Stable Isotope Ratios of Herbs and Spices Commonly Used as Herbal Infusions in the Italian Market.

Stable isotope ratio analysis has been widely used for traceability and authenticity purposes in relation to various food commodities, but only in a limited number of herb and spice species. This study explored the stable isotope ratios of carbon, nitrogen, sulfur, oxygen, and hydrogen (δ13C, δ15N, δ34S, δ18O, and δ2H) of 119 herbs and spices belonging to 116 plant species and 57 plant families collected from the Italian market for the first time. The characteristic value ranges of δ13C, δ15N, δ34S, δ18O, and δ2H of the herbs and spices went from -31.0 to -11.6, -4.7 to 12.0, -5.0 to 22.0, 14.7 to 46.0, and -158 to -12‰, respectively. The isotopic profiles within and between common botanical families and their similarity/dissimilarity between herbs and spices belonging to the common botanical families are also discussed here. The results of this exploratory work highlight the possibility of characterizing herbs and spices and suggest widening the scope of the survey through more extensive sampling and focusing on specific plant species.

Introduction

Herbs and spices have long been used as food supplements to impart aroma, flavor, and color. They are rich sources of natural antioxidants and beneficial for human health.[1,2] Many herbs and spices are also consumed as infusions, synonymously called herbal infusion/tea, or tisane.[3] They are preferable to traditional tea and are coffee substitutes because they have comparably less caffeine and theine and provide a therapeutic or recreational value. Furthermore, previous studies on the clinical efficacy and safety of herbal teas have shown that the intake of herbal tea (lavender, chamomile, fenugreek, stinging nettle, spearmint, hibiscus, etc) has positive effects on female health, diabetes, heart disease, and weight loss.[1] For these reasons, herbal infusions are becoming more and more popular beverages.[3]

The use of herbs and spices is expanding all over the world. Though the largest share of production and consumption of herbs and spices is occupied by the Asia-Pacific regions, the market in North America and Europe has also significantly increased in the last few decades.[4,5] For instance, Italy has tripled the production and export of herbs and spices and doubled their import since the beginning of the millennium and is currently one of the top five producing countries in Europe and the world’s first user of herbs for the cosmetic sector. It is estimated that the Italian national market of herbs and spices is worth about $750 million in a global market that exceeds $6 billion.[6]

Generally, herbs and spices are high price commodities and are often sold as dried plant parts or in powder form. They have a long and complex supply chain.[7] For this reason, they are often a target of adulteration, mislabeling, or substitution with cheaper products. For instance, a survey of 78 retail and online oregano samples in the United Kingdom resulted in 24% of oregano samples having been adulterated with olive or myrtle leaves. The amount of adulteration ranged from 30% to over 70% of sample weight and two samples had almost no oregano leaves.[8] The substitution or adulteration with one or more other plant materials could adversely affect the consumer’s health because some plant species can accumulate a higher quantity of alkaloids[9] or trace elements[10] than other species. Thus, adulteration of herbs and spices has more than economic repercussions, as it is also a direct threat to public health due to the potentially high content of toxic phytochemicals, synthetic compounds, heavy metals, or allergens. Therefore, one of the biggest challenges for the herbs and spices supply chain is to control such frauds and economically motivated adulteration.

Several studies were conducted in the past to investigate different methods to control herbs and spices fraud and adulteration.[11−13] These studies included adulteration of limited plant species such as saffron (Crocus sativus), turmeric (Curcuma domestica), black pepper (Piper nigrum), ginger (Zingiber officinale), rosemary (Salvia rosmarinus), thyme (Thymus vulgaris), oregano (Origanum vulgare), cinnamon (Cinnamomum sp.), cumin (Cuminum cyminum), sage (Salvia officinalis), and heath speedwell (Veronica officinalis). However, hundreds of herb and spice species are used for herbal infusions[3,14] and we have limited knowledge about their traceability and authenticity. Stable isotope ratio analysis is a widely used analytical method to test the traceability and authenticity of various food commodities.[15,16] Stable isotope ratios of carbon (δ13C), nitrogen (δ15N), sulfur (δ34S), oxygen (δ18O), and hydrogen (δ2H) are influenced by parameters such as climate, rainfall, physiological/metabolic pathway (i.e., photosynthesis, evapotranspiration), geographical characteristics such as altitude, latitude, coastline proximity, as well as soil properties, cultivation practices, and mycorrhizal associations.[17−21] Thus, stable isotope ratios of bioelements are able to discriminate plants based on their biological and geographical origin and cultivation practices.[22]

This analytical approach has also been applied to a few herb and spice species to investigate the cultivation practices of chicory roots (Cichorium intybus)[23] and pepper (Capsicum annuum),[24] geographical origin of saffron[25,26] and Asian ginseng (Panax ginseng),[19,27] botanical origin of vanillin,[28] N fluxes of bladderwrack (Fucus vesiculosus),[29] and photosynthetic pathways of various medicinal plants.[30] To date, the stable isotope ratios of other herbs and spices have not been extensively explored. This study was developed to characterize herb and spice species that are commonly used for herbal infusion in the Italian market, based on the stable isotope ratio analysis of bioelements. Therefore, this paper aims to start exploring and determining the characteristic value ranges for the isotope ratios of different types of herbs and spices, focusing on those commonly available on the Italian market.

Result and Discussions

Here, we present the isotopic ranges of each element studied in 119 samples from 116 plant species and 57 botanical families and describe them from a general point of view. Then, we compare the isotopic profiles of the herbs and spices belonging to the common botanical families within and between botanical families.

General Descriptions of Stable Isotope Ratios in Herb and Spice Samples

Carbon Isotope Ratios

The stable isotope ratio of carbon (δ13C) in all herbs and spices (n = 119) ranges from −31.1 to −11.6‰ with the median value of −28.0‰ (Table ). The δ13C value of terrestrial plants varies primarily according to their photosynthetic pathways; the carbon assimilation in plants occurs through the Calvin cycle (i.e., C3 plants), the Hatch–Slack C assimilation pathway (C4 plants), or the Crassulacean acid metabolism (CAM plants)[22] with representative δ13C values between −35 and −21, −14 and −10, and −20 to −10‰, respectively.[22] The C3 plants presented lower δ13C values due to the significant isotopic discrimination against 13CO2 by rubisco enzyme (∼20‰ compared to atmospheric CO2) during CO2 fixation. In contrast, the isotopic discrimination against 13CO2 by rubisco is limited in C4 plants due to the morphological variation of bundle sheath cells. CAM plants present δ13C values intermediate between C3 and C4 plants, allowing for the CO2 fixation through both pathways.

Table 1

Summary of Stable Isotope Ratios of Carbon, Nitrogen, Sulfur, Oxygen, and Hydrogen in Herbs and Spices (n = 119)

	δ13C (‰)	δ15N (‰)	δ34S (‰)	δ18O (‰)	δ2H (‰)
min	–31.1	–4.7	–5.1	14.7	–158
1st Qu.	–28.7	–0.3	2.7	21.0	–102
median	–28.0	2.0	4.5	22.8	–91
mean	–27.5	1.9	5.1	23.7	–86
3rd Qu.	–27.0	3.7	6.5	25.6	–71
max	–11.6	12.4	22.2	46.3	–12

In this study, the δ13C value of 116 herbs and spices fell between −31 and −24‰ within the range of C3 plants (Table ). The two plant species couch grass (Agropyron repens) and maize (Zea mays) showed a δ13C value −13.2 and −11.6‰, respectively, representing C4 plants (Table ). In addition, the δ13C value of a seaweed, bladderwrack −16‰ (Table ), corresponded to that of macrophytes because the source of C in bladderwrack is dissolved inorganic C with δ13C values around 0‰ by definition, and the δ13C value of such macrophytes is usually between −13 and −20‰ depending upon the source and spatial variability.[29]

Table 2

Stable Isotope Ratios of Carbon, Nitrogen, Sulfur, Oxygen, and Hydrogen of Herbs and Spices in Detail

family/species	common name	δ15N	δ13C	δ34S	δ18O	δ2H
Acoraceae
Acorus calamus	calamus	3.5	–27.1	6.0	25.3	–36
Adoxaceae
Sambucus nigra	elder	4.0	–27.0	2.0	27.1	–69
Anacardiaceae
Schinus molle	peruvian pepper	3.0	–28.1	13.5	27.1	–76
Araliaceae
Eleutherococcus senticosus	Siberian ginseng	–0.1	–28.2	5.6	18.4	–116
Panax ginseng	ginseng	–0.2	–27.1	5.9	21.8	–77
Aspleniaceae
Ceterach officinarum	rustyback	–3.0	–30.0	3.8	23.0	–93
Asteraceae
Achillea millefolium	yarrow	–0.3	–29.1	4.6	24.8	–96
Anthemis nobilis	chamomile	5.4	–28.8	3.0	23.1	–87
Arctium lappa	bardana	–0.5	–24.8	5.2	23.4	–71
Artemisia abrotanum	southernwood	1.8	–26.9	–1.9	22.8	–111
Artemisia absinthium	wormwood	2.9	–26.9	8.6	24.7	–98
Calendula officinalis	marigold	2.3	–29.3	4.5	30.9	–61
Cichorium intybus	chicory	0.8	–28.4	1.4	22.7	–94
Cynara scolymus	Mexican arnica	4.5	–29.3	3.6	22.5	–99
Grindelia robusta	gumweed	8.4	–29.9	3.9	20.6	–98
Heterotheca inuloides	Mexican arnica	–0.7	–25.6	1.8	27.9	–91
Hieracium pilosella	hawkweed	–1.0	–28.2	4.6	19.4	–92
Silybum marianum	milk thistle	4.5	–29.9	2.1	19.4	–114
Solidago virgaurea	Solidago	1.0	–27.7	4.4	23.7	–111
Taraxacum officinale	dandelion	2.2	–27.0	–0.9	24.3	–80
Berberidaceae
Berberis vulgaris	barberry	–2.6	–27.6	3.1	25.0	–71
Betulaceae
Betula pendula	silver birch	–0.9	–28.4	4.0	22.4	–137
Bignoniaceae
Tabebuia avellanedae	pink trumpet tree	5.4	–26.4	7.2	22.5	–52
Boraginaceae
Symphytum officinale	comfrey	1.2	–28.5	–5.1	21.6	–77
Burseraceae
Boswellia serrata	Indian frankincense	0.4	–27.4	8.8	14.7	–158
Cannabaceae
Humulus lupulus	common hop	4.7	–26.3	3.4	19.3	–87
Caprifoliaceae
Valeriana officinalis	valerian	2.4	–26.0	1.1	21.7	–81
Crassulaceae
Rhodiola rosea	roseroot	–2.8	–28.1	2.8	18.9	–99
Cruciferae
Brassica alba	white mustard	4.9	–27.8	5.7	25.7	–130
Capsella bursa-pastoris	Shephard’s purse	3.5	–27.6	4.4	24.4	–103
Sisymbrium officinale	hedge mustard	2.6	–28.0	6.9	26.4	–83
Equisetaceae
Equisetum arvense	field horsetail	4.5	–27.6	2.5	21.7	–114
Ericaceae
Arctostaphylos pungens	European blueberry	–0.7	–27.4	2.2	27.9	–117
Vaccinium myrtillus	bilberry manzanita	–4.7	–30.1	5.3	26.8	–86
Fagaceae
Quercus petraea	sessile oak	–2.1	–26.7	2.7	14.7	–98
Fucaceae
Fucus vesiculosus	rockweed	7.2	–16.1	22.2	21.6	–93
Getianaceae
Gentiana lutea	genziana	–0.5	–26.3	6.7	22.3	–94
Ginkoaceae
Ginko biloba	ginkgo	2.0	–28.9	3.0	20.1	–102
Graminacee
Agropyron repens	couch grass	–4.2	–13.2	2.1	28.9	–66
Zea mays	maize	0.3	–11.6	4.6	20.7	–64
Hypericaceae
Hypericum perforatum	St. Jonh’s-wort	–1.2	–29.3	5.9	23.1	–114
Juglandaceae
Juglans sps.	walnut	1.3	–27.1	6.2	26.7	–46
Lamiaceae
Hyssopus officinalis	hyssop	–1.0	–27.2	2.4	23.3	–106
Lavandula hybrid	lavender	1.0	–28.6	4.7	26.0	–93
Melissa officinalis	balm mint	3.4	–28.3	4.5	20.7	–112
Mentha piperita	peppermint	5.3	–27.3	4.2	20.9	–113
Ocymum basilicum	basil	5.5	–28.3	7.9	27.1	–59
Origanum majorana	marjoram	1.6	–29.0	5.7	29.2	–56
Origanum vulgare	oregano	4.0	–27.4	7.9	22.7	–120
Orthosiphon stamineus	java tea 1	1.1	–29.8	4.2	20.2	–90
Orthosiphon stamineus	java tea 2	1.3	–30.4	3.9	18.0	–101
Thymus serphyllum	wild thyme	0.4	–29.1	4.2	21.8	–108
Salvia officinalis	sage	–1.4	–28.7	6.2	21.4	–106
Lauraceae
Cinammomum zeylanicum	true cinnamon	–0.6	–31.1	19.6	21.9	–35
Leguminosae
Cassia angustifolia	Alexandria senna	2.1	–27.2	7.6	35.0	–92
Galega officinalis	galego	–1.0	–27.0	–4.4	23.0	–112
Glycyrrhiza glabra	liquorice	0.3	–26.1	–3.0	22.8	–74
Ononis spinosa	spiny restharrow	0.4	–28.3	0.8	23.7	–89
Trigonella foenum graecum	fenugreek	1.2	–25.8	12.8	34.3	–72
Liliaceae
Ruscus aculeatus	butcher’s broom	–3.3	–27.4	1.6	28.3	–53
Smilax medica	common salix	–1.0	–25.8	0.5	26.0	–46
Loranthaceae
Viscum album	mistletoe	2.4	–30.2	1.7	19.5	–75
Malvaceae
Althaea officinalis	marsh mallow	3.8	–29.4	3.2	19.3	–115
Malva sylvestris	cheeses	6.6	–28.9	9.3	24.3	–95
Monimiaceae
Peumus boldus	boldo	–1.3	–30.5	10	30.9	–71
Myristicaceae
Myristica fragrans	nutmeg	3.6	–27.3	16.2	23.2	–79
Myrtaceae
Pimenta officinalis	allspice	1.5	–28.2	3.7	23.8	–61
Syzygium aromaticum	clove	2.1	–30.4	18	25.4	–37
Oleaceae
Olea europaea	olive	3.5	–27.1	1.5	23.7	–91
Onaeraceae
Epilobium parviflorum	hoary willow herb	6.5	–27.0	0.2	21.1	–107
Papaveraceae
Chelidonium majus	greater celidona	5.7	–28.3	4.8	21.4	–101
Fumaria officinalis	fumitory	3.2	–26.1	7.1	24.7	–95
Papaveraceae
Eschscholtzia californica	California puppy	5.1	–28.1	5.1	22.9	–87
Passifloraceae
Passiflora incarnata	Passionflower	2.5	–27.4	7.7	20.8	–75
Pedaliaceae
Harpagophytum procumbens	Devils claw	2.6	–26.8	10.4	24.8	–73
Piperaceae
Piper nigrum	black pepper	3.2	–29.1	4.3	26.7	–26
Plantaginaceae
Plantago lanceolata	ribwort plantain	3.1	–29.5	4.7	19.2	–113
Plantago psyllium	Psyllium	4.2	–24.0	0.4	33.1	–65
Veronica officinalis	heath speedwell 1	–0.1	–28.9	4.0	22.4	–101
Veronica officinalis	heath speedwell 2	0.0	–28.6	3.1	22.1	–107
Polygonaceae
Rumex crispus	curly dock	4.0	–26.8	7.2	23.8	–66
Polygonum aviculare	common knotgrass	6.5	–28.7	2.6	22.3	–97
Rheum rhaponticum	false rhubarb	–0.3	–28.1	2.7	18.6	–87
Rhamnaceae
Rhamnus frangula	alder buckthorn	–2.6	–29.7	5.5	19.9	–83
Rosaceae
Agrimonia eupatoria	agrimony	0.8	–28.3	4.9	19.0	–109
Alchemilla vulgaris	lady’s mantle	–1.2	–27.4	6.3	20.8	–103
Crataegus oxyacantha	hawthorns	–2.3	–28.6	3.2	22.7	–93
Filipendula ulmaria max	meadowsweet	3.2	–28.5	4.5	20.0	–108
Prunus cerasus	sour cherry	2.0	–26.8	3.3	24.4	–66
Rubus idaeus	red raspberry	–3.7	–29.2	3.2	22.8	–99
Rubiaceae
Asperula odorata	woodruff	–4.5	–31.1	3.9	21.2	–90
Cinchona succirubra	cinchona	–2.7	–28.2	3.3	17.0	–78
Galium aparine	cleavers	7.2	–27.6	4.3	22.2	–94
Rutaceae
Citrus aurantium var bigaradia	bitter orange	12.4	–26.4	11.7	34.6	–16
Ruta graveolens	common rue	1.8	–27.1	2.0	22.3	–98
Salicaceae
Salix alba	white willow 1	2.7	–29.3	1.4	16.7	–102
Salix alba	white willow 2	1.4	–28.7	–1.8	17.5	–91
Sapindaceae
Paullina cupana	guarana	5.0	–27.3	13.9	27.3	–18
Schisandraceae
Schisandra chinensis	five-flavor berry	–3.3	–27.7	6.0	23.7	–85
Scrophulariaceae
Scrophularia nodosa	common figwort	–0.6	–28.4	4.8	22.8	–104
Theaceae
Thea sinensis	tea	4.7	–27.1	1.1	21.4	–70
Ulmaceae
Ulmus campestris	field elm	1.3	–27.1	5.5	15.2	–77
Umbelliferae
Anethum graveolens	dill	3.3	–27.5	0.2	30.0	–89
Angelica archangelica	angelica	3.5	–25.2	–1.4	22.7	–60
Carum carvi	caraway	2.9	–25.3	13.4	46.3	–51
Coriandrum sativum	coriander	2.0	–28.4	4.6	23.6	–111
Cuminum cyminum	cumin	1.9	–24.2	6.3	37.9	–98
Daucus carota	wild carrot	0.6	–28.1	6.0	24.7	–106
Hydrocotyle asiatica	pennywort	3.3	–28.3	14.4	18.8	–106
Petroselinum crispum	parsley	7.6	–28.2	6.8	26.6	–73
Pimpinella anisum	anise	3.3	–27.3	10.5	35.0	–62
Urticaceae
Parietaria officinalis	litchwort	–2.9	–29.4	4.5	19.1	–97
Urtica dioica	nettles	6.7	–27.6	3.1	19.1	–100
Verbenaceae
Lippia citriodora	lemon verbena	7.1	–28.3	8.4	24.6	–93
Violaceae
Viola tricolor	wild pansy	2.3	–28.1	4.2	23.9	–100
Zingiberaceae
Aframomum melegueta	grains of paradise	4.9	–28.4	11.4	26.7	–35
Curcuma domestica	turmeric	4.4	–25.8	7.5	29.2	–44
Elettaria cardamomum	green cardamom	1.1	–27.7	8.4	26.4	–12
Zingiber officinale	ginger	2.3	–26.6	1.0	21.5	–83

Besides the photosynthetic pathways, other factors such as stomatal conductance, environmental indicators (temperature, solar radiation, relative humidity, and rainfall), nutrient availability, and pathological stress also contribute to the plant’s δ13C value.[17] For instance, reduced photosynthesis due to a limited supply of nutrients and water induces a partial closure of stomata, which reduces the intercellular to ambient CO2 concentrations (ci/ca) and thus leads to a high plant δ13C value. Conversely, the higher concentration of ci/ca dilutes heavier C isotopes (13C) due to C assimilation, which results in low δ13C values. For example, some plant species native to coastal areas or mountainous regions such as true cinnamon (Cinnamomum zeylanicum), woodruff (Asperula odorata), boldo (Peumus boldus), java tea (Orthosiphon stamineus), mistletoe (Viscum album), clove (Syzgium aromaticum merril eugenia), bilberry (Vaccinium myrtillus), and rustyback (Ceterach officinarum) presented δ13C values lower than −30‰ (Table ). On the other hand, the tropical/subtropical climate could be the possible reason of the higher δ13C value of psyllium (Plantago psyllium, −24.0‰), cumin (−24.2‰), and bardana (−24.8‰) (Table ). Therefore, the plant δ13C variation noticed in this study was expected, as these plants were possibly grown/cultivated in different regions with different environmental or agricultural regimes.

Nitrogen Isotope Ratios

The δ15N determined in this study resulted within the range of −4.7 to 12.4‰ (Table ). In most cases, the δ15N value of a plant reflects the available sources of nitrogen, and lower δ15N values (close to 0‰) correspond to the nitrogen sourced from either synthetic fertilizer inputs or symbiotic N fixation, while higher δ15N values are due to organic fertilizer inputs.[22] Furthermore, drought, stress, and distance from the sea lead to an enrichment of 15N atoms in terrestrial plants.[31] Here, most of the plant species (more than 80% of samples) presented δ15N values in a narrow range between −3.0 and 5.0‰. Only 17 species had a δ15N value higher than 5‰, including sour orange (Citrus aurantium var bigaradia, 12.4‰), gumweed (Grindelia robusta, 8.4‰), parsley (Petroselinum crispum, 7.6‰), stickywilly (Galium aparine, 7.2‰), bladderwrack (7.2‰), lemon verbena (Lippia citriodora, 7‰), nettle (Urtica dioica, 6.7‰), cheeses (Malva sylvestris, 6.6‰), hoary willowherb (Epilobium parviflorum, 6.5‰), and so forth (Table ). It seems that these plant species derive N from organic fertilizer inputs because N volatilization during the N mineralization process enhances heavy nitrogen isotope (15N) enrichment in the N sources available for plants and results in a high plant δ15N value.[32] Another possible explanation for higher δ15N values could be that these plant species grew closer to sea or in an arid climate.[29,31] For instance, gumweed and bladderwrack are species native to coastal areas, and the δ15N values of such plants correspond to the sources of N (i.e., agricultural N or sewage waste water)[29] or sea spray (oceanic nitrate).[31] Besides, some plant species such as field horsetail (Equisetum arvense) and gentian (Gentiana lutea) enriched their 13C and 15N isotopes content as compared to adjacent plant species due to the fully or partial mycorrhizal association, respectively.[33,34]

On the other hand, six plant species had δ15N values lower than −3.0‰ such as bilberry (−4.7‰), woodruff (−4.5‰), couch grass (−4.2‰), red raspberry (Rubus idaeus, −3.7‰), Mongolia berry (Schisandra chinensis, −3.3‰), and butcher’s broom (Ruscus aculeatus, −3.3‰) (Table ). Besides, other factors, such as dry atmospheric depositions, physiology and metabolism (uptake, transport, and acquisition), cultivation regime and so forth, may also influence the plant’s δ15N value.[19,20] The slightly negative δ15N values of these species might be the result of substantial isotopic fractionations during plant physiology and metabolism (uptake, transport, and acquisition)[20] if an excess of N is available to plants or plants receive NOx pollutants that are depleted in 15N content as sources of N.

Sulfur Isotope Ratios

The sulfur isotope ratio (δ34S) of the plant species was between −5.1 and 22.2‰ (Table ). Because of the reduction in atmospheric pollution, a major source of S for plants is soil S or fertilizer S inputs and the δ34S of soil and fertilizers is then reflected in the plants.[35] No significant isotopic fractionation occurs during plant uptake and metabolism,[18] so most of the cultivated plant species should present a δ34S value close to that of the soil or fertilizer. The soil δ34S varies largely with the sources; the range is between −50 and 40‰ from sedimentary rock to seawater sulfate,[36] whereas the δ34S value of the most common fertilizer is between −10 and 10‰.[37] On the other hand, the oxidation of S minerals (i.e., pyrite) results in a low δ34S value.[18]

Here, most of the plant species had a δ34S value of less than 10‰. It seems that most of the herbs in the present study get their S from fertilizers or the oxidation of mineral soils. For instance, comfrey (Symphitum officinale, −5.1‰), galego (Galega officinalis, −4.4‰), and liquorice (Glycyrrhiza glabra, −3.0‰) are species native to river basins or natural grasslands (Table ). Only 13 plant species exceeded δ34S values of 10‰ (Table ). The highest δ34S values were observed in plant species native to the coastal areas such as bladderwrack (22.2‰), true cinnamon (19.6‰), clove (18‰), and nutmeg (Myristica fragrans, 16.2‰) (Table ) due to the so called “sea spray effect”.[38]

Oxygen Isotope Ratios

The oxygen isotope ratios (δ18O) of the herb samples were in the range of 14.7 to 46‰, with median values of about 23‰ (Table ). The δ18O of the plant organic matter is the sum of the isotopic signature of source water, its isotopic enrichment through transpiration pool (leaf water), and the isotope exchange between leaf water and source water, CO2, and biomolecules.[21,39,40] The isotopic signature of source water (e.g., precipitation) is influenced by climatic (i.e., temperature and relative humidity) and geographical (i.e., proximity to evaporative sites and altitude) factors and as there is no isotopic fractionation during water uptake, the δ18O of leaf water reflects the geographical origin of the plant, even in the presence of evapotranspiration processes. In general, a plant’s δ18O value is about 28‰ higher than the δ18O value of source water.[21] The plant’s highest δ18O values correspond to plants from tropical climates and/or close to coastal areas, whereas the low values represent plants from temperate climates or those grown at higher altitudes.

In the present study, 76 plant species presented δ18O values of less than 25‰ (Table ). The lowest values of δ18O among all herbs have been observed in herb species native to temperate regions such as sessile oak (Quercus petraea, 14.7‰), field elm (Ulmus campestris, 15.2‰), white willow (Salix alba, 16.7‰), cinchona (Cinchona succirubra, 17.0‰), and mountainous species such as Indian frankincense (Boswellia serrata, 14.7‰) and Siberian ginseng (Eleutherococcus senticosus, 18.7‰) (Table ). On the other hand, nine species showed δ18O values of more than 30‰, mostly native species of tropical or subtropical climate such as caraway (Carum carvi, 46‰), cumin (37‰), Alexandra senna (Cassia angustifolia, 35.0‰), aniseed (Pimpinella anisum, 35.0‰), sour orange (34.6‰), fenugreek (Trigonella foenum graecum, 34.3‰), psyllium (33.0‰), pot marigold (Calendula officinalis, 30.9‰), and boldo (30.9‰) (Table ).

Hydrogen Isotope Ratios

As expected, δ2H values varied from −158 to −12‰ on a wider range than that of δ18O, with most plant species having δ2H values lower than −71‰ (Table ). Similar to δ18O, the plant δ2H is affected by source water and isotopic enrichment through the transpiration pool.[100] Unlike δ18O, the δ2H value is also influenced by various kinetic isotope fractionations during different biosynthetic pathways and isotope exchange with biomolecules,[41] which is supported by the larger variations in the δ2H value as compared to the δ18O (Table ). In the present study, the δ2H values of most plant species lie between −60 and −130‰ (Table ). Eleven plant species which are native to tropical/coastal climate, such as green cardamom (Elettaria cardamomum, −12‰), sour orange (C. aurantium var bigaradia, −16‰), guarana (Paullinia cupana, −18‰ and black pepper (P. nigrum, −26‰), grains of paradise (Aframomum melegueta, −35‰), cinnamon (Cinammomum zeylanicum, −35‰), calamous (Acorus calamus, −36‰), clove (−37‰), turmeric (Curcuma domestica, −44‰), common salix (Smilax medica, −46‰), and walnut (Juglans sp., −46‰) (Table ) showed the highest δ2H values.

On the other hand, the resin of Indian frankincense presented the lowest δ2H −158‰ (Table ). Resins are constituted by terpenoids and phenolic compounds, and due to their biosynthetic pathway, it was already determined that terpenoids undergo 13C depletion in the bulk plant tissues.[42,43] We can suppose that the same occurs also with δ2H. Furthermore, the low δ2H value of Indian frankincense can be expected due to the altitude effect, as it originates in the dry mountainous regions of India, Northern Africa, and the Middle East.[44] Other two low values of δ2H, −130 and −137‰, were observed in plant species native to temperate climates such as white mustard (Brassica alba) and silver birch (Betula pendula), respectively (Table ).

Common Botanical Families and Herbs/Spices Classification

Figure A–E illustrates the stable isotope ratio profiles of common botanical families (Asteraceae, Leguminosae, Lamiaceae, Plantaginaceae, Rosaceae, Umbelliferae, and Zingiberaceae). The common botanical families had a narrow interquartile range of the δ13C value; a difference of less than 1.5‰ was observed between the third and first quartiles, except in Leguminosae and Umbelliferae specimens (Figure A). The δ15N value close to 0‰ in the Leguminosae (Figure B) was associated with N fixation.[22] The median value of δ15N seemed to be lower in the Rosaceae family than other families (Figure B). In particular, the two species red raspberry and Hawthorns produced slightly negative δ15N values (Table ). However, the difference to the other families was not statistically significant (p < 0.05). Moreover, Umbelliferae and Zingiberaceae tended to have slightly high δ15N, δ34S, and δ18O values, but there was no statistically significant difference among them at 95% confidence (Figure B–D). Though the median δ34S of leguminous herbs tended to be the lowest, the values were dispersed from −4.5 to 12‰ (Figure C). Three families, Lamiaceae, Rosaceae, and Plantaginaceae, showed comparable values of both δ18O and δ2H (Figure C, D). Zingiberaceae had significantly higher δ2H than other families (p < 0.05) (Figure D).

Figure 1

Comparison of stable isotope ratios of bioelements; (A) carbon, (B) nitrogen, (C) sulfur, (D) oxygen, and (E) hydrogen among common herbs/spices families. Abbreviate labels of the families; Ast = Asteraceae, Lgm = Leguminosae, Lmc = Lamiaceae, Pln = Plantagoniaceae, Rsc = Rosaceae, Umb = Umbelliferae, and Zng = Zingiberaceae. Boxes in the graph indicate interquartile ranges, horizontal lines inside the boxes are medians, and red dots are outliers. The letter “n” corresponds to the number of samples.

Despite the variabilities in isotopic profiles among the common botanical families, it was not possible to classify herbs and spices. The impact of plant parts isotopic composition on classifying the botanical families was checked. In particular, seeds and roots generally showed slightly higher isotopic values than leaves and other plant parts, but this tendency was not statistically significant (p < 0.05) within each botanical family. A dendrogram of hierarchical clustering also provided an overview of how herbs and spices of a family are dissimilar to each other (Figure S1). It seems that herbs and spices which had seeds as plant parts formed a partially separated cluster (Figure S1). However, the hierarchical cluster was mainly influenced by δ2H and δ18O values, the high values formed a cluster at the left side and low values at the right side. The result indicated that the geographical origin of herbs and spices probably limits the possible classification of botanical families of herbs and spices based on their isotopic profiles. In addition, the plant parts can be considered as a co-factor having some influence on the isotopic variability. Further research with extensive herb and spice samples of different plant parts and with a known geographical origin is recommended.

Materials and Methods

Sampling and Sample Preparation

One-hundred and nineteen dried herb and spice samples were purchased in 2017 from a drug/herbal store in Northern Italy. All the plant samples were commercially certified products according to the botanical origin and belonged to 116 plant species and 57 plant families. A previous study has already explored the alkaloid content of the same samples.[9] The samples included different plant parts such as leaves, foliage (whole above-ground biomass), stem, bark, flower, stigma, fruits (or berries), seed, and roots. Plant samples were grounded and homogenized with a mechanical shaker (Cryomill, Retsch GmbH, Germany).

Stable Isotope Ratio Analysis

About 0.25 ± 0.05 mg of duplicate plant samples was weighed into silver capsules for oxygen and hydrogen isotope ratio analysis and introduced into the TC-IRMS (a pyrolizer—Finnigan TC/EA, high-temperature conversion elemental analyzer, Thermo Scientific, Bremen, Germany—coupled with an isotope ratio mass spectrometer IRMS—Finnigan DELTA XP, Thermo Scientific, Bremen, Germany). Isotope values were corrected with the International standard, USGS 54 (U.S. Geological Survey, Reston, VA, USA, certified values δ2H; −150.4‰, δ18O; +17.79‰) and USGS 56 (U.S. Geological Survey, certified values δ2H; −44‰, δ18O; +27.23‰). The value precisions (i.e., standard deviation) were <5‰ and <1‰ for the hydrogen and oxygen isotope ratio, respectively.

The sample weight for C, N, and S isotope ratio analysis was calculated based on the test results. For the sample weight test, ∼ 2.5 mg of triplicate plant samples of different plant parts (leaves, foliage, flower, bark, and root) were weighed in a tin capsule and introduced into an EA-IRMS (elemental analyzer—VARIO CUBE, Elementar Analysensysteme GmbH, Germany—coupled with an IRMS - Isoprime, Isoprime Ltd., UK). About 2–4 mg triplicate plant samples were introduced into an EA-IRMS for the CNS isotope ratio analysis. The isotopic values for δ13C, δ15N, and δ34S were calculated against in-house working standards, which were themselves calibrated against international reference materials: potassium nitrate IAEA-NO3 (IAEA-International Atomic Energy Agency, Vienna, Austria) for δ15N, l-glutamic acid USGS 40 (U.S. Geological Survey, Reston, VA, USA) for 13C/12C and 15N/14N, fuel oil NBS-22, IAEA-CH-6 for 13C/12C, barium sulfates IAEASO-5, NBS 127 (IAEA), and USGS 42 and USGS 43 for δ34S. Through the development of a linear equation, the values were calculated against two working standards. For the δ13C, δ15N, δ34S, δ18O, and δ2H, the uncertainty of measurement (1 Std Dev) was 0.1, 0.2, 0.3, 0.3, and 1‰, respectively.

Data Analysis

Due to the limited information on the sample labeling and the presence of only a few samples of herb and spice species in some families, sample sizes were narrowed down to common botanical families (at least four samples in each family) for a statistical approach. Normal distribution and homogeneity of variances for each isotopic ratio were tested in RStudio software version 1.3.1073. One-way analysis of variance for normally distributed data (i.e., the δ13C, δ15N, and δ34S) and a nonparametric test (the Kruskal–Wallis test) for the data without normal distribution (δ18O and δ2H) were applied to check significant differences according to the plant family or plant parts. Partial least square regression discriminant analysis and classification and regression tree were tested in XLSTAT Microsoft excel but produced low accuracy in the classification of botanical families. A dendrogram of the hierarchical clustering (Euclidean distance based on Ward’s minimum variance method) was created to illustrate the dissimilarities of botanical families. All the graphics were prepared in RStudio software version 1.3.1073.