Leaf senescence is a complex phenomenon occurring in all plant species, but it is still poorly understood in plants grown in Mediterranean field conditions and well-adapted to harsh climatic conditions. To better understand the physiological processes underlying leaf senescence in mastic trees (Pistacia lentiscus L.), we evaluated leaf growth, water and N content, photosystem II (PSII) photochemistry, lipid peroxidation and levels of photosynthetic pigments, antioxidants, abscisic acid, and salicylic acid and jasmonic acid during the complete leaf lifespan, from early expansion to late senescence in relation to natural climatic conditions in the field. While mature leaves suffered from water and N deficit during late spring and summer, both young (emerging) and old (senescing) leaves were most sensitive to photo-oxidative stress, as indicated by reductions in the Fv/Fm ratio and enhanced lipid peroxidation during late autumn and winter. Reductions in the Fv/Fm ratio were associated with low α-tocopherol (vitamin E) levels, while very old, senescing leaves additionally showed severe anthocyanin losses. We have concluded that both young (emerging) and old (senescing) leaves suffer oxidative stress in mastic trees, which may be linked in part to suboptimal temperatures during late autumn and winter as well as to low vitamin E levels.
Leaf development, from bud break to death, is a finely controlled process at the molecular, biochemical and physiological levels. The two extremes of leaf lifespan are characterized by rapidly growing, young emerging leaves on one hand, and very old, senescing leaves on the other. While studies of senescing leaves are common, their physiology is still poorly understood. Furthermore, there are few experiments to better understand the processes underlying what happens in emerging young leaves. Leaf senescence is characteristic of the latest stages of development and involves nutrient remobilization to other plant parts and some degenerative changes that lead these organs to death. It is usually characterized by three phases: (i) an initiation phase triggers the process in fully-expanded young leaves (also called mature, non-senescing leaves). This is followed by (ii) a remobilization phase, which is governed at the molecular level and allows nutrient remobilization accomplishing one of the most important functions of the senescing process in leaves. Finally, (iii) a terminal phase occurs when nuclei and the nucleic acid-related machinery is destroyed and the leaf can no longer accomplish any physiological role (Munné-Bosch and Alegre 2004; Lim et al. 2007; van Doorn 2011; Fischer 2012).
Reactive oxygen species (ROS) generation is common to all aerobic organisms including plants, and occurs in the plant's response to various stresses, including water deficit, high light, low temperatures or salinity, etc., as well as during leaf senescence. The process of protein degradation during leaf senescence, which occurs to a high extent in chloroplasts, is initiated by ROS and involves the action of proteolytic enzymes (Bhattacharjee 2005; Khanna-Chopra 2011). Chloroplasts function at high oxygen tensions and in the light, therefore displaying a high photoprotective demand and a strong photo-oxidative potential. They can produce superoxide anions as a consequence of the direct reduction of oxygen by ferredoxin in the so-called Mehler reaction in PSI, but can also give rise to the more reactive singlet oxygen as a result of the interaction of oxygen with triplet excited chlorophylls (3Chl*). Indeed, it has recently been shown that singlet oxygen is one of the most potentially damaging ROS molecules in chloroplasts (Triantaphylidès et al. 2008). Superoxide anions can be rapidly converted by superoxide dismutases to hydrogen peroxide, which is in turn metabolized to water by the so-called ascorbate-glutathione cycle (Wang et al. 2010; Foyer and Noctor 2011). In contrast, singlet oxygen can be destroyed by the concerted action of α-tocopherol (vitamin E) and carotenoids, mainly β-carotene (pro-vitamin A), in photosystem II (PSII) reaction centers (Trebst 2003), therefore avoiding photodamage to PSII. Another mechanism that is thought to protect leaves from photo-oxidative stress is the accumulation of anthocyanins, which might act as a light-barrier and therefore avoid over-excitation of the photosynthetic apparatus (Steyn et al. 2002). However, the direct role of anthocyanin accumulation, which occurs generally in emerging and senescing leaves of several species, is presently under debate (Gould et al. 2009).
On the other hand, abscisic acid (ABA), jasmonic acid (JA) and salicylic acid (SA), which are known to play a major role in plant responses to environmental stress, are also generally involved in senescing processes in the leaves of several species. It has been shown both in the model plant Arabidopsis thaliana and in other plant species that ABA, JA, and SA are involved in promoting the senescing process, both in developmentally-regulated and stress-induced leaf senescence (He et al. 2002; Buchanan-Wollaston et al. 2005; Lim et al. 2007; Schippers et al. 2007; Abreu and Munné-Bosch 2009; Lee et al. 2011; Fischer 2012). However, little is known about the possible role of these phytohormones in very young, emerging leaves, although it has been known for a long time that bud break is associated with a physiological stress. And that a decrease in ABA levels is needed in dormant buds so that leaves can emerge (Wright 1975).
We have previously shown that mastic trees activate mechanisms of photo– and antioxidative protection, presumably for maintaining chloroplast function during the first stages of leaf senescence, while antioxidant defenses are lost during the latest stages of senescence (Munné-Bosch and Peñuelas 2003). On the other hand, it has been shown in other species that young, emerging leaves show a high photoprotective demand and are therefore very sensitive to photo-oxidative stress (Jiang et al. 2005). However, it is still unknown whether this is the case in young leaves accumulating high anthocyanin levels, such as those of mastic trees. The study of the physiology of mastic tree leaves at the two extremes of the leaf lifespan is very important and is a good example to better understand the ecophysiology of Mediterranean species well-adapted to survive harsh climatic conditions. In this study, we aimed at evaluating to what extent (i) young, emerging leaves might be as sensitive as old, senescing leaves to oxidative stress in this species, (ii) whether this is linked to specific climatic variables that change over the seasons under Mediterranean field conditions, and (iii) can this be associated with intrinsic changes in the levels of some photoprotectants, such as anthocyanins, carotenoids and tocopherols, or to endogenous levels of the phytohormones ABA, SA and JA.
Climatic conditions during the experimental period (from November 2009 to November 2010) were typical of the Mediterranean climate (Figure 1). Suboptimal growth temperatures (mean monthly maximum and minimum temperatures below 20 °C and 10 °C, respectively) were combined with abundant precipitation (387 mm between October and March) during autumn and winter, while the summer (particularly between June and August) was characterized by scarce rainfalls (94 mm during this period) and maximum mean monthly temperatures above 30 °C in July. The maximum diurnal photosynthetically-active photon flux density (PPFD) and monthly precipitation occurred in May (Figure 1). Taking into account these changing climatic conditions over the course of the seasons, we followed the physiological processes underlying the development of mastic tree leaves from very young (emerging) leaves in November 2009 to very old, senescing leaves in November 2010. Very small, emerging leaves were marked and the same leaves were followed throughout the seasons until their deaths. Therefore, observed changes in physiological parameters were caused by the interaction of both the leaf's age and the climatic conditions that the plants were exposed to over the year. The lifespan of leaves was of approximately 12 months, with some variability between plants. It is worth noting that sampling during November 2010 was performed on the plants for which leaves still persisted, since some of the leaves had already dropped by that time and could not be sampled. Therefore, this sampling point represents leaves in a very advanced developmental stage, just before leaf abscission occurred (here called very old, senescing leaves).
Leaf growth, water, and nutrient content
Leaf growth (estimated as dry mass) increased sharply from 14.1 mg to 43.8 mg during the first stages, particularly between January and May, but only very slightly between November and January. Maximum leaf biomass was attained in May, after which it remained constant (Figure 2). The leaf mass per area (LMA) ratio increased sharply during early stages of growth from November to May, and most notably between January and March, to decline later in July and increase again in November of the next year. The relative water content (RWC) increased slightly during the first stages of development to decline later in May. RWC values remained above 90% in January and March and close to 80% between May and September, with a slight decline at the end of the experiment (Figure 2). The marked decrease in the RWC between March and May was accompanied by a 35% decrease in the N content of leaves and a 52% increase in the C/N ratio (Figure 3). Indeed, N content decreased and the C/N ratio increased from the start of leaf development in November until May, and remained constant thereafter. C content remained unchanged throughout the experiment, with values ranging between 50.3% and 53.8% (data not shown).
PSII photochemistry and photosynthetic pigments
The Fv/Fm ratio followed a marked variation throughout the seasons, with minimum values at the two extremes of the leaf lifespan. Values below 0.75 were attained both at the beginning of leaf development (between November and March, with the lowest values of 0.69 attained during November and January), and at the very end of the leaf lifespan (with values of 0.68 during November 2010, Figure 3). The Fv/Fm ratio remained above 0.75 between May and September, with maximum values of 0.81 attained in September. It is interesting to note that the lowest Fv/Fm values were observed during late autumn and winter and at the two extremes of the leaf lifespan. However, these variations did not correlate with the contents of photosynthetic pigments. Chlorophyll a+b levels decreased from the onset of leaf development, and most particularly during the first months of leaf development. The minimum chlorophyll levels were attained in July and thereafter remained constant until the end of the leaf lifespan (Figure 4). In contrast, the chlorophyll a/b ratio followed a different pattern. This ratio increased during the early stages of leaf growth to decrease sharply from March to May, and later remained constant until the end of the leaf lifespan. Carotenoid levels followed a similar pattern of variation of chlorophyll, but the decline started later (Figure 4).
Antioxidants and lipid peroxidation
Aside from the lipid-soluble carotenoids (Figure 4), we analyzed other antioxidant molecules including tocopherols, and anthocyanins, and evaluated the extent of lipid peroxidation by estimating the levels of malondialdehyde (MDA) (Figure 5). Among the tocopherols analyzed (α, β, γ and δ-tocopherol), only the α homolog could be properly quantified in mastic tree leaves. α-Tocopherol levels increased gradually with leaf development, attaining maximum levels of 63.3 μg (per g dry weight (DW)) during September, 6.2-fold higher than at the beginning of the experiment. However, the levels of this antioxidant were drastically reduced (by 93%) in very old, senescing leaves (from September to November). The values in these leaves were even 55% lower than those observed at the beginning of the experiment in recently emerged leaves. The lowest values in anthocyanin levels were also observed in very old, senescing leaves. However, in this case, recently emerged leaves accumulated large amounts of these compounds, 4.6-fold higher than in very old, senescing leaves. It is worth noting that anthocyanin accumulation in young leaves was maintained throughout late autumn and winter. On the other hand, anthocyanin levels suffered a drastic decrease not only during spring (between March and May), but also during late autumn in old leaves (September to November), concomitantly with drastic reductions in tocopherol levels (Figure 5). Interestingly, MDA levels peaked at the two extremes of the leaf lifespan (Figure 5), opposite to the variations in the Fv/Fm ratio (Figure 3). The highest MDA levels correlated with the lowest Fv/Fm values, except for September, in which lipid peroxidation increased (Figure 5) without any effect on the Fv/Fm ratio (Figure 3). It is also worth noting that the extent of lipid peroxidation gradually decreased in developing young leaves, reaching minimum levels during May and July. Furthermore, when comparing the two extremes of the lifespan, we observed that MDA levels were 1.7-fold higher in recently emerged leaves when compared with very old, senescing leaves (Figure 5).
The levels of antioxidants per unit of chlorophyll (Figure 6) revealed a different pattern of variation compared to that given when values were expressed in dry mass (Figures 4 and 5), particularly for carotenoids and anthocyanins. Recently emerged leaves showed a 28% increase in carotenoids per chlorophyll unit from November to January (due to the reduction of chlorophylls but not of carotenoids), concomitantly with increases in tocopherol and anthocyanins per chlorophyll unit, although in these two latter increases lasted until March. Young, emerging leaves (November 2009) showed low levels of antioxidants (tocopherol, carotenoids and anthocyanins) per chlorophyll unit, at least compared to fully expanded leaves (May), while the lowest levels of antioxidants (in this case tocopherol and anthocyanins, but not carotenoids) per chlorophyll unit were observed again in very old, senescing leaves (November 2010).
Endogenous ABA, SA, and JA content were also followed during the entire leaf lifespan throughout the year (Figure 7). The most striking differences were again observed in young, emerging leaves, particularly between November and January during which ABA, SA, and JA decreased. However, the major changes were observed for ABA, which decreased by 74% during this period. It is also interesting to note a 7-fold increase in SA levels between March and May, which occurred concomitantly with a 50% increase in ABA levels. Finally, it is worth mentioning that ABA, SA, and JA all showed no significant increase during autumn in very old, senescing leaves (Figure 7).
Mastic tree (P. lentiscus) is a deep-rooted, evergreen, dioecious shrub or small tree of up to 5 m tall that produces new leaves both in spring and autumn. Autumn leaves were selected here to monitor physiological changes throughout the year, and most particularly at the two extremes of their lifespan, when leaves are exposed to similar climatic conditions of autumn/winter and can therefore suffer photo-oxidative stress due to suboptimal temperatures – an aspect poorly investigated thus far in Mediterranean plants. It should be noted that the age-related changes in physiological parameters shown here reflect the changes in the so-called autumn leaves, which appear in late summer/early autumn. Spring leaves (appearing during late winter/early spring), which were not investigated here, may not necessarily follow the same physiological changes during leaf development, an aspect that warrants further research. Another important point is that the effects of leaf age cannot be separated from the influence of climatic conditions. Although this is somewhat limiting in establishing a causal relationship, we believe this approach is essential to understand the ecophysiology of this species. With this approach we can observe what really occurs to the leaves of this plant species during its complete lifespan in natural field conditions. Of particular interest, it was shown that leaves at the two extremes of their lifespan show stress symptoms (most particularly oxidative stress), and also important differences in terms of phytohormones (mainly ABA) and antioxidants (mainly tocopherols and anthocyanins).
Clear symptoms of oxidative stress were observed in emerging leaves, as reflected by MDA accumulation, an indicator of the extent of lipid peroxidation (Hodges et al. 1999), and Fv/Fm values below 0.75, which are generally indicative of damage to PSII photochemistry (Takahashi and Murata 2008). However, it is possible that low Fv/Fm ratios were associated with an immature chloroplast, and not necessarily a damaged PSII in rapidly growing, emerging leaves, which requires further investigations. Interestingly, the Fv/Fm ratio stayed close to 0.70 throughout late autumn and winter (between November and March), while MDA concentrations decreased gradually during the same period, suggesting that reduced Fv/Fm values are probably linked to low temperatures and/or immature chloroplasts, and the degree of lipid peroxidation is reduced as leaves expand and develop. In other words, it appears that the youngest, emerging leaves are the most sensitive to lipid peroxidation under the suboptimal low temperatures typical of late autumn and winter. It is interesting to note that the Fv/Fm ratio also decreased in very old, senescing leaves, which also encountered similar low temperatures typical of late autumn in the Mediterranean climate, suggesting a direct effect of low temperatures on the induction of photoinhibition in this species. Indeed, this is in agreement with Nikiforou et al. (2011), who showed that despite accumulating anthocyanins, leaves of this species are sensitive to cold-induced photoinhibition during the winter. Therefore, it appears that accumulating anthocyanins, which might act as light screens or directly as antioxidants (Gould et al. 2009), does not confer young, emerging leaves complete photoprotection. In this study, it is shown that recently emerged leaves accumulate anthocyanins, but have low levels of carotenoids and tocopherols per chlorophyll unit, therefore displaying a low photoprotective capacity in relation to the potential amount of light absorbed by chlorophyll molecules, an aspect considered very important to counteract excess light (Kyparissis et al. 1995; Munné-Bosch and Alegre 2000; Munné-Bosch et al. 2001). It therefore appears that recently emerged leaves are not only in a physiological state with reduced photosynthetic capacity as previously shown (Nikiforou et al. 2011), but also exhibit a reduced capacity for photoprotection by carotenoids and tocopherols, which are essential to maintain the integrity of PSII (Trebst 2003). It is still unknown whether or not these young, emerging leaves of mastic trees would suffer a higher degree of photo-oxidative stress if they did not accumulate anthocyanins.
Similar to recently emerging leaves, senescing leaves also experience oxidative stress, but with some important differences. In this case, lipid peroxidation increased first, well before the Fv/Fm ratio was significantly reduced. This indicates that oxidative stress is induced in senescing leaves and that photo-oxidative damage occurs in these leaves, but at the latest stages only, concomitantly with a strong depletion of tocopherols and anthocyanins. Oxidative stress therefore appears not to be induced (or at least not to be solely induced) by high light, since mature leaves exposed to very high PPFDs showed the lowest MDA accumulation over the entire leaf lifespan. It has previously been shown that this species is very resistant to summer drought typical of the Mediterranean climate, by inducing low osmotic potential and by reducing transpiration rates, partly due to the presence of thick leaf cuticles (Zohary 1962). It is therefore not surprising that plants could maintain RWC values above 80% throughout the leaf lifespan, except in very old, senescing leaves, in which values fell just slightly below this threshold which is considered essential to keep full cell turgor. It is likely that the combination of drought and high light typical of the Mediterranean summer are triggering the senescing process, as suggested previously (Munné-Bosch and Peñuelas 2003). In this previous study, however, drought during the summer was more severe, which induced a higher degree of oxidative stress and accelerated the senescence process compared to the present study. It is also worth mentioning that the MDA increases in senescing leaves occurred before α-tocopherol levels dropped, thus indicating that a decrease in this antioxidant is not the only cause of oxidative stress, but probably the consequence of increased oxidative stress within the senescing cell. Further research is therefore required to unravel the signaling events and molecular players leading to MDA accumulation in senescing leaves of this species.
Abscisic acid levels were very high in recently emerged leaves, decreasing sharply from November to January. It is worth noting that ABA decreases did not correlate with leaf biomass increases; the depletion of ABA levels occurred earlier than the increase in leaf biomass. It is also noteworthy that the highest MDA levels coincided with the lowest Fv/Fm ratio and the highest ABA levels in recently emerged leaves only, but not during leaf expansion. A decrease in ABA levels may be a pre-requisite for leaf growth, since ABA is generally considered a growth inhibitor (Zhou et al. 2003). Furthermore, it is well known that phytohormones are involved in the regulation of leaf senescence, but their specific activities can vary depending on the species and very little is known about leaf senescence in natural field conditions, despite its importance to better understand the ecophysiology of plants. ABA, SA, JA levels did not increase during the progression of senescence in mastic tree leaves. This is not in agreement with previous findings that support an effect of ABA, SA, and JA as promoters of leaf senescence in several species (Morris et al. 2000; Buchanan-Wollaston et al. 2005; Abreu and Munné-Bosch 2008, 2009). It therefore appears that leaf senescence in this species is not triggered by any of these senescence-promoting compounds.
We conclude that both young (emerging) and old (senescing) leaves suffer oxidative stress in mastic trees, which may be linked to suboptimal temperatures during late autumn and winter as well as to low vitamin E levels, among other factors, in both leaf types.
Materials and Methods
Plant material, growth conditions and sampling
This study was conducted using Pistacia lentiscus L., an evergreen species widely distributed along the Mediterranean basin. Fifty-five juvenile plants with a height between 40 and 110 cm were purchased in Bioriza (Cornellà de Terri, Girona, Spain) in the spring of 2009 and were homogeneously transplanted in an area of 30 m2 to the experimental fields of the Faculty of Biology at the University of Barcelona (Barcelona, Spain). Plants were grown under Mediterranean climatic conditions and received water exclusively from rainfall during the study period. All new, emerging leaves of four newly emerged shoots of each plant were labeled (28 and 29 October 2009) to know the leaf age at the moment of the measurements. Leaf samples were collected from the labeled leaves every 2 months from November 2009 to November 2010 at midday (at maximum incident diurnal PPFD). For analysis of MDA, photosynthetic pigments, anthocyanins, tocopherols and phytohormones, samples were collected, immediately frozen in liquid nitrogen, and stored at −80 °C until analysis. Despite destructive analyses, the large number of leaves labeled at the beginning of the experiment allowed us to follow the exact age of the leaves throughout the experiment.
Leaf biomass, RWC, LMA and elemental analyses
Leaf biomass was measured by weighing the samples before and after drying to constant weight at 80 °C. The relative water content (RWC) was determined as (FW-DW)/(TW-DW), where FW is the fresh matter, TW is the turgid weight after hydrating the leaves with distilled water for 24 h at 4 °C, and DW is the dry matter after oven-drying the samples. Leaf area was estimated by using a flatbed scanner (model CX- 5400; Epson Stylus, Nagano, Japan) and an image-processing program. Leaf mass area (LMA) was measured as the ratio of leaf area per DW. Total C and N concentrations in leaves were measured by the Dumas elemental analysis method, using a Thermo EA 1108 analyzer (Thermo Scientific, Milan, Italy).
Photosynthetic pigments, anthocyanins, and tocopherols
Leaf samples (100 mg) were ground in liquid nitrogen and extracted with 100% methanol using ultrasonication. After centrifuging at 16 000 ×g for 10 min at 4 °C, the pellet was re-extracted with the same solvent and supernatants were pooled. Chlorophylls and carotenoids were estimated spectrophotometrically, and specific absorption coefficients for chlorophyll a, chlorophyll b, and total carotenoids reported by Lichtenthaler (1987) were used. A molecular weight of 570 for carotenoids was used for calculations. Anthocyanin content was determined following extract acidification with concentrated HCl, and an absorption coefficient of 30/mM per cm at 530 nm was used according to Gitelson et al. (2001).
For the extraction of tocopherols, leaf samples (50 mg) were ground in liquid nitrogen and extracted with 100% methanol using ultrasonication for 45 min at 4 °C. The samples were then centrifuged for 15 min at 4 °C and the supernatants were transferred to vials for analysis. The high performance liquid chromatography (HPLC) analysis was carried out as described (Amaral et al. 2005). In brief, the HPLC equipment consisted of an integrated system with a Waters 600 controller pump, a Waters 714 plus auto-sampler, and an FP-1520 fluorescence detector (Jasco, Essex, UK). Tocopherols were separated on an Inertsil 100A (30 × 250 mm, 5 μm, GL Sciences Inc. Torrance, CA, USA) normal-phase column operating at room temperature. The mobile phase used was a mixture of n-hexane and 1,4-dioxane (95.5:4.5, v/v) at a flow rate of 0.7 mL/min, and the injection volume was 10 μL. Detection was carried out for excitation at 295 nm and emission at 330 nm. Among the analyzed tocopherols (α-, β-, γ- and δ-tocopherol) only α-tocopherol could be quantified based on the fluorescence signal response compared with an authentic standard (Sigma-Aldrich, St. Louis, MO, USA).
Lipid peroxidation and chlorophyll fluorescence
The extent of lipid peroxidation was estimated by measuring the amount of MDA in leaves by the method described by Hodges et al. (1999), which takes into account the possible influence of interfering compounds in the thiobarbituric acid-reactive substances (TBARS) assay.
Measurements of the maximum efficiency of photosystem II photochemistry (Fv/Fm ratio) were made by using a pulse modulated fluorimeter Imaging PAM (Walz, Effeltrich, Germany) after 2 h of dark adaptation in leaves collected at midday. The Fv/Fm ratio was calculated as (Fm-F0)/Fm, where Fm and F0 are the maximum and basal fluorescence yields, respectively, of dark adapted leaves as described earlier (van Kooten and Snel 1990).
The extraction and analyses of endogenous concentrations of ABA, SA, and JA were carried out as described by Müller and Munné-Bosch (2011). Deuterium labeled phytohormones (d6-ABA, d4-SA, and d5-JA) were used as internal standards.
Seasonal variations were evaluated using the analysis of variance (anova) and were considered significant at a probability level of P < 0.05.
(Co-Editor: Hai-Chun Jing)
We are very grateful to the Serveis Científico-Tècnics and Serveis dels Camps Experimentals (Universitat de Barcelona) for technical assistance. This work was supported by the Spanish Government (BFU2009–07294). Support for the research was also received through the prize ICREA Academia given to S.M.-B., funded by the Generalitat de Catalunya.