Nikolic et al. 2021, Biologica Nyssana 12(1) 12 (1) September 2021: 23-32 DOI: 10.5281/zenodo.5522967 Variability of headspace volatiles in native population of Abies x borisii-regis from the central Rhodopes Original Article Jelena S. Nikolić Department of Biology and Ecology, Faculty of Sciences and Mathematics, University of Niš, Višegradska 33, Niš, Serbia jelenanikolic9311@gmail.com (corresponding author) Snežana Č. Jovanović Department of Chemistry, Faculty of Sciences and Mathematics, University of Niš, Višegradska 33, Niš, Serbia coka.sslagalica@gmail.com Bojan Zlatković Department of Biology and Ecology, Faculty of Sciences and Mathematics, University of Niš, Višegradska 33, Niš, Serbia bojanzlat@yahoo.com Gordana S. Stojanović Department of Chemistry, Faculty of Sciences and Mathematics, University of Niš, Višegradska 33, Niš, Serbia gordanastojanovic2411960@gmail.co Zorica Mitić Department of Biology and Ecology, Faculty of Sciences and Mathematics, University of Niš, Višegradska 33, Niš, Serbia saraczorica@gmail.com Received: March 31, 2021 Revised: July 05, 2021 Accepted: July 07, 2021 Abstract: The present study investigates variability of headspace needle volatiles in native population of Abies x boisii-regis (King Boris fir) from the central Rhodopes (southern Bulgaria) as well as the relationships among this and seventeen previously studied fir populations from the presumed zones of natural hybridization in the Balkan Peninsula. According to multivariate statistical analyses, all four populations identified as A. x borisii-regis from southern Bulgaria, northern and central Greece included volatile profiles of both supposed parent species (A. alba and A. cephalonica), where profile frequencies changed clinally along the latitudinal gradient. Considering that one of the indicators of the presence of a hybrid zone is the clinical variation of all or a larger number of characters located in the same area, the observed geographic distribution of volatile entities supports the hypothesis that studied A. x borisii-regis populations are of secondary origin due to hybridization between A. alba and A. cephalonica. Key words: Abies x borisii-regis, Rhodopes, headspace, needle volatiles, chemodiversity Apstract: Varijabilnost hedspejs isparljivih jedinjenja u prirodnoj populaciji Abies x borisii-regis sa centralnih Rodopa Ovo istraživanje ispituje varijabilnost hedspejs isparljivih jedninjenja četina prirodne populacije Abies x borisii-regis (jela kralja Borisa) sa centralnih Rodopa (južna Bugarska), kao i odnose između ove i 17 prethodno istraživanih populacija jela iz pretpostavljene zone prirodne hibridizacije na Bakanskom poluostrvu. Na osnovu multivarijantnih statističkih analiza, sve četiri populacije identifikovane kao A. x borisii-regis iz južne Bugarske, severne i centralne Grčke uključivale su ispaljive profile obe pretpostavljene roditeljske vrste (A. alba and A. cephalonica), gde su se učestalosti profila klinalno menjale po gradijentu georgrafske širine. Uzimajući u obzir da je jedan od pokazatelja prisustva hibridne zone klinalno variranje svih ili većeg broja karaktera lociranih na istom području, dobijena geografska distribucija isparljivih entiteta podržava hipotezu da su ispitivane populacije A. x borisii- regis sekundarnog porekla nastale hibridizacijom A. alba i A. cephalonica. Ključne reči: Abies x borisii-regis, Rodopi, hedspejs, isparljiva jedinjenja četina, hemodi- verzitet Introduction Conifers of the genus Abies Mill. (Pinaceae) play an important ecological role in forest ecosystems of the northern hemisphere (Liepelt et al., 2010). Most of the Abies species are in the temperate zone, with a few of them on high mountains in subtropical areas (Xiang et al., 2007). The Mediterranean area is considered as one of the Abies distribution centers with about eight species depending on particular classification scheme. Mediterranean Abies spp. are weakly reproductively isolated and relatively easily hybridize in natural and artificial conditions as shown by crossability studies (Kormutak, 1985; 2004; Moulalis, 1986). © 2021 Nikolić et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and build upon your work non-commercially under the same license as the original. 23 On the Balkan Peninsula, there are two native species of the section Abies: A. alba Mill. (Silver fir), widely distributed across Central Europe and some parts of Southern and Eastern Europe, and A. cephalonica Loudon (Greek fir), endemic to the mountain massifs of central and southern Greece (Panetsos, 1975). Except them, A. x borisii-regis Matff. (King Boris fir) was described from the central Rhodopes in Bulgaria by Mattfeld (1926) as a hybrid between A. alba and A. cephalonica. Subsequently, Turrill (1937) considered as equally possible that King Boris fir is an ancient lineage from which A. alba diversified towards the north and A. cephalonica towards the south. In addition, there are a few more mutually exclusive hypotheses about its origin (Krajmerová et al., 2015). Nevertheless, in most previous classifications, A. x borisii-regis has the status of a hybrid between A. alba and A. cephalonica (Chater, 1964; Christensen, 1986; Christensen, 1997). Namely, it is considered that A. alba and A. cephalonica have experienced secondary contact in the refuges during periods of glacial expansion in Pleistocene. These events probably led to the exchange of genetic material between these two species in the glacial refugia which contributed to appearance of A. x borisii-regis (Liepelt et al., 2010). The reports of A. x borisii-regis distribution range 24 BIOLOGICA NYSSANA ● 12 (1) September 2021: 23-32 Nikolić et al. ● Variability of headspace volatiles in native population of Abies x borisii-regis from the central Rhodopes Fig. 1. Geographical position of analyzed populations. For the description of the location and habitat conditions of the populations, cf. Tab. 1. 25 are highly controversial: according to one concept A. x borisii-regis appears parapatrically along with the putative parent species, i.e. in its northern distribution with A. alba and in southern with A. cephalonica (Jalas & Suominen, 1973; Christensen, 1997; Andreev, 1992). Thus, it occurs in southern Albania, almost the whole Greece (except the Cephalonia Island and the southern half of the Peloponnese), the southernmost parts of Northern Macedonia and across southwestern and southern Bulgaria (Jalas & Suominen, 1973). On the other hand, based on the Atlas of the Word’s Conifers (Farjon & Filer, 2013), A. alba, A. x borisii-regis, and A. cephalonica are mostly allopatric, i.e., King Boris Fir is distributed in southern Albania, northern and northeastern Greece, and Rhodope Mt. in southern Bulgaria. Finally, Fady (1993) considered only the populations from Pindos Mt. and central Greece as A. x borisii-regis. The usefulness of terpenes as volatile chemophenetic markers has been addressed in detail by Hanover (1992), who emphasized their importance in research of biodiversity, geographic variability, evolution, and systematics, particularly in studies of conifer order Pinales. Static headspace (HS) extraction is a common technique for analyzing volatile organic compounds, enabling the analyst to study a variety of sample matrices while avoiding the costly and time-consuming preparation involved with traditional gas chromatography (Kolb & Ettre, 2006). The advantage of combining of HS and GC- MS/FID (gas chromatography-mass spectrometry/ flame ionization detection) methods in rapid and effective extraction and identification of highly volatile compounds as well as for chemotyping conifers has already been reported (Mitić et al., 2020; Nikolić et al., 2021). In our previous study, seventeen native fir populations from the presumed zones of natural hybridization in the Balkan Peninsula were analyzed using HS needle volatiles as chemophenetic markers (Nikolić et al., 2021). Abies alba and A. cephalonica were characterized by distinct volatile profiles, while three populations from northern and central Greece identified as A. x borisii-regis shared the volatile profiles with both of the supposed parent species (Nikolić et al., 2021). Although King Boris fir was described from the central Rhodopes in Bulgaria (Mattfeld, 1926), recent molecular study did not provide evidence on hybrid origin of any Bulgarian fir populations including populations from this mountain massif (Krajmerová et al., 2015). Therefore, the aim of the present study was to shed more light on chemotaxonomic status of controversial Bulgarian fir populations. Hence, in this article we presented new results for one population of A. x borisii-regis from the southern Bulgaria (the central Rhodopes) regarding diversity of HS needle volatiles. For multivariate statistical analyses (canonical discriminate analysis and agglomerative hierarchical clustering), in addition to the results obtained in this study, we used our previously published data relating to the broader transition zone between A. alba and A. cephalonica (Nikolić et al., 2021). In this way, HS needle volatiles of 18 populations of three Abies taxa from the Balkan Peninsula have been analyzed for the first time (Fig. 1). Materials and Methods Plant material Plant material of A. x borisii-regis Mattf. was collect- ed in the southern Bulgaria (the central Rhodopes, Pamporovo). Three-year-old needles were sampled during September 2021 from 14 individuals, which were randomly selected and distant at least 30 m. Plant material, deposited in labeled polyethylene bags (sample plot, date of collection, and locality), was immediately transferred to a freezer and stored at -20 °C prior to further analysis. Voucher specimen was deposited in the “Herbarium Moesiacum Niš” (HMN) of the Department of Biology and Ecology, Faculty of Sciences and Mathematics, University of Niš under the acquisition number 14389. Loca- tions of the presently studied population, as well as 17 populations previously analyzed by Nikolić et al. (2021) are presented in Fig. 1. The corresponding geographic and geologic data, date of collection, voucher information as well as the number of stud- ied individuals for every population are listed in the Tab. 1. Headspace isolation Three-year-old needles of every individual tree were cut into pieces of 2-3 mm length. Five hundred milligrams of chopped plant material was placed into 20-mL headspace vials and 1 mL of distilled water was added in each. The prepared vials were placed into a tray for further automated procedure. Each sample was heated at 80 °C for 20 minutes with the following program: shaking for 5 seconds, pausing for 2 seconds. After equilibration, the gas vapor of HS volatiles above the aliquot of volatiles was sampled with a gas syringe and subsequently injected into a GC analyzer. GC-MS/FID analyses Qualitative analysis of the HS volatiles was performed by Gas Chromatography-Mass Spectrometry (GC-MS), while the quantitative data were obtained by Gas Chromatography-Flame Ionization Detector (GC-FID) analysis. Samples BIOLOGICA NYSSANA ● 12 (1) September 2021: 23-32 Nikolić et al. ● Variability of headspace volatiles in native population of Abies x borisii-regis from the central Rhodopes 26 BIOLOGICA NYSSANA ● 12 (1) September 2021: 23-32 Nikolić et al. ● Variability of headspace volatiles in native population of Abies x borisii-regis from the central Rhodopes Ta bl e 1. L oc at io n an d ha bi ta t d es cr ip tio n of th e se le ct ed p op ul at io ns o f t hr ee A bi es ta xa Ta xo n L oc al it y P op ul at io n co de L at it ud e (N ) L on gi tu de (E ) A lt it ud e (m .a .s .l. ) Su bs tr at um N um be r of in di vi du al s an al yz ed D at e of co lle ct io n N um be r of va uc he rs A . a lb a R om an ia , S ou th er n C ar pa th ia ns , T ra ns fã gã rã şa n A R C 45 °2 3' 30 " 24 °3 7' 56 " 95 7 Si lic at e 5 A ug us t, 20 18 13 95 0 H M N A . a lb a Se rb ia , M t. Ta ra , P re do v K rs t A ST 40 °1 1' 47 " 22 °1 0' 48 " 10 96 L im es to ne 15 Ju ne , 2 01 8 13 94 9 H M N A . a lb a Se rb ia , M t. K op ao ni k, K ad ije va c A SK 43 °1 9' 43 " 20 °4 5' 41 " 13 50 Si lic at e 15 Ju ne , 2 01 7 13 94 6 H M N A . a lb a Se rb ia , M t. St ar a pl an in a, A rb in je A SS 43 °1 8' 25 " 22 °4 7' 18 " 12 95 Si lic at e 15 A ug us t, 20 18 13 94 4 H M N A . a lb a N or th M ac ed on ia , M t. Ša r- pl an in a, L eš ni ca A M S 42 ° 1' 17 " 20 °4 7' 04 " 15 05 L im es to ne 15 A ug us t, 20 17 13 94 5 H M N A . a lb a N or th M ac ed on ia , M t. N id že A M N 40 °5 8' 54 " 21 °4 7' 31 " 19 00 Si lic at e 15 A ug us t, 20 18 13 94 8 H M N A . a lb a B ul ga ri a, M t. Pi ri n, a bo ve B an sk o A B P 23 °2 8' 33 " 12 90 Si lic at e 15 O ct ob er , 2 01 8 13 94 2 H M N A . a lb a B ul ga ri a, M t. R ila , a bo ve R ila M on as te ry A B R 42 °0 9' 17 " 23 °2 4' 49 " 15 36 Si lic at e 15 O ct ob er , 2 01 8 13 94 7 H M N A . x b or is ii- re gi s B ul ga ri a, M t. R ho do pe s, Pa m po ro vo B B R 41 °3 9' 01 " 24 °4 1' 40 " 16 20 Si lic at e 14 Se pt em be r, 20 20 14 38 9 H M N A . x b or is ii- re gi s G re ec e, M t. Pi er ia , F te ri B G P 40 °1 1' 48 " 22 °1 1' 72 " 92 1 Si lic at e 15 O ct ob er , 2 01 8 13 94 3 H M N A . x b or is ii- re gi s G re ec e, M t. O ly m pu s, a bo ve L ep to ka ry a B G O 40 ° 01 '5 4" 22 °3 0' 25 " 64 0 L im es to ne 15 O ct ob er , 2 01 8 13 95 1 H M N A . x b or is ii- re gi s G re ec e, M t. Ty m fr is to s, S el lo s B G T 38 °5 4' 33 " 21 °5 4' 36 " 97 1 L im es to ne 15 Se pt em be r, 20 19 13 99 7 H M N A . c ep ha lo ni ca G re ec e, M t. O iti C G O 38 °4 4' 15 " 22 °2 2' 54 " 89 5 L im es to ne 15 Se pt em be r, 20 19 14 00 2 H M N A . c ep ha lo ni ca G re ec e, M t. Pa rn as so s, Pl ei ad es C G P 38 °3 3' 06 " 22 °3 4' 10 " 16 60 L im es to ne 15 Se pt em be r, 20 19 14 00 1 H M N A . c ep ha lo ni ca G re ec e, P el op on ne se , M eg a Sp ila io M on as te ry C G M 38 °0 5' 17 " 22 °1 0' 23 " 86 0 L im es to ne 15 Se pt em be r, 20 19 14 00 0 H M N A . c ep ha lo ni ca G re ec e, P el op on ne se , M t. C he lm os , K al av ry ta C G K 38 °0 0' 41 " 22 °0 9' 52 " 13 68 L im es to ne 15 Se pt em be r, 20 19 13 99 8 H M N A . c ep ha lo ni ca G re ec e, P el op on ne se , M t. C he lm os C G C 38 °0 0' 56 " 22 °1 0' 57 " 15 50 L im es to ne 15 Se pt em be r, 20 19 13 99 9 H M N A . c ep ha lo ni ca G re ec e, C ep ha lo ni a, M t. A in os C G A 38 °0 9' 37 " 20 °3 7' 13 " 10 05 L im es to ne 15 Ju ly , 2 01 9 13 99 6 H M N were analyzed on Agilent Technologies 7890B GC equipped with a fused silica capillary column (HP- 5MS, 250 μm×25 m, film thickness 0.25 μm, Agilent Technologies, Santa Clara, CA, USA) and coupled with a 7890A flame ionization detector (FID) and 7000B MS/MS spectrometer (operating in MS1 scan mode) from the same company. The GC was operated under the following conditions: injector temperature 250 °C; GC-MS interface temperature 300 °C; oven temperature programmed from 50 °C for 2.25 minutes, then to 200 °C at 4 °C/minute (carrier gas He, 1.0 mL/min, constant flow mode). MS conditions were as follows: ionization voltage of 70 eV; acquisition mass range 40-440; scan time 0.32 seconds. HS volatiles were identified from TIC (Total Ion Chromatogram) by comparison of their linear retention indices relative to C8-C20 n-alkanes recorded on the same column/temperature program with literature values (NIST MS Search 2.0; NIST Chemistry WebBook SRD69; Adams, 2007) and their mass spectra with those of standards from databases (Wiley 6, NIST02, Adams) by the application of the AMDIS software (the Automated Mass Spectral Deconvolution and Identification System, Ver. 2.7, distributed within the software package for 7890- 7000 BGC-MS/MS triple quadrupole system). The percentage composition of the HS volatiles was computed from the GC-FID peak areas without any corrections. Statistical analyses Statistical data processing was carried out by STATISTICA 8 software (Statsoft, Inc., Tulsa, OK, USA). Statistical matrices included HS volatiles that were detected in the content ≥0.5% in at least one of the analyzed taxa (12 components in total: santene, tricyclene, α-pinene, camphene, β-pinene, myrcene, α-phellandrene, limonene+β- phellandrene, terpinolene, borneol, bornyl acetate, (E)-caryophyllene). Multivariate analysis included canonical discriminant analysis (CDA) and agglomerative hierarchical clustering (AHC). Overall differences between the compared groups are presented by Mahalanobis distances. The calculated matrix distance was used for Agglomerative Hierarchical Clustering (AHC), using the Ward’s method. Results and discussion Variability of HS needle volatiles GC-MS/FID analyses of HS needle volatiles in A. x borisii-regis population from the central Rhodopes revealed the presence of 34 compounds (Tab. 2). The experimental and literary retention indices, identification methods, and classes to which identified compounds belong are also presented in Tab. 2. As the most volatile compounds, monoterpene hydrocarbons represented the major compound class (95.3%), while contents of oxygenated monoterpenes and sesquiterpene hydrocarbons were significantly lower (2.2 and 2.4%, respectively). Further, the phytochemical profile of the main HS volatiles for studied population could be presented as follows: β-pinene >>> α-pinene > camphene > limonene+β- phellandrene (where, based on Petrakis et al. (2001), =, >, >>, and >>> represent difference of 0.1-1.0%, 1.1- 5.0%, 5.1-15.0%, and more than 15.1%, resp.). According to Nikolić et al. (2021), the phytochemical profiles of the dominant HS volatiles of A. alba, A. x borisii-regis, and A. cephalonica were: A. alba: β-pinene >> limonene+β-phellandrene > camphene > α-pinene A. x borisii-regis: β-pinene >> α-pinene >> camphene > limonene+β-phellandrene A. cephalonica: β-pinene > α-pinene >>> camphene > limonene+β-phellandrene Thus, presently reported volatile profile of Bul- garian fir population from the central Rhodopes fits well with described profile of Greek A. x borisii- regis populations. Multivariate analyses (CDA and AHC) of HS needle volatiles at the population level The CDA based on 259 individuals from 18 populations of A. alba, A. x borisii-regis, and A. cephalonica showed that the first two canonical axes participated in 62.9% of the total discrimination, of which the first axis (CA1) accounted for 44.5% (Fig. 2 and Tab. 3). Four compounds, i.e. β-pinene, limonene+β-phellandrene, camphene, and α-pinene had significant impact on both axes, while bornyl acetate significantly affected only the CA2 (Tab. 3). Based on the obtained scatterplot (Fig. 2), all populations of A. alba showed positive values for the CA1 (yellow ellipse), while A. cephalonica populations formed a group at the negative part of CA1 (blue ellipse). According to the results of the post-hoc test that was done in the study of Nikolić et al. (2021), the “alba” profile was characterized by higher content of limonene+β-phellandrene and camphene and lower content of α-pinene compared to the second profile; vice versa the “cephalonica” profile was characterized by lower amount of limonene+β-phellandrene and camphene and higher 27 BIOLOGICA NYSSANA ● 12 (1) September 2021: 23-32 Nikolić et al. ● Variability of headspace volatiles in native population of Abies x borisii-regis from the central Rhodopes 28 BIOLOGICA NYSSANA ● 12 (1) September 2021: 23-32 Nikolić et al. ● Variability of headspace volatiles in native population of Abies x borisii-regis from the central Rhodopes Table 2. Relative percentage composition of HS needle volatiles of Bulgarian A. x borisii-regis population from the central Rhodopes Entry Compounds RI1 LI2 Content (%)3 n=14 Class of compounds Identification method 1. Hexanal 801 801 tr O RI, MS 2. (2E)-Hexenal 849 846 0.1±0.2 O RI, MS 3. (3Z)-Hexenol 850 850 tr O RI, MS 4. Hexanol 866 863 tr O RI, MS 5. Santene 884 884 0.5±0.2 MH RI, MS 6. Tricyclene 921 921 3.3±1.2 MH RI, MS 7. α-Pinene 934 932 19.7±2.5 MH RI, MS 8. Camphene 949 946 16.3±4.2 MH RI, MS 9. β-Pinene 980 974 40.3±5.5 MH RI, MS 10. Myrcene 990 988 1.2±0.3 MH RI, MS 11. α-Phellandrene 1004 1002 0.3±0.2 MH RI, MS 12. Limonene+β- Phellandrene 1029 1024/1025 13.3±2.1 MH RI, MS 13. 2-Heptyl acetate 1040 1038 tr O RI, MS 14. γ-Terpinene 1057 1054 tr MH RI, MS 15. Terpinolene 1087 1086 0.4±0.1 MH RI, MS 16. 2-Nonanone 1089 1087 tr O RI, MS 17. Borneol 1166 1165 1.0±0.9 MO RI, MS 18. Linalyl acetate 1254 1254 tr MO RI, MS 19. Bornyl acetate 1287 1287 1.2±1.0 MO RI, MS 20. α-Longipinene 1353 1350 0.1±0.1 SH RI, MS 21. α-Ylangene 1373 1373 tr SH RI, MS 22. α-Copaene 1377 1374 tr SH RI, MS 23. Sibirene 1404 1400 tr SH RI, MS 24. Longifolene 1409 1407 tr SH RI, MS 25. (E)-Caryophyllene 1420 1417 1.6±1.2 SH RI, MS 26. Himachala-2,4-diene 1429 14292a 0.1±0.2 SH RI, MS 27. Humulene 1456 1452 0.4±0.3 SH RI, MS 28. β-Selinene 1490 1489 tr SH RI, MS 29. δ-Selinene 1494 1492 0.1±0.1 SH RI, MS 30. α-Selinene 1498 1498 tr SH RI, MS 31. β-Himachalene 1503 1500 tr SH RI, MS 32. δ-Amorphene 1509 1511 tr SH RI, MS 33. γ-Cadinene 1516 1513 tr SH RI, MS 34. δ-Cadinene 1525 1522 tr SH RI, MS Total 100.0±0.2 Monoterpene hydrocarbons (MH) 95.3±1.9 Oxygenated monoterpenes (MO) 2.2±1.8 Sesquiterpene hydrocarbons (SH) 2.4±1.8 Others (O) 0.2±0.2 1RI: Experimental linear retention indices relative to C8-C20 alkanes. 2LI:Literature indices-Adams’ retention indices and 2aaccording to NIST data base. 3Contents are given as percentages (mean±standard deviation) of the total compounds content; n: the number of analyzed individuals; tr - trace<0.05%; (-) not detected compounds; the most abundant HS volatiles are in boldface. amount of α-pinene compared to the “alba” profile. Accordingly, eastern A. alba lineage and A. cephalonica may be undoubtedly characterized by volatile markers, while populations of A. x borisii- regis were positioned in the intermediate position. Specifically, all four populations identified as A. x borisii-regis included volatile profiles of both supposed parent species, whereby profile frequencies changed clinally along the latitudinal gradient (Fig. 2). In the northernmost population of A. x borisii- regis from southern Bulgaria (Rhodopes Mt. - BBR) most individuals belonged to the “alba” profile (the largest number of green diamonds is situated within the yellow ellipse), while individuals from two northern Greece populations were more evenly distributed on both sides of CA1. Nevertheless, in a slightly more northern population (Pieria Mt. - BGP) a larger number of individuals belonged to the “alba” profile (green squares more abundant within the yellow ellipse), while individuals from a somewhat southern population (Olympus Mt. - BGO) were evenly distributed on both sides of CA1, indicating approximately equal frequencies of both profiles (green circles with approximately equal representation within both ellipses). Finally, in the southernmost A. x borisii-regis population from the central Greece (Tymfristos Mt. - BGT) most individuals were located in the negative part of CA1, closer to the “cephalonica” profile (the largest number of green triangles is situated within the blue ellipse). Therefore, all four studied A. x borisii-regis populations included both the A. alba and A. cephalonica volatile profiles, but distributed along a latitudinal gradient connecting the ranges of presumed parent species, which confirms the hypothesis proposed in our previous study (Nikolić et al., 2021). Considering that one of the indicators of the presence of a hybrid zone is the clinical variation of all or a larger number of characters located in the same area, the observed geographic distribution of volatile entities supports the hypothesis that studied A. x borisii-regis populations are of secondary origin due to hybridization between A. alba and A. cephalonica. In addition, clinal (north-south) pattern in the distribution of morpho-anatomical (Mitsopoulos & Panetsos, 1987) and genetic traits (Krajmerová et al., 2015) over the A. x borisii-regis range was previously reported. The dendrogram obtained by AHC analysis has also shown a tendency to divide analyzed fir populations in similar way (Fig. 3). Namely, populations of A. alba and A. cephalonica were divided in two separate clusters, while populations of A. x borisii-regis appeared heterogeneous. Actually, the northernmost population of A. x borisii- 29 BIOLOGICA NYSSANA ● 12 (1) September 2021: 23-32 Nikolić et al. ● Variability of headspace volatiles in native population of Abies x borisii-regis from the central Rhodopes Variables CA1 CA2 Santene -0.105 -0.164 Tricyclene 0.827 -0.905 α-Pinene 1.123 1.526 Camphene 2.193 2.934 β-Pinene 4.207 2.859 Myrcene 0.063 0.056 α-Phellandrene -0.089 0.203 Limonene+β-Phellandrene 3.690 2.987 Terpinolene 0.066 -0.177 Borneol 0.468 -0.712 Bornyl acetate 0.610 1.239 (E)-Caryophyllene 0.730 0.436 Eigenvalue 4.197 1.729 % explained variation 0.445 0.629 Table 3. Standardized coefficients for the first two canonical axes (CA) of variation in 12 needle vola- tiles from the discriminant functional analysis of 18 a priori groups. The most significant coefficients are in boldface Fig. 2. Canonical discriminant analysis (CDA) based on contents of 12 HS needle volatiles isolated from 259 individuals of 18 populations. Abbreviations and symbols refer to populations as indicated in Tab. 1. and Fig 1. Abbreviation colors: yellow – A. alba, green – A. x borisii-regis, and blue – A. cephalonica. The yellow and blue ellipses encompass all individuals of supposed parent species. regis from southern Bulgaria (BBR) fitted within the “alba” and all Greek populations of A. x borisii-regis (BGP, BGO, and BGT) within the “cephalonica” cluster. Furthermore, the northernmost (BBR) and southernmost (BGT) population of A. x borisii- regis showed the greatest resemblance to the geographically closest populations of the parent species: southwestern Bulgarian populations of A. alba (ABR and ABP) and population of A. cephalonica from central Greece (CGO), respectively (Fig. 3). The history of fir populations from Rhodope Mt. is largely unknown; fluctuating presence of fir pollen was recorded in Rhodope sites since the end of the Pleniglacial until the early Holocene and an expansion occurred only after 8500 BP (Terhürne- Berson et al., 2004). Although King Boris fir was described from the central Rhodopes (Mattfeld, 1926), recent molecular study did not provide evidence on hybrid origin of any of 10 analyzed populations from this mountain massif, based on maternally inherited mitochondrial (mtDNA) markers (Krajmerová et al., 2015). Moreover, these authors found exclusively A. alba mitotype in all fir populations from Bulgaria, North Macedonia, and northern Greece, while only two populations from central Greece (Tymfrostos and Rentina Mts.) were mixed and contained predominantly the A. cephalonica mitotype, with a lower proportion of the A. alba mitotype. Even though Bella et al. (2014) analyzed the same mtDNA marker (nad5-4) as in the study of (Krajmerová et al., 2015), they found a mixture of maternal lineages (mitotypes of both parent species) in two northern Greek populations (Pieria and Pindos Mts.). Nevertheless, the second study did not include any of fir populations from central Greece, Bulgaria, or North Macedonia. According to phytochemical markers selected in the present study, presence of both volatile profiles was recorded in a somewhat wider geographical area. i.e. four populations from the southern Bulgaria, northern and central Greece, where the northernmost population contained predominantly A. alba and the southernmost A. cephalonica profile (Fig. 2). Thus, based on this concept, A. alba and A. cephalonica were co-occurring at least in the southern Bulgaria, northern and central Greece, suggesting that this region may have served as a refugium for A. alba during the Pleistocene glacial cycles. Bearing in mind several proposed distribution concepts of King Boris fir, our results are mostly in agreement with the concept given in the Atlas of the Word’s Conifers (Farjon & Filer, 2013) with the addition of central Greece. Conclusions GC-MS/FID analyses of HS needle volatiles in population A. x borisii-regis from the southern Bulgaria (the central Rhodopes) revealed presence BIOLOGICA NYSSANA ● 12 (1) September 2021: 23-32 30 Nikolić et al. ● Variability of headspace volatiles in native population of Abies x borisii-regis from the central Rhodopes Fig. 3. Dendrogram obtained by agglomerative hierarchical clustering (AHC) of 18 populations based on 12 HS needle volatiles. Abbreviations and symbols refer to populations as indicated in Tab. 1. and Fig 1. Abbreviation colors: yellow - A. alba, green - A. x borisii-regis, and blue - A. cephalonica of 34 compounds. The major HS volatiles were β-pinene (40.3%), α-pinene (19.7%), camphene (16.3%) and limonene+β-phellandrene (13.3%). The presently reported phytochemical profile of the main HS volatiles fits well with previously described profile of Greek A. x borisii-regis populations. Based on multivariate statistical analyses, all four populations identified as A. x borisii-regis from southern Bulgaria, northern and central Greece included volatile profiles of both supposed parent species, where profile frequencies changed clinally along the latitudinal gradient connecting the ranges of parent species. Thus, based on HS needle volatiles, A. alba and A. cephalonica were co-occurring at least in the southern Bulgaria, northern and central Greece, suggesting that this region may have served as a refugium for A. alba during the Pleistocene glacial cycles. As a natural hybrid, populations of King Boris fir can be interesting from the point of view of commercial forestry for the reason that offer the possibility to cope with the predicted forest decline in Southern Europe as a result of climate change and can be used for reforestation. Acknowledgements. This research was supported by Grants No 451-03-9/2021-14/200124 by the Ministry of Education, Science and Technological Development of the Republic of Serbia. References Andreev, N. 1992: Pinophyta, in: Kozuharov, S. (ed.), Field Guide to the Vascular plants in Bulgaria, 81-85, Naouka & Izkoustvo, Sofia (in Bulgarian). Bella, E., Liepelt, S., Parducci, L., Drouzas, A. 2014: Genetic insights into the hybrid origin of Abies x borisii-regis Mattf. Plant Systematics and Evolution, 301: 749-759. Chater, A.O. 1964: Abies Miller. In: Tutin, T.G., Heywood, V.H., Burges, N.A., Valentine, D.H., Walters, S.M., Webb, D.A. (eds.), Flora Europaea, 1: 37-38, Cambridge University Press, Cambridge. Christensen, K.I. 1986: Abies Miller. In: Strid, A. (ed.), Mountain flora of Greece, 1: 38-41, Cambridge University Press, Cambridge. Christensen, K.I. 1997: Abies Miller. In: Strid, A., Tan, K. (eds.), Flora Hellenica, 1: 1-3, Koeltz Scientific Books, Königstein. Fady, B. 1993: Caractéristiques écologiques et sylvicoles des sapins de Grèce dans leur aire naturelle et en plantation dans le Sud de la France. Perspectives pour le reboisement en règion méditerrané enne. Revue Forestière Française, 45: 119-133. Farjon, A., Filer, D. 2013: An Atlas of the World’s Conifers: an Analysis of Their Distribution, Biogeography, Diversity and Conservation Status. Koninklijke Brill NV, Leiden. 512 p. Hanover, J.W. 1992: Applications of terpene analysis in forest genetics. New Forests, 6: 159-178. Jalas, J., Suominen, J. 1973: Atlas Florae Europaeae. 2 Gymnospermae (Pinaceae to Ephedraceae). The Committee for Mapping the Flora of Europe & Societas Biologica Fennica Vanamo, Helsinki, 40 p. Kolb, B., Ettre, L.S. 2006: Static Headspace-Gas Chromatography: Theory and Practice. Wiley- Interscience, New York. 349 p. Kormutak, A. 1985: Study on species hybridization within the genus Abies. Acta dendrobiologica. VEDA, Bratislava. 127 p. Kormutak, A. 2004: Crossability relationships between some representatives of the Mediterranean, Northamerican and Asian firs (Abies sp.).VEDA, Bratislava. Krajmerová, D., Paule, L., Zhelev, P., Voleková, M., Evtimov, I., Gagov, V., Gömöry, D. 2015: Natural hybridization in eastern-Mediterranean firs: The case of Abies borisii-regis. Plant Biosystematics, 150: 1189-1199. Liepelt, S., Mayland-Quellhorst, E., Lahme, M., Ziegenhagen, B. 2010: Contrasting geographical patterns of ancient and modern genetic lineages in Mediterranean Abies species. Plant Systematics and Evolution, 284: 141-151. Mattfeld, J. 1926: Die Europäischen und Mediterranen Abies Arten. Die Pflanzenareale, 1: 22-29 (in German). Mitić, Z.S., Jovanović, S.Č., Zlatković, B.K., Milanovici, S.J., Nikolić, B.M., Petrović, G.M., Stojanović G.S., Marin, P.D. 2020: Variation of needle volatiles in native populations of Pinus mugo- evidence from multivariate statistical analysis. Plant Biosystems-An International Journal Dealing with all Aspects of Plant Biology, 155(4): 700-710. Mitsopoulos, D.J., Panetsos, C.P. 1987: Origin of variation in fir forests of Greece. Silvae Genetica, 36: 1-15. Moulalis, D. 1986: Diagnosis and characterization of fir hybrids. Scientific Annals of the Department of Forestry and Natural Environment, Aristotle University of Thessaloniki, 29: 371-404. Nikolić, J.S., Zlatković, B.K., Jovanović, S.Č., Stojanović, G.S., Marin, P.D., Mitić, Z.S. 2021: Needle volatiles as chemophenetic markers in 31 BIOLOGICA NYSSANA ● 12 (1) September 2021: 23-32 Nikolić et al. ● Variability of headspace volatiles in native population of Abies x borisii-regis from the central Rhodopes BIOLOGICA NYSSANA ● 12 (1) September 2021: 23-32 32 Nikolić et al. ● Variability of headspace volatiles in native population of Abies x borisii-regis from the central Rhodopes differentiation of natural populations of Abies alba, A. x borisii-regis, and A. cephalonica. Phytochemistry, 183: 112612. Panetsos, C.P. 1975: Monograph of Abies cephalonica Loudon. Annals of Forest Science, 7: 1-18. Petrakis, P.V., Tsitsimpikou, C., Tzakou, O., Couladis, M., Vagias, C., Roussis, V. 2001: Needle volatiles from five Pinus species growing in Greece. Flavour and Fragrance Journal, 16: 249-252. Terhürne-Berson, R., Litt, T., Cheddadi, R. 2004: The spread of Abies throughout Europe since the last glacial period: Combined macrofossil and pollen data. Vegetation History and Archaeobotany, 13: 257-268. Turrill, W.B. 1937: XI-On the flora of the near east: XVIII. New species, new records and notes. Kew Bulletin, 2: 79-86. Xiang, X.G., Cao, M., Zhou, Z.K. 2007: Fossil history and modern distribution of the genus Abies (Pinaceae). Frontiers of Forestry in China, 2: 355- 365.