‘Defence lignin’ and hydroxycinnamyl alcohol dehydrogenase activities in wounded Eucalyptus gunnii

Lignins are classically considered as complex three-dimensional phenolic polymers resulting from the oxidative polymerization of hydroxycinnamyl alcohols (p-coumaryl alcohol, coniferyl alcohol and sinapyl alcohol) ( and 1990), although other monomeric units can also be incorporated ( et al. 1994; et al. 1997). These molecules are important in several aspects of plant development. For example, their hydrophobicity waterproofs the conducting cells of the xylem and the rigidity that they impart to cell walls strengthens the supporting fibre cells in both phloem and xylem tissues. The chemical composition of these ‘developmental lignins’ varies according to a number of parameters such as plant group, species, organ, tissue, cell-type and even cell-wall layer. For example, gymnosperm lignin is typically composed of mainly guaicyl units (G-units derived from coniferyl alcohol) whereas typical angiosperm lignin is composed of a mixed guaicyl–syringyl lignin (G- and S-units, derived from coniferyl alcohol and sinapyl alcohol, respectively). The pathway leading to the biosynthesis of developmental lignins has been – and continues to be – extensively studied; while most of the structural enzymes have been purified and the corresponding genes cloned (- and 1999), the metabolic pathways linking the enzymes are continually being revised in the light of new experimental results ( et al. 2001; et al. 2001; et al. 2001). Nevertheless, the basic pathway is relatively well understood, and starts with the deamination of phenylalanine followed by a number of enzymatic steps leading to the production of cinnamoyl-coenzyme A-esters. These esters are the precursors for a wide range of phenolic molecules such as flavonoids and coumarins, which play important roles in plant development and defence (D and 1995). In the case of angiosperm lignin synthesis, the cinnamoyl-coenzyme A-esters are converted into the three types of hydroxycinnamyl alcohols (monolignols) through the action of the two enzymes of the lignin-specific pathway – cinnamoyl-coenzyme A reductase (CCR) and cinnamyl alcohol dehydrogenase (CAD). Lignins also play an important role in plant defence, and indeed, one of the characteristic responses of plants to injury and disease is the reinforcement of cell walls in the vicinity of the wound/infection site with lignin, as well as with suberin ( et al. 1980; and 1992). The lignin deposited in this case is also known as ‘lignin-like material’ ( et al. 2001), or ‘defence lignin’, so as to distinguish it from developmental lignin. The reinforcement of cell walls with such material results in the formation of an impermeable barrier that protects the healthy tissue from water loss and opportunistic infection, as well as limiting the spread of the original pathogen ( et al. 1980; 1992; 1994). Defence lignin formation has also been implicated in the responses of plants to abiotic stresses such as ozone ( et al. 1998) and water deprivation ( et al. 1998). A significant amount of data ( et al. 1985; and 1990; and 1993; et al. 1995) suggests that the chemical composition of ‘defence lignin’ is different to that of ‘developmental lignin’. For example, in cucurbits, it appears that defence lignin is extremely rich in p-coumaraldehyde units (H-units) as opposed to cucurbit developmental lignin, which is a typical angiosperm guaiacyl–syringyl (G–S) lignin ( et al. 2001). Similarly, in wounded poplar and almond trees, defence lignin is apparently associated with G-lignin ( et al. 1984; and 1987) as opposed to the normal G–S developmental lignin found in these species. While the enzymatic activities involved in the biosynthesis of defence lignins appear to be similar to those involved in developmental lignin biosynthesis, the signal transduction pathways involved in the production of these two lignin types appear to be different. For example, different PAL genes are regulated differentially during development or in response to elicitation ( et al. 1995), and a similar situation occurs in the case of O-methyltransferase genes ( et al. 1992; et al. 1999). Such regulation presumably enables the plant to direct its pool of phenolic metabolites to where they are needed (defence phenolic molecules or developmental lignin precursors). While the situation was considered to be less complex in the case of the lignin-specific pathway genes, the recent isolation of two CCR genes (AtCCR1 and AtCCR2) from Arabidopsis thaliana ( et al. 2001) showing different expression patterns during development, or in response to pathogen attack, would seem to suggest that the production of developmental versus defence lignin is also controlled at this level of the biosynthetic pathway. Similarly, the isolation of two lucerne CAD cDNAs (MsaCAD1 and MsaCAD2), which are differentially regulated during development and plant defence ( et al. 1999), would seem to support the hypothesis of separate signal transduction pathways for the formation of developmental and defence lignins. In contrast to herbaceous plants, little information is available concerning the same processes in woody plants. In order to understand more about the chemical nature of defence lignin in trees, we undertook detailed microscopic and histochemical analyses of the wounding response in 3.5-month-old Eucalyptus gunnii plantlets. Plantlets were used for this part of the study because we wished to develop a potential model wounding system for subsequent in situ molecular studies of defence lignin formation. In addition, and in order to provide preliminary data on enzymatic activities associated with the wound response in E. gunnii, we measured hydroxycinnamyl alcohol dehydrogenase activities over a 17-day period following wounding in 6.5-year-old, field-grown E. gunnii trees. Trees, rather than plantlets, were used for this part of the investigation because the characterization of enzymatic activity required more protein than could be provided by the plantlets. E. gunnii Hook. plants used for the histochemical analyses were grown from seed kindly provided by AFOCEL (Association Fôret Cellulose) as described previously ( and 1996). Plants used for the wounding experiment were 3.5 months old, approximately 20 cm high, and had between six and eight pairs of leaves. Plants were decapitated using a fresh sterile razor blade (Fig. 1a) and the wounded zone excised at days 0, 1, 3, 5 and 7 post-wounding. Samples were fixed in FAA (formol-acetic-alcohol), embedded in paraffin (paraplast plus) and semi-thin transversal and longitudinal sections (10–15 μm) were prepared as described previously ( and 1996). Diagram of wounding system and morphological response of Eucalyptus gunnii plantlets to wounding. (a) Diagram indicating position of wound (decapitation, A–B). (b) Diagram (longitudinal section) illustrating the position of the wound zone in a decapitated stem 7 days after wounding. (c) Detail of wound zone from (b). PRZ = pith reaction zone; V = vascular tissue The presence of phenolic material was revealed by its autofluorescence under UV light, and Sudan black ( 1984) was used to indicate the presence of suberin (and/or cutin). Phloroglucinol-HCl (P-HCl) was used to indicate the presence of lignin, and the Mäule reaction was used to distinguish lignins rich or poor in syringyl monomers ( 1987). In order to differentiate wall-bound (ester-linked) phenolic acids from polyphenolic wall material, control sections were saponified ( and 1979) with 0.5 m NaOH on a hot plate (60°C, 30 min) prior to histochemical analysis. Slides were examined and photographs taken under bright-field and fluorescent light conditions using a Leitz Laborlux D light microscope as described previously ( and 1996). For the determination of CAD activities, samples were collected from 6-year-old E. gunnii trees (Clone No. 800859X) growing in AFOCEL plantations at Longages (Lat. 43°22′ N, Long. 1°05′ E) in the south of France. Ten trees were selected at random, and 10 wounds were made in the trunk of each tree by removing a 1-cm diameter disc of bark tissue with a 1-cm diameter cork borer (Fig. 2a). Samples for analysis were then harvested at 0, 1, 2, 3, 5, 8, 10, 13, 15 and 17 days following wounding using a 2-cm cork borer to remove a ring of bark tissue around the original wound zone (Fig. 2b). One ring was removed from each of the 10 trees, thereby giving a total of 10 rings that were pooled for each sampling date. Control samples from unwounded trees were taken at days 0, 10 and 17. Samples were dropped directly into liquid nitrogen and then stored at −80°C until protein extraction. Photograph of wounding protocol used with field-grown Eucalyptus gunnii trees. (a) Wound (arrow) made using a 1-cm diameter cork borer (CB). Bar = 6 cm. (b) Tissue ring (arrow) removed for enzymatic analyses using a 2-cm diameter cork borer Between 5 and 9 g (f. wt.) of frozen tissue for each sampling date was ground to a fine powder in a pre-chilled coffee mill and then extracted in cold extraction buffer (1 : 3 w/v) as described previously ( and 1994). Following a centrifugation step (18 000 × g, 45 min) to eliminate cell debris, 500 μl of the crude extract was removed for desalting on a sephadex G 25 column ( and 1994) and analysed for hydroxycinnamyl alcohol dehydrogenase activity. Hydroxycinnamyl alcohol dehydrogenase activities were determined spectrophotometrically with coniferyl alcohol and sinapyl alcohol by monitoring the increase in A400 due to the formation of coniferaldehyde from coniferyl alcohol, and sinapylaldehyde from sinapyl alcohol, respectively, as described previously ( and 1994). Chronologically, the development of the wound response in decapitated plants was similar to that reported previously ( and 1996) with the formation between 5 and 7 days after wounding of a well-established ‘barrier’ sealing-off underlying tissues from the environment. This barrier formed in two main groups of tissues – (a) the ground tissues (cortex and pith) and dermal tissues, and (b) the vascular tissues. Following decapitation, the ‘soft’ tissues of the pith shrink, leaving the harder vascular tissue protruding (Fig. 1b,c). The walls of the pith tissue cells then become reinforced via the deposition of polymeric material, which, in conjunction with the deposition of similar polymeric material in the cortex and vascular tissues, results in the formation of the barrier. Figure 3a shows a transverse section of a non-wounded control stem stained with the Mäule reaction. Only phloem fibres and differentiated secondary xylem react positively with this stain giving a red colouration. Primary xylem, immature secondary xylem and other tissues do not react positively with this reagent indicating that only the phloem fibres and differentiated secondary xylem contain significant amounts of syringyl lignin. Histochemical characterization of the wound zones in Eucalyptus gunnii seedlings as viewed in transverse (a–d) and longitudinal (e, f) sections for controls (a) and 7 days after wounding (b–f). Different stains and lighting conditions were used: Mäule (a, d); phloroglucinol-HCl (c); phloroglucinol-HCl and Sudan black (e); UV (b, f). (a) A positive red reaction indicating the presence of syringyl (S) units is observed in mature secondary xylem (s) and phloem fibres (p), while primary xylem (px) and developing secondary xylem (sx) show a negative brown reaction indicating the absence of S-units. Bar = 100 μm. (b) Xylem cell blockages (b), cortex- (c) and vascular cambium-cell walls (v) autofluoresce under UV illumination indicating the presence of phenolic material. Bar = 40 μm. (c) Cortex cell walls (c) are stained red indicating the presence of lignin-like material. Bar = 40 μm. (d) Cortex- (c) and vascular cambium-cell walls (v) are stained brown indicating that the lignin-like material observed in Fig. 2c is poor in S-units. Bar = 20 μm. (e) Cortex cell walls (c) are stained red indicating the presence of lignin-like material. (x) = xylem cell blockage. Bar = 30 μm. (f) Cell walls of pith parenchyma cells (p) autofluoresce indicating the presence of phenolic material. Bar = 20 μm Figure 3b–e shows the cortex barrier (see Fig. 1) 7 days after wounding. Cell walls in the cortex barrier zone (c, Fig. 3b) fluoresce under UV light while the comparable region in non-wounded control sections (data not shown) does not fluoresce, thereby suggesting that phenolic material has been deposited in the cell walls in response to wounding. Cortex cell walls gave a positive red reaction with P-HCL (c, Fig. 3c) and a negative brown reaction with the Mäule reagent (c, Fig. 3d) suggesting that the phenolic material is composed of lignin poor in syringyl units. Alkali hydrolysis did not change the colour reactions suggesting that polymeric material rather than ester-linked phenolic acids are involved (data not shown). The position and extent of the cortex barrier (c, Fig. 3e) is more clearly illustrated in a stem longitudinal section stained with P-HCL and Sudan black. 3, 4 show the pith reaction zone (see Fig. 1) 7 days after wounding. A well-defined zone (p, Fig. 3f) can be clearly identified by the fluorescence of cell walls under UV light indicating the deposition of cell wall phenolic material. As with the cortex barrier, a positive red reaction (p, Fig. 4a) was obtained with P-HCl and a negative reaction with the Mäule reaction (p, Fig. 4b), indicating that the ‘lignin-like’ material deposited in the cell walls of this zone is relatively poor in syringyl units. Histochemical characterization of the wound zones in Eucalyptus gunnii seedlings as viewed in transverse (c, d) and longitudinal (a, b, e, f) sections 7 days after wounding. Different stains and lighting conditions were used: Mäule (b, d); phloroglucinol-HCl (a, c, e); Safranin O/Fast Green (f). (a) Cell walls of pith parenchyma cells (p) are stained red indicating the presence of lignin-like material. Bar = 20 μm. (b) Pith parenchyma cell walls (p) are stained brown indicating that the lignin-like material observed in Fig. 2g is poor in S-units. Bar = 10 μm. (c) Vascular cambium cell walls (v) are stained red indicating the presence of lignin-like material. (b) = blockages of xylem cells. Bar = 10 μm. (d) Developing xylem cell walls (x) adjacent to the vascular cambium (v) are stained red indicating the presence of a lignin-like material rich in S-units, while protoxylem vessel walls (m) are stained brown. Bar = 20 μm. (e) Metaxylem cell blockages (b) are stained red indicating the presence of lignin-like material. Bar = 200 μm. (f) Xylem vessel- (b) and fibre cell-blockages (f) are stained red indicating the presence of phenolic material, uf = fibre cell without blockage. Bar = 40 μm Figure 4c–f shows the wound response in the xylem 7 days after wounding. The walls of all xylem cells gave a positive red colouration with P-HCl (Fig. 4c) and some cells (vessels or axial parenchyma, but not radial parenchyma) were blocked by a bright-red-stained phenolic material that also fluoresced under UV light (b, Fig. 3b). A negative brown reaction with the Mäule reagent (Fig. 4d) suggested that the blockage material was poor in syringyl units. The walls of immature xylem cells (x; Fig. 4d) adjacent to the vascular cambium (v, Fig. 4c) gave a positive reaction with the Mäule reagent suggesting the deposition of syringyl lignin in these cells. Such an observation is interesting because these cells do not usually produce this lignin until in their development Fig. Primary xylem cells (b, Fig. and more mature xylem Fig. are blocked by a material that positively with P-HCl (Fig. and Safranin (Fig. fibre cells Fig. also appear to be blocked by a similar material, while other adjacent fibres Fig. The characterization of CAD activities required more protein than could be provided by the E. gunnii to this part of investigation using 6-year-old field-grown E. gunnii trees that were wounded as described in and A level of CAD activity using coniferyl alcohol as was in bark samples (Fig. activity using sinapyl alcohol could be in these CAD activity then approximately a at days showing a at days 15 and 17. CAD activities in bark from unwounded trees at days 10 and 17 were comparable to those observed at (Fig. activity could be until days after wounding (Fig. and then relatively after 10 a at 17 activities in unwounded trees at days 10 and 17 were comparable to those observed at (Fig. The activity (Fig. that activity to CAD activity after 10 days and approximately of the CAD activity at 17. of alcohol dehydrogenase activity using coniferyl alcohol and sinapyl alcohol as in samples harvested from wounded Eucalyptus gunnii trees growing under (a) activity during the 17-day experimental CAD = activity using coniferyl alcohol as = activity using sinapyl alcohol as = control (b) of activity ( et al. 1980; and 1992; et al. 2001) has that lignin or ‘lignin-like’ are involved in plant defence and detailed analyses are the underlying and molecular in herbaceous plants, less information is available concerning the same in woody plants. In order to provide detailed histochemical data about the response of trees to we had previously examined lignin formation in wounded E. gunnii plantlets ( and 1996). In that the wound was made by a 1-cm in the thereby the phloem and part of the xylem tissues. trees, other plants, are to different types of wounding and it be that the of response according to the of wound provide histochemical information concerning lignin formation in response to a different of we examined the response to wounding in plantlets that were decapitated so as to the by In this of the of the vascular tissues is and the plant in order to opportunistic pathogen and In and in order to provide information about the underlying involved in defence lignin we hydroxycinnamyl alcohol dehydrogenase activities in wounded field-grown E. gunnii trees. microscopic analyses indicate as previously reported ( and wounding the production of polymeric phenolic material that cell walls in the cortex barrier, vascular tissue and pith reaction In addition, polymeric material, or at material to and other xylem cell types in the vascular tissues. As a a barrier is formed to tissues from and opportunistic pathogen were used to the nature of the barrier. While be taken using such as they are less than chemical used for they have the of characterization of the ‘lignin-like’ material 7 days after wounding that the lignin deposited in the cell walls of the cortex barrier, vascular tissue and pith reaction zone was poor in S-units. the cell walls of formed also positively with the Mäule reagent indicating the presence of lignin in cells that do not usually contain this lignin until more analyses of the xylem blockages also that the lignin-like material formed was poor in syringyl units. As E. gunnii developmental xylem and tissues contain more than = = and the histochemical results in this suggest that the ‘defence lignin’ at during the wound response is different in from that of ‘developmental lignin’. the histochemical used in this do not to the lignin formed is or do they the between and cinnamyl alcohols because the (P-HCl) with both of these ( and two types of reaction could be in the microscopic (a) the reinforcement of cell walls and (b) the of xylem and other xylem cell suggest that in E. gunnii, the reinforcement of cell walls the deposition of polymeric lignin, while the blockage of xylem the deposition of a units ( and in xylem This hypothesis would be in with the of the cells parenchyma cells are presumably of and lignin in their cell while xylem are and of such a In this the for the blockage of these cells be provided by other in the of an reinforced by units. Nevertheless, (- showing that cell walls of xylem lignin precursors following the of the corresponding could suggest that the cell walls of blocked xylem cells can also be reinforced by lignin precursors from parenchyma cells. In support of this it is interesting to that the of has that several are in poplar xylem axial and radial parenchyma cells suggesting that these cells provide for in the cell walls of xylem ( et al. et al. et al. et al. et al. The of sections under UV light (data not shown) that a cell blockages This of P-HCl ( 1984) suggests that some other phenolic material – suberin – is also implicated in the of these cells. formation is associated with expression and ( et al. 1992; et al. et al. has that different of genes in the pathway are differentially with some being in development while other are wounding or ( et al. 2001) suggests that CAD also be differentially The CAD the reaction in the formation of hydroxycinnamyl alcohols and CAD cDNAs have been from a number of et al. 1998). other cDNAs and showing a of approximately to that of the CAD cDNAs have also been these cDNAs were all as defence ( et al. has been to a CAD with a range of than that of typical CAD while has been to a alcohol dehydrogenase ( et al. 1996). of these enzymes are during plant defence reactions in response to Such results would seem to suggest that in the signal transduction pathways for developmental lignin and defence lignin also at the hydroxycinnamyl alcohol little information is available concerning expression in trees. The results of enzymatic on in of activities from field-grown E. gunnii trees, that hydroxycinnamyl alcohol dehydrogenase activity in bark tissues as a response to wounding. two different types of CAD activity – a coniferyl alcohol dehydrogenase activity and a sinapyl alcohol dehydrogenase activity using coniferyl alcohol and sinapyl alcohol, respectively, as CAD activity was in control samples and to increase until approximately after wounding. In activity was not until after wounding after which it to increase of the of to CAD activity that it from 10 after wounding to the of the experimental activity for more than of the total activity at 17. These are interesting in the light of the results of et al. a with a more CAD from a developing xylem from with and less with other genes ( et al. 1999), and analyses indicate that to a of appears that these two genes are during the formation of developmental lignin in angiosperm tissues the typical G–S mixed lignin of this plant is not known is in response to pathogen or preliminary characterization of the CAD and enzymatic activities suggests the of differentially regulated E. gunnii CAD and In histochemical characterization of the defence lignin formed in E. gunnii plantlets that its chemical composition from that of E. gunnii developmental lignin. While the is a relatively typical mixed lignin, the is relatively poor in syringyl units. In addition, results suggest as in herbaceous species, and biosynthesis in trees are differentially for AFOCEL is for plant material. and are for during the of this