Glutamate Dehydrogenase of Tobacco Is Mainly Induced in the Cytosol of Phloem Companion Cells When Ammonia Is Provided Either Externally or Released during Photorespiration

Abstract

Glutamate (Glu) dehydrogenase (GDH) catalyses the reversible amination of 2-oxoglutarate for the synthesis of Glu using ammonium as a substrate. This enzyme preferentially occurs in the mitochondria of companion cells of a number of plant species grown on nitrate as the sole nitrogen source. For a better understanding of the controversial role of GDH either in ammonium assimilation or in the supply of 2-oxoglutarate (F. Dubois, T. Tercé-Laforgue, M.B. Gonzalez-Moro, M.B. Estavillo, R. Sangwan, A. Gallais, B. Hirel [2003] Plant Physiol Biochem 41: 565–576), we studied the localization of GDH in untransformed tobacco (Nicotiana tabacum) plants grown either on low nitrate or on ammonium and in ferredoxin-dependent Glu synthase antisense plants. Production of GDH and its activity were strongly induced when plants were grown on ammonium as the sole nitrogen source. The induction mainly occurred in highly vascularized organs such as stems and midribs and was likely to be due to accumulation of phloem-translocated ammonium in the sap. GDH induction occurred when ammonia was applied externally to untransformed control plants or resulted from photorespiratory activity in transgenic plants down-regulated for ferredoxin-dependent Glu synthase. GDH was increased in the mitochondria and appeared in the cytosol of companion cells. Taken together, our results suggest that the enzyme plays a dual role in companion cells, either in the mitochondria when mineral nitrogen availability is low or in the cytosol when ammonium concentration increases above a certain threshold.

Ammonium is the ultimate form of inorganic nitrogen available to the plant. It can originate from a wide variety of metabolic processes such as nitrate reduction, photorespiration, phenylpropanoid metabolism, utilization of nitrogen transport compounds, amino acid catabolism, symbiotic nitrogen fixation (Hirel and Lea, 2001), and insect digestion in carnivorous plants (Schulze et al., 1997). It is then incorporated into an organic molecule, 2-oxoglutarate, by the combined action of the enzymes Gln synthetase (GS) and Glu synthase (Gln-2-oxoglutarate aminotransferase; GOGAT) to allow the synthesis of Gln and Glu. Both amino acids are further used as amino group donors for the synthesis of virtually all the nitrogen-containing molecules including the other amino acids needed for protein synthesis and nucleotides used as basic molecules for RNA and DNA synthesis (Hirel and Lea, 2001). The GS/GOGAT cycle is the major mechanism of ammonium assimilation in higher plants regardless of the various sources of ammonium listed above. However, it has often been argued that other enzymes have the capacity to assimilate ammonium, leading to the hypothesis that alternative pathways might operate under particular physiological conditions when the GS/GOGAT pathway may not be able to fulfill its function (Harrison et al., 2003).

One of these alternative pathways is the reaction catalyzed by the mitochondrial NAD(H)-dependent Glu dehydrogenase (GDH; EC 1.4.1.2), which possesses the capacity to assimilate ammonium in vitro utilizing the organic molecule 2-oxoglutarate to synthesize Glu. This observation led a number of authors to propose that GDH could operate in the direction of ammonium assimilation (Yamaya and Oaks, 1987; Oaks, 1995; Melo-Oliveira et al., 1996) although all the ¹⁵N labeling experiments performed in vivo on a variety of plants demonstrated that GDH operates in the direction of Glu deamination (Robinson et al., 1992; Fox et al., 1995; Stewart et al., 1995; Aubert et al., 2001). It was concluded that GDH is involved in the supply of 2-oxoglutarate rather than in assimilation of ammonium when carbon becomes limiting (Robinson et al., 1992; Aubert et al., 2001; Miflin and Habash, 2002). The physiological role of GDH in the whole plant context remains speculative given the recent finding that the majority of the GDH protein is located in the mitochondria of companion cells (CCs; Dubois et al., 2003).

Studies on tobacco (Nicotiana tabacum) leaf source/sink relationships have shown that GDH is induced when plants are grown on ammonium as the sole nitrogen source (Tercé-Laforgue et al., 2004), a situation in which leaf protein nitrogen remobilization is limited. In contrast, both GDH transcripts and activity remain at a very low level when nitrogen remobilization is maximal as the result of nitrogen starvation. It was therefore concluded that GDH does not play a direct role during nitrogen remobilization but is rather induced following an accumulation of ammonium into the leaves when applied externally or released in the sieve tubes as the result of protein hydrolysis during senescence (Tercé-Laforgue et al., 2004). It was therefore hypothesized that the induction of GDH was not physiologically relevant since the enzyme does not assimilate ammonium in vivo (Robinson et al., 1992; Aubert et al., 2001). As proposed before (Loulakakis and Roubelakis-Angelakis, 2001), the ammonium-dependent induction of the enzyme was the result of a more general response to a metabolic stress due to the toxic and pleiotropic effect of ammonium on cellular metabolism (Gerendás et al., 1997).

To further investigate the putative role of externally applied ammonium in the regulation of GDH protein synthesis, the subcellular localization of the enzyme was studied in ammonium fed (NH₄⁺) and nitrogen limited (NL) plants with a particular emphasis on highly vascularized organs where the enzyme is mostly present. In addition, the subcellular localization of GDH was also investigated in transformed tobacco plants expressing a partial ferredoxin-dependent GOGAT (Fd-GOGAT) cDNA in the antisense orientation during the transition from CO₂ enrichment (where photorespiration is inhibited or greatly reduced) to air (where photorespiration is a major process of ammonium production in leaves; Ferrario-Méry et al., 2000). In air, the leaves of low Fd-GOGAT expressors accumulate more ammonium than the untransformed controls (Ferrario-Méry et al. 2000). This flexible biological system therefore allows the changes in the subcellular localization of GDH when ammonium is released internally from the leaf mesophyll to be studied.

RESULTS

Relative Ammonium Concentration in the Phloem Sap

To determine the concentration of ammonium and amino acids in the phloem, samples of sap were collected in NH₄⁺ and NL plants, in Fd-GOGAT antisense plants, and in untransformed control plants using the same young leaf in all experiments. This young leaf was used in both NH₄⁺ and NL plants because older leaves were not able to exude. In Fd-GOGAT antisense plants, photorespiration and thus ammonium production in the mitochondria has been shown to be higher in young photosynthetically active leaves (Leegood et al., 1995). The ammonium concentration of the phloem sap was 16 times higher in NH₄⁺ plants compared to NL plants (Table I). In Fd-GOGAT antisense plants transferred from high CO₂ to air for 3 d, there was a 2-fold increase in ammonium compared to the untransformed control plants (Table I). The total concentration of free amino acids in the phloem sap was 3 times higher in NH₄⁺ plants due to a preferential accumulation of Pro and Ser in response to the excess of ammonium (Brugière et al., 1999). An accumulation of Gln was observed in the phloem sap of Fd-GOGAT antisense plants as the result of a lower enzyme activity in these transgenic plants (Ferrario-Méry et al., 2000).

Table I.

Ammonium and amino acid concentrations in the phloem sap

	NL	NH4+	UC	L15
NH4+	14.0 ± 3.7	236.1 ± 30.1	41.2 ± 13.1	74.2 ± 14.2
Asp	14.9 ± 2.9	7.7 ± 2.5	6.6 ± 1.5	8.6 ± 1.8
Asn	nda	8.4 ± 0.4	7.4 ± 1.2	2.4 ± 0.4
Ser	25.6 ± 2.7	116.2 ± 16.2	11.1 ± 1.6	13.0 ± 2.1
Glu	15.3 ± 2.2	30.4 ± 1.9	5.7 ± 1.1	7.6 ± 0.7
Gln	1.0 ± 0.1	34.7 ± 4.4	33.6 ± 4.4	56.6 ± 9.5
Pro	nda	111.5 ± 15.6	30.2 ± 5.4	26.1 ± 4.2
Others	74.8 ± 16.4	115.8 ± 17.4	37.8 ± 6.4	39.5 ± 6.4
Total	137.1 ± 29.8	433 ± 53.1	132.4 ± 22.8	143.8 ± 23.1

Ammonium and aminoacids concentrations (nmol mL⁻¹) in the phloem sap of NH₄⁺ plants, NL plants, Fd-GOGAT antisense plants (L15), and untransformed control plants (UC). Values are the means ± sd.

^and, Not detectable.

GDH Activity and Protein Levels in Stems and Midribs of Plants Grown on Low Nitrogen and Ammonium and in Fd-GOGAT Antisense Plants

In previous publications we have shown that ammonium is one of the main effectors responsible for the increase or the induction of GDH at the activity, protein, and gene expression levels (Tercé-Laforgue et al., 2004), and that the protein is mostly if not exclusively localized in the CCs (Dubois et al., 2003).

Both GDH aminating and deaminating activities were therefore measured in stems of NH₄⁺ and NL plants following manual separation of the central cylinder (containing most of the phloem tissues) from the cortical parenchyma cell layers. Using this technique, it was possible to demonstrate that GDH aminating activity was always higher in the central cylinder compared to the cortical cell layers, along the stem length in both NH₄⁺ and NL plants (Table II). In addition, highest activity was measured in the central cylinder in the basal stem part of NH₄⁺ plants, whereas at the top of the stems its activity was 3 times lower regardless of the mode of nitrogen nutrition. Although this technique was not fully accurate in terms of tissue separation, it demonstrated that the enzyme is more active in the zone of the stem containing vascular tissue. A similar pattern was observed for GDH deaminating reaction, but its activity was approximately 10 times lower compared to its aminating counterpart (data not shown). GDH aminating activity was also much higher in the leaf midrib of NH₄⁺ plants (Table II).

Table II.

NADH-GDH activity in organs containing vascular tissue

		Stems
	Midribs	Cortex		Central Cylinder			Petioles	Midribs
		Base	Top	Base	Top
NH4+:	218.8 ± 31.4	102.6 ± 2.9	35.0 ± 10.8	321.4 ± 39.3	88.4 ± 11.6	L15:	256.8 ± 37.5	188.2 ± 51.3
NL:	77.6 ± 26.3	72.3 ± 3.2	47.4 ± 2.6	186.5 ± 12.6	71.4 ± 4.3	UC:	106.2 ± 37.8	138.3 ± 42.5

NADH-GDH activity (nmol min⁻¹ mg protein⁻¹) in midribs and stems of NH₄⁺ plants, NL plants, and in petioles and midribs of Fd-GOGAT antisense plants (L15) and untransformed control plants (UC). Values are the means ± sd.

In young leaves of Fd-GOGAT antisense plants in which photorespiratory ammonia accumulates following transfer from high CO₂ to ambient atmosphere (Ferrario-Méry et al., 2000), an increase in GDH activity was found to occur both in the leaf midrib and in the petiole compared to the untransformed control plants (Table II).

Using grapevine (Vitis vinifera) antibodies, already successfully employed for the immunolocalization of GDH in a number of plant species (Dubois et al., 2003), western-blot experiments were conducted to determine if the amount of protein was higher in organs containing phloem tissues. In the leaf midrib, the amount of GDH protein was much higher in NH₄⁺ plants compared to NL plants (Fig. 1A).The amount of GDH protein isolated from the basal part of the stem was substantially higher in the central cylinder compared to the cortical parenchyma cell layers in both NH₄⁺ and NL plants although, as already observed for the midrib, larger amounts of protein were present in NH₄⁺ plants. At the top of the stem there were no major differences in the GDH protein content of the cortex whether the plants were nitrogen starved or grown on ammonium as the sole nitrogen source (Fig. 1A).

Figure 1.

Protein gel-blot analysis of GDH in phloem-containing tissues. A, GDH protein content in midribs and stem sections of NH₄⁺ and NL plants. B, Amount of GDH protein in petioles and midribs of untransformed tobacco control plants (UC) and Fd-GOGAT antisense plants (L15). The amount of protein loaded in each track was calculated on a similar DW basis for each leaf sample (0.5 mg). The relative amount of GDH protein was quantified following densitometric scanning using the highest protein content (100%) as a maximum value.

A slight increase in the amount of GDH protein was observed in both the petiole and the midrib of Fd-GOGAT antisense plants following a 3-d transfer from high CO₂ to air compared to untransformed plants (Fig. 1B).

The variations in GDH activity and in the amount of the corresponding protein in leaf mesophyll tissue are not presented because it was not possible to remove all the minor veins containing some phloem tissue (see below).

Subcellular Localization of GDH by Immunogold Transmission Electron Microscopy

To refine the localization of GDH in phloem-rich tissues such as stems, petioles, and midribs and to determine whether ammonium provided externally or internally released during photorespiration had an impact on the subcellular distribution of the protein, immunocytochemical electron microscopy experiments were conducted. In previous physiological studies experiments were performed on leaves that still contained some minor veins in the mesophyll tissue (Masclaux et al., 2000; Tercé-Laforgue et al., 2004). Therefore, in this study the localization of GDH was investigated in deribbed leaf fragments.

Figure 2A shows a partial view of a mature leaf mesophyll cell of NH₄⁺ plants after incubation with the GDH antiserum. The section was devoid of gold particles, indicating the absence of GDH protein. However, it is possible that very low amounts of protein were present in the mesophyll cells and therefore at the limit of detection of the immunocytochemical technique. No gold particles were observed when a similar section was treated with preimmune serum (Fig. 2B). Similar results were obtained in NL plants (data not shown). Quantification of gold particles confirmed the absence of GDH both in the mitochondria and cytosol of leaf mesophyll, considering that the background level is around 4 particles/μm². For the cytosol, the labeling was close to the background level, and the changes in GDH protein content observed both in the mitochondria and in the cytosol of NH₄⁺ plants were not significant (Table III).

Figure 2.

Immunolocalization of GDH in mature tobacco leaf tissues of NH₄⁺ and NL plants. A, Mesophyll cell of a mature leaf of NH₄⁺ plants. B, Mesophyll cell of a mature leaf of NH₄⁺ plants treated with preimmune serum. C, CC of a leaf minor vein of NH₄⁺ plants. Arrows indicate labeling in the vacuolar material. D, CC of a leaf minor vein of NL plants. Cy, Cytosol; m, mitochondria; P, plastids; S, starch granule; V, vacuole. Bars = 1 μm.

Table III.

Quantification of GDH protein in different tissue sections of tobacco plants

			Number of Gold Particles/μm2
	Mitochondria		Cytosol		Vacuole
	NL Plants	NH4+ Plants	NL Plants	NH4+ Plants	NL Plants	NH4+ Plants
Leaf mesophyll cells	3.83 ± 1.3	5.1 ± 1.3b	0.54 ± 0.1	0.87 ± 0.26b	ndc	ndc
Minor veins CC	28 ± 4.1	42 ± 6.9b	4.74 ± 0.8	12.6 ± 3a	7 ± 2	26 ± 8a
Midribs CC	21 ± 3.6	45 ± 3.6a	5.5 ± 0.45	13 ± 3.5a	20 ± 7	61 ± 13a
Base stems CC	52 ± 7.6	104 ± 4.3a	5.7 ± 0.7	20 ± 4.2a	14 ± 5	31 ± 10a
Top stems CC	30 ± 4.5	54 ± 8.2a	6.2 ± 1	9.5 ± 0.34a	ndc	ndc

Immunolocalization of the enzyme was performed using transmission electron microscopy on plants grown on NH₄⁺ and NL plants. Values are the mean ± sd of gold particles counted on 3 to 20 different sections. See footnotes for induction levels of GDH protein.

^aSignificant.

^bNot significant at 0.05 probability level.

^cnd, Not detectable.

In contrast, in the CCs of the minor veins of NH₄⁺ plants gold particles were mostly present in the mitochondria. However, some labeling was detected in the cytosol and the vacuolar material (Fig. 2C). In the CCs of minor veins of NL plants the labeling was weak and mostly visible in the mitochondria (Fig. 2D). Quantification of gold particles showed that in the minor vein CCs the increase in GDH protein content occurred both in the cytosol and in the vacuole of NH₄⁺ plants (Table III).

In the leaf midrib (Fig. 3A) of NH₄⁺ plants, GDH protein was present both in the cytosol and the mitochondria of the CCs, but the strongest labeling was detected inside the vacuole. In the vacuole, the variations observed between organs and treatments were not only due to the density of labeling in the vacuolar material but also dependent on the proportion of cellular material present inside the vacuole. Therefore, the data presented in Table III correspond to an average density of gold particle present in the whole vacuolar area. Although much lower than in the cytosol and the vacuole, an increase in GDH protein in the mitochondria of midribs CCs (Table III) was also observed. Similar results were obtained with a petiole section (data not shown).

Figure 3.

Immunolocalization of GDH in a leaf midrib and petiole of NH₄⁺ plants. A, CC of a leaf midrib. Large arrows indicate labeling inside the vacuole (V) and arrowheads indicate the presence of a scarce labeling area in this vacuolar material. B, Control section of a leaf midrib treated with preimmune serum. Arrowheads indicate only a few gold particles. C, Control section of a CC in a leaf midrib of NH₄⁺ plants treated with GS antibodies (arrows indicate the presence of the protein in the cytosol of CC). D, Leaf mesophyll cell treated with GS antibodies. E, Control section of a CC in the leaf midrib of NH₄⁺ plants treated with IDH antibodies. Cy, Cytosol; m, mitochondria; P, plastids; V, vacuole. Bars = 1 μm.

Only a few gold particles were visible when a similar leaf midrib section was treated with preimmune serum (Fig. 3B). Controls for the specificity of the labeling were also performed with GS and isocitrate dehydrogenase (IDH) antiserum. Figure 3D shows a partial view of a mature leaf mesophyll cell of NH₄⁺ plants after incubation with the GS antiserum. Gold particles were only present in the plastids. When minor vein or midrib sections of NH₄⁺ plants were treated with the GS antiserum, gold particles were only detected in the cytosol of the CCs (Fig. 3C). The other control using antibodies against IDH clearly shows that the protein is only present in the mitochondria (Fig. 3E). In the three controls we did not observe any labeling above the background level in the vacuolar material of the CCs (Fig. 3, B, C, and E). Moreover, the GDH antiserum appears to be highly specific since only one spot could be detected following two-dimensional western-blot analysis (Tercé-Laforgue et al., 2004).

At the base of the stem of NH₄⁺ plants large amounts of GDH protein were present both in the mitochondria and the cytosol of the CCs (Fig. 4, A and B). Some labeling was also detected in the vacuolar material of these cells (Fig. 4, C and D). However, the density of gold particles was much lower compared to that found in the vacuole of the leaf midrib CCs (see Fig. 3A and Table III for comparison). In an equivalent stem section of NL plants GDH protein was mostly present in the mitochondria of the CCs. However, compared to NH₄⁺ plants the labeling was much weaker (Fig. 4E). Quantification of gold particles in the basal stem part CCs of plants fed with NH₄⁺ confirmed a stronger induction of GDH in the cytosol compared to that seen in the mitochondria (Table III). This quantification also shows that there is more protein in the stems compared to midribs. A weak background of gold particles could be seen in the basal stem part cortical parenchyma cells of NH₄⁺ plants (Fig. 4F). In these cells, the number of gold particles was very low compared to CCs containing tissue and not significantly different in NL and NH₄⁺ plants. Moreover, their density was similar to that measured in leaf mesophyll cells (data not shown).

Figure 4.

Immunolocalization of GDH in the CCs of NH₄⁺ and NL basal tobacco stem internodes. A, CC of a basal internode of NH₄⁺ plants. B, Detail of a basal internode CC of NH₄⁺ plants. C, CC of a basal internode of NH₄⁺ plants showing the presence of labeling in the vacuolar material (arrows). D, Detail of a basal internode CC in which labeling was detected in the vacuolar material (arrows). E, CC of a basal internode of NL plants. F, Cortical parenchyma cell of NH₄⁺plants. Cy, Cytosol; m, mitochondria; P, plastids; V, vacuole. Bars = 2 μm (A), 1 μm (B–F).

A similar investigation was conducted using the top stem part of both NH₄⁺ and NL plants. A very weak labeling was only observed in the top stem CCs of NH₄⁺ plants in both the cytosol and in the mitochondria (Fig. 5A). At the top of the stem of NL plants, the labeling was mostly present in the mitochondria of CCs and was much weaker in comparison to that of NH₄⁺ plants (Fig. 5B; Table III). Compared to NL plants there was a significant increase in the number of gold particles both in the mitochondria and the cytosol of NH₄⁺ plants (Table III). However, the signal was much lower when compared to that observed in the basal part of the stem, and the vacuole was unlabeled (see Fig. 4, A and B; Table III). Immunolocalization of GDH was also conducted using leaves of Fd-GOGAT antisense plants. When Fd-GOGAT antisense plants were transferred from high CO₂ to air, some labeling was detected both in the mitochondria and in the cytosol of the CCs present in the midrib (Fig. 5C). When untransformed control plants were subjected to the same treatment, GDH protein was detected only in the mitochondria (Fig. 5D). Since in the mesophyll parenchyma cells of Fd-GOGAT antisense plants no gold particles were visible either in the cytosol or in the mitrochondria (Fig. 5E), it can be concluded that ammonium released from photorespiration is able to induce the synthesis of GDH specifically in the cytosol of the CCs.

Figure 5.

Immunolocalization of GDH in the CCs of top stem parts of NH₄⁺ and NL plants and in leaves of Fd-GOGAT antisense plants. A, CC of top stem part of NH₄⁺ plants. Arrows indicate the presence of labeling in vacuolar inclusions originating from cytoplasmic protrusions into the vacuole. B, CC of top stem part of NL plants. C, CC of a leaf of Fd-GOGAT antisense plants. D, CC of a leaf of untransformed control plants. E, Parenchyma cell of a leaf of Fd-GOGAT antisense plants. Cy, Cytosol; m, mitochondria; P, plastids; V, vacuole. Bars = 1 μm.

DISCUSSION

Previous investigations showed that high concentrations of ammonium either provided externally (Cammaerts and Jacobs, 1985; Srivastava and Singh Rana, 1987; Lea and Ireland, 1999) or released into the sieve tube from protein hydrolysis (Masclaux et al., 2000; Limami et al., 2002) generally lead to an increase in leaf GDH activity.

To our knowledge, this is the first demonstration that GDH increase occurs mainly in stems and midribs and is restricted to the CCs. Subcellular localization of the enzyme using immunogold labeling demonstrates that the amount of GDH increases mainly in the cytosol and vacuolar material and in the mitochondria of the CCs. This phenomenon occurs when ammonia is applied externally to the root system or internally released during the photorespiratory process. This conclusion was corroborated by higher amounts of ammonium translocated through the phloem stream, irrespective of its origin. The increase in both NADH and NAD-dependent activity and GDH content was proportional to the increase in ammonium content of the phloem sap and not directly correlated to the concentration of a particular amino acid.

It is well established that cycling of nitrogen molecules from xylem to phloem largely contributes to nitrogen transport via the phloem (Jeschke and Pate, 1992). Although there is no evidence that ammonium is transferred from the xylem to the phloem, a similar route is plausible for ammonium applied to a root medium. In contrast, photorespiratory ammonium that is produced in the leaf mesophyll cells is likely to be loaded in the sieve tubes and accumulated along the phloem pathway into the CCs (van Bel, 2003). Both deductions imply that ammonium itself or one of its derivates induces GDH production in the CCs. Since there is an apoplastic diffusion of ammonium within the leaf (Mattsson and Schoerring, 2002), it is likely that the extracellular flux of ammonium is not able to induce GDH in the mitochondria or in the cytosol of the mesophyll cells.

Under standard plant growth conditions, the basic level of GDH activity in the CCs, which remains confined to the mitochondria, may have a house-keeping role in a tissue that has to cope with a low oxygen concentration (van Dongen et al., 2003). Under these conditions, the lack of reducing equivalents such as NADH is in favor of the enzyme having a function in oxidizing Glu as a respiratory substrate (Freebairn and Remmert, 1957), as already demonstrated to occur in vivo (Robinson et al., 1992; Aubert et al., 2001). In ammonium-fed plants, the increase in GDH observed in the mitochondria of CCs may be related to a higher demand in reducing equivalents when inorganic nitrogen and particularly NH₄⁺ become available in large amounts. The appearance of GDH in the cytosol of CCs of NH₄⁺ plants suggests that the enzyme is able to assimilate ammonium when its concentration reaches a certain level. This hypothesis will be tested using ¹⁵NH₄⁺-labeling experiments on isolated vascular tissues.

The picture arising from the immunogold labeling studies is that the increase in the cytosolic GDH is not only due to a higher ammonium level but is also dependent of the anatomy of the organ. GDH induction was observed to be higher in basal parts of the stem, which may be explained by the fact that they are the main site for xylem-phloem exchange of nitrogen (Atkins, 2000). Consistently, we have previously shown that both the ammonium and the free amino acid content is higher in ammonium-fed plants at the base of the plant. Conversely, at the top of the stem where ammonium is loaded directly into the phloem, we observed a lower induction of GDH in both the cytosol and the mitochondria of the CCs. This observation is consistent with the finding that lower amounts of ammonium were present in the younger parts of the plant (Tercé-Laforgue et al., 2004).

Interestingly, we also found some organ specificities in the distribution of GDH protein, particularly in the leaf midrib, in which the vacuolar material was strongly labeled. It is unlikely that this labeling is an artifact since two positive controls performed either with GS antibodies or IDH antibodies and a negative control performed with preimmune serum confirmed the specificity of the labeling in the cytosol (GS) and in the mitochondria (IDH) of CCs but not in the vacuole. In addition, other experiments showed that the vacuolar labeling was specific to the CCs in the leaf midrib because little vacuolar labeling was observed in other cell types and CCs in minor veins, top stem internodes, and mesophyll cells.

The high levels of GDH in the vacuole remain enigmatic, although GDH has been found previously localized inside multivesicular bodies associated with vacuolar autophagic activity. This lytic activity of the vacuole already reported in the senescing flowers of grapevine (Paczek et al., 2002) and possibly in senescing leaves (Dubois et al., 2003) is the result of carbon starvation (Matile and Winkenbach, 1971; Robinson et al., 1992; Aubert et al., 1996). These results suggest that GDH is sensitive to the C status of the plant via the vascular tissue if we consider the well-established fact that the expression of GDH is repressed when the cellular level of sugars is high (Turano et al., 1997, Masclaux-Daubresse et al., 2002).

In conclusion, our results suggest that GDH plays a dual role in CCs, either in the mitochondria when mineral nitrogen availability is low or in the cytosol when ammonium concentration increases above a certain threshold. An attractive hypothesis is that GDH, when induced by ammonium in the cytosol, may act as a sensor to evaluate the carbon/nitrogen status of the plant particularly with respect to ammonium and sugar concentration and/or fluxes through the phloem stream. A putative sensing role for GDH is in keeping with the current view that the continuity between CCs and sieve tubes is one of the key elements for metabolite translocation and signaling during plant growth and development (van Bel, 2003; van Bel et al., 2003). Currently, further experiments are being carried out to examine the potential regulatory role of GDH.

MATERIALS AND METHODS

Plant Material and Growth

Tobacco (Nicotiana tabacum) cv xanthi XHFD8 (Institut National de la Recherche Agronomique [INRA], Versailles, France) was grown on coarse sand. From the bottom of the seedlings, each emerging leaf was numbered and tagged. From a batch of 6-week-old plants, 12 plants of uniform development and numbering 7 leaves were selected. These were transferred to a controlled environment growth chamber (16 h light, 350–400 μmol photons m⁻² s⁻¹, 25°C; 8 h dark, 18°C) and watered with either a NH₄⁺ solution (5 mm NH₄⁺) or an NL solution (0.1 mm NO₃⁻). For both NH₄⁺ and NL solutions the basic mixture contained 1.25 mm K⁺, 0.25 mm Ca²⁺, 0.25 mm Mg²⁺, 1.25 mm H₂PO₄⁻, 0.75 mm SO₄²⁻, 21.5 μm Fe²⁺ (Sequestrene; Syngenta Agro S.A.S, Rueil Malmaison, France), 23 μm B³⁺, 9 μm Mn²⁺, 0.3 μm Mo²⁺, 0.95 μm Cu²⁺, and 3.5 μm Zn²⁺. For the NH₄⁺ solution ammonium was supplied as 1 mm (NH₄)₂SO₄ plus 3 mm NH₄Cl, and for the NL solution NO₃⁻ was supplied as 0.1 mm KNO₃. Plants were automatically watered for 1 min (flow rate for each plant, 50 mL min⁻¹) every 2 h. Four plants were used for each nitrogen feeding condition. Four weeks after sowing, leaves were numbered 8, 9, 11, 13, 15, and 20 (from bottom to top) for both NH₄⁺ and NL plants as described by Tercé-Laforgue et al. (2004). Leaf 10 that corresponds to an old source leaf (in which GDH induction is maximum) and leaf 18 that corresponds to a young sink leaf were used for both biochemical and cytoimmunochemical experiments. The stem sections were collected between leaf 16 and 20 for the top and between leaf 9 and 12 for the bottom. Stem, midrib, and mesophyll tissue (in which the midrib and small veins were removed) were harvested and pooled in two groups. One was weighed and then lyophilized to determine fresh and dry weights. The other was weighed, frozen, and used to determine both the quantity of protein and GDH activity. These tissues were frozen in liquid nitrogen and immediately reduced to a homogenous powder that was stored at −80°C and used for all the further experiments. All the harvesting of fresh material was done between 1 and 5 pm.

The growth and culture of transformed and untransformed tobacco plants were similar to those described in Ferrario-Méry et al. (2000). Plants were grown in pots in a growth chamber with an atmosphere enriched in CO₂ (4,000 μL L⁻¹) to diminish photorespiration and to allow plants with greatly decreased Fd-GOGAT activities to survive. The plants were watered 3 times/h with 25 mL of a complete nutrient solution containing 10 mm NO₃⁻ and 2 mm NH₄⁺ (Coic and Lesaint, 1971). Under these growing conditions the accumulation of ammonium in the leaf was solely due to photorespiration and not the ammonium present in the nutrient solution. (Ferrario-Méry et al., 2000). The plants were provided with 300 to 400 μmol photons m⁻² s⁻¹ irradiance and a 16-h-light (25°C), 8-h-dark (18°C) cycle. Two-month-old plants with 20 leaves were used for the following experiments. Line 15 was used in all subsequent experiments (Ferrario-Méry et al., 2001). A young mature leaf (leaf 18) was harvested in the middle of the photoperiod from 3 plants of line 15 (L15) and untransformed control (high CO₂). The atmospheric CO₂ content was then decreased from 4,000 μL L⁻¹ to 350 μL L⁻¹. Again the first youngest mature leaf was harvested from 3 other plants of line 15 and untransformed control 3 d after transfer.

Metabolite Extraction and Analyses

Lyophilized plant material was used for metabolite extraction. Free ammonium was extracted with 2% (w/v) 5-sulfosalicylic acid (1 mL/10 mg dry weight [DW]) as described by Ferrario-Méry et al., (1998) and its concentration determined by the phenol hypochlorite assay (Berthelot reaction). Total amino acids were analyzed by the Rosen colorimetric method (Rosen, 1957). Free amino acid composition of phloem exudates was determined by ion-exchange chromatography (Rochat and Boutin, 1989) following pH adjustment to 2.1 with 0.1 n HCl. Suc, Glc, Fru, and starch were extracted with 1 m HClO₄ (1 mL/5–10 mg DW of plant material) as described by Ferrario-Méry et al. (1998). The soluble sugars (Glc, Fru, and Suc) were measured enzymatically using a commercially available kit assay (Boehringer Mannheim, Mannheim, Germany). Starch content was determined as described by Ferrario-Méry et al. (1998).

Phloem Sap Collection

Phloem exudates were collected from leaf 18 of NH₄⁺, NL, and Fd-GOGAT antisense plants grown as described above. Phloem exudates were obtained using the technique described by King and Zeevaart (1974). The leaves were cut off and petioles were recut under water before rapid immersion in the collection buffer. For each experiment, petioles of fully expanded leaves were placed separately in a solution of 10 mm HEPES, 10 mm EDTA (adjusted to pH 7.5 with NaOH) in a humid chamber (relative humidity >90%) and in the dark. Exudates were collected during 6 h from 10 am to 4 pm and stored at −80°C. Phloem exudates (in the EDTA solution) were adjusted to pH 2.1 and centrifuged to remove debris and EDTA that precipitate at that pH.

Enzymatic Assay and Determination of Total Soluble Protein

Enzymes were extracted from frozen leaf material stored at −80°C. All extractions were performed at 4°C. GDH was measured as described by Turano et al. (1996) except that the extraction buffer was the same as for GS (Tercé-Laforgue et al., 2004). Soluble protein was determined using a commercially available kit (Coomassie Protein assay reagent, Bio-Rad, Munich, Germany) using bovine serum albumin (BSA) as a standard.

Statistics

For metabolite analyses and measurement of enzyme activities, results are presented as mean values for four plants with ses (se = sd/, where sd is the standard deviation and n the number of replicates).

Gel Electrophoresis and Protein-Blot Analysis

Proteins were extracted from frozen leaf material in cold extraction buffer containing 50 mm Tris-HCl, pH 7.5, 1 mm EDTA, 1 mm MgCl₂, 0.5% (w/v) polyvinylpyrrolidone, 0.1% (v/v) β-mercaptoethanol and 4 μm leupeptine, and separated by SDS-PAGE (Laemmli, 1970). Equal amounts of protein (equivalent to 0.5 mg DW) were loaded in each track. The percentage of polyacrylamide in the running gels was 10%.

Proteins were scanned on the western-blot membranes with a Power Look II scanner (UMAX Data Systems, Taipei, Taiwan) and quantified with the software the public domain NIH Image program (developed at the United States National Institutes of Health and available on the Internet at http://rsb.info.nih.gov/nih-image).

Cytoimmunochemical Studies

Leaf fragments (2–3 mm²), mesophyll, midribs, or stems were fixed in freshly prepared 1.5% (w/v) paraformaldehyde in phosphate buffer 0.1 m, pH 7.4, for 4 h at 4°C. For immunolocalization, material was dehydrated in an ethanol series (final concentration 90% [v/v] ethanol) then embedded in London Resin white resin (Polysciences, Warrington, PA). Polymerization was carried out in gelatin capsules during 10 h at 54°C. For immunotransmission electron microscopy studies, ultra thin sections were mounted on 400-μm mesh nickel grids and allowed to dry at 37°C. Sections were first incubated with 5% (v/v) normal goat serum in T1 buffer (0.05 m Tris-HCl buffer containing 2.5% [w/v] NaCl, 0.1% [w/v] BSA, and 0.05% [v/v] Tween 20, pH 7.4) for 1 h at room temperature then with anti-GDH rabbit serum (Loulakakis and Roubelakis-Angelakis, 1990) diluted 70 times in T1 buffer for 6 h at room temperature. Sections were then washed 5 times with T1 buffer, 2 times with T2 buffer (0.02 m Tris-HCl buffer containing 2% [w/v] NaCl, 0.1% [w/v] BSA, and 0.05% [v/v] Tween 20, pH 8) and incubated for 2 h at room temperature with 10 nm colloidal gold-goat anti-rabbit immunoglobulin complex (Sigma, St. Louis) diluted 50 times in T2 buffer. After several washes, grids were treated with 5% (w/v) uranyl acetate in water and observed with a Philips CM12 electron microscope (Philips, Eindhoven, The Netherlands) at 80 kV.

Polyclonal antiserum raised against tobacco GS2 (Hirel et al., 1984) and tobacco IDH (Lancien et al., 1998) were used as controls for cytoimmunochemistry. For both techniques, additional controls were conducted either by omitting the primary antibody or by substituting it with normal rabbit serum.