INTRODUCTION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION CONCLUSIONS MATERIALS AND METHODS LITERATURE CITED

Ascorbic acid plays an essential role in the survival of plant organisms. This role seems to be closely related to its antioxidant properties, providing an appropriate redox status in both symplastic and apoplastic compartments. At the extracellular level, ascorbate (ASC) is involved in defense against pathogen attack (see Noctor and Foyer, 1998) and in the regulation of cell elongation (Córdoba and González-Reyes, 1994; González-Reyes et al., 1998). Intracellular ASC has been demonstrated to be involved in the regulation of cell division and proliferation (Arrigoni, 1994; Potters et al., 2002). In both cases, ASC does not seem to exert its action directly; instead, it is used as a substrate for enzymes that regulate these processes, such as ASC peroxidase (APX), or it inhibits some other cell wall peroxidases (Takahama and Oniki, 1992, 1994; Takahama, 1993a, 1993b).

The reduced form of ASC can undergo oxidation because it is used in the different reactions yielding monodehydroascorbate (MDHA), a semi-oxidized form also known as ASC free radical, and dehydroascorbate (DHA), the fully oxidized form. Higher plants have developed different mechanisms to ensure optimal concentrations of these metabolites at the different tissues. These mechanisms include reduction of oxidized or semioxidized forms (by DHA reductase [DHAR] and MDHA reductase [MDHAR], respectively), synthesis and transport to the different cell compartments including the extracellular matrix, and transport of extracellular DHA through the plasmalemma for cytosolic reduction and recycling (Horemans et al., 2000; Paciolla et al., 2001; Smirnoff et al., 2001). Therefore, it is expected that those metabolic situations leading to significant changes in ASC content, would result in appreciable changes in the activity pattern of the enzymes and factors implicated in ASC metabolism. For example, the response of ASC and the enzymes involved in its metabolism against oxidative stress or pathogen attack has been clearly demonstrated in leaves from many species (Ranieri et al., 1996; Vanacker et al., 1998a, 1998b; Hernández et al., 2001). In these cases, apoplastic and symplastic ASC and DHA and enzymes such as SOD, catalase, APX, MDHAR, DHAR, and glutathione reductase (GR) change their activity patterns with specific features on each case. Changes have been also reported to occur in roots submitted to different stress conditions (see Schützendübel et al., 2001; Shalata et al., 2001).

Symplastic ASC has been demonstrated to play a relevant role in the control of cell division and proliferation (Arrigoni, 1994; De Pinto et al., 1999; Horemans et al., 2000). It has been shown that ASC controls the activity of prolyl hydroxylase, an enzyme that catalyzes the hydroxylation of cell wall proteins (De Gara et al., 1991). Its inhibition results in abnormal cell walls and in delayed cell cycle progression in onion (Allium cepa) roots (De Tullio et al., 1999). Also, the importance of ASC during the logarithmic growth phase in tobacco (Nicotiana tabacum) cultured cells has been emphasized (Kato and Esaka, 1999; De Pinto et al., 2000).

Apoplastic ASC has been involved in the regulation of cell expansion and elongation, although its action mechanism is not completely understood. Hidalgo et al. (1989) and González-Reyes et al. (1994) reported the stimulation of onion root growth by MDHA or ASC, and a mechanism involving redox reactions at the plasma membrane was suggested. However, a more complex role of ASC on this phenomenon is very likely. Today, it is generally accepted that ASC facilitates elongation via inhibition of enzymes involved in cell wall stiffening. In this regard, Takahama (1993a, 1993b) and Takahama and Oniki (1992, 1994) showed the inhibition of cell wall peroxidases by ASC in vitro. Because these enzymes catalyze the cross-linking of structural components of the cell wall leading to its hardening and consequent cessation of cell elongation, ASC was proposed to stimulate cell growth through its inhibitory action in cell wall peroxidases. Further evidence came from the work of Córdoba-Pedregosa et al. (1996) showing that 24- to 48-h treatments with ASC and/or L-galactono--lactone, the immediate precursor in the ASC synthesis pathway (see Smirnoff et al., 2001), resulted in decreased cell wall peroxidase activity in parallel to enhanced root growth, thus providing in vivo evidence for the correlation of peroxidase activity, ascorbic acid, and cell growth.

Fry (1998) recently proposed a nonenzymatic scission of cell wall polysaccharides mediated by hydroxyl radical, which can be produced at the cell wall from hydrogen peroxide formed by a reaction involving ASC, Cu²⁺, and oxygen. This hypothesis has been recently tested in vitro (Miller and Fry, 2001; Schopfer, 2001) and in vivo (Schopfer, 2001; Schopfer et al., 2002) and provides not only a role of active oxygen species in elongation by enhancing cell wall loosening, but also an additional explanation on the role of ASC as a regulator of cell expansion.

From the literature cited above, multiple possible interactions between ASC system and other factors such as enzymatic activities or hydrogen peroxide can be deduced. However, despite the works of De Gara et al. (1997) and Tommasi et al. (2001) who showed changes in ASC and DHA content and related enzymes activities during seed germination, there are no exhaustive studies on these interactions and their relationship with plant growth.

When cultured at constant conditions, onion roots grow under steady-state kinetics, in which cell proliferation and elongation rate are constant during several days, constituting an excellent material to investigate growth and development (for example, see Córdoba-Pedregosa et al., 1996; González-Reyes et al., 1994). In this paper, we have analyzed possible correlations between apoplastic and intracellular ASC and DHA, the activity of enzymes related to both forms, the presence of hydrogen peroxide and the process of elongation in onion roots. To achieve this goal, we have divided roots into different zones from the tip to the onion base. These zones also represent tissues in different stages of differentiation, because this process occurs from the meristem, which contains non-differentiated cells, to the onion crown, in which tissues have reached a fully differentiated state. The results included here show large differences in ASC content, ASC-related enzyme activities, and hydrogen peroxide along the root axis, indicating that each zone of the root has different and specific requirements for these metabolites and that these requirements are related to the degree of differentiation.

	RESULTS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION CONCLUSIONS MATERIALS AND METHODS LITERATURE CITED

Fraction Purity

Apoplastic fluids (AF) obtained in our experiments showed scarce contamination from the cytosol as deduced from the controls made with Glc-6-phosphate dehydrogenase (G6PDH) activity, a marker for cytosolic contamination. This activity was assayed for each apoplast extraction, and the results are given in Table I. All of the following data concerning apoplastic constituents have been corrected accordingly for cytosolic contamination.

ASC and DHA Content in Onion Roots

In total homogenates, ASC content was higher at the root tip and gradually decreased toward the onion base. At every zone, the reduced form was significantly higher than DHA, but the redox status of the molecule (ASC/ASC+DHA ratio) remained similar along the root axis (Table II).

Onion root AF contained different ASC and DHA concentrations depending on the zone. Both ASC and DHA contents were significantly lower than in total homogenates. Unlike results obtained for total homogenates (see above), DHA was the predominant form in AF. Although there was not a clear content gradient for these molecules, both forms were more abundant at zones II and III. The redox ratio was especially low (i.e. higher concentration of DHA) in apoplasts obtained from zone I (Table II).

Determination of ASC and DHA content in the tissue remaining after centrifugation for AF obtained resulted in a significant (15%-25%) loss of both forms (not shown). This was probably because of mechanical alterations of the tissue, which was found squashed against the syringe bottom after the centrifugation. However, this fact had no significant effect on enzymatic determinations. Thus activities calculated using total homogenates were nearly identical to those obtained from AF plus ISF.

ASC-Related Enzyme Activities along the Root Axis

Enzymatic activities assayed in ISF and AF varied depending on the zone of the root. In both fractions, peroxidase activity against guaiacol (GPX) was higher in zone I and then decreased in zones II and III (Fig. 1A). The activity recovered in AF represented about 3% of the ISF activity in all the three zones. An in vivo detection of peroxidase in whole roots, revealed a pattern strongly similar to that described above: higher activity in zone I and a significant decrease in zones II and III (Fig. 2A). However, the staining was not uniform along the zone I: The root cap showed intense staining, whereas the next 1 or 2 mm, corresponding to the meristem and the beginning of the elongation zone, remained practically unstained. After the elongation zone, the staining reappeared gradually (see Fig. 2B).

View larger version (29K): [in this window] [in a new window]   Figure 1.   Guaiacol-dependent peroxidase (A) and APX (B) activities in ISF and AF from the different zones of onion roots. The inset in B shows an immunoblot of ISF and AF proteins from the root zones stained with anti-APX antibody. Values are means ± SE of five independent experiments. a, P < 0.01 versus zones II and III. b, P < 0.05 versus zone II. c, P < 0.01 versus zone I. d, P < 0.01 versus zones I and II.

View larger version (24K): [in this window] [in a new window]   Figure 2.   In vivo staining of peroxidase activity in onion roots. A, The activity in a whole root. Black lines divide the root in 2-cm-length zones. B, Detail of the root tip. EZ, Elongation zone; M, meristem; RC, root cap. The black lines divide the tip in 1-mm-long zones.

In ISF APX activity was higher in zones I and III and significantly lower in zone II. However, in apoplast, this activity was practically undetected in zone I but increased gradually in zones II and III (Fig. 1B). In these zones APX ranged between 0.4% and 0.6% of the ISF activity. APX was also detected in both fractions by immunoblot using an antibody against cytosolic APX, and the results are displayed in Figure 3, insert. In this case, apoplasts from zones II and III and ISF from all the three zones yielded a band of about 28 kD. In AF from zone I, the band was nearly undetectable. These results fit very well with those obtained for APX activities.

View larger version (24K): [in this window] [in a new window]   Figure 3.   Antioxidative enzyme activities in ISF and AF from the different zones of onion roots. A, MDHAR; B, DHAR; C, GR; D, catalase. Data are mean values ± SE from five independent experiments. a, P < 0.01 versus zones II and III. b, P < 0.01 versus zone I. c, P < 0.01 versus zones I and II.

Activities of ASC-recycling enzymes (DHAR and MDHAR) also showed different patterns along the root axis. DHAR was very poorly represented in AF (0.05%-0.3% of ISF), whereas MDHAR activity ranged between 0.1% in zone I to 10% of ISF in zone III. Both activities also changed at the different root zones and according to the enzyme source (Fig. 3, A and B). In AF, MDHAR and DHAR were low in zone I and significantly increased in zones II and III. However, in ISF, the activities showed different patterns: Whereas MDHAR decreased from zone I toward the onion base, DHAR activity remained constant along the root axis.

The pattern of GR activity (Fig. 3C) was strongly similar to MDHR: higher values in zone I and subsequent decrease for ISF, and a gradual increase from zone I to zones II and III in AF. In this case, apoplasts contained between 0.3% and 4% of the ISF activity (see Fig. 3C).

Finally, we measured the distribution of catalase activity in ISF and AF along the root axis, and the results are depicted in Figure 3D. In both fractions, the highest values were obtained in zone I and decreased significantly in zones II and III. Here, the apoplast from zone I contained about 1% of the ISF activity, whereas in zones II and III, this proportion was lesser (between 0.3% and 0.4%).

Subcellular Localization of Hydrogen Peroxide in Onion Root

Pattern of hydrogen peroxide localization also varied at the different zones of the root and the results are summarized in Figure 4 and Table III. Hydrogen peroxide was detected as cerium perhydroxide electron-dense spots at the cell walls and intercellular spaces. In most of the cases, the reaction was associated with plasma membrane and middle lamellae (Fig. 4, E and G). Deposits inside the cells were not detected. In the meristematic zone, cerium deposits were found in epidermal and cortical cell walls, whereas more internal cell walls showed less reaction. In epidermis, the reaction was located in radial, tangential, and external walls (Fig. 4A). Although a significant number of cell walls from internal cells showed the electron-dense spots as well, the intensity of the reaction was weaker (Fig. 4B). At the middle of zone I (i.e. at about 1 cm from the root apex) the pattern was similar, although the reaction was less intense than in the meristematic zone. In this case, all epidermal and about a 50% of cortical cells showed the reaction. However, few walls with cerium deposits were observed in internal cells (Fig. 4C). At the middle of zone II (i.e. at about 3 cm from the root apex), all of the epidermal radial cell walls showed a very weak reaction. Also, a very low number of cortical cell walls showed a weak reaction (Fig. 4D). However, some internal cell showed a more intense reaction at the cell wall (Fig. 4E). Finally, at the middle of zone III (at about 5 cm from the root apex), the reaction was very weak or undetectable for epidermal, cortical, and internal cell walls (Fig. 4F). In Figure 4G, we show a detail of a meristematic cell wall in which the reaction seems to be directly related to the plasma membrane and middle lamellae. Preincubation of roots with sodium pyruvate, a H₂O₂ scavenger, prevented the precipitation of cerium perhydroxide (Fig. 4H). A quantitative analysis on the results described above is shown in Table III.

View larger version (115K): [in this window] [in a new window]   Figure 4.   Subcellular localization of hydrogen peroxide detected as cerium perhydroxide deposits (arrows) in onion roots. A, An epidermal cell from the meristematic zone. External (ecw), tangential (tcw), and radial (rcw) cell walls are shown. B, Cortical meristem section. C and D, Examples of cortical cell walls from MZI and MZII, respectively. E, An internal cell of MZII. F, Cell wall from the internal region of MZIII. G, Cell wall in which the reaction is found in plasma membranes (PM) as well as in the middle lamellae (ML). H, Cell wall between two meristematic cells from a tissue preincubated with sodium pyruvate as control. From A to F, bar = 2 µm; for G and H, bar = 0.2 µm.

	DISCUSSION

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION CONCLUSIONS MATERIALS AND METHODS LITERATURE CITED

The present results show significant differences in ASC content, ASC-related enzyme activities, and H₂O₂ location along the root axis in onion and that these differences can be related to the tissue dynamics of each root zone. Furthermore, our results also show differences between symplastic and apoplastic fractions, suggesting that both compartments have different metabolite requirements and probably are subjected to specific control mechanisms. To our knowledge, this is the first exhaustive report on zonal differences for these parameters in higher plant roots. On the other hand, it is important to note that our study has been developed under controlled conditions so that roots were growing at constant rate, and no symptoms of pathogen presence were observed during the experiments. Therefore, our results can be considered as a control situation of healthy growing onion roots.

ASC/DHA in Onion Roots

ASC content in onion roots growing hydroponically varied depending on the zone. A marked decreasing gradient from the root cap to the onion base was found for total homogenates. This gradient seems to occur for both reduced (ASC) and oxidized (DHA) forms, but being the first the predominant form along the root axis. In whole roots, total content was similar to that reported previously in the same material (Córdoba-Pedregosa et al., 1996; De Tullio et al., 1999).

As far as we know, no information is available about ASC and DHA contents in root apoplasts, and our data showed several interesting features. First, content of both metabolites is low in this compartment, especially in zone I, compared with other organs. For example, higher contents have been reported in leaves from different species (Vanacker et al., 1998a, 1998b; van Hove et al., 2001; Kollist et al., 2001; Veljovic-Jovanovic et al., 2001). In leaves, apoplastic content is about 10% of total ASC. In our case, ASC+DHA represent between 2% and 8% of total homogenate contents. The standard extraction of apoplast with previous vacuum infiltration of the tissue led to a fluid with no detectable ASC or DHA (Córdoba-Pedregosa et al., 1996). This is probably attributable to the diffusion of both forms to the medium during the vacuum period. We partially avoided this problem by submitting the root segments to centrifugation without previous infiltration (see "Materials and Methods"). Moreover, it should be noted that hydroponic culture can facilitate the diffusion of soluble and small molecules (such as ASC and DHA) to the medium, making their detection in the apoplast difficult.

Second, ASC and DHA did not show the same distribution pattern in onion root apoplasts as in total homogenates. Instead, the zones closer to the onion base (zones II and III) showed higher concentration of both forms compared with zone I. Also, DHA concentration was higher than ASC, with mean redox status ratios from 0.13 to 0.27 depending on the root zone. This is in accordance with the observation of Vanacker et al. (1998a, 1998b) and Kollist et al. (2001) in leaves from different species, and confirms the high oxidation rate of ASC in root apoplast as well.

It is also interesting to note that apoplastic ASC+DHA content did not represent a constant proportion of whole ASC+DHA. In zone I, apoplastic ASC+DHA was about 2% of total content, whereas for zone III this proportion was about 8%. This result could indicate a difference in ASC transport from the symplast to the cell wall or a different rate of consumption/regeneration of these metabolites at the different zones of the root.

ASC-Related Enzymatic Activities in Onion Roots

In this study, we have determined a number of enzymatic activities related to ASC metabolism and the antioxidative response. These enzymes have been detected in the apoplastic compartment in other organs and species, especially in leaves (Vanacker et al., 1998a, 1998b; Blokhina et al., 2001; Veljovic-Jovanovic., 2001), and occupy only a small part of total activities in apoplast, similar to catalase, GPX, APX, and GR in onion roots. The more striking differences are a low apoplastic DHAR activity (especially in zone I) and high MDHAR and GR activities found in AF from zones II and III. Our results of total activities in ISF were similar to those reported for whole roots in tomato (Lycopersicum pennellii; Shalata et al., 2001) and pine (Pinus sylvestris; Schützendübel et al., 2001) with the exception of DHAR, which was not detected in these plants. The very low apoplastic DHAR activity found along the root axis could be attributable to ISF contamination. However, our results do not necessarily mean that AF from onion root lack DHAR because this enzyme has proven to be very unstable and might not withstand the extraction procedure to obtain AF samples (Hossain and Asada, 1984a; Foyer and Mullineaux, 1998).

In onion roots, each enzyme showed a specific and differential activity pattern depending on its apoplastic or symplastic origin and the zonal location. ASC-related enzymes have been exhaustively reported to change in different organs when plants are submitted to pathogen attack or to experimental conditions leading to oxidative stress or abnormal growth. For example, changes in ASC/DHA contents and in apoplast and cytosolic enzymatic activities have been reported in leaves from different species submitted to mildew attack or oxidative stress. In these cases, the plant response was not uniform, and significant differences were found even when comparing different cultivars from the same species (for example, see Ranieri et al., 1996; Vanacker et al., 1998, 1998b; Hernández et al., 2001).

In roots, the ASC system and related enzymes also change under stress situation. In this way, Shalata et al. (2001) have reported a high efficiency of the antioxidative system against salt-induced stress in tomato, showing that the wild tolerant variety L. pennellii Corr. (D'Arcy) increases superoxide dismutase, catalase, MDHAR, APX, and reduced ASC and glutathione. On the other hand, Schützendübel et al. (2001) have reported changes in these metabolites in roots of pine submitted to cadmium. However, as suggested by these authors, many of these changes can be explained by an acceleration of differentiation and consequently accelerated aging of root tissues. In this regard, our results can be also explained in terms of root aging, because zone I represents the younger zone of the root, where meristem and elongation zones (i.e. non-differentiated cells) are located, and zone III (fully differentiated cells) includes the oldest cells of the root.

GPX activity has been proposed to be involved in cell wall stiffening and cessation of growth (Sánchez et al., 1996; De Souza and MacAdam, 1998). Our data show that GPX activity is higher in zone I, which includes meristem and elongation zone. The in vivo staining of peroxidase activity confirmed that zone I has a high activity, but also revealed that the meristem and the elongation zone remained practically unstained. These results, together with high content of ASC at the meristem (see RC+M in Table II), are compatible with the proposed role of peroxidases and ASC in cell expansion (see Córdoba and González-Reyes, 1994; González-Reyes et al., 1998).

From a metabolic point of view, it is clear that zone I is more dynamic than zones II and III. For example, cell proliferation and elongation are restricted to zone I, whereas zones II and III sequentially represent older and more differentiated cells. Also, zone I includes the root cap and the quiescent center in which very low levels of intracellular ASC have been reported (Kerk and Feldman, 1995; Kerk et al., 2000). These authors suggested that an appropriate interaction between the ASC system and auxin is crucial for the maintenance of the quiescent center and consequently for organization of the meristem. In this respect, our results could represent a specific status quo of each root zone and could lead us to hypothesize a delicate balance of the ASC system and related enzymes at every root segment.

Cell proliferation occurs exclusively at the meristem. In this zone, ASC seems to play an essential role because its synthesis inhibition results in a significant decrease of the mitotic index (Arrigoni, 1994). Its precise role on proliferation seems to be related to its presence as a cofactor in the synthesis and transport of Hyp-rich proteins, which are essential for cell cycle progression (De Tullio et al., 1999). However, other possible roles of ASC on cell proliferation cannot be discarded (Potters et al., 2002). Interestingly, the highest amount of intracellular ASC occurs at this zone of the root (see Table II, RC+M) and the high MDHAR activity found in ISF from zone I may have a role in the maintenance of a high redox status for ASC at this root zone.

Cell elongation is the second phenomenon contributing to growth. In onion roots, elongation has been proposed to depend in part on intracellular and extracellular ASC (Córdoba and González-Reyes, 1994; Córdoba-Pedregosa et al., 1996). In roots, cell elongation is restricted to the meristem and the so-called "elongation zone," which extends some millimeters from the meristem toward the bulb base but is fully included in our zone I. Surprisingly, we did not find significant amounts of ASC or DHA at AF in zone I. The possible diffusion of these metabolites during tissue handling (see Materials and Methods") together with the oxidative nature of the apoplastic compartment (Vanacker et al., 1998a, 1998b; Blokhina et al., 2001) can contribute to maintaining low ASC content in AF. Nevertheless, the possibility exists that a high amount of apoplastic ASC could be concentrated at the meristematic and elongation zones. This fact could contribute to explain the low peroxidase activity found in the in vivo staining and the high rates of proliferation and elongation in these root regions. Extraction of clear AF from isolated meristems and elongation zones should help to solve this problem. However, the size of these regions (about 1-2 mm length) makes such an extraction unpractical.

Hydrogen Peroxide in Onion Roots

According to results obtained with cerium chloride technique, we have shown that in onion roots, hydrogen peroxide is located mainly in cell walls, and very few spots of cerium precipitates were located inside the cells. In these few cases, distribution of spots was not uniform and did not follow any specific pattern. Similar results have been reported by Bestwick et al. (1997, 1998) and Pellinen et al. (1999). In some cases (Bestwick et al., 1998; Pellinen et al., 1999), intracellular detection of hydrogen peroxide was possible only after several hours of pathogen infection. These results suggest that, in control conditions, the amount of hydrogen peroxide produced inside the onion root cells is low and is not detectable by cytochemistry. In our material, cerium perhydroxide spots were found in cell walls from epidermis, in meristematic cells, and in other cells in differentiation and were mainly associated with the plasma membrane and the middle lamellae.

Location of hydrogen peroxide in cell walls has been correlated to the presence of pathogens or to cell differentiation phenomena, including wall lignification (Ogawa et al., 1997; Wojtaszek, 1997; Potikha et al., 1999), and seems to be absent in cells undergoing elongation (see Wojtaszek, 1997). However, the examination of our samples at the electron microscope did not reveal the presence of pathogens in any case. On the other hand, in our experiments we have found a high number of meristematic cells showing discrete amount of cerium spots in their cell walls. Similar results have been reported by Blokhina et al. (2001) in wheat (Triticum aestivum) and rice (Oryza sativa) roots. Rodríguez et al. (2002) have very recently detected hydrogen peroxide in apoplast from elongation zone of maize leaves and suggest that it is necessary for leaf elongation.

Fry (1998) and Miller and Fry (2001) have proposed that extracellular H₂O₂ is necessary for the induction of xyloglycan breakage. In this model, apoplastic ASC is involved in the production of hydroxyl radicals via a nonenzymatic reaction (Miller and Fry, 2001), with these radicals being responsible for polysaccharide scission and the consequent cell wall relaxation (Schopfer, 2001, 2002). Our results showing cerium spots in cell wall from meristems and elongation zones, are compatible with the hypothesis of hydroxyl radical-mediated wall loosening and partially explain the low amount of ASC found in apoplast of zone I. The higher catalase activity detected in that zone could also be involved in the control regulation of H₂O₂ in zone I. The reaction found in deeper and thicker cell walls at zones II and III most probably corresponds to elements in different stages of differentiation. Similar results have been reported by Pellinen et al. (1999) and Potikha et al. (1999).

	CONCLUSIONS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION CONCLUSIONS MATERIALS AND METHODS LITERATURE CITED

The present results show a close relationship among ASC content, peroxidase activity, ASC-related enzyme activities, and H₂O₂ localization at the different zones of onion roots in apoplastic and symplastic compartments. In the apoplast, these constituents are most probably related to the dynamics of extracellular matrix to provide the optimal conditions for cell wall loosening during elongation, as occurs in meristems and elogation zone, or for wall stiffening, as occurs in differentiation zones and toward the onion base. The symplastic compartment is the site of synthesis of all of these constituents and the transport of each one to the cell wall will depend in great extent on the function and physiological status of each region of the root. Therefore, a strict regulation of the synthesis, transport, and activity of these constituents in both compartments, is necessary for the maintenance of the differential functionality of cells and tissues at every root zone.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION RESULTS DISCUSSION CONCLUSIONS MATERIALS AND METHODS LITERATURE CITED

Growth Conditions

Onion (Allium cepa) roots were grown hydroponically in the dark at 25°C. Once roots had reached 6 cm length, they were detached from the bulbs and cut into three zones of 2 cm length each. The zone size was the minimum possible able to be handled without appreciable damage to the root and yielding AFs with a low contamination of cytosol (see below).

Isolation of AFs and ISFs

About 2 g of each type of segment was quickly washed in distilled water, placed in petri dishes in 10 mM sodium phosphate, pH 6, containing 1.5% (w/v) polyvinylpolypyrrolidone, 1 mM EDTA, and 0.5 mM phenylmethylsulfonyl fluoride, and submitted to vacuum (60 kPa) for 5 min at 4°C. Afterward, root zones were carefully dried with filter paper and placed in syringes, which were placed in centrifugation tubes. Roots were centrifuged at 150g for 5 min, and the AF recovered at the bottom of the tubes. With this procedure, we obtained 70 to 110 µL of AF for 1 g fresh weight of each zone. The remaining roots were used to obtain the ISF after homogenization in the same medium with an Ultraturrax T-25 (IKA Labortecnik, Staufen, Germany) and centrifugation at 15,000g for 30 min. Cytosolic contamination of AF was monitored by assaying G6PDH activity as marker.

ASC and DHA Determination

For determination of apoplastic ASC and DHA, root zones were not vacuum-infiltrated, because previous experience (Córdoba-Pedregosa et al., 1996) had showed a significant loss of ASC and DHA during the infiltration process. Instead, zones were obtained, washed, and quickly blotted dry onto filter paper and placed into the syringes. In this cases, AFs were collected in centrifuge tubes containing a concentrated solution of metaphosphoric acid, so that its final concentration was 5% once apoplast had been obtained. With this procedure, we obtained 60 to 100 µL of apoplast for 1 g fresh weight of each zone. In another set of experiments, whole intact roots were homogenized in 5% (w/v) metaphosphoric acid to determine total ASC and DHA contents at each root zone.

ASC content was estimated using the bipyridyl method described by Knörzer et al. (1996). An ASC standard calibration curve was previously run, and an extinction coefficient of 16.5 mM¹ was obtained. For the determination of ASC contents, samples of AF and total homogenates (125 µL) were neutralized with 25 µL of 1.5 M triethanolamine, and after mixing, 150 µL of 150 mM sodium phosphate (pH 7.4) and 150 µL of water were added. For the determination of total ASC (ASC+DHA), the samples were neutralized, phosphate buffer and water were added, and then 75 µL of 10 mM dithiothreitol (DTT) was added and incubated for 15 min at room temperature. To remove excess DTT, 75 µL of 0.5% (w/v) N-ethylmaleimide were added. Samples were then mixed and incubated for at least 30 s at room temperature. For further procedures, both samples (without and with DTT) were treated identically as described by Knörzer et al. (1996). Reading of absorbance was at 525 nm. DHA contents were calculated from the difference of ASC content measured with and without DTT preincubation.

Finally, ASC and DHA contents were also determined in homogenates obtained from root cap and meristems as well as in elongation zone. For this purpose, roots were detached from the bulbs and two small portions of about 2 mm lengths were obtained for each root. The first piece contains the root cap and the meristem, whereas the second one contains the elongation zone. By this procedure we collected about 0.1 g of each part.

Enzymatic Activities

Enzymatic activities were spectrophotometrically assayed for AF and ISF obtained from each root zone. Except where noted, reactions were developed at 25°C for 5 min, with stirring, in a final volume of 1 mL containing 25 to 35 µg of protein.

GPX was determined according to Zheng and Van Huystee (1992). The reaction mixture contained 10 mM sodium phosphate (pH 6), 0.1 mL of 0.3% (v/v) H₂O₂, and 0.1 mL of 1% (v/v) guaiacol. Reaction was initiated by the addition of H₂O₂ and followed at 470 nm (extinction coefficient of guaiacol = 26.6 mM¹ cm¹) at 30°C.

For APX determination, root zones were vacuum-infiltrated in phosphate buffer containing 5 mM ASC. For ISF, homogenization buffer also contained 5 mM ASC. The activity was measured by the method of Nakano and Asada (1981). The reaction mixture contained 50 mM phosphate buffer (pH 7), 1 mM sodium ASC, and 2.5 mM H₂O₂. After the addition of ASC to the mixture, the reaction was followed at 290 nm (extinction coefficient of ASC = 2.8 mM¹ cm¹).

MDHAR was assayed following the method of Hossain et al. (1984). The reaction mixture contained 50 mM Tris-HCl (pH 7.5), 0.2 mM NADH, 2.5 mM ASC, and 0.15 unit of ASC oxidase. The reaction was initiated by adding ASC oxidase to the mixture, thus generating the substrate MDHA. Activity was measured as the ASC oxidase-induced oxidation of NADH. The reaction was monitored at 340 nm (extinction coefficient for NADH = 6.2 mM¹ cm¹).

DHAR was assayed according to Hossain and Asada (1984b) by measuring the reduction of DHA to ASC in a reaction mixture containing 50 mM potassium phosphate (pH 7), 0.5 mM DHA, and 2.5 mM GSH. The reaction was followed at 265 nm (extinction coefficient of ASC = 14 mM¹ cm¹).

Catalase activity was estimated using the method of Aebi (1983). The reaction mixture contained 50 mM potassium phosphate (pH 7) and 10 mM H₂O₂. After enzyme addition, the reaction was monitored by following decomposition of H₂O₂ at 240 nm (extinction coefficient of H₂O₂ = 43.6 mM¹ cm¹).

GR was measured according to Foyer and Halliwell (1976). The reaction was developed in 50 mM Tris-HCl (pH 7.5) containing 2.5 mM MgCl₂, 0.5 mM GSSG, and 0.2 mM NADPH. Oxidation of NADPH was followed at 340 nm (extinction coefficient = 6.2 mM¹ cm¹).

G6PDH assay was developed in 100 mM Tris-HCl (pH 8) containing 1 mM MgCl, 0.2 mM NADP⁺, and 1 mM Glc-6-phosphate. Generation of NADPH was measured at 340 nm, and the extinction coefficient was 6.2 mM¹ cm¹ (Weimar and Rothe, 1986).

In Vivo Detection of Peroxidase Activity

We used the method of De Pinto and Ros-Barceló (1997). Onions growing under normal conditions and showing roots of about 6 cm long were transferred to a medium consisting of 0.1 M Tris-acetate (pH 5), 0.1 mM 4-chloro-naphtol, and 0.9 mM H₂O₂, at pH 5. After several minutes of incubation, a darkening reaction began to be appreciable in the roots, which were then immediately photographed.

Electrophoresis and Western Blotting

After obtaining apoplasts, samples (15 µg protein) were submitted to SDS-electrophoresis in 12% (w/v) acrylamide gels. After separation, proteins were transferred to nitrocellulose filters (pore size, 0.45 µm; Immobilon-NC, Millipore, Bedford, MA) and incubated with an anti-APX antibody raised against cytosolic peroxidase from spinach (Spinacia oleracea; 1:500, v/v) for 4 h. Afterward, blots were incubated for 45 to 60 min with anti-IgG-alkaline phosphatase-conjugated secondary antibody diluted 1:2,000 (v/v) and then revealed in a mixture of nitroblue tetrazolium and 5-bromo-4-chloro-3-indolyl phosphate. For immunodetection of APX, AF and ISF were obtained in the same way as for determination of APX activity.

Electron Microscopy

Hydrogen peroxide was detected by cytochemistry. Roots were detached from the bulbs, and five to six pieces (0.5 mm length) were processed. The detection of H₂O₂ is based in the formation of cerium perhydroxide from exogenous cerium chloride and endogenous H₂O₂, as described by Bestwick et al. (1997). In brief, pieces from different zones of the root were obtained and preincubated in 50 mM MOPS buffer, pH 7, containing 5 mM CeCl for 1 h. Then, samples were quickly washed in the buffer and fixed in 2.5% (w/v) glutaraldehyde-2% (w/v) paraformaldehyde mixture in 0.1 M sodium cacodylate buffer (pH 6.8) for 4 h at 4°C. Then, samples were washed in buffer and post-fixed in 1% (w/v) osmium tetroxide, dehydrated in an ethanol series, treated with propylene oxide, and embedded in Epon 812. After curing, sections of about 60 nm thickness were obtained in an ultramicrotome, mounted on Ni grids, observed unstained, and photographed in an electron microscope (300 EM, Philips, Eindhoven, The Netherlands). In some cases, semithin sections (0.5-1 µm thick) were obtained and stained with toluidine blue. For control, some pieces were preincubated for 15 min in 10 mM sodium pyruvate (Sigma-Aldrich, St. Louis) because this molecule has been reported to be a strong hydrogen peroxide scavenger (Li et al., 1998).

Protein Determination

Protein was determined by the dye-binding method of Bradford (1976), using -globulin as a standard.

Statistical Analysis

In all experiments, mean values were compared using Student's t test. Significance levels of 95% (P < 0.05) or 99% (P < 0.01) are indicated in table and figure legends.