Università degli Studi di Ferrara

Transcription

1 Università degli Studi di Ferrara DOTTORATO DI RICERCA IN BIOLOGIA EVOLUZIONISTICA E AMBIENTALE CICLO XXVI COORDINATORE Prof. Guido Barbujani Biochemical and molecular analyses of key enzymes controlling salt stress response in Italian rice (Oryza sativa L.) varieties Settore Scientifico Disciplinare BIO/04 Dottorando Dott. Michele Bertazzini Tutore Prof. Giuseppe Forlani Anni 2011/2014

2 ABSTRACT Rice (Oryza sativa L.) is a vascular monocotyledon that belongs to the family of Poaceae. Two are the most important rice species for human nutrition: O. glaberrima, that grows mainly in West Africa, and O. sativa, grown worldwide. Both of them are diploid. This cereal has an undisputed economic and social value, and played one of the most important role in Asia s customs and traditions. The early domesticated rice faced a series of ecological and climate conditions that led to the selection of some new traits. Other selective pressure came from farmers and traders, which selected and crossed new varieties. This scenario caused the development of two main variety groups: indica and japonica. The first group includes varieties adapted to grow in tropical areas, the second one consists of genotypes typical of tropical uplands and temperate regions. Nowadays rice is cultivated in all continents with the exception of Antarctica and its production takes place in a wide range of locations and climatic conditions. Rice is a staple food for more than one half world population. FAO statistics report that in 2013 a total amount of 745 million tonnes of paddy rice were produced in more than 100 countries. Italy is the only country in Europe where a significant land area is used for rice production, and Italian germplasm comprises not less than varieties belonging to the japonica ssp. Soil salinity is one of the main constraints to crop yield. The area affected by excess salt continuously increases because of the climate change and wrong agricultural practices. Nowadays about 6% of the Earth s total surface area is affected by salt. By definition, a soil is considered saline when ionic concentration exceeds an electrical conductivity of 4 ds m -1, a value that corresponds to about mm. Intensive efforts have therefore been carried out during recent years to increase salt tolerance of crop species. Unfortunately, our knowledge of natural mechanisms that allow some plants to withstand salt stress conditions is still limited. Plants can cope with moderate salinity conditions either by limiting uptake at the root level, by compartmentalizing/extruding ions in the vacuole/apoplast, or by counteracting the consequent water withdrawal through the intracellular accumulation of compatible osmolytes, among which the amino acid proline has the widest occurrence. At increasing soil conductivity, salinity stress causes a progressive reduction of the photosynthetic rate and stimulates the production of reactive oxygen species (ROS), which in turn lead to oxidative damages at the cellular level. To overcome salt-induced damages, plants detoxify ROS by upregulating antioxidative enzymes. The essential role of antioxidative systems to maintain a balance between ROS overproduction and their scavenging to keep them at signaling level for reinstating metabolic homeostasis has been well established. Intracellular accumulation of high proline levels as a stress protectant has long been reported in many plant species. Although not conclusively analyzed experimentally, under high nitrogen availability proline seems to be synthesized mainly from ornithine, which is converted to δ1-pyrroline-5-carboxylate (P5C) by an ornithine-δ-aminotransferase. In contrast, under (hyper)osmotic stress conditions and/or nitrogen starvation P5C is produced from glutamate by a bifunctional P5C synthetase. The two pathways share the last reaction, in which P5C is reduced to proline by a P5C reductase. Rice is very sensitive to excess salt, mainly at the seedling level, as well as during the transition to the reproductive stage, when salinity can severely disrupt grain formation and yield. Early papers showed that a conductivity value as low as 3 ds m -1 was already inhibitory. More recently, however, experimental evidence accounted for an even higher sensitivity, with a damage threshold of 1.9 ds m -1. Because of the agronomical relevance, many studies investigated the occurrence of a differential sensitivity to salt stress among rice cultivars. However, virtually all studies dealt with Asian rice genotypes, whereas no information has been made available to date with respect to the European rice germplasm. In order to fill this and other gaps of knowledge, a research program named RISINNOVA was launched in 2012 and funded by Ager-Fondazioni in rete per la ricerca agroalimentare, a collaborative project among Bank Foundations targeted to support excellence in research in the agriculture and agro-food area. Its general objective was the development of genetic/genomic approaches to afford development of an innovative Italian rice producing chain, to provide producers and industry with more competitive varieties suitable for inner as well as international 1

3 markets. In the frame of this program, the present work aimed at a biochemical analysis of key enzymes controlling salt stress response in a group of three to five Italian rice cultivars with a contrasting capability to cope with excess salt during the vegetative and reproductive stages. The levels and the properties of selected enzymes playing a pivotal role in antioxidant defense would have been studied in these genotypes, as well as the resulting levels of ROS, reduced ascorbate/gsh and free proline in tissues of osmotically-stressed seedlings. With the aim to identify suitable genotypes for assessing whether properties and/or activity levels of enzymes involved in proline or antioxidative pathways may contribute to increased salt tolerance, a group of 17 Italian rice cultivars were screened for the ability to germinate and grow in the presence of increasing conductivity values. Instead of using NaCl as the osmoticum, their sensitivity was assessed by complementing MS medium with a salt mixture mimicking the composition of the soil solution. Seed germination was affected only over a threshold of about 8.5 ds m -1. Seedling growth was much more sensitive, with a significant reduction of both fresh and dry biomass at conductivity values as low as 4 ds m -1. Results allowed the calculation of the concentrations able to inhibit growth by 50% (ID 50 ), and of the threshold of damage, defined as the conductivity value at which 10% reduction of plant growth was evident. Data showed the occurrence of a significant variability among Italian rice germplasm with respect to the susceptibility to salt stress. To obtain further information, the effect of salt stress conditions was investigated also at the plant cell cultures level. Interestingly, in this system cell growth was mildly affected at conductivity values at which the growth of seedlings had been totally abolished. Data seem to suggest that in the presence of a non-limiting energy source the cell is able to withstand highest external concentrations of salts. Moreover, even though the number of genotypes tested is limited, the relative susceptibility to salt was exactly the opposite than that previously pointed out at the seedling level. The comparison of the inhibition brought about by the addition of the salt mixture with that exerted by an equivalent amount of NaCl showed completely not overlapping patterns, with a strikingly higher ability to withstand the latter conditions. On the whole, results suggest that different (or supplemental) mechanisms are involved in planta to confer the ability of growing in the presence of excess salt in the soil, strengthening a possible pivotal role for the regulation of intertissutal transport, and compartmentalization or exclusion strategies. Whatever the meaning, results ruled out the possibility of using in vitro systems to elucidate the biochemical basis of a differential tolerance to hyperosmotic stress. On the other hand, data prompted us to verify whether significantly different results would be attained by using NaCl alone instead of the (more realistic) salt mixture utilized during the initial screening. With this aim, the same plate assay was applied to increasing NaCl concentrations. Data obtained in the case of 6 representive genotypes showing differential susceptibility to salt stress were normalized with respect to actual conductivity values, and compared to those previously found with salts. Data clearly showed that, at least in some cases, the resulting patterns are significantly different. A general lower sensitivity toward NaCl alone was evident, and differences as high as 40% were found with respect to growth reduction under the two experimental conditions. Since the presence of NaCl only in salt-affected environments is quite uncommon, this implies that an exclusively NaCl-based screening for tolerant cultivars could lead to misleading inferences, and to a serious overestimation of the ability of a given genotype to grow in the present of excess salt in the soil. Finally, the possibility that the results obtained with the simplified plate assay method employed could be at least partially inconsistent with those occurring with plants grown under field conditions was considered. To rule out this possibility, the sensitivity to increasing salt concentration of a subset of 6 rice cultivars was measured on seedlings grown in a growth chamber up to the three leaf stage. On the whole, data showed a high concordance, and strengthened the reliability of the approach based upon the rapid plate assay. On this basis, genotypes were divided into three groups: highly salt sensitive (threshold below 3 ds m -1 ), sensitive (ranging from 3 to 6 ds m -1 ) and mildly tolerant (over 6 ds m -1 ). Two representative cultivars from each group were selected as the material to be used with the aim to identify the molecular determinants of the differential susceptibility. 2

4 These genotypes were at first characterized with respect to cation content under normoosmotic and hyperosmotic stress conditions, to verify whether salt tolerance may depend upon exclusion mechanisms. In collaboration with the Laboratory of Prof. G.A. Sacchi at the University of Milan, Na, K, Mg, Ca, Mn, Fe, Cu, Zn and Mo levels were determined in shoots from individual plantlets grown in the absence or in the presence of inhibitory, yet sublethal levels of the salt mixture. Shoot analysis yielded consistent results, and allowed us to compare cation concentration in genotypes differing in salt susceptibility under the same conditions in which their ability to withstand salt stress had been evaluated. No striking differences were evident with respect to Na + content, that increased similarly in all rice cultivars. Statistically significant variations were indeed found among genotypes, but these were completely unrelated to the relative tolerance to salt stress. The experimental approach employed does not allow us to distinguish between cations present in the cytoplasm from those confined into the vacuole, where for instance the replacement of K + by Na + does not induce toxic effects. In any case, results suggested that the tested cultivars do not owe their differential ability to cope with excess salt to mechanism(s) for reduced Na + translocation to the aerial part, or for its increased transport back to the roots mediating Na + exclusion from leaves. The same experimental approach was applied to study plants that had been grown under normoosmotic conditions and then exposed to excess salt. Very similar results were obtained 24 and 48 h after the treatment. Once again neither this pattern, nor the consequent sodium-to-potassium ratio showed a significant relationship with the relative salinity tolerance of rice cultivars. High Na + concentration inside the cell can lead to ROS formation and oxidative damage. To verify this possibility, seedlings of the same genotypes were grown at increasing conductivity values, and the consequent lipid peroxidation was determined by measuring the thiobarbituric acid reactive substances (TBARS). Results showed that a significant increase of TBARS content indeed occurs in salt-stressed plants. Quite unexpectedly, such an increase was similar in all cultivars, showing a doubled TBARS content at the highest dose tested. These results seem to suggest that in all cases rice genotypes are able to maintain lipid peroxidation within a certain limit, irrespectively of the ability to grow under salt stress. To overcome damages of salt-mediated ROS, plants up-regulate antioxidative enzymes. However, the determination of specific activity levels in extracts from control and stressed plants is hampered by the presence of numerous isoenzymatic forms with different expression pattern in planta. To obtain a first set of data and evaluate whether increased levels of a given antioxidative enzyme do occur in tissues of salt-stressed rice seedlings, an immunological approach was chosen. Roots and shoots from five uniformly-grown seedlings of the representative rice cultivars were extracted. After denaturing PAGE, Western blot analysis allowed to obtain protein expression patterns. No significant variations were found with respect to GSH reductase and Fe-SOD; in the case of ascorbate peroxidase, a contrasting pattern was evident for the plastidial (increasing) and the cytosolic (decreasing) isozyme, but the overall content remained unchanged. A slight decrease was found for catalase isozyme A. On the contrary, an increase up to 70% of the levels in untreated controls was evident for Cu/Zn-SOD and Mn-SOD that resulted the most responsive enzymes. Similar results were obtained with extracts from roots. A preliminary statistical analysis did not show any significant difference among genotypes concerning these patterns. However, the experimental approach used provides only semi-quantitative data, and the possibility that enzyme activities would not correspond to protein levels (for instance because of a different catalytic efficiency) cannot be ruled out. Taking one further step forward, the activity of the two most responsive enzymes was measured in extracts from seedlings grown under salt stress conditions. Unexpectedly, only minor changes in Cu/Zn- and Mn-dependent SOD activity levels were found, without any relationship with either the severity of the stress or the relative sensitivity to excess salt of a given rice cultivar. Activities were completely unrelated to the corresponding protein levels. Such a discrepancy might be due to a partial inactivation of the enzyme during catalysis, one that can lead to a transient accumulation of inactive protein. Some papers indeed reported that in the presence of H 2 O 2 the Cu/Zn-dependent SOD can function as a peroxidase, leading to its auto-inactivation. This hypothesis is further strengthened by the significant lowering of activity levels that occur in some genotypes at conductivity values at which TBARS content showed the presence of high ROS 3

5 concentrations inside the cell. However, to obtain a conclusive evidence as to this point, the measurement of gene expression levels should be also performed under the same experimental conditions. As an alternative, ROS scavenging ability may depend on the regeneration of antioxidant molecules, among which ascorbic acid and glutathione play a major role. To gain some information, ascorbic (AsA) and dehydroascorbic (DHA) acid level, and reduced (GSH) and oxidized (GSSG) glutathione content were measured in shoots of rice seedlings. Results showed that in both cases the amount of the oxidized form increases with the severity of the stress. However, such an increase was relatively low, and no relationship was evident between these traits and the corresponding ability to withstand excess salt, although statistical differences were found among genotypes. More interestingly, a remarkable increase of total glutathione and total ascorbate content was found, that in the former case was strictly proportional to the conductivity value at which the plants had been grown. No clear relationships were evident in the present study with respect to ascorbate. On the contrary, a correlation was found between the percent increase of total glutathione and the resistance to excess salt: the more resistant the cultivar, the higher the increase. Indeed, it has been hypothesized that glutathione metabolism may play a key role in determining the degree of expression of defense genes controlled by several signalling pathways both before and during stress. Therefore the possibility exists that an increased total glutathione content may confer an increased ability to withstand salt stress. On the other hand, plants can cope with moderate salinity conditions also through the intracellular accumulation of compatible osmolytes, the most common of which is the amino acid proline. The possibility that the differential tolerance to excess salt of the selected Italian rice cultivars may rely on a different ability to accumulate proline was then considered. In preliminary experiments performed on a limited number of genotypes, intermediate and mildly tolerant cultivars showed higher proline accumulation in the leaves and were responsive to lower salt concentrations than sensitive cultivars. On this basis, a more comprehensive experiment was planned, in which all the 17 cultivars available were analysed. In a completely randomized design, seedlings were germinated and grown for two weeks under either normoosmotic conditions or in the presence of a salt level corresponding to about 16 ds m -1. The resulting concentration of free proline in shoots was measured in extracts prepared independently from 10 uniformly-grown plantlets. Significant differences were evident among cultivars both in stressed and in non-stressed seedlings, suggesting that a natural variability is present in rice germplasm concerning the mechanisms underlying proline homeostasis. However, when these levels were plotted against an index of the susceptibility to salt stress, such as IC 50 values, no plain relationship was found considering either the basal or the salt-induced proline content. This notwithstanding, when the same data were plotted as the salt-induced increase in free proline, a completely different pattern was evident. A highly significant relationship (P < 0.001) was found between this trait and the tolerance of rice cultivars to salt stress expressed as the damage threshold. This suggests that the higher the ability to accumulate the imino acid in response to hyperosmotic stress conditions, the higher the capability to withstand the negative effects potentially rising from the presence of a high salt concentration in the substrate. The existence of a causal relationship between proline accumulation and salt tolerance has been hypothesized in several studies, also on rice. However, in most studies a limited number of cultivars was considered, and contrasting results were obtained. To the best of our knowledge, this is the first case in which such a correlation has been obtained on a large set of genotypes. In order to better understand the possible functional significance of proline accumulation, similar experiments were conducted either under drought (simulated through the addition of PEG 6000 to the culture medium) or early after the establishment of salt stress conditions. As a term of comparison, the same trials were performed with tobacco seedlings, a species that is well known using proline accumulation to achieve osmotic adjustment. In rice plants grown under normoosmotic conditions, proline levels rapidly raised following the treatment with increasing NaCl concentrations. However, also the level of other free amino acids increased likewise, and the percentual content of proline showed only slight variations, if any. On the contrary, under similar experimental conditions tobacco seedlings showed a marked and specific increase of free proline, 4

6 which accounted for up to 25% of total amino acids, with a 5-fold increase over the basal level. On the whole, data suggested that while in tobacco a specific synthesis of proline does occur following the exposure to stress conditions, in rice an aspecific increase of proline content may depend on a general decrease of protein synthesis and/or an increase of proteolysis. On the contrary, in rice seedlings directly sown in the presence of inhibitory concentrations of salt the homeostatic level of most amino acids was maintained, with the only exception of proline, which increased from 7% to 16% of total amino acids. These results therefore suggest that, contrary to other plant species, proline accumulation in rice is not a rapid mechanism for osmotic adjustment, but it represents a longer-term adaptation to cope with the effects of excess salt in the environment. Since proline has been reported to play a role as a radical scavenger, it could protect the cell from the oxidative damage derived by salt-induced ROS generation. A better comprehension of the protective role of proline under salt stress conditions would benefit from a proper characterization of the enzymes involved in the metabolic pathways responsible for its synthesis and re-oxidation. With this aim, the available cdna clones for the enzymes of interest were obtained from the Rice Genome Resource Center, National Institute of Agrobiological Sciences DNA Bank, Japan. They were subcloned in the expression vector pet151 by directional TOPO cloning during a stage in the laboratory of Dr. Dietmar Funck (Department of Plant Physiology and Biochemistry, University of Konstanz, Germany). For two of them, namely those coding for P5C dehydrogenase and P5C reductase, heterologous expression allowed to obtain functional proteins, which were affinity-purified and characterized at the functional level. In order to assess whether changes of ion homeostasis in plants grown in saline environment may influence proline metabolism, the effect of increasing concentrations of various salts on their catalytic rate was evaluated. In the case of P5C dehydrogenase, activity was progressively inhibited by Na + and Mg 2+ cations in the 10-2 to 10-1 M range, whereas the same levels of K + ions were ineffective. Contrary to what it is believed, Cl - anions were inhibitory only at concentrations exceeding 150 mm. Interestingly, the enzyme was found to be susceptible also to micromolar concentrations of other mono- and divalent cations. Zinc, copper and manganese have in fact been detected at micromolar levels in higher plant mitochondria, and some evidence has been described supporting the possibility that their fluctuations may modulate activity of mitochondrial enzymes. For P5C dehydrogenase a preferential usage of NAD + was found, but the same susceptibility to a given ion was evident when NADP + was the electron acceptor. On the contrary, in the case of P5C reductase the effect of ions was strongly dependent on the use of NADH or NADPH as the electron donor. When increasing levels of either NaCl, MgCl 2 and KCl were added to the standard reaction mixture, a significant inhibition of the NADHdependent activity was found. Conversely, when NADPH was the electron donor a remarkable stimulation was evident. Interestingly, the stimulatory effect of low salt concentrations on the NADPH-dependent reaction was much stronger than the inhibitory effect on the NADH-dependent reaction. For instance, the presence of either NaCl or KCl at 100 mm caused a 40%-inhibition of the NADH-driven reaction, but a 150%-stimulation of the NADPH-dependent activity. Being NADPH the preferred co-substrate, this could result in increased proline synthesis in salt-loaded cells. On the whole, the results herein described provide a first insight into the mechanisms adopted by Italian rice varieties to cope with salt stress conditions. Some general conclusions can be drawn: a rapid screening method in Petri dishes can provide reliable results in order to screen for salttolerant rice genotypes: a good concordance between the results obtained and the current trend of rice cultivar use in salt-affected regions of Northern Italy was indeed observed the use of a complex salt mixture is strongly advisable to study the plant response to salt stress, since a NaCl-based screening for tolerant cultivars could lead to an overestimation of the ability of a given genotype to grow in the present of excess salt 5

7 cell suspension cultures cannot be used as an alternative experimental system, since the response to salt stress at the undifferentiated tissue level differs both quantitatively and qualitatively from that at the seedling level considering the Italian germplasm, differences in rice salt tolerance are almost negligible at high conductivity values, whereas higher variations are detectable at low salt concentrations, under realistic field conditions. To attempt the identification of the molecular determinants of the differential tolerance to excess salt, some representative varieties were characterized with respect to several parameters potentially able to counteract the detrimental effects of salt stress conditions, and some conclusions can be inferred: at least in the genotypes examined, the differential tolerance does rely upon neither the ability to limit ion uptake at the root level, nor to reduce transport rates to the aerial part by means of transporters capable of retrieving Na + from the transpiration stream concerning the function of selected antioxidant enzymes, under salt stress conditions the protein level of the Mn-dependent and Cu/Zn-dependent SOD isozymes significantly increased with the severity of the stress, suggesting a role in quenching ROS and minimizing the deriving cell damage; however, enzyme activity data were not consistent, suggesting that a partial inactivation may occur as a consequence of ROS detoxification; in any case, no clear patterns were evident that can explain the differential tolerance interestingly, a trend was observed under stress with respect to glutathione and proline levels: the more salt-resistant is the cultivar, the higher is the increase that occur inside the cell as to these metabolites; considering that both can have antioxidative effect, this result may suggest that a higher ability to counteract oxidative damages corresponds to an increased tolerance to excess salt this possibility was strengthened in the case of proline by the results obtained with a much larger group of rice varieties contrary to what is commonly believed, proline accumulation in rice seems to play a role in a long-term adaptation to salt stress conditions, and does not represent a short-term mechanism for osmotic adjustment; this conclusion is also strengthened by the amount of free amino acid that is accumulated inside the cell: although highly significant from a statistic point of view, such an increase seems in fact not enough to counteract the presence of high Na + concentrations in the apoplast although still preliminary, the results of the characterization of two proline-metabolizing enzymes are consistent with this overall picture, in that the activity of the anabolic P5C reductase is stimulated by the presence of salts, whereas that of the catabolic P5C dehydrogenase is inhibited. These results represent the first dataset ever produced with the aim to clarify the mechanisms for salt tolerance inside the Italian rice germplasm. Data are susceptible to provide the breeders with new molecular markers for the development of more resistant varieties with better performances under salt stress, a condition that is forecasted to increase in the next future in response to climate changes. 6