Chemical defence responses of Norway spruce to two fungal pathogens

Constitutive and inducible terpene production is involved in conifer resistance against insects and fungal infestations. To gain knowledge about local defence responses of Norway spruce bark against pathogens and to find potential chemical markers for resistance breeding, we inoculated the stem of 8-year-old Norway spruce ( Picea abies ) clonal

beetle-associated fungi are usually vectored to trees by bark beetles (Kirisits, 2004), and the actions of both bark beetles and their associated fungi contribute to tree mortality.In addition, the Heterobasidion species complex represents some of the most serious pathogens of Norway spruce in Scandinavia by causing root and stem rot and rendering the timber defective for sawing and pulping (Thor, 2005).These decay fungi are commonly spread by airborne spores that upon landing on freshly cut stump surfaces give rise to mycelia that grow further to the neighbouring living trees through root contacts.
The ability of conifers to react to various stressors by production of high amounts of defence chemicals likely contributes to their long survival.Norway spruce relies on both constitutive and induced defences to repel invading insects and restrict the spread of invading fungal pathogens.The first line of defence in Norway spruce trees is the bark, where the constitutive defence chemicals deter invasion by their physical properties and chemical toxicity (Franceschi et al., 2005).Once an injury or an infection is recognized by the tree, induced defences, including the formation of traumatic resin ducts and polyphenolic parenchyma cells (Franceschi et al., 2005), are activated.The trees may increase terpene concentrations a hundred times or more, which can dose-dependently inhibit the colonization of the spruce bark beetle (Zhao, Borg-Karlson, et al., 2011a;Zhao, Krokene, et al., 2011b) and its associated fungi (Novak et al., 2014;Zeneli et al., 2006).Certain terpenes inhibit the growth of decay fungi (Kusumoto et al., 2014).
Several studies have shown that genetically controlled host characteristics partly determine the susceptibility of Norway spruce to Heterobasidion infection (Arnerup et al., 2010;Gorriz et al., 2016).Some genotypes are less infected by root rot fungi than others (Karlsson & Swedjemark, 2006;Skrøppa et al., 2015;Steffenrem et al., 2016;Swedjemark & Karlsson, 2004;Swedjemark et al., 1998).Interestingly, the genetic components appear to play an even larger role in resistance to E. polonica than H. parviporum in Norway spruce (Skrøppa et al., 2015;Skrøppa et al., 2015;Steffenrem et al., 2016), and lesion length following inoculation with E. polonica varied extensively between Norway spruce genotypes (Skrøppa et al., 2015;Steffenrem et al., 2016;Zeneli et al., 2006).These results indicate Norway spruce has genetic resistance against the two fungi, which can be used to select genotypes with resistance to both pathogens for a breeding programme.
By investigating natural infections and performing artificial inoculations in clonal trees, the susceptibility of a large number of Norway spruce genotypes to Heterobasidion spp.has been determined in Sweden (Swedjemark & Karlsson, 2004;Swedjemark et al., 1998).However, the susceptibility of these genotypes to bark beetles and their associated fungal pathogens has not been investi-

gated. A deeper understanding of conifer defence responses to both
Heterobasidion species and the bark beetle-fungi complex is needed to select optimal Norway spruce genotypes with resistance to multiple pests.
In this pilot study, we investigated clone-specific terpene defence against two fungal pathogens.The primary aim of the study was to test whether Norway spruce has a similar chemical defence against the two major pathogens.In addition, we also aimed to search for potential biomarkers that could assist tree-breeding programmes in selecting Norway spruce genotypes with less susceptibility to bark beetle-fungi complex and Heterobasidion species.Four spruce clones, two showing the highest and two showing the lowest susceptibility to Heterobasidion in field inoculations in prior studies (for details, see Materials and Methods), were chosen for the experiment.Six ramets of each clone were inoculated with Heterobasidion parviporum and Endoconidiophora polonica.
The production of terpenes in intact and fungi-inoculated bark was analysed using GC-MS.The chemical differences found between the more and the less susceptible trees in this study could stimulate further investigation to verify the relationship between putative-resistant markers and individual tree performance when subjected to multiple stressors.

| Plant and fungal materials
The experiments were carried out in the nursery at Ekebo, Sweden, on potted 8-year-old Norway spruce clonal trees with 100-150 cm in height and ca 30 mm stem diameter at the inoculation sites.The parents of the clones come from Germany and were used in controlled crosses to generate 15 families.The experimental clones are rooted cuttings of representative plants from these families (Rosvall, 1999) The fungal growth in sapwood was measured 34 days after inoculating Heterobasidion annosum on the stem of 4-year-old Picea abies clonal trees (Swedjemark et al., 1998).

TA B L E 1
Origin and fungal growth of the clones used in this study growth after inoculation of Heterobasidion annosum in tree stem (Swedjemark & Karlsson, 2004;Swedjemark et al., 1998).In this study, four clones with different susceptibility were used.Two clones (132 and 329) with the most fungal growth and two clones (393 and 457) with the least fungal growth in previous studies (Swedjemark & Karlsson, 2004;Swedjemark et al., 1998) were assumed to be representative of trees with high and low susceptibility to Heterobasidion, respectively (Table 1).

| Inoculation procedure
Trees were inoculated with fungi in July 2014.Six ramets of each clone were inoculated with H. parviporum on one side and with E.
polonica on the other side of the stem, with a vertical distance of ca 30 cm between the inoculation sites.Fungal inoculations were performed using two 4-mm cork borers, one for Ep inoculation and the other for Hp inoculation to avoid any cross-contamination.The inoculum consisted of mycelium that had been growing on malt agar (2% malt, 1.5% agar) for 1 week.After removing the tree bark, a fungal plug with agar inoculum was inserted into the hole, and then parafilm ® M (Bemis Company) was wrapped around the stem to close the wound.
Three weeks after inoculation, two bark plugs were taken from the border (one from the upper and one from the lower border) of each inoculation hole using a 4-mm cork borer.The two plugs were then pooled as one sample to characterize the induced terpene response as well as possible fungal metabolites in the different clones.
An additional sample was taken 30 cm below the Ep inoculation to measure the constitutive terpene level in the clones.Previous trials have shown that no terpene induction occurs in this area after inoculation (unpublished data).After sampling, the bark samples were individually wrapped in aluminium foil, frozen and stored in liquid nitrogen until processing in the laboratory.In the laboratory, the bark plugs were cut into 1 × 1 mm pieces and extracted at room temperature in 0.5 ml n-hexane containing 0.158 mg/ml pentadecane (Lancaster Synthesis) as an internal standard and 0.12 mg/ml 3-tert-butyl-4-hydroxy-anisol (Fluka, Switzerland) as an antioxidant.
After 24h, the extracts were transferred to new vials and stored at −25°C before GC-MS analyses.The bark samples were dried at 80°C for 6 hr and then weighed on a Sartorius electronic balance for calculation of absolute amounts of the terpenes.

| GC-MS analyses
The terpenes and fomannoxin (a volatile toxin produced by Heterobasidion species) in the hexane extracts were separated, identified and quantified using a Varian 3,400 GC connected to a Finnigan SSQ 7,000 MS and a DB-wax column (J&W USA, 30 m, 0.25 mm id, 0.25 µm film thickness).The temperature program was set to 40°C for 1 min, followed by an increase of 4°C/min to 235°C, and held at the final temperature for 29 min.One-microlitre hexane extract was injected into a split/splitless injector with a 30-s splitless injection at 230°C.The Aux temperature was set to 240°C.Helium was used as the carrier gas at a flow of 1 ml/min.The temperature of the ion source was 150°C, the mass detector was operated with a mass range of 30-350 m/z, and the electron impact ionization was 70 eV.Thunbergol was identified by comparing with a sample isolated and characterized from Norway spruce (Kimland & Norin, 1968).The absolute amounts of terpenes were calculated relative to the internal standards and expressed as mg/g dry wt.The relative amounts of terpenes were calculated as the ratio of the area of each peak to the sum of all the areas of the volatiles in a defined GC fraction and expressed as percentages.

| Statistical analyses
All the statistics were carried out using Statistica 9.0 software (StatSoft Inc.).One-way ANOVA was used to test the differences in terpene levels in intact or fungal-infested bark between clones.
If the treatments were significantly different (p < 0.05), means were separated using LSD at p = 0.05.Terpene quantities were log(y + 1)-transformed, and proportional data were arcsin-transformed before ANOVA to correct for unequal variance and departures from normality.

| Terpene profile in intact bark
The six ramets within each clone had similar terpene profiles.However, the four experimental clones showed clearly different terpene hydrocarbon composition in the intact bark (Figure A1; Table A1).The constitutive levels of ß-phellandrene in clones 132 and 329, limonene in clones 393 and 457, ß-pinene in clones 132 and 393, myrcene in clones 132 and 457, and (+)-3-carene and sabinene in clone 393, were significantly higher than those of the other clones (p = 0.012-0.048) (Figure 1; Table A1).In addition, the total terpene amounts were higher in the two clones regarded as less susceptible to H. parviporum (clone 393 and 457) compared with the two more susceptible clones (329 and 132) (Figure 2).Interestingly, the less susceptible clones (393 and 457) also had a higher proportion of diterpenes in their terpene profile than the susceptible ones (329 and 132) (Figure 2).The level of thunbergol was significantly higher in the two less susceptible clones compared with the two susceptible clones (p <0.02, Figure 1).A1).All the clones responded strongly to Hp infection.

| Clonal
However, the responses of the four experimental clones to Ep were

| Clonal differences in fomannoxin content
In addition to the plant-produced chemicals, the reactive fungal toxin fomannoxin was detected in living Hp-inoculated bark.There were only small amounts of fomannoxin in the infested bark of the clones, with the lowest amount in the most susceptible clone 132 and the highest amount in the less susceptible clone 457 (p < 0.05) (Figure 3).
The six ramets within each clone showed a very similar chemical profile in intact control bark.Each ramet within a clone also demon- (+)-3-Carene and terpinolene were found to be associated with resistance in Sitka spruce (Picea sitchensis) genotypes originating from an area that may have been subject to high weevil pressure (Hall et al., 2011;Robert, 2010;Robert et al., 2010).In a feeding bioassay, an extremely resistant clone with a high (+)-3-carene content reduced white pine weevil (Pissodes strobi) feeding by 50% compared with an extremely susceptible clone with trace (+)-3-carene (Robert & Bohlmann, 2010).In Norway spruce, (+)-3-carene was observed to be highly induced by Ep and Hp inoculation (Danielsson et al., 2011;Zhao et al., 2010).Increased levels of (+)-3-carene in Norway spruce trees are correlated with decreased larval survival of the great spruce bark beetle (Storer & Speight, 1996) and inhibited growth of E. polonica in malt agar medium (Novak et al., 2014).In this study, (+)-3-carene was one of the characteristic monoterpenes for the least susceptible clone 393.These observations suggest that 3-carene might be involved in conifer resistance against insects and pathogens, and that the high 3-carene content in clone 393 bark may contribute to its low susceptibility to Heterobasidion fungi.
Several diterpene acids, including abietic and dehydroabietic acids, are known to have a significant impact on the growth of Hp (Kusumoto et al., 2014) and bark beetle-associated fungal pathogens (Kopper et al., 2005).Dehydroabietic acid has also been identified as a strong indicator of resistance of Sitka spruce against white spruce weevil (Robert, 2010;Robert et al., 2010).Here, we note that the two less susceptible clones (393 and 457) had higher constitutive diterpene levels in the intact bark, and higher induced levels of thunbergene, thunbergol and a few other diterpenes after fungal infection compared with the two susceptible clones.These findings are in accordance with the observation from maritime pine (Pinus pinaster) genotypes with a varied resistance against the pine weevil (Hylobius abietis).Higher amounts of diterpenes were detected in the stem bark of more resistant genotypes after methyl jasmonate treatment (López-Goldar et al., 2014).Thus, the variation in diterpene profiles might also correlate with the degree of resistance to pathogens.
In a previous study, we found that the absolute amount of thunbergol and the total amount of all the quantified diterpenes were significantly higher in putatively resistant trees with shorter lesions compared with putatively susceptible trees with long lesions after Ep inoculation (Zhao et al., 2010).We discovered a similar pattern here.
The amount of thunbergol was much higher in the two less susceptible clones compared with the two more susceptible clones, both in intact bark and in bark inoculated with Ep or Hp.It is also noteworthy that the production of thunbergol was absent in the most susceptible clone 132.In contrast, the amount of geranyllinalool, a diterpene that only appeared in Hp-infected Norway spruce bark, was significantly higher in the two most susceptible clones compared with the two less susceptible clones.Thus, these observations suggest that these two diterpenes are of interest when searching for chemical markers that are associated with differential susceptibility of Norway spruce clones to fungal infection.
It is known that Hp can produce a complex mixture of toxic compounds related to fomannoxin (Hansson et al., 2012(Hansson et al., ,2014)).
Fomannoxin is volatile and less polar than those compounds identified by HPLC (Hansson et al., 2012(Hansson et al., ,2014) ) and could be detected using GC-MS.Kusumoto et al. (2014) found that the fungus produced a significantly lower amount of fomannoxin when growing in a medium with a high concentration of diterpene (abietic acid or dehydroabietic acid).Here, we found the highest amount of fomannoxin in the least susceptible clone, but only trace amounts of this compound in the most susceptible clone.This finding is not in agreement with Kusumoto et al. (2014).The reason for this difference is not known, but a possible explanation could be that the fungus produced more phytotoxin to counteract the stronger defence when inoculated on a less susceptible tree, or H. parviporum has strain-specific variation in phytoxin production.
In conclusion, and in accordance with earlier observations (Danielsson et al., 2011;Zeneli et al., 2006) The compounds were identified by comparing retention times and mass spectra with available authentic standards, or by comparing retention indexes (RIs) and mass spectra with MassFinder 3 (Hochmuth Scientific Consulting, Germany) and the reference libraries of NIST (National Institute of Standards and Technology).Most terpene standards were purchased from Sigma-Aldrich, Lancaster Synthesis or Bedoukian, but a few were received as gifts from Firmenich, or isolated from Norway spruce and Scots pine turpentine at KTH.
-specific terpene response of Norway spruce to the two fungi Three weeks after fungal inoculation, most experimental clones showed clear terpene responses to inoculation by the two fungal pathogens.In general, the trees showed a stronger response to Hp inoculation compared with Ep inoculation.The level of monoterpenes, sesquiterpenes and diterpenes in tree bark increased threefold, 4.4fold and 6.4-fold 3 weeks after Ep inoculation, whereas these three classes of terpenes showed ninefold, 6.9-fold and 16.6-fold increase, respectively, in Hp-inoculated bark.The six ramets within each clone responded similarly to either Hp or Ep inoculation, indicating consistency in inoculation, sampling and chemical analysis.The four experimental clones showed a different terpene response to the inoculation of the two pathogens (Table highly variable.One clone (457) showed a strong response, two clones (393 and 132) only showed an intermediate response, and the last clone (329) did not show a clear quantitative terpene response to Ep inoculation (Figure 2).Whereas the levels of all the three-terpene classes (monoterpene, sesquiterpene and diterpene) were elevated after fungal inoculation, the increase in diterpenes in response to fungal inoculation was the most pronounced.The proportion of diterpenes in the terpene mixture F I G U R E 1 The amount of 3-carene, thunbergol and geranyllinalool in tree bark of the four tested Norway spruce clones with differential susceptibility to Heterobasidion.The levels of thunbergol in the bark of the two less susceptible clones were higher than in the two more susceptible clones, both in intact bark (CT) bark or bark inoculated with Heterobasidion parviporum (Hp) and Endoconidiophora polonica (Ep).Geranyllinalool, a diterpene detected only in Hp-inoculated bark, showed an opposite pattern.Error bars denote standard error (n = 6).Bars with different letters donate significant difference within each treatment at p = 0.05 level increased from 36.3% in control bark to 54.3 and 51.9%, respectively, in Ep-and Hp-inoculated bark.Interestingly, fungal inoculations induced variable increases in the proportion of different diterpene hydrocarbons, and extensive differences in diterpene content were observed between clones after fungal inoculation.The amount of thunbergol in the bark of the two less susceptible clones (clones 393 and 457) was clearly higher than in the two susceptible clones (clones 329 and 132), both in Ep-and Hp-inoculated bark (Figure 1).The most susceptible clone (clone 132) did not produce thunbergol, while the least susceptible clone (457) produced high quantities of thunbergol in both intact and fungal-inoculated bark.Geranyllinalool was not present in the control bark but was highly induced by Hp but not Ep inoculation.The level of geranyllinalool was much higher in the two susceptible clones after Hp inoculation (Figure 1), suggesting that it might be a marker of Heterobasidion spp.susceptibility in Norway spruce.

E 2
The absolute amount (top) and relative proportion (bottom) of monoterpenes (MT), sesquiterpenes (ST) and diterpenes (DT) in the intact bark (CT), or bark inoculated with Endoconidiophora polonica (Ep) and Heterobasidion parviporum (Hp) of the four tested Norway spruce clones with differential susceptibility to Heterobasidion, 3 weeks after fungal inoculation.Error bars denote standard error (n = 6).Bars with different letters donate significant differences across all treatments at p = 0.05 level F I G U R E 3 The amounts of fomannoxin present in the phloem of the four Norway spruce clones, 3 weeks after Heterobasidion parviporum inoculation (n = 6).Error bars denote standard error.Bars with different letters donate significant difference at p = 0.05 level strated a surprisingly similar defence response towards either Ep or Hp inoculation, indicating reliable inoculation, sample extraction and chemical analysis.However, the four selected clones had different chemical composition in intact bark and exhibited a varied terpene accumulation after fungal inoculation.The two less susceptible clones had higher total terpene concentrations as well as a larger diterpene proportion in the intact bark than the two more susceptible clones.Terpenes were strongly induced by Hp in all the clones, but only the clone 457 showed a similar response to both Hp and Ep inoculation, whereas the clone 329 only responded to Hp inoculation.Our observations suggest that Norway spruce has both clone-and fungi-specific terpene responses to Ep and Hp inoculation.
, we found differing chemical responses among Norway spruce clones towards Hp and Ep inoculation.Induced defence in Norway spruce includes a variety of chemical compounds, but their effects on fungi and insects are only partly understood (Biedermann et al., 2018).Our observations suggest that Norway spruce has both clone-and fungi-specific terpene responses to Ep and Hp inoculation.Tree-breeding programmes should include trees with strong defence response to a variety of fungi and other stressors.Our results demonstrate that certain Norway spruce clones can develop resistance to several pathogens-further studies with larger experimental material are necessary to improve the understanding of underlying traits.ACK N OWLED G EM ENTS Dr. Bo Karlsson is greatly acknowledged for providing space at Ekebo for the experiment and necessary information about the clonal material used in this study.The study was funded by Ferdowski University of Mashhad, Iran, to AZ; Formas grant to TZ (grant 229-2011-890); the Department of Chemistry, KTH to KA; SSF (Paretree) and Mobilitas top researcher grant MT4 Chemical Ecology Estonia to AKBK.We thank Dr. Douglas Jones for linguis-The peer review history for this article is available at https://publo ns.com/publo n/10.1111/efp.12640.U R E A 1 PCA plot based on the amounts of all the quantified terpenes in control Norway spruce phloem without fungal inoculation.The distance between the samples/trees in the plot indicates their similarity in terpenoid composition.The positions of the terpenes indicate their contribution to the principal components.The first principal component (PC1) explained 61.5% and the second component (PC2) 27.5% of the sample variation TA B L E A 1 Absolute amounts of major terpene hydrocarbons (μg/g.drywt equivalent to pentadecane) in control Norway spruce bark, or bark inoculated with either Heterobasidion parviporum or Endoconidiophora polonica, 3 weeks after fungal inoculation.Data are expressed as means ± 1 SE (n = 6 trees)