Leaf trait dissimilarities between Dutch elm hybrids with a ...

Leaf trait dissimilarities between Dutch elm hybrids with acontrasting tolerance to Dutch elm disease

Jaroslav Durkovic1,*, Ingrid Canova1, Rastislav Lagana2, Veronika Kucerova3, Michal Moravcık1,Tibor Priwitzer4, Josef Urban5, Milon Dvorak5 and Jana Krajnakova5

1Department of Phytology, Technical University, 960 53 Zvolen, Slovakia, 2Department of Wood Science, Technical University,960 53 Zvolen, Slovakia, 3Department of Forest Protection and Game Management, Technical University, 960 53 Zvolen,Slovakia, 4Forest Research Institute, National Forest Centre, 960 92 Zvolen, Slovakia and 5Faculty of Forestry and Wood

Technology, Mendel University, 613 00 Brno, Czech Republic* For correspondence. E-mail [email protected]

Received: 20 September 2012 Returned for revision: 11 October 2012 Accepted: 13 November 2012 Published electronically: 21 December 2012

† Background and Aims Previous studies have shown that Ophiostoma novo-ulmi, the causative agent of Dutchelm disease (DED), is able to colonize remote areas in infected plants of Ulmus such as the leaf midrib and sec-ondary veins. The objective of this study was to compare the performances in leaf traits between two Dutch elmhybrids ‘Groeneveld’ and ‘Dodoens’ which possess a contrasting tolerance to DED. Trait linkages were alsotested with leaf mass per area (LMA) and with the reduced Young’s modulus of elasticity (MOE) as a resultof structural, developmental or functional linkages.† Methods Measurements and comparisons were made of leaf growth traits, primary xylem density components,gas exchange variables and chlorophyll a fluorescence yields between mature plants of ‘Groeneveld’ and‘Dodoens’ grown under field conditions. A recently developed atomic force microscopy technique, PeakForcequantitative nanomechanical mapping, was used to reveal nanomechanical properties of the cell walls oftracheary elements such as MOE, adhesion and dissipation.† Key Results ‘Dodoens’ had significantly higher values for LMA, leaf tissue thickness variables, trachearyelement lumen area (A), relative hydraulic conductivity (RC), gas exchange variables and chlorophyll a fluores-cence yields. ‘Groeneveld’ had stiffer cell walls of tracheary elements, and higher values for water-use efficiencyand leaf water potential. Leaves with a large carbon and nutrient investment in LMA tended to have a greater leafthickness and a higher net photosynthetic rate, but LMA was independent of RC. Significant linkages were alsofound between the MOE and some vascular traits such as RC, A and the number of tracheary elements per unitarea.† Conclusions Strong dissimilarities in leaf trait performances were observed between the examined Dutch elmhybrids. Both hybrids were clearly separated from each other in the multivariate leaf trait space. Leaf growth,vascular and gas exchange traits in the infected plants of ‘Dodoens’ were unaffected by the DED fungus.‘Dodoens’ proved to be a valuable elm germplasm for further breeding strategies.

Key words: Adhesion, atomic force microscopy, gas exchange, leaf mass per area, modulus of elasticity,Ophiostoma novo-ulmi, tracheary element.

INTRODUCTION

Ophiostoma novo-ulmi, the causative agent of current Dutchelm disease (DED) pandemics, is highly pathogenic to bothnative European and North American elm trees. This ascomy-cetous fungus is polytypic, spreading in the form of two sub-species, subsp. novo-ulmi and subsp. americana, previouslyreferred to as the Eurasian and North American races, respect-ively (Brasier and Kirk, 2001). A rapid emergence of hybridsbetween these two subspecies has recently been reported, andit is likely that complex hybrid swarms are now expandingacross the European continent (Brasier and Kirk, 2010). Theinitial elm breeding programmes, launched in TheNetherlands, were focused on the identification of resistantelms to replace the popular but susceptible cultivar Ulmus ×hollandica ‘Belgica’, and emphasized the native Europeanelms, especially U. glabra and U. minor. Asian elms,

particularly U. wallichiana, proved to be a useful additionalsource of DED resistance genes with the advent of thesecond disease pandemics in Europe during the 1970s(Heybroek, 1983; Smalley and Guries, 1993). The Dutch elmcultivar releases of the 1960s and 1970s, such as‘Groeneveld’, ‘Commelin’, ‘Dodoens’, ‘Lobel’, ‘Plantyn’and others, were widely planted in western Europe and showa varying degree of resistance to O. novo-ulmi isolates. Forbetter control of this disease, current elm breeding pro-grammes also integrate biotechnological approaches (Pijutet al., 1990; Fenning et al., 1996; Corredoira et al., 2002).

Fungal metabolites associated with vascular wilt diseaseaffect the plasma membrane function, and cause an interfer-ence with the stomatal regulation of transpiration and a reduc-tion of flow through the stem due either to vascular plugging orto an increase in the viscosity of the xylem sap (Van Alfen andTurner, 1975; Van Alfen, 1989). The fungus produces

# The Author 2012. Published by Oxford University Press on behalf of the Annals of Botany Company. All rights reserved.

For Permissions, please email: [email protected]

Annals of Botany 111: 215–227, 2013

doi:10.1093/aob/mcs274, available online at www.aob.oxfordjournals.org

Downloaded from https://academic.oup.com/aob/article-abstract/111/2/215/255587by gueston 07 April 2018

mailto:[email protected]

hydrophobin cerato-ulmin (a parasitic fitness factor) (Templeet al., 1997), phytotoxic peptidorhamnomannan (Strobelet al., 1978; Sticklen et al., 1991), cell wall-degradingenzymes such as xylanases, laccases (Binz and Canevascini,1996a, b), exo-glycanases and glycosidases (Svaldi andElgersma, 1982), and tissue-invading structures, which arethought to be involved in cavitation of the water column andalteration of parenchyma cells (Ouellette et al., 2004).Several species of elm bark beetles, primarily in the genusScolytus and Hylurgopinus, have been recognized as themajor transmission vectors of DED (Webber, 2004).

Leaf traits are often closely associated with plant growth, sur-vival and light requirement, and thus may be good predictors ofplant performance across a diverse range of plant communities(Poorter and Bongers, 2006; Sterck et al., 2006). Althoughleaves vary considerably in area, thickness, shape, nutrient con-centrations and their capacity for gas exchange, the intercorrel-ation of many traits places some bounds on this diversity,particularly among those related to carbon economy, includingleaf mass per area (LMA), leaf life span, nitrogen concentrationper mass and net photosynthetic rate per mass. The LMA is akey trait in plant growth and a significant descriptor of plantstrategies that can contribute to the success of a given plantspecies in the field (Westoby et al., 2002; Poorter et al.,2009). In its own right, LMA is a hub trait interlinked with adisproportionate number of other traits (Sack et al., 2003;Sack and Holbrook, 2006). Variation in leaf traits of Ulmusspecies has been the focus of several studies with respect totheir resistance to elm leaf beetle defoliation (Young andHall, 1986; Bosu and Wagner, 2007, 2008). However, not sowidely covered is the research aimed at leaf trait variation inelms with a varying degree of resistance to DED.

The mechanical properties of the cellular microenviron-ment, notably its rigidity and stiffness, play a critical regula-tory role for a variety of fundamental cell behaviours andresponses (Janmey et al., 2009). In the case of trachearyelements of leaf primary xylem – highly specialized cellsfor transporting water and nutrients to the leaf lamina –knowledge of the in situ stiffness of the cell wall quantifiedby the modulus of elasticity (MOE) should be of great import-ance because it can be used either to assess stress values result-ing in cell wall deformations or to evaluate the risk oftracheary element implosion when an embolism spreads andcavitation occurs under stressful conditions. Although the in-ternal organization of the cell wall architecture of trachearyelements has been the focus of structural studies (Nakashimaet al., 1997; Lacayo et al., 2010), information about the nano-mechanical properties of the cell walls of tracheary elements islargely lacking.

The objective of this study was to compare the performancesin leaf traits, including an in situ assessment of nanomechani-cal properties of cell walls of tracheary elements, between twoDutch elm hybrids (‘Groeneveld’ and ‘Dodoens’) whichpossess a contrasting tolerance to DED. ‘Dodoens’ plants tol-erant to DED respond to O. novo-ulmi inoculation with analtered pattern of secondary xylem annual ring organization.These plants tolerant to DED show the formation of many nar-rowed vessels with small lumen areas in the successive annualrings (Fig. 1A–C). Thus, we hypothesized that performancesin leaf traits of infected plants of ‘Dodoens’ related to leaf

growth (leaf area, LMA), nanomechanical properties ofprimary xylem cell walls (MOE), and gas exchange andchlorophyll a fluorescence (net photosynthetic rate, transpir-ation, stomatal conductance, variable-to-initial fluorescenceratio) will be significantly decreased due to the altered morph-ology of vessels and the occasional occurrence of fungalhyphae in water-conducting cells, when compared with thenon-infected plants. Additionally, we tested trait relationshipswith LMA and MOE. We expected correlations with thesetraits as a result of structural, developmental or functional lin-kages, including leaf dimensions, net photosynthetic rate andother vascular traits (Sack and Holbrook, 2006; Dunbar-Coet al., 2009; Bartlett et al., 2012; Durkovic et al., 2012).

MATERIALS AND METHODS

Plant material and study site

The experiments were conducted on clonally micropropagated(Krajnakova and Longauer, 1996), mature flowering plantsof the Dutch elm hybrid cultivars ‘Groeneveld’ [(Ulmus ×hollandica 49) × U. minor ssp. minor 1] and ‘Dodoens’(open pollinated U. glabra ‘Exoniensis’ × U. wallichianaP39) growing in the experimental field plot at Banska Bela,Slovakia (48828′N, 18857′E, 590 m a.s.l.). According to themeteorological station at Arboretum Kysihybel in BanskaStiavnica (540 m a.s.l.), located 3.6 km south-west of thestudy site, the climate of the area is characterized by a meanannual temperature of 7.7 8C and a mean annual precipitationof 831 mm. The study site soil which has a silt loam texture isidentified as an Eutric Cambisol formed from theslope deposits of volcanic rocks (andesite and pyroclasticmaterials).

Fungus identification and inoculation, and plant source selection

Five plants of each cultivar, at least 10 years of age, wereinoculated with O. novo-ulmi ssp. americana × novo-ulmiisolate M3 according to the procedure of Solla et al. (2005).The spore suspension (1 × 107 spores mL21) was inoculatedinto the current annual ring, 20 cm above the base of thestem. Hybrid isolate M3 belongs to mating type B, and wasisolated from an infected elm tree in Brno, Czech Republic.This isolate proved to be ssp. americana in the fertility testand had a cerato-ulmi (cu) gene profile of ssp. americanabut a ssp. novo-ulmi colony type (col1) gene profile (Konradet al., 2002; Dvorak et al., 2007).

The infected plants of ‘Groeneveld’ do not show any toler-ance toward O. novo-ulmi ssp. americana × novo-ulmi, andthey die at the end of the growing season after a fungus inocu-lation (Dvorak et al., 2009). Thus, leaf traits coming from thisplant source were not investigated in our study. However, theinfected plants of ‘Dodoens’ do show a high tolerance toDED. Although scanning electron microscopy (SEM) con-firmed the occasional occurrence of fungal hyphae insideboth the secondary xylem vessels of successive annual ringsafter a fungus inoculation (Fig. 1A) and the primary xylemtracheary elements of current-year leaves (Fig. 1F), novisible evidence of DED was observed on the infected plantsof ‘Dodoens’ in the following growing seasons. Thereby,

Durkovic et al. — Leaf traits in Dutch elm hybrids with contrasting disease resistance216


due to their high tolerance to the pathogen, the infected plantsof ‘Dodoens’ were included in the experiments describedherein. SEM images of the plant material used in this studyare presented in Fig. 1D–L. The experiments were conductedon fully expanded leaves which were measured directly in thefield and sampled for the further laboratory procedures andanalyses in June 2011.

Scanning electron microscopy of leaf and wood samples

Leaf cross-sections from the centre of the leaf blade wereimmersed in 5 % glutaraldehyde in a 0.1 M cacodylate bufferat pH 7.0, dehydrated in ethanol and acetone, and dried inliquid CO2 using a Leica EM CPD030 critical point drier(Leica Microsystems, Wetzlar, Germany). Leaf and wood

A B C

D E F

G H I

J K L

FI G. 1. Scanning electron microscopy images of leaf and wood samples of infected plants of ‘Dodoens’ (A–F), non-infected plants of ‘Dodoens’ (G–I) andnon-infected plants of ‘Groeneveld’ (J–L) used in this study. Ophiostoma novo-ulmi ssp. americana × novo-ulmi hyphae (white arrows) inside the secondaryxylem vessel, radial section (A). The formation of narrowed secondary xylem vessels as a response to the fungus incoculation, radial section (B) and cross-section(C). Mesophyll tissue, cross-section (D, G, J). Midrib and primary xylem area, cross-section (E, H, K). Fungal hyphae inside primary xylem tracheary elements,cross-section (F). Unaffected, natural secondary xylem vessel grouping, cross-section (I, L). Scale bars: (A) ¼ 50 mm, (B, D, G, J) ¼ 100 mm,

(C, E, H, I, K, L) ¼ 500 mm, (F) ¼ 10 mm.

Durkovic et al. — Leaf traits in Dutch elm hybrids with contrasting disease resistance 217


sections were mounted on specimen stubs, sputter-coated withgold, and observed by high-vacuum SEM using a VEGA TS5130 instrument (Tescan, Czech Republic) operating at 15 kV.

Leaf growth

Leaf growth characteristics were assessed on the third fullyexpanded leaf from the apex that was sampled from five-leavedcurrent-year shoots. Leaf sizes (length, width, area) were mea-sured with an LI-3000A leaf area meter (LI-COR, Lincoln,NE, USA). Leaf slenderness was calculated as the ratiobetween leaf length and leaf width. Leaves were dried at65 8C for 72 h, then weighed. The variable LMA was calcu-lated as the ratio between leaf dry mass and leaf area.Measurements were performed on eight plants of the non-infected ‘Groeneveld’ (five sun leaves per plant), four plantsof the non-infected ‘Dodoens’ (five sun leaves per plant) andfour plants of the infected ‘Dodoens’ (five sun leaves perplant). Two repetitions of the measurements were carried out.

Leaf histology

Leaf lamina samples were cut with a razor blade, fixed in5 % glutaraldehyde in a 0.1 M cacodylate buffer at pH 7.0,dehydrated in ethanol and propylene oxide, and embedded inSpurr embedding medium. Sections approx. 1.5 mm thickwere cut using the automated rotary microtome LeicaRM2255 (Leica Biosystems, Nussloch, Germany) and glassknives, and stained with toluidine blue and basic fuchsin asdescribed by Lux (1981). Sections were observed using anOlympus BX50F light microscope (Olympus Europa,Hamburg, Germany). The thickness of the leaf, mesophylland palisade parenchyma was measured using NIS-ElementsAR 3.0 image analysis software (Laboratory Imaging,Prague, Czech Republic). Measurements were performed oneight plants of the non-infected ‘Groeneveld’ (two sunleaves per plant, two sections per leaf, one measurementper section), four plants of the non-infected ‘Dodoens’ (twosun leaves per plant, two sections per leaf, one measurementper section) and four plants of the infected ‘Dodoens’ (twosun leaves per plant, two sections per leaf, one measurementper section). Two repetitions of the measurements werecarried out.

Primary xylem conduit density and relative conductivity

Conduit density characteristics of the midrib primary xylem[tracheary element lumen area (A) and tracheary element dens-ities (N) per 0.1 mm2 of the primary xylem area] were deter-mined using NIS-Elements AR 3.0 image analysis software.Measurements of A and calculations of N were made on thepopulation of tracheary elements, which ranged from 118(the non-infected ‘Dodoens’) to 373 (the non-infected‘Groeneveld’) tracheary elements per 0.1 mm2 of the primaryxylem area within the examined midrib sections. In addition,the tracheary element lumen fraction (F) and the trachearyelement size:number ratio (S) were calculated as described inZanne et al. (2010). Total theoretical relative conductivity(RC) per 0.1 mm2 of the primary xylem area was calculatedas the sum of individual RCs divided by the area of a

cross-section of primary xylem (Durkovic et al., 2012),whereas the individual RC was calculated according toZimmermann (1983) as the fourth power of the equivalentcircle diameter of the tracheary element lumen. Lignin auto-fluorescence in cell walls of tracheary elements was detectedby excitation at 360 nm using a barrier filter with a transmis-sion cut-off at 470 nm, and photographed using a LeicaDM4000 B microscope equipped with a Leica DFC490digital colour CCD camera (Leica Microsystems). Primaryxylem conduit density and RC measurements were performedon eight plants of the non-infected ‘Groeneveld’ (two sunleaves per plant, one section per leaf midrib), four plants ofthe non-infected ‘Dodoens’ (two sun leaves per plant, onesection per leaf midrib) and four plants of the infected‘Dodoens’ (two sun leaves per plant, one section per leafmidrib). Two repetitions of the measurements were carried out.

Atomic force microscopy and nanomechanical propertiesof cell walls of tracheary elements

Deparaffinized midrib cross-sections, approx. 15 mm thick,were mounted on glass slides coated with (3-aminopropyl)-triethoxy-silane and allowed to air-dry in sterile Petri dishes.PeakForce QNM (quantitative nanomechanical) measurementswere done using a MultiMode 8 atomic force microscope(AFM) with a Nanoscope V controller (Bruker NanoSurfaces, Santa Barbara, CA, USA). Cell walls of trachearyelements were indented by silicon cantilevers MPP-12120,model TAP150A (Bruker AFM Probes, Camarillo, CA, USA)with a spring constant between 4.7 and 5.3 N m21,deflection sensitivity between 33.1 and 42.8 nm V21, andresonance frequency between 141 and 144 kHz, at 25 8C andambient air pressure. Prior to the measurements, the tipradius and geometry were controlled using a commercialgrid for 3-D visualization. PeakForce QNM measurements ofthe reduced Young’s MOE, adhesion and dissipation wereperformed at low approach tip velocities of 0.3–0.5 mm s21.To achieve accurate and reliable calculations of MOE, a suffi-cient sample deformation of 2 nm was ensured by adjusting thePeakForce setpoint to 25 nN. Measurements were performedon eight plants of the non-infected ‘Groeneveld’ (one sunleaf per plant, four cell walls of tracheary elements perleaf midrib), four plants of the non-infected ‘Dodoens’(one sun leaf per plant, four cell walls of tracheary elementsper leaf midrib) and four plants of the infected ‘Dodoens’(one sun leaf per plant, four cell walls of tracheary elementsper leaf midrib). Two repetitions of the measurements werecarried out.

Atomic force microscopy data processing

The initial data of MOE, coming from the PeakForce QNMmapping (represented by 256 × 256 matrices), were analysedby NanoScope Analysis software, version 1.40r2 (BrukerAXS, Santa Barbara, CA, USA), which uses the DMT model(Derjaguin et al., 1975). The raw data were subsequentlyimported into the MATLAB software, version 7(MathWorks, Natick, MA, USA). A slippery effect on thesteep surface topography strongly influenced the measurementof MOE. Thus, the height gradient was calculated for each



image pixel. Values corresponding to ‘steep’ points, where thesurface slope exceeded the value of 30 8, were not used(Durkovic et al., 2012).

Gas exchange

An open portable photosynthesis system with infra-red gasanalyser LI-6400 XTR (LI-COR) was used for in situ measure-ments of gas exchange characteristics. Net photosynthetic rate(PN), transpiration rate (E), stomatal conductance (gs) andintercellular CO2 concentration (ci) were measured on leavesfully exposed to the sun under a saturating photosyntheticphoton flux density of 1200+ 5 mmol m22 s21 and anambient CO2 concentration of 370+ 5 mmol mol21 usingthe 6400-08 standard leaf chamber with the 6400-02B LEDred/blue light source (LI-COR). Instantaneous water-use effi-ciency (WUE), giving information on the photosyntheticcarbon gain per unit transpirational water loss, was calculatedas the ratio of PN to E (Campbell et al., 2005). During thedescribed measurements, microclimatic conditions inside theassimilation chamber were kept constant (leaf temperatureTL 21+ 1 8C, relative air humidity 70+ 5 %). The vapourpressure deficit ranged from 0.9 to 1.3 kPa. Measurementswere performed on eight plants of the non-infected‘Groeneveld’ (five sun leaves per plant, three measurementsper leaf), four plants of the non-infected ‘Dodoens’ (five sunleaves per plant, three measurements per leaf) and fourplants of the infected ‘Dodoens’ (five sun leaves per plant,three measurements per leaf). Two repetitions of the measure-ments were carried out.

Chlorophyll fluorescence and chlorophyll content

Chlorophyll a fluorescence yields were measured on bothadaxial and abaxial surfaces of sun-exposed leaves using aportable fluorometer Plant Efficiency Analyser (HansatechInstruments Ltd, Kings Lynn, UK). Leaves were kept for30 min under leaf clamps for dark adaptation. After theinitial measurement of dark-adapted minimum fluorescence(F0), leaves were exposed to a saturating irradiance of2100 mmol m22 s21 for 1 s to measure the maximal fluores-cence of dark-adapted foliage (Fm). Maximal photochemicalefficiency of photosystem II [Fv/Fm ¼ (Fm – F0)/Fm],variable-to-initial fluorescence ratio (Fv/F0) and potential elec-tron acceptor capacity – ‘area’ (i.e. area above the inductioncurve between F0 and Fm) were determined. Measurementswere performed on eight plants of the non-infected‘Groeneveld’ (ten sun leaves per plant), four plants of the non-infected ‘Dodoens’ (ten sun leaves per plant) and four plants ofthe infected ‘Dodoens’ (ten sun leaves per plant). Two repeti-tions of the measurements were carried out.

Relative chlorophyll content was estimated with a portablechlorophyll meter CL-01 (Hansatech Instruments Ltd), andthe results were expressed as the chlorophyll index (Cassolet al., 2008). Measurements were performed on eight plantsof the non-infected ‘Groeneveld’ (seven sun leaves per plant,three measurements per leaf ), four plants of the non-infected‘Dodoens’ (seven sun leaves per plant, three measurementsper leaf) and four plants of the infected ‘Dodoens’ (sevensun leaves per plant, three measurements per leaf). Two repe-titions of the measurements were carried out.

Water potential

Leaf water potential (CL) was measured using theScholander pressure chamber technique. After cutting, theleaves were immediately placed into a sealed plastic bag andthen quickly measured (no more than 10 min elapsedbetween the time of leaf collection and measurement). Leafsamples were stuck through the head of a pressure chamber,model 1000 (PMS Instrument Co., Albany, OR, USA),which was slowly pressurized at a rate ,0.02 MPa s21 untila droplet of liquid occurred on the leaf cut surface. Chamberpressure at the moment of liquid emergence was recorded asa negative value of leaf water potential. Measurements weredone from 0600 to 1800 h in 2 h resolution on eight plantsof the non-infected ‘Groeneveld’ (one sun leaf per plant ateach measuring interval), four plants of the non-infected‘Dodoens’ (one sun leaf per plant at each measuring interval)and four plants of the infected ‘Dodoens’ (one sun leaf perplant at each measuring interval). Two repetitions of the mea-surements were carried out.

Statistical analysis

Data showed a normal distribution and thus were subjectedto a one-way analysis of variance (ANOVA). Duncan’s mul-tiple range tests were used for pairwise comparisons ofmeans. The effects of leaf surface on the yields of chlorophylla fluorescence were tested by two-way ANOVA. The Pearsoncorrelation coefficients were calculated for trait–trait linkages.The relationships were considered significant if P , 0.05.

Multivariate association of the 27 examined leaf traits wasanalysed with a principal component analysis (PCA) to de-scribe patterns of covariation among leaf growth, vascularand ecophysiological traits.

RESULTS

Leaf growth

Data on the leaf growth traits are presented in Table 1. For leaflength, width, area and dry mass, there were no statisticallysignificant differences found between non-infected plants of‘Groeneveld’ and ‘Dodoens’. ‘Dodoens’ performed betterthan ‘Groeneveld’ for LMA, leaf thickness, mesophyll thick-ness and palisade parenchyma thickness. Infected plants of‘Dodoens’ had a higher value of leaf slenderness than non-infected plants of ‘Dodoens’. For the remaining eight leafgrowth variables, however, the differences between infectedand non-infected plants of ‘Dodoens’ were negligible.

Primary xylem conduit density and nanomechanical properties ofcell walls of tracheary elements

‘Dodoens’ had significantly higher values than ‘Groeneveld’for primary xylem conduit density components such as A andS, as well as for RC per unit area (Table 2). With regard to thecomponents N and F, there were no significant differencesfound between the two Dutch elm hybrids. Interestingly, thedifferences between infected and non-infected plants of‘Dodoens’ were negligible for all the examined primaryxylem conduit density components.



‘Groeneveld’ had significantly higher values than ‘Dodoens’for MOE of cell walls of midrib tracheary elements. However,for this trait, there was no significant difference found betweeninfected and non-infected plants of ‘Dodoens’. In addition, cellwalls shared the same similarities for other nanomechanicalproperties, i.e. differences between non-infected plants of‘Groeneveld’ and ‘Dodoens’, as well as between infectedand non-infected plants of ‘Dodoens’, were negligible forboth adhesion and energy dissipation (Table 2). Lignin auto-fluorescence imaging of the primary xylem structure in themidrib, and AFM peak force error and height images of thecell wall surfaces of tracheary elements are presented in Fig. 2.

Ecophysiological traits

‘Dodoens’ had significantly higher values than ‘Groeneveld’for PN, E, gs and ci (Table 3). ‘Groeneveld’ performed betterthan ‘Dodoens’ for WUE and CL. There were no significantdifferences found between non-infected and infected plantsof ‘Dodoens’ for the above five gas exchange variables andleaf water potential. The highest relative chlorophyll contentwas determined in non-infected plants of ‘Dodoens’. Theinfected plants of ‘Dodoens’ still had a significantly higher

value of chlorophyll index than non-infected plants of‘Groeneveld’.

The results of chlorophyll a fluorescence yields for both leafsurfaces are given in Fig. 3A–F. Two-way ANOVA showedthat elm cultivar and leaf surface significantly contributed tothe differences that were found for the three examined vari-ables (Table 4). Abaxial surfaces had significantly highervalues of Fv/Fm and Fv/F0 ratios than adaxial surfaces. Withregard to the variable ‘area’, higher values were recorded onadaxial surfaces. Taken together for both leaf surfaces,‘Dodoens’ performed better than ‘Groeneveld’ for Fv/Fm andFv/F0 ratios as well as for the area above the induction curvebetween F0 and Fm. Interestingly, the infected plants of‘Dodoens’ had significantly higher values of the three chloro-phyll fluorescence variables than non-infected plants of‘Dodoens’.

Correlated and independent traits

As expected, several traits were correlated with LMA andMOE, respectively. Leaves with a large carbon and nutrient in-vestment in LMA tended to have a shorter leaf length (r ¼–0.58, P ¼ 0.018), and a greater leaf thickness (r ¼ 0.70, P ¼0.002, Fig. 4A), mesophyll thickness (r ¼ 0.73, P ¼ 0.001)

TABLE 1. Leaf growth characteristics in the examined Dutch elm hybrids

Trait ‘Groeneveld’ non-infected ‘Dodoens’ non-infected ‘Dodoens’ infected

LL (cm) 8.36+0.19a 7.79+0.20a 7.96+0.17a

LW (cm) 4.97+0.12a 4.83+0.12ab 4.57+0.11b

LS (cm cm21) 1.69+0.01b 1.62+0.02c 1.76+0.03a

LA (cm2) 27.6+1.3a 24.4+1.2ab 22.9+1.0b

LDM (g) 0.12+0.01a 0.14+0.01a 0.13+0.01a

LMA (g m22) 44.9+1.0b 55.8+1.7a 57.3+1.3a

LT (mm) 136+3b 166+6a 165+5a

MT (mm) 102+2b 127+5a 129+4a

PPT (mm) 54.5+1.6b 64.9+3.0a 68.4+3.3a

LL, leaf length; LW, leaf width; LS, leaf slenderness; LA, leaf area; LDM, leaf dry mass; LMA, leaf mass per area; LT, leaf thickness; MT, mesophyllthickness; PPT, palisade parenchyma thickness.

Data represent means+ s.e. Mean values followed by the same letters within the same row across hybrids are not significantly different at P , 0.05.

TABLE 2. Primary xylem conduit density characteristics and nanomechanical properties of cell walls of tracheary elements in theexamined Dutch elm hybrids


A (1025 mm2) 13.8+0.7b 17.3+0.9a 17.3+0.9a

N 202+13a 168+15a 165+8a

F (1023 mm2) 26.1+0.4a 27.8+1.1a 27.8+1.0a

S (1027 mm4) 7.8+0.7b 11.4+1.3a 11.0+0.9a

RC (1026 mm4) 7.27+0.44b 9.62+0.50a 9.78+0.77a

MOE (MPa) 2809+185a 1960+199b 2097+235b

ADH (nN) 9.6+0.6a 10.1+1.3a 10.6+0.9a

DIS (eV) 485+41a 515+69a 521+45a

A, tracheary element lumen area; N, number of tracheary elements per 0.1 mm2 of the primary xylem area; F, tracheary element lumen fraction; S, trachearyelement size:number ratio; RC, total theoretical relative conductivity per 0.1 mm2 of the primary xylem area; MOE, modulus of elasticity; ADH, adhesion;DIS, energy dissipation.




and palisade parenchyma thickness (r ¼ 0.67, P ¼ 0.005,Fig. 4B), as well as higher rates of PN (r ¼ 0.83, P , 0.001,Fig. 4C), E (r ¼ 0.82, P , 0.001), gs (r ¼ 0.87, P , 0.001)and ‘area’ (r ¼ 0.65, P ¼ 0.007). On the other hand, LMAwas independent of RC per unit area (r ¼ 0.18, P ¼ 0.51,Fig. 4D).

Significant linkages were also found between cell wall stiff-ness of tracheary elements and other vascular traits. RC perunit area (r ¼ –0.74, P ¼ 0.001, Fig. 4E), tracheary elementlumen area (r ¼ –0.76, P , 0.001, Fig. 4F) and trachearyelement size to number ratio (r ¼ –0.75, P , 0.001) weregreater in midribs with lower MOE values of cell walls of

tracheary elements. A positive correlation has been foundbetween MOE and the number of tracheary elements perunit area (r ¼ 0.68, P ¼ 0.004, Fig. 4G). We found no signifi-cant support for a negative correlation of MOE with leaf area(r ¼ –0.08, P ¼ 0.78, Fig. 4H).

Associations among leaf traits

A PCA was done to evaluate how leaf traits were associated(Fig. 5). The first axis explained 37 % of the variation andshowed strong positive loadings for E, gs, PN, A, S and the

‘Dodoens’ non-infected

‘Dodoens’ infected

‘Groeneveld’ non-infected Peak force error Height

FI G. 2. Lignin autofluorescence images of the primary xylem in the midrib (left images), AFM peak force error images (middle images) and AFM flatten heightimages (right images) of the cell wall surfaces of tracheary elements. Scale bars for non-infected plants of ‘Groeneveld’ and ‘Dodoens’: 100 mm for fluorescencemicroscopy images and 1.7 mm for AFM images. Scale bars for infected plants of ‘Dodoens’: 100 mm for the fluorescence microscopy image and 1.6 mm for

AFM images.



leaf tissue thickness traits. The negative side of the axis indi-cated strong loadings for N, MOE, WUE and CL. Thesecond axis explained 18 % of the variation and showedstrong positive loadings for leaf growth variables such asleaf area, length, width and dry mass. The negative side ofthe axis indicated strong loadings for WUE, LMA, MOEand N. In addition, PCA showed that both examined hybrids

formed compact homogeneous clusters, clearly separatedfrom each other, except for the single outlier specimen fromeach hybrid cultivar. These two specimens were placedoutside their own clusters, extending to the clusters of theircounterparts. The positions of infected plants of ‘Dodoens’were placed almost within the range of non-infected plantsof ‘Dodoens’.

TABLE 3. Ecophysiological characteristics in the examined Dutch elm hybrids


PN (mmol CO2 m22 s21) 14.6+0.2b 21.5+0.5a 22.0+0.2a

E (mmol H2O m22 s21) 2.33+0.03b 3.76+0.08a 3.72+0.03a

gs (mmol H2O m22 s21) 200+4b 380+14a 392+6a

ci (mmol CO2 mol21) 238+2b 253+1a 255+2a

WUE (mmol CO2 mmol H2O21) 6.34+0.09a 5.70+0.05b 5.94+0.08b

CL (MPa) 1.76+0.05a 1.55+0.03b 1.54+0.05b

CHLI 10.4+0.2c 14.1+0.2a 11.9+0.3b

PN, net photosynthetic rate; E, transpiration; gs, stomatal conductance; ci, intercellular CO2 concentration; WUE, instantaneous water-use efficiency; CL,leaf water potential; CHLI, chlorophyll index.


c

cb

c

b

a

a

ba a

b

a a

b

a a

b

a

0‘Groeneveld’non-infected

‘Dodoens’non-infected

‘Are

a’ (

Mb

s–1)

Fv/F

0F

v/F

m

Adaxial surface Abaxial surface

‘Dodoens’infected

‘Groeneveld’non-infected

‘Dodoens’non-infected

‘Dodoens’infected

10

20

30

40

2

3

4

5

0·76

0·78

0·80

0·82

A B

C D

E F

FI G. 3. Chlorophyll a fluorescence yields in the examined Dutch elm hybrids determined on adaxial (A, C, E) and abaxial (B, D, F) leaf surfaces. (A, B) Resultsof the maximal photochemical efficiency of photosystem II (Fv/Fm). (C, D) Results of the variable-to-initial fluorescence ratio (Fv/F0). (E, F) Results of thepotential electron acceptor capacity (‘area’). Histograms represent means+ s.e. Mean values followed by the same letters across hybrid cultivars are not signifi-

cantly different at P , 0.05.



DISCUSSION

We found definite differences in leaf traits between the twoDutch elm hybrids with a contrasting tolerance to DED.Furthermore, we found an unexpected number of similaritiesbetween the infected and non-infected plants of ‘Dodoens’.We relate these findings to previously published works, high-lighting the novel findings pertinent to nanomechanical prop-erties of the cell walls of midrib tracheary elements.

Leaf trait dissimilarities between the examined Dutch elm hybrids

For the whole complex of leaf growth, vascular and ecophy-siological traits, trait dissimilarities reached a frequency of70.4 %. Dissimilarities were particularly dominant amongecophysiological traits, where 100 % of the examined traitsshowed significant differences. Dissimilarities were alsofound, but to a lesser extent, among leaf growth (55.6 %)and vascular traits (50 %). We found that ‘Dodoens’leaves had significantly higher gas exchange rates than‘Groeneveld’ leaves. Thus, the enhanced photosyntheticcapacity accumulated more carbon biomass per unit leaf area(LMA), which would contribute to the faster relative growthrate of ‘Dodoens’ (Shipley, 2002; Kruger and Volin, 2006;Pasquet-Kok et al., 2010). In addition, the greater thicknessof leaf, mesophyll and palisade parenchyma may also contrib-ute to the faster growth of ‘Dodoens’ due to their positive cor-relation with LMA that was found in this study (Niinemets,1999; Vendramini et al., 2002).

Vascular traits

Tracheary element size and number are the primary indica-tors of leaf vascular strategy. Tracheary element lumen areastrongly affects hydraulic conductivity, whereas conduitdensity influences bulk xylem composition (Preston et al.,2006). The influence of tracheary elements on xylem densitycan be decomposed into two additional significant componentsof vascular strategy, tracheary element lumen fraction andtracheary element size to number ratio (Zanne et al., 2010).These primary xylem conduit density components were usedin the previous comparative study of leaf vascular strategyamong parental and hybrid species of Sorbus, where theyexhibited parental-like or transgressive phenotypic expressionin the examined hybrids (Durkovic et al., 2012). In thisstudy, ‘Dodoens’ performed better than ‘Groeneveld’ for thecomponents A and S. Moreover, the significantly lower valuefor relative hydraulic conductivity per unit area found in

‘Groeneveld’ limited its performance for the examined gas ex-change variables such as PN, E, gs and ci (Tombesi et al.,2010).

Nanomechanical properties of cell walls of tracheary elements

There are a few studies in which AFM mapping has beenused to examine the stiffness of vascular cells, e.g. stiffnessof smooth muscle cells in monkey (Qiu et al., 2010) and cor-onary atherosclerotic plaque components in mouse (Tracquiet al., 2011). In a study carried out by Lacayo et al. (2010),AFM images were used to characterize the topography of thecell wall surface of single tracheary elements of Zinniaelegans after the transdifferentiation of cultured leaf mesophyllcells into tracheary elements. Unfortunately, no nanomechani-cal data for cell walls were provided. In this study, we wereable to provide unique quantitative nanomechanical datathrough the application of PeakForce QNM measurements.The data of the reduced Young’s MOE for Dutch elmhybrids lay fully within the range that was reported for cellwalls of midrib tracheary elements of five Sorbus species(Durkovic et al., 2012). The cell walls of tracheary elementsof ‘Groeneveld’ were stiffer as their MOE values were signifi-cantly higher than those of ‘Dodoens’. Possible indirectreasons for the above observations could be that the MOEwas inversely correlated with tracheary element lumen area.‘Groeneveld’ had significantly smaller free cavity spacefor water flow than ‘Dodoens’, hence ‘Groeneveld’ shouldhave a thicker cell wall mass for water-conductive cells. Thehigher stiffness is usually caused by the thicker cell wall mass.

In addition, de Farias Viegas Aquije et al. (2010) examinedthe adhesion forces in leaves of Ananas comosus cultivarsinfected by the fungus Fusarium subglutinans using theAFM contact mode. The fusariosis-susceptible cultivar had asignificantly higher frequency of high adhesion force measure-ments for soft mesophyll tissue cell walls than the resistantcultivar. However, this was not the case for the stiff, lignifiedcell walls of the Ulmus hybrid cultivars in this study. We foundno significant support for the more adhesive surface of cellwalls of tracheary elements in infected plants of ‘Dodoens’.This result may reflect a different nature and response of softand stiff cells to fungus infection, as well as different effectsof the fungi examined. Adhesion is a measure of a force inter-action between the tip and the sample cell wall surface.Changes in adhesion may be a good indicator of alterationsin cell wall biopolymer distribution and orientation aswell as chemical alterations in cell walls. Dissipation deals

TABLE 4. Two-way ANOVA of chlorophyll a fluorescence variables in the examined Dutch elm hybrids

Source of variation

Degrees of freedom Sum of squares Mean square F-test

Fv/Fm Fv/F0 ‘Area’ Fv/Fm Fv/F0 ‘Area’ Fv/Fm Fv/F0 ‘Area’ Fv/Fm Fv/F0 ‘Area’

Elm 2 2 2 0.026 17.753 0.210 0.0132 8.877 0.105 71.84*** 74.09*** 44.53***Leaf surface 1 1 1 0.001 0.584 0.655 0.0010 0.584 0.655 5.91* 4.88* 277.48***Elm × surface 2 2 2 0.001 0.718 0.046 0.0005 0.359 0.023 2.79 NS 3.00 NS 9.73***Error 314 314 314 0.058 37.623 0.741 0.0001 0.120 0.002

***P , 0.001; *P , 0.05; NS, non-significant.



with a deformation resistance of a material. Energy dissipatedbetween the tip and the sample during each tap on the cell wall

surface is related to the toughness of a material, and thus may

be used to evaluate a degree of cell wall degradation by fungal

hyphae. Biodegradation rapidly decreases toughness of sec-

ondary xylem measured by dynamic tests (Clausen, 2010).

To the best of our knowledge, this study is the first to report

quantitative data for adhesion and dissipation of the walls of

conductive cells of leaves.

Leaf trait similarities between the infected and non-infectedplants of ‘Dodoens’

For the whole complex of leaf growth, vascular and ecophy-siological traits, trait similarities reached a frequency of81.5 %. Strong similarities were found among vascular traitswhere 100 % of the examined traits showed non-significantdifferences. The proportion of trait similarities was also veryhigh among leaf growth traits (88.9 %), with the only excep-tion being the infected plants of ‘Dodoens’ with their more

250A E

B F

C G

D H

Leaf

thic

knes

s(µ

m)

200

150

100

50

0

y = 40·989 + 2·166xR2 = 0·495

y = 4·415 + 1·107xR2 = 0·446

y = –1·895 + 0·396xR2 = 0·695

y = 6·269 × 10–6 + 4·377 × 10–8xR2 = 0·031

y = 15·088 × 10–6 – 0·027 × 10–7xR2 = 0·550

y = 25·653 × 10–5 – 0·417 × 10–7xR2 = 0·573

y = 55·468 + 0·053xR2 = 0·457

y = 27·958 – 0·977 × 10–3xR2 = 0·006

100

Pal

isad

e pa

renc

hym

ath

ickn

ess

(µm

)

80

60

40

20

0

30

Net

pho

tosy

nthe

sis

(µm

ol C

O2

m–2

s–1

) 25

20

15

10

5

0

20

Rel

ativ

e co

nduc

tivity

per

unit

area

(10

–6 m

m4 )

15

10

5

0

20

Rel

ativ

e co

nduc

tivity

per

unit

area

(10

–6 m

m4 )

15

10

5

0

25

Trac

hear

y el

emen

tlu

men

are

a (1

0–5 m

m2 )

20

15

10

5

0

400

Num

ber

of tr

ache

ary

elem

ents

per

uni

t are

a

300

200

100

0

50

Leaf

are

a(c

m2 )

40

30

20

10

030 40 50

Leaf mass per area (g m–2) Modulus of elasticity (MPa)

60 70 80 1000 1500 2000 2500 3000 3500 4000

Non-infected ‘Groeneveld’Non-infected ‘Dodoens’Infected ‘Dodoens’

FI G. 4. Trait linkages with (A–D) leaf mass per area (LMA) and with (E–H) modulus of elasticity (MOE), identified in the examined Dutch elm hybrids.Relationships of LMA to leaf thickness (A), palisade parenchyma thickness (B), net photosynthetic rate (C) and relative hydraulic conductivity per 0.1 mm2

of the primary xylem area (D). Relationships of MOE to relative hydraulic conductivity per 0.1 mm2 of the primary xylem area (E), tracheary elementlumen area (F), number of tracheary elements per 0.1 mm2 of the primary xylem area (G) and leaf area (H). Non-infected plants of ‘Groeneveld’, non-infected

plants of ‘Dodoens’ and infected plants of ‘Dodoens’ are as indicated in the key in (E).



slender leaves. Although some differences were found amongecophysiological traits, the similarities were proportionallygreater (60 %). Previous studies have shown that DEDfungus is able to colonize remote areas in the plant such asthe leaf midrib and secondary veins (Pomerleau and Mehran,1966; Nasmith et al., 2008). Thus, we hypothesized that leaftrait performances of infected plants of ‘Dodoens’ related togas exchange rates, chlorophyll fluorescence yields, leafgrowth and nanomechanical properties of primary xylem cellwalls would be significantly decreased. Unexpectedly, theinfected plants had significantly higher values for chlorophylla fluorescence variables, but these higher chlorophyll fluores-cence yields did not have a direct impact on the more effectivebiological functioning of photosystem II in comparison withthe non-infected plants. The reaction centres of photosystemII were intact functionally in both types of ‘Dodoens’; more-over, their Fv/Fm ratios were far higher than the thresholdvalue of 0.725 that indicates the onset of reversible changesin reactions centres of photosystem II (Canova et al., 2012).In addition, the infected plants had a significantly lowerchlorophyll index, but, again, no impact on the reduced ratesof PN was observed in comparison with the non-infectedplants. In U. minor plants inoculated with O. novo-ulmi ssp.americana, Oliveira et al. (2012) observed a significant de-crease in chlorophyll content that was accompanied by a de-crease in the Fv/Fm ratio. This discrepancy may be explainedby the plant material used in the experiment and the

experimental conditions. The authors used 2-month-old invitro plants sensitive to DED, and the experiment was com-pleted 42 d after inoculations with a suspension of blastos-pores. However, no work with the tolerant or resistant elmclones accompanied the above experiment. Also, no informa-tion about the destiny and the performances of infectedplants in the next growing season was provided. In our experi-ment, however, mature plants of ‘Dodoens’ survived inocula-tions without severe damage, their leaves were fullyfunctional in the following growing seasons, and no decreasesin the Fv/Fm ratio and PN rate were observed in comparisonwith the non-infected plants.

Taken together, except for two traits (leaf slenderness andrelative chlorophyll content), we found no evidence of a de-crease in leaf trait performances among infected plants of‘Dodoens’, despite the persistence of O. novo-ulmi hyphae inthe lumens of midrib tracheary elements. This result impliesthat leaf growth, vascular and gas exchange traits in matureplants of ‘Dodoens’ were unaffected by the DED fungus.

Correlated and independent traits

The LMA, an important plant carbon economy trait, showsfrequent linkages with other leaf functional ecology traits thattogether shape the performance of plants. In this study, wenoted that LMA was interlinked with leaf lamina thicknessand other examined leaf growth traits (Niinemets, 1999;

–0·5 –0·4 –0·3 –0·2 –0·1

WUE

MOE

ADH

'Area'PPT

LMAMT

LT

DIS

LS

FS

E

A

ci

gs

PN

Fv/F0Fv/Fm

CHLI

LDM

RC

LW

LLNon-infected ‘Groeneveld’Non-infected ‘Dodoens’Infected ‘Dodoens’LA

N

0PCA axis 1 (37 %)

PC

A a

xis

2 (1

8 %

)

0·1 0·2 0·3 0·4 0·5

–0·4

–0·3

–0·2

–0·1

0

0·1

0·2

0·3

0·4

–6 –5 –4 –3 –2 –1 0 1 2 3 4 5 6

–5

–4

–3

–2

–1

0

1

2

3

4

5

YL

FI G. 5. Positions of 27 leaf traits on the first and second axes of the principal component analysis (PCA). Trait abbreviations are found in Tables 1–3and Fig. 3. Non-infected plants of ‘Groeneveld’, non-infected plants of ‘Dodoens’ and infected plants of ‘Dodoens’ are as indicated in the key. The bottom

and left-hand axes refer to the examined leaf traits, whereas the top and right-hand axes refer to the examined trees.



Vendramini et al., 2002). Plants with higher values of LMA(here it was ‘Dodoens’) usually have a better leaf persistenceand defence against herbivores and physical hazards (Poorteret al., 2009). We also observed that LMA strongly influencedgas exchange variables, mostly the net photosynthetic rate(Shipley, 2002; Wright et al., 2004). Thus, photosynthetic cap-acity scales linearly to the biomass investment in the leaf,making leaf anatomy the main driver of photosynthesis(Poorter et al., 2009). In addition, water flux-related traits in-cluding RC are also a potentially fundamental determinantof species performance differences (Sack and Holbrook,2006). In this study, the RC of the examined Dutch elmhybrids was orthogonal to LMA, which supports the independ-ence of these two hub traits (Brodribb et al., 2005; Sack et al.,2005).

Furthermore, we noted that MOE of the cell walls oftracheary elements was inversely correlated with trachearyelement lumen area. Woodrum et al. (2003) and Jacobsenet al. (2005) observed a similar trend between the MOE ofbulk secondary xylem and fibre lumen diameter. The lesserstiffness was caused by a thinner mass of supportive cellwalls. In our case, the higher tracheary element lumen areacaused higher stress in the cell walls due to the leaf laminamass, and consequently caused a decrease in the MOE ofcell walls, an observation also reported by Kern et al.(2005). On the other hand, MOE was positively related tothe number of tracheary elements per unit area. Thus, the stiff-ness of the primary xylem, weakened by the higher trachearyelement lumen area and the lower MOE of cell walls, was sup-ported by the higher number of tracheary elements to achieve atrade-off. Lastly, in contrast to the previously reported negativecorrelation of MOE with leaf area in diverse Sorbus species(Durkovic et al., 2012), no relationship was found in thisstudy due to insufficient variation in leaf area between theexamined Dutch elm hybrids.

Conclusions

Strong dissimilarities in leaf trait performances (70.4 % intotal) were observed between the examined Dutch elmhybrids. Both hybrids were clearly separated from each otherin the multivariate leaf trait space. ‘Dodoens’ had significantlyhigher values for LMA, leaf tissue thickness variables,tracheary element lumen area, RC, gas exchange variables,relative chlorophyll content and chlorophyll a fluorescenceyields. On the other hand, ‘Groeneveld’ had stiffer cell wallsof tracheary elements, and also higher values for WUE andCL. Unexpectedly, we found a very high proportion of leaftrait similarities between the infected and non-infected plantsof ‘Dodoens’ (81.5 % in total). Leaf growth, vascular andgas exchange traits in the infected plants of ‘Dodoens’ wereunaffected by the DED fungus. Leaves with a large carbonand nutrient investment in LMA tended to have a greaterleaf thickness and higher rates of PN, but LMA was independ-ent of RC. Significant linkages were also found between theMOE of cell walls of tracheary elements and some vasculartraits such as RC, tracheary element lumen area and thenumber of tracheary elements per unit area.

ACKNOWLEDGEMENTS

The authors thank Professor D. Gomory for statistical advice,Dr A. Cicak and Mr M. Mamon for technical assistance,and Mrs E. Ritch-Krc for language revision. This workwas financed by the Slovak scientific grant agency VEGA(1/0132/12).

LITERATURE CITED

Bartlett MK, Scoffoni C, Sack L. 2012. The determinants of leaf turgor losspoint and prediction of drought tolerance of species and biomes: a globalmeta-analysis. Ecology Letters 15: 393–405.

Binz T, Canevascini G. 1996a. Xylanases from the Dutch elm disease patho-gens Ophiostoma ulmi and O. novo-ulmi. Physiological and MolecularPlant Pathology 49: 159–175.

Binz T, Canevascini G. 1996b. Differential production of extracellular laccasein the Dutch elm disease pathogens Ophiostoma ulmi and O. novo-ulmi.Mycological Research 100: 1060–1064.

Bosu PP, Wagner MR. 2007. Effects of induced water stress on leaf trichomedensity and foliar nutrients of three elm (Ulmus) species: implications forresistance to the elm leaf beetle. Environmental Entomology 36:595–601.

Bosu PP, Wagner MR. 2008. Anatomical and nutritional factors associatedwith susceptibility of elms (Ulmus spp.) to the elm leaf beetle(Coleoptera: Chrysomelidae). Journal of Economic Entomology 101:944–954.

Brasier CM, Kirk SA. 2001. Designation of the EAN and NAN races ofOphiostoma novo-ulmi as subspecies. Mycological Research 105:547–554.

Brasier CM, Kirk SA. 2010. Rapid emergence of hybrids between the twosubspecies of Ophiostoma novo-ulmi with a high level of pathogenicfitness. Plant Pathology 59: 186–199.

Brodribb TJ, Holbrook NM. 2005. Leaf physiology does not predict leafhabit; examples from tropical dry forest. Trees 19: 290–295.

Campbell DR, Galen C, Wu CA. 2005. Ecophysiology of first and secondgeneration hybrids in a natural plant hybrid zone. Oecologia 144:214–225.

Canova I, Durkovic J, Hladka D, Lukacik I. 2012. Changes in stomatalcharacteristics and photochemical efficiency during leaf development insix species of Sorbus. Photosynthetica 50: 635–640.

Cassol D, De Silva FSP, Falqueto AR, Bacarin MA. 2008. An evaluation ofnon-destructive methods to estimate total chlorophyll content.Photosynthetica 46: 634–636.

Clausen CA. 2010. Biodeterioration of wood. In: Forest Products Laboratory.Wood handbook—wood as an engineering material. General TechnicalReport FPL-GTR-190, Centennial edn. Madison, WI: US Departmentof Agriculture, Forest Service, 14-1–14-16.

Corredoira E, Vieitez AM, Ballester A. 2002. Somatic embryogenesis inelm. Annals of Botany 89: 637–644.

Derjaguin BV, Muller VM, Toporov YP. 1975. Effect of contact deforma-tions on the adhesion of particles. Journal of Colloid and InterfaceScience 53: 314–326.

Dunbar-Co S, Sporck MJ, Sack L. 2009. Leaf trait diversification and designin seven rare taxa of the Hawaiian Plantago radiation. InternationalJournal of Plant Sciences 170: 61–75.

Durkovic J, Kardosova M, Canova I, et al. 2012. Leaf traits in parental andhybrid species of Sorbus (Rosaceae). American Journal of Botany 99:1489–1500.

Dvorak M, Tomsovsky M, Jankovsky L, Novotny D. 2007. Contribution toidentify the causal agents of Dutch elm disease in the Czech Republic.Plant Protection Science 43: 142–145.

Dvorak M, Jankovsky L, Krajnakova J. 2009. Characterisation of CzechOphiostoma novo-ulmi isolates. Suleyman Demirel University Faculty ofForestry Journal A: 233–237.

de Farias Viegas Aquije GM, Zorzal PB, Buss DS, Ventura JA, FernandesPMB, Fernandes AAR. 2010. Cell wall alterations in the leaves offusariosis-resistant and susceptible pineapple cultivars. Plant CellReports 29: 1109–1117.



Fenning TM, Tymens SS, Gartland JS, Brasier CM, Gartland KMA.1996. Transformation and regeneration of English elm using wild-typeAgrobacterium tumefaciens. Plant Science 116: 37–46.

Heybroek HM. 1983. Resistant elms for Europe. In: Burdekin DA. ed.Research on Dutch elm disease in Europe. London: Her Majesty’sStationery Office, 108–113.

Jacobsen AL, Ewers FW, Pratt RB, Paddock WA III, Davis SD. 2005. Doxylem fibers affect vessel cavitation resistance? Plant Physiology 139:546–556.

Janmey PA, Winer JP, Murray ME, Wen Q. 2009. The hard life of softcells. Cell Motility and the Cytoskeleton 66: 597–605.

Kern KA, Ewers FW, Telewski FW, Koehler L. 2005. Mechanical perturb-ation affects conductivity, mechanical properties and abovegroundbiomass of hybrid poplars. Tree Physiology 25: 1243–1251.

Konrad H, Kirisits T, Riegler M, Halmschlager E, Stauffer C. 2002.Genetic evidence for natural hybridization between the Dutch elmdisease pathogens Ophiostoma novo-ulmi ssp. novo-ulmi andO. novo-ulmi ssp. americana. Plant Pathology 51: 78–84.

Krajnakova J, Longauer R. 1996. Culture initiation, multiplication and iden-tification of in vitro regenerants of resistant hybrid elms. Lesnictvı–Forestry 42: 261–270.

Kruger EL, Volin JC. 2006. Reexamining the empirical relation betweenplant growth and leaf photosynthesis. Functional Plant Biology 33:421–429.

Lacayo CI, Malkin AJ, Holman H-YN, et al. 2010. Imaging cell wall archi-tecture in single Zinnia elegans tracheary elements. Plant Physiology 154:121–133.

Lux A. 1981. A rapid method for staining semi-thin sections of plant material.Biologia 36: 753–757 [In Slovak].

Nakashima J, Mizuno T, Takabe K, Fujita M, Saiki H. 1997. Direct visu-alization of lignifying secondary wall thickenings in Zinnia elegans cellsin culture. Plant and Cell Physiology 38: 818–827.

Nasmith C, Jeng R, Hubbes M. 2008. A comparison of in vivo targeted geneexpression during fungal colonization of DED-susceptible Ulmus ameri-cana. Forest Pathology 38: 104–112.

Niinemets U. 1999. Components of leaf dry mass per area – thickness anddensity – alter leaf photosynthetic capacity in reverse directions inwoody plants. New Phytologist 144: 35–47.

Oliveira H, Sousa A, Alves A, Nogueira AJA, Santos C. 2012. Inoculationwith Ophiostoma novo-ulmi subsp. americana affects photosynthesis, nu-trition and oxidative stress in in vitro Ulmus minor plants. Environmentaland Experimental Botany 77: 146–155.

Ouellette GB, Rioux D, Simard M, Cherif M. 2004. Ultrastuctural and cyto-chemical studies of host and pathogens in some fungal wilt diseases:retro- and introspection towards a better understanding of DED.Investigacion Agraria: Sistemas y Recursos Forestales 13: 119–145.

Pasquet-Kok J, Creese C, Sack L. 2010. Turning over a new ‘leaf’: multiplefunctional significances of leaves versus phyllodes in Hawaiian Acaciakoa. Plant, Cell and Environment 33: 2084–2100.

Pijut PM, Domir SC, Lineberger RD, Schreiber LR. 1990. Use of culturefiltrates of Ceratocystis ulmi as a bioassay to screen for disease tolerantUlmus americana. Plant Science 70: 191–196.

Pomerleau R, Mehran AA. 1966. Distribution of spores of Ceratocystis ulmilabelled with phosphorus-32 in green shoots and leaves of Ulmus ameri-cana. Le Naturaliste Canadien 93: 577–582.

Poorter H, Niinemets U, Poorter L, Wright IJ, Villar R. 2009. Causes andconsequences of variation in leaf mass per area (LMA): a meta-analysis.New Phytologist 182: 565–588.

Poorter L, Bongers F. 2006. Leaf traits are good predictors of plant perform-ance across 53 rain forest species. Ecology 87: 1733–1743.

Preston KA, Cornwell WK, DeNoyer JL. 2006. Wood density and vesseltraits as distinct correlates of ecological strategy in 51 California coastrange angiosperms. New Phytologist 170: 807–818.

Qiu H, Zhu Y, Sun Z, et al. 2010. Vascular smooth muscle cell stiffness as amechanism for increased aortic stiffness with aging. Circulation Research107: 615–619.

Sack L, Cowan PD, Jaikumar N, Holbrook NM. 2003. The ‘hydrology’ ofleaves: co-ordination of structure and function in temperate woodyspecies. Plant, Cell and Environment 26: 1343–1356.

Sack L, Holbrook NM. 2006. Leaf hydraulics. Annual Review of PlantBiology 57: 361–381.

Sack L, Tyree MT, Holbrook NM. 2005. Leaf hydraulic architecture corre-lates with regeneration irradiance in tropical rainforest trees. NewPhytologist 167: 403–413.

Shipley B. 2002. Trade-offs between net assimilation rate and specific leaf areain determining relative growth rate: relationship with daily irradiance.Functional Ecology 16: 682–689.

Smalley EB, Guries RP. 1993. Breeding elms for resistance to Dutch elmdisease. Annual Review of Phytopathology 31: 325–352.

Solla A, Bohnens J, Collin E, et al. 2005. Screening European elms for resist-ance to Ophiostoma novo-ulmi. Forest Science 51: 134–141.

Sterck FJ, Poorter L, Schieving F. 2006. Leaf traits determine the growth–survival trade-off across rain forest tree species. American Naturalist 167:758–765.

Sticklen MB, Bolyard MG, Hajela RK, Duchesne LC. 1991. Molecular andcellular aspects of Dutch elm disease. Phytoprotection 72: 1–13.

Strobel G, Van Alfen N, Hapner KD, McNeil M, Albersheim P. 1978.Some phytotoxic glycopeptides from Ceratocystis ulmi, the Dutch elmdisease pathogen. Biochimica et Biophysica Acta 538: 60–75.

Svaldi R, Elgersma DM. 1982. Further studies on the activity of cellwall degrading enzymes of aggressive and non-aggressive isolatesof Ophiostoma ulmi. European Journal of Forest Pathology 12: 29–36.

Temple B, Horgen PA, Bernier L, Hintz WE. 1997. Cerato-ulmin, a hydro-phobin secreted by the causal agents of Dutch elm disease, is a parasiticfitness factor. Fungal Genetics and Biology 22: 39–53.

Tombesi S, Johnson RS, Day KR, DeJong TM. 2010. Relationships betweenxylem vessel characteristics, calculated axial hydraulic conductance andsize-controlling capacity of peach rootstocks. Annals of Botany 105:327–331.

Tracqui P, Broisat A, Toczek J, Mesnier N, Ohayon J, Riou L. 2011.Mapping elasticity moduli of atherosclerotic plaque in situ via atomicforce microscopy. Journal of Structural Biology 174: 115–123.

Van Alfen NK. 1989. Reassessment of plant wilt toxins. Annual Review ofPhytopathology 27: 533–550.

Van Alfen NK, Turner NC. 1975. Influence of a Ceratocystis ulmi toxin onwater relations of elm (Ulmus americana). Plant Physiology 55:312–316.

Vendramini F, Dıaz S, Gurvich DE, Wilson PJ, Thompson K, HodgsonJG. 2002. Leaf traits as indicators of resource-use strategy in floraswith succulent species. New Phytologist 154: 147–157.

Webber JF. 2004. Experimental studies on factors influencing the transmis-sion of Dutch elm disease. Investigacion Agraria: Sistemas y RecursosForestales 13: 197–205.

Westoby M, Falster DS, Moles AT, Vesk PA, Wright IJ. 2002. Plant eco-logical strategies: some leading dimensions of variation betweenspecies. Annual Review of Ecology and Systematics 33: 125–159.

Woodrum CL, Ewers FW, Telewski FW. 2003. Hydraulic, biomechanical,and anatomical interactions of xylem from five species of Acer(Aceraceae). American Journal of Botany 90: 693–699.

Wright IJ, Reich PB, Westoby M, et al. 2004. The worldwide leaf economicsspectrum. Nature 428: 821–827.

Young CE, Hall RW. 1986. Factors influencing suitability of elms for elm leafbeetle, Xanthogaleruca luteola (Coleoptera: Chrysomelidae).Environmental Entomology 15: 843–849.

Zanne AE, Westoby M, Falster DS, et al. 2010. Angiosperm woodstructure: global patterns in vessel anatomy and their relation to wooddensity and potential conductivity. American Journal of Botany 97:207–215.

Zimmermann MH. 1983. Xylem structure and the ascent of sap. Berlin:Springer-Verlag.



Leaf trait dissimilarities between Dutch elm hybrids with a ...

Documents

Transcript of Leaf trait dissimilarities between Dutch elm hybrids with a ...