- Top of page
- Materials and Methods
Response mechanisms of plants to either excess soil salinity (for recent reviews see Hasegawa et al., 2000; Munns, 2002, 2005) or high solar radiation (particularly UV-radiation, Rozema et al., 1997; Jansen et al., 1998; Caldwell et al., 2003) have been investigated in great details over the past three decades. To date, less is known about how plants perform when exposed to the concomitant action of root-zone salinity stress and high solar irradiance (Lu et al., 2003; Munns & James, 2003) although world's land area afflicted by soil salinity increases because of clearing (Munns, 2005). The knowledge of response mechanisms of both cultivated and wild species, particularly those distributed in Mediterranean-type ecosystems (Margaris, 1981), to the concerted action of excess soil salinity and high solar radiation is therefore of increasing significance from both a horticultural and an ecological point of view.
Basically, salt-induced decreases in leaf net CO2 assimilation rate because of stomatal and mesophyll limitation to CO2 diffusion (Evans & von Caemmerer, 1996), and because of reduced carboxylation efficiency (Allakhverdiev et al., 2000; Tattini et al., 2003), may induce severe conditions of ‘excess-light’ stress (Demmig-Adams & Adams, 1992) in plants growing at clearing sites, as the leaf's potential to use sunlight in photosynthetic processes is drastically reduced (Lu & Zhang, 2000; Wang et al., 2001). Furthermore, the problem of allocating potentially toxic ions out of highly sensitive cellular compartments (Niu et al., 1995; Hasegawa et al., 2000) may be aggravated in plants – particularly evergreen species (Munns, 2002) – growing at sunny sites, as transpiration and growth rates and, hence, the transport of salt in the shoot may increase because of high solar irradiance (Munns & James, 2003). As a consequence, species exposed to salty and sunny environments have to highly integrate mechanisms primarily intended to control (1) both the penetration and distribution of solar wavelengths in the leaf (Evans, 1999; Tattini et al., 2005) and (2) the allocation of ‘toxic ions’ at whole-plant and organismal level (Cheeseman, 1988; Colmer et al., 1995). The relative efficiency and the integration (Lambers et al., 1998; Tattini et al., 2005) of these suites of morpho-anatomical and physiological traits will be largely responsible for interspecific and intraspecific ‘tolerance’ to salinity and excess light stress under field conditions, as it may reduce both the generation of reactive oxygen species (ROS) (Hasegawa et al., 2000; Mittler, 2002) and damage to the photosynthetic apparatus (Allakhverdiev et al., 2000; Lu & Zhang, 2000; Tattini et al., 2005).
The increase in ROS content coupled with changes in net carbon gain because of light- and salt-stress may also strongly affect the biosynthesis of carbon-based secondary compounds, particularly leaf polyphenols (Mackerness et al., 2001; Babu et al., 2003; Tattini et al., 2004), which serve key functions in the ‘biochemical’ (Semerdjieva et al., 2003) strategies of most species to cope with harsh environments (Aerts, 1995; Tattini et al., 2005). Slow-growing, long-lived evergreen sclerophylls, such as those typically distributed in the Mediterranean basin, make a great investment in leaf phenolics (Margaris, 1981; Romani et al., 2002) to effectively counter damages from both pathogens and predators (Aerts, 1995; Dixon & Paiva, 1995), and at the same time, protect photosynthetic organs from the detrimental effects of highly energetic solar wavelengths (Tattini et al., 2000; Semerdjieva et al., 2003; Tattini et al., 2005).
Here we report data of an experiment conducted on Myrtus communis and Pistacia lentiscus, two sclerophyll evergreen shrubs (Brosse, 1979; Gucci et al., 1998; Mendes et al., 2001) widely used in herbal medicine in Mediterranean countries, and both recently determined to contain very high amounts of hydrolysable tannins and flavonoid glycosides (Baratto et al., 2003; Romani et al., 2004). Both species largely occur in Mediterranean coastal areas suffering from both drought stress and high solar radiation (Gucci et al., 1998; Mendes et al., 2001), but M. communis is sparsely distributed on seashore dunes, which are also frequently exposed to excess soil salinity, over the long summer season. To date, experiments aimed at elucidating response mechanisms of M. communis and P. lentiscus to either root-zone salinity stress or high solar radiation are surprisingly scarce (Mendes et al., 2001). To draw a comprehensive picture of the integrated mechanisms of response to the combined action of salinity stress and high solar radiation we estimated several traits related to morpho-anatomy, physiology and biochemistry, in plants exposed to 20% (referred as to shade plants throughout the paper) or 100% (sun plants) solar radiation, and supplied with 0 or 200 mm NaCl. In detail, we analysed: (1) leaf water and ionic relations; (2) gas exchange features estimated at both saturating light and on a daily basis in situ, and whole-plant growth rate; (3) basic leaf morphology (leaf size, leaf angle and leaf mass per area) and anatomy (whole-leaf and tissue layer thickness) together with leaf optical features; (4) the content and composition of photosynthetic pigments, the photochemical reflectance index (PRI) to estimate the radiation use efficiency, and the maximum potential efficiency of photosystem II (PSII) photochemistry (Fv/Fm); and (5) the leaf lipid peroxidation, the whole-leaf concentration and the leaf CO2-based accumulation of hydrolysable tannins and flavonoids, together with their tissue-specific distribution.
- Top of page
- Materials and Methods
Myrtus communis and P. lentiscus significantly differed in 84% of traits examined (see the Appendix, Table A1), except for root Na+ concentration (Naroot, on a tissue water molar basis), RWC, and both Pn and total chlorophyll on a leaf area basis (Chltot cm−2). We note, however, that both Pn and Chltot cm−2 underwent significant species–salt (as also observed for RWC) and species–light interactions. Changes in root zone salinity concentration significantly affected 69% of the examined traits, mostly ionic and water relations, and, consistently, gas exchange features. Salinity did not significantly alter morpho-anatomical and optical features of leaves, except for PRI. Finally, the total chlorophyll content (Chltot), both the lipid peroxidation (estimated on the basis of leaf MDA content) and the content of hydrolysable tannins varied significantly because of salinity stress (see the Appendix, Table A1). Overall, a highly significant effect of changes in solar irradiance was seen in 75% of the traits examined, particularly leaf morpho-anatomy and leaf optics, and both the content and composition of photosynthetic pigments (see the Appendix, Table A1). Noticeably, light treatment also dramatically affected ionic relations: both the Na+ concentration (on a tissue water molar basis) in different plant organs and the leaf K+ : Na+ ratio (K/Naleaf). Overall, changes in solar irradiance did not significantly affected gas exchange performances estimated at saturating light, as a result of significant salt–light and species–light interactions on most examined features. The light treatment affected the concentration of leaf flavonoids to a significantly greater degree than that observed for hydrolysable tannins (see the Appendix, Table A1). Finally, significant species–salt–light interactive effects were seen in: (1) both the leaf Na+ concentration (Naleaf, F = 225.1, df = 39) and the amount of Na+ allocated to the leaf tissue (Naleaf, F = 165.7, df = 39), which varied at a substantially greater degree in P. lentiscus than in M. communis (Figs 1–3); (2) gas exchange features (F = 10.4, 27.6 and 8.4 for Pn, gs and E, respectively, df = 159), PRI (F = 454.2, df = 63) and both the content (Chltot g−1 DW, F = 16.8 for df = 63) and the composition of photosynthetic pigments (Car : Chl, F = 37.5 for df = 63), which, in contrast, underwent larger changes in M. communis than in P. lentiscus (Fig. 1; see Table 2).
Figure 1. Normalized index of variation to changes in root zone salinity concentration (a, NIVsalinity) or solar irradiance (b, NIVsunlight) for 26 parameters (see Table A1 and text) related to (1) ionic and water relations; (2) photosynthetic features and growth rate; (3) both morpho-anatomical and optical traits, and photosynthetic pigments; (4) lipid peroxidation and polyphenol metabolism in Myrtus communis (open triangles) and Pistacia lentiscus (closed circles). Naorgan represents the amounts of Na+ in different plant organs (i.e. moles per organ), whereas K/Naleaf has been calculated on a tissue water cation molar basis. NIVs × 10 have been reported for Fv/Fm for illustrative purposes only.
Download figure to PowerPoint
Figure 2. Na+ fluxes (JNa, µmol g−1 root DW d−1) at the level of whole-plant (a) or leaf tissue (b) in Myrtus communis and Pistacia lentiscus exposed to 20% or 100% solar irradiance and supplied with 0 (open bars) or 200 mm NaCl (closed bars), over an 8-wk experimental period. Data are means ± SD, n = 5.
Download figure to PowerPoint
Figure 3. The Na+ concentration (on a tissue water molar basis) in leaf (circle) and stem (triangle) tissues of different age (apical to basal tissues) in Myrtus communis (a) and Pistacia lentiscus (b) exposed to 20% or 100% solar irradiance and supplied with 0 (open symbols) or 200 mM NaCl (closed symbols), over an 8-wk experimental period. Data are means ± SD, n = 5.
Download figure to PowerPoint
Species-specific responses to changes in root zone salinity concentration
The species-specific response to changes in root zone salinity stress has been reported on the basis of the normalized index of variation (NIVsalinity) for a total of 26 parameters (Fig. 1a). The amount of Na+ allocated to the leaf (moles per organ, Naleaf) in salt-treated P. lentiscus (NIV = 0.69) largely exceeded that in salt-treated M. communis (NIV = 0.35), whereas the amounts of Na+ in stem (Nastem) and roots (Naroot) varied similarly in both species due to changes in root zone salinity. In agreement with this, we observed that JNa,plant in salt-treated P. lentiscus exceeded by only 30% that in salt-treated M. communis (particularly in plants at the sun site, Fig. 2a), whereas the rate of Na+ transport to the leaf (i.e. JNa,leaf) was 3.5 times greater in the former than in the latter (Fig. 2b). These results were mostly due to markedly different leaf Na+ concentration (Naleaf, on a tissue water molar basis) in the species examined (Fig. 3), as the root to shoot ratio varied from 0.32 to 0.44 and from 0.37 to 0.49 in M. communis and P. lentiscus, respectively, because of salinity stress (data not shown). Indeed, Naleaf in salt-treated M. communis (Fig. 3a) accounted for just 42% Naleaf in salt-treated P. lentiscus, irrespective of solar irradiance (Fig. 3b). Furthermore, in salt-treated M. communis Na+ stem concentration (Nastem) was greater (+28%) than Naleaf, and the latter slightly superior (+22%) in basal than in apical leaves. By contrast, Naleaf did not vary because of leaf age in salt-treated P. lentiscus, and dramatically exceeded Nastem (Fig. 3). No species-specific response to salinity stress was detected for leaf the K+ : Na+ ratio, which decreased sharply (NIV = −0.69 for both species) from controls to salt-treated plants (Fig. 1a).
Myrtus communis and P. lentiscus also largely differed in their ability to counter salt-induced osmotic unbalance, as the salt-induced decrease in leaf ψw was substantially greater in M. communis (NIV = 0.50) than in P. lentiscus (NIV = 0.35), but changes in ψπ caused by salinity stress were similar in both species over the whole experimental period (Fig. 1a). Consequently, the loss of leaf turgor (ψp) was severe passing from controls (ψp = 1.15 MPa) to salt-treated M. communis (ψp = 0.61 MPa), whereas leaf ψp varied to a much lesser degree in P. lentiscus, from 1.23 MPa in controls to 1.02 MPa in salt-treated plants (Fig. 1a). Furthermore, salt-induced decreases in ψπ (Table 1) were mostly due to dehydration in M. communis (D accounted for 57% to salt-induced changes in ψπ) or net solute accumulation in P. lentiscus (D = 38%), since nonosmotic volume change (Gucci et al., 1997) contributed little (less than 5% in both species) to salt-induced decreases in ψπ. Moreover, the per cent contributions of and to ψπFT in salt-treated P. lentiscus (43%) largely exceeded those in salt-treated M. communis (27%), whereas the contribution of K+ to ψπFT of salt-treated leaves was similar in both species (Table 1).
The effect of salinity stress on gas exchange features measured at saturating light (i.e. at 600 and > 900 µmol m−2s−1 PAR for shade and sun leaves, respectively) was clearly species-specific (see the Appendix, Table A1). In detail, net assimilation rate (Pn), and both stomatal conductance (gs) and transpiration rate (E) decreased to a much greater extent in M. communis (on average NIV = −0.52) than in P. lentiscus (on average NIV = −0.28) in response to salinity stress (Fig. 1a, Table 2). Similarly, both net carbon gain (CO2tot) and transpiration rates (Etot), estimated on a daily basis, were also more severely depressed in M. communis , than in P. lentiscus and by salinity stress (Fig. 1a, Table 2). Salt-induced inhibition in the efficiency of PSII photochemistry (here estimated on the basis of the maximum potential efficiency of PSII in dark-adapted leaves, Fv/Fm) was appreciable only in M. communis, and Fv/Fm of salt-treated M. communis was 88% of that of P. lentiscus at the sunny site (Table 2). Finally, relative growth rate decreased more in M. communis (NVIRGR = −0.53) than in P. lentiscus (NVIRGR =−0.32) because of salinity stress, irrespective of solar irradiance (Fig. 1a, Table 2).
Table 2. Net CO2 assimilation rate at saturating light (Pn), net carbon gain over the experimental period (CO2tot), maximum efficiency of photosystem II (PSII) photochemistry (Fv/Fm), photochemical reflectance index (PRI) and relative growth rate (RGR) in Myrtus communis and Pistacia lentiscus in response to changes in root zone salinity concentration and solar radiation
|Species||NaCl (mm)||Sunlight (%)||Pn (µmol m−2 s−1)||CO2tot (mol m−2)||Fv/Fm||PRI||RGR (d−1 × 103)|
|M. communis|| 0|| 20||15.8 ± 1.0||16.5 ± 1.3||0.818 ± 0.008|| 0.015 ± 0.003||23.6 ± 3.1|
|200|| || 6.1 ± 3.6|| 5.6 ± 1.1||0.783 ± 0.019|| 0.006 ± 0.002|| 8.5 ± 1.8|
| 0||100||14.6 ± 0.9||20.8 ± 1.6||0.764 ± 0.017||−0.041 ± 0.006||29.4 ± 1.9|
|200|| || 4.2 ± 1.8|| 6.3 ± 0.9||0.693 ± 0.048||−0.079 ± 0.010|| 7.8 ± 2.4|
|P. lentiscus|| 0|| 20||11.6 ± 0.9||14.3 ± 0.8||0.824 ± 0.011|| 0.014 ± 0.002||17.6 ± 2.4|
|200|| || 6.9 ± 1.2|| 8.1 ± 1.5||0.821 ± 0.009|| 0.010 ± 0.002||11.5 ± 2.6|
| 0||100||14.9 ± 1.1||25.0 ± 1.8||0.808 ± 0.015||−0.009 ± 0.002||28.5 ± 2.2|
|200|| || 7.0 ± 2.4||12.6 ± 2.0||0.785 ± 0.027||−0.016 ± 0.004||12.3 ± 1.5|
Salinity stress poorly affected morpho-anatomical traits (see the Appendix, Table A1), as well as the absorption efficiency over the visible waveband (AVis) and the efficiency ((R/T)850) to scatter light (Fig. 1a; Table 3). Conversely, the photochemical reflectance index (which is negatively related to the content of xanthophyll cycle pigments, and positively related to the photosynthetic radiation use efficiency, Guo & Trotter, 2004) decreased to a much greater extent in M. communis than in P. lentiscus due to salinity stress, as also occurred for the net carbon gain (CO2tot, Tables 2 and 3). Moreover, both species had similar PRI when supplied with good-quality water at the shade site, but PRI was dramatically smaller in M. communis (−0.079) than in P. lentiscus (−0.016), when plants were concomitantly exposed to salinity stress and high solar radiation. Finally, the total chlorophyll content (on a tissue dry weight basis, Chltot g−1) did not appreciably vary in P. lentiscus, whereas substantially decreased in M. communis because of salinity stress (Fig. 1a).
Table 3. Morpho-anatomical and optical features in Myrtus communis and Pistacia lentiscus leaves in response to changes in root zone salinity concentration and solar radiation
|Species||NaCl (mm)||Sunlight (%)||LMA (g DW m−2)||Whole-leaf thickness (µm)||Cuticle thickness (µm)||Palisade to spongy ratio||AVis (%mg−1 DW)||(R/T)850 (%mg−1 DW)|
|M. communis|| 0|| 20|| 58.3 ± 3.4||184.8 ± 8.2||4.1 ± 0.4||0.41 ± 0.03||9.98 ± 0.15||0.14 ± 0.02|
|200|| || 64.1 ± 2.6||200.6 ± 12.9||4.1 ± 0.3||0.44 ± 0.03||9.01 ± 0.25||0.13 ± 0.02|
| 0||100||112.3 ± 6.7||309.8 ± 10.4||5.2 ± 0.2||0.54 ± 0.05||4.92 ± 0.32||0.09 ± 0.02|
|200|| ||113.5 ± 4.3||314.5 ± 25.2||5.3 ± 0.5||0.56 ± 0.02||4.75 ± 0.27||0.08 ± 0.01|
|P. lentiscus|| 0|| 20||110.5 ± 2.9||258.2 ± 11.6||7.6 ± 0.6||0.56 ± 0.04||4.92 ± 0.32||0.07 ± 0.01|
|200|| ||111.2 ± 2.4||251.1 ± 10.7||7.4 ± 0.3||0.54 ± 0.02||4.85 ± 0.14||0.07 ± 0.01|
| 0||100||183.5 ± 8.6||380.4 ± 23.6||8.1 ± 0.4||0.67 ± 0.03||2.82 ± 0.28||0.05 ± 0.01|
|200|| ||182.2 ± 9.8||372.0 ± 27.1||8.3 ± 0.4||0.64 ± 0.03||2.88 ± 0.24||0.05 ± 0.01|
Figure 4. The content of malondialdehyde (MDA, nmol g−1 DW) and the CO2-based accumulation of hydrolysable tannins (, mmol mol−1 CO2) and flavonoids (, mmol mol−1 CO2) in leaves of Myrtus communis and Pistacia lentiscus supplied with 0 (open bars) or 200 mm NaCl (closed bars) and exposed to 20% or 100% solar irradiance, over an 8-wk experimental period. Data are means ± SD, n = 4.
Download figure to PowerPoint
Species-specific responses to changes in solar radiation
Overall, changes in solar irradiance affected the whole set of examined parameters to a much greater extent in M. communis (mean NIV1,26 = 0.24 using absolute NIVs) than in P. lentiscus (mean NIV1,26 = 0.14), with the remarkable exceptions of Naleaf (mm), RGR, and both CO2tot and Etot (Fig. 1b). The light-induced increases in Naleaf were closely related with JNa,leaf, which varied from 10.4 to 20.1 µmol g−1 root DW d−1 in plants growing at the shade or sun site, respectively (Fig. 2a). As a consequence, the contribution of Na+ and Cl− to ψπFT was greater in sun (43%) than in shade (34%) salt-treated plants (Table 3), particularly in P. lentiscus (‘toxic ions’ accounted for 50% and 38% of ψπFT in sun and shaded P. lentiscus leaves, respectively, Table 1). Overall, water relations and gas exchange features at saturating light, underwent minor variations in both species because of high solar irradiance, as Pn, gs and E were more severely depressed by salinity stress at the sun than at the shade site (see the Appendix, Table A1; Fig. 1b), particularly in M. communis (Table 2, and data not shown). Similarly, light-induced reductions in both Fv/Fm and photosynthetic radiation use efficiency (i.e. PRI) were greater in M. communis than in P. lentiscus , particularly in plants concomitantly exposed to root zone salinity stress (Fig. 1b, Table 2). Both CO2tot and RGR (Fig. 1b; Table 2) consistently increased less in M communis (, NIVRGR = 0.07) than in P. lentiscus (, NIVRGR= 0.17), because of high solar radiation.
Pistacia lentiscus and M. communis also strikingly differed in their ‘constitutive’ (i.e. in leaves at the shade site) morpho-anatomical traits (see the Appendix, Table A1; Table 3). The LMA, the whole-leaf thickness and the thickness of the cuticular layer (data reported in Table 3 are means of adaxial and abaxial cuticle thickness) in M. communis accounted for 53, 71 and 54%, respectively, of corresponding features in P. lentiscus. Generally, leaf morpho-anatomical features mostly varied in response to changes in solar irradiance (see the Appendix, Table A1; Fig. 1b). Leaf size decreased and leaf angle increased to a much greater extent in M. communis than in P. lentiscus, as also occurred for the increases in both whole-leaf and cuticular thickness (and in the ratio of palisade to spongy parenchyma thickness), in response to high sunlight (Fig. 1b, Table 3). The efficiency to both absorb visible wavelengths (AVis) and to scatter light [(R/T)850] consistently decreased more in M. communis (on average NIV = −0.28) than in P. lentiscus (NIV = −0.20) because of high solar radiation, but still remained significantly higher in the former than in the latter (Fig. 1b, Table 3). Finally, both the content (Chltot g−1) and the composition of photosynthetic pigments (i.e. Chla : Chlb and Car : Chltot ratios) (1) were larger in M. communis than in P. lentiscus (Fig. 1b) and (2) underwent significant species–salt–light interactive effects. We found that Chltot g−1 ranged from 6.1 to 2.9 (Car : Chltot from 0.20 to 0.30) or from 3.1 to 2.2 mg g−1 DW (Car : Chltot from 0.23 to 0.24) from shade ‘control’ to sun ‘salt-treated’ plants of M. communis or P. lentiscus, respectively.
The increase in leaf lipid peroxidation – which was markedly greater in M. communis (+56%) than in P. lentiscus (+ 21%) – was accompanied by a parallel increase in the share of newly assimilated carbon allocated to secondary metabolites in response to high solar radiation (Figs 1b and 4a,b), although the polyphenol content did not appreciably vary between sun and shade leaves (NIV = 0.02, Fig. 1b; see the Appendix, Table A1). Light-induced accumulation (on the basis of assimilated CO2) was greater for flavonoids than for hydrolysable tannins (the : ratio varied from 0.05 in the shade to 0.09 in full-sun leaves, respectively), particularly in leaves of M. communis (Figs 1b and 4c). There was a close relation between the fluorescence profiles in both the blue (F440) and yellow-red waveband (F580), and light-induced increases of hydrolysable tannins and flavonoid glycosides (Fig. 4a,b). Indeed, the contribution of epidermal flavonoids to F580 was slightly greater in P. lentiscus than in M. communis, but in the latter, mesophyll flavonoids accumulated to a dramatically higher degree than in the former (Figs 5a,b). Profiles of F440 also showed that hydrolysable tannins accumulated to a greater extent throughout the whole depth in P. lentiscus (798 µmol g−1 DW as measured by high-pressure liquid chromatography (HPLC-DAD) than in M. communis leaves (572 µmol g−1 DW; see also the Appendix, Table A1, Fig. 4). However, we note that our fluorescence spectroscopy technique did not allow us compare the actual amounts of hydrolysable tannins and flavonoid glycosides, as the contribution of NR-stained hydrolysable tannins and NR-stained flavonoids at F440 and F580, respectively, may have differed sharply owing to markedly different absorption coefficients at the excitation wavelength (λexc = 365 nm) as well as to large variations in fluorescence quantum yields of individual polyphenol classes.
Figure 5. The tissue-specific distribution of hydrolysable tannins (F440) and flavonoids (F580) in Myrtus communis and Pistacia lentiscus leaves exposed to 100% solar radiation, over an 8-wk experimental period. Measurements were conducted on 50-µm-thick cross-sections stained with Naturstoff reagent and excited at 365 nm. Profiles of fluorescence intensity (from adaxial through abaxial tissues) were recorded by a charge-coupled device (CCD) camera, at the maximum emission wavelengths of tannins (440 nm, F440) and flavonoid glycosides (580 nm, F580), respectively. Ad and ab indicate adaxial and abaxial epidermal (including cuticle) layers, respectively; palisade and spongy refer to palisade parenchyma and spongy parenchyma layers, respectively.
Download figure to PowerPoint
- Top of page
- Materials and Methods
The data of our experiment may offer new insights in the mechanisms of response, which highly integrate morpho-anatomical, physiological and biochemical adjustments (Tattini et al., 2005) of woody species to excess soil salinity and high solar irradiance, and may help to explain the ecological distribution of M. communis and P. lentiscus in Mediterranean-type ecosystems.
First, we highlight the strikingly different strategies adopted by the species examined to cope with excess root zone salinity concentration, which mostly concern the allocation of salt at organismal level (see the Appendix, Table A1; Fig. 1a) and strongly affect leaf water relations (Gucci et al., 1997) and carbon acquisition (Cheeseman, 1988; Tattini et al., 2003). Pistacia lentiscus and M. communis may be roughly classified (Munns, 2002) as salt-including or a salt-excluding species, respectively, based on the relative contributions of JNa,leaf (68% and 28% in P. lentiscus and M. communis, respectively) to JNa,plant (Fig. 2). We note, however, that (1) the actual Na+ concentrations in the xylem sap (which may conclusively estimate the actual degree of ‘Na+ exclusion’; e.g. Munns, 2002) were not measured in the present experiment; and (2) the Na+ flux at whole-plant level differed no more than 30% in the species examined (Fig. 2a). Nevertheless, a massive allocation of Na+ to the leaf tissue taken together with (1) the lack of a gradient in Na+ concentration between leaves of different age (Fig. 3b) and (2) a leaf Na+ concentration (650 mm in apical leaves at the full-sun site) dramatically greater than that in both the stem tissue (on average 150 mm at the full-sun site; Fig. 3b) and in soil solution, led us to hypothesize that a very active secretion of Na+ in the leaf vacuole effectively operated in P. lentiscus (Niu et al., 1995; Hasegawa et al., 2000). Indeed, neither P. lentiscus nor M. communis have specialized organs to excrete toxic ions out of the leaf (Brosse, 1979). Conversely, the efficiency of mechanisms aimed to sequester potentially toxic ions in ‘less sensitive’ organs, such as the stem and the old leaves in M. communis (which largely contribute to confer ‘salt-tolerance’ in ‘salt-excluding’ species; see Colmer et al., 1995; Tattini & Gucci, 1999; Tattini et al., 2003; Munns, 2005), was markedly smaller (particularly at the full-sun site, Fig. 3a) than that previously reported for ‘salt-tolerant’, salt-excluding, Mediterranean evergreens, such as Olea europaea and Phillyrea latifolia (Gucci & Tattini, 1997; Tattini et al., 2003). Furthermore, nonstructural carbohydrates (mostly sucrose, glucose and fructose) (Lucchesini et al., 2001) in M. communis contributed little to salt-induced decreases in ψπFT (data not shown), in contrast with findings previously reported for salt-tolerant, mannitol-synthesizing Oleaceae (Tattini et al., 1996; Tattini et al., 2003).
Strikingly different abilities in using Na+ and Cl− to counter salt-induced osmotic unbalance (which, in turn, likely depended upon the relative abilities to sequester potentially toxic ions out of highly sensitive cellular compartments) (Niu et al., 1995; Hasegawa et al., 2000) were likely responsible for the markedly greater reductions in both photosynthetic performance and relative growth rates detected in M. communis than in P. lentiscus (Tattini et al., 1997, 2003; James et al., 2002). Indeed, although Naleaf increased to a dramatically greater degree in P. lentiscus than in M. communis (Figs 1b and 3), both the efficiency of PSII photochemistry (Lu et al., 2002; Lu et al., 2003) and the photosynthetic radiation use efficiency (Wang et al., 2001) were more severely depressed in the latter when both species suffered from concomitant conditions of salinity stress and high solar radiation (Table 2). We note that an increase in the mesophyll limitation to CO2 diffusion (Evans & von Caemmerer, 1996) may have been also responsible for significant salt–light interactive effects on photosynthetic performance in both species (Table 2; see the Appendix, Table A1), as sun leaves were thicker, had a major proportion of palisade to spongy tissue (Table 3), and reduced intercellular spaces (data not shown) than the shade counterparts (Miyazawa & Terashima, 2001; Niinemets et al., 2005).
Overall, reductions in net carbon gain and RGR taken together with (1) decreases in both the chlorophyll content (Greenberg et al., 1997) and the PSII efficiency (Lu & Zhang, 2000; Lu et al., 2002) and (2) increases in leaf oxidative damage (Shalata & Neumann, 2001) because of root zone salinity, allow us conclude that M. communis was actually ‘less tolerant’ than P. lentiscus to excess soil salinity (we note that individual parameters may fail in estimating ‘salt-tolerance’, which is both a multicomponent and a qualitative descriptor; see Cheeseman, 1988; Munns, 2002; Munns & James, 2003; Tattini et al., 2003).
Second, data presented here may allow drawing a comprehensive picture of how the suite of ‘constitutive’ (i.e. in leaves at the shade site) leaf morpho-anatomical traits (see the Appendix, Table A1; Table 3) in the examined species (which probably reflects the evolutionary adaptations of P. lentiscus and M. communis in areas of contrasting light and water availability, Lambers et al., 1998) may have actually determined the extent to which physiological and biochemical traits underwent adjustments in response to high solar irradiance (Gutschick, 1999; Semerdjieva et al., 2003; Tattini et al., 2005). In detail, the greater cuticle thickness, coupled with both a greater thickness and compactness (data not shown) of the palisade parenchyma tissue (Table 3), probably gave P. lentiscus a greater efficiency in both limiting the absorption of visible light – absorption over the visible waveband on a leaf area basis, i.e. AVis cm−2, was also smaller in P. lentiscus (85.1%) than in M. communis (90.7%) – and decreasing the pathlength of photons within the leaf (Table 2, Knapp et al., 1988). At the same time, mesophyll anatomy in P. lentiscus should have allowed a more equal distribution of highly collimated solar radiation in deeper tissues compared with M. communis (Evans, 1999). As a consequence, morphological adjustments primarily intended to avoid light from reaching leaf surface, such as the decrease in leaf size and the increase in leaf angle (Ehleringer & Comstock, 1987; Werner et al., 1999), operated to a much greater degree in M. communis than in P. lentiscus in response to high solar radiation (see the Appendix, Table A1; Fig. 1b). Similarly (1) loss of chlorophyll, which may also effectively help to reduce centres of light absorption and photo-damage to photosynthetic apparatus (Havaux & Tardy, 1999), (2) the increase in the Car : Chltot ratio (particularly xanthophyll cycle pigments; see PRI values in Table 3) aimed to dissipate thermally excess radiant energy (Demmig-Adams, 1998) and (3) the increase in the Chla : Chlb ratio probably linked to the reduction of the light-harvesting complex II (Kirchgeβner et al., 2003), served prominent roles in the acclimatization mechanisms of M. communis (McKinnon & Mitchell, 2003) to high solar radiation. The superior light-induced plasticity (Valladares et al., 2000) in morpho-anatomical traits, and in the content and composition of photosynthetic pigments (which probably contributed to limit the usage of solar radiation in photosynthetic processes and new growth to a greater extent in M. communis than in P. lentiscus at the full-sun site (Fig. 1b, Table 2, Falster & Westboy, 2003), did not preserve M. communis from a more severe oxidative damage to leaf membrane lipids (i.e. MDA) and photosynthetic apparatus (i.e. Fv/Fm) compared with P. lentiscus, particularly when plants were also exposed to salinity stress (Fig. 4, Table 2). The lack of a direct relationship between light-induced plasticity in morpho-anatomical traits and adaptation to excess solar radiation has been widely discussed in the very recent past (for a review, see van Kleunen & Fischer, 2004) and previously reported in some species differing in light demand (Valladares et al., 2000; Tattini et al., 2005).
Overall, species-specific responses to light treatment may allow one conclude that M. communis is ‘less tolerant’ than P. lentiscus to high solar radiation. This finding, taken together with significant species–salt–light effects on most photosynthetic features may offer a conclusive explanation for the very sparse distribution of the former in seashore dunes of the Mediterranean basin, where excess soil salinity may frequently occur in concomitance with clear days, over the long summer season.
Our final discussion concerns the potential roles served by hydrolysable tannins and flavonoids in the biochemical strategies adopted by M. communis and P. lentiscus to cope with high solar radiation and salinity stress (Figs 4 and 5). First, we show that the share of newly assimilated carbon allocated to polyphenols markedly increased in response to both salinity stress and high solar radiation (Fig. 4), although neither stressors appreciably affected the whole-leaf polyphenol content (see the Appendix, Table A1; Tattini et al., 2004). These findings fit with a widely reported plant behaviour to allocate carbon preferentially for defence than for new growth when exposed to environmental constraints of different origin (Lavola et al., 2000; Kandil et al., 2004). We suggest, therefore, that the whole-leaf polyphenol content seems inadequate to estimate environment-induced alterations in leaf biochemistry (Tattini et al., 2004), although it is a key determinant of plant–environment interactions (Aerts, 1995; Kandil et al., 2004).
The salt- and light-induced greater increases in the allocation of carbon to flavonoids than to hydrolysable tannins (Figs 1 and 4) appear of particular interest. The light-induced preferential accumulation of flavonoids over that of hydrolysable tannins (Tegelberg et al., 2004) may have depended simply upon relative UV-spectral features, as galloyl derivatives have absorption maxima in the low UV-B waveband (270–285 nm, Romani et al., 2002) and offer a negligible protection against UV-A wavelengths (which mostly accounts for the UV-waveband of solar radiation), compared with that due to both myricetin and quercetin glycosides, which effectively absorb wavelengths in the range 330–370 nm (Tattini et al., 2004). We also hypothesize that, the greater concentration (and CO2-based accumulation) of flavonoids in the mesophyll of M. communis than in P. lentiscus (Fig. 5) probably depended on a reduced ability of the former to limit the flux of UV-radiation in the leaf because of both a thinner cuticle and a smaller contribution from epidermal flavonoids (Day, 1993; Krauss et al., 1997). However the fact that (1) flavonoid glycosides accumulated to a greater extent than did hydrolysable tannins because of salinity stress (Fig. 1a); and (2) there was a close relationship between leaf oxidative damage and preferential accumulation of flavonoids, irrespective of stressful conditions (Figs 1 and 4), supports the idea that myricetin and quercetin glycosides probably played antioxidant roles in M. communis and P. lentiscus exposed to salt and light stress (Tattini et al., 2004; Tattini et al., 2005). Flavonoids have the potential for serving antioxidant functions in tissues exposed to both high UV-radiation (Rice-Evans et al., 1996; Ryan et al., 1998) and oxidative damages of different origin (Schoch et al., 2001; Babu et al., 2003). However, flavonol 3-O-glycosides and galloyl derivatives do not actually differ in their in vitro abilities to scavenge free-radicals, such as the hydroxyl or the superoxide radicals (Rice-Evans et al., 1996; Baratto et al., 2003), and, hence, the high ‘constitutive’ content of hydrolysable tannins (we note that carbon allocated to hydrolysable tannins was at least one order of magnitude greater than that allocated to flavonoids irrespective of growing conditions; Fig. 4a,b) in the mesophyll cells should have been in large excess of that necessary to quench light- or salt-induced highly reactive oxygen species in both species. However, tannins (Bussotti et al., 1998) and flavonoid glycosides (Yamasaki et al., 1997; Takahama, 2004) have been shown to have a nearly exclusive vacuolar distribution (wall-bound flavonoid glycosides have been also reported in some species, Schnitzler et al., 1996; Agati et al., 2002), and therefore, both polyphenol classes have merely to reduce hydrogen peroxide, which may freely diffuse from cellular organelles to enter the vacuole under conditions of severe oxidative damage (Yamasaki et al., 1997; Takahama, 2004). We therefore hypothesize that the preferential accumulation of flavonoids with respect to that of hydrolysable tannins might be linked to a superior ability of the former class to serve as substrates for class III peroxidases, which reduce hydrogen peroxide in the presence of phenolics to form more stable phenoxyl radicals in cell vacuoles (Yamasaki et al., 1997; Takahama, 2004). The electron-donor ability of flavonol 3-O-glycosides, however, has been recently questioned by Pearse et al. (2005), as 3-O-glycosylation may both preserve flavonoids from autoxidation and increase their solubility in the vacuolar milieu, but, at the same time, reduces sharply their abilities to be oxidized via the hydrogen peroxide–peroxidase system compared with those of corresponding aglycones (Hodnick et al., 1988; Yamasaki et al., 1997). Future experiments should be aimed at addressing the complex issue of the functional roles of polyphenols in light of the chemical structure–electron-donor ability relationship for class III peroxidases.