Next Article in Journal / Special Issue
The Essential Oil Composition of Trachymene incisa Rudge subsp. incisa Rudge from Australia
Previous Article in Journal
Impact of Rising Temperature in the Deposition Patterns of Bioactive Compounds in Field Grown Food Barley Grains
Previous Article in Special Issue
The Volatile Phytochemistry of Monarda Species Growing in South Alabama
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Plants
Volume 10
Issue 3
10.3390/plants10030600
plants-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles thumb_up
...
Endorse textsms
...
Comment
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Chemical Characterization of Marrubium vulgare Volatiles from Serbia

by
Milica Aćimović
1,*,
Stefan Ivanović
2,
Katarina Simić
2,
Lato Pezo
3,
Tijana Zeremski
1,
Jelena Ovuka
1 and
Vladimir Sikora
1
1
Institute of Field and Vegetable Crops Novi Sad, Maksima Gorkog 30, 21000 Novi Sad, Serbia
2
Institute of Chemistry, Technology and Metallurgy, University of Belgrade, 11000 Belgrade, Serbia
3
Institute of General and Physical Chemistry, University of Belgrade, 11000 Belgrade, Serbia
*
Author to whom correspondence should be addressed.
Plants 2021, 10(3), 600; https://doi.org/10.3390/plants10030600
Submission received: 28 February 2021 / Revised: 10 March 2021 / Accepted: 16 March 2021 / Published: 23 March 2021
(This article belongs to the Special Issue Essential Oils of Plants (Chemical Composition, Variation and Properties))
Browse Figures
Versions Notes

Abstract

:
Marrubium vulgare is a cosmopolitan medicinal plant from the Lamiaceae family, which produces structurally highly diverse groups of secondary metabolites. A total of 160 compounds were determined in the volatiles from Serbia during two investigated years (2019 and 2020). The main components were E-caryophyllene, followed by germacrene D, α-humulene and α-copaene. All these compounds are from sesquiterpene hydrocarbons class which was dominant in both investigated years. This variation in volatiles composition could be a consequence of weather conditions, as in the case of other aromatic plants. According to the unrooted cluster tree with 37 samples of Marrubium sp. volatiles from literature and average values from this study, it could be said that there are several chemotypes: E-caryophyllene, β-bisabolene, α-pinene, β-farnesene, E-caryophyllene + caryophyllene oxide chemotype, and diverse (unclassified) chemotypes. However, occurring polymorphism could be consequence of adaptation to grow in different environment, especially ecological conditions such as humidity, temperature and altitude, as well as hybridization strongly affected the chemotypes. In addition, this paper aimed to obtain validated models for prediction of retention indices (RIs) of compounds isolated from M. vulgare volatiles. A total of 160 experimentally obtained RIs of volatile compounds was used to build the prediction models. The coefficients of determination were 0.956 and 0.964, demonstrating that these models could be used for predicting RIs, due to low prediction error and high r2.

1. Introduction

Marrubium vulgare L., also known as white horehound, is a perennial species from the Lamiaceae family. It is indigenous to the region between the Mediterranean Sea and Central Asia; however, today it is found worldwide, apart from the coldest regions and high altitudes [1]. This plant is highly resistant to drought and due to this it grows well in semiarid areas [2]. Additionally, as it is a moderate salt-tolerant species this medicinal plant could be grown on saline soil [3]. The surface of M. vulgare vegetative and generative organs is densely clothed with glandular and nonglandular trichomes which accumulate secondary metabolites [4]. M. vulgare produces structurally highly diverse groups of secondary metabolites, thus represents a valuable source of bioactive compounds and preparations with health-promoting effects: antioxidant, hepatoprotective, antiproliferative, anti-inflammatory, antidiabetic, and antimicrobial [5]. The use of this herb in traditional medicine is recorded worldwide for ameliorating chronic cough and cold, numerous conditions related to skin, liver, gastric, heart, and immune system [6]. Generally, M. vulgare is poor in essential oil, and the major compounds are diverse [1,7,8,9,10,11,12,13,14,15,16,17]. This proves that there are various chemotypes of M. vulgare. The lack of information in this field is pointed out by Yabrir [1,18]. The studies about genus Marrubium are mainly focused on taxonomical, morphological, and genetic diversity [4,19,20,21,22,23,24,25,26].
The main aim of this investigation was to determine volatiles composition of M. vulgare grown in Serbia during two years and to compare its chemical composition with literature data not only of M. vulgare but with other species from this genus as well (M. anisodon, M. aschersonii, M. astracanicum, M. crassidens, M. deserti, M. duabense, M. parviflorum, M. peregrinum, M. persicum, M. propinquum, M. velutinum). Another goal was to establish the new quantitative structure retention relationship (QSRR) models for anticipating the retention indices (RIs) of certain compounds in M. vulgare volatiles obtained by GC–MS chromatography utilizing the genetic algorithm (GA) variable selection method and the boosted trees regression. Furthermore, we gather information about the volatile compounds of species from Marrubium genus in order to classify the chemotype of M. vulgare from this study according to unrooted cluster tree.

2. Results

The main components in M. vulgare volatiles were E-caryophyllene with 24.6% and 23.0%, followed by germacrene D with 9.6% and 17.0%, α-humulene with 5.2% and 5.3% as well as α-copaene with 3.3% and 6.1% in 2019 and 2020, respectively. All these compounds are from the sesquiterpene hydrocarbons class which was dominant in both years of the investigation, 52.0% in 2019 and 67.8% in 2020. This variation in volatiles composition could be a consequence of weather conditions, as in case of other aromatic plants [27,28,29,30,31,32,33].
However, some of the components detected in M. vulgare volatiles during the two-year research have not been detected yet in this species, while other components have not been detected in other species of this genus. ScienceDirect Elsevier, SpringerLink, PubMed, Scopus, Scifnder, Web of Science, Wiley Online, and Google Scholar databases were reviewed and scientific publications from 1990 until 2020 that deal with chemical composition of volatiles species from genus Marrubium were summarized and shown in Table 1.
The predicted RIs are shown in Table 1, and confirm the good quality of the constructed BRT model by showing the relationship between the predicted and experimental RI values. Graphical comparison between experimentally obtained RIs of M. vulgare volatiles composition (RIa), the retention time indices found in NIST database (RIb) and the retention time indices predicted by the two BRT models (RIpred.) are presented in Figure 1.
In order to calculate the molecular descriptors, the PaDel-descriptor was used in this investigation. Due to a great amount of data that was obtained, it was required to select the most important set of descriptors to build the adequate model which would be able to predict the RIs [55]. The factor analysis was done before the GA calculation, and only ca. 320 uncorrelated descriptors remained in the GA calculation [56,57]. The seven most significant molecular descriptors chosen by GA are as follows: four 2D autocorrelation descriptors (AATSC4e, AATSC2p, GATS6v and MATS5v), two Barysz matrix descriptors (VR1_Dzs and SM1_Dzv) and Vertex adjacency information (magnitude) descriptor (VAdjMat).
The predicted RIs and molecular descriptors are presented in Table 1. Seven molecular descriptors were utilized for predictions of RIs in the two BRT models. The predicted RIs are presented in Figure 2, and visually confirm the adequate prediction capabilities of the constructed BRT by showing the relationship between the predicted and experimental retention values.
Separation of compounds in GC–MS and their RIs is linked to their affinity towards mobile and stationary phases. Affinity and solubility of separated molecules directly depend on their chemical structure and physico-chemical properties, which could be expressed by molecular descriptors. According to Pearson’s correlation coefficients, there was a rather poor correlation between all 3D autocorrelation descriptors (Table 2). Therefore, utilized molecular descriptors were appropriate to predict RIs of compounds in M. vulgare by the two multivariate BRT models [58].
Detailed explanations about the descriptors were found in the Handbook of Molecular Descriptors [59]. These descriptors encode different aspects of the molecular structure and were applied to develop the QSRR model. According to Pearson’s correlation, there was a rather poor correlation between all molecular descriptors. Hence, utilized descriptors were appropriate to predict RIs of compounds isolated from M. vulgare volatiles by the two multivariate BRT models. The calibration and predictive capability of a QSRR model should be tested through model validation. The most widely used squared correlation coefficient (r2) can provide a reliable indication of the fit of the model, thus, it was employed to validate the calibration capability of a QSRR model.
In order to explore the nonlinear relationship between RIs and the descriptors selected by GA, BRT technique was used to build the two predictive models. Two BRT models were constructed to predict the retention time of compounds isolated from M. vulgare volatiles, respectively. The coefficients of determination were 0.956 and 0.964, respectively, indicating that these models could be used for prediction of RIs, due to low prediction error and high r2. The tests of the two BRT models fit (2019 and 2020) are shown in Table 3, with the higher r2 values and lower χ2, MBE, RMSE, and MPE values showing the better fit to the experimental results [60,61].
Obtained results reveal the reliability of the BRT models for predicting the RIs of compounds in M. vulgare volatiles obtained by GC–MS analysis. The influence of the seven most important molecular descriptors, identified by using genetic algorithm on the RIs was studied in this section. According to the Figure 3, VAdjMat was the most important molecular descriptor for chemical compounds’ RIs calculation in M. vulgare, during 2019, while VR1 Dzs was the most important variable during 2020.

3. Discussion

According to the cluster analysis (unrooted cluster tree) with 37 samples of Marrubium sp. volatiles from literature and average values from this study (Figure 4), it could be said that there are several chemotypes, but only E-caryophyllene chemotype [9,12,36,38,47,50,51] is clearly segregated. However, these are samples of M. vulgare, M. incanum, M. parviflorum, M. peregrinum, and M. crassidens grown in Serbia, Poland, Slovakia, Egypt, and Iran. This indicated that genus Marrubiumis very diverse in the case of volatiles composition.
In addition, E-caryophyllene is a compound which is occurring in all samples (except M. vulgare from Eastern Algeria [15]), but in E-caryophyllene chemotype its content ranges between 15.6% and 45.8%. Other chemotypes can be classified as β-bisabolene (13.1–28.3%) [11,17,34,47], α-pinene (21.5–28.9%) [16,53], β-farnesene (20.2–24.2%) [37,44], E-caryophyllene + caryophyllene oxide chemotype [44,46,54], and diverse (unclassified) chemotypes [7,8,10,13,15,35,38,40,41,43,45,49,52].
Occurring polymorphism could be a consequence of adaptation to grow in different environments [19,62], especially ecological conditions such as humidity, temperature and altitude [22] as well as hybridization [20] strongly affected the chemotypes, as well as biotic and abiotic stresses (including temperature, light, water, salt, and oxidative stresses) [63].
Detected compounds in M. vulgare volatiles obtained by GC–MS analysis were used for QSRR analysis. The following seven molecular descriptors that characterize the RIs of obtained compounds were suggested by the genetic algorithm. Selected molecular descriptors were not autocorrelated which was suggested by a correlation coefficient matrix; thus, descriptors were suitable for QSRR analysis. These descriptors were utilized as inputs for the boosted trees regression models, for estimating the RIs using a set of GC–MS data from a series of 160 compounds found in M. vulgare volatiles. Statistical models that quantify the relation between the structure of molecules and their chromatographic RIs were represented by the quantitative structure retention relationship (QSRR) model [58,64]. Numerous publications related to the QSRR analysis in plants from Lamiaceae family could be found in the literature: Thymus vulgaris [65], T. serpyllum [66], Satureja kitaibelii [55], Salvia officinalis [67], as well as Stachys sp. [68]. The connection between the molecular descriptors and the retention time can be established by artificial neural network, machine learning algorithms [69,70,71,72,73] or by boosted trees regression (BRT) [74].

4. Materials and Methods

4.1. Plant Material

M. vulgare was grown in the Institute of Field and Vegetable Crops (IFVCNS) collection garden of medicinal and aromatic plants in Bački Petrovac (45°21′ N; 19°35′ E), confirmed by Milica Rat, PhD, and deposited at the Herbarium BUNS (Department of Biology and Ecology, Faculty of Sciences, University of Novi Sad) under the voucher number 2-1409. After seed maturation (August 2018), it was collected and sown in field conditions in September 2018 and 2019. The experimental plot was 10 m long and 5 m wide, with a 70 cm spacing between rows. From the seven rows, only three central rows were used for collecting plant material to avoid edge effects (one row one sample).

4.2. Volatiles Isolation and Analysis

Flowering aerial parts of M. vulgare (Marrubii herba) were collected during July 2019 and 2020, dried in a solar dryer at a temperature of 40° with air circulation. After drying, plant material was fragmented, and volatiles was extracted by Clevenger apparatus. Taking in account that M. vulgare produces trace amounts of essential oil, it was trapped in n-hexane. This process was performed in tree repetition for both years, as well as analysis of chemical composition.
GC–MS analysis was carried out using an Agilent 7890A apparatus equipped with a 5975 C MSD, FID and a nonpolar HP-5MS fused-silica capillary column (30 m × 0.25 mm, film thickness 0.25 μm). The carrier gas was helium, and its inlet pressure was 19.6 psi and linear velocity of 1 mL/min at 210 °C. The injector temperature was 250 °C, injection volume was 1 μL, split ratio, 10:1. Mass detection was carried out under source temperature conditions of 230 °C and interface temperature of 315 °C. The EI mode was set at electron energy, 70 eV with mass scan range of 40–600 amu. Temperature was programmed from 60 to 300 °C at a rate of 3 °C/min. The components were identified based on their linear retention index relative to C8-C32 n-alkanes, compared with data reported in the literature (Adams4 and NIST11 databases). The relative percentage of the oil constituents was expressed as percentages by FID peak area normalization.

4.3. QSRR Analysis

PaDel-descriptor software was used to calculate specified molecular descriptors [75], as described in our previous investigation [66]. Factor analysis and genetic algorithm (GA) were used to determine the most important descriptors [76,77]. The relationship between the chosen descriptors was examined and collinear descriptors were excluded. Statistica 10 software was used for the statistical investigation of the data [78].

4.4. BRT Model

In order to relate and to predict categorical or continuous dependent variables the BRT model could be used [79,80], as it does not require transformation or outliers [81]. The BRT method calculation is connected to the boosting methods enforced to regression trees [82]. The main idea is to calculate a set of simple trees, where each successive tree is built for the prediction residuals of the preceding tree [83]. This method builds binary trees such as partition the data into two samples at each split node [78].
The decision trees are combined through a cross-validation or “boosting” procedure in order to acquire the single computational model [84]. BRT modeling consists of the following steps: (a) an initial regression tree is defined according to a minimum loss function; (b) the other trees are engaged in the iterative process in which several new regression trees were developed and selected to the subsequent according to the StatSoft Statistica’s recommendation—the least square error (LSE); (c) step (b) is repeated until a stopping criterion is reached (for instance, the value of LSE).
In this study, several regularization parameters were set in order to optimize the fit between experimental results and computing model: the number of trees (between 100 and 1000), learning rate (between 0.0005 and 0.1), random test data proportion (0.1–0.9) and subsample proportion (0.1–0.9). According to Statistica’s manual, prior to computation, a subsample of data is created, according to random test data proportion of the cases, and these data are treated as test samples used to evaluate the appropriate fit of the model. The remaining set of data is used for the analyses via stochastic gradient boosting (for the selection of samples for consecutive boosting steps).

4.5. Cluster Analysis

The cluster analysis (CA) was used to evaluate intra- and interpopulation variability and differentiation of volatile constituents of Marrubium samples collected in different locations and/or taken from literature reports. The phylogenetic tree diagram for Marrubium samples was calculated and plotted using R software 4.0.3 (64-bit version). The R package “ape” (Analysis of Phylogentics and Evolution) was used for calculation, applied as a graphical tool to represent the arrangements of similar volatiles concentration (evaluated in the cluster analysis). The obtained experimental results were collected in the matrix, after which the hierarchical cluster analysis was performed. The distance matrix was determined using the Euclidean method, while the cluster analysis was performed using the “complete” method.

5. Conclusions

The main components in M. vulgare volatiles were E-caryophyllene (24.6% and 23.0%), followed by germacrene D (9.6% and 17.0%), α-humulene (5.2% and 5.3%) and α-copaene (3.3% and 6.1%) in 2019 and 2020, respectively. All these compounds are from sesquiterpene hydrocarbons class, which was dominant in both years of the investigation, 52.0% in 2019 and 67.8% in 2020.
The results demonstrated that the boosted trees regression models were adequate in predicting the RIs of the compounds in M. vulgare volatiles obtained by GC–MS analysis on a HP-5MS column. The coefficients of determination were 0.956 and 0.964 (for compounds found in M. vulgare volatiles, during the years 2019 and 2020, respectively), which is a good indication that these models could be used as a fast mathematical tool for prediction of RIs, due to low prediction error and moderately high r2. Suitable models with high statistical quality and low prediction errors were derived, and it could be further used for estimation of RIs of newly detected compounds.
According to the unrooted cluster tree with 37 samples of Marrubium sp. volatiles from literature and average values from this study, it could be said that there are several chemotypes: E-caryophyllene, β-bisabolene, α-pinene, β-farnesene, E-caryophyllene + caryophyllene oxide chemotype, and diverse (unclassified) chemotypes. However, occurring polymorphism could be a consequence of adaptation to grow in different environments, especially ecological conditions such as humidity, temperature and altitude, as well as hybridization which strongly affected the chemotypes. Further research on M. vulgare chemotypes needs to be focused on genetic markers, because evaluation of genetic diversity has key importance in improving the quality of raw material used for industrial purposes.

Author Contributions

Conceptualization, M.A. and J.O.; methodology, S.I. and K.S.; software, L.P.; validation, S.I., K.S. and T.Z.; formal analysis, L.P.; investigation, M.A.; resources, M.A. and V.S.; data curation, L.P.; writing—original draft preparation, M.A.; writing—review and editing, T.Z. and V.S.; visualization, L.P.; supervision, J.O.; project administration, T.Z.; funding acquisition, M.A., T.Z., J.O. and V.S. All authors have read and agreed to the published version of the manuscript.

Funding

This research was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia, grant number: 451-03-9/2021-14/200032.

Data Availability Statement

Not applicable.

Acknowledgments

We thank Vele Tešević, Marina Todosijević, Jovana Stanković Jeremić, Jovana Ljujić, and Mirjana Cvetković for participating in this research.

Conflicts of Interest

The authors declare no conflict of interest.

References

  1. Yabrir, B. Essential oil of Marrubium vulgare: Chemical composition and biological activities. A review. Nat. Prod. Sci. 2019, 25, 81–91. [Google Scholar] [CrossRef] [Green Version]
  2. Lippai, A.; Smith, P.A.; Price, T.V.; Weiss, J.; Lloyd, C.J. Effects of temperature and water potential on germination of horehound (Marrubium vulgare) seeds from two Australian Localities. Weed Sci. 1996, 44, 91–99. [Google Scholar] [CrossRef]
  3. Nedjimi, B.; Souissi, Z.E.; Guit, B.; Daoud, Y. Differential effects of soluble salts on seed germination of Marrubium vulgare L. J. Appl. Res. Med. Aromat. Plants 2020, 17, 100250. [Google Scholar] [CrossRef]
  4. Dmitruk, M.; Haratym, W. Morphological differentiation of non-glandular and glandular trichomes on Marrubium vulgare L. Mod. Phytomorphol. 2014, 6, 85. [Google Scholar]
  5. Aćimović, M.; Jeremić, K.; Salaj, N.; Gavarić, N.; Kiprovski, B.; Sikora, V.; Zeremski, T. Marrubium vulgare L.: A phytochemical and pharmacological overview. Molecules 2020, 25, 2898. [Google Scholar] [CrossRef]
  6. Lodhi, S.; Vadnere, G.P.; Sharma, V.K.; Usman, M.R. Marrubium vulgare L.: A review on phytochemical and pharmacological aspects. J. Intercult. Ethnopharmacol. 2017, 6, 429–452. [Google Scholar] [CrossRef]
  7. Mahmoud, A.A.; Gendy, A.S.H.; Said-Al Ahl, H.A.H.; Grulova, D.; Astatkie, T.; Abdelrazik, T.M. Impact of harvest time and water stress on the growth and essential oil components of horehound (Marrubium vulgare). Sci. Hortic. 2018, 232, 139–144. [Google Scholar] [CrossRef]
  8. Said-Al Ahl, H.A.H.; Gendy, A.S.H.; Mahmoud, A.A.; Mohamed, H.F.Y. Essential oil composition of Marrubium vulgare L. cultivated in Egypt. Int. J. Plant. Sci. Ecol. 2015, 1, 138–141. [Google Scholar]
  9. El-Leithy, A.S.; El-Hanafy, S.H.; Omer, E.A.; El-Sayed, A.A.A. Effect of nitrogen and potassium biofertilization on growth, yield and essential oil production of the white horehound, Marrubium vulgare L. plant. J. Hortic. Sci. Ornam. Plants 2013, 5, 46–59. [Google Scholar]
  10. Weel, K.G.C.; Venskutonis, P.R.; Pukalskas, A.; Gruzdiene, D.; Linssen, J.P.H. Antioxidant activity of horehound (Marrubium vulgare L.) grown in Lithuania. Lipid 1999, 101, 395–400. [Google Scholar] [CrossRef]
  11. Morteza-Semnani, K.; Saeedi, M.; Babanezhad, E. The essential oil composition of Marrubium vulgare L. from Iran. J. Essent Oil Res. 2008, 20, 488–489. [Google Scholar] [CrossRef]
  12. Golparvar, A.R.; Hadipanah, A.; Mehrabi, A.M.; Armin, A. Essential oil composition of Marrubium vulgare L. from Iran. Flavour Herb. Drug 2015, 6, 1–5. [Google Scholar]
  13. Zarai, Z.; Kadri, A.; Chobba, I.B.; Mansour, R.B.; Bekir, A.; Mejdoub, H.; Gharsallah, N. The in-vitro evaluation of antibacterial, antifungal and cytotoxic properties of Marrubium vulgare L. essential oil grown in Tunisia. Lipids Hum. Health Dis. 2011, 10, 161. [Google Scholar] [CrossRef] [Green Version]
  14. Kadri, A.; Zarai, Z.; Bekir, A.; Gharsallah, N.; Damak, M.; Gdoura, R. Chemical composition and antioxidant activity of Marrubium vulgare L. essential oil from Tunisia. Afr. J. Biotechnol. 2011, 10, 3908–3914. [Google Scholar]
  15. Abadi, A.; Hassani, A. Essential oil composition and antioxidant activity of Marrubium vulgare L. growing wild in Eastern Algeria. Int. Lett. Chem. Phys. Astron. 2013, 9, 17–24. [Google Scholar] [CrossRef]
  16. Bayir, B.; Gunduz, H.; Usta, T.; Sahin, E.; Ozdemir, Z.; Kayir, O.; Sen, O.; Aksit, H.; Elmastas, M.; Erenler, R. Chemical composition of essential oil from Marrubium vulgare L. leaves. J. New Results Sci. 2014, 6, 44–50. [Google Scholar]
  17. El-Hallous, E.I.; Alsanie, W.F.; Ismail, I.A.; Dessoky, E.S. Utilization of Marrubium vulgare extract as a therapeutic to hepatic damage induced by Carbon Tetrachloride in rats. Int J. Pharm. Res. Allied Sci. 2018, 7, 168–178. [Google Scholar]
  18. Yabrir, B. Chemical composition and biological activities of some Marrubium species essential oil: A review. Chem. J. Mold. 2018, 13, 8–23. [Google Scholar] [CrossRef]
  19. Salehi, N.; Kharazian, N.; Shiran, B. Genetic diversity of Marrubium species from Zagros Region (Iran), using inter simple sequence repeat molecular marker. J. Sci. Islamic Repub. Iran. 2018, 29, 7–19. [Google Scholar]
  20. Kharazian, N.; Hashemi, M. Chemotaxonomy and morphological studies in five Marrubium L. species in Iran. Iran. J. Sci. Technol. Trans. Sci. 2017, 41, 17–31. [Google Scholar] [CrossRef]
  21. Akgul, G.; Ketenoglu, O.; Pinar, N.M.; Kurt, L. Pollen and seed morphology of the genus Marrubium (Lamiaceae) in Turkey. Ann. Bot. Fenn. 2008, 45, 1–10. [Google Scholar] [CrossRef]
  22. Ahvazi, M.; Jamzad, Z.; Balali, G.R.; Saeidi, H. Trichome micro-morphology in Marrubium L. (Lamiaceae) in Iran and the role of environmental factors on their variation. Iran. J. Bot. 2016, 22, 39–58. [Google Scholar] [CrossRef]
  23. Ahvazi, M.; Balali, G.R.; Jamzad, Z.; Saeidi, H. A taxonomical, morphological and pharmacological review of Marrubium vulgare L., an old medicinal plant in Iran. J. Med. Plants 2018, 17, 7–24. [Google Scholar]
  24. Martin, E.; Cetin, O.; Akgul, G.; Hilal, A.Y. New chromosome numbers in the genus Marrubium (horehound) from Turkey. Biol. Divers. Conserv. 2011, 4, 185–188. [Google Scholar]
  25. Buyukkartal, H.N.; Colgecen, H.; Akgul, G. Comparative leaf, stem and root anatomies of taxa Marrubium bourgaei and Marrubium heterodon (Lamiaceae). Aust. J. Crop Sci. 2016, 10, 1516–1522. [Google Scholar] [CrossRef]
  26. Marzouk, R.I.; El-Darier, S.M.; Nour, I.H.; Kamal, S.A. Numerical taxonomic study of Marrubium L. (Lamiaceae) in Egypt. Catrina 2015, 13, 25–35. [Google Scholar]
  27. Sgarbossa, J.; Schmidt, D.; Schwerz, F.; Schwerz, L.; Prochnow, D.; Caron, B.O. Effect of season and irrigation on the chemical composition of Aloysia triphylla essential oil. Rev. Ceres 2019, 66, 85–93. [Google Scholar] [CrossRef] [Green Version]
  28. Castelo, A.V.M.; Del Menezzi, C.H.S.; Resck, I.S. Seasonal variation in the yield and the chemical composition of essential oils from two Brazilian native arbustive species. J. Appl. Sci. 2012, 12, 753–760. [Google Scholar] [CrossRef]
  29. Yavari, A.; Nazeri, V.; Sefidkon, F.; Hassani, M.E. Influence of some environmental factors on the essential oil variability of Thymus migricus. Nat. Prod. Comm 2010, 5, 943–948. [Google Scholar] [CrossRef] [Green Version]
  30. Zouari-Bouassida, K.; Trigui, M.; Makni, S.; Jlaiel, L.; Tounsi, S. Seasonal variation in essential oils composition and the biological and pharmaceutical protective effects of Mentha longifolia leaves grown in Tunisia. Biomed. Res. Int. 2018, 7856517. [Google Scholar] [CrossRef] [Green Version]
  31. Gosztola, B.; Sarosi, S.; Nemeth, E. Variability of the essential oil content and composition of chamomile (Matricaria recutita) affected by weather conditions. Nat. Prod. Commun. 2010, 5, 465–470. [Google Scholar] [CrossRef] [Green Version]
  32. Acimovic, M.; Cvetkovic, M.; Stankovic, J. Effect of weather conditions, location and fertilization on coriander fruit essential oil quality. Essent Oil Bear Plants 2016, 19, 1208–1215. [Google Scholar]
  33. Aziz, E.E.; Badawy, E.M.; Zheljazkov, V.D.; Nicola, S.M.; Fouad, H. Yield and chemical composition of essential oil of Achillea millefolium L. as affected by harvest time. Egypt J. Chem. 2019, 62, 533–540. [Google Scholar] [CrossRef]
  34. Hamdaoui, B.; Wannes, W.A.; Marrakchi, M.; Brahim, N.B.; Marzouk, B. Essential oil composition of two Tunisian horehound species: Marrubium vulgare L. and Marrubium aschersonii Magnus. J. Essent Oil Bear Plants 2013, 16, 608–612. [Google Scholar] [CrossRef]
  35. Laouer, H.; Yabrir, B.; Djeridane, A.; Yousfi, M.; Beldovini, N.; Lamamra, M. Composition, antioxidant and antimicrobial activities of the essential oil of Marrubium deserti. Nat. Prod. Commun. 2009, 4, 1133–1138. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  36. Nagy, M.; Svajdlenka, E. Comparison of essential oils from Marrubium vulgare L. and M. peregrinum L. J. Essent Oil Res. 1998, 10, 585–587. [Google Scholar] [CrossRef]
  37. Kırımer, N.; Kurkcuoglu, M.; Akgul, G.; Baser, K.H.H.; Mahmoud, A.A. Composition of the essential oil of Marrubium anisodon C. Koch of Turkish origin. Rec Nat. Prod. 2015, 9, 234–236. [Google Scholar]
  38. Teimori, M.; Khavari-Nejad, R.A.; Yassa, N.; Nejadsatari, T. Analysis of the essential oil of Marrubium crassidens Bioos. and M. astracanicum Jacq. J. Appl. Sci. 2008, 8, 1793–1795. [Google Scholar] [CrossRef] [Green Version]
  39. Hamedeyazdan, S.; Asnaashari, S.; Fathiazad, F. Characterization of non-terpenoids in Marrubium crassidens Bioss. essential oil. Adv. Pharm. Bull. 2013, 3, 429–432. [Google Scholar]
  40. Golmakani, H.; Rabbani Nasab, H.; Sharifan, M.; Kamali, H.; Yadollahi, A. The essential oil composition and antibacterial activity of Marrubium dubanese Murata from North Khorassan Province, Iran. J. Essent Oil Bear Plants 2016, 19, 963–971. [Google Scholar] [CrossRef]
  41. Hamedeyazdan, S.; Zarei, M.; Salem, A.; Asnaashari, S.; Fathiazad, F. Essential oil of two Iranian horehound species: Marrubium propinquum and Marrubium parviflorum. Pharm. Sci. 2017, 23, 143–149. [Google Scholar] [CrossRef] [Green Version]
  42. Bal, Y.; Kaban, S.; Kirimer, N.; Baser, K.H.C. Composition of the essential oil of Marrubium parviflorum Fisch. et Mey. subsp. oligodon (Boiss.) Seybold. J. Essent Oil Res. 1999, 11, 300–302. [Google Scholar] [CrossRef]
  43. Kaurinovic, B.; Vlaisavljevic, S.; Popovic, M.; Vastag, D.; Djurendic-Brensel, M. Antioxidant properties of Marrubium peregrinum L. (Lamiaceae) essential oil. Molecules 2010, 15, 5943–5955. [Google Scholar] [CrossRef] [Green Version]
  44. Lazari, D.; Skaltsa, H.; Constantinidis, T. Essential oils of Marrubium velutinum Sm. and Marrubium peregrinum L., growing wild in Greece. Flavour Fragr. J. 1999, 14, 290–292. [Google Scholar] [CrossRef]
  45. Hamedeyazdan, S.; Fathiazad, F.; Asnaashari, S. Chemical composition of the essential oil from Marrubium persicum C.A. Mey. (Lamiaceae). Pharm. Sci. 2013, 19, 35–38. [Google Scholar]
  46. Nik, B.Z.; Mirza, M. Composition of the essential oil of Marrubium astracanicum Jacq. J. Essent Oil Res. 2003, 15, 342–343. [Google Scholar] [CrossRef]
  47. Khanavi, M.; Ghasemian, L.; Motlagh, E.H.; Hadjikhoondi, A.; Shafiee, A. Chemical composition of the essential oils of Marrubium parviflorum Fisch. & C.A. Mey. and Marrubium vulgare L. from Iran. Flavour Fragr. J. 2005, 20, 324–326. [Google Scholar]
  48. Sarikurkcu, C.; Ozer, M.S.; Calli, N.; Popović-Djorđević, J. Essential oil composition and antioxidant activity of endemic Marrubium parviflorum subsp oligodon. Ind. Crops Prod. 2018, 119, 209–213. [Google Scholar] [CrossRef]
  49. Miloudi, K.; Hamimed, A.; Benmimoun, Y.; Bellebna, Y.; Taibi, A.; Tilmatine, A. Intensification of essential oil extraction of the Marrubium vulgare using pulsed electric field. J. Essent Oil Bear Plants 2018, 21, 811–824. [Google Scholar] [CrossRef]
  50. Zawislak, G. Comparison of chemical composition of the essential oil from Marrubium vulgare L. and M. incanum Desr. during the second year of cultivation. Acta Agrobot. 2015, 68, 59–62. [Google Scholar] [CrossRef]
  51. Petrovic, S.; Pavlovic, M.; Maksimovic, Z.; Milenkovic, M.; Couladis, M.; Tzakou, O.; Niketic, M. Composition and antimicrobial activity of Marrubium incanum Desr. (Lamiaceae) essential oil. Nat. Prod. Commun. 2009, 4, 431–434. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  52. Chemsa, A.E.; Zellagui, A.; Ozturk, M.; Erol, E.; Ceylan, O.; Duru, M.E.; Gherraf, N. Antibiofilm formation, antioxidant and anticholinesterase activities of essential oil and methanol extract of Marrubium deserti de Noe. J. Mater. Env. Sci. 2016, 7, 993–1000. [Google Scholar]
  53. Mohammadhosseini, M. First report of screening of the profiles of the essential oils and volatiles from the aerial parts of Marrubium persicum using classical and advanced methods prior to Gas Chromatographic Mass Spectrometric determination. J. Med. Plants Prod. 2016, 2, 169–180. [Google Scholar]
  54. Agryropoulou, C.; Skaltsa, H. Identification of essential oil components of Marrubium thessalum Boiss. & Heldr., growing wild in Greece. Nat. Prod. Res. 2012, 26, 1–7. [Google Scholar]
  55. Aćimović, M.; Pezo, L.; Tešević, V.; Čabarkapa, I.; Todosijević, M. QSRR Model for predicting retention indices of Satureja kitaibelii Wierzb. Ex Heuff. essential oil composition. Ind. Crops Prod. 2020, 154, 112752. [Google Scholar] [CrossRef]
  56. Nekoei, M.; Salimi, M.; Dolatabadi, M.; Mohammadhosseini, M. Prediction of antileukemia activity of berbamine derivatives by genetic algorithm–multiple linear regression. Mon. Chem. 2011, 142, 943. [Google Scholar] [CrossRef]
  57. Nekoei, M.; Mohammadhosseini, M.; Pourbasheer, E. QSAR study of VEGFR-2 inhibitors by using genetic algorithm-multiple linear regressions (GA-MLR) and genetic algorithm-support vector machine (GA-SVM): A comparative approach. Med. Chem. Res. 2015, 24, 3037–3046. [Google Scholar] [CrossRef]
  58. Azar, P.A.; Nekoei, M.; Riahi, S.; Ganjali, M.R.; Zare, K. A quantitative structure-retention relationship for the prediction of retention indices of the essential oils of Ammoides atlantica. J. Serb. Chem Soc. 2011, 76, 891–902. [Google Scholar] [CrossRef]
  59. Todeschini, R.; Consonni, V. Handbook of Molecular Descriptors, Methods and Principles in Medicinal Chemistry; Wiley-VCH Verlag GmbH: Weinheim, Germany, 2000. [Google Scholar]
  60. Bakić, V.V.; Pezo, M.L.; Jovanović, M.P.; Turanjanin, V.M.; Vučićević, B.S.; Mirkov, N.S. Technical analysis of photovoltaic/wind systems with hydrogen storage. Science 2012, 16, 865–875. [Google Scholar]
  61. Arsenović, M.; Pezo, L.; Stanković, S.; Radojević, Z. Factor space differentiation of brick clays according to mineral content: Prediction of final brick product quality. Appl. Clay. Sci. 2015, 115, 108–114. [Google Scholar] [CrossRef]
  62. Boulila, A.; Sanaa, A.; Salem, I.B.; Rokbeni, N.; Mrabet, Y.; Hosni, K.; Fernandez, X. Antioxidant properties and phenolic variation in wild populations of Marrubium vulgare L. (Lamiaceae). Ind. Crops Prod. 2015, 76, 616–622. [Google Scholar] [CrossRef]
  63. Mahdavi, A.; Moradi, P.; Mastinu, A. Variation in terpene profiles of Thymus vulgaris in water deficit stress response. Molecules 2020, 25, 1091. [Google Scholar] [CrossRef] [PubMed] [Green Version]
  64. Noorizadeh, H.; Farmany, A.; Noorizadeh, M. Quantitative structure-retention relationships analysis of retention index of essential oils. Química Nova 2011, 34, 242–249. [Google Scholar] [CrossRef]
  65. Driouche, Y.; Messadi, D. Quantitative structure-retention relationship model for predicting retention indices of constituents of essential oils of Thymus vulgaris (Lamiaceae) (Short communication). J. Serb. Chem Soc. 2019, 84, 405–416. [Google Scholar] [CrossRef] [Green Version]
  66. Acimovic, M.; Pezo, L.; Stankovic Jeremic, J.; Cvetkovic, M.; Rat, M.; Cabarkapa, I.; Tesevic, V. QSRR model for predicting retention indices of geraniol chemotype of Thymus serpyllum essential oil. J. Essent Oil Bear Plants 2020, 23, 464–473. [Google Scholar] [CrossRef]
  67. Pavlić, B.; Teslić, N.; Kojić, P.; Pezo, L. Prediction of the GC-MS retention time for terpenoids detected in sage (Salvia officinalis L.) essential oil using QSRR approach. J. Serb. Chem. Soc. 2020, 85, 9–23. [Google Scholar] [CrossRef] [Green Version]
  68. Noorizadeh, H. Linear and nonlinear quantitative structure linear retention indices relationship models for essential oils. Eurasian J. Anal. Chem 2013, 8, 50–63. [Google Scholar]
  69. Héberger, K. Quantitative structure–(chromatographic) retention relationships. J. Chromatogr A 2007, 1158, 273–305. [Google Scholar] [CrossRef] [Green Version]
  70. Kaliszan, R.; Ba̧czek, T.; Buciński, A.; Buszewski, B.; Sztupecka, M. Prediction of gradient retention from the linear solvent strength (LSS) model, quantitative structure-retention relationships (QSRR), and artificial neural networks (ANN). J. Sep. Sci 2003, 26, 271–282. [Google Scholar] [CrossRef]
  71. Khodadoust, S.; Ghaedi, M.; Hadjmohammadi, M.R. Dispersive nano solid material-ultrasound assisted microextraction as a novel method for extraction and determination of bendiocarb and promecarb: Response surface methodology. Talanta 2013, 116, 637–646. [Google Scholar] [CrossRef]
  72. Wolfender, J.L.; Martia, G.; Thomas, A.; Bertranda, S. Current approaches and challenges for the metabolite profiling of complex natural extracts. J. Chromatogr. A 2015, 1382, 136–164. [Google Scholar] [CrossRef] [PubMed]
  73. Zisi, C.; Sampsonidis, I.; Fasoula, S.; Papachristos, K.; Witting, M.; Gika, H.G.; Nikitas, P.; Pappa-Louisi, A. QSRR modeling for metabolite standards analyzed by two different chromatographic columns using multiple linear regression. Metabolites 2017, 7, 7. [Google Scholar] [CrossRef] [Green Version]
  74. Deconinck, E.; Zhang, M.H.; Petitet, F.; Dubus, E.; Ijjaali, I.; Coomans, D.; Vander Heydena, Y. Boosted regression trees, multivariate adaptive regression splines and their two-step combinations with multiple linear regression or partial least squares to predict blood–brain barrier passage: A case study. Anal. Chim. Acta 2008, 609, 13–23. [Google Scholar] [CrossRef]
  75. Yap, C.W. PaDEL-descriptor: An open source software to calculate molecular descriptors and fingerprints. J. Comput. Chem. 2011, 32, 1446–1474. [Google Scholar] [CrossRef]
  76. Goldberg, D.E. Genetic Algorithms in Search, Optimisation and Machine Learning; Addison-Wesley: Boston, MA, USA, 1989. [Google Scholar]
  77. Gramatica, P. Principles of QSAR models validation: Internal and external. Qsar Comb. Sci 2007, 26, 694–701. [Google Scholar] [CrossRef]
  78. Statistica 10 Software (StatSoft, Inc. STATISTICA, ver. 10, Data Analysis Software System). Available online: https://www.statsoft.de/en/software/tibco-statisticatm (accessed on 15 December 2018).
  79. Foroughia, M.; Hossein, M.; Azqhandic, A.; Kakhki, S. Bio-inspired, high, and fast adsorption of tetracycline from aqueous media using Fe3O4-g-CN@PEI-β-CD nanocomposite: Modeling by response surface methodology (RSM), boosted regression tree (BRT), and general regression neural network (GRNN). J. Hazard. Mater. 2020, 388, 121769. [Google Scholar] [CrossRef]
  80. Cheong, Y.L.; Leitão, P.J.; Lakes, T. Assessment of land use factors associated with dengue cases in Malaysia using Boosted Regression Trees. Spat. Spatiotemporal Epidemiol. 2014, 10, 75–84. [Google Scholar] [CrossRef] [PubMed]
  81. Froeschke, J.T.; Froeschke, B.F. Spatio-temporal predictive model based on environmental factors for juvenile spotted seatrout in Texas estuaries using boosted regression trees. Fish. Res. 2011, 111, 131–138. [Google Scholar] [CrossRef]
  82. Sut, N.; Simsek, O. Comparison of regression tree data mining methods for prediction of mortality in head injury. Expert Syst. Appl. 2011, 38, 15534–15539. [Google Scholar] [CrossRef]
  83. Mazaheri, H.; Ghaedi, M.; Azqhandi, M.A.; Asfaram, A. Application of machine/ statistical learning, artificial intelligence and statistical experimental design for the modeling and optimization of methylene blue and Cd (II) removal from a binary aqueous solution by natural walnut carbon. Phys. Chem. Chem. Phys. 2017, 19, 11299–11317. [Google Scholar] [CrossRef]
  84. Salonen, J.S.; Luoto, M.; Alenius, T.; Heikkilä, M.; Seppä, H.; Telford, R.J.; Birks, H.J. Reconstructing palaeoclimatic variables from fossil pollen using boosted regression trees: Comparison and synthesis with other quantitative reconstruction methods. Quat Sci. Rev. 2014, 88, 69–81. [Google Scholar] [CrossRef]
Plants 10 00600 g001 550
Figure 1. Retention indices (RIs) of the M. vulgare volatiles composition, from experimentally obtained GC–MS data on a HP-5MS column (RIa) and NIST database (RIb).
Figure 1. Retention indices (RIs) of the M. vulgare volatiles composition, from experimentally obtained GC–MS data on a HP-5MS column (RIa) and NIST database (RIb).
Plants 10 00600 g001
Plants 10 00600 g002 550
Figure 2. Comparison of experimentally obtained RIs on a HP-5MS column (RIa) with BRT pre-dicted values (RIpred.) in 2019 (a) and 2020 (b).
Figure 2. Comparison of experimentally obtained RIs on a HP-5MS column (RIa) with BRT pre-dicted values (RIpred.) in 2019 (a) and 2020 (b).
Plants 10 00600 g002
Plants 10 00600 g003 550
Figure 3. Predictor importance of the molecular descriptors on RI in 2019 (a) and 2020 (b).
Figure 3. Predictor importance of the molecular descriptors on RI in 2019 (a) and 2020 (b).
Plants 10 00600 g003
Plants 10 00600 g004 550
Figure 4. Unrooted cluster tree for different Marrubium samples.
Figure 4. Unrooted cluster tree for different Marrubium samples.
Plants 10 00600 g004
Table
Table 1. Chemical composition of Marrubium vulgare during two years (2019 and 2020).
Table 1. Chemical composition of Marrubium vulgare during two years (2019 and 2020).
NoCompound/ClassCycleRIpred.20192020Reference
RIa%RIa%
12E-Hexenal OTrain892.915--8470.2M. aschersonii [34], M. deserti [35], M. peregrinum [36], M. vulgare [10,12,15,16,34]
2Furan, 2,5-diethyltetrahydro OValidation853.684--8970.1
31-Octen-3-ol OValidation965.8189760.29740.6M. anisodon [37], M. astracanicum [38], M. crassidens [39], M. deserti [35], M. duabense [40], M. parviflorum [41,42], M. peregrinum [43,44], M. persicum [45], M. propinquum [41], M. velutinum [44], M. vulgare [7,8,10,15]
42-Pentyl furan OTrain1059.803--9890.1
53-Octanol OTest962.233--9920.1M. anisodon [37], M. astracanicum [46], M. duabense [40], M. peregrinum [36,44], M. velutinum [44]
6Linalool OMNTrain1106.04111020.110980.1M. aschersonii [34], M. astracanicum [46], M. parviflorum [41,42,47,48], M. peregrinum [36,43,44], M. persicum [45], M. velutinum [44], M. vulgare [8,10,12,17,34,36,47,49]
7n-Nonanal OTrain1078.484--11020.1M. aschersonii [34], M. deserti [35], M. duabense [40], M. peregrinum [43,44], M. persicum [45], M. velutinum [44], M. vulgare [34]
8E-Thujone OMNTrain1118.307--11140.1M. peregrinum [43], M. vulgare [8,15]
9NI-1----11320.1-
10Geijerene OTrain1192.30111430.111390.6M. incanum [50,51], M. parviflorum [42,47], M. peregrinum [43], M. vulgare [50]
112E-Nonen-1-al OValidation1097.602--11560.1
12β-Cyclocitral OTrain1216.889--12190.1M. peregrinum [44], M. velutinum [44], M. vulgare [10]
13Cogeijerene OTrain1203.235--12830.1
14Pregeijerene OTrain1149.85712900.112870.2M. astracanicum [38], M. crassidens [38], M. parviflorum [42,47], M. peregrinum [43]
15Thymol ARTest1209.01712920.3--M. deserti [52], M. vulgare [7,8,10,15,50]
162-Undecanone OTrain1269.22812950.11292TraceM. vulgare [15]
17Carvacrol ARValidation1179.07213020.1--M. duabense [40], M. incanum [50], M. parviflorum [42], M. peregrinum [43], M. vulgare [7,8,10,49,50]
18δ-Elemene STTest1512.436--13360.1M. anisodon [37], M. astracanicum [38], M. crassidens [38], M. deserti [35,40], M. duabense [40], M. incanum [50,51], M. parviflorum [47], M. peregrinum [44], M. persicum [53], M. thessalum [54], M. velutinum [44], M. vulgare [47,50]
19α-Cubebene STTrain1491.202--13480.1M. astracanicum [38], M. crassidens [39], M. deserti [35,40], M. duabense [40], M. parviflorum [47], M. peregrinum [44], M. persicum [45], M. vulgare [8,47]
20Eugenol ARTrain1372.624--13570.4M. aschersonii [34], M. peregrinum [36,43,44], M. persicum [53], M. velutinum [44], M. vulgare [10,12,34,36,47]
21α-Copaene STTrain1475.87813773.313776.1M. anisodon [37], M. aschersonii [34], M. astracanicum [38], M. crassidens [38], M. deserti [35], M. duabense [40], M. incanum [50,51], M. parviflorum [42,47,48], M. peregrinum [36,43,44], M. persicum [53], M. thessalum [54], M. velutinum [44], M. vulgare [8,9,10,11,12,13,36,47,50]
22β-Bourbonene STTrain1487.97613850.813841.2M. anisodon [37], M. astracanicum [38], M. crassidens [38], M. deserti [35,52], M. incanum [50,51], M. parviflorum [41,42,47,48], M. peregrinum [43,44], M. persicum [45,53], M. thessalum [54], M. velutinum [44], M. vulgare [9,10,13,50]
23NI-2----13880.1-
24β-Cubebene STTest1475.61013900.113890.2M. aschersonii [34], M. deserti [35], M. peregrinum [43,44], M. parviflorum [42], M. velutinum [44], M. vulgare [12,13,34,47]
25β-Elemene STTrain1475.50613920.413911.0M. anisodon [37], M. astracanicum [38], M. crassidens [38], M. deserti [35,52], M. duabense [40], M. incanum [50,51], M. parviflorum [42,47], M. peregrinum [44], M. persicum [53], M. thessalum [54], M. velutinum [44], M. vulgare [47,50]
26Z-Caryophyllene STTrain1463.16114070.114060.2
27α-Z-Bergamotene STTrain1428.21514160.2--
28E-Caryophyllene STValidation1463.161142224.6142323.0M. anisodon [37], M. aschersonii [34], M. astracanicum [38,46], M. crassidens [38,39], M. deserti [35,52], M. duabense [40], M. incanum [50,51], M. parviflorum [41,42,47,48], M. peregrinum [36,43,44], M. persicum [45,53], M. propinquum [41], M. thessalum [54], M. velutinum [44], M. vulgare [7,8,9,10,11,12,13,16,17,34,36,47,49,50]
29β-Copaene STTest1459.62314300.414301.3M. incanum [50], M. vulgare [50]
30α-E-Bergamotene STTrain1428.21514360.114350.1M. anisodon [37], M. astracanicum [46], M. crassidens [38], M. parviflorum [42,47], M. peregrinum [44], M. velutinum [44], M. vulgare [47]
31NI-3--14450.214440.6-
32α-Humulene STValidation1503.99914545.214555.3M. anisodon [37], M. aschersonii [34], M. astracanicum [38,46], M. crassidens [38,39], M. duabense [40], M. incanum [50,51], M. parviflorum [42,47], M. peregrinum [36,43], M. persicum [45], M. thessalum [54], M. velutinum [44], M. vulgare [8,9,10,12,13,15,34,36,47,50]
33Sesquisabinene STTrain1442.740--14570.9
34E-β-Farnesene STTest1431.41914571.3--M. anisodon [37], M. aschersonii [34], M. crassidens [39], M. parviflorum [41,42,47], M. peregrinum [43,44], M. persicum [45], M. propinquum [41], M. thessalum [54], M. velutinum [44], M. vulgare [8,10,12,16,17,34,36,47]
35C16H34 ATrain1573.43614621.514620.2
36NI-4---Trace-0.2-
37Z-Muurola-4(14),5-diene STTrain1482.433--14660.1
38NI-5--14690.1---
39NI-6--14720.1---
40E-Cadina-1(6),4-diene STTrain1481.465--1475TraceM. vulgare [15]
41γ-Muurolene STTest1450.20314790.1--M. incanum [50], M. peregrinum [43,44], M. parviflorum [42], M. velutinum [44]
42Germacrene D STTest1450.18814839.6148717.0M. anisodon [37], M. aschersonii [34], M. astracanicum [38], M. crassidens [38,39], M. deserti [35,52], M. incanum [50,51], M. parviflorum [41,42,47,48], M. peregrinum [36,43,44], M. persicum [45,53], M. propinquum [41], M. thessalum [54], M. velutinum [44], M. vulgare [9,10,11,12,13,15,16,17,34,36,47,50]
43E-β-Ionone OTest1471.73514860.41489TraceM. anisodon [37], M. aschersonii [34], M. duabense [40], M. incanum [51], M. parviflorum [42], M. peregrinum [43,44], M. thessalum [54], M. vulgare [12]
44NI-7----14890.1-
45epi-Cubebol OSTTrain1622.285--14950.2
46Viridiflorene STValidation1507.44714970.1--
47Bicyclogermacrene STValidation1493.69714980.214980.2M. astracanicum [38], M. crassidens [38,39], M. deserti [35,52], M. duabense [40], M. incanum [50,51], M. parviflorum [41,42,47,48], M. peregrinum [36,43,44], M. persicum [45], M. propinquum [41], M. thessalum [54], M. velutinum [44], M. vulgare [10,11,17,50]
48NI-8----14990.7-
49Pentadecane ATest1486.88415000.21500Trace
50α-Muurolene STTrain1465.65015010.115010.2M. aschersonii [34], M. deserti [35,52], M. incanum [51], M. peregrinum [43], M. velutinum [44], M. vulgare [12,13,34]
51Germacrene A STTrain1450.18815080.115060.1M. incanum [50], M. parviflorum [47,48]
52β-Bisabolene STValidation1425.13915110.215070.2M. anisodon [37], M. aschersonii [34], M. crassidens [38], M. parviflorum [47], M. peregrinum [44], M. persicum [45], M. propinquum [41], M. thessalum [54], M. velutinum [44], M. vulgare [11,12,13,17,34,47,49]
53γ-Cadinene STTest1450.20315130.215150.4M. deserti [52], M. incanum [50], M. parviflorum [47,48], M. peregrinum [43,44], M. persicum [53], M. velutinum [44], M. vulgare [7,10,15,47]
54δ-Cadinene STTest1475.07015234.715289.7M. deserti [52], M. incanum [50], M. parviflorum [42,47], M. peregrinum [43,44], M. persicum [53], M. velutinum [44], M. vulgare [7,10,15,47]
55E-Cadina-1,4-diene STTrain1471.52115330.115330.1M. vulgare [15]
56α-Cadinene STTrain1465.650--15370.1M. peregrinum [43,44], M. velutinum [44], M. vulgare [47]
57α-Calacorene STTrain1540.123--15430.1M. deserti [52], M. vulgare [12,15]
58NI-9--15550.215520.2-
59E-Nerolidol OSTValidation1567.13615613.515641.5M. anisodon [37], M. deserti [52], M. parviflorum [42], M. peregrinum [43,44], M. thessalum [54], M. velutinum [44], M. vulgare [9,36]
60NI-10----15710.1-
61NI-11--15770.215750.9-
62NI-12----15820.3-
63Caryophyllene oxide OSTTest1636.61215801.015831.8M. anisodon [37], M. astracanicum [46], M. crassidens [38,39], M. deserti [52], M. duabense [40], M. incanum [50,51], M. parviflorum [41,42,47,48], M. peregrinum [36,43], M. persicum [45,53], M. propinquum [41], M. thessalum [54], M. velutinum [44], M. vulgare [8,9,10,12,36,47,50]
64NI-13----15870.1-
65Viridiflorol OSTValidation1573.43615970.1--M. aschersonii [34], M. astracanicum [38], M. crassidens [38], M. incanum [51], M. parviflorum [47], M. peregrinum [43], M. vulgare [10,12,34,47]
66Hexadecane ATrain1594.57616020.1--M. duabense [40], M. velutinum [44]
67Humulene epoxide II OSTTrain1626.95916070.216070.2M. anisodon [37], M. incanum [51], M. thessalum [54], M. vulgare [10]
68Muurola-4,10(14)-dien-1-β-ol OSTTrain1605.330--16270.3
69NI-14--16280.1---
704,4-dimethyl-Tetracyclo [6.3.2.0(2,5).0(1,8)]tridecan-9-ol OValidation1605.030--16310.2
71NI-15--16320.1---
72Caryophylla-4(12),8(13)-dien-5-α-ol OSTTrain1605.03016360.116350.3
73epi-α-Muurolol (=tau-muurolol) OSTTest1605.03016420.216410.6M. astracanicum [38], M. deserti [35], M. incanum [51], M. parviflorum [42], M. peregrinum [44], M. velutinum [44]
74α-Muurolol (=Torreyol) OSTTrain1652.148--16450.1
75α-Cadinol OSTTrain1682.93416540.316540.9M. crassidens [38], M. deserti [35,52], M. incanum [50,51], M. parviflorum [42], M. persicum [45], M. vulgare [12,50]
76NI-16--16580.216560.2-
77NI-17--16620.116620.1-
78E-Calamenen-10-ol OSTTrain1608.844--16690.1
79NI-18--16680.2---
80NI-19----16700.2-
818-Heptadecene OTrain1607.164--16730.2
821-Tetradecanol OTrain1702.77116750.1--
83Germacra-4(15),5,10(14)-trien-1-α-ol OSTTrain1700.00316820.116850.2
84Heptadecane AValidation1726.88616960.316960.2M. anisodon [37], M. parviflorum [42,47], M. vulgare [10,47]
85Pentadecanal OValidation1581.92817100.117110.1M. anisodon [37]
86Mint sulfide STTrain1778.77717330.117360.1
87NI-20--17340.1---
88NI-21--17420.1---
89NI-22--17430.417440.1-
90E-3-Octadecene OTrain1722.391--17770.1
91n-Pentadecanol OTrain1787.02217780.1--M. parviflorum [42]
92NI-23----17820.1-
93Octadecane AValidation1950.09317960.1--M. parviflorum [47], M. peregrinum [43], M. vulgare [47]
94NI-24--18190.1---
956,10,14-trimethyl-2-Pentadecanone OTrain1915.81818444.818420.5M. peregrinum [44], M. velutinum [44], M. vulgare [10]
96NI-25--18490.1---
97NI-26--18530.2---
98NI-27--18880.1---
99NI-28--18910.818910.1-
100Nonadecane ATest1869.34618970.218970.2M. duabense [40], M. parviflorum [47], M. peregrinum [43], M. vulgare [10,15,47]
101NI-29--19040.11906Trace-
1025E,9E-Farnesyl acetone OSTTrain1956.28919160.31915TraceM. thessalum [54], M. vulgare [15]
103NI-30--1918Trace1917Trace-
104NI-31--19240.1---
105NI-32----1926Trace-
106NI-33--19250.1---
107NI-34--19290.1---
108NI 19380.11940Trace
109Hexadecanoic acid OValidation1995.49119603.9--M. parviflorum [42], M. peregrinum [36], M. vulgare [36,47]
110NI-35--19730.11974Trace-
111Eicosane ATrain2034.56019970.219940.1M. parviflorum [48]
112NI-36--20010.1---
113E,E-Geranyl linalool ODTrain2028.64520271.6--M. aschersonii [34], M. parviflorum [42], M. vulgare [12,34]
1143,7,11,15-tetramethyl-(E,E)-1,6,10,14-Hexadecatetraen-3-ol OD --20280.9
115Manool ODTrain2064.19620570.3--
116NI-37--20610.1---
117NI-38--20670.1---
118NI-39--20840.1---
119NI-40--20960.1---
120Heneicosane ATrain2120.28421011.621001.3M. parviflorum [42,47], M. peregrinum [43], M. propinquum [41], M. vulgare [10]
121NI-41--21080.321050.2-
122NI-42--21120.221100.3-
123Phytol ODTest2124.81821161.421130.4M. anisodon [37], M. incanum [51], M. parviflorum [41,42], M. peregrinum [36], M. propinquum [41], M. vulgare [10,15]
124NI-43--21310.2---
125NI-44--21430.221430.1-
126NI-45--21470.2---
127NI-46--21640.221600.2-
128NI-47--21670.121720.3-
129NI-48--21750.621760.4-
130NI-49--21810.921790.2-
131NI-50--21830.4---
132NI-51--21982.421952.4-
133Docosane AValidation2194.42122050.921980.6M. crassidens [39], M. parviflorum [47]
134NI-52----22010.1-
135NI-53----22090.3-
136NI-54--22150.3---
137NI-55--22250.322210.1-
138NI-56--22460.3---
139NI-57--22580.222530.3-
140NI-58--22700.122650.1-
141NI-59--22770.222740.2-
142NI-60--22933.822881.7-
143Tricontane ATrain2381.64223053.623022.6
144NI-61--23090.223050.2-
145NI-62--23440.223410.3-
146NI-63--23800.123770.1-
147NI-64--23830.123820.1-
148Tetracosane ATrain2493.49124010.323950.2M. deserti [52], M. parviflorum [41,42,47], M. propinquum [41]
149NI-65--24540.424470.2-
150NI-66--24880.224830.2-
151Pentacosane ATest2510.08725030.824970.6M. anisodon [37], M. parviflorum [42,47]
152Heptacosane ATrain2730.53727020.626960.5M. anisodon [37], M. aschersonii [34], M. incanum [51], M. parviflorum [42], M. vulgare [34]
153NI-67----27660.1-
154Octacosane A 28010.12791TraceM. crassidens [39], M. parviflorum [42], M. persicum [45]
155Squalene TTrain2870.67328350.128230.1
156NI-68--28680.128550.1-
157Nonacosane ATest2930.73229050.728920.6M. anisodon [37], M. crassidens [39], M. persicum [45]
158Untriacontane AValidation3150.67331050.430950.3
159NI-69----32120.1-
160Tritriacontane ATrain3319.75333000.13301Trace
Oxygenated monoterpenes OMN 0.1 0.2
Sesquiterpene hydrocarbons ST 52.0 67.8
Oxygenated sesquiterpenes OST 5.8 6.2
Oxygenated diterpenes OD 3.3 1.3
Triterpene T 0.1 0.1
Aromatics AR 0.4 0.4
Alkanes A 11.7 7.4
Other O 9.9 3.4
NI 16.7 12.5
Total 100 99.3
RIpred.—BRT calculated retention index; RIa—retention index experimentally obtained on a HP-5MS column; Other—aliphatic hydrocarbons, aliphatic aldehydes and alcohols, aliphatic acids, their esters and aldehydes, aromatic esters with aliphatic acids, alkyl-aromatic alcohols, or aryl esters of aromatic acids; NI—not identified compound.
Table
Table 2. The correlation coefficient matrix for the selected descriptors by GA.
Table 2. The correlation coefficient matrix for the selected descriptors by GA.
AATSC2pMATS5vGATS6vSM1_DzvVR1_DzsVAdjMat
AATSC4e0.031−0.138−0.1350.0300.2050.224
AATSC2p −0.265−0.1310.036−0.231−0.010
MATS5v 0.212−0.0080.0660.109
GATS6v 0.0720.1310.214
SM1 Dzv 0.0580.084
VR1 Dzs 2.339
Table
Table 3. The “goodness of fit” tests for the developed BRT model.
Table 3. The “goodness of fit” tests for the developed BRT model.
Boosted Tree Modelχ2RMSEMBEMPEr2
20194455.27266.160−13.0633.2850.956
20203975.75162.581−7.6983.2410.964
χ2—reduced chi-square, MBE—mean bias error, RMSE—root mean square error, MPE—mean percentage error.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Share and Cite

MDPI and ACS Style

Aćimović, M.; Ivanović, S.; Simić, K.; Pezo, L.; Zeremski, T.; Ovuka, J.; Sikora, V. Chemical Characterization of Marrubium vulgare Volatiles from Serbia. Plants 2021, 10, 600. https://doi.org/10.3390/plants10030600

AMA Style

Aćimović M, Ivanović S, Simić K, Pezo L, Zeremski T, Ovuka J, Sikora V. Chemical Characterization of Marrubium vulgare Volatiles from Serbia. Plants. 2021; 10(3):600. https://doi.org/10.3390/plants10030600

Chicago/Turabian Style

Aćimović, Milica, Stefan Ivanović, Katarina Simić, Lato Pezo, Tijana Zeremski, Jelena Ovuka, and Vladimir Sikora. 2021. "Chemical Characterization of Marrubium vulgare Volatiles from Serbia" Plants 10, no. 3: 600. https://doi.org/10.3390/plants10030600

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop