ARTICLE IN PRESS

Journal of Plant Physiology 167 (2010) 526533

Contents lists available at ScienceDirect

Journal of Plant Physiology

journal homepage: www.elsevier.de/jplph

Specic leaf areas of the tank bromeliad Guzmania monostachia perform distinct functions in response to water shortage
Luciano Freschi a, Cassia Ayumi Takahashi a, Camila Aguetoni Cambui a, Thais Ribeiro Semprebom a, Aline Bertinatto Cruz a, Paulo Tamoso Mioto a, Leonardo de Melo Versieux a, Alice Calvente a, Sabrina Ribeiro Latansio-Aidar b, Marcos Pereira Marinho Aidar b, Helenice Mercier a,n
a b

~ Paulo, SP, Brazil ~ o Paulo, CEP 05508-900, Sao Department of Botany, Institute of Biosciences, University of Sa ~ o Paulo, SP, Brazil Institute of Botany, C.P. 4005, 01061-970 Sa

a r t i c l e in f o
Article history: Received 18 September 2009 Received in revised form 28 October 2009 Accepted 29 October 2009 Keywords: Crassulacean acid metabolism CAM-idling Drought stress Epiphytes Photosynthesis

a b s t r a c t
Leaves comprise most of the vegetative body of tank bromeliads and are usually subjected to strong longitudinal gradients. For instance, while the leaf base is in contact with the water accumulated in the tank, the more light-exposed middle and upper leaf sections have no direct access to this water reservoir. Therefore, the present study attempted to investigate whether different leaf portions of Guzmania monostachia, a tank-forming C3-CAM bromeliad, play distinct physiological roles in response to water shortage, which is a major abiotic constraint in the epiphytic habitat. Internal and external morphological features, relative water content, pigment composition and the degree of CAM expression were evaluated in basal, middle and apical leaf portions in order to allow the establishment of correlations between the structure and the functional importance of each leaf region. Results indicated that besides marked structural differences, a high level of functional specialization is also present along the leaves of this bromeliad. When the tank water was depleted, the abundant hydrenchyma of basal leaf portions was the main reservoir for maintaining a stable water status in the photosynthetic tissues of the apical region. In contrast, the CAM pathway was intensied specically in the upper leaf section, which is in agreement with the presence of features more suitable for the occurrence of photosynthesis at this portion. Gas exchange data indicated that internal recycling of respiratory CO2 accounted for virtually all nighttime acid accumulation, characterizing a typical CAM-idling pathway in the drought-exposed plants. Altogether, these data reveal a remarkable physiological complexity along the leaves of G. monostachia, which might be a key adaptation to the intermittent water supply of the epiphytic niche. & 2009 Elsevier GmbH. All rights reserved.

Introduction The epiphytic habitat represents a highly dynamic environment, subject to temporal and spatial variations in irradiation, nutrient and water supply. Among these abiotic factors, sporadic or seasonal periods of water shortage are perhaps one of the most common challenges even for epiphytes occurring in humid tropics, which are characterized by a high annual rainfall (Zotz and Thomas, 1999). Some of these epiphytes, such as the

Abbreviations: CAM, Crassulacean acid metabolism; Cars, carotenoids; Chls, chlorophylls; DW, dry weight; DH + , dawn-dusk titratable acidity; MDH, malate dehydrogenase; PEPC, phosphoenolpyruvate carboxylase; RWC, relative water content. n Corresponding author at: Departamento de Bot anica, Instituto de Bioci encias, ~ o Paulo, Rua do Mata ~ o, 277 CEP 05508-900 Sa ~ o Paulo, SP, Brasil. Universidade de Sa Fax: +55 11 30917547. E-mail address: hmercier@usp.br (H. Mercier). 0176-1617/$ - see front matter & 2009 Elsevier GmbH. All rights reserved. doi:10.1016/j.jplph.2009.10.011

bromeliads, lack an absorptive root system and completely depend on direct precipitation for their water supply (Benzing, 1990). Therefore, a suite of adaptations must be employed by these plants in order to cope with an intermittent water supply (Zotz and Hietz, 2001). Approximately half of all bromeliads are epiphytic, and the success of this family in the epiphytic niche is frequently associated with the development of strategies to intercept, absorb and store rainwater more efciently (Benzing, 2000). In the tankforming bromeliads, for example, the rainwater accumulates in external tanks (phytotelma) formed by the overlapping of the leaf bases, allowing the plant to draw upon their water reservoir during periods of drought (Schmidt and Zotz, 2001). Moreover, the presence of epidermal trichomes on the leaf surfaces ensures an efcient way for water absorption in most of these epiphytic bromeliads (Benzing, 2000). In addition to these morphological specializations, a large number of epiphytic bromeliads display Crassulacean Acid

ARTICLE IN PRESS
L. Freschi et al. / Journal of Plant Physiology 167 (2010) 526533 527

Metabolism (CAM), which is a specialized photosynthetic pathway that minimizes the plant evaporative demand by opening its stomata mainly throughout the night when atmospheric vapor pressure decits are lower (Luttge, 2004). In fact, it is well-known that CAM bromeliads are especially abundant in epiphytic communities of dry forests (Grifths and Smith, 1983), which reinforces the ecological importance of the CAM pathway for the survival of vascular epiphytes in water-limited environments. Besides water economy, another key feature of the CAM is the remarkable plasticity provided by this mode of photosynthesis, especially among the C3-CAM facultative species, which can perform either C3 or CAM depending on the environmental conditions (Cushman, 2001). Among the bromeliads, the best recognized facultative CAM species is Guzmania monostachia (Medina et al., 1977; Grifths and Smith, 1983; Maxwell et al., 1992). This epiphytic tank bromeliad is widespread throughout the middle to upper canopy in neotropical forests and shows a high capacity of acclimation in response to variations in the natural environment (Maxwell et al., 1994). Water availability and light intensity seem to be the major environmental factors controlling the degree of CAM expression in G. monostachia (Medina et al., 1977; Maxwell et al., 1994). Medina et al. (1977) observed that this bromeliad exhibits atmospheric CO2 uptake only during the day with a small day/ night malate uctuation when water is not limited. On the other hand, after approximately one week of water shortage, the CO2 uptake occurred predominantly during nighttime, and larger acid uctuations were observed. Moreover, re-watering droughtexposed G. monostachia plants resulted in a reversible shift from CAM to C3 photosynthesis (Medina et al., 1977). By analyzing the gas exchange of G. monostachia growing as an epiphyte, Luttge et al. (1986) found that nocturnal recycling of respired CO2 is particularly important under eld conditions. Afterwards, studies carried out by Maxwell et al. (1994) also indicated that the nocturnal acid accumulation in droughtexposed G. monostachia resulted mainly from re-xation of respiratory CO2. This nocturnal regeneration of respiratory CO2 can be considered an important CAM response to severe stress conditions since the removal of this internal CO2 source signicantly increased the susceptibility to photoinhibition as demonstrated for Pyrrosia piloselloides (Grifths et al., 1989). Typical for most bromeliads, G. monostachia exhibits longlived leaves arranged in rosettes (Grifths and Smith, 1983). This rosette-habit creates a longitudinal light gradient along the leaf as the top portion receives more light than the leaf bases during the entire life span (Popp et al., 2003). Moreover, in leaves of tank-forming bromeliads there is the establishment of additional gradients along the leaf blade. In these plants, only the leaf bases (also called leaf sheaths) are in direct contact with the water accumulated in the tank and have the ability to absorb this resource (Schmidt and Zotz, 2001). On the other hand, the presence of water in the tank also limits the capacity of the leaf bases to exchange gas with the atmosphere (Benzing, 2000). Consequently, we should expect strong functional differences along the leaves of bromeliads, especially among the tankforming species. However, only few studies have analyzed the existence of physiological gradients along the bromeliad leaves. In the CAM species Aechmea aquilega, for instance, the highest values of nocturnal acid accumulation (DH + ) and CO2 uptake were found at the distal third of the leaf (Luttge et al., 1986). Also, signicant differences in the content of nitrogen and lipids and in the d13C values were detected between the chlorophyll-containing tissues and the non-chlorophyllous leaf bases of pineapple leaves (Medina et al., 1994). Further, Popp et al. (2003) observed that the leaf bases of ve CAM bromeliads possessed the lowest

levels of DH + when compared to their middle and upper leaf portions. Based on these studies, the present work was designed to evaluate whether different leaf portions of G. monostachia, a tank-forming C3-CAM bromeliad, play distinct physiological roles during the modulation of the photosynthetic pathway in response to changes in water availability. The level of CAM was evaluated in basal, middle and apical leaf portions of this bromeliad by analyzing the changes in DH + and in the activities of key enzymes of the CAM pathway. Additionally, internal and external morphological features, relative water content and chlorophylls and carotenoids levels were analyzed in these portions in order to allow the establishment of correlations between the structure and the functional importance of each leaf region. Finally, gas exchange analyses were performed in the leaf portions to obtain more information about possible differences in the photosynthetic capacity along the leaf blade.

Material and methods Plant growth and treatments Adult plants of Guzmania monostachia (L.) Rusby ex Mez var. monostachia (see Supplementary Table S1 for growth stage characterization) were taken from stocks maintained in a green~ o Paulo, house of the Department of Botany at the University of Sa ~ o Paulo, Brazil, and transferred to controlled environment Sa chambers. Throughout the experiments, and for the preceding three weeks of acclimation, the plants were maintained in a growth chamber at a photosynthetic ux density (PFD) of about 250 mmol m 2 s 1 supplied by uorescent lamps (Sylvania, Germany), 12 h photoperiod, day/night air temperature of 25/20 1C, and day/night relative humidity of 60/70%. PFD inside the growth chamber was monitored with an LI-190 quantum sensor connected to an LI-250 A meter (LI-COR Instruments, USA). All plants were cultivated in pots containing vermiculite, with one plant per pot. They were watered with distilled water on a daily basis and received a 10% (v:v) dilution of Hoaglands solution once a week (Hoagland and Arnon, 1938). After the acclimation period, plants were separated into three experimental groups, each one submitted to a different condition of watering: (a) control plants watered daily for a week, (b) water stress treatment, in which water was withheld for a week and (c) rewatered treatment, which consisted of plants subjected to drought for a week and, then, watered daily for another week.

Leaf sampling After each treatment, the 2nd to the 8th youngest fully developed leaves from ve individual plants were collected 1 h after the onset of illumination and divided in three portions: (a) basal, corresponding to the part of the leaf that forms the tank; (b) middle, corresponding to the lower half of the green portion of the leaf blade; and (c) apical, corresponding to upper half of the green part of the leaf blade. All samples corresponding to each leaf portion were fragmented into small pieces of about 5 mm length, weighed, frozen in liquid nitrogen and stored at 20 1C until use in the biochemical analyses. Additional samples for titratable acidity determination were taken 1 h before the end of the light period, frozen in liquid nitrogen and also stored at 20 1C until use.

ARTICLE IN PRESS
528 L. Freschi et al. / Journal of Plant Physiology 167 (2010) 526533

Anatomical analyses Freehand transverse sections were taken from fresh samples of basal, middle and apical leaf portions of four individual well-watered plants and analyzed by optical microscopy (Zeiss Standard 25 ICS, Carl Zeiss, Germany) in order to evaluate mesophyll, hydrenchyma and chlorenchyma thickness as well as the hydrenchyma cell diameter. For each parameter, about 80 measurements per leaf portion were carried out. Furthermore, stomatal and trichomes densities were assessed using epidermal silicon imprints prepared from the abaxial surfaces of each leaf portion (10 elds of 1 mm2 per leaf portion). All anatomical measurements were examined using KS100 3.0 software (Carl Zeiss, Germany). Measurements of tissue water content The tissue water content of the leaf portions was determined according to Martin and Schmitt (1989). After determining the fresh weight (FW), the samples were dried to a constant weight at 65 1C and allowed to cool down for 3 h before determining the dry weight (DW). Tissue water content was calculated using the formula ((FW DW)/DW) 100. Measurements were made in triplicates. Chlorophylls and carotenoids determination Chlorophylls (Chls) and carotenoids (Cars) were extracted by homogenizing 1 g of fresh leaf tissue in 7 mL of cold 80% (v/v) aqueous acetone. The homogenate was ltered, and the total volume of ltrate was completed to 20 mL with cold 80% acetone. Chls a and b and total Cars were determined spectrophotometrically according to Lichtenthaler (1987). Measurements were made in triplicates and the results were expressed as mg g 1 DW. The relative levels of Chl a and Chl b (Chl a/b) as well as the ratio of total Chls to total Cars (Chls/Cars) were also calculated for each leaf portion. Titratable acidity To determine the nocturnal accumulation of acidity (DH + ), leaf samples collected before dawn and dusk were ground in liquid nitrogen and then homogenized with 20 volumes (v/w) boiled distilled water. The crude extracts were boiled for 20 min, ltered and then allowed to cool down to room temperature. Titration of the water extracts was performed with 20 mM NaOH to an end point of pH 11. Titration up to pH 11 covered the ionization range of malic, citric and isocitric acids. Measurements were made in triplicates and the results were expressed as mmol H + g 1 DW. PEPC and MDH extraction and assay To measure phosphoenolpyruvate carboxylase (PEPC, EC 4.1.1.31) and malate dehydrogenase (MDH, EC 1.1.1.37) activities, leaf samples stored in liquid nitrogen were ground to a ne powder and extracted in ve volumes (v/w) of buffer containing 200 mM TrisHCl (pH 8.0), 1 mM EDTA, 5 mM dithiothreitol (DTT), 10 mM MgCl2, 10% (v/v) glycerol, 0.5% (w/v) bovine serum albumine (BSA). The homogenate was centrifuged for 5 min at 15.000g, and the supernatant was immediately used for the enzymatic assays. The PEPC activity was assayed in a 2 mL standard reaction medium containing 50 mM TrisHCl (pH 8.0), 1 mM DTT, 10 mM MgCl2, 10 mM NaHCO3, 200 mM NADH, 3 mM phosphoenolpyruvate (PEP) and 10 units MDH (Nievola et al., 2005). The MDH activity was assayed in the OAA-reducing

direction in a 2 mL reaction medium containing 50 mM TrisHCl (pH 8.0), 2 mM OAA, 5 mM MgCl2 and 200 mM NADH (Cuevas and , 2000). For both enzymatic determination, the reaction Podesta was started by adding an aliquot of enzyme extract and change in absorbance was continuously measured at 340 nm. All reported rates are from linear portions of absorbance vs. time curves (usually between 0 and 10 min). The enzymes were assayed at 30 1C.

Leaf gas exchange and incident light measurements Gas-exchange measurements were made continuously on the middle and apical portions of the same leaf over 24 h by using an infra-red gas exchange system (LI-6400, Li-Cor, USA). Every leaf portion was enclosed in a chamber, which tracked the environmental conditions inside the growth cabinet. A CO2 cylinder was used to keep CO2 concentration constant under 380 mmol m 2 s 1. Three plants were analyzed for each treatment and all measurements were carried out in the second youngest fully developed leaf of each plant. Incident photosynthetic ux density (PFD) on the adaxial surfaces of basal, middle and apical portions of these leaves was also determined by using an LI-190 quantum sensor connected to an LI-250 A meter (LI-COR Instruments, USA).

Statistical analysis All data are presented as mean 7 SD. One-way ANOVA was used to analyze the results and the means were compared by the Tukey test at 5% probability level.

Results Longitudinal differences in the leaf structure and pigment composition In order to investigate the existence of morphological gradients along the leaf blade of G. monostachia, anatomical analyses were carried out on the basal, middle and apical leaf portions of this bromeliad. These analyses revealed an inverse relationship between the densities of trichomes and stomata along the leaf blade (Table 1). Trichome density was twofold higher in basal than in middle and apical leaf portions. In contrast, the middle and apical leaf regions showed the highest values of stomatal density, while in the basal portion there were almost no stomata (Table 1). Moreover, internal anatomical analyses demonstrated that the mesophyll and hydrenchyma thickness increased signicantly from the apical to the basal region of the leaves (Table 1). The chlorenchyma thickness, on the other hand, was relatively uniform along the three leaf portions. In addition, while the apical and middle portions exhibited hydrenchyma cells of approximately the same diameter, these cells were signicantly larger in the leaf bases (Table 1). Besides these anatomical differences, the content of Chls and Cars also varied markedly along the leaf blade. As expected, the lowest content of total Chls and Cars were found in the leaf bases (Table 2). Moreover, the leaf bases exhibited the lowest values of Chl a/b ratio (Chl a/b) and the highest weight ratio of total Chls to total Cars (Chls/Cars), as is usually found in leaves adapted to low light conditions (Lichtenthaler et al., 2007). On the other hand, much higher values of Chl a/b ratio, total Chls and Cars contents were observed in the apical leaf portion, suggesting higher photosynthetic capacity in this region (Table 2). Incident PAR was also signicantly higher in the leaf top

ARTICLE IN PRESS
L. Freschi et al. / Journal of Plant Physiology 167 (2010) 526533 529

Table 1 Anatomical characteristics of different leaf portions of well-watered plants of G. monostachia. Leaf portions Top Middle Base Stomatal density (mm 2) 22( 7 4.9) a 11( 7 4.0) b 1( 7 0.5) c Trichome density (mm 2) 23( 7 4.6) b 26( 7 4.2) b 53( 7 11.9) a Mesophyll thickness (lm) 303.5( 7 20.8) c 357.9( 7 24.3) b 429.8( 7 25.2) a Hydrenchyma thickness Hydrenchyma cell (lm) diameter (lm) 190.7( 7 25.2) c 225.9( 7 22.5) b 304.1( 7 28.5) a 59.1( 7 11.2) b 60.5( 7 9.3) b 66.6( 7 9.3) a Chlorenchyma thickness (lm) 112.8( 7 22.1) c 132.3( 7 18.2) a 123.8( 7 26.8) b

Data for stomatal and trichomes densities were derived from light microscopy of 10 replicates per leaf portion. Internal anatomical data were determined from transverse leaf sections mounted in water and examined by light microscopy (n =80 per leaf portion). Data are expressed as the mean 7 SD. Different letters indicate a signicant difference between the leaf portions (P o 0.05%; Tukey test).

Table 2 Total chlorophylls, total carotenoids, chlorophyll a/b ratio (Chl a/b) and chlorophylls/carotenoids ratio (Chls/Cars) in different leaf portions of G. monostachia plants maintained under well-watered conditions or exposed to 7 d of drought. Treatment Well-watered Leaf portion Top Middle Base Top Middle Base Total chlorophylls (lg g 1 DW) 1602.5( 7 194.9) a 1211.7( 7 123.6) b 357.3( 7 101.1) c 2162.5( 7 186.3) a 1184.8( 7 81.3) b 199.8( 7 59.9) c Total carotenoids (lg g 1 DW) 564.0( 7 93.9) a 420.8( 7 59.8) a 89.9( 7 36.1) b 762.7( 7 79.2) a 407.2( 7 34.4) b 54.1( 7 11.5) c Chl a/b ratio 2.7( 7 0.1) a 2.7( 7 0.1) a 1.8( 7 0.1) b 2.5( 7 0.1) a 2.6( 7 0.1) a 1.9( 7 0.2) b Chls/cars ratio 2.8( 7 0.2) b 2.8( 7 0.1) b 3.8( 7 0.8) a 2.8( 7 0.1) b 2.9( 7 0.1) b 3.7( 7 0.4) a

Drought-exposed

Data are expressed as the mean ( 7 SD) of three replicate samples. Different letters indicate a signicant difference between the leaf portions (P o 0.05%; Tukey test).

1000 base Relative Water Content 800 Aa Ba 600 Ca middle top Aa ABb Bc Bb Ca Aa

400

hand, basal and middle portions of drought-exposed plants showed a reduction in their RWC of about 34% and 25%, respectively. Consequently, these results indicated that the RWC in the apical portion was maintained unchanged during the water stress treatment, probably at the expense of water translocation from the basal and middle leaf sections. In these more basal leaf regions, drought caused a reduction in leaf thickness due to shrinkage of the hydrenchyma but not of the chlorenchyma (data not shown). As also presented in Fig. 1, the drought-induced reduction in the RWC of the basal and middle leaf portions was completely reverted after one week of re-watering.

200
CAM expression

0 well-watered drought-exposed Treatments

Fig. 1. Relative water content of different leaf portions of well-watered, droughtexposed (7 d of drought) and rewatered (7 d of drought followed by 7 d of recovery) plants of G. monostachia. Means followed by the same letter are not signicantly different. Small letters represent the comparison among the treatments and capital letters the comparison among the leaf portions (P o 0.05%; Tukey test).

rewatered

(235 7 4 mmol m 2 s 1) than in the middle (168 7 6 mmol m 2 s 1) and basal (68 7 5 mmol m 2 s 1) leaf regions. Water relations along the leaf blade Leaves of G. monostachia also showed a marked longitudinal gradient in terms of relative water content (RWC). Under wellwatered condition, basal and middle leaf portions exhibited RWC values signicantly higher than those detected in the apical region (Fig. 1). On the other hand, under drought conditions, these differences practically disappeared, indicating that the reduction in RWC caused by water shortage did not occur equally along the leaf blade. In fact, the values of RWC observed in the leaf top portion after the drought treatment were statistically similar to those detected in the same region of well-watered plants. On the other

Since reductions in the RWC are believed to inuence the degree of CAM expression in C3-CAM facultative species (Cushman, 2001), we further investigated whether the differential changes in the water content along the leaf blade of G. monostachia were correlated with modications in the photosynthetic pathway (C3 or CAM) performed by each leaf region. As shown in Fig. 2, the analyses revealed a marked increase in the activities of PEPC and MDH in the drought-exposed plants, which occurred specically in the apical portion of the leaves (Fig. 2). Increase in the activity of these key enzymes of the CAM cycle was positively correlated with a signicant rise in the levels of nocturnal accumulation of titratable acidity (DH + ) also in the upper leaf region (Fig. 2). In these water stressed plants, total Chls and Cars also increased particularly in the leaf top, even though both Chl a/b and Chls/Cars ratios remained virtually unchanged (Table 2). When compared with the leaf top, basal and middle leaf portions of drought-exposed plants showed no or much less marked changes in pigment composition, DH + and PEPC and MDH activities. After re-watering, the values of both enzymatic activity and DH + returned to levels similar to or even lower than those observed for well-watered plants (Fig. 2). However, it is interesting to mention that a signicant level of nocturnal acidication was detected in middle and apical leaf portions of both well-watered and rewatered individuals of G. monostachia,

ARTICLE IN PRESS
530 L. Freschi et al. / Journal of Plant Physiology 167 (2010) 526533

12
PEPC activity (mol min-1 g-1 DW)

base middle top

Aa

considerable levels of nocturnal acidity accumulation observed in the apical and middle regions of drought-exposed plants (Fig. 2C) occurred in almost complete absence of atmospheric CO2 uptake and, therefore, could only be explained by re-xation of the respiratory CO2 produced during the dark period.

6 Ba 3 Bb Bb Ab Ca Bc
drought-exposed

Discussion

Ab Bc

0
well-watered rewatered

500 Aa
MDH activity (mol min-1 g-1 DW)

400 300 200 100 0

well-watered drought-exposed rewatered

Ba Cb Bb Ab Ca Cc Bb

Ab

Aa 600
H+ ( mol H+ g-1 DW)

400 Aa 200 Bb 0

Ab Ba Ba

Ab

Ca
drought-exposed

Cb
rewatered

well-watered

Treatments
Fig. 2. Activities of PEPC (A), MDH (B), and titratable acidity (C) of different leaf portions of well-watered, drought-exposed (7 d of drought) and rewatered (7 d of drought followed by 7 d of recovery) plants of G. monostachia. Means followed by the same letter are not signicantly different. Small letters represent the comparison among the treatments and capital letters the comparison among the leaf portions (P o 0.05%; Tukey test).

suggesting the occurrence of some degree of CAM even under these conditions (Fig. 2). Striking differences in the diurnal pattern of CO2 assimilation were also observed between well-watered and drought-treated plants (Fig. 3). In well-watered plants, CO2 assimilation occurred primarily during early morning and late afternoon in the middle and apical leaf portions, both showing a midday depression. In these plants, the apical leaf portion exhibited diurnal levels of gas exchanges slightly higher than those found in the middle region, revealing a positive correlation with the highest stomatal density also detected in the leaf top (Table 1). In contrast to the wellwatered plants, those submitted to water stress exhibited virtually no CO2 exchange in the middle and apical leaf portions during the entire day and night (Fig. 3). Therefore, the

Leaves compose most of the vegetative body of epiphytic tank bromeliads and are responsible not only for photosynthesis but also for water absorption and storage (Benzing, 2000; Popp et al., 2003). Therefore, besides anatomical modications, functional differences are also expected to occur along the leaf blade of these bromeliads (Benzing, 2000; Takahashi et al., 2007). In agreement, the data obtained here demonstrated that different leaf regions of G. monostachia can perform distinct roles during the physiological responses triggered by changes in one of the most signicant variables in the epiphytic niche, the water availability (Zotz and Hietz, 2001). During sporadic or seasonal periods of water deprivation, epiphytic tank bromeliads can rely basically on only two sources of water storage, the external tank and the leaf water-storing tissues (e.g. hydrenchyma). As pointed out by Zotz and Thomas (1999), the presence of relatively large tanks does not imply that bromeliads such as G. monostachia are continuously supplied with water since without resupply, the water accumulated in this structure will dry out after little more than one week under natural conditions. Therefore, given the intermittence in the water availability typical of the epiphytic niche, the water-storing tissues of the leaf can be considered a key adaptation for the survival of these plants during medium to long-term periods of water shortage (Benzing, 2000). As demonstrated in the present study, these internal water reservoirs are not uniformly distributed along the leaf blade of a tank bromeliad, being especially abundant in basal leaf regions (Table 1). In fact, the gradual increase in the mesophyll thickness observed from the top to the leaf bases can be explained almost exclusively by a progressive increment in the abundance of the hydrenchyma tissues. Therefore, the predominance of water-storing tissues closer to the leaf bases correlates well with the high capacity of water absorption at this region, as indicated by the elevated density of trichomes (Table 1) and the direct contact with the water accumulated in the tank (Benzing, 2000). Moreover, the drought-induced changes in the relative water content also occurred unevenly along the leaf blade once signicant reductions in the RWC occurred in the basal and middle portions, while no considerable variations were found in the water content of the leaf top (Fig. 1). Given the fact that the drought treatment was imposed by removing the water supply from both roots and tanks, the maintenance of a stable water status in the leaf apex during drought was probably due to water translocation from tissues closer to the leaf bases. Therefore, when water is no longer available in the external tank, the water stored in the hydrenchyma of these basal leaf regions may represent the main reservoir for maintaining a favorable water status in the photosynthetic tissues. In agreement with these results, several studies have demonstrated that water movement between distinct leaves or from tissue to tissue within a leaf are commonly observed in succulent plants exposed to drought (Schulte and Nobel, 1989). In most cases, a preferential movement of water from chlorophyllless hydrenchyma to the chlorenchyma is maintained by an osmotic gradient in which the hydrenchyma, with lower solute concentration, provides water to the chlorenchyma tissue (Schulte and Nobel, 1989). For instance, Nowak and Martin

ARTICLE IN PRESS
L. Freschi et al. / Journal of Plant Physiology 167 (2010) 526533 531

3 top Net photosynthetic rate (mol m-2 s-1) middle 2

-1 18:00 20:00 22:00 3

0:00

2:00

4:00

6:00

8:00

10:00 12:00 14:00

16:00 18:00

Net photosynthetic rate (mol m-2 s-1)

-1 18:00 20:00 22:00

0:00

2:00

4:00 6:00 8:00 Time of the day

10:00 12:00 14:00

16:00 18:00

Fig. 3. Diurnal pattern of CO2 assimilation in the middle and top leaf portions of G. monostachia plants maintained under well-watered conditions (A) or exposed to 7 d of drought (B). The shaded areas indicate the dark period. (n = 3).

(1997) demonstrated that water movement from the hydrenchyma to the chlorenchyma was responsible, at least in part, for the maintenance of high physiological activity in the atmospheric CAM bromeliad Tillandsia ionantha even after 20 d of extreme drought conditions. Taking into account the direct contact of the base leaf portion with the water retained in the tank, even if intermittently, and the considerably lower light intensities received by this region, the leaf bases of tank bromeliads are assumed to carry out reduced photosynthetic rates (Benzing, 2000). Although the partial superposition of the leaves impeded direct measurements of gas exchanges in intact leaf bases of G. monostachia, the almost complete absence of stomata and the low levels of Chls and Cars in fact suggested little photosynthetic activity at this leaf portion. On the other hand, the leaf regions localized outside the tank area and, consequently, more exposed to light and with less restriction to carry out gas exchange exhibited a set of characteristics indicative of a higher photosynthetic capacity, such as abundant stomatal densities, increased contribution of the chlorenchyma to the mesophyll thickness (Table 1) and higher total Chls and Cars contents (Table 2). In general, similar diurnal patterns of gas exchange were observed in the middle and apical leaf portions of well-watered plants, with maximal photosynthesis restricted to the beginning and end of the photoperiod (Fig. 3). In agreement with previous laboratory and eld studies (Luttge et al., 1986; Maxwell et al., 1994, 1995; Maxwell, 2002), G. monostachia under well-watered conditions exhibited low values of diurnal atmospheric CO2 (Fig. 3), which may be attributed, at least in part, to

the presence of several leaf morphological adaptations to drought stress, such as hypostomaty, presence of sunken stomata and a mesophyll dominated by a tightly packed hydrenchyma. Although crucial for the survival in the water-limited epiphytic niche, these leaf morphological adaptations may limit CO2 diffusion from the atmosphere to inside the leaf and, therefore, restrict carbon gain and growth (Maxwell, 2002). Another key adaptation of G. monostachia to withstand the water shortage periods is the capacity to perform CAM photosynthesis under adverse conditions (Medina et al., 1977; Grifths and Smith, 1983; Maxwell et al., 1992). Studies carried out by Maxwell et al. (1995) demonstrated that CAM is rapidly induced in this bromeliad at the start of the dry season and is completely reverted to C3 photosynthesis as the conditions become more favorable. Afterwards, by analyzing the CAM activity in G. monostachia over an entire year, Zotz and Andrade (1998) revealed that considerable nocturnal acidication in the leaf tissues of this bromeliad can occur even during the rainy periods with just a small, but signicant, increase in the dry season. In agreement with these eld data, our results also indicated signicant levels of nocturnal H + accumulation in middle and apical leaf portions of well-watered plants of G. monostachia (Fig. 2), suggesting that some level of CAM can occur in this bromeliad even when water supply is abundant and fairly constant. However, the data obtained also reinforce the occurrence of a high photosynthetic plasticity in G. monostachia and provide evidence demonstrating that the intensication of the CAM in response to drought does not occur homogenously along

ARTICLE IN PRESS
532 L. Freschi et al. / Journal of Plant Physiology 167 (2010) 526533

the leaf blade of this bromeliad. In fact, the drought treatment induced signicant increases in DH + exclusively in the upper leaf portion where most of the activities of CAM enzymes were also detected (Fig. 2). On the other hand, the levels of DH + and activities of CAM enzymes in the middle and basal leaf portions exhibited little or no changes in response to water stress. Therefore, the intensication of CAM in response to drought was detected exclusively in the leaf top, which is the most lightexposed leaf portion (Table 2) and also the only region that retained almost unchanged values of RWC (Fig. 1). Since the presence of large vacuoles in the mesophyll cells are considered to be critical for overnight acid storage in CAM plants (Cushman, 2001), it seems plausible to assume that, by keeping a relatively stable water content, the capacity of C4 acid accumulation at the leaf top remained unaffected during the drought, which may explain the preference for the operation of the CAM cycle at this leaf portion. In addition, the higher nocturnal carboxylation capacity of the leaf apex might also be associated with a possibly higher availability of carbohydrates at this region since it is considered a limiting resource for nocturnal CO2 xation in CAM plants (Borland and Dodd, 2002). Moreover, the occurrence of drought-induced increases in total Chls and Cars exclusively in the leaf top of G. monostachia (Table 2) seems to indicate that the intensication of the CAM at this leaf region was also associated with a general increment in the photosynthetic capacity of the tissue. Irrespective of the reasons for the preferential establishment of the CAM cycle in the leaf top, these data are in agreement with previous observations in obligate CAM bromeliads, which usually show higher values of DH + from the leaf bases to the tip (Luttge et al., 1986; Popp et al., 2003). Virtually no exogenous CO2 uptake was observed during the entire diurnal cycle in both middle and upper leaf portions of drought-exposed plants (Fig. 3) therefore, we can conclude that all nighttime accumulation of organic acids detected in these tissues (Fig. 2) was originated by nocturnal re-xation of respiratory CO2. This physiological state, known as CAM-idling, is considered a strong mode of CAM which is characterized by a substantial reduction in water loss by maintaining the stomata closed during day and night (Rayder and Ting, 1983; Luttge, 2004). Although plants in CAM-idling mode do not show net carbon gain or growth, they are able to maintain a basal metabolic activity even during severe drought conditions and can rapidly recover the photosynthetic capacity when favorable water conditions are re-established (Rayder and Ting, 1983). Additionally, in these plants, the internal CO2 recycling may alleviate photoinhibition by providing a sink for electron transport. Therefore, although a number of studies under eld and laboratory conditions have already reported that re-xation of respiratory CO2 forms a signicant proportion of the nocturnal acid accumulation in G. monostachia (Luttge et al., 1986; Maxwell et al., 1994; Maxwell, 2002), the data obtained here demonstrate that a typical CAM-idling cycle can also be induced in this bromeliad in response to water deprivation. In addition, the drought-induced establishment of CAM-idling in G. monostachia occurred extremely rapid when compared to other bromeliads. For example, in the bromeliad Aechmea Maya, a typical CAMidling state occurred only after 150 d of water stress (Ceusters et al., 2009), while in G. monostachia practically no CO2 uptake was detected after one week of water shortage (Fig. 3). Therefore, this rapid acclimation of the photosynthetic metabolism to abrupt changes in the water supply may help to explain the widespread distribution of this bromeliad across a wide range of exposed and shaded habitats in the epiphytic niche (Medina et al., 1977; Grifths and Smith, 1983). In summary, the present study demonstrated that different leaf sections of the tank C3-CAM bromeliad G. monostachia can

perform distinct functional roles in response to water shortage, which may represent a signicant adaptation to the survival of this species in the water-limited epiphytic niche. The main anatomical, biochemical and physiological differences observed along the leaves of this bromeliad are summarized in the Supplementary Figure S1, which also presents the internal structure of each leaf region. According to the results, when water is no longer available in the external tank, the leaf base seems to represent the main reservoir of this resource for maintaining the photosynthetic activity in more apical leaf regions. On the other hand, another critical drought adaptation, the CAM pathway, was intensied specically in the upper leaf section, which was the only region showing an almost constant water status during water deprivation. Finally, gas exchange data indicated that under drought conditions practically all nighttime accumulation of organic acids resulted from the re-xation of respiratory CO2, characterizing, therefore, the establishment of a typical CAM-idling pathway. Altogether, these results reinforce the remarkable physiological complexity of tank bromeliad leaves and highlight the importance of further studies regarding the existence of functional gradients along the leaf blade of other highly specialized plants.

Appendix A. Supplementary material Supplementary data associated with this article can be found in the online version at doi:10.1016/j.jplph.2009.10.011.

References
Benzing DH. In: Vascular Epiphytes: General Biology and Related Biota. Cambridge: Cambridge University Press; 1990. Benzing DH. In: Bromeliaceae: Prole of an Adaptive Radiation. Cambridge: Cambridge University Press; 2000. Borland AM, Dodd AN. Carbohydrate partitioning in Crassulacean acid metabolism plants: reconciling potential conicts of interest. Funct Plant Biol 2002;29: 70716. Ceusters J, Borland AM, Londers E, Verdoodt V, Godts C, De Proft MP. Differential usage of storage carbohydrates in the CAM bromeliad Aechmea Maya during acclimation to drought and recovery from dehydration. Physiol Plant 2009;135:17484. FE. Purication and physical and kinetic characterization of an Cuevas IC, Podesta NAD + -dependent malate dehydrogenase from leaves of pineapple (Ananas comosus). Physiol Plant 2000;108:2408. Cushman JC. Crassulacean acid metabolism. A plastic photosynthetic adaptation to arid environments. Plant Physiol 2001;127:143948. Grifths H, Smith JAC. Photosynthetic pathways in the Bromeliaceae of Trinidad: relations between life-forms, habitat preference and the occurrence of CAM. Oecologia 1983;60:17684. Grifths H, Ong BL, Avadhani PN, Goh CJ. Recycling of respiratory CO2 during Crassulacean acid metabolism: alleviation of photoinhibition in Pyrrosia piloselloides. Planta 1989;179:11522. Hoagland DR, Arnon DI. The water culture method for growing plants without soil. Calif Exp Stn Circ 1938;347:139. Lichtenthaler HK. Chlorophylls and carotenoids: pigments of photosynthetic biomembranes. Method Enzymol 1987;148:35082. Lichtenthaler HK, Ac A, Marek MV, Kalina J, Urban O. Differences in pigment composition, photosynthetic rates and chlorophyll uorescence images of sun and shade leaves of four tree species. Plant Physiol Biochem 2007;45:57788. Luttge U. Ecophysiology of crassulacean acid metabolism (CAM). Ann Bot 2004;93:62952. Luttge U, Stimmel KH, Smith JAC, Grifths H. Comparative ecophysiology of CAM and C3 Bromeliads. 2. Field-measurements of gas-exchange of CAM bromeliads in the humid tropics. Plant Cell Environ 1986;9:37783. Martin CE, Schmitt AK. Unusual water relations in the CAM atmospheric epiphyte Tillandsia usneoides L (Bromeliaceae). Bot Gaz 1989;150:18. Maxwell K. Resistance is useful: diurnal patterns of photosynthesis in C3 and Crassulacean acid metabolism epiphytic bromeliads. Funct Plant Biol 2002;29:67987. Maxwell C, Grifths H, Borland AM, Broadmeadow MSJ, McDavid CR. Photoinhibitory responses of the epiphytic bromeliad Guzmania monostachia during the dry season in Trinidad maintain photochemical integrity under adverse conditions. Plant Cell Environ 1992;15:3747.

ARTICLE IN PRESS
L. Freschi et al. / Journal of Plant Physiology 167 (2010) 526533 533

Maxwell C, Grifths H, Young AJ. Photosynthetic acclimation to light regime and water stress by the C3-CAM epiphyte Guzmania monostachia: gas-exchange characteristics, photochemical efciency and the xanthophyll cycle. Funct Ecol 1994;8:74654. Maxwell C, Grifths H, Borland AM, Young AJ, Broadmeadow MSJ, Fordham MC. Short-term photosynthetic responses of the C3-CAM epiphyte Guzmania monostachia var monostachia to tropical seasonal transitions under eld conditions. Aust J Plant Physiol 1995;22:77181. Medina E, Delgado M, Troughton JH, Medina JD. Physiological ecology of CO2 xation in Bromeliaceae. Flora 1977;166:13752. Medina E, Ziegler H, Luttge U, Trimborn P, Francisco M. Light conditions during growth as revealed by d13C values of leaves of primitive cultivars of Ananas comosus, an obligate CAM species. Funct Ecol 1994;8: 298305. Nievola CC, Kraus JE, Freschi L, Souza BM, Mercier H. Temperature determines the occurrence of CAM or C3 photosynthesis in pineapple plantlets grown in vitro. In Vitro Cell Dev Biol Plant 2005;41:8327. Nowak EJ, Martin CE. Physiological and anatomical responses to water decits in the CAM epiphyte Tillandsia ionantha (Bromeliaceae). Int J Plant Sci 1997;158: 81826.

Popp M, Janett HP, Luttge U, Medina E. Metabolite gradients and carbohydrate translocation in rosette leaves of CAM and C3 bromeliads. New Phytol 2003;157:64956. Rayder L, Ting IP. CAM-idling in Hoya carnosa (Asclepiadaceae). Photosynth Res 1983;4:20311. Schmidt G, Zotz G. Ecophysiological consequences of differences in plant size: in situ carbon gain and water relations of the epiphytic bromeliad, Vriesea sanguinolenta. Plant Cell Environ 2001;24:10111. Schulte PJ, Nobel PS. Responses of a CAM plant to drought and rainfall: capacitance and osmotic pressure inuences on water movement. J Exp Bot 1989;40: 6170. Takahashi CA, Ceccantini GCT, Mercier H. Differential capacity of nitrogen assimilation between apical and basal leaf portions of a tank epiphytic bromeliad. Braz J Plant Physiol 2007;19:11926. Zotz G, Andrade J-L. Water relations of two co-occurring epiphytic bromeliads. J Plant Physiol 1998;152:54554. Zotz G, Hietz P. The physiological ecology of vascular epiphytes: current knowledge, open questions. J Exp Bot 2001;52:206778. Zotz G, Thomas V. How much water is in the tank? Model calculations for two epiphytic bromeliads. Ann Bot 1999;83:18392.