latitude 19° 40′ 39″ N; Longitude 97° 01′ 07″ W, elevation 1,813 m asl.

Voucher samples: The botanical materials for O. capitatus (I. Acosta, 4044), O. echinops (I. Acosta, 4045 & 4046), and O. flaccidus (I. Acosta, 4047) were deposited at the Herbarium XAL of the Instituto de Ecología, A. C. (Xalapa, Veracruz, Mexico). Sampled leaves came from four individuals of each species in two localities.

Extraction (essential oils preparation). All the chemicals used in this procedure were purchased directly from Sigma-Aldrich (St. Louis, MO, USA) and were used without additional purification. The essential oil from the freshly aerial parts of each species was obtained by a hydrodistillation process. Fresh plant material was ground briefly in a food processor (Nutribullet^®). Afterwards, 250 g were transferred to a 1 L flat bottom flask equipped with a magnetic stirrer and distilled water (600 mL) was added. The material was hydrodistilled for three hours once boiling started. The distilled liquors were transferred to a separatory funnel, and the aqueous phase was extracted with dichloromethane twice (100 mL × 2). The organic layers were collected, combined, and filtered through anhydrous Na₂SO₄. The solvent was evaporated in a rotatory evaporator (R-II, BÜCHI, Flawil, Switzerland) at 600 mbar, 25-30 °C. The oily residue was weighted, and yield was obtained as follows (^{Soto-Armenta et al. 2017}):

where:

The essential oils were stored at -20 °C until further analysis; four replicates were prepared for each species.

Analysis by Gas Chromatography Coupled to Mass Spectrometry (GC-MS). One microliter of each essential oil was injected into the GC port, and the chromatographic separation and analysis were carried out in a gas chromatograph coupled to a mass analyzer (Shimadzu, Single Quadrupole QP2010 Ultra) as previously described by ^{Lascurain-Rangel et al. (2018)}. Briefly, helium gas was used as carrier gas (1.2 mL/min, constant flow), and a ZB-5MSi column (30 length × 0.25 mm inner diameter × 0.25 µm film thickness; Zebron, Phenomenex Inc.) was used as a stationary phase. A split injector at a 16.7 rate and temperature of 250 °C was used to introduce the sample to the GC column. GC oven temperature for compound separation was adjusted at an initial temperature of 50 °C was held for 4 minutes, then it increased at a rate of 15 °C/min up to 250 °C, which was held for 5 minutes. The MS was operated in electron impact (EI, 70 eV) mode with a source temperature of 230 °C, interface temperature of 250 °C, and a continuous scan from 30 m/z to 500 m/z. The mass spectrum data of volatile compounds present in the Oreopanax oils were compared with those in the NIST/EPA/NIH Mass Spectral Library, NIST 11, Software version 2.0 (National Institute of Standards and Technology, www.nist.gov), using a range of 84-100 % similarity values, with the Lab solutions GCMS solutions 2.72 software (Shimadzu, Japan) and by co-elution with authentic standards under the same analytical conditions above described. The analysis of essential oils was carried out in quadruplicate, and all standards used for comparison were purchased in Sigma-Aldrich (St. Louis, MO, USA) at a GC grade purity (> 95 %). With the spectrometric dataset of the three species (m/z_Rt pair values) and NIST identification, a heatmap with hierarchical ordering was constructed using the Metaboanalyst bioinformatics platform (https://www.metaboanalyst.ca/MetaboAnalyst/home.xhtml) for comparative purposes. Raw spectrometric data were Log (10) transformed, autoscaled, and normalized by quantiles.

Leaf anatomy study. Sampling.- To study the leaf anatomy of the Oreopanax leaves, sections of approximately 0.5 × 0.5 cm were taken from each leaf, from the center to the left margin, and at the apex. Those sections were fixed in a mixture of formaldehyde, acetic acid, and 70 % alcohol (5:5:90 per volume) for several days. Then, samples were washed in water several times until the smell of formaldehyde became imperceptible.

Dehydration.- Samples were dehydrated in a Tertiary Butyl Alcohol (TBA) series (^{Ruzin 1999}) till reaching pure TBA.

Paraffin embedding.- Samples were transferred to glass vials with pure TBA, and a few paraffin shavings were added. The vials were kept at room temperature. When the TBA dissolved the paraffin shaves, more shaves were added. This procedure was repeated three more times, and then the vials were put in an oven at 60 ^oC for 24 hours. The mixture of paraffin-TBA was discarded, and fresh, melted paraffin was added to each vial. After two changes in pure paraffin, molds were cast with two or three sections in each mold, following the procedure of ^{Ruzin (1999)}. After trimming the paraffin molds and mounting them on a wooden support, sections 18-20 µm thick were obtained with a rotary microtome Leica RM 2125RTS using a disposable blade.

Staining.- The paraffin with the tissue sections was removed with 100 % xylol, then, tissues were rehydrated by passing them through a series of decreasing ethanol concentrations down to pure water. Sections were stained with 0.05 % methyl blue in water for 2-3 hours. Afterward, sections were washed in three water changes and dehydrated in increasing series of ethanol, up to 100 %. Then, sections were transferred to a mixture of equal parts of ethanol and methyl salicylate (as a clearing agent). After two changes in pure methyl salicylate, sections were mounted with synthetic resin dissolved in xylene and covered with a glass coverslip of 2.5 × 4 cm. Images of the most remarkable features were taken with a microscope Nikon Eclipse E600, with bright field or ultraviolet light.

Results

Chemical characterization. Given that consumers value traditional xoco tamales because of the characteristic flavor provided by the leaves of Oreopanax, the study of the chemical composition of the leaves included the analysis of the essential oils to identify distinctive metabolites associated with each species contributing to the differentiation of this food complex. The extraction by hydrodistillation allowed us to obtain the essential oil from each species of Oreopanax as colorless oils with a spiced but pleasant odor. In Table 2, it can be observed that the three species analyzed presented a similar yield (%) of essential oil. So, there were no differences in terms of essential oil abundance that could be associated with a given species.

Later, the same essential oils from Oreopanax spp. were analyzed by GC-MS combined with co-elution with a set of authentic standards to increase the accuracy of some identifications. From these analyses, a total of 44 volatile compounds were identified in the leaves of Oreopanax spp. (Table 3), based on their spectrometric fingerprints compared with the reference compounds or with those reported in the NIST database. As expected for those plants used as spices in traditional food, chemical composition was complex in the three studied species and among the compounds found in their leaves as it can be observed in the corresponding chromatograms (Figure 3). We are reporting for the first time the presence of several aliphatic and aromatic alcohols and terpenoids (mono, sesqui, and diterpenes) as some of the major volatiles in xoco leaves.

Figure 3 Comparative analysis of volatile profiles obtained by GC-MS with the essential oils of xocos leaves. A) Oreopanax equinops, B) Oreopanax capitatus, C) Oreopanax flaccidus. 

To distinguish among the species of the xoco complex, the essential oils obtained by GC-MS were ordered hierarchically in a heatmap (Figure 4). The heatmap revealed two well-defined clusters among species, O. echinops being the most distinct species when compared to O. capitatus and O. flaccidus. O. echinops contained the most significant accumulation (major abundance) of a total of 10 volatiles that were absent in the other species and includes β-ionone, δ-cadinene, (-)-kaurene, α-copaene, caryophyllene, among others. These metabolites were classified as sesquiterpenoids and could be considered as distinctive chemical markers in this species. Interestingly, O. capitatus and O. flaccidus were more similar and share approximately 17 volatiles that were absent in O. echinops, such as acetaldehyde, phenol, 1-heptadecene, nerol, α-terpineol, geraniol, linalool, being some of these metabolites from a monoterpenoid origin. However, although O. capitatus and O. flaccidus were more like each other, they both had also distinctive compounds. For example, the essential oil of O. capitatus contained five compounds [(-)-spathulenol, cis-1-chloro-9-octdecene, trans-neorolidol, α-cadinol and (Z)-4-Decen-1-ol] that does not contained the essential oil of O. flaccidus, and this comprised seven compounds (1-heptacosanol, bornyl alcohol, dihydrodehydro-β-ionone, p-cymen-8-ol,1-nonadecene, n-hexadecanoic acid, and 1,3-benzenedicarboxylic acid) that were not present in the essential oils of the O. capitatus and O. echinops (Figure 4).

Figure 4 Heatmap with hierarchical clustering of Oreopanax spp. The plot was built using the spectrometric dataset obtained by GC-MS. In X-axis the replicates of each essential oil per species are indicated, and Y-axis contains the names of the volatile compounds identified by co-analysis with standards or through NIST database assignments. The three species are colored-coded by the three classes labeled at the top of the heatmap. 

Leaf anatomy. Leaves of Oreopanax spp. shared some anatomical features (Table 4): They were hairy (except O. capitatus, which was glabrous) in the abaxial surfaces, with long, three-to-four-branched trichomes (Figures 5A, B). The most remarkable common feature was the presence of resin canals along the veins and in the cortical parenchyma of the main veins (Figures 6A-F). Those canals were lined with six to eight epithelial cells (Figure 6D) in one or two layers and had diameters from 15.5-23 µm in the midvein and 15-25 µm in minor veins. In the lamina, resin canals were formed in connection with the minor veins, on the top and at the bottom of each vein, between the palisade parenchyma (top) or between the spongy parenchyma (bottom) (Figure 6E). In the midvein, the resin canals were formed in the cortex (Figures 6B, D) or between the vascular bundles (Figure 6C). Lamina thickness varied from 430 to 580 µm. O. capitatus was the thickest with 580 µm, O. flaccidus had 520 µm, while the thinner was O. echinops, with only 430 µm of thickness. The thickest parts of each leaf were the midveins of the three species.

Figure 5 Leaf surfaces of Oreopanax leaves. A) O. echinops, adaxial surface. B) O. flaccidus, adaxial surface. C) O. capitatus, abaxial surface. A) and B) show articulated trichomes (arrows), while C) is glabrous. Scale bars: 10 mm. 

Figure 6 A) O. echinops midvein, cross-section. Resin ducts look like dark areas (red arrows) under UV light. Lignin fluorescence is orange/red in the xylem secondary walls; the fibers forming the sheath surrounding the vascular tissue fluoresce in green. B) O. flaccidus cross-section through midvein. Bright field. Most of the canals form in the cortical tissue, both in the abaxial (low arrow) and adaxial (upper arrow) sides (arrows). Smaller canals can be observed in the center of the midvein (yellow arrow). C) O. capitatus midvein in cross-section. Resin ducts are associated with the phloem rays (arrows). D) O. echinops. Detail of two large resin canals formed by six to eight epithelial cells (arrows). E) O. flaccidus. Lamina cross-section through a minor vein. An upper canal can be seen (arrow) in close contact with the palisade parenchyma. F) O. flaccidus. Cross-section of the lamina comprising a portion of the midvein, showing several resin canals (arrow). Scale bars: A) = 300 μm; B) = 500 μm; C) = 200 μm; D) 25 μm; E) = 50 μm; F) = 260 μm. 

Discussion

Our chemical analysis and anatomical studies of the Oreopanax of the xoco complex indicate that these three species share volatile compounds. However, they also have important differences in their volatile profiles that could be related to their flavor and distinguishable by local consumers of xoco tamales. The resin canals of O. capitatus and O. flaccidus are formed by one to two layers of epithelial cells. These three species have in common the formation of resin canals along the minor leaf veins, or in the cortex of midveins.

The extraction by hydrodistillation indicates no differences in essential oil content that could be associated with a given species. The essential oils from the leaves of Oreopanax spp. were analyzed by GC-MS, indicating a total of 44 volatile compounds among these species, highlighting the presence of several aliphatic and aromatic alcohols, sesqui and monoterpenoids as some of the most significant volatiles in xoco leaves.

It is well known that essential oils from medicinal and food plants are related to a wide spectrum of biological activities, and industrial and technological applications, this is mainly because of their volatile compounds’ composition. Table 5 summarizes the bioactivities reported for the major distinctive volatiles found in each Orepanax species. This information provides added value to the use of xoco leaves to traditional cuisine since these oils contain compounds that are beneficial for the consumer's health.

The odor profiles of essential oils obtained by GC-MS revealed two well-defined clusters among species (Figure 3, O. echinops being the most distinct species when compared to O. capitatus and O. flaccidus. Oreopanax echinops contains the greatest accumulation of 10 volatiles that are absent in the other species and includes β-ionone, δ-cadinene, (-)-kaurene, α-copaene, and caryophyllene, among others. These compounds are classified as sesquiterpenoids and could be considered distinctive chemical markers in this species. Sesquiterpenoids are biosynthesized by the mevalonic acid pathway that occurs mainly at the cytosol in plant cells (^{Dewick 2009}). Interestingly, O. capitatus and O. flaccidus are more similar and share approximately 17 volatiles that are absent in O. echinops, such as acetaldehyde, phenol, 1-heptadecene, nerol, α-terpineol, geraniol, linalool, among others. These volatiles are monoterpenoid-type compounds, whose biosynthetic origin is throughout the 1-deoxy- D-xylulose 5-phosphate (DXP) or non-mevalonate pathway that takes place at the chloroplast level in plants (^{Dewick 2009}), suggesting a noticeable difference in the active biosynthetic and enzymatic machinery among the studied xoco plants. However, although O. capitatus and O. flaccidus are more similar to each other, they both have also distinctive compounds. Perhaps, conducting deeper molecular studies on these xoco species such as transcriptomics targeting the expression levels of gene clusters for key enzymes in the mevalonic acid biosynthesis such as 3-hydroxy-3-methylglutaryl-CoA synthase and 3-hydroxy-3-methylglutaryl-CoA reductase in O. echinops or DXP synthase for O. capitatus and O. flacciuds will allow corroborating these metabolic differences. Untargeted metabolomics using total crude extracts from leaves instead of essential oils could be also useful in distinguishing these species. Nonetheless, to the best of our knowledge, there are no reports of the traditional use of xoco leaves in the form of powder or ground material as occur for other spices.

Leaves of Oreopanax spp. share some anatomical features; for instance, they are hairy (except O. capitatus, which is glabrous) on both surfaces, with long, three-to-four-branched trichomes. The most remarkable common feature is the presence of resin canals along the veins and in the cortical parenchyma of the main veins. In the lamina, resin canals are formed in connection with the minor veins, on the top and bottom of each vein, between the palisade parenchyma or the spongy parenchyma. In the midvein, the resin canals are formed in the cortex or between the vascular bundles. Lamina thickness varies: O. capitatus is the thickest, with 580 µm, O. flaccidus 520 µm, and O. echinops 430 µm.

Mexican xoco tamales food complex (Oreopanax spp.) has O. capitatus as the signature species (^{Linares & Bye 1987}) that characterizes this ethnobotanical complex because it is the most common one in regional markets as well as the preferred species. The taxa O. echinops and O. capitatus are used and traded in the same distribution range, at least in central Veracruz, except for O. flaccidus, which has not been observed for local sale (^{Lascurain-Rangel et al. 2017}). So far, it is unknown if there are substitutes for these species by other local ones that could be included in this complex. Studies will be necessary to identify the uses of Oreopanax leaves in other regions of the country.