POTENTIAL OF THE PHYSIOLOGICAL RESPONSE OF PEA PLANTS ( Pisum sativum L . ) TO IRON DEFICIENCY ( DIRECT OR LIME-INDUCED )

Pea (Pisum sativum L.) is an important food crop in Tunisia, where calcareous soils represents the major limiting factor for agriculture production. In the present study a greenhouse experiment was conducted to assess the effects of direct and bicarbonateinduced iron deficiency on plant growth, chlorophyll fluorescence, photosynthesis, spad index and iron nutrition in two Tunisian pea genotypes (Pisum sativum L.). Plants were grown hydroponically and iron deficiency was induced for 3 weeks. Iron deficiency decreased all the above physiological parameters. The direct Fe deficiency is more drastic than bicarbonateinduced Fe deficiency. A close relationship between plant growth, photosynthesis and SPAD index was observed. Fe use efficiency for plant growth and Fe use efficiency for photosynthesis discriminates clearly the studied genotypes and seems to be the main reason of the tolerance of Kelvedon, as compared to Lincoln.


INTRODUCTION
In calcareous soils, which constitute the major part of cultivated land, the soil solution does not provide more than 10% of the plant requirements for Fe (MORTVEDT, 1991).The total Fe content in these soils is high but the available fraction for the plant is insufficient.This is caused by the very low solubility of iron oxides at the alkaline pH conditions that are buffered by the presence of bicarbonate in these soils.Ammari and Mengel (2006) indicated that Fe concentration found in calcareous soil was high enough to meet the plant's demand and prove that Fe chlorosis on such soils is not a question of the Fe availability in the soil.
Iron (Fe) is known to be essential for many physiological and biochemical processes such as: photosynthesis, respiration, DNA synthesis, nitrogen assimilation and symbiotic nitrogen fixation and participates in the electron transfer through reversible redox reactions, cycling between Fe 2+ and Fe 3+ .Fe deficiency affects the structure, development and function of the entire photosynthetic apparatus (ABADIA et al., 1999).It has been shown that Fe deficiency decreases lightharvesting pigments, particularly chlorophylls (MORALES et al., 2000), and promotes antenna disconnection in PSII (MORALES et al., 2001).Furthermore, Fe-deficient leaves show lower PSII efficiency and a decrease in the proportion of open PSII reaction centers (qp ) (LARBI et al., 2006).Salahi et al. (2017) demonstrated that foliar iron sprays improve the performance of oriental plane tree in calcareous soil better than soil treatments.
Differences among species and genotypes in plant response to iron deficiency have been reported.Krouma et al. (2008) and Slatni et al. (2009) reported clear genotypic differences in the response of common bean to iron deficiency.Soybean seed yield decreased by 20% per unit of visual leaf chlorosis (FeDC) when grown on calcareous soil (FROEHLICH;FEHR, 1981).Various authors reported that Fe resupply to deficient plants restores many plant functions.For instance, it leads within a few days to increases in chlorophyll concentration and photosynthetic activity in several annual species, including sugar beet (LARBI et al., 2004) and tobacco (PUSHNIK; MILLER, 1989).Paula et al. (2016) observed in Lotus japonicus that interveinal chlorosis was associated with a reduced Fe 2+ shoot content in all sensitive ecotypes and a decline in photosynthesis rate and PSII performance compared to the control.In some apple cultivars, bicarbonate treatment reduced active and total Fe and total chlorophyll concentrations, and FCR (leaf ferric chelate reductase) activity (SAHIN et al., 2017) The objective of this study was to examine the physiological parameters associated with the tolerance of pea to iron deficiency in order to establish useful test for screening program, and to assess the relationships between plant growth, SPAD index, photosynthesis, chlorophyll fluorescence and iron nutrition.

Plant material and experimental conditions
Two pea genotypes were used, Kelvedon and Lincoln largely cultivated in the North and North West of Tunisia.Seeds were disinfected with 2% hypochlorite calcium solution then rinsed in deionized water.Germination was made in Petri dishes containing moistened filter paper for 6 days at 20 °C.Seven-day-old seedlings were then transferred to a half strength aerated nutrient solution for 7 days and then similar sized seedlings were selected and cultured as groups of 10 plants in 10 L of full strength aerated nutrient solution.
After 21 days of treatments, SPAD index, chlorophyll fluorescence and gas exchange parameters were measured, and then plants were separated into shoots and roots, dried at 60 °C for 72 hours, and then pulverized into a fine powder.

SPAD Index
The degree of chlorosis was estimated nondestructively in the youngest fully expanded apical leaves from five plants of each treatment using a portable SPAD-502 meter (Minolta, Osaka, Japan).Five SPAD readings were recorded for each leaf, homogeneously distributed from the apex to the base of the leaf, to obtain a representative degree of leaf chlorosis.

Active iron
Measurements of active iron (Fe 2+ ) were performed according to Köseoglu and Acikgöz (1995) The extraction was made in 25 mg of leaves fine powder shacked in 10 ml of 1N HCl.

Gas exchange measurement
Gas exchange measurements were made with an LI-6400 (LI-COR, Inc.) portable gas exchange system.Measurements were made on the 3 youngest fully expanded leaves.Photosynthesis was induced with saturating light (1000 µmol m -2 s - 1 ).This light was fitted to the standard 6-cm 2 clamp on the leaf chamber.Sample pCO 2 , flow rate, and temperature were kept constant at 362 mbar, 500 µmol.s -1 , and 25 °C, respectively.

Chlorophyll fluorescence measurements
Prior to the measurements, the attached leaves were dark adapted for 30min in leaf-clips.Values for maximum fluorescence (F m ) and initial fluorescence (F 0 ) from the fluorescence induction curve were measured with a portable chlorophyll fluorometer (OS1-FL).Photosynthetic photon flux density (PPFD) was lower than 0.4 µmol m -2 s -1 at the leaf surface.F m was measured at 20 kHz with a 0.8 s pulse of 6000 µmol m -2 s -1 of white light (Morales et al., 1998).

Calculations
Fe use efficiency for plant growth (FeUEDW) was expressed as the ratio of biomass production to Fe accumulation in leaves [g dry weight.µmol -1 Fe] and Fe use efficiency for net assimilation (FeUEAn) was expressed as the ratio of net assimilation to Fe accumulated in leaves [(µmol CO 2 m -2 s -1 ).µmol -1 Fe].

Statistical analysis
Variance analysis of data (one-way ANOVA) was performed using the SPSS 10.0 program, and means were separated to Duncan's test at p≤ 0.05.Data shown are means of five repetitions (photosynthetic parameters, SPAD index and chlorophyll fluorescence) or nine (Biomass and Fe content) replicates for each treatment.

Plant growth and SPAD index
All plants subjected to iron deficiency exhibited a clear decrease of biomass production.This effect was more important when plants are cultivated on Fe-free medium.Nevertheless, the negative effect of iron deficiency on plant growth is more pronounced in Lincoln than Kelvedon.The decrease of biomass production was estimated to 8% and 40% in Lincoln subjected to direct or bicarbonate-induced Fe deficiency, respectively; and 9% and 17% in Kelvedon subjected to direct or bicarbonate-induced Fe deficiency, respectively (fig 1).
The SPAD index values (table 1) followed the same scheme of variation of plant growth parameters.Iron deficiency decreased SPAD index, the effect is more drastic in direct than indirect Fe deficiency.Kelvedon remain the less affected genotype as compared to Lincoln.The values of SPAD index decreased with 11 % and 16 % in Kelvedon and with 24 % and 28 % in Lincoln, respectively subjected to direct or bicarbonateinduced iron deficiency.     .The calculation of non-photochemical quenching (NPQ) demonstrates an important effect of iron deficiency in Lincoln (-10% in bicarbonateinduced and -21% in direct iron deficiency) but not in Kelvedon (-6% only in direct iron deficiency) (table 2).

Iron nutrition
The analysis of the extractible active fraction of Fe in leaves (fig 5) demonstrated that iron deficiency induced a significant decrease of this micronutrient.When subjected to induced Fe chlorosis, this decrease was estimated to 35% and 33%, respectively in Kelvedon and Lincoln, while reaching 54% and 52%, respectively in Kelvedon and Lincoln, subjected to direct iron deficiency.This result indicated a clear problem of iron allocation to leaves.The calculation of the Fe use efficiency for plant growth and photosynthetic activity demonstrated that this parameter increased with iron deficiency in Kelvedon and remain without significant changes in Lincoln (table 3).The first genotype expressed higher efficiency of Fe use in Fe depletion conditions as compared to Lincoln.

DISCUSSION
Typically, Fe and P are the two main nutrients that limit plant growth on calcareous soils (MARSCHNER, 1995).Our results show that iron deficiency decreased biomass production and SPAD index in the two studied genotypes and significantly inhibit active iron (FeII) accumulation.Direct Fe deficiency is more drastic than bicarbonate-induced Fe deficiency and Lincoln is usually more affected than Kelvedon.As documented by several authors (ZOCCHI et al., 2007;KROUMA et al., 2008) deficiency adversely affected plant growth and shoot length in several plant species.In fact, Fe is known to be essential for many physiological and biochemical processes such as: photosynthesis, respiration, DNA synthesis, nitrogen assimilation and symbiotic fixation.iron is shown indispensable element for chlorophyll and carotenoids biosynthesis (THOIRON et al., 1997), the photosynthesis (JELLELI et al., 2011) as well as the metabolism of plastidial proteins (SPENCE et al., 1991) and therefore, the induced iron chlorosis and the decrease of SPAD index observed in this study can be explained by a drastic decrease of iron availability for these organs.İncesu et al. ( 2015) observed significant differences in SPAD and iron chlorosis scale reading between rootstocks and Fe treatments.In fact, fig 6 which correlates the leaves Fe content with their SPAD index demonstrated a close relationship between these two parameters (R 2 = 0.99).The genotypic variability observed in this study was found in common bean subjected to iron deficiency (KROUMA et al., 2008).It has been shown that Fe deficiency decreased light-harvesting pigments, particularly chlorophylls (MORALES et al., 2000); and promotes antenna disconnection in PSII (MORALES et al., 2001;MOSELEY et al., 2002).Furthermore, many micronutrients (e.g.Fe, Mn, Cu and Zn) that are freely available in acid soils are only sparingly available in calcareous soils, due to their poor solubility at high pH (BRADY and WEIL, 1999).Experiments have shown calcifuges plants (those which cannot establish well on calcareous soils e.g.pea and common bean) to be primarily excluded from growth in calcareous soils due to poor P and Fe use efficiency (KERLEY et al., 2001).The calculation of the Fe use efficiency for plant growth (FeUEDW, table 3) expressed as the ratio of biomass production to Fe accumulation in leaves (g DW. µmol -1 Fe) screened clearly the studied genotypes.The values of FeUEDW where 2 times more important in bicarbonate-induced Fe deficiency plants as compared to control one's and 3 times in direct Fe deficiency.The most attractive result at this level is that Kelvedon develop more important efficiency of iron use than Lincoln with the same quantities of iron accumulated in leaves when subjected to iron deficiency.The FeUEDW is 1.2 and 1.4 times higher in Kelvedon than Lincoln when subjected to induced and direct Fe deficiencies, respectively.This efficiency seems to be the origin of Kelvedon tolerance.The measurements made on gas exchange parameters and chlorophyll fluorescence show that independently of its origin, iron deficiency decreased net assimilation and other photosynthetic parameters.The maximum quantum yield of PSII is significantly affected by iron deficiency in Lincoln but not in Kelvedon.
Genotypic differences previously observed were maintained and Kelvedon develops a better preservation of its photosynthetic apparatus.Previous works demonstrated that the maximum quantum yield of photosystem II decreased in Fe deficient leaves of citrus (PESTANA et al., 2005), pear (DONNINI et al., 2009) and peach rootstocks (MOLASSIOTIS et al., 2006).Similar results were found by Donnini et al. (2009) showing different reorganization of the photosynthetic apparatus between tolerant and sensitive genotypes of pear and quince cultivated in the presence of bicarbonate.In addition, Sharma (2007) investigated the adaptation of photosynthesis under Fe deficiency in maize plants and suggested an involvement of nuclear-chloroplast signaling in mediating adaptive changes in the photosynthetic machinery triggered by redox status and possibly, accumulation of chlorophyll biosynthesis intermediates.Kara (2016) concluded that low temperature, high net photosynthesis rate, high internal CO 2 concentration/ambient CO 2 ratio and low transpiration rate might be used as reliable selection criteria in further triticale breeding programs.In the present study, the calculation of Fe use efficiency for photosynthesis (FeUEAn, calculated as the ratio of net assimilation to Fe accumulated in leaves, µmol CO 2 m -2 s -1 .µmol -1 Fe) (table 3), demonstrated a clear increase of this parameter in plants subjected to iron deficiency.Kelvedon maintain its performance, as compared to Lincoln, in the two Fe deficiency origin with values of FeUEAn 1.4 times more important than lincoln.
It appears clearly that FeUEDW and FeUEAn discriminates the two studied pea genotypes.The better efficiency of Kelvedon genotype, compared to Lincoln one, gives us a new explanation of its performance in a limiting iron availability condition like calcareous soil.In fact, our results suggested that the tolerance of Kelvedon is probably linked to two parameters: firstly a better ability to allocate more iron to shoots to maintain photosynthetic activity and plant growth; and secondly, to its efficiency of iron use.We suggest that Fe use efficiency for plant growth and Fe use efficiency for photosynthesis might be used as reliable selection criteria in further plant breeding programs.

Figure 1 .
Figure 1.Biomass production in pea plants subject to bicarbonate-induced or direct iron deficiency.Vertical bars represent ± standard errors of means of 9 replicates, p ≤ 0.05.

Figure 2 .
Figure 2. Net assimilation (An) in two pea genotypes subjected to direct (-Fe) or induced Fe chlorosis (+Fe +biC) as compared to control treatment (+Fe).Vertical bars represent ± standard errors of means of 5 replicates, p ≤ 0.05.

Figure 4 .
Figure 4. Internal CO 2 concentration in two pea genotypes subjected to direct (-Fe) or induced Fe chlorosis (+Fe +biC) as compared to control treatment (+Fe).Vertical bars represent ± standard errors of means of 5 replicates, p ≤ 0.05.Measurements made on chlorophyll fluorescence demonstrate important genotypic differences in the maximum quantum yield of PSII in dark as in light test (table 2).In fact, no significant effect of direct and induced Fe chlorosis

Figure 6 .
Figure 6.Relationship between active iron (FeII) concentrations in leaves and SPAD index in two pea genotypes subjected to induced or direct iron deficiency.

Table 1 .
SPAD index in leaves of two pea genotypes subjected direct (-Fe) or bicarbonate-induced (+Fe +biC) Fe deficiency compared to control treatment (+Fe).Standard errors of means of 9 replicates, p ≤ 0.05.

Table 3 .
Fe use efficiency for plant growth (FeUEDW, expressed as the ratio of biomass production to Fe accumulation in leaves, g DW. µmol -1 Fe) and Fe use efficiency for net assimilation (FeUEAn, calculated as the ratio of net assimilation to Fe accumulated in leaves), in two pea genotypes subjected to direct or induced Fe deficiency.