Phytologia (August2012) 94(2) 199 GEOGRAPHIC VARIATION IN LEAF ESSENTIAL OILS OF DOUGLAS FIR(PSEUDOTSUGA MENZIESII) Robert P. Adams BiologyDepartment, BaylorUniversity, Box 97388, Waco, TX 76798, USA [email protected] J. Jesus Vargas-Hernandez ProgramaForestal - Campus Montecillo, Colegio de Postgraduados km 65.5 CarretaraMexico - Texcoco, Mexico CP. 56230 Mexico M. Socorro Gonzalez Elizondo CIIDIRUnidad Durango, Instituto PolitecnicoNacional, Sigma 119, Fracc. 20 deNoviembre II, Durango, Dgo., 34220 Mexico Gay Hunter Museum Curator, OlympicNational Park WA 600 EastParkAve., PortAngeles, 98362 Thomas A. Fairhall WA 1136 Water St., Suite 101, PortTownsend, 98368 David Thornburg 220 W. WinchesterLane, Cottonwood AZ and FrankCallahan Callahan Seeds, P. O. Box 5531, Central Point, OR97502 ABSTRACT The volatile leaf oils of Douglas fir {Pseudotsuga menziesii) were analyzed from throughout its range. The major differentiation foundwasthe divergence ofthe inlandpopulations (var.glaucd) fromthe coastal and Sierra Nevada populations (var. menziesii). The oils ofvar. menziesii differed from var. glauca in theirmajor components: camphene (0.4-17, 25-30%), p-pinene (25-38, 2.5-12%), sabinene (4-12, trace- 0.5%), a-terpinene (2.1-3.3, 0.2-0.0%), y-terpinene (3.4-5.5, 0.1-0.3%), 200 Phytologia (August2012) 94(2) terpinolene (9.5-14.6, 1.2-1.6%), cis-p-menth-2-en-l-ol (0.7-0.8, 0.0- trace), terpinen-4-ol (10-12.1, trace-0.5%), and bomyl acetate (0.2-1.3, 14.6-44.7%). The oil of the Sierra Nevada population (Sierra Nevada race of Snajberk and Zavarin, 1976) was quite similar to typical coastal Doug Fir from WA and OR and not as distinct as found in the oleoresin oils ofSnajberk and Zavarin. Overall the oils ofthe inland variety (var. glauca) varied from Yellowstone southward with the AZ - NM oils forming a group along with the southern Mexico oils except for the population at Cerro Potosi that had a different oil. The Cerro Potosi oil differed from other inland (var. glauca) oils in having larger amounts of a-pinene, P-pinene, 6-3-carene and smaller amounts of camphene, limonene and bomyl acetate. The leafoil ofvar. oaxacana did not differ from var. glauca from nearby populations in southern Mexico. The leaf essential oils ofP. macrocarpa are also reported. Phytologia 94(2):199- 218 (August 1, 2012). KEY WORDS: Pseudotsuga menziesii, var. menziesii, var. glauca, var. oaxacana, leafessential oils, geographic variation, Douglas Fir, Douglas Fir [Pseudotsuga menziesii (Mirb.) Franco] is a wide ranging, common forest tree in North America (Fig. 1). The nomenclatural history of the name is a morass (see http://www.plantsystematics.org/reveal/pbio/LnC/dougfir.html), but seems to be settled by James Reveal. In a recent treatment, Eckenwalder(p. 572, 2009) recognizes two varieties: var. menziesii and var. glauca (Mayr) Franco [cited as (Beissn.) Franco in Eckenwalder, 2009]. Eckenwalder (2009) did not recognize var. oaxacana Debreczy & Racz, described from Oaxaca(Debreczy and Racz, 1995). The leafessential oils ofP. menziesii have been exhaustively studied by von Rudloff (von Rudloff, 1972, 1973, 1984, von Rudloff and Rehfeldt, 1980) who carefully documented the large differences in oils between coastal (var. menziesii) and inland (var. glauca) in many compounds including camphene (0.3-8, 20-30%), p-pinene (15-30, 5- 10%), sabinene (2-12, 0.1-0.5%), a-terpinene 1-3, 0-0.3%), y-terpinene (2-8, 0.1-1%), terpinolene (5-15, 0.5-3%), terpinen-4-ol (5-15, 0.5-3%), and bomyl acetate (0.5-5, 20-30%). However, von RudlofPs studies Phytologia (August2012) 94(2) 201 were limited to the Pacific Northwest and northern US-Canada (south to Wyoming). A second team from USDA, Forest Products, Richmond, CA (Snajberk and Zavarin), made extensive studies ofthe terpenoids from the oleoresin of Douglas Fir (Snajberk, Lee and Zavarin, 1974, Snajberk and Zavarin, 1976, Zavarin and Snajberk, 1973, 1975). In their most comprehensive study (Snajberk and Zavarin, 1976), they found four chemical races: coastal, northern inland, southern inland and Sierra Nevada. These are shown in Fig. 1, alongwith populations used in the present study. Notice the populations sampled in the present study are from each race: coastal (2), northern inland (1), SierraNevada (1), southern inland (2), plus Mexico (6) as well as P. macrocarpa. Douglas Fir in Mexico is principally found in small restricted populations, except for a larger area in Chihuahua in northern Mexico (Fig. 1). In spite ofthe very exhaustive studies ofleafand wood oils in the United States and Canada, nothing has been reported about variation in Douglas Firoils in Mexico. However, there has been work at the molecular level on Douglas Fir in Mexico. Li and Adams (1989) reported that allozymes dividedthe Douglas Fir into northern coastal (var. menziesii) and inland (var. glauca) with two subgroups (northern and southern inland). They did not find evidence of a subgroup of Sierra Nevada Douglas Fir as Snajberk and Zavarin (1976) found, based on the oleoresin oils. In addition, Li and Adams (1989) found a distinct pattern inthe allozymes MX from population 103 at General Cepeda, Coah., and speculatedthat it might be P. flahaultii Flous (also recognized by Martinez, 1963). However, a nearby collection (104, La Encantada, near Zaragoza, NL) clustered closely with P. menziesii from New Mexico. So that case seems unresolved. Gugger et al. (2010) examined mtDNA and cpDNA sequences and found support for coastal (var. menziesii) and inland (var. glauca) in the United States and Canada. No evidence was found for a Sierra Nevada taxon, but mtDNA suggested the inland (var. glauca) might be divided into northern and southern groups. In a subsequent study, Gugger et al. (2011) examined Douglas Fir from Mexico. They found considerable divergence in cpDNA from Cerro Potosi,NL and Jame, 202 Phytologia (August2012) 94(2) ^ / Pseudotsuga menziesii ^ / chemical races of \. SnaiberK and Zavarin, 1976 / ^\ V.'^^L / coastal ^\ N^r northern, I \ f r*^ I Iinland I • Cj^^f^ = populations sampled, 2011, present study Figure 1. Distribution ofP. menziesii with chemical races ofSnajberk andZavarin, 1976 andpopulations sampled inthepresent study. Coah. from other Mexico populations. CpSSRs supported two clades in Mexico (Gugger, Fig. 4c), but that pattern was not supported in mtDNA (Gugger, Fig. 4a) or cpDNA (Gugger, Fig. 4b) data; in summary, they concluded that "Mexican populations were genetically Phytologia (August2012) 94(2) 203 distinct from USA and Canadian populations, but more closely related to the Rocky Mountain variety than the coastal variety". As Gugger et al. (2010) did not show data from Mexico, and Gugger et al. (2011) showed only data from that country, it is difficult to ascertain the relationship ofMexicanpopulations to those ofthe USA. Phenotypic analyses (Reyes Hernandezet al. 2006) revealed thatPseudotsuga populations ofnorthernMexico are morphologically similarto P. menziesiivar. glauca from southwestern USA, butthe populations from central Mexico differed. They also found apopulation ofNE Mexico (San Francisco) morphologically separated fromthe rest, even from those ofthe same geographical region, suggesting an effect ofisolation. This population isjust 15 kmNWfromthe one from El Potosi analyzedhere. The purpose this paper is to report on geographic variation in the leaf essential oils of Pseudotsuga menziesii from the USA and Mexico. MATERIALS AND METHODS Plant material: P. menziesii var. menziesii (coastal/ Sierra Nevada): Adams J2918-12922, Olympic National Forest, Port Angeles, WA, 48° 02' 48.1" N, 123° 25' 04.08" W, 525 m; Adams 12745-12757, on serpentine, Oregon Mtn., OR, 41° 59' 59.1" N, 123° 47' 10.2" W, 895 m; Adams 12779-12783, 6 km e ofBuck Meadows, CA, 21 km w of Yosemite NP on US 120, 37° 49.579' N, 119° 58.421' W, 1150 m. var. glauca: Adams 12556-12560, 13 km w of Cimarron, NM on US 64, 36.54684° N, 105.03321° W, 2125 m; Adams 12744-12748 (ex D. Thornburg, 1-5), 9 km ne ofPine, AZ on Hwy 87, 34° 27.422' N, 111° 24.115' W, 2250 m; Adams 12818-12822, 20 km e ofYellowstone NP, onus 14 atthe Palisades, 44.45448°N, 109.78182°W, \9\() m;Adams 13056-13060, (ex M. Socorro Gonzalez Elizondo 7777a-e), Cerro Potosi, NL, 24° 53' 9" N, 100° 13'14" W, 3141 m.; Adams 13061- 13066, (ex Martha Gonzalez Elizondo 4408-4409, 4413-4416) Los Altares, Dur., 25°2'56" N, 105°59'48" W, 2310 m; Adams 13082- 13087 (ex Vargas-HernandezJ1-J6), El Chico Natl. Park, Mineral del Chico, Hgo., 20° 10' 16" N, 98° 43' 55" W, 2,765m, Nov. 4, 2011; Adams 13088-13094 (ex Vargas-Hernandez C1-C7), Cuatexmola, 204 Phytologia (August2012) 94(2) Ixtacamaxtitlan, Puebla, 19° 30' 22" N, 9r 50' 21" W, 2,980m, Oct. 30, 2011; Adams 13095-13100 (ex Vargas-Hernandez T1-T6), Ejido Paso National, Tlachichuca, Puebla, 19M7' 47" N, 97° 19' 45" W, 3,000m, Sep. 30, 2011; var. oaxacana: Adams 13101-13103, 13105-13106 (ex Vargas-Hernandez 11-16, Paraje Pena Prieta, Oaxaca, 17° 09' 38" N, 96° 38' 07" W, 2,700m, Oct. 21, 201 1. P. macrocarpa: Adams 12776- 12778, USPS Eddy Arboretum, Placerville, CA. Voucher specimens are deposited in the Herbarium, Baylor University (BAYLU, CIIDIR and CHAPA). Isolation ofOils - Fresh leaves (200 g) were steam distilled for 2 h using a circulatory Clevenger-type apparatus (Adams, 1991). The oil samples were concentrated (ether trap removed) with nitrogen and the samples stored at -20°C until analyzed. The extracted leaves were oven dried(100°C, 48 h) fordetermination ofoil yields. Chemical Analyses - Oils from 10-15 trees of each ofthe taxa were analyzed and average values reported. The oils were analyzed on a HP5971 MSD mass spectrometer, scan time 1/ sec, directly coupled to a HP 5890 gas chromatograph, using a J & W DB-5, 0.26 mm x 30 m, 0.25 micron coating thickness, fused silica capillary column (see Adams, 2007 for operating details). Identifications were made by library searches of the Adams Essential Oil library (Adams, 2007), using the HP Chemstation library search routines, coupled with retention time data of authentic reference compounds. Quantitation was by FID on an HP 5890 gas chromatograph using a J & W DB-5, mm 0.26 x 30 m, 0.25 micron coating thickness, fiised silica capillary columnusingtheHP Chemstation software. Data Analysis - Terpenoids (as per cent total oil) were coded and compared among the species by the Gower metric (1971). Principal coordinate analysis was performed by factoring the associational matrix using the formulation of Gower (1966) and Veldman(1967). RESULTSAND DISCUSSION The leafessential oils ofP. menziesii var. menziesii (coastal) were found (Table 1) to be nearly identical to the reports of von Phytologia (August2012) 94(2) 205 Rudloff (1972, 1973). This is remarkable considering the changes in gas chromatography from packed to fused capillary columns. The oils of var. menziesii differed (Table 1) from var. glauca in their major components: camphene (0.4-17, 25-30%), p-pinene (25-38, 2.5-12%), sabinene (4-12, trace-0.5%), a-terpinene (2.1-3.3, 0.2-0.0%), y- terpinene (3.4-5.5, 0.1-0.3%), terpinolene (9.5-14.6, 1.2-1.6%), cis-p- menth-2-en-l-ol (0.7-0.8, 0.0-trace), terpinen-4-ol (10-12.1, trace- 0.5%), and bomyl acetate (0.2-1.3, 14.6-44.7%). The oil from the Yosemite NP, Sierra Nevada (Yose in Table 1) population (cf Sierra Nevada race of Snajberk and Zavarin, 1976) was quite similar to WA typical coastal DougFirfrom (ONF, Table 1) andnot as distinct as found in the oleoresin oil of Snajberk and Zavarin. However, comparing coastal (ONF) and Sierra Nevada (Yose, Table 1) oils reveals the coastal oil is higher in sabinene (12.9, 4.0%), 6-3-carene (1.8, 0.5%), y-terpinene (5.4, 3.4%), terpinolene (14.6, 9.5%) and geranyl acetate (2.1, 0.9%), but lower in P-pinene (25.5, 38.0%), bomyl acetate (0.2, 1.3%) and citronellyl acetate (1.2, 2.8%). The inland group (var. glauca. Yell, AZ, NM, A Du, El C, Oax, Table 1) shows some differences between the northern inland (Yell) and southern inland oils (excluding Cerro Potosi) for camphene (17.0 vs. 24-30%), a-pinene (2.5 vs 3.2-9.1%) andbomyl acetate (44.7 vs. 16-32%). The Cerro Potosi oil (C Po, Table 1) differed from other inland (var. glauca) oils in having larger amounts of a-pinene, P- pinene, 5-3-carene and smaller amounts of camphene, limonene and bomyl acetate. The leaf oil of var. oaxacana (Oax, Table 1) did not differfrom var.glauca oils innearbypopulations in southemMexico. Principal Coordinate (PCO) analysis of 24 terpenoids (boldface. Table 1) excluded one (Z-p-ocimene) whose maximum value was too small (0.4%). PCO ofthe resulting 23 terpenoids gave eigenroots that accounted for 48.7, 15.4, 7.9, 7.5 and 5.0% of the variation before the eigenroots asymptoted. The overall trends among the oils are seen in Fig. 2 where the major Principal Coordinate (PCO) accounted 49% ofthe variance among the populations. The first axis separated coastal (var. menziesii) and inland (var.glauca) populations. Notice that Pseudotsuga macrocarpa is well resolved (Fig. 2). 206 Phytologia (August2012) 94(2) To examine variation among the P. menziesii populations, P. macrocarpa was removed form the data and a new PCO performed. PCO analysis of 24 terpenoids (boldface, Table 1) excluded Z-p- ocimene (max. value 0.4%) and humulene epoxide II (max. value 0.0). PCO ofthe resulting 22 terpenoids gave eigenroots that accounted for 57.6, 9.4, 8.4, 5.9 and 4.2% of the variation before the eigenroots asymptoted. 2(15%) PCO 23 terpenes p. macrocarpa 1(49% U/ ) / P. menziesii/ coastal NA Figure 2. PCO ofP. menziesiipopulations andP. macrocarpa based on 23 terpenes. This resulted in an even greater amount ofvariance removed by PCO axis 1 (58%, Fig. 3). One can now see some differentiation among the inland populations. Note particularly the much lower similarity ofCerro Potosi oil to other inland populations (0.774), than seen between other inlandpopulations (0.892, 0.888). There appears to NM be a slight north - south cline from Yellowstone - AZ, - Mexico populations (see tricyclene, a-pinene, camphene, P-pinene and bomyl acetate. Table 1). Phytologia (August2012) 94(2) 207 2(9%) PCO 22 terpenes P. menziesii inland NA and Mexico Yellowstone AZ, NM -888 Mexico 1(58% .512 Cerro R menziesii Potosi coastal NA 3(8%) Figure 3. PCO of 12 populations of P. menziesii using 22 teq^enes. Dashed lines are minimum spanning links. Numbers next to the lines are similarities. Another way to visualize the clustering and similarities is by use of a phenogram. Figure 4 shows the clustering of populations based on 22 terpenes. The coastal - inland split is the majortrend. PseudotSLiga menziesii ElChico. Hgo Minimum Spanning TIach., Pue. 22 terpenes Cuat., Pue. Altares, Dur. Oaxac. Oax. NewMexico var. glauca(inland) Arizona Yellowstone CerroPotosi OlympicNF var.menziesii(coastal, Sierra Nev) Oregon Mtn. „YoSemiteNP Figure 4. Minimum spanningdiagram basedon 22 terpenes for populations ofDouglas Fir. 208 Phytologia (August2012) 94(2) One can also see that the oil from the Cerro Potosi population is the most unusual oil in the inland (var. glauca) group (Fig. 4). In addition, the divergence of the Yosemite NP (Sierra Nevada) population is clearly seen. The Mexico populations (excluding Cerro Potosi) are very uniform in their oils, including var. oaxacana (Oaxac. in Fig. 4). There is no support forthe recognition ofvar. oaxacana, but itmay be divergent in molecular characters. To better understand the variation, geographic clustering was preformed. The resulting diagram (Fig. 5) clearly shows the uniformity ofthe oils in the inland group and the divergence ofthe oil of Cerro Potosi from other Mexico populations. Again, the oil ofvar. oaxacana from Oaxaca is shown to be very similar to nearby populations ofvar. glauca. These oils data are similar to the cpDNA sequencing data of Gugger et al. (2011), showing Cerro Potosi population to be different from other Mexicopopulations. A recent study (Wei et al. 2011) found a similar pattern in mtDNA (coastal and inland groups), but they found the cpDNA of inland Mexico populations to differ from inland USA populations. Additional molecular studies (in progress) are needed to clarify the taxonomic relationship of Cerro Potosi and inland USA populations. ACKNOWLEDGEMENTS Thanks to Tonya Yanke for lab assistance and to Martha Gonzalez and Abraham Torres for field assistance. This research was supported inpartwith funds from Baylor University. LITERATURE CITED Adams, R. P. 1991. Cedarwood oil - Analysis andproperties, pp. 159- 173. in: Modem Methods ofPlantAnalysis,New Series: Oil and Waxes. H.-F. Linskens and J. F. Jackson, eds. Springier- Verlag, Berlin. Adams, R. P. 2007. Identification ofessential oil components by gas chromatography/mass spectrometry. 4th ed. AlluredPubl., Carol Stream, IL.