Introduction

The Atlantic Forest is the fourth hottest biodiversity hotspot and houses more than 60% of the terrestrial species on the planet, although it only covers 11.7% of the surface of the globe. Moreover, the forest houses 21,361 species of vascular plants, amphibians, reptiles, birds and mammals, and among those 8,567 are endemic species (Myers et al. 2000). The high level of endemism in this ecosystem demonstrates that many species have become extinct and that others will also meet the same fate due to intense fragmentation (Brooks and Balmford 1996; Rodrigues et al. 2009; Silva and Tabarelli 2000).

These endemic species are not distributed at random but occur in particular portions of the Atlantic Forest. The species are preferentially distributed in at least six centers of endemism, including the Pernambuco Center for Endemism (Silva and Casteleti 2003). The center includes the coastal forests situated north of the River San Francisco, or between the states of Alagoas and Rio Grande do Norte. Although the original area of the Pernambuco Center for Endemism represents just over 4% of the original area of the entire Atlantic Forest, this region hosts about 8% of vascular plant flora of this forest (Porto et al. 2006). Currently, a few stretches of forest north of the River San Francisco have unique characteristics, since the block was reduced to forest archipelagos of small forest fragments (Silva and Tabarelli 2000). Pernambuco is one of the most degraded areas of the entire Atlantic Forest (CI-Brasil 1993) and, as a consequence, several species endemic to this region are on the verge of global extinction (Brooks and Rylands 2003). This is one of the regions of the planet where conservation efforts are most urgently required (Rodrigues et al. 2004). However, reforestation must follow the strategies established by Alves-Costa et al. (2008). Among their criteria, the authors stated that the appropriate proportion between diversity and filling species must be observed. Filling species are those that have rapid growth and high canopy cover, permitting rapid shading of the planted area (Chazdon et al. 1996; Souza et al. 2004). Species classified as pertaining to the diversity group do not show rapid growth and generally have low tree canopy coverage (Ribeiro et al. 2004; Souza et al. 2004; Swaine and Whitmore 1988); however, they are essential to ensure the perpetuation of the planted area. Species from this group will gradually replace the filling species when they come into senescence, finally occupying the area restored and ensuring its continuity.

Native species are widely used and perform well, even in highly disturbed environments (Rodrigues et al. 2009). Restoration takes time, the costs of monitoring are high and success is often hard to assess (Parker 1997; Rodrigues et al. 2009). Moreover, global climate changes are altering the natural characteristics of many ecosystems around the world (Intergovernmental Panel on Climate Change 2007). To the northeast of Brazil the most optimistic forecast is an increase of 2.0°C in the mean temperature, in the worst situation 4.0°C. However, in both cases is expected to uneven rainfall with dry period of about greater than today. Thus, plants need to adjust their metabolism to the changing scenario, such that those with greater plasticity will most likely be able to establish successfully. Within this scenario of global climate change, several studies have used plant ecophysiology to understand the behavior of species in different ecosystems (Gulías et al. 2009; Ogaya and Peñuelas 2003).

A study involving two species from different functional groups indicated that, in the absence of water constraints, the higher CO2 assimilation rates of filling group species are supported by the efficiency of the entire dissipative structure, involving both degradation and the dissipative process (Souza et al. 2004). Thus, the higher photosynthetic performance of filling group species was associated with higher values for stomatal conductance and transpiration rate (Ribeiro et al. 2004; Vincent 2001).

In this study, the behavior of four native species of Brazilian Atlantic Forest was evaluated under different light and water availability schemes. Species plasticity in relation to gas exchange, chlorophyll fluorescence and certain aspects of leaf metabolism was investigated. The main goal was to determine whether the species presented filling or diversity characteristics using ecophysiological tools.

Materials and methods

Study site, plant species and growth conditions

The species studied were from the Cucaú, which is private property located in the city of Rio Formoso, Pernambuco State, northeastern Brazil. The seeds were collected at this location, which has argil soils, a tropical climate and an annual rainfall of 2,788 mm, with the wettest period between May and June. The annual mean temperature ranges between 24 and 29°C. However, in some years of severe drought, the current annual rainfall is concentrated within a few days of the year.

The study included only species endemic to the Brazilian Atlantic Forest: Inga sp. of the family Leguminosae and subfamily Mimosaceae; Cinnamomum zeylancium Blume of the family Lauraceae, Brosimum guianensis Aubl., which is included in the family Moraceae and Tapirira guianensis Aubl. of Anacardiaceae.

The seeds of the species studied were collected in fragments of local Atlantic Forest. After germination and the acclimatization period of 4 months, the plants were transplanted into 9-L pots and maintained under greenhouse conditions for 70 days. During this period the mean temperature alternated between 24 and 32°C. The pots contained 7 kg of soil, a mixture of soil and washed sand (3:1), respectively.

Water treatment

Irrigation was suspended for 7 days during stress treatment (drought) and remained continuous in the control treatment. Measurements were performed on day 7 of treatment. Before the suspension of irrigation and during treatment, the control pots, was added 2.0 L of water to keep the substrate fully moistened.

Light treatment

This experiment was conducted during the same period as the water-deficit experiment. All the plants were well watered the entire time. Half of them were kept under 10% of natural light (shade treatment) and the remainder under natural light (control). Photosynthetic active radiation (PAR) ranged from 1,500 to 2,000 μmol m−2 s−1 during this period. Measurements were taken after 20 days of acclimation, on the same day as the measurements performed on the plants under water deficit.

Leaf water status

Under water deficit, water tension in the xylem was measured after 8 days without irrigation and on the same day in the shade and light treatments, using a Scholander pressure chamber (Soilmoisture Equipment Corp., Santa Barbara, CA, USA) on a fully expanded leaf from the shoot middle of the plant, a mature but not senescent leaf. These measurements were assumed to determine the leaf water potential (Ψl).

Gas exchange

Gas exchange measurements were performed on day 7 of drought, from 10:00 to 11:30 h, and on the same day in the shade and light treatments, using a Portable Infrared CO2 Gas Analyser (IRGA), ADC, model Lci (Hoddesdon, UK). Leaf gas exchange was measured on a mature but not senescent leaf in the middle of plant with a maximum PAR of 1,080 μmol m−2 s−1 in the leaf cuvette of the IRGA. The local PAR at the time of measurement was 1,500 μmol m−2 s−1, and mean temperature ranged from 27.1 to 29.2°C. The vapor pressure deficit (VDP) was approximately 2.1 kPa. The intrinsic water use efficiency (IWUE) was calculated by the relationship A/g s (Osmond et al. 1980).

Chlorophyll a fluorescence

Chlorophyll a fluorescence was measured using an ADC fluorometer (Hoddesdon, UK). Maximal (F m) and basal (F o) fluorescence yields were measured in dark-adapted (30 min) leaves, and steady-state (F s) and maximal (\( F_{\text{m}}^{\prime } \)) fluorescence yields were determined in a light-adapted state (Schreiber et al. 1994; Van Kooten and Snel 1990). The following parameters were assessed (Schreiber et al. 1994): the maximal quantum yield of PSII [F v/F m = (F m − F o)/F m] and the excitation capture efficiency of open centers [\( F_{\text{v}}^{\prime } /F_{\text{m}}^{\prime } \) = (\( F_{\text{m}}^{\prime } \) − \( F_{\text{o}}^{\prime } \))/\( F_{\text{m}}^{\prime } \)].

Leaf soluble sugar, free amino acid, protein and chlorophyll contents

Leaf samples were collected in the middle of the afternoon, from 14:00 to 15:00 h. Sample leaves were immediately frozen in liquid nitrogen and stored at −80°C in humidity proof containers until analysis.

Soluble carbohydrates were determined according to Dubois et al. (1956), using d(+)-glucose as standard. For free amino acid determination, 0.5 mL of 10% trichloroacetic acid was added to an aliquot of 0.5 mL of the water extract, and the mixture was maintained at 25°C for an hour. This mixture was then centrifuged at 12,000×g for 5 min, and the supernatant was used for amino acid determination (Yemm and Cocking 1955), using l-leucine as standard. Leaf protein content was determined according to Bradford (1976). Chlorophyll content was determined according to Lichtenthaler and Wellburn (1983).

Experimental design and statistical analysis

The experimental design was completely randomized with two water levels (water-deficit experiment), two light levels (light availability experiment) and three replicates. The data were subjected to analysis of variance (ANOVA) and means were compared and segregated by the Student–Newman–Keuls test (p < 0.05), when significance was detected.

Results

Effect of light availability

Leaf water potential (Ψl) was higher (p < 0.05) in the treatment under low light availability (shade) in the species Inga sp. and T. guianensis. However, B. guianensis showed a lower Ψl value in plants under shade (p < 0.05) (Table 1).

Table 1 Leaf water potential (Ψl, MPa) in young plants of tree species under different light availability of full light (control) and shaded (shade) or water deficit (drought)
Full size table

The values of stomatal conductance (g s), CO2 assimilation (A) and transpiration (E) determined for B. guianensis were lower (p < 0.05) compared with the other three species. Moreover, Inga sp. showed high values of g s, A and E, especially in the control treatment (light) compared with B. guianensis and T. guianensis (Fig. 1A, C). In general, IWUE was higher in the shade treatment, except for Inga sp. (Fig. 1D).

Fig. 1
figure 1

Changes in stomatal conductance, g s (A), net photosynthesis, A (B), transpiration, E (C) and intrinsic water use efficiency, IWUE (D) of young plants of tree species after 15 days of treatment: full light 1,080 μmol m−2s−1(control) and shaded 167 μmol m−2 s−1 (shade). Values followed by different letters in each column differ by test average Student–Neuman–Keuls (p ≤ 0.05). Values represent the mean of three replications

Full size image

The values of fluorescence parameters changed under shade, showing diminished basal fluorescence (F o) (p < 0.05) in all species (Fig. 2). Furthermore, except for Inga sp., all species showed an increase (p < 0.05) in maximum quantum yield of PSII, as determined by the ratio F v/F m. The species B. guianensis and T. guianensis showed a reduction (p < 0.05) in the values of excitation capture efficiency of open centers \( F_{\text{v}}^{\prime } /F_{\text{m}}^{\prime } \), under control conditions. For the fluorescence parameters, significant differences among species (p < 0.05) occurred only when exposed to full light, except for the species C. zeylancium which showed a difference in relation to the others when under shade (Fig. 2).

Fig. 2
figure 2

Fluorescence parameters responses of young plants of tree species after 15 days of treatment: full light 1,080 μmol m−2 s−1 (control) and shaded 167 μmol m−2 s−1 (shade). F o, fluorescence basal; F v/F m, PSII maximum quantum yield; \( F_{\text{v}}^{\prime } /F_{\text{m}}^{\prime } \), excitation capture efficiency of open centers. Values followed by different letters in each column differ by test average Student–Neuman–Keuls (p ≤ 0.05). Values represent the mean of three replications

Full size image

The determination of leaf soluble sugar, amino acid and protein (Table 2) in plants under light and shade revealed that only minimal alterations occurred between the treatments. B. guianensis showed the highest value of soluble sugars under shade.

Table 2 Leaf soluble sugars, amino acids and protein of young plants of tree species under different light availability of full light (control) and shaded (shade), after 15 days of treatment
Full size table

Effect of water deficit

Water deficit induced by 7 days of irrigation suspension reduced the Ψl in all the species studied (p < 0.05) (Table 1), except C. zeylancium. This finding was not accompanied by the rates of g s, A, E and IWUE (p > 0.05) (Fig. 3), except for B. guianensis which showed an A rate of zero (Fig. 3B). The gas exchange values obtained tended to be lower in plants under stress; however, T. guianensis plants under drought showed a high value for g s (p < 0.05) and A (Fig. 3A, B). No significant differences were observed for E and IWUE (p > 0.05), except for B. guianensis, which showed high g s and E values with a low A rate; thus, the IWUE for this species was the lowest determined (p < 0.05) (Fig. 3C, D).

Fig. 3
figure 3

Changes in stomatal conductance, g s (A), net photosynthesis, A (B), transpiration, E (C) and intrinsic water use efficiency, IWUE (D) of young plants of tree species after 7 days of treatment: well hydrated (control) and water deficit (drought). Values followed by different letters in each column differ by test average Student–Neuman–Keuls (p ≤ 0.05). Values represent the mean of three replications

Full size image

After 7 days under drought, the fluorescence parameters F o, F v/F m and \( F_{\text{v}}^{\prime } /F_{\text{m}}^{\prime } \) showed no significant difference between the treatments (Fig. 4A, C); however, B. guianensis showed the lowest values of F v/F m and \( F_{\text{v}}^{\prime } /F_{\text{m}}^{\prime } \) among the species studied.

Fig. 4
figure 4

Fluorescence parameters responses of young plants of tree species after 7 days of treatment: well-hydrated plants (control) and water deficit (drought). F o fluorescence basal; F v/F m PSII maximum quantum yield; \( F_{\text{v}}^{\prime } /F_{\text{m}}^{\prime } \) excitation capture efficiency of open centers. Values followed by different letters in each column differ by test average Student–Neuman–Keuls (p ≤ 0.05). Values represent the mean of three replications

Full size image

Leaf metabolism parameters showed no significant difference between the treatments; however, Inga sp. and T. guianensis showed lower values compared with the other two species (Table 3).

Table 3 Leaf soluble sugars, amino acids and protein of young plants of tree species under water-deficit (drought) and well-hydrated conditions (control), after 7 days of treatment
Full size table

Discussion

Gas exchange and chlorophyll fluorescence are two good parameters for understanding the behavior of different species (Gulías et al. 2009; Ogaya and Peñuelas 2003; Ribeiro et al. 2004; Souza et al. 2004) and are used worldwide. This study aimed to determine the role that ecophysiological tools can play in the difficult task of restoring areas of the Brazilian Atlantic Forest, where many species have become extinct (Rodrigues et al. 2009; Silva and Tabarelli 2000).

Analysis of the results showed that the four species behaved differently both under distinct light availabilities and under water deficit. Under shade, both Inga sp. and T. guianensis showed a high value for leaf water potential (Ψl) (p < 0.05) (Table 1). Such behavior may have been a consequence of lower stomatal conductance (g s) (Fig. 1A) when exposed to low light, which may be indicative of a low tolerance to shade, typical of the filling species group (Alves-Costa et al. 2008). Furthermore, lower values of gas exchange were obtained for B. guianensis in both treatments compared to the other species, especially to Inga sp. (Fig. 1A, C). Plant response to light availability is an important characteristic for growth, directly affecting the forest succession process. Thus, species range from the early successional light demanding species (filling group) to the very shade tolerant (diversity group) (Chazdon et al. 1996). The results obtained support the argument that under different light availabilities, Inga sp. and B. guianensis can be characterized as species pertaining to different functional groups.

After 15 days under distinct light availabilities, the basal fluorescence (F o) was significantly reduced (p < 0.05) in all species under shade treatment (Fig. 2A); this variable is related to excitation energy loss during its transfer from the pigment bed to reaction centers (Yordanov et al. 1997). This result verified that low light availability was sufficient to reduce the photochemical reactions of shaded plants, on the other hand, was not enough to modify the content of chlorophyll a, b and carotenoids (p > 0.05) (data not shown). However, the maximum quantum efficiency of PSII photochemistry (F v/F m) was increased in all species under shade (Fig. 2B), except Inga sp. The largest difference between treatments was determined in B. guianensis, followed by C. zeylancium. Similar behavior was observed for B. guianensis when evaluating PSII maximum efficiency, which estimates the maximum quantum efficiency of PSII under given light conditions (Fig. 2B, C). These data again suggest the same division between functional groups: Inga sp. pertaining to the filling group and B. guianensis to the diversity group, as suggested by the gas exchange data.

Soluble sugar, amino acid and protein contents (Tables 2, 3) were evaluated in this study; however, it was not possible to determine any typical behavior among species using these variables; i.e., separation into different functional groups could not be achieved using these data. However, the contrasting behavior between Inga sp. and B. guianensis was further emphasized by the determination of the lowest and the highest (p < 0.05) contents of leaf soluble sugars, respectively, among species studied. Data for amino acids and protein showed no distinct pattern among these species (Tables 2, 3), in disagreement with results obtained for the common bean, in which it was shown that the daily balances of leaf soluble sugars and free amino acids were good indicators for evaluating the stage of water deficit and metabolic capacity (Santos and Pimentel 2009). In addition, Ennajeh et al. (2009) showed the role of sugars in osmoprotection and osmoregulation in leaf cell of two olive cultivars under severe drought stress.

After 7 days without irrigation, lower Ψl values (p < 0.05) were obtained for all species compared with controls (Table 1). Among the four species, the lowest values were presented by C. zeylancium followed by Inga sp. and the greatest potential was shown by B. guianensis. Thus, as occurred when exposed to different light availability (Fig. 1A), B. guianensis showed the lowest g s value (p < 0.05) among the four species, regardless of treatment, and Inga sp. presented the highest (p < 0.05) value. Seven days without water supply was sufficient to reduce the g s of Inga sp. and C. zeylancium, but was not sufficient to reduce the g s of B. guianensis and T. guianensis. Stomatal closure and the consequent decrease in E responses to water deficit are expected in plants to prevent excessive loss of water (Kramer and Boyer 1995; Souza et al. 2004). However, these species showed no significant difference in A and E between treatments (Fig. 3B, C), except for the IWUE of B. guianensis, which showed a reduction (p < 0.05) when exposed to severe drought (Fig. 3D). IWUE is a good parameter for assessing drought tolerance (Pimentel et al. 1999). However, it must be evaluated along with other variables, as g s is dependent on several factors, including root activity, when it remains high under low soil water potential provides a high g s, thus decreasing the IWUE (Fig. 3A, D).

Moreover, without any interfering environmental factor (control treatment), Inga sp. showed higher g s, A and E than B. guianensis (Figs. 1A, C, 3A, C), supporting its classification as a filling group species (Alves-Costa et al. 2008). The gas exchange performance of the majority of the filling plant group can be affected under adverse environmental conditions, such as water deficit in the soil (Ribeiro et al. 2004; Souza et al. 2004) and low light conditions (Vincent 2001).

The stomatal behavior of the species supported the changes in A values (Figs. 1A, B, 3A, B), except for the pattern presented by B. guianensis under water deficit (Fig. 3A, B). High g s value was obtained for this species compared to controls, even though it showed A equal to zero. In tropical regions, plant species often experience substantial daily and seasonal changes in light availability (Chazdon et al. 1996). For some species, such as the diversity group, high light availability works as a stress factor rather than as a resource that can be used by the leaves (Chazdon et al. 1996; Ribeiro et al. 2004; Souza et al. 2004). This seems to be the case with B. guianensis, as revealed by the contrasting behavior between the control and stressed plants (Figs. 1A, B, 3A, B), showing high g s and low A. This indicates distinctive behavior in photochemical reactions (Baker and Rosenqvist 2004; Björkman and Powles 1984), including low photochemical efficiency of PSII, which was confirmed by lower values (p < 0.05) of \( F_{\text{v}}^{\prime } /F_{\text{m}}^{\prime } \) and \( F_{\text{v}}^{\prime } /F_{\text{m}}^{\prime } \) for B. guianensis compared to other species under the environmental conditions studied (Figs. 2B, C, 4B, C).

The present results confirm that certain ecophysiology parameters help to classify plant species within functional groups, as reported by Ribeiro et al. (2004) and Souza et al. (2004). The importance of correct classification is essential for reforestation programs, especially in northeastern Brazil. In this region, most of the Atlantic Forest area has been converted into agricultural land, with only 2% of the original forest remaining (Silva and Tabarelli 2000). Forest remnants are dispersed as small patches surrounded by open fields; thus, protected areas in this region are mostly small and isolated (Rodrigues et al. 2009). Determining the functional group to which a species belongs may increase the chance of success to establish reforested areas. Reforestation with high diversity is a chance to maintain or recover environmental functions, species and ecological interactions that occur or originally occurred in the northeastern Atlantic Forest. This is one way to definitively reduce the chances of extinction of several plant and animal species that are disappearing in this region and the world (Alves-Costa et al. 2008).

Among the species studied, two species can be characterized as clearly as pertaining to different functional groups: Inga sp., filling group; and B. guianensis, diversity group. Under the conditions determined in this study, C. zeylancium and T. guianensis showed characteristics of species of both groups, suggesting the possibility of their widespread use in reforestation programs.