ECOPHYSIOLOGICAL RESPONSES OF MEDIUM MORPHOTYPE OF Paubrasilia echinata (Lam.)Gagnon,H.C.Lima and G.P.Lewis RAISED UNDER FULL SUNLIGHT AND NATURAL SHADE

The existing inconsistent data on the irradiance needs of Brazilwood plants Paubrasilia echinata (Lam.) Gagnon,H.C.Lima and G.P.Lewis. can be explained by their phenotypical variations. While small-leaf morphotypes grows better in the shade, not tolerating environments with high irradiance, a recently discovered population of medium morphotype presents diff erent behavior: better performance under direct sunlight and limited growth in the shade. In order to understand the physiological mechanisms of the medium morphotype in response to the available irradiance, this study was performed to characterize the oxidative stress metabolism, photochemical and biochemical photosynthesis effi ciency, as well as anatomical adjustments of leaves of the medium morphotype of P. echinata under diff erent intensities of irradiance. The analyses were performed at direct sunlight condition (2000 μmol m.s) and within a dense ombrophilous forest with 80% shading (192 μmol m-s). Growth, leaf anatomy, chloroplast pigments, photochemical (chlorophyll a fl uorescence), and biochemical (gaseous exchanges) effi ciency, soluble carbohydrates, and antioxidants were measured. The results showed that the higher effi ciency in light energy uptake, paired with better photochemical performance and better CO 2 fi xation in plants under direct sunlight resulted in higher concentration of soluble sugars and growth. The energy that should have been used in photochemical and/or biochemical reactions of shaded plants was dissipated in the form of heat, re-emitted as fl uorescence or translocated to the production of antioxidant defense compounds of the secondary metabolism. Therefore, the medium morphotype of P. echinata presents an ecological profi le of sun-tolerant or pioneer species, and as such, it is recommended its planting in full sunlight. These results diff er from previous studies on small morphotype of P. echinata and suggest the need for a taxonomic reconstruction of this species, which is essential to adequate management practices in Atlantic rainforest recovery programs.


1.INTRODUCTION
Recognized worldwide by its mega diversity, the Brazilian Atlantic Forest shelters a number of species that are highly important economically, and culturally, such as Paubrasilia echinata Lam Gagnon,H.C.Lima and G.P.Lewis (Brazilwood) (Gagnon et al., 2016).
Even though the plant is well studied at taxonomic, phytochemical and propagation levels, little is known about Brazilwood's ecophysiology, especially its ecological habit regarding natural light. For Budowski (1965), this is a climax species (ombrophilous), whereas Lima (1992) and Lorenzi (2002) categorize it as heliophilous. For Mengarda et al. (2009), this species has semi-heliophilous or intermediate characteristics, because of its higher growth, photosynthetic capacity, and water-use effi ciency of 50% of photosynthetically active radiation (PAR). Investigating the acclimatization of P. echinata plants under contrasting irradiance, Mengarda et al. (2012) found a rapid reduction in the maximum quantum yield of PSII primary photochemical reactions (φP0 = F V /F M ) and performance index (PI ABS ) followed by total leaf abscission during the fi rst week of exposure to direct sunlight. The authors concluded that the species is shade-tolerant or late in forest succession. Thus, these two Mengarda's study showed that P. echinata plants are intolerant to the full sunlight.
These confl icting data concerning ecological habits may be related to morphological variations of Brazilwood. Three ecotypes have been reported by Juchum et al. (2008), in which the most common has smaller leafl ets and light orange core and is found on the Brazilian, coast from Rio de Janeiro to Rio Grande do Norte. The second ecotype diff ers from the fi rst one in that the latter has slightly larger leafl ets and orange-reddish core. Natural populations of this ecotype occur in the states of Rio de Janeiro, Espírito Santo, and southern Bahia. The third ecotype does not have subleafl ets, only pines made of large leafl ets and dark red core, found only in Bahia.
With the recent spread of this information, plants with semi-heliophilous or intermediate habits described by Mengarda et al. (2009) and shade tolerant or late succession (2012) were found to be small ecotypes, according to analyses of pictures in these authors' studies. This leads us to the conclusion that heliophilous (Lima, 1992;Lorenzi, 2002) or ombrophilous (Budowski, 1965) descriptions are related to other ecotypes. Recently, a population of medium ecotype planted in the Brazilian tablelands in the state of Espírito Santo was shown to have strong heliophilous habits (Gama, 2013).
The semi-heliophilous habits of the small ecotype and the heliophilous habit of the medium ecotype suggest deep changes in their primary and secondary metabolism. The small ecotype, which is sensitive to high irradiance, showed marginal leafl et burns, inhibition of gaseous exchanges, defi ciency in water use (Mengarda et al., 2009), decrease in chloroplast pigment contents and increase in chlorophyll a fl uorescence (Mengarda et al., 2012) under direct sunlight. These photodamage symptoms were followed by an increase in glucose, fructose, sucrose and raffi nose contents (Mengarda et al., 2012) associated to antioxidant metabolism (Terashima et al., 2006;Anjum et al., 2017;Portela et al., 2019).
The aim of this study is to characterize medium ecotype antioxidant metabolism and its implications in photosynthesis raised under full sunlight and natural shade exposure to reinforce the need of setting a new ecological structure of P. echinata in its possible subspecifi c taxa. We hypothesize that the medium ecotype of Brazilwood may present a divergent response to the contrast of luminosity compared to other previously reported ecotypes. This echophysiological variation is likely to be hindering a precise classifi cation of this species as well as its reintroduction in natural environments. Such information is essential for handling this species in Atlantic Forest recovery programs.

Study area and vegetal material
This study was carried out in a fragment of the Atlantic Forest tableland at the Reserva Natural Vale, located in the municipality of Sooretama, ES,Brazil (19º11'30'' S and 40º05'46'' W,altitude 58 m). The experiments used non-clonal seeds of medium ecotype of Paubrasilia echinata Lam Gagnon,H.C.Lima and G.P.Lewis. (Leguminosae, Caesalpiniaceae) from the same mother-plant at the same experiment site. Seeding was carried out in a greenhouse using plastic bags with irrigation and controllable environmental conditions (50% photosynthetically active radiation intercepted on the site, 80% relative humidity and 28ºC). After two months of germination, seedlings were planted in March 2010 in two areas: under direct sunlight (2000 µmol m -2 .s -1 ) and natural shading for Cariniana legalis (Mart.) Kuntze (192 µmol m -2 .s -1 ). At the beginning of the experiment, the seedlings had a mean height of 20 cm and 100 sample seedlings for each light condition without fertilization and irrigation as well as ten seedlings were selected for the analyses using the criterion of morphological homogeneity. The area occupied by the experiment comprised 100 square meters. Biometric analyses and leaf collection were performed on February 2012, when the plants were 23 months old. The sampling period was marked by a very dry summer, with precipitation and average temperature of 10.4 mm and 24ºC, respectively (INMET, 2012). The analyses were carried out from the third to the fourth node of fully expanded leaves. The leaf samples were collected, frozen in liquid nitrogen for transport and stored in ultrafreezer (−80ºC).

Growth analyses
For growth analyses, the height and diameter of the stem of ten (10) plants in each irradiance environment were measured. Leaf area was measured using Area Meter, LI-COR 3100, Nebraska, USA. From these data, we calculated specifi c leaf mass (SLM = LFM/LA) as per Hunt (1982), and water content or succulence (Teor H 2 O = LFM-LDM/LA), according to Parida et al. (2004), where LFM = leafl et fresh mass, LA = leafl et area, and LDM = leafl et dry mass.

Leaf anatomy
Fourth node leaves of six (6) plants in each irradiance environment were collected and fi xed in FAA 70 (formaldehyde, acetic acid, and ethanol 70%) for 48 hours and stored in alcohol 70% (v/v). Samples of the limb middle third in the internervural regions and midrib leaf areas were sliced by hand using table microtome with steel blades. The slices were colored with safrablau and the histological blades were mounted with glycerin gelatin (Bukatsch, 1972). For stomatal quantifi cation, abaxial leaf epidermis was printed on glass blade using cyanoacrylate ester (Super Bonder®, USA).

Photosynthetic pigments
Chlorophyll a (Chl. a), Chlorophyll b (Chl. b), total chlorophyll and carotenoid (Carot.) contents and Chlorophyll a/b (Chl. a/b) and Chlorophyll/Carotenoid (Chl./Carot.) ratios were determined using 4 leaf discs of 0.45 cm in diameter each of six (6) plants in each irradiance environment. The discs were stored in amber tubes with 7 mL of dimethyl sulfoxide (DMSO) and incubated in a water bath at 65ºC for 24 hours (Hiscox and Israelstam, 1979). The readings were carried out in spectrophotometer (Genesys 10S UV-Vis, Thermo Fisher Scientifi c, Waltham, USA) at the wavelengths of 470, 645 and 663 nm, and pigment concentration was determined using Arnon (1949) and Lichtenthaler (1987) equations.

Kinetics of transient or polyphase fl uorescence emission (OJIP)
Fluorescence emission kinetics was measured between 8:00 am and 9:30 am, in the same subleafl et collected to extract pigments, after acclimatization in the dark for 30 minutes using leaf clips (Strasser et al., 2004). Then, a saturating light beam of 3,000 µmol m -2 .s -1 was induced in a 4 mm diameter area of the subleafl et. OJIP Fluorescence transient (10 µs at 1 s) was measured using fl uorimeter Handy-PEA (Plant Effi ciency Analyzer, Hansatech Instruments Ltd, King's Lynn Norfolk, UK). The results were tabulated using the application software Biolyzer (Laboratory of Bioenergetics, University of Geneva, Switzerland) for an electronic spreadsheet (Strasser et al., 2004).
The OJIP fl uorescence transient curves were normalized based on the relative variable between points O and P [V t = (F t -F 0 )/(F M -F 0 )]. The defi nitions and equations of the OJIP test adopted followed Strasser et al. (2004), from which the following parameters were chosen: RC/ABS Index, which represents the quantity of active reaction centers (RCs) per energy absorbed; quantum yield of photochemical reactions during absorption until reduction of primary electron acceptor of photosystem II -PSII (Q A -) (TR 0 /ABS = F V /F M ); quantum yield in electron transport from QA-to pool of plastoquinone (PQ) (ET 0 /ABS = 1-F J /F M ); quantum yield in the reduction of fi nal acceptors of photosystem I -PSI (RE 0 /ABS = 1-F I /F M ); effi ciency of electron movement through electron transport chain transporters from Q A -to PQ (ET 0 /TR 0 ) and reduced PQ for PSI fi nal acceptors (RE 0 /ET 0 ); photosynthetic performance indices from excitation to reduction of intersystem electron acceptors (PI ABS ) and from excitation to reduction of PSI fi nal acceptors (PI TOTAL ); and specifi c fl ow of energy dissipation at antenna chlorophyll level, DI 0 /RC. We use twenty (20) plants per treatment.

Gaseous exchanges
Analyses of gas exchange were carried out along with fl uorescence analyses, in a closed system using a portable infrared gas analyzer (IRGA) model LCi (ADC BioScientifi c LCi Analyzer), under atmospheric concentration (400 CO 2 ppm) and natural lighting: direct sunlight of 1,000 µmol m -2 .s -1 irradiance and natural shading of 192 µmol m -2 .s -1 . We evaluated carbon photosynthetic assimilation (A), stomatal conductance (g s ), transpiration (E) and leaf inner carbon (C i ) of ten plants per treatment at the same time fl uorescence emission kinetics was measured (between 8:00 am and 9:30 am). We also calculated water-use effi ciency (WUE) by equation A/E, intrinsic water-use effi ciency (iWUE) by equation A/g s and instantaneous carboxylation effi ciency (P n /C i ) by equation A/C i . We used twenty (20) plants per treatment.

Extraction and quantifi cation of soluble carbohydrates
From the same leaf which chloroplast pigments were extracted, we removed 1.0 g of the fresh mass of the central internervural region of six (6) plants. The samples were boiled in 10 mL ethanol 80% for 3 minutes for enzymatic inactivation. They were then macerated and submitted to soluble carbohydrate extraction in a water bath at 80ºC for 15 minutes. The extract was centrifuged for 15 minutes at room temperature of 25ºC, the supernatant was removed and operation repeated twice more. Supernatants were put together, homogenized and concentrated in a rotary evaporator (Quimis®, Q344B1, Diadema, Brazil) at 40ºC. The carbohydrates were resuspended in 10 mL of deionized water, and the fi nal extract was stored in a freezer at −20ºC for later analysis.
Total concentration of soluble sugars was determined by the phenol sulfuric acid method (Dubois et al., 1956); of free and combined fructose, by acid hydrolysis of anthrone 0.2% in sulfuric acid (Jermyn, 1956); and of sucrose, by degradation of reducing carbohydrates by potassium hydroxide (KOH 5.4 N) as described by Riazi et al. (1985). Fructose rates from 0 to 50 µg were used for the standard curve and absorbance readings conducted using spectrophotometer (Genesys 10S UV-Vis, Thermo Fisher Scientifi c, Waltham, USA) at 620 nm.
Glucose in its free form was estimated with the use of the enzymatic method and a BioSystem kit. The reaction consists in the action of glucose oxidase and peroxidase over reagents glucose and aminoantipirin+phenol, respectively, which release quinone-imine using standard solution of 10 µg glucose/urea/creatinine. Absorbancy readings were carried out at 500 nm.

Determination of antioxidant activity
From the same leaves used for analyzing pigments and carbohydrates (six plants in each irradiance environment), 0.4 g of fresh mass were macerated with polyvinylpolypyrrolidone (PVPP) 1% (p/v) in order to obtain the enzymatic extract of superoxide dismutase (SOD, E.C. 1.15.1.1). The solution was homogenized in 2 mL phosphate buff er (50 mM, pH 7.5), EDTA-Na 1 mM, NaCl 50 mM, and ascorbic acid 1 mN. The extract was centrifuged for 25 minutes at 2ºC (Bulbovas et al., 2005). The supernatant was separated in plastic threaded tubes for later analyses.
For dosing the activity of superoxide dismutase, a cocktail of the enzyme reaction medium was prepared. Methionine and ribofl avin were prepared in the dark and kept in containers wrapped in aluminum foil to avoid photo-oxidation. Nitro blue tetrazolium (NBT) was the last reagent to be prepared, that is, only at the time of reaction. It was made with the same care as before to avoid photo-oxidation. The cocktail was made up of 0.5 mL of EDTA-Na 2 0.54 mM, 0.8 mL of potassium phosphate buff er (0.1 M, pH 7.0), 0.5 mL of methionine 0.13 mM, 0.5 mL of p-Nitro blue tetrazolium (NBT) 0.44 mM and 0.2 mL of ribofl avin 1 mM, and the resulting solution was exposed to fl uorescent light (80 W) for 20 minutes. Extracts prepared following the same procedure were kept in the dark. Absorbancy of the solution was measured in a spectrophotometer (Genesys 10S UV-Vis, Thermo Fisher Scientifi c, Waltham, USA) at 560 nm (Bulbovas et al., 2005).
Guaiacol peroxidase activity (POD. EC. 1.11.1.7) was determined with 0.3 g of fresh leaf mass homogenized with phosphate buff er (0.1 M, pH 7.0) and polyvinylpolypyrrolidone (PVPP) 2%. Right after that, the extract was centrifuged for 30 minutes at 2ºC. After adding 2 mL of phosphate buff er (0.1 M, pH 5.5), 0.3 mL of guaiacol (1%) and 0.05 mL of hydrogen peroxide (0.3%) to the supernatant, as per Klumpp et al. (1989), we carried out an absorbancy reading in a spectrophotometer (Genesys 10S UV-Vis, Thermo Fisher Scientifi c, Waltham, USA) at 485 nm. The absorbancy of the complex H 2 O 2 -POD formed was measured at two stages within a linear range of the reaction curve. Delta of absorbency (ΔE) was divided by the time in which both measurements were registered (at 2.0 and 3.5 minutes), which shows the POD activity during the reduction of hydrogen peroxide (Bulbovas et al., 2005).
Determination of phenolic compounds was carried out using the same alcoholic extract from carbohydrate analyses, and the calibration equation was obtained from the gallic acid standard curve. During the absorbency reading stage, 0.02 mL of an alcoholic extract solution were mixed to 0.2 mL of Folin-Ciocalteu. After 5 minutes of reaction, 3.23 mL of distilled water and 50 µL of sodium carbonate were added to the contents in the vortex. Next, the mixture was allowed to rest for two hours at room temperature (Hossain and Rahman, 2011). The readings were performed at 760 nm using a spectrophotometer (Genesys 10S UV-Vis, Thermo Fisher Scientifi c, Waltham, USA).
Determination of total anti-oxidant capacity (ABTS method) was carried out using the same alcoholic extract of the carbohydrate analyses. This analysis was adapted based on the method proposed by Lako et al. (2007). Cationic radical ABTS•+ was produced from the reaction of 7 mM 2.20-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS) and 2.45 mM of potassium persulfate. It was subsequently diluted with distilled water and the content transferred to an amber fl ask was allowed to rest at room temperature for 24 hours. The readings were performed with 10 mL of the ABTS solution and 5 µL of alcoholic sample at 734 nm using a spectrophotometer (Genesys 10S UV-Vis, Thermo Fisher Scientifi c, Waltham, USA). In order to calculate total anti-oxidant capacity, we carried out a standard curve calibration of Trolox equivalents (6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid).

Statistical analysis
The experimental design was completely randomized (CRD) for each treatment. The results were subjected to analysis of variance (ANOVA) at a 5% level of signifi cance (P<0.05). All analyses were carried out using Assistat, version 7.6 beta (Federal University of Campina Grande, Campina Grande, PB, Brazil).

Morphology and leaf anatomy
P. echinata plants under direct sunlight had growth values signifi cantly higher compared to those under natural shading (Table 1). Under direct sunlight, average height was 140 cm and stem diameter 26 cm, whereas plants in the natural shading population had growth 1/3 lower: 45 cm high and 8 cm of stem diameter. Specifi c leaf mass (SLM) and leaf water content were also higher among the population grown under direct sunlight. The radiation diff erence did not interfere in the unit leaf area (ULA) in both treatments.
For anatomic variables, we verifi ed higher tissue thickness among plants in full sunlight, based on cellular growth, not increase in cellular layers (Table 1). Parenchyma tissues, cuticle and epidermis were thicker under direct sunlight, which was refl ected by increased leaf limb thickness. Also, stomatal density was higher in the same treatment.

Chloroplast pigments
Irradiance intensity infl uenced pigment content (Table 1). Under natural shading, pigment concentrations Chl. a, Chl. b, Chl. Total and Carot. were 50% higher compared to plants in full sunlight. These pigment diff erences between treatments did not infl uence signifi cantly the ratios Chl. a/b and Chl./Carot.

Kinetics of transient or polyphase fl uorescence emission (OJIP)
The plants in natural shading showed higher levels of fl uorescence in phases O-J-I ( Figure 1A).
These high levels of energy dissipation were refl ected by the low values of parameters shown in Figure 1B for plants in natural shading. The density of active reaction centers of PSII (RC/ABS), the energy capture yields (TR 0 /ABS) and the probability of a Q A electron entering the transport chain (ET 0 /TR 0 ) were lower among plants in natural shading. However, the index that shows the effi ciency of reduced plastoquinone electron movement to fi nal PSI acceptors (RE 0 /ET 0 ) was higher among shaded plants. This result can be confi rmed at the (I-P) stage of these plants, the only stage that did not show discrepant levels compared to directly sunlit plants ( Figure 1A). * Asterisks indicate statistical diff erences between treatments by analysis of variance (ANOVA) at 5% probability (n=6). ET 0 /TR 0 is the effi ciency of electron movement through electron transport chain transporters from Q A -to plastoquinone (PQ); RE 0 /ET 0 is the effi ciency with which an electron cam move from the reduced; RE 0 /TR 0 is the effi ciency with which a trapped exciton can move an electron into the electron transport chain from Q A − to the PSI end electron acceptors; and PI TOTAL is the photosynthetic performance indices from excitation to reduction of intersystem electron acceptors and from excitation to reduction of PSI fi nal acceptors. Asterisks (*) indicate statistical diff erences between treatments by analysis of variance (ANOVA) at 5% probability (n=10). The total performance index (PI TOTAL ) (a parameter that shows the total performance of photochemical reactions from light capture in the collecting complexes, transport of electrons between photosystems, to the fi nal reduction in photosystem I) was kept in higher levels among directly sunlit plants ( Figure 1B).

Gaseous exchanges
Besides the superiority in the photochemical phase of photosynthesis, the biochemical phase was no diff erent: the plants under direct sunlight showed a higher capacity of carbon assimilation (A) (Figure 2A), stomatal conductance (g s ) ( Figure 2D) and transpiration (E) ( Figure 2C). These higher values in gaseous exchanges were positively refl ected by water-use effi ciency (WUE) ( Figure 2E) and apparent carboxylation effi ciency (P n /C i ) ( Figure 2G) compared to those in the shade. Intrinsic water-use effi ciency (iWUE) was higher in sunlight ( Figure 2F). The substomatal internal carbon concentration (C i ) was higher among plants exposed to natural shading ( Figure 2B), which confi rms the higher assimilation of this gas (A) in plants exposed to direct sunlight and, consequently, its reduction inside the mesophyll ( Figure 2G).

Soluble carbohydrates
The plants exposed to direct sunlight showed higher leaf concentrations of total soluble sugar, fructose, and glucose, but there was no diff erence in sucrose between treatments (Table 1). Carbohydrate concentration followed a decreasing proportion: total soluble sugar > sucrose > fructose > glucose for both levels of irradiance. For the direct-sunlight treatment, glucose was the sugar that stood out most signifi cantly, with values twice as high compared to the shaded plants (Table 1).

Antioxidants
The plants in natural shade had a total antioxidant capacity (ABTS) 1.6 higher than those exposed to direct sunlight ( Figure 3D), as well as higher phenol contents ( Figure 3B) in the same proportion. For enzymatic antioxidants, low irradiance resulted in more guaiacol peroxidase (POD) activity 1.6 higher in relation to full sunlight ( Figure 3C). Superoxide dismutase (SOD) activity showed a more pronounced diff erence between the contrasting irradiances ( Figure  3A). The SOD activity in natural shade was 3.8 higher than in sunlight.

4.DISCUSSION
The signifi cant morphology of P. echinata plants under direct sunlight combined with thickening of their leaves, cuticle, mesophyll cell elongation and higher stomatal density, which are typical of pioneer species (Sabbi et al., 2010), confi rm the heliophilous habit of this medium ecotype. These morphological adjustments increase resistance to water loss and control leaf temperature under high irradiance (Rossatto and Kolb, 2010).
Although palisade parenchyma elongation under direct sunlight was 70% higher than that of shaded plants, this strategy did not prevent oxidation of chloroplast pigments. However, this damage did not aff ect the carbon fi xation reaction of plants under direct sunlight, where A, gs, and E were higher than those among the shaded group. Consequently, WUE and Pn/Ci were higher under direct sunlight, which supports the idea that heliophilous species have higher water use effi ciency than ombrophilous species when exposed to high irradiance (Silva et al., 2010;Yang et al., 2018). Higher water-use effi ciency under direct sunlight can be attributed to high stomatal density and adapted anatomical structure, which facilitates atmospheric CO 2 spread to substomatal cavities and stimulate carboxylation reaction (Fini et al., 2010;Portela et al., 2019).
Higher concentrations of chlorophylls and carotenoids among the medium ecotype population in natural shading can be interpreted as an attempt to optimize photon absorption, regardless of wavelength, because the Chl. a/b ratio did not vary between the two lighting conditions. Overall, ombrophilous species have low Chl. a/b ratio in shaded areas (Lichtenthaler et al., 2007;Gaburro et al., 2017;Portela et al., 2019) because Chl. b is specialized in capturing irradiance in wavelengths shorter than those of Chl. a (Lima et al., 2010). Anyhow, investment in the system of antennas in the shaded population did not increase carbon fi xation, probably due to weak sunlight intensity or damage in the photochemical phase by oxygen reactive species.
This damaging reaction takes place in the oxidation complex of the water molecule during the photochemical phase, forming oxygen reactive species (ROS) incapable of being oxidized under intense shading (Sielewiesiuk, 2002;Gaburro et al., 2015). One of the ROS reactions consists in oxidizing the active reaction centers of PSII, and drastically aff ecting photochemical effi ciency (Favaretto et al., 2011), which is shown by higher Chl. a fl uorescence among medium ecotypes in natural shading. Under high shading, the electron transport chain is aff ected, which compromises the synthesis of energetic compounds during the photochemical phase of photosynthesis (Sielewiesiuk, 2002). This is refl ected by the low capacity to assimilate carbon (Baker, 2008;Mengarda et al., 2012;Zani et al., 2017;Portela et al., 2019), as seen among the medium ecotype population under natural shading.
The low photosynthetic capacity of medium ecotype under natural shading can also be confi rmed by the lower concentrations of photo-assimilated (glucose, fructose, and sucrose) compared to the population under direct sunlight. Besides working as energy sources, these soluble carbohydrates can also sequester ROS (Terashima et al., 2006;Gaburro et al., 2015;Zani et al., 2017;Portela et al., 2019). For the small ecotype (semi-heliophilous), sensitive to high irradiance, tolerance to direct sunlight was overcome by increasing the concentrations of glucose, fructose, sucrose, and raffi nose during the signalization phase and recovery of light stress (Mengarda et al., 2012). These carbohydrates also work as osmoregulators, attracting water molecules under contrasting irradiance. A is net CO 2 assimilation, (A), C i is CO 2 concentration in the intercellular airspaces (B), E is transpiration rate (C), gs is stomatal conductance to water vapor (D), WUE is water-use effi ciency (E), iWUE is intrinsic water use effi ciency (F) and Pn/C i is apparent carboxylation effi ciency (G) of the medium ecotype of Paubrasilia echinata Lam. grown in full sunlight (o) and natural shade (n). Asterisks (*) indicate statistical diff erences between treatments by analysis of variance (ANOVA) at 5% probability (n = 20).
from the extracellular media into the vacuole (Kakani et al., 2011;Mengarda et al., 2012;Portela et al., 2019), which causes signifi cant elongation of the palisade parenchyma in new leaves of the P. echinata small ecotype after exposure to direct sunlight (Mengarda et al., 2012). This osmoregulation action was also seen in the medium ecotype of the present study, in which leaves exposed to direct sunlight show 100% water content, higher than that of plants grown in shaded areas. This information confi rms the role of non-structural carbohydrates as a promising criterion for successional classifi cation (Li et al., 2016).  The heliophilous habit of medium ecotype can also be confi rmed by the increased antioxidant enzyme POD and SOD activity under natural shading, which is typical of pioneer trees in rain forests when exposed to 10% irradiance (Favaretto et al., 2011;Gaburro et al., 2015). In addition to their involvement in free radical neutralization, antioxidant enzymes also have an antagonistic eff ect in auxin activity. POD enzyme catalyzes the oxidation of this class of phytohormones recognized as a powerful regulator of growth and cellular expansion (Sofo et al., 2004). This can also explain the inhibited growth of medium ecotype of Brazilwood under natural shading.
Although the plants have shown higher POD and SOD activities in these conditions, the enzymatic mechanism was not enough to neutralize the ROS that destabilized the photochemical phase, which resulted in higher fl uorescence rates and lower carbon fi xation capacity. Consequently, instead of being used in carbon fi xation, the reducing compounds may have deviated to the secondary metabolism, which is more active under high shading levels (Favaretto et al., 2011;Gaburro et al., 2017). This includes the synthesis of phenolic compounds, which are in higher concentrations in the medium ecotype under natural shading. Phenols and fl avonoids work synergistically with antioxidant compounds, getting accumulated in leaf tissues so as to increase resistance against adverse environmental conditions (Sankari et al., 2017) such as low irradiance.
The results in this study show that the heliophilous habit of medium ecotype of P. echinata owes much more to its effi cient antioxidant metabolism since elongation of mesophyll cells did not prevent oxidation of chloroplast pigments under direct sunlight. Therefore, enzymatic (POD and SOD) and non-enzymatic (phenols and ABTS) antioxidant mechanisms of medium ecotype have shown to be very effi cient under direct sunlight. Both photochemical and carboxylation phases worked eff ectively, with no signs of photodamage, which refl ects the signifi cant growth under direct sunlight. On the other hand, even with increased POD and SOD activity under natural shading, the enzymatic mechanism was not enough to stop the oxidative damage shown by higher Chl. a fl uorescence and lower photosynthesis.

5.CONCLUSION
We conclude that the medium ecotype of P. echinata has a sun-tolerant or pioneer species profi le, as and such it is recommended that it be planted under full sunlight. The contradictions regarding ecological habits of P. echinata may be related to morphological and phylogenetic diff erences in the three ecotypes of this species. The physiological and biochemical characterization of these three morphological variations according to irradiance and other environmental factors are extremely important for attempts of new ecological structuring of P. echinata in its possible subspecifi c taxa. This information is essential for handling this species in Atlantic Forest recovery programs.