Dracaena (Langenheim, 2003)

Vaníčková, Lucie, Pompeiano, Antonio, Maděra, Petr, Massad, Tara Joy & Vahalík, Petr, 2020, Terpenoid profiles of resin in the genus Dracaena are species specific, Phytochemistry (112197) 170, pp. 1-8 : 2-6

3.2. Volatile compounds in Dracaena resin

Our comparison of terpenoid profiles was based on an evaluation of 20 compounds, 13 of which are described in Dracaena resin volatiles for the first time here. These compounds include five monoterpenes, namely α -thujene, α -pinene, camphene, β -pinene, and δ -2-carene, and eight sesquiterepenes, namely (−)-isodauca-6,9-diene, γ -elemene, trans- muurola-3,5-diene, γ -humulene, γ -himachelene, ε - and ω -amorphene, and α -muurolene. Furthermore, this is the first report of the volatile composition of D. serrulata and D. ombet resins. Many terpenoid compounds identified in Dracaena resins are common plant volatiles ( El-Sayed, 2012) and have valuable medicinal properties, including anti-carcinogenic, antimalarial, antiulcer, antimicrobial, antiseptic, nematicidal, larvicidal, anti-inflammatory and diuretic properties ( Schwab et al. 2008). Nevertheless, the majority of studies on Dracaena are focused on the isolation and identification of flavonoids and sterols ( Baumer and Dietemann, 2010; Gupta et al. 2008; Masaoud et al. 1995; Yi et al. 2011), while only a limited number of reports describe the composition of terpenoid volatiles from Dracaena resins . Santos et al. (2011) used SPME-GC-MS to analyze leaf extracts of D. draco (from the Azores, Portugal) and detected 31 components, eight of which were terpenes. Both our work and that of Santos et al. (2011) document the presence of limonene, α -calacorene, and caryophyllene oxide in Dracaena spp. resins. Twenty-six volatiles were also identified from D. cochinchinensis resin extracted in hexane and run on a GC-MS; these include three sesquiterpenes (τ -cadinol, τ -muurolon and α-cadinol; Teng et al. 2015). GC-MS analyses of D. reflexa leaf volatiles revealed 16 terpenoid compounds, 13 of which are monoterpenes and three of which are sesquiterpenes ( Gurib-Fakim and Demarne, 1994). Two of the D. reflexa monoterpenes, δ -3-carene and p -cymene, were also identified in the present study. In comparison with previous reports on Dracaena volatile organic compounds (VOCs), our method allows for the identification of a wider spectrum of terpenoids. This may be due to the combination of the analytical approaches we used, including the optimized SPME technique for VOCs collection and our very sensitive GC × GC-MS method.

3.3. Species specificity in Dracaena resins

We demonstrated that the composition of terpenoids in Dracaena is species specific. We therefore propose that monoterpene profiles of Dracaena resins may be evaluated in future studies as chemotaxonomic traits that allow for the identification of the species origin of Dracaena resins . In pines, the composition of monoterpenes in cortical oleoresin changes with location and season ( Mita et al. 2002). In addition, high variation in the presence/absence of particular volatile compounds was found within Boswellia species (Burseraceae); this variability could have been caused by different environmental features associated with the trees sampled, by differences in the timing of sampling, or by differences in the part of the trees sampled (stem base vs. annual shoots; Maděra et al. 2017). Nonetheless, studies collectively suggest the composition of Dracaena resin is species specific. In previous reports on Dracaena marker determination, Edward et al. (2001) found differences among resin from two species, D. draco and D. cinnabari . Key vibrational spectroscopic marker bands were identified in the Raman spectra of the resins, and they were used for species identification and determination of the geographical origin of samples ( Edward et al., 2001). Gonzalez et al. (2004) compared the composition of VOCs in the resin of D. draco and D. tamaranae , finding that the chemical composition of resins from D. draco subsp. draco (Canary Islands and Cape Verde) and D. draco subsp. ajgal . Together with our data and the results of Gonzalez et al. (2004) and Sousa et al. (2008), these findings all demonstrate the chemical composition of Dracaena resin is an accurate and useful species identifier.

Currently D. draco is found only in a very restricted area of southern Morocco. It is noteworthy that these Moroccan populations have been designated as their own subspecies, D. draco subsp. ajgal , an indication that the taxon is in the early stages of speciation ( Marrero et al. 1998). If chemotypes align with phylogeny, our results would suggest that the most closely related species are D. draco subsp. draco from the Canary Islands and subsp. ajgal from Morocco followed by D. ombet from Ethiopia and D. cinnabari from Socotra. Dracaena serrulata from Oman is the most distinct from the other four. However, Lu and Morden (2014) investigated the phylogenetic relationship among Dracaenoid genera and species using chloroplast DNA loci and have come to different conclusions. According to their results the most closely related species are D. draco and D. serrulata , whereas D. cinnabari and D. ombet form distinct clusters and are also distant from each other. In contrast, morphological work suggests that D. draco is closely related to the Socotra species, D. cinnabari ( Marrero et al., 1998) . Work with other groups of plants has similarly demonstrated that chemotypes do not always match phylogenies. Becerra (1997) found only a weak relationship between phylogeny and chemical similarity for Bursera species, common trees in the dry forests of Mexico. Likewise, Kursar et al. (2009) found a weak correlation between phylogenetic and chemical distances within the Neotropical tree genus, Inga . Overall chemical similarity between species was also not associated with phylogeny in the genus Protium ( Salazar et al., 2018) . This lack of phylogenetic signal in the expression of specialised metabolites suggests divergent selection on antiherbivore defences, such that closely related species do not necessarily produce similar defences. This should make it more difficult for herbivores to track hosts over evolutionary time, thereby reducing herbivore pressure on plants and resulting in the evolutionary lability of defensive traits ( Endara et al., 2015). Specialised plant metabolites, such as the terpenes studied here, play important ecological and evolutionary roles, most notably in the deterrence of natural enemies. Detailed studies of herbivores and chemical diversity should be undertaken in the genus Dracaena in order to better understand its phytochemical richness and contribute to ecological investigations of chemically mediated plant-insect interactions.