Intraspecies diversity of the bioactive compounds of wild and cultivated Cinnamomum species in Sri Lanka

Abstract

Background

Cinnamomum verum has significant biochemical diversity across Sri Lanka, encompassing over 500 distinct chemotypes. Additionally, six wild relatives of Cinnamomum found in the region exhibit substantial morphological, genetic, and phytochemical variability, often shaped by their adaptation to isolated habitats. This research aims to explore the inter and intraspecies phytochemical diversity, volatile oil profiles, and environmental influences on wild Cinnamomum species, providing valuable insights for conservation, sustainable utilization, and breeding programs to enhance cinnamon's economic and ecological resilience. This study analyzed the inter and intraspecies diversity of six bioactive compounds in leaf and bark extracts using HPLC, the leaf volatile oil profiles (LVOP) through GCMS, and the leaf oil yield (LOY) of six wild Cinnamomum species. Samples were collected from at least two distinct agroecological zones, including both domesticated plots and natural habitats.

Results

HPLC analysis revealed that none of the species had detectable coumarin contents. Notably, C. sinharajaense from a domesticated plot in Matara presented significantly greater LOY (1.53%), leaf-eugenol content (30.38 mg/g), and bark-cinnamaldehyde content (8.20 mg/g) than its wild counterpart (Sinharaja forest: 0.32%, 4.99 mg/g and 1.67 mg/g). This performance was also superior to C. verum-wild and competitive with cultivated varieties. Cinnamomum capparu-coronde presented competitive bark-cinnamaldehyde levels at both locations (8.18/8.34 mg/g), with no significant difference, whereas its LVOP presented intraspecies variation in the relative abundances of eugenol (8.36/6.24%), benzyl benzoate (28.92/5.62%), bicyclogermacrene (5.54/19.77%), and gamma-muurolene (10.03/5.37%). GCMS analysis also revealed unique LVOPs for C. litseifolium, C. capparu-coronde, C. citriodorum, and C. dubium with high intraspecies diversity in terms of relative abundance, the total number of compounds, and prominent compounds. The intraspecies variability in leaf-eugenol, bark-cinnamaldehyde, and LVOP was correlated with the mean annual rainfall and temperature.

Conclusions

The intraspecies biochemical diversity observed in C. sinharajaense, C. capparu-coronde, and C. dubium highlights their potential to adapt to diverse environmental conditions. Additionally, the observed inter- and intraspecies variations in all wild species emphasize the critical importance of conserving these underutilized cinnamon species through both ex-situ and in-situ conservation strategies.

Introduction

The genus Cinnamomum Schaeff. is a member of the family Lauraceae. This group consists of more than 250 species distributed in Asia, Australia, and the Pacific Islands [1]. Among those, Cinnamomum verum J. Presl (syn. Cinnamomum zeylanicum Blume), Cinnamomum burmanni (Nees & T.Nees) Blume, Cinnamomum tamala (Buch.-Ham.) T. Nees & C.H. Eberm., Cinnamomum tazia (Buch.-Ham.) Kosterm. ex M. Gangop and Cinnamomum camphora (L.) J. Presl. are recognized as economically viable species [1]. In addition to C. verum, seven other Cinnamomum species have been reported in Sri Lanka. They are Cinnamomum capparu-coronde Blume, Cinnamomum citriodorum Thwaites, Cinnamomum dubium Nees, Cinnamomum litseifolium Thwaites, Cinnamomum ovalifolium Wight, Cinnamomum rivulorum Kosterm, and Cinnamomum sinharajaense Kosterm. Cinnamomum citriodorum and C. rivulorum are categorized as endangered, whereas other wild species are included in the vulnerable category, except for C. dubium, categorized as a near-threatened species [2]. Furthermore, some of these species, such as C. sinharajaense, C. rivulorum, C. litseifolium, and C. ovalifolium, are naturally grown in isolated habitats restricted to specific geographical locations and conditions [3].

Cinnamon is among the earliest known spices used in human civilization for its intensely aromatic, sweet, warm smell and taste. The demand for true cinnamon is increasing worldwide because of recent and vital scientific evidence on its medicinal and pharmacological properties [4]. While historical evidence suggests that C. verum has grown naturally in the up-country rainforests in Sri Lanka before being cultivated by Dutch invaders, its natural populations still remain in such locations. This germplasm would be extremely useful in future breeding efforts to improve the traits of cultivated cinnamon [4].

A substantial amount of literature also exists on wild relatives of cultivated cinnamon species in other countries. For example, the essential oils of five Cinnamomum species growing wild in north-central Vietnam were studied via gas chromatography (GC) and screened for antimicrobial and mosquito larvicidal activity [5]. These species include C. damhaensis, C. longipetiolatum, C. ovatum, C. polyadelphum, and C. tonkinense. Leela et al. [6] investigated the essential oil composition of C. citriodorum growing in India from leaves and petioles via gas chromatography-mass spectrometry (GCMS) analysis [6]. Furthermore, Vairappan et al. [7] investigated the essential oil composition of six species of wild cinnamon discovered in Borneo. The essential oils of the studied samples contained high levels of oxygenated monoterpenes, and 65 volatile constituents were identified [7].

Similarly, several morphological, molecular, and biochemical studies have been conducted on wild cinnamon species in Sri Lanka. Liyanage [3] studied the distributional ecology of wild Cinnamomum in Sri Lanka [3]. Bandusekera et al. (2020) studied the leaf morphology of wild species in detail and introduced a key for the identification of those species under field conditions. Furthermore, the study revealed high intraspecies morphological variation in C. verum, C. sinharajaense, and C. dubium [8]. A recent study reported high intraspecies genetic diversity of the cultivated species C. verum [9], whereas considerable intraspecies diversity was reported in wild species such as C. sinharajaense, C. litseifolium, C. ovalifolium, C. capparu-coronde and C. dubium [10]. Moreover, several studies have indicated that the bark-cinnamaldehyde and leaf-eugenol contents of C. sinharajaense and C. capparu-coronde are competitive with C. verum [11, 12]. This is the first study to focus on the intraspecies biochemical diversity and leaf volatile oil profiles (LVOPs) of wild Cinnamomum species collected from various locations and under various environmental conditions. Additionally, the study specifically examined samples of C. verum collected from the wild, which has not been previously reported in any other research.

This study is driven by the significant ecological, medicinal, and economic importance of Cinnamomum species, alongside the increasing global demand for true cinnamon and the urgent need for sustainable utilization of its genetic resources. Despite their immense value, wild Cinnamomum species in Sri Lanka remain largely underexplored, particularly in terms of their intraspecies phytochemical diversity. Previous studies have primarily focused on interspecies phytochemical diversity, leaving a critical gap in understanding the adaptability and potential of wild relatives in different environmental settings for breeding, conservation, and commercial applications.

To ensure the sustainable use and conservation of these wild cinnamon species, it is vital to employ species-specific chemical fingerprinting and investigate intraspecies diversity while considering environmental influences. While Cinnamomum species are well known for their medicinal and aromatic properties, their phytochemical composition has not been sufficiently examined across diverse agroecological zones. This study addresses this gap by exploring the biochemical diversity and environmental adaptability of wild Cinnamomum species from multiple agroecological zones, focusing specifically on bioactive compounds, volatile oil profiles, and leaf oil yields.

The findings provide crucial insights into how environmental factors influence both the therapeutic efficacy and commercial value of these species, with potential applications in pharmaceuticals, cosmetics, and culinary industries. Furthermore, this research lays a foundation for informed conservation strategies, promoting the sustainable utilization of wild Cinnamomum species and supporting the development of climate-resilient cinnamon varieties. By identifying unique phytochemical profiles and exploring their potential in breeding programs, this study advances our understanding of Cinnamomum diversity, with practical implications for improving disease resistance, pest tolerance, and crop performance. Ultimately, this research is vital for ensuring the resilience, quality, and economic viability of Cinnamomum species globally, while also contributing to their conservation in the face of climate change and growing commercial demand.

Methods

Sample collection

Samples were collected from a minimum of two different agroecological zones in Sri Lanka for each species on the basis of availability. Major soil types, mean annual rainfall, mean annual temperature and soil fertility levels were well described in each agroecological zone by the Natural Resource Management Centre Soil Map in Sri Lanka [13]. While some of the species were collected from domesticated research plots, others were previously identified by Liyanage [3]. The details of the samples and different locations are listed in Table 1. All wild species were taxonomically identified utilizing an authenticated reference collection housed at the National Herbarium, Peradeniya, Sri Lanka. Specimens of wild Cinnamomum species were collected and subsequently deposited in the National Herbarium. Detailed information regarding the collection sites and voucher numbers for each specimen is provided in Supplementary Table 1.

Table 1 HPLC biochemical compounds of bark and leaf methanol-extracts from seven Cinnamomum species collected from various locations in Sri Lanka

Sample preparation

Bark and leaf samples were collected separately from the first secondary branch of selected plants. For uniformity, all harvested branches were 9–11 cm in circumference and pest- and disease-free. The bark was hand-peeled as chips and air-dried for ten days. Only mature, healthy leaves from harvested branches were air-dried for five days. To obtain fine, even-sized particles, dried leaves, and bark were ground into a powder and sieved through US sieve mesh number eight (typical nominal wire diameter = 1.00 mm).

HPLC analysis of bark and leaf methanol extracts

Commercial standards of trans-cinnamaldehyde (PubChem CID: 637,511), eugenol (PubChem CID: 3314), coumarin (PubChem CID: 323), coumaric acid (PubChem CID: 637,542), cinnamyl alcohol (PubChem CID: 5,315,892), cinnamyl acetate (PubChem CID: 5,282,110), and all HPLC-grade reagents includes acetonitrile (ACN) (PubChem CID: 6342), orthophosphoric acid (PubChem CID: 73,357,664) and methanol (PubChem CID: 887) for extraction and analysis were purchased from Sigma‒Aldrich (St. Louis, MO, USA).

HPLC was used to analyze the bark and leaf samples separately. The methanol extraction was conducted according to the method described by Liyanage et al. [14, 9], with minimal modifications [14]. Finely powdered leaf and bark samples (25 mg) were placed in clean 2 mL centrifuge tubes with 1.5 mL of 100% methanol. After sonicating for 30 min at room temperature, centrifuged at 10,000 rpm for 10 min, the supernatant was filtered through a 0.45 μm nylon filter and analyzed.

The separation and quantitative determination of selected compounds were performed via an Agilent 1260 chromatography system (Agilent Technologies) with a diode array detector (DAD). A Zorbax Eclipse Plus C18 column (150 mm × 4.6 mm, particle size 5 µm) was used, and the column temperature was set at 20 °C. Ten-microliter sample was injected and separated using a solvent gradient. The gradient flows for the two-solvent-system include 0.1% (v/v) orthophosphoric acid in water and a mixture of 50–100% (v/v) acetonitrile (ACN). Orthophosphoric acid (0.1% v/v in water) continuously flows the whole time (45 min). The ACN-time combinations were 20% (v/v) ACN for 20 min, 50% (v/v) ACN for 10 min, and 100% (v/v) ACN for 15 min. The flow rate of the mobile phase was 1.0 mL/min. The highest concentrated peaks for trans-cinnamaldehyde, cinnamyl alcohol, cinnamyl acetate, eugenol, coumarin, and coumaric acid were detected at 290 nm, 265 nm, 265 nm, 285 nm, 280 nm, and 290 nm, respectively. The sample was tested in triplicate to assess any technical error in the process. Representative HPLC profiles standards and samples are shown in Supplementary Fig. 1.

Each standard was prepared in a series of concentrations ranging from 0.5 ppm to 1000 ppm in methanol to obtain standard curves. The limit of detection (LODs) and the limit of quantifications (LOQs) were calculated via the calibration curve (Supplementary Fig. 2) linearity and residual standard deviation for the measured chemical compounds described by Fan and Stewart [15].

Determination of leaf volatile oil yield and moisture content

The essential oil was extracted by hydrodistillation in a Clevenger-type apparatus, according to the method of Demirci et al. [16] with minor modifications. Fifty grams of powdered leaf sample was placed in a 500 mL round bottom flask, and 300 mL of deionized water was added to immerse the leaf sample completely in water. The flask was placed on a heating device and connected to the Clevenger apparatus for 5 h to extract the volatile leaf oil. The extracted oil was trapped in a Clevenger's arm using a 3 mL mixture of n-hexane and petroleum ether (2:1). The oil accumulated in the arm was separated from the water and transferred into clean glass bottles, and the water droplets at the bottom of the oil layer were removed. The oil was weighed after removing the organic solvent, and traces of water saturated in the oil were taken to determine the oil yield. The extracted oil samples were stored at 4 °C.

The moisture content of the samples was determined according to the standardized procedure AOCS (1993) [17] via the Dean and Stark distillation method. Five grams of powdered leaf sample was transferred into a round bottom flask, and 80 mL of toluene was added to immerse the sample entirely in toluene. After the apparatus was assembled, the flask containing the leaf sample was continuously heated for 2 h and allowed to cool to room temperature, and the volume of water accumulated in the graduated tube was measured as the moisture content of the sample.

GC‒MS analysis of leaf volatile oils

Gas chromatography-mass spectrometry (GCMS) analysis of oil (1.0 µl) was performed on an HP-5 ms Ultra Inert column: 30 m × 250 μm × 0.25 μm (0 °C—325 °C −350 °C) using an Agilent 19091S-433UI GC system equipped with an MSD-5977A single quadrupole mass spectrometer. The injector and mass spectra transfer line temperatures were set at 200 °C and 220 °C, respectively, and the oven temperature was programmed at an initial temperature of 50 °C with an increasing rate of 5 °C/min up to 120 °C (hold time of 2 min.) and at a final temperature of 300 °C (hold time of 4 min.) with the same increasing rate of 5 °C/min. The compounds were identified via mass Hunter/LIBRARY/NIST14. L and Mass Hunter\ LIBRARY\ W9N11. L libraries. The results of the quantitative analysis expressed as area percentages, were carried out via peak area normalization.

Statistical analysis

The oil yield and HPLC analyses were carried out in triplicate, and technical replicates were used. The analysis was subjected to the GLM procedure followed by Duncan's Multiple Range mean separation at the 0.05 probability level. The correlation coefficients of the environmental factors and biochemical properties were also determined. The analysis was conducted via the SAS 8.0 statistical package. The dendrogram was constructed according to the "similarity distance of single linkage and squared Euclidean distance" using the HPLC concentrations of six major chemicals present in the leaf and bark methanol extracts, the relative abundances of twenty selected leaf oil compounds, and the leaf oil yield. The resulting clusters indicate the biochemical similarity of the members. The statistical analyses were performed via the statistical package Minitab 18 version (State College, PA: Minitab, Inc.).

Results

HPLC analysis of bark and leaf methanol extracts

Coumarin content

The HPLC analysis of methanol extracts from the bark and leaves of Cinnamomum species collected from various locations in Sri Lanka is presented in Table 1. None of the species exhibited detectable levels of coumarin or coumaric acid in the leaf or bark samples. The present LOD value for coumarin (0.0014 g/kg) indicates that both cultivated and wild species of the genus Cinnamomum in Sri Lanka comply with the daily coumarin intake recommendations, suggesting these species possess more health attributes compared to Cassia cinnamon.

Bark-cinnamaldehyde and leaf-eugenol contents

The intraspecies variation in bark-cinnamaldehyde and leaf-eugenol contents across different locations is shown in Table 1. Bark-cinnamaldehyde content was varied from 0.09 mg/g to 18.31 mg/g while leaf-eugenol varied from 0.32 mg/g to 38.13 mg/g. Cinnamomum verum commercial varieties Sri gemunu and Sri wijaya presented significantly higher bark-cinnamaldehyde and leaf-eugenol contents from all the studied species. For C. verum cultivated varieties, the bark-cinnamaldehyde content ranged from 18.31 mg/g to 12.89 mg/g, with no significant intraspecies variation among the different locations (Table 1). The leaf-eugenol content ranged from 20.44 mg/g to 38.13 mg/g, indicating significant intraspecies diversity across different locations. Cinnamomum verum-wild samples reported very low bark-cinnamaldehyde (2.13—4.20 mg/g) and leaf-eugenol content (13.05—23.56 mg/g) compared to it cultivated varieties.

Cinnamomum capparu-coronde presented the second-highest bark-cinnamaldehyde content, with similar values at two locations: 8.34 mg/g in Delpitiya and 8.18 mg/g in Matara, indicating no significant intraspecies variation. However, there was a notable difference in the leaf-eugenol content between the locations, ranging from 1.75 mg/g in Delpitiya to 4.57 mg/g in Matara. Such intraspecies variations in leaf-eugenol were also recorded from C. litseifolium (Delpitiya: 4.36 mg/g and Matara: 8.71 mg/g) and C. sinharajaense. The bark-cinnamaldehyde and leaf-eugenol contents of C. sinharajaense differed significantly between the two locations. Bark-cinnamaldehyde varied from 8.20 mg/g (Matara) to 1.67 mg/g (Sinharaja), whereas it was 30.38 mg/g and 4.99 mg/g for leaf-eugenol (Table 1).

Notably, C. capparu-coronde and C. sinharajaense (collected from Matara) exhibited significantly higher bark-cinnamaldehyde contents compared to the C. verum-wild samples collected from Bibila (4.20 mg/g) and Norwood (2.13 mg/g). Cinnamomum citriodorum and C. litseifolium presented relatively low levels of bark-cinnamaldehyde content, which was not detectable in C. dubium or C. ovalifolium (Table 1). Leaf-eugenol was not detected in C. ovalifolium, whereas C. dubium and C. citriodorum were present at low levels.

Other chemical compounds

The intraspecies variation in other bark and leaf biochemical compounds at different locations is shown in Table 1. The highest level of bark-eugenol content was reported in C. litseifolium collected from Delpitiya (5.11 ± 0.06 mg/g), followed by Matara (2.72 ± 0.04 mg/g). Cinnamomum verum cultivated varieties/wild samples, C. sinharajaense, and C. capparu-coronde had very low bark-eugenol content but significantly differed among different locations (Table 1). Cinnamomum ovalifolium did not significantly differ in terms of bark-eugenol content compared with the two different locations, Hakgala and H’Plains, whereas C. dubium and C. citriodorum presented no detectable level of bark-eugenol (Table 1).

The highest bark-cinnamyl-acetate content reported in the C. litseifolium sample collected from Delpitya was 1.06 ± 0.02 mg/g, which was significantly lower than that reported for the Hakgala sample (0.02 ± 0.01 mg/g). In addition, C. capparu-coronde, C. sinharajaense, and C. verum varieties presented lower bark-cinnamyl-acetate contents but presented significant intraspecies diversity for different locations. The remaining species had no detectable level of bark-cinnamyl-acetate content (Table 1). The highest level of bark-cinnamyl-alcohol was reported in C. ovalifolium, which did not significantly differ between the H’Plains (1.57 ± 0.23 mg/g) and Hakgala (1.68 ± 0.34 mg/g) locations. The second-highest bark-cinnamyl-alcohol content was reported from the C. litseifolium sample collected from Hakgala, which was not significantly different from that of C. ovalifolium (Table 1). The Cinnamomum capparu-coronde sample collected from Delpitiya showed a significantly higher bark-cinnamyl-alcohol content compared to the sample from Matara. In contrast, no intraspecies variation in bark-cinnamyl-alcohol content was observed among the cultivated and wild samples of C. verum or in C. sinharajaense. Cinnamomum dubium and C. citriodorum had no detectable level of bark-cinnamyl-alcohol content.

Leaf-cinnamaldehyde content also varied within species but was reported in only the samples of cultivated C. verum, C. capparu-coronde Matara sample, and the C. sinharajaense Matara sample, while the remaining samples were not detectable (Table 1). Leaf-cinnamyl acetate was completely absent in C. verum-wild, C. sinharajaense, C. capparu-coronde, and C. ovalifolium while it was present in C. litseifolium and C. citriodorum (Table 1). Leaf-cinnamyl-alcohol was completely absent in C. verum-wild, C. citriodorum, C. ovalifolium, and C. dubium, while it was present in C. sinharajaense and C. capparu-coronde (Table 1).

Leaf oil yield and organoleptic properties

The leaf oil yield (LOY) varied from 0.12% to 3.21% at the genus level, while the lowest value was recorded for C. ovalifolium. The significantly highest LOY was reported from the C. verum cultivated variety Sri gemunu, followed by the C. sinharajaense Matara sample, at 3.21% and 1.53%, respectively. Compared with the cultivated variety Sri gemunu, C. verum-wild sample (0.92%) had a significantly lower LOY, whereas the C. sinharajaense sample from Sinharaja (0.32%) also had a significantly lower LOY than the sample from Matara (1.53%) (Table 2). Such intraspecies variation in LOY was also reported in C. citriodorum (Matara: 0.42%, Norwood: 0.32%) and C. dubium (Matara: 0.72%, Sinharaja: 0.23%) (Table 2). Cinnamomum litseifolium, C. ovalifolium, and C. capparu-coronde did not show intraspecies variation in LOY between different locations (Table 2). In organoleptic properties, C. verum can be distinguished easily from other wild species with unique bark tastes, bark fragrances, and leaf aromas. Cinnamomum citriodorum and C. capparu-coronde leaf and their oils have unique citronella, and the Kapuru aroma could be useful for distinguishing both species from the rest. The oil colour and aroma of other species, including C. litseifolium, C. ovalifolium, and C. dubium, did not differ at the intra- or interspecies level.

Table 2 Leaf volatile oil yield and composition in seven Cinnamomum species from various locations in Sri Lanka

GCMS analysis of leaf oil volatile composition

Table 2 presents the compositions and relative abundances of major volatile oil compounds in the leaves of wild and cultivated Cinnamomum species. The present study revealed that over 50 different biochemical compounds from the genus and 20 were economically important. The GCMS profiles revealed that while some species, such as C. sinharajaense and C. verum, have similar chemical compositions (Tables 2 and Supplementary Fig. 3–9), others, such as C. citriodorum, C. dubium, C. ovalifolium, C. rivulorum, and C. litseifolium, exhibit unique characteristics. However, the relative abundances of similar compounds varied between C. verum and C. sinharajaense (Table 2). From the genus, the lowest number of compounds was observed from Sri gemunu (8), while the highest number of compounds was observed from the C. dubium Matara sample (32).

A total of 34 different volatile compounds were identified from the essential leaf oil of C. capparu-coronde: 31 in Delpitiya and 29 in Matara. The major and mutual volatile compounds recorded in Delpitiya and Matara were in decreasing order: benzyl benzoate (28.92%, 5.26%), bicyclogermacrene (5.54%, 19.77%), and 3-carene (12.81%, 8.55%). alpha-copaene (0.28%, 11.14%), gamma-muurolene (10.03%, 5.37%) and alpha-linalool (6.21%, 4.23%). The most abundant compounds for Delpitiya were benzyl benzoate (28.92%) and 3-carene (12.81%), while it was bicyclogermacrene (19.77%) and alpha-copaene (11.14) for the Matara sample (Table 2 and Supplementary Fig. 3).

A total of 26 volatile compounds were detected in C. citriodorum, 19 in Matara and 18 in Norwood. The major and mutual volatile compounds recorded in the Matara and Norwood samples in decreasing order were citronellal (42.16%, 15.53%), citronellol (26.55%, 19.46%), 3-carene (3.21%, 13.19%), alpha-guaiene (2.64%, 4.89%), and eugenol (1.23%, 4.76%). Six compounds were recorded only in the Matara sample, whereas four were recorded only in the Norwood sample (Table 2 and Supplementary Fig. 4).

A total of 39 volatile compounds were detected in C. dubium, with 35 in Matara and 19 in Sinharaja, indicating high intraspecies variation in the number of compounds. The major volatile compounds recorded at both locations (Matara and Sinharaja) were in decreasing order: alpha-phellandrene (18.01%, 19.66%), beta-santalol (0.32%, 14.22%), terpineol (1.59%, 13.57%), benzyl benzoate (6.58%, 2.04%), 3-carene (6.37%, 0.41%), and alpha-pinene (4.57%, 0.24%). Six compounds, such as o-cymene (7.56%), cis-muurola-3,5-diene (6.44%), naphthalene (6.10%), gamma-elemene (4.63%), germacrene D (4.62%) and beta-caryophyllene (4.02%), were detected only in the Matara sample. Some economically important compounds, such as eucalyptol, terpineol, and beta-santalol, were highly abundant in the Sinharaja sample (Table 2 and Supplementary Fig. 5).

A total of 35 different compounds were detected from C. litseifolium: 12 in Delpitiya and 32 in Hakgala. The major and mutual volatile compounds recorded in Delpitiya and Hakgala were methyl eugenol (61.02%, 11.56%), eugenol (23.86%, 0.72%), beta-linalool (4.56%, 15.42%) and 3-carene (9.19%, 7.48%). Furthermore, some compounds, such as cis-calamenene (11.89%), humulene (6.23%), azulene (4.17%), beta-caryophyllene (3.82%), benzaldehyde (2.43%), alpha-copaene (2.39%) and o-cymene (2.34%), were detected only in Hakgala (Table 2 and Supplementary Fig. 6).

A total of 34 different compounds were detected from C. ovalifolium: 25 in Hakgala and 23 in H’plains. The major and mutual volatile compounds recorded in the Hakgala and H’plains were naphthalene (34.21%, 48.98%), alpha-muurolene (3.07%, 6.63%), alpha-calacorene (4.22%, 2.92%) and tetradecanal (2.73%, 2.98%). Furthermore, some compounds were recorded only in Hakgala: aristolene (8.12%), methyl eugenol (4.70%), gamma-muurolene (4.36%), and eugenol (1.47%) (Table 2 and Supplementary Fig. 7).

A total of 22 different compounds were detected from C. sinharajaense: 18 in Matara and 12 in Sinharaja. The major and mutual volatile compounds recorded in Matara and Sinharaja were eugenol (90.53%, 94.71%), benzyl benzoate (4.69%, 1.23%), beta-caryophyllene (0.67%, 1.89%), and 3-carene (1.09%, 0.72%). Both samples presented very high relative abundances (area percentage) of eugenol (Table 2 and Supplementary Fig. 8).

A total of 15 compounds were recorded from the C. verum-wild and C. verum variety Sri gemunu. The most abundant compound in Sri gemunu is eugenol, with a 95.26% relative abundance. Interestingly, when all the species were considered, Sri gemunu presented the lowest number of compounds (08). In addition to eugenol, several other compounds, such as beta-caryophyllene (1.44%), 3-carene (1.00%) and humulene (0.26%), were present. The most abundant compounds for C. verum-wild are eugenol (28.14%), naphthalene (21.56%) and gamma-muurolene (17.05%) (Table 2 and Supplementary Fig. 9).

Correlations between environmental factors and biochemical compositions

Table 3 shows the correlations among the mean annual temperature (MAT), mean annual rainfall (MAR), leaf-eugenol content, bark-cinnamaldehyde content, LOY, total number of biochemical compounds, and major biochemical compounds. Both the leaf-eugenol and bark-cinnamaldehyde contents were positively correlated with MAT and LOY. The total number of biochemical compounds in the leaf oil was negatively correlated with the MAR and leaf-eugenol content.

Table 3 Correlation coefficients between the environmental factors and biochemical properties of Cinnamomum species

The present study also confirmed that the bark-cinnamaldehyde content was positively correlated with the mean annual temperature, whereas the mean annual rainfall was negatively correlated with the number of major biochemical compounds. For example, C. verum and C. sinharajaense are effective when they are available at high mean annual temperatures and low rainfall.

Combined cluster analysis

A clustering pattern was revealed based on the concentrations of four major chemicals measured via HPLC and their total number of compounds, the relative abundance of twenty leaf oil compounds, and the leaf oil yield. The species were grouped into two main groups at a 57.4% similarity level (Fig. 1). Cinnamomum verum cultivated variety and wild species were grouped with both samples of C. sinharajaense (Matara and Sinharaja) (Fig. 1). Cinnamomum verum variety Sri gemunu was more closely clustered with C. sinharajaense than with C. verum-wild. This clustering pattern could be mostly due to the leaf-eugenol and bark-cinnamaldehyde content which are prominently available in this cluster (Fig. 1). The rest of the species were clustered together as a second cluster and C. ovalifolium separated from this cluster at a similarity distance of 71.6%. Even though C. capparu-coronde had significant bark-cinnamaldehyde which was clustered with C. litseifolium could be a result of the similarity of other chemical compounds available in the two species.

Fig. 1

Dendrogram illustrating the intraspecies biochemical relationships of seven Cinnamomum species in Sri Lanka. Note: The dendrogram was constructed according to the “similarity distance of single linkage and squared Euclidean distance” using the HPLC concentrations of four major chemicals present in the leaf and bark methanol extracts, the relative abundance of twenty leaf oil compounds and the leaf oil yield

Discussion

Coumarin content

Coumarin content in cinnamon influences its global trade and industrial value. Coumarin, a compound considered carcinogenic when present in relatively high quantities, has significant implications for the cinnamon industry [18]. The European Parliament and the Council of the European Union (EPCEU) have established limits for the maximum acceptable coumarin content, ranging from 0.05 g/kg in traditional and/or seasonal bakery products to 0.005 g/kg in desserts (REGULATION (EC) No 1334/2008) [19]. The absence of detectable coumarin in Sri Lankan Cinnamomum species underscores their compatibility with these regulations and highlights their potential as safer alternatives in the cinnamon market. Furthermore, since the samples were collected from different agroecological zones of Sri Lanka, the presence or absence of coumarin may be more strongly controlled by genetic factors than by environmental factors.

Intraspecies variation in bark-cinnamaldehyde and leaf-eugenol contents

Bark cinnamaldehyde and leaf eugenol are key compounds driving cinnamon's global trade and industrial value. Investigating and studying these economically significant compounds is essential for advancing the industry. A recent study in Sri Lanka investigated 500 chemotypes of C. verum, analyzed via HPLC, revealing an average bark-cinnamaldehyde content of 8.912 ± 5.220 mg/g and a leaf-eugenol content of 21.706 ± 12.014 mg/g [14]. Interestingly, the results of the present study demonstrated competitive levels of bark-cinnamaldehyde content from both C. sinharajaense and C. capparu-coronde, as well as leaf-eugenol content from C. sinharajaense. Poole and Poole [20] revealed that the amount of leaf-eugenol and bark-cinnamaldehyde is a good indicator for determining the botanical origin of Cinnamomum [20]. This finding also suggests that the botanical origin of C. sinharajaense is much closer to that of the cultivated species C. verum. A recent genetic study also revealed that the C. verum cultivated variety Sri gemunu and the wild species C. sinharajaense Matara presented the lowest sequence divergence, suggesting the genetic relatedness of these two species [10]. Moreover, the present study reported high interspecies diversity in back-cinnamaldehyde and leaf-eugenol content of Cinnamomum species in Sri Lanka. Recent studies have also reported the interspecies biochemical [11, 12] and morphological diversity [8] of wild and cultivated Cinnamomum species. Notably, the present study marked the inaugural investigations into the intraspecific biochemical diversity of Cinnamomum in Sri Lanka, shedding light on the nuanced variations existing within a single species in different locations.

Most of the study species have shown intraspecies variation in bark-cinnamaldehyde and leaf-eugenol except for C. dubium, C. ovalifolium, and C. citriodorum. Cinnamomum verum and C. sinharajaense exhibit significant intraspecies variation in bark-cinnamaldehyde, while C. sinharajaense, C. verum, and C. litseifolium display high intraspecies variation in leaf-eugenol. Cinnamomum verum exhibits significant intraspecies molecular diversity across Sri Lanka, with over 500 chemotypes varying in leaf-eugenol and bark-cinnamaldehyde content [14]. This variation is influenced by both genetic and environmental factors. However, vegetatively propagated C. verum cultivars also display notable variation in these compounds, highlighting the substantial impact of environmental factors on their chemical composition. Therefore, the intraspecies biochemical diversity observed in C. sinharajaense and C. litseifolium, as determined by HPLC analysis, cannot be completely attributed to intraspecies genetic diversity. Environmental factors such as mean annual rainfall, temperature, soil conditions, and shade levels, which vary significantly across locations, also contribute to this intraspecies variation. The biochemical composition of cinnamon, specifically the synthesis of bioactive compounds such as cinnamaldehyde and eugenol, is intricately influenced by the mean annual temperature and rainfall. Temperature plays a vital role in plants' enzymatic activity and metabolic processes, impacting the production of biochemical compounds [21]. Higher temperatures often lead to increased metabolic rates and altered enzymatic activity, influencing the quantity of biochemical compounds present in cinnamon. The present finding also indicates greater intraspecific variation in the leaf-eugenol content than in the bark-cinnamaldehyde content across locations. The greater variability in leaf-eugenol content is attributed to leaves being more exposed to environmental factors such as light, temperature, and water stress, whereas the bark is more protected, resulting in stable bark-cinnamaldehyde production than leaf-eugenol.

Apart from bark-cinnamaldehyde and leaf-eugenol, other chemicals had considerable variability among the genus Cinnamomum, which was previously reported by a recent study [11, 12]. The present study confirms those findings and further investigates the substantial level of intraspecies variation, especially in terms of bark-eugenol, bark/leaf-cinnamyl acetate, and bark/leaf-cinnamyl alcohol content. Such a variation could be mainly manipulated by environmental factors rather than genetics.

GCMS analysis of leaf oil volatile composition

Interspecies variation in leaf oil composition and relative abundance was evident across all the species. While some species, such as C. sinharajaense and C. verum, have similar LVOP, others, such as C. citriodorum, C. dubium, C. ovalifolium, C. rivulorum, and C. litseifolium, exhibit unique characteristics. Such unique compounds in LVOP could result from variations in gene expression, gene mutations, and genetic recombination. This contributes to the production of unique compounds in specific cinnamon species [22]. This could be due not only to genetic effects but also to environmental effects, both at the macro and micro levels. Environmental factors play a pivotal role in shaping compound composition. Varying climates, soil conditions, sunlight exposure, and altitude affect the expression of genes involved in compound biosynthesis [23]. Cinnamon species adapt to their respective environments, leading to variations in compound production and leaf oil accumulation [22]. Furthermore, Cinnamon species such as C. sinharajaense, C. ovalifolium and C. litseifolium have evolved in response to specific ecological niches, leading to adaptations and variations in compound composition. Environmental pressures, such as interactions with pathogens, herbivores, and competitors, influence the production of defensive compounds in leaf oils [24]. These adaptations account for the species-specific compositions observed.

High intraspecies variation in LVOP was observed across most of the species while C. capparu-coronde and C. citriodorum exhibiting slight variation. The observed intraspecies variation encompassed the total number of biochemical compounds, the number of major biochemical compounds, the relative abundance of economically important biochemical compounds, and the presence of unique compounds. Such variation confirms the influence of genetic and environmental conditions on the intraspecies biochemical variation of cinnamon.

The present study further reveals that species located in warmer, low-altitude conditions exhibited a greater relative abundance of eugenol in LVOP, higher LOY, and bark-cinnamaldehyde content. Environmental factors significantly influence the relative abundance and composition of LVOP, potentially resulting in notable variations in controlled or domesticated plots. Specifically, C. verum, C. sinharajaense, C. citriodorum, C. litseifolium, C. capparu-coronde, and C. dubium have diverse LVOP in response to different environmental conditions. These variations could be reflected in the compounds present in their leaf oils. Moreover, the investigation highlights the capacity of wild cinnamon species to exhibit distinct performances when subjected to domesticated conditions. This underscores their potential for future breeding initiatives, particularly in response to environmental changes and associated challenges. Consequently, conservation efforts employing both ex-situ and in-situ methods are imperative for the preservation of these underutilized wild species, ensuring enhanced utilization in the future. Introducing selected species such as C. sinharajaense, C. citriodorum, C. litseifolium, C. capparu-coronde, and C. dubium to different environmental conditions for ex-situ conservation could benefit the industry by potentially altering and enhancing their biochemical profiles.

Despite its comprehensive approach, this study has certain limitations that should be addressed in future research. The limited sample size and geographic representation for each species and agroecological zone may restrict the generalizability of the findings to broader populations. Furthermore, the biochemical variability observed may have been influenced by uncontrolled environmental and seasonal factors, such as microclimatic conditions and soil properties. While HPLC and GCMS analyses provided detailed biochemical insights, the study did not incorporate advanced molecular approaches, such as metabolomics or genomics, which could reveal the underlying molecular mechanisms driving the observed diversity. To address these limitations, future research should focus on increasing sample sizes and expanding geographic coverage to include a broader range of agroecological zones. Implementing effective conservation approaches will be essential to support these efforts. Genetic investigations, including genome-wide association studies and transcriptomic analyses, could elucidate the genetic architecture and pathways regulating metabolite biosynthesis. Additionally, both in-situ and ex-situ conservation strategies should be implemented to safeguard the genetic and biochemical diversity of wild cinnamon species. Research on propagation methods and agronomic practices should prioritize the preservation of genetic diversity while improving yield and quality. Exploring the pharmaceutical, nutraceutical, and industrial applications of the unique chemical profiles of underutilized cinnamon species could enhance their economic potential and support their sustainable utilization.

Conclusions

In conclusion, this study highlights that both wild and cultivated species of the Cinnamomum genus in Sri Lanka do not contain detectable levels of coumarin. Significant intraspecies variation was observed, particularly in the leaf-eugenol and bark-cinnamaldehyde contents, with C. sinharajaense and C. capparu-coronde. While all the species presented unique leaf oil compositions, C. verum, and C. sinharajaense presented a similar biochemical profile. The leaf oil GCMS analysis identified over 50 biochemical compounds, 20 of which hold economic importance, with their relative abundance varying due to environmental and genetic factors. Furthermore, domesticated plants of C. sinharajaense and C. capparu-coronde demonstrated competitive performance compared with cultivated varieties. The observed intraspecies diversity holds potential for future breeding efforts and suggests that wild species may exhibit competitive performance when cultivated under domesticated conditions.