alteration of foliar flavonoid chemistry induced by enhanced uv-b radiation in field-grown pinus...

Journal of Photochemistry and Photobiology B: Biology 66 (2002) 125–133www.elsevier.com/ locate / jphotobiol

Alteration of foliar flavonoid chemistry induced by enhanced UV-Bradiation in field-grown Pinus ponderosa, Quercus rubra and

Pseudotsuga menziesiia a , b b*Jeffrey M. Warren , John H. Bassman , D. Scott Mattinson , John K. Fellman ,

c dGerald E. Edwards , Ronald RobberechtaDepartment of Natural Resource Sciences, Washington State University, Pullman, WA 99164-6410, USA

bDepartment of Horticulture and Landscape Architecture, Washington State University, Pullman, WA 99164-6414, USAcSchool of Biological Sciences, Washington State University, Pullman, WA 99164-4236, USA

dDepartment of Rangeland Ecology, University of Idaho, Moscow, ID 83844-1135, USA

Accepted 15 January 2002

Abstract

Chromatographic analyses of foliage from several tree species illustrate the species-specific effects of UV-B radiation on both quantityand composition of foliar flavonoids. Pinus ponderosa, Quercus rubra and Pseudotsuga menziesii were field-grown under modulatedambient (13) and enhanced (23) biologically effective UV-B radiation. Foliage was harvested seasonally over a 3-year period, extracted,purified and the flavonoid fraction applied to a mBondapak/C column HPLC system sampling at 254 nm. Total flavonoid concentrations18

in Quercus rubra foliage were more than twice (leaf area basis) that of the other species; Pseudotsuga menziesii foliage had intermediatelevels and P. ponderosa had the lowest concentrations of total flavonoids. No statistically significant UV-B radiation-induced effects werefound in total foliar flavonoid concentrations for any species; however, concentrations of specific compounds within each speciesexhibited significant treatment effects. Higher (but statistically insignificant) levels of flavonoids were induced by UV-B irradiation in 1-and 2-year-old P. ponderosa foliage. Total flavonoid concentrations in 2-year-old needles increased by 50% (13 ambient UV-B radiation)or 70% (23 ambient UV-B radiation) from that of 1-year-old tissue. Foliar flavonoids of Q. rubra under enhanced UV-B radiation tendedto shift from early-eluting compounds to less polar flavonoids eluting later. There were no clear patterns of UV-B radiation effects on1-year-old P. menziesii foliage. However, 2-year-old tissue had slightly higher foliar flavonoids under the 23 UV-B radiation treatmentcompared to ambient levels. Results suggest that enhanced UV-B radiation will alter foliar flavonoid composition and concentrations inforest tree species, which could impact tissue protection, and ultimately, competition, herbivory or litter decomposition. 2002 ElsevierScience B.V. All rights reserved.

Keywords: Pinus ponderosa; Quercus rubra; Pseudotsuga menziesii; Ultraviolet radiation; Phenolic; HPLC; Secondary metabolites

1. Introduction the ozone column before the year 2050 [6–8]. It nowappears there may be negative synergistic effects with

Enhanced ultraviolet-B (UV-B; 280–320 nm) radiation global warming, denitrification, or other atmospheric pro-(increased UV-B irradiance and a shift towards shorter, cesses that could exacerbate stratospheric O depletion and3

more actinic wavelengths) is being observed at temperate delay recovery of the ozone column [9,10]. Data for theand polar regions of the Earth’s surface as a result of year 2000 show a return to polar O depletions not3

depletion in stratospheric ozone (O ) [1–5]. Although observed since the mid-1990s [9]. Consequently, world3

there has been some success in reducing causal chloro- biomes will likely experience enhanced UV-B radiation forfluorocarbon (CFC) emissions based upon the Montreal some time to come.Protocol, optimistic estimates do not suggest recovery of The effects of short-term exposure of UV-B radiation on

higher plants, especially herbaceous agronomic crops, havebeen well-documented (e.g. Refs. [11–15]) and include*Corresponding author. Tel.: 11-509-335-5296; fax: 11-509-335-morphological, physiological and biochemical changes7862.

E-mail address: [email protected] (J.H. Bassman). with concomitant changes in photosynthesis and carbon

1011-1344/02/$ – see front matter 2002 Elsevier Science B.V. All rights reserved.PI I : S1011-1344( 02 )00230-0

126 J.M. Warren et al. / Journal of Photochemistry and Photobiology B: Biology 66 (2002) 125 –133

allocation. Very few studies have been conducted in the USA (468449 N, 1178109 W, elevation 777 m). Three treefield under realistic UV-B irradiation exposure regimes species were evaluated over a 3-year period: Pinus ponder-[16]. In these studies, the main effect of enhanced UV-B osa Dougl. ex Laws., Pseudotsuga menziesii var. glaucaradiation appears to be subtle changes in carbon partition- (Beissn.) Franco and Quercus rubra L. These species wereing rather than massive growth effects (e.g. Refs. [16–19]). selected because of their contrasting leaf morphology andOf substantial significance is the shift in carbon allocation shoot growth characteristics. They represent both angio-from growth pools to secondary metabolic pathways whose sperms and gymnosperms and three different genera.products may act in a protective or defensive capacity. Leaves of angiosperms do not screen UV-B radiation asIndeed, the production of phenolics and other UV-absorb- effectively as do the needles of gymnosperms [34,35]. Ining compounds appears to be the most consistent plant angiosperms, UV-absorbing compounds occur primarily inresponse to enhanced UV-B radiation [16]. the vacuoles of epidermal cells, allowing considerable

UV-B radiation stimulates gene transcription and expres- penetration through anticlinal cell walls to underlyingsion of key enzymes in the flavonoid biosynthetic pathway mesophyll tissues. In gymnosperms, UV-absorbing com-[20–22]. These enzymes lead to the induction of phenolic pounds are found both in the vacuoles and conjugated inUV-absorbing compounds (primarily in epidermal tissue) cell walls, resulting in a more continuous UV screen.[23,24] that reduce penetration of UV radiation to sensitive Consequently, a large fraction of incoming UV-B radiationmesophyll tissues, thus mitigating potential damage to is attenuated by the epidermis [35].plant tissue under UV-B radiation stress [25,26]. An The leaf structure of Q. rubra is typical of woodyimportant adjunct to this role is that many of the phenolic angiosperms, i.e. leaves are laminar with an upper epi-secondary metabolites induced by UV radiation also act as dermis, palisade parenchyma, spongy parenchyma, andantioxidants and anti-herbivore compounds, and affect lower epidermis. Quercus rubra is semi-determinate ordecomposition. Consequently, UV-B radiation effects at multiple flushing, so a particular compliment of leavesthe subcellular level could be translated into significant may be exposed to high levels of UV-B radiation for longeffects at higher trophic levels within ecosystems. periods of time. In P. ponderosa, the epidermis is com-

Many studies have reported UV-induced increases in prised of two layers, the internal one being the hypo-absorbance of methanol-soluble extracts from leaves, dermis. The mesophyll lacks palisade parenchyma and thepresumed to be primarily UV-absorbing flavonoids (recent- arrangement of this tissue is homogeneous. The leaves ofly reviewed in Ref. [27]). Comparatively little effort has P. menziesii are unique because they are laterally flattenedbeen devoted to isolation, identification, and quantification and their internal anatomy has attributes of both angio-of specific absorption profiles of these compounds [28,29], sperms and members of the Pinaceae. The hypodermalespecially in UV-B-treated tree species [30,31]. It is layer is present, but may be poorly developed and dis-important to do so because individual flavonoids, even continuous. The mesophyll is differentiated into palisadeclosely related molecules, have varying absorption spectra parenchyma and spongy parenchyma in an arrangementin the UV wavebands and do not attenuate UV-B radiation similar to angiosperm leaves.to the same degree [30]. Also, different flavonoids may Geographic origin of the seed was as follows: P.increase or decrease with enhanced UV-B radiation [31] ponderosa north of Spokane, WA, USA (488209 N, 1178309

and some may offer more protection than others [32]. W, 914 m); P. menziesii west of Yakima, WA, USA (468409

Thus, reports of UV-B radiation damage, despite the N, 1218109 W, 1219 m); Q. rubra from bulked sources inpresence of inducible flavonoids, may result from poor Pennsylvania, USA (|418 N, 788 W). Seed was coldabsorption characteristics as well as spatial variation in stratified according to procedures as outlined by the Unitedthese compounds [32,33]. States Department of Agriculture Forest Service [36].

The objectives of our study were to quantify the specific, Plastic pots (3 l) were filled with either a standard nursery-direct effects of long-term exposure to enhanced UV-B potting medium comprised of peat and vermiculite (coni-radiation on foliar flavonoid quantity and composition in fers) or a topsoil:compost:sand (1:1:1) mix. Stratifiedseveral important North American forest tree species with conifer seeds were sown 3–5 mm into the plantingcontrasting leaf morphologies and shoot growth patterns. medium and covered with 10 mm of white sand toOur hypotheses were that supplemental UV-B radiation minimize infection from damping-off fungi (e.g.would (a) increase total flavonoids in all species and (b) Phytophthora spp.). Quercus rubra seeds were sown 25alter flavonoid composition, and that (c) accumulation mm into the soil medium. Pots containing seeds were thenpatterns would be species- and leaf age-specific. placed under the UV-B lamps and seeds were germinated

under the respective treatments (see below). Pots wereburied to ground level beneath lamp frames and mulched

2. Materials and methods with sand to attain near-ambient soil temperatures. Ap-proximately 100 individual plants for each species for each

2.1. Plant materials and growth conditions treatment were established, each plant serving as a repli-cate. Plants were watered as necessary to maintain soil

The study was conducted in the field at Pullman, WA, near field capacity and fertilized at regular intervals with

J.M. Warren et al. / Journal of Photochemistry and Photobiology B: Biology 66 (2002) 125 –133 127

N, P, K (20-20-20) commercial liquid fertilizer containingmicronutrients. The nutrient solution for conifers wasacidified as described in Wenny and Dumroese [37].During the dormant period after the first growing season,plants were repotted into 11-l containers.

2.2. UV-B radiation treatments

Two treatments were examined: ambient biologicallyeffective UV-B radiation (UV-B ; 13) or twice-ambientBE

(23) UV-B radiation. A modulated, constant additionBE

system (Ashurst Designs, Logan, UT, USA), similar to thatdescribed by Caldwell et al. [38], was used to provide

Fig. 1. Spectral irradiance under ambient (13) and twice-ambient (23)enhanced UV-B radiation. Biologically effective UV-Bbiologically effective UV-B radiation measured beneath lamp frames of

radiation (UV-B ) was determined based upon theBE the modulated system at solar noon on the summer solstice in Pullman,generalized plant action spectrum of Caldwell [39], nor- WA, USA.Values were determined using a spectroradiometer with 0.5 nm

discrimination (Model 754, Optronics Laboratories, Orlando, FL, USA).malized to one at 300 nm. The system tracked ambientUV-B radiation levels and adjusted lamp output to com-pensate for solar angle, cloud cover, filter degradation,lamp aging and temperature. Our system was modified in ambient UV-B . The 23 UV-B radiation treatmentBE BE

from earlier versions so that the system could be operated was maintained based on below rack ambient radiationduring cold weather. Fluorescent lamps (Q-Panel UV-B- conditions and ranged from 1.6 to 2.2 times ambient solar313, Cleveland, OH, USA) supplied UV-B radiation. The UV-B throughout the day. The system was seasonallyBE

lamp fixtures were mounted in 1.232.4 m metal frames recalibrated to produce 2.0 times ambient conditions atthat were adjusted to maintain a constant 0.35 m height solar noon.above the plant canopies. The long axis of each frame wasoriented east–west. Lamp spacing was 0.35 m. Six lamp 2.3. Flavonoid extraction and HPLC analysisracks, each consisting of eight lamps, were used to providethe different UV-B radiation treatments. For the enhanced Conifer blocks were sampled in October at the end ofUV-B radiation treatments, lamps were filtered with 0.13- each growing season. Quercus rubra was sampled duringmm-thick cellulose diacetate film (transmission $292 nm). the third growing season in mid-July, just as the secondIn the control treatments, lamps were filtered with 0.13- growth flush was emerging. Main stem foliage was har-mm-thick polyester film (transmission $320 nm), such vested from four randomly selected trees per treatment plotthat plants received only ambient levels of solar UV-B and separated into cohorts (1- or 2-year-old foliage for theradiation. Filters required replacement at about 3–4-week conifers, first or second growth flush for the angiosperm).intervals. The lamp system was monitored using the Projected leaf area of each cohort of leaves was quantifiedOptronics Model 754 spectroradiometer (0.5-nm discrimi- using a leaf area meter (Delta-T Devices, Cambridge, UK).nation), which was calibrated periodically against a 200-W The foliage was immediately frozen in liquid nitrogen,NIST traceable standard lamp (OL 752-10). Before each then lyophilized, weighed, ground and stored at 240 8Cuse, wavelength alignment was checked against mercury until biochemical analysis. Based on foliar weight and leaf

22vapor emission lines (OL 752-150). Continuous moni- area, specific leaf masses (SLM; g cm ) were calculatedtoring of solar UV radiation, as input for the modulated for every tree sampled to express the flavonoid response onsystem, was made using a calibrated R-B meter (Solar a leaf area basis. Leaf area was not quantified in 2-year-oldLight Co., PA, USA). foliage of P. menziesii in the third year; SLM was assumed

The UV-B radiation treatments were initiated prior to to be 11% greater than 1-year-old foliage, based onseedling germination before the first growing season, and previous SLM increases with age and maximal values ofexcept during one winter season (in which the system SLM sampled.failed due to environmental conditions), treatments were Modified procedures of Miller [40] were used to purifymaintained continuously through the entire study. and quantify foliar flavonoids. Lyophilized, ground plant

Ambient UV-B irradiance for Pullman, WA, USA was tissue (0.500 g) was blended with an acidified extractionBE

measured 2 m above the ground at the study area and 0.35 solution (80% ethanol, 1% formic acid) in a homogenizerm below the lamps (Fig. 1) at hourly intervals on the until tissue was thoroughly macerated. The sample wassummer solstice. Open measurements resulted in a daily extracted (15 ml at 60 8C for 30 min with periodic

22dose rate of 6.72 kJ m UV-B . Measurements taken mixing); the heat and acidity used to reduce phenolic-BE

beneath the modulated system yielded daily dose rates on glycosides to their respective aglycones to allow for22 22the solstice of 5.94 kJ m (13) and 11.72 kJ m (23): adequate compound resolution. After filtering and rinsing,

shading by the lamp frames resulting in the 12% decrease the ethanol was removed under vacuum at 60 8C and the


sample cleaned three times with diethyl ether (to removethe lipids and non-polar pigments), each time retaining theaqueous fraction. The sample was then solvent-partitionedthree times with ethyl acetate and a phenolic extractionsolution (20% ammonium sulfate, 20% ethanol, 2%metaphosphoric acid) into the aqueous phase (antho-cyanins) and the acetate phase (flavonoids). The acetatewas removed under vacuum at 60 8C, leaving a sticky,yellow residue. The concentrated flavonoids were dis-solved in 2 ml of methanol (MeOH) and stored below 0 8Cuntil analysis.

Flavonoid extracts (10 ml) were injected onto a HPLCsystem (Varian, Walnut Creek, CA, USA) equipped with a5m C column (25034.6 mm with detection at 254 nm).18

The mobile phase consisted of a solvent gradient of80:10:10 (distilled water, 0.5% phosphoric acid, MeOH)increased linearly to 20:10:70 over 40 min, maintained at20:10:70 for 5 min, cleaned with 100% MeOH for 13 min,and then re-equilibrated to 80:10:10 for 2 min, all at a flow

21rate of 1.0 ml min .Quantification was accomplished with detection at 254

nm by comparing sample retention times and usingresponse factors generated from standards. Many flavonoidcompounds have a bimodal absorption pattern in MeOH,with the first peak in the UV-B or UV-C range (230–300nm) and the second peak often above 300 nm. Measure-ment at 254 nm resulted in a strong response for most ofthe standards used. Standards used included specific flavo-noids and related phenolic compounds (Fig. 2), all ob-tained from commercial sources (Indofine Chemical Com-pany, Belle Mead, NJ, USA; Sigma-Aldrich Corporation,St. Louis, MO, USA). Relative quantification of sampleflavonoid /phenolic compounds can be expressed as equiv-alent quantities of purified standards for each species /treatment on a leaf area basis. Flavonoid equivalents werecalculated by comparison of sample peak HPLC signalresponses at 254 nm with the response of pure standardsunder the same elution conditions.

2.4. Statistical analysis

Fig. 2. Typical C HPLC chromatographs for purified foliar flavonoids18Individual trees beneath each paired, calibrated lampof (a) Quercus rubra, (b) Pseudotsuga menziesii and (c) Pinus ponderosa

rack were treated as individual replicates. Analysis of are depicted by the solid curves. Overlaid are standards (dotted bars withvariance was performed on the total mean HPLC flavonoid numbers) representing the relative response factor (absorption) in metha-

nol at 254 nm using (a) 3 mg, (b) 0.15 mg or (c) 0.04 mg of each appliedarea output using the SAS statistical software packagestandard: (1) (1)catechin (eluting at 12.8 min), (2) chlorogenic acid(SAS Institute, Cary, NC, USA) to test for overall treat-(14.8), (3) (2)epicatechin (17.3), (4) 4-coumaric acid (20.6), (5) ferulicment differences across compound peaks. For the P.acid (21.9), (6) sinapic acid (22.5), (7) phloridzin (27.5), (8) rutin (28.1),

ponderosa and Q. rubra samples, four individual trees (9) t-cinnamic acid (29.9), (10a, b) myricetin (30.0, 34.2), (11) quercetinwere used for each plot. Low biomass of P. menziesii (34.0) and (12) kaempferol (37.9).foliage necessitated combining foliage into a single sampleof four trees (first year), two samples of two trees (secondyear) or not combining (third year) for each treatment. Aone-way analysis of variance (by compound) was con- paired t-tests [41]. Differences were considered significantducted to test each specific compound for significant at P#0.10.treatment effects. Equality of variance was tested using Due to the high cost of the modulated lamp system,


Table 1 treatment effect, and (2) equal variances existed between2 21Mean specific leaf area (SLA) (cm g ) for Q. rubra, P. menziesii and P. all trees within a species, as confirmed by paired t-tests.

ponderosa exposed to 13 and 23 ambient UV-B radiation treatmentsBE

Species Tree age Leaf age Specific leaf area2 21(cm g ) (6S.E.)

13 UV-B 23 UV-BBE BE 3. ResultsQ. rubra 3 years 1st flush 186.2 (611.4) 188.2 (64.2)

2 213 years 2nd flush 149.6 (68.5) 154.0 (64.7) Specific leaf area (cm g ) generally decreased (notP. menziesii 1 year 1 year 86.3 (64.9) 77.8 (62.9) statistically significant, NSS) in both conifers with 23

2 years 1 year 42.5 (63.3) 37.1 (62.4) ambient UV-B radiation treatments as compared with theBE2 years 2 years 68.9 (62.8) 65.4 (66.2)a a 13 ambient UV-B treatment, while no such trend wasBE3 years 2 years 38.3 33.5

exhibited by the angiosperm (Table 1). Typical HPLCP. ponderosa 1 year 1 year 56.9 (616.9) 43.1 (614.3)2 years 1 year 42.5 (63.3) 37.1 (62.4) chromatograms depicting the patterns of flavonoid /phen-2 years 2 years 50.4 (66.3) 40.5 (63.1) olic compound separation on an equivalent weight basis,

Individual tree SLA values were used to express the foliar HPLC and the relative response factors of the standards scaled toflavonoid quantitation on a leaf area basis. Shown by treatment are mean the individual species response, are illustrated in Fig. 2.SLA values with standard error (6S.E.) for each foliage class. Mean flavonoid concentrations for all samples, based ona These values represent an estimate of SLA based on an assumed

the average response of the 12 standards, yielded 268 mgdecrease in SLA from year-old leaves (see text description).22flavonoid equivalents (fe) cm for Q. rubra foliage, 96

22 22mg fe cm for P. menziesii and 46 mg fe cm for P.

treatment blocks (lamp frames) were not replicated. It was ponderosa with detection at 254 nm. The equivalences ofassumed that all differences between lamp frames resulted two specific standards, chlorogenic acid and rutin trihy-from the UV-B treatments. This assumption was based on drate (quercetin-3-rutinoside) are quantified by species-two conditions [42]: (1) the repeated calibrations and close treatment and age class for reference (Table 2). Overall,physical position of the paired treatment racks were similar the enhanced UV-B radiation treatments increased (.15%)enough to minimize the variance that would exist between foliar flavonoid and related phenolic compounds in P.true blocks (replicates) of one species, such that individual ponderosa and P. menziesii; however, these changes weretrees under a single lamp rack experienced primarily a not statistically significant. Q. rubra foliage showed a

Table 2Quantification of total HPLC separated flavonoids and related phenolics in foliage of Q. rubra, P. menziesii and P. ponderosa exposed to 13 and 23

UV-B radiation treatmentsBE

Species Tree Leaf UV-B Equivalence of applied standards at 254 nm Mean change in flavonoidsBE22age age treatment (mg standard equivalents cm dry foliage) with 23 UV-BBE

Chlorogenic acid Rutin (.15% difference) P.F

Q. rubra 3 years 1st flush 13 344 (654) 260 (640) 0 0.67All peaks 23 316 (672) 240 (654)

13 266 (636) 200 (628) 2 0.33Peaks 1–18 23 206 (642) 156 (630)

13 68 (618) 50 (614) 1 0.11Peaks 19–24 23 100 (630) 74 (622)

P. menziesii 2 years 1 year 13 116 (620) 88 (614) 2 0.3023 74 (616) 56 (612)

2 years 2 years 13 120 (624) 90 (620) 1 0.7523 140 (620) 106 (616)

3 years 2 years 13 110 (612) 84 (68) 1 0.4523 138 (614) 106 (612)

P. ponderosa 1 year 1 year 13 44 (612) 33 (610) 0 0.9623 44 (68) 34 (66)

2 years 1 year 13 47 (612) 35 (610) 1 0.2723 56 (614) 42 (612)

2 years 2 years 13 65 (618) 49 (614) 1 0.3023 75 (616) 57 (612)

Partial HPLC responses of early- or late-eluting compounds (HPLC peaks) of Q. rubra foliage are included. The values are expressed as the equivalent22chlorogenic acid and rutin standard HPLC response in mg standard equivalents cm dry tissue (6S.E.). Mean increase or decrease (.15%) in total

average concentration of flavonoids with 23 UV-B as compared with 13 UV-B is depicted by 1, 2, or 0 (no change); means were not significantlyBE BE

different (P,0.1), with P-values shown for tests of equal means.


distinct (NSS) compositional shift in flavonoid concen- cantly lower in the 23 treated tissue (Fig. 4a). Totaltrations (Table 2). For each species, specific flavonoid flavonoid concentration of the same 23 foliage cohort thecompounds exhibited statistically significant treatment next year was 89% higher than the 1-year-old 23 foliage.effects. Foliar flavonoids of this 2-year-old cohort after the third

treatment season were generally higher in the 23 ambient3.1. Quercus rubra UV-B radiation treatment than in the ambient treatment,BE

with a significant increase in three specific compoundsTotal flavonoid concentrations in foliage of Q. rubra (Table 2; Fig. 4b).

were more than twice that of the other species (by leafarea), however, the predominant compound was almost 20 3.3. Pinus ponderosatimes (by weight) that of any eluted compound from anyspecies. There were no statistically significant differences Total foliar flavonoids in P. ponderosa harvested afterbetween the 13 and 23 treatments in total flavonoids for the first growing season showed no significant change infully mature first-flush foliage after the third year of response to enhanced UV-B radiation, although one in-UV-B irradiance. Three individual compounds had sig- dividual compound decreased significantly under 23BE

nificantly lower and one compound significantly higher ambient UV-B radiation (Table 2; Fig. 5a). Foliage fromBE

concentrations under the 23 treatment (Fig. 3). In general, the same 23 cohort the following year had higher overallthere were fewer early-eluting flavonoid compounds flavonoid concentration (NSS), with two compounds ex-(peaks 1–18), and more late-eluting compounds (peaks hibiting significant increases (Fig. 5b). A new compound19–27) in the 23 UV-B radiation treatment (Table 2). appeared (eluting at 41.4 min) while another compoundBE

disappeared (12.9 min), possibly related to slight shifts in3.2. Pseudotsuga menziesii quantities of related phenolics. Foliage exposed to the 13

treatment for two growing seasons increased its totalNo clear patterns were exhibited by 1-year-old P. flavonoid quantity by 50%, and that of the 23 treatment

menziesii foliage emerging in the first and second year; increased by 70% over first-year levels. One-year-oldhowever, 2-year-old tissue exhibited an increase (NSS) in foliage sampled after the second year of treatment hadtotal foliar flavonoid levels under the 23 ambient UV-B higher (NSS) concentrations of flavonoids in the 23BE

radiation treatment (Table 2). Main stem foliage harvestedafter the first growing season had similar quantities offlavonoids in both treatments with 10 peaks higher in the13 treatment and 16 higher in the 23 treatment. In thesecond year, total 2-year-old tissue flavonoid concentra-tions were 15% higher than first year foliage. Flavonoidlevel in foliage of 23 treated P. menziesii that emerged inthe second year of treatment generally had lower (NSS)total flavonoid concentrations than in the ambient treat-ment (Table 2). Three specific compounds were signifi-

Fig. 4. Flavonoid and related phenolic compound peaks of 13 and 23

Fig. 3. Flavonoid and related phenolic compound peaks of 13 and 23 ambient UV-B irradiated Pseudotsuga menziesii foliage as eluted viaBE

ambient UV-B irradiated Quercus rubra foliage as eluted via reverse- reverse-phased C HPLC. Illustrated is the second foliage cohortBE 18

phased C HPLC. Illustrated is the first foliage flush just after second- followed through two treatment years: (a) year-old foliage, (b) 2-year-old18

flush emergence during the third year of treatments. Values represent the foliage. Values represent the mean (6S.E.) of n52 (2nd year) or 4 (3rdmean (6S.E.) of n54. Bars with different letters for the same elution year). Bars with different letters for the same elution time are significantlytime are significantly different (P#0.10). Standard error bars are absent different (P#0.10). Standard error bars are absent where single datawhere single data points are represented. points are represented.


rubra, and probable vacuolar isolation of flavonoids [35],resulted in much higher total foliar flavonoid concen-trations than the other species. Leaf trichomes in Quercuscould contribute an additional flavonoid component [43].Foliage of P. menziesii, whose needles have anatomicalattributes similar to both P. ponderosa and Q. rubra, hadtotal flavonoid concentrations intermediate between theother species. The conifers appeared to accumulate addi-tional flavonoids in response to enhanced UV-B radiation,but changes were not statistically significant. Quercusrubra shifted the composition of its constitutive pools offlavonoids, with no overall increase in total concentration.Other research with Q. rubra shows increased palisadeparenchyma thickness with UV-B treatment [44], whichcould affect spatial distribution of these flavonoids. Theability of conifers to attenuate and dissipate most UV-Bradiation in the epidermal layers (regardless of UV-B-induced changes in thickness), where flavonoids are moreevenly dispersed, should result in a lower total concen-tration of flavonoids necessary for protection [30,34,35].Our results for total flavonoid content of the three speciessubstantiate this observation (Fig. 2).

Leaf and plant age both affected the accumulation ofphenolic compounds in our study. In P. ponderosa, agingneedles accumulated higher concentrations of total foliarflavonoids, with the greatest changes attributed to only afew specific compounds. In P. sylvestris, leaf age also wasimportant in the accumulation of two specific flavonolmonoglycosides [30]. The similar relative flavonoid re-sponse between the P. ponderosa 1- and 2-year-old needleFig. 5. Flavonoid and related phenolic compound peaks of 13 and 23

ambient UV-B irradiated Pinus ponderosa foliage as eluted via reverse- cohorts after the second year of treatments (as opposed toBE

phased C HPLC. Two foliage cohorts are illustrated: (a, b) the first18 patterns in the first year) illustrates that seedling age is ascohort followed through two treatment years, and (c) the second cohort important as the foliage age in determining relative flavo-after one growing season. Values represent the mean (6S.E.) of n54.

noid composition (Fig. 5b and c).Bars with different letters for the same elution time are significantlyAs expected, there were specific UV-B-induced effectsdifferent (P#0.10).

on individual compounds for each species tested. In Q.rubra, however, there was a distinct, though statistically

ambient UV-B radiation treatment as compared with the insignificant, shift in flavonoid composition from early- toBE

ambient treatment (Table 2). Two specific compounds late-eluting compounds under enhanced UV-B radiation. Inshowed significant increases with the 23 treatment (Fig. Populus trichocarpa under the same laboratory conditions,5c). phenolic glycosides elute from the HPLC column between

27.5 and 32 min [45], including rutin at 28.1 min (Fig. 2),which is the same range where the 23 UV-B radiation-BE

4. Discussion treated Q. rubra had increased compound presence. Thesephenolic compound shifts in Q. rubra from more polar to

The phenolic compound profiles of foliage from two less polar compounds (likely flavonoid glycosides) mayconifers and one angiosperm in this study confirm our result in stronger UV-B radiation absorbance potential, andhypothesis that these responses are species specific, and better protection for tissues. In Petunia spp., a UV-Bsuggest that there are different mechanisms acting to radiation-induced increase in the quercetin:kaempferolprotect foliage from enhanced UV-B radiation: leaf mor- ratio has been reported [46]. This slight shift in phenolicphology, foliar display and leaf age. Foliage of P. ponder- composition could allow for better absorption of UV-Bosa had the lowest concentration of flavonoids, likely due radiation and dissipation of damaging energy.to incorporation of these compounds in the epidermal cell The predominant component of purified P. menziesiiwalls and in vacuoles, the added hypodermal layer, and foliage eluted at 27.9 min (also probably a flavonoidmore vertically displayed foliage than the other species. glycoside), and had significantly higher concentrations inThe laminar and horizontally displayed foliage of Q. 23 treated foliage after the third year of treatments (Fig.


4b). The unexpected decline in 23 P. menziesii total Acknowledgementsphenolics seen in the previous year (Table 2, Fig. 4a) maybe a result of other external stressors. The terminal shoots This project was funded with the support of the Unitedin those plots were exposed through the snow in the first States Environmental Protection Agency Science toyear causing extensive tip mortality during a particularly Achieve Results (STAR) graduate fellowship program.cold winter, which may have influenced the phytochemis- Funding was also provided by the Cooperative Statetry of new growth, confounding the results. Exposure to Research Service, United States Department of Agricultureenhanced UV-B radiation induced increases in specific under agreement 94-37100-0312 made to J.H.B., G.E.E.flavonoids, hydroxycinnamic acids (including chlorogenic and R.R., and by the Agricultural Research Center, Collegeacid) and proanthocyanidins in Betula spp. (another an- of Agriculture and Home Economics, Washington Stategiosperm) [47]. These were found to be species- and University under Project 0113 (J.H.B.). Any opinions,population-specific. Our results agree with other studies findings, or recommendations are those of the authors andthat suggest that the specific composition of phenolics are do not necessarily reflect the view of the United Statesaffected by enhanced UV-B radiation, and that these effects Department of Agriculture.may be as important as phenolic concentration in determin-ing UV-radiation screening, tissue palatability or otherbiological activities.

ReferencesThe specificity of responses among the species dictatethat we need to understand the relationships between these

[1] R.A. Kerr, The ozone hole reaches a new low, Science 262 (1993)compounds more clearly; how differences in concentra-501.tions of individual compounds contribute to UV-radiation

[2] G.B. Seckmeyer, B. Mayer, R. Erb, G. Bernhard, UV-B in Germanyabsorption and dissipation, and herbivore and higher higher in 1993 than in 1992, Geophys. Res. Lett. 21 (1994)trophic level interactions. In Picea abies, seasonal shifts of 577–580.epidermal flavonoids to more effective UV-B screening [3] C.S. Zerefos, A.F. Bais, C. Meleti, I.C. Ziomas, A note on the recent

increase of solar UV-B radiation over northern middle latitudes,compounds have been reported [48] with significantlyGeophys. Res. Lett. 22 (1995) 1245–1247.higher compound concentrations in UV-B-exposed tissue.

[4] J.R. Herman, P.K. Bhartia, J. Ziemke, Z. Ahmad, D. Larko, UV-BOther studies have also shown flavonoid-specific responses increases (1979–1992) from decreases in total ozone, Geophys. Res.to UV-B radiation in trees: e.g. Pinus sylvestris [23] and Lett. 23 (1996) 2117–2120.Betula pendula [47]. In our study, all species did not [5] N.R.P. Harris, G. Ancellet, L.H.D.J. Bishop, J.B. Kerr, R.D.

McPeters, M. Prendez, W.J. Randel, J. Stahelin, B.H. Subbaraya, A.demonstrate the expected increase in total foliar phenolicVolz-Thomas, J. Zawodny, C.S. Zerefos, Trends in stratospheric andconcentrations reinforcing the evidence that total tissuetropospheric ozone, J. Geophys. Res. 102 (1997) 1570–1590.phenolic content does not provide enough information to

[6] J.W. Elkins, T.M. Thompson, T.H. Swanson, J.H. Butler, B.D. Hall,adequately describe UV-B-induced changes in specific S.O. Cummings, D.A. Fisher, A.G. Raffo, Decrease in the growthphenolics. Indeed, as seen in Q. rubra, the total con- rates of atmospheric chlorofluorocarbon-11 and chlorofluorocarbon-

12, Nature 364 (1993) 780–783.centration was not affected, however, there were nonsig-[7] Anonymous, Antarctic ozone hole shows beginning of recoverynificant changes in groups of compounds. Crude tissue

process, Bull. Am. Meteorol. Soc. 81 (2000) 116.MeOH extracts and spectrophotometric measurements can[8] M. Schrope, Successes in fight to save ozone layer could close holes

provide a measure of total UV absorption, but provide little by 2050, Nature 408 (2000) 627.information concerning the specific compounds affecting [9] P. Aldhous, Global warming could be bad news for Arctic ozone

layer, Nature 404 (2000) 531.that absorption.[10] D.L. Hartmann, J.M. Wallace, V. Limpasuvan, D.W.J. Thompson,In this study, foliar flavonoid composition and con-

J.R. Holton, Can ozone depletion and global warming interact tocentration was affected by enhanced UV-B radiation, andproduce rapid climate change?, Proc. Natl. Acad. Sci. USA 97

those effects were age- and species-dependent. Flavonoids (2000) 1412–1417.and related phenolic compounds have many biological [11] M. Tevini, A.H. Teramura, UV-B effects on terrestrial plants,functions in plants, in addition to the attenuation of Photochem. Photobiol. 50 (1989) 479–487.

[12] M.M. Caldwell, S.D. Flint, Stratospheric ozone reduction, solardamaging UV-B radiation [27]. When plants are exposed toUV-B radiation and terrestrial ecosystems, Climatic Change 28UV-B radiation, vegetative and reproductive growth may(1994) 375–394.

be slightly reduced as a consequence of increased carbon [13] V.C. Runeckles, S.V. Krupa, The impact of UV-B radiation andpartitioning into these relatively stable secondary com- ozone on terrestrial vegetation, Environ. Pollut. 83 (1994) 191–213.pounds. Alteration in species-specific flavonoid accumula- [14] J.H. Sullivan, J. Rozema, UV-B effects on terrestrial plant growth

and photosynthesis, in: J. Rozema (Ed.), Stratospheric Ozonetion resulting from enhanced UV-B radiation may result inDepletion: the Effects of Enhanced UV-B Radiation on Terrestrialchanges in tissue attractiveness or palatability to herbi-Ecosystems, Backhuys, Leiden, The Netherlands, 1999, pp. 39–57.

vores, changes in litter decomposition processes, and [15] J. Rozema, UV-B radiation and terrestrial ecosystems: processes,ultimately changes in competition and ecosystem structure structure and feedback loops, in: J. Rozema (Ed.), Stratospheric[12,15,49]. Ozone Depletion: the Effects of Enhanced UV-B Radiation on


Terrestrial Ecosystems, Backhuys, Leiden, The Netherlands, 1999, [32] L.G. Landry, C.S. Chapple, R.L. Last, Arabidopsis mutants lackingpp. 101–115. phenolic sunscreens exhibit enhanced ultraviolet-B injury and

[16] P.S. Searles, S.D. Flint, M.M. Caldwell, A meta-analysis of plant oxidative damage, Plant Physiol. 109 (1995) 1159–1166.field studies simulating stratospheric ozone depletion, Oecologia 127 [33] T.A. Day, T.C. Vogelmann, Alterations in photosynthesis and(2001) 1–10. pigment distributions in pea leaves following UV-B exposure,

[17] J.H. Sullivan, A.H. Teramura, The effects of ultraviolet-B radiation Physiol. Plant. 94 (1995) 433–440.on loblolly pine 2. Growth of field-grown seedlings, Trees 6 (1992) [34] T.A. Day, Relating UV-B radiation screening effectiveness of115–120. foliage to absorbing-compound concentration and anatomical

[18] A. Singh, Growth, physiological, and biochemical responses of three characteristics in a diverse group of plants, Oecologia 95 (1993)tropical legumes to enhanced UV-B radiation, Can. J. Bot. 74 (1996) 542–550.135–139. [35] T.A. Day, G. Martin, T.C. Vogelmann, Penetration of UV-B radia-

[19] C.T. Ruhland, T.A. Day, Effects of ultraviolet-B radiation on leaf tion in foliage: evidence that the epidermis behaves as a non-elongation, production and phenylpropanoid concentrations of De- uniform filter, Plant Cell Environ. 16 (1993) 735–741.schampsia antarctica and Colobanthus quitensis in Antarctica, [36] USDA Forest Service, Seeds of Woody Plants in the United StatesPhysiol. Plant. 109 (2000) 244–251. [Handbook 450], US Department of Agriculture, Washington, DC,

[20] E. Wellmann, UV dose-dependent induction of enzymes related to 1974.flavonoid biosynthesis in cell suspension cultures of parsley, FEBS [37] D.L. Wenny, R.K. Dumroese, A Growing Regime for Container-Lett. 51 (1975) 105–107. grown Douglas-fir Seedlings, University of Idaho (College of˚[21] A. Strid, Alteration in expression of defence genes in Pisum sativum Forestry, Wildlife and Range Sciences), Moscow, ID, 1992.after exposure to supplementary ultraviolet-B radiation, Plant Cell [38] M.M. Caldwell, W.G. Gold, G. Harris, C.W. Ashurst, A modulatedPhys. 34 (1993) 949–953. lamp system for solar UV-B (280–320 nm) supplementation studies

[22] G. Fuglevand, J.A. Jackson, G.I. Jenkins, UV-B, UV-A and blue in the field, Photochem. Photobiol. 37 (1983) 479–485.light signal transduction pathways interact synergistically to regulate [39] M.M. Caldwell, Solar UV irradiation and the growth and develop-chalcone synthase gene expression in Arabidopsis, Plant Cell 8 ment of higher plants, in: A.C. Giese (Ed.), Photophysiology, Vol. 6,(1996) 2347–2357. Academic Press, New York, 1971, pp. 131–177.

¨[23] J. Schnitzler, T.P. Jungblut, C. Feicht, M. Kofferlein, C. Langebar- [40] T.W. Miller, Comparative Studies of the Relationship betweentels, W. Heller, H. Sandermann Jr., UV-B induction of flavonoid Aroma Volatile Biogenesis and Anthocyanin Synthesis in ‘Delicious’biosynthesis in Scots pine (Pinus sylvestris L.) seedlings, Trees 11 Apples, University of Idaho, Moscow, ID, 1995, Ph.D. dissertation.(1997) 162–168. [41] S.D. Schlotzhauer, R.C. Littell, SAS System for Elementary Statisti-

[24] R. Robberecht, M.M. Caldwell, Protective mechanisms and acclima- cal Analysis, SAS Institute, Cary, NC, 1991.tion to solar ultraviolet-B radiation in Oenothera stricta, Plant Cell [42] S.H. Hurlbert, Pseudoreplication and the design of ecological fieldEnviron. 6 (1983) 477–485. experiments, Ecol. Monogr. 54 (2) (1984) 187–211.

[25] E.M. Middleton, A.H. Teramura, The role of flavonol glycosides [43] H. Skaltsa, E. Verykokidou, C. Harvala, G. Karabourniotis, Y.and carotenoids in protecting soybean from ultraviolet-B damage, Manetas, UV-B protective potential and flavonoid content of leafPlant Physiol. 103 (1993) 741–752. hairs of Quercus ilex, Phytochemistry 37 (1994) 987–990.

[26] J.J. Sheahan, Sinapapte esters provide greater UV-B attenuation than [44] L.M. Nagel, J.H. Bassman, G.E. Edwards, R. Robberecht, V.R.flavonoids in Arabidopsis thaliana (Brassicaceae), Am. J. Bot. 83 Franceshi, Leaf anatomical changes in Populus trichocarpa, Quer-(1996) 679–686. cus rubra, Pseudotsuga menziesii and Pinus ponderosa exposed to

[27] B. Meijkamp, R. Aerts, J. van de Staaij, M. Tosserams, W.H.O. enhanced ultraviolet-B radiation, Physiol. Plant. 104 (1998) 385–Ernst, J. Rozema, Effects of UV-B on secondary metabolites in 396.plants, in: J. Rozema (Ed.), Stratospheric Ozone Depletion: the [45] J.M. Warren, unpublished data, 2001.Effects of Enhanced UV-B Radiation on Terrestrial Ecosystems, [46] K.G. Ryan, K.R. Markham, S.J. Bloor, J.M. Bradley, K.A. Mitchell,Backhuys, Leiden, The Netherlands, 1999, pp. 71–79. B.R. Jordan, UVB radiation induced increase in quercetin:kaem-

¨[28] D. Strack, J. Heilemann, M. Momken, V. Wray, Cell wall-conjugated pferol ratio in wild-type and transgenic lines of Petunia, Photochem.phenolics from coniferae leaves, Phytochemistry 27 (1988) 3517– Photobiol. 68 (3) (1998) 323–330.3521. [47] A. Lavola, Accumulation of flavonoids and related compounds in

¨[29] R. Julkunen-Tiitto, M. Rousi, J. Bryant, S. Sorsa, M. Keinanen, H. birch induced by UV-B irradiance, Tree Physiol. 18 (1998) 53–58.Sikanen, Chemical diversity of several Betulaceae species: com- [48] R.J. Fischbach, B. Kossmann, H. Panten, R. Steinbrecher, W. Heller,parison of phenolics and terpenoids in northern birch stems, Trees: H.K. Seidlitz, H. Sandermann, N. Hertkorn, J.-P. Schnitzler, Season-Struct. Funct. 11 (1996) 16–22. al accumulation of ultraviolet-B screening pigments in needles of

[30] T.P. Jungblut, J.-P. Schnitzler, W. Heller, N. Hertkorn, J.W. Metzger, Norway spruce (Picea abies (L.) Karst.), Plant Cell Environ. 22W. Szymczak, H. Sandermann Jr., Structures of UV-B-induced (1999) 27–37.

´sunscreen pigments of the Scots pine (Pinus sylvestris L.), Angew. [49] M.C. Rousseaux, C.L. Ballare, A.L. Scopel, P.S. Searles, M.M.Chem. 34 (1995) 312–314. Caldwell, Solar ultraviolet-B radiation affects plant–insect interac-

[31] A. Lavola, R. Julkunen-Tiitto, P. Aphalo, T. de la Rosa, T. Lehto, tions in a natural ecosystem of Tierra del Fuego (southern Argen-The effect of u.v.-B radiation on u.v.-absorbing secondary metabo- tina), Oecologia 116 (1998) 528–535.lites in birch seedlings grown under simulated forest soil conditions,New Phytol. 137 (1997) 617–621.

alteration of foliar flavonoid chemistry induced by enhanced uv-b radiation in field-grown pinus...

Documents