SEARCH

SEARCH BY CITATION

Keywords:

  • ATP imaging;
  • energy status;
  • photosynthesis;
  • metabolic gradient;
  • cell differentiation;
  • seed development

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

To analyse the energy status of Vicia faba embryos in relation to differentiation processes, we measured ATP concentrations directly in cryosections using a quantitative bioluminescence-based imaging technique. This method provides a quantitative picture of the ATP distribution close to the in vivo situation. ATP concentrations were always highest within the axis. In pre-storage cotyledons, the level was low, but it increased strongly in the course of further development, starting from the abaxial region of cotyledons and moving towards the interior. Greening pattern, chlorophyll distribution and photosynthetic O2 production within embryos temporally and spatially corresponded to the ATP distribution, implicating that the overall increase of the energy state is associated to the greening process. ATP patterns were associated to the photosynthetic capacity of the embryo. The general distribution pattern as well as the steady state levels of ATP were developmentally regulated and did not change upon dark/light conditions. The major storage protein legumin started to accumulate in abaxial regions with high ATP, whereas starch localized in regions with relatively lower ATP levels. This suggests a role of the energy state in the partitioning of assimilates into the different storage-product classes. Highest biosynthetic rates occurred when cotyledons became fully green and contained high ATP levels, implicating that a photoheterotrophic state was required to ensure high fluxes. Based on these data, we propose a model for the role of embryonic photosynthesis to improve the energy status of the embryo.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Embryos of faba bean (Vicia faba) differentiate from a meristem-like tissue into a highly specialized storage organ. The process has been divided into seven stages (Borisjuk et al., 1995). Organogenesis and morphogenesis are completed at stage III. Stages IV–VII describe early-to-late cotyledon development. At stage V, differentiation starts gradually from the inner adaxial face progressing towards the outer region, thus forming developmental gradients across the cotyledonary cell population. Starch accumulation relates positively to cell expansion, but is negatively correlated to mitotic activity (Borisjuk et al., 1998). How these developmental patterns are coordinated is unclear, but there is evidence for a metabolic regulation (Wobus and Weber, 1999) and the involvement of sugars (Borisjuk et al., 2002). A switch in the presence of the principal sugars from hexoses to sucrose coincides with cotyledonary differentiation (Weber et al., 1995, 1997).

The energy state of a given tissue is an indicator of the overall metabolic condition and can be described by the adenylate energy charge (AEC), which refers to the relative saturation of the adenylate pool in phosphoanhydrid bonds (Pradet and Raymond, 1983). The AEC is an important physiological parameter regulating metabolic activity. AEC is directly correlated to the ATP pool by the activity of adenylate kinase. Therefore, the ATP pool can serve as an indicator of the energy state. During seed development, levels of ATP and other nucleotides change characteristically as shown for rape seed (Ching et al., 1974), soybean (Quebedeaux, 1981) and faba bean (Rolletschek et al., 2002). ATP levels and AEC are especially low in very young embryos. However, with the onset of the storage phase, both ATP levels and AEC are maintained at a higher and relatively constant value (Rolletschek et al., 2003). Varying ATP levels are caused by changes in both production and demand. Indeed, respiratory activity of pea embryos was shown to vary considerably during development (Harvey et al., 1976; Kollöffel and Matthews, 1983). Energy demand also changes between distinct growth phases like the meristematic growth at early stages and the subsequent seed-filling phase with high biosynthetic activity. However, a regulatory function of the overall energy state or pool sizes of nucleotides in legume seeds was so far not regarded in detail in growing legume embryos.

Growing embryos are predominantly heterotrophic. In order to perform biosynthesis, the plastids in plant storage organs must be supplied with substrates and energy (Neuhaus and Emes, 2000). There is evidence that storage product biosynthesis is energy-limited. ATP transport into the amyloplasts is a limiting step for starch synthesis in potato tubers (Tjaden et al., 1998). Unlike tubers, legume embryos become green and photosynthetically active (Eastmond et al., 1996; Smith et al., 1990). In isolated soybean plastids, light influences the partitioning of assimilates into different storage products (Willms et al., 1999). However, embryonic photosynthesis is of minor importance in terms of carbon budgets (Atkins and Flinn, 1978; Eastmond et al., 1996; Flinn and Pate, 1970; Harvey et al., 1976), but can play a role to provide ATP and NADPH for storage product synthesis (Browse and Slack, 1985). The importance of photosynthetically produced energy equivalents is controversially discussed, but seems to be lower when compared to mitochondrial respiration (Eastmond et al., 1996).

We assumed that in growing legume embryos, the energy state changes in relation to differentiation processes. However, appropriate methods to measure ATP distribution within tissues on a spatial level were not available so far. Existing methods required tissue homogenization and/or extraction procedures. Thereby, always a mixture of cell types, which differ in respect to physiological and differentiation states, is analysed. We recently used a method for high-resolution mapping of glucose and sucrose in tissue sections by quantitative bioluminescence and single photon imaging with a spatial resolution near the single cell level (Borisjuk et al., 1998, 2002). In this study, we performed ATP imaging on sections of faba bean embryos. The aim was to perform a spatial and temporal analysis of ATP concentrations in relation to developmental and physiological parameters. We found that the patterns of ATP distribution changed in a specific temporal and spatial manner during differentiation. Higher levels were related to the greening pattern and photosynthetic capacity of the embryo. Based on these data, we propose a model for the role of photosynthesis related to respiration. We have already provided experimental evidence for this model by investigating the effects of light and dark conditions on the tissue levels of oxygen and metabolites and on metabolic activity (Rolletschek et al., 2003).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Determination of local ATP concentrations using bioluminescence-based imaging

For quantitative estimations of ATP distributions within the embryo, we applied a metabolite imaging method. This technique is based on bioluminescence reactions in cryosections using firefly luciferase, where the intensity of the light emission is proportional to the ATP concentration. The emission of single photons is registered in real time and integrated by a photon-counting camera mounted on a microscope (Figure 1a). Finally, a two-dimensional map is obtained showing the spatial ATP concentrations within the cryosection. An analogous technique has been applied to measure ATP in the millimolar concentration range in animal tissues (Levin et al., 1999). However, for plant embryos, the method had to be adapted to measurements in the nanomolar range using tissue standards with known concentrations of ATP. Combining the calibration procedure and the imaging photon counting allowed to calculate the regional ATP levels in absolute concentrations within tissue sections presented as a colour code (Figure 1b). Sensitivity and resolution of the technique depend on the intensity of light emission, the thickness of sections and the temperature as well as the microscopic magnification and the sensitivity of the photon-counting unit (Mueller-Klieser and Walenta, 1993).

image

Figure 1. Measuring local ATP concentrations in tissue sections using bioluminescence.

(a) Experimental set-up used for bioluminescence imaging of ATP within cryosections. The reaction module includes the tissue cryosection attached to the cover glass and a glass chamber filled with the enzymatic solution.

(b) ATP calibration curve (see Experimental procedures).

Download figure to PowerPoint

Mean ATP levels in V. faba embryos have also been measured by conventional HPLC technique. These data were plotted against those obtained from the bioluminescence method (Figure 2a). There is quite a good correlation between the values derived from both methods (Spearman rank correlation coefficient 0.85; P < 0.01), demonstrating the reliability of the imaging method. To analyse whether ATP levels of cotyledon cells reflect the adenylate energy, charge values were plotted against each other (Figure 2b). Up to about 200 nmol g−1, the ATP levels were linearly quite well correlated to the AEC; at higher values, the AEC did not further increase (Spearman rank correlation coefficient 0.75; P < 0.01). This means that regions containing low ATP contents are generally energy-limited. At higher ATP contents, the AEC remains constantly high, indicating an increase of all adenylate nucleotides. For comparison, in Figure 2(c), the ATP levels measured by HPLC are also shown in embryos of different stages along with the AEC values (data are adapted from Rolletschek et al., 2003).

image

Figure 2. ATP and AEC levels in growing V. faba embryos

(a) Mean ATP levels in V. faba embryos measured by HPLC plotted against the values obtained by bioluminescence.

(b) Mean ATP levels in V. faba embryos measured by HPLC plotted against AEC values calculated as described by Rolletschek et al. (2002).

(c) ATP levels measured by HPLC and AEC values in embryos of different stages. The ATP levels measured by bioluminescence were calculated on the basis of average grey values of randomly chosen images used for ATP mapping (Borisjuk et al., 2002). AEC values were calculated as described by Rolletschek et al. (2002) from data adapted from Rolletschek et al. (2003).

Download figure to PowerPoint

The ATP distribution pattern changes during development

The ATP distribution was determined within sections of whole embryos covering the cotyledon stages IV–VII (Borisjuk et al., 1995). The spatial distribution of ATP is shown in the top panel of Figure 3, presented as a colour code within the tissue section directly used for the measurements (lower panel). In pre-storage-phase embryos of stage IV (approximately 50 mg seed FW), ATP levels were highest in the axis (approximately 400 µm, arrowhead), whereas in the cotyledon levels were much lower (approximately 100 µm). This stage of development is characterized by overall mitotic activity with highest frequencies in the axis. For stage V embryos (approximately 150 mg seed weight), the ATP distribution is shown within a longitudinal section through the axis and cotyledons (left side) and within a transverse section through the cotyledons only (right side). Again, highest ATP levels were present within the axis (arrowhead). In the cotyledons, a gradient was detected from the outer abaxial to the inner adaxial region. Highest concentrations (approximately 300 µm) were detected within a tissue layer covering the out-warded part of the cotyledons, whereas in the inner region, the levels were lower than 50 µm. During this stage, mitotic activity decreased within the inner regions of the cotyledons and cells started to elongate (Borisjuk et al., 1995). These events led to a heterogeneous cell population first recognizable at this stage. Stage VI embryos, which already entered the storage phase (approximately 300 mg seed FW), were characterized by generally higher ATP levels with higher contents within the outer abaxial region (approximately 300 µm) compared to the inner side (approximately 200 µm). In mid-storage-phase embryos (stages VI and VII, approximately 400 mg), ATP was generally high in the whole cotyledonary tissue (approximately 300 µm) without gradients, whereas again, slightly increased levels were present in the axis (arrowhead). At the end of the storage phase (stage VII, approximately 800 mg), the overall ATP concentration was high (approximately 400 µm). However, the inner region, where storage processes had already been terminated, contained very low levels of only approximately 50 µm ATP.

image

Figure 3. ATP pattern during embryogenesis of V. faba.

ATP was measured by bioluminescence (upper panel) within cryosections of embryos at different stages (lower panel, contours of sections drawn by dotted lines). Local concentrations of ATP are shown in colour code. Bars = 2 mm.

Download figure to PowerPoint

In summary, using a novel metabolite imaging technique, we analysed the ATP distributions within the Vicia embryos. ATP concentrations were always high within the axis. Within growing cotyledons, the ATP concentrations changed temporally and spatially. In pre-storage cotyledons, ATP levels were very low but increased in the course of differentiation. This increase proceeded in a distinct spatial pattern starting from the outer (abaxial) to the inner (adaxial) region. At the end of maturation, ATP dramatically dropped within regions where storage activity had stopped.

The ATP concentration pattern corresponds to the greening pattern

Vicia embryos undergo a greening process during development. Microscopical studies under UV light were used to analyse this process by detecting chlorophyll fluorescence. In stage IV embryos, a fluorescence signal was observed mainly in the axis with much lower intensities in the cotyledons (Figure 4a). Within stage V embryos, a gradient of chlorophyll fluorescence was visible with higher intensities in the outer region (Figure 4b). At mid-storage phase, a strong signal was detected within the whole cotyledonary parenchyma (Figure 4c). Because the fluorescence signal revealed a gradient in chlorophyll distribution within cotyledons of stage VI embryos, we analysed the chlorophyll content in stage V embryos separated into axis, adaxial and abaxial regions after extraction and subsequent spectrophotometric determination. Chlorophyll content was about twofold higher in the outer abaxial region of cotyledon and axis as compared to its inner parts. The chlorophyll a:b ratio was found to be around 2 (Figure 4d,e).

image

Figure 4. Chlorophyll distribution in V. faba embryo determined as chlorophyll fluorescence within tissue sections under UV light (a–c).

(a) Fluorescence covers the embryo axis and the closely attached adaxial parts of cotyledons at stage IV.

(b) At stage V, higher fluorescence intensity is present in the outer region of a cotyledon.

(c) During stage VI, the fluorescence signal covers the whole cotyledon without any visible gradient.

(d) Analysis of chlorophylls a and b in tissue homogenates isolated from dissected embryos at early stage VI as schematically shown in (e). Bars = 0.5 mm.

Download figure to PowerPoint

These results show that the greening process in Vicia embryos, measured as chlorophyll fluorescence, becomes first obvious in the axis. Within cotyledons, the greening process spreads from the outer abaxial to the adaxial region. The pattern temporally and spatially corresponds to that of the ATP concentrations.

Highest ATP levels are present in photosynthetically active tissue

To analyse photosynthetic activity related to the chlorophyll gradients, we measured photosynthesis directly within the cotyledons under in vivo conditions. An O2-sensitive microsensor was inserted into the embryo of intact stage V seeds to different depths from the outer (abaxial) surface towards the interior (along the α-axis; Figure 5). Photosynthetic O2 production was highest directly underneath the abaxial surface (approximately 38 nmol O2 g−1 min−1), but dropped about twofold within approximately 0.5 mm towards the interior (Figure 5a). A similar pattern was observed for embryos at stage VI and, with a less steep gradient, at stage VII (data not shown). The spatial pattern of photosynthesis corresponded to the chlorophyll distribution visualized by confocal microscopy. The highest chlorophyll level (fluorescent signal, respectively) was detected in the abaxial region (Figure 5b). ATP content was highest directly underneath the outer surface (approximately 400 µm) and decreased to about 100 µm at 1 mm depth (Figure 5c). The graph represents the ATP concentration profile along the α-axis. Thus, spatial gradients of ATP concentration, chlorophyll distribution and photosynthetic activity were found to be correlated within the embryo.

image

Figure 5. Spatial assessment of photosynthetic activity, chlorophyll content and ATP concentration in cotyledons at stage VI.

(a) Photosynthetic oxygen production measured by a microsensor (y-scale) at different depths towards the interior (x-scale along α).

(b) Confocal imaging of chlorophyll distribution within the tissue (the same as in (a)) measured by excitation and emission.

(c) Local concentration of ATP within the same tissue region, represented as a colour-code map and as a graph along the α-axis (see (a, b)).

Download figure to PowerPoint

ATP distribution pattern and pool size do not change during day/night

Our results suggest that the actual photosynthetic activity affects the ATP pattern. Therefore, ATP was measured in cryosections of embryos harvested during day and night. Neither the distribution pattern nor levels of ATP were different (Figure 6). This was obvious for stages IV–VI analysed, indicating that the general distribution pattern as well as the steady state levels of ATP did not change upon day/night, but were developmentally controlled.

image

Figure 6. ATP patterns within the embryo during day and night.

Local ATP levels were measured in corresponding tissue sections from embryos at stages IV (a, b), V (c, d) and end of stage VI (e, f) at day and night. ATP concentrations are given in the same colour code as in Figure 1. Bars = 1 mm.

Download figure to PowerPoint

Constant levels of ATP do not necessarily indicate constant metabolic fluxes. To get a comprehensive view of the metabolic situation in the embryo requires investigation of metabolic fluxes and their potential diurnal changes. This is not possible using the bioluminescence imaging technique, but was performed by isotope partitioning experiments reported by Rolletschek et al. (2003).

ATP patterns are related to respiration and tissue differentiation

To study the regulation of the ATP level, we analysed respiration and parameters of embryo differentiation: mitotic and cell elongation index and starch accumulation rates (Figure 7). Respiration was measured as O2 uptake by isolated embryos. It was highest in very young embryos of stage IV, with actively proliferating cells and high mitotic indices (Figure 7a), and with low rates of cell expansion and no detectable starch accumulation (Figure 7b). Respiration decreased during stages V and VI. The latter represents the main storage phase characterized by low mitotic activity, as well as high cell expansion and starch accumulation rates. Stage VII embryos were fully differentiated with no mitotic activity but high cell elongation and starch synthesis. Respiration was nearly constant but at a rate of only 50% of that of the early stages. Thus, respiration decreased in the course of differentiation. Figure 7(c) shows schematically the accumulation of starch and total nitrogen during embryo growth.

image

Figure 7. Physiological and biochemical characteristics of developing embryos.

(a) Mitotic index (black bars for adaxial and grey bars for abaxial regions of cotyledons) and respiration during seed growth.

(b) Cell expansion index (bar) and starch accumulation rates (graph).

(c) Schematic presentation of starch and nitrogen accumulation in growing V. faba embryos. Contents of starch and protein in mature dry seeds of V. faba are around 40–45% and 28–32%, respectively.

Download figure to PowerPoint

The temporal sequence of differentiation events during stages V and VI was reflected in spatial developmental gradients within the cotyledons. Mitotic activity was lower in the adaxial region and cell elongation was higher as compared to the outer abaxial region (Figure 7a). The number and size of accumulated starch grains were higher in the adaxial region. Therefore, the cells localized at the abaxial (outer) region were developmentally younger than the adaxial cells. It was not possible to measure respiration on a spatial level. But, because respiration decreases in the course of differentiation, a gradient in respiration can be assumed with higher rates in abaxially, developmentally younger regions. Figure 3 shows that during stages V and Vl, the highest ATP concentration covered the developmentally younger tissues that were located outside. Thus, high ATP is obviously associated with high respiratory activity. Therefore, regions with both highest respiratory and photosynthetic activity possess highest ATP concentrations.

The spatial patterns of starch and protein accumulation are inversely oriented

We showed that the distribution of ATP within growing V. faba embryos was tightly regulated in a specific temporal and spatial manner. During stages V and VI, the local concentrations were arranged in gradients with higher levels within the outer abaxial region and decreasing towards the inner adaxial side (Figure 3). Embryos of V. faba started to accumulate starch and storage proteins at approximately the same time. We therefore compared the spatial accumulation patterns of starch and legumin storage protein as well as transcripts in relation to the described ATP distribution pattern. Using sections of stage V embryos (a toluidine blue stain is shown in Figure 8a), the expression levels of legumin-mRNA as well as ADP-Glc pyrophosphorylase (AGP)-mRNA were analysed by in situ hybridization. Legumin and starch were analysed by immuno- and iodine staining, respectively. Legumin-mRNA accumulation was detected in both the outer abaxial and the inner adaxial regions (Figure 8b). Legumin protein was present mainly within the outer epidermal cells and within the parenchyma cell rows directly underlying the epidermis, whereas the more inner cells of the parenchyma were only weakly labelled (shown as a blue-coloured labelling, Figure 8c,f). In contrast, a higher number of large starch grains were detected within these inner cells (Figure 8d,g) compared to the outer ones. Accordingly, AGP gene expression, which is a marker for starch synthesis, started from the inner adaxial regions (Figure 8e). In mid-cotyledon-stage (VI) embryos, these inversely oriented patterns of legumin protein and starch accumulation were even more evident. Starch content was higher within the adaxial region (Figure 8i,k), whereas legumin accumulation was predominantly detected within the outer abaxial area (Figure 8h,j).

image

Figure 8. Spatial distribution of storage processes within the embryo during developmental stages V (a–g) and VI (h–k).

(a) Toluidine blue staining of tissue sections at stage V.

(b) In situ localization of a legumin transcript (white labelling).

(c) Immunostaining of legumin (dark blue labelling).

(d) Distribution of starch by iodine staining (dark brown stain).

(e) In situ localization of ADP-Glc pyrophosphorylase-mRNA (white labelling).

(f) Immunolabelled protein bodies in cotyledonary epidermis enlarged from (c). Shown by arrowheads.

(g) Starch granules enlarged from (d). Shown by arrowheads.

(h) Abaxial localization of legumin protein shown by immunostaining.

(i) Gradient of starch accumulation.

(j) Immunolabelled protein bodies in epidermis and parenchyma cells, enlarged from (h). Shown by arrowheads.

(k) Distribution of starch grains enlarged from (i).

Bars: 0.5 mm (a–e); 40 µm (f, g, j, k), 1 mm (h, i).

Download figure to PowerPoint

The results show that the pattern of starch accumulation at stage V was inversely oriented to that of legumin. Moreover, legumin protein accumulation was initiated in regions enriched in ATP, whereas starch was detected first in regions with lower ATP levels.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We present here a method to quantitatively measure ATP directly in cryosections based on bioluminescence imaging. Thereby, it is possible to analyse the energy state within a heterogeneous tissue as growing seeds. Bioluminescence-based metabolic imaging provides information close to the in vivo situation and represents a most suitable tool to assess the ATP status within embryos. Such results can be combined with other tissue characteristics obtained by in situ methods like histo- and immunostaining and in situ hybridization, and allows analysing the physiological conditions of the tissue. We show here that the pattern of ATP distribution changes in a specific temporal and spatial manner and is related to embryo differentiation.

Local ATP concentrations are correlated to the greening process of the embryo

In stage IV embryos, ATP was found to be approximately fourfold higher within the axis compared to cotyledons. The generally higher level in the axis were evident throughout further development and may be necessary to provide this highly mitotically active tissue with energy. Within the undifferentiated cotyledons, ATP concentrations were low. At stage V, ATP levels increased resulting in approximately fivefold higher overall values when the embryo entered the storage phase at stage VI, as measured in tissue homogenates (Rolletschek et al., 2003). The rise occurred as a distinct pattern starting from the abaxial face, thereby forming a gradient with a four- to fivefold concentration difference towards the adaxial side. This gradient disappeared towards the later stage VI, and ATP became uniformly distributed at high levels all over the storage parenchyma. ATP levels remained high throughout the storage phase. At the end of maturation, the concentration decreased up to 10-fold in regions where the storage activity had already been terminated. Obviously, because of decreased overall metabolic activity, development switches from storage to desiccation (Kollöffel and Matthews, 1983). A similar pattern of ATP, but only on a temporal scale, has been reported for embryos of oilseed rape (Ching et al., 1974) and soya bean (Quebedeaux, 1981). Therefore, highest AEC and ATP values are temporally associated with embryo maturation.

Using the imaging technique, we showed that the ATP distribution pattern during stages V–VII was in opposite orientation to that of cell differentiation. Whereas the ATP gradient decreased from abaxial to adaxial regions, cell differentiation was initiated within the inner adaxial face characterized by lower mitotic activity and higher cell elongation.

We found an apparent correlation between ATP levels and the greening process. Similar to the rise of ATP, greening started in the axis at stage IV, then proceeded within the cotyledons from outer to inner sides. Gradients in chlorophyll concentration as well as for ATP were present at stage V, and both disappeared thereafter. Accordingly, photosynthetic activity measured within a transect revealed a similar gradient along with ATP concentrations. We therefore conclude that the local increase of ATP within differentiating Vicia embryos is associated with the greening process. Both events are in opposite orientation to the cell differentiation gradient. The higher ATP content within green tissues is not simply an effect of photosynthetic ATP production because after 4 h of darkness, the ATP concentration did not change (Figure 6). However, oxygen immediately responded to light and therefore, unlike ATP, is directly related to photosynthesis (Rolletschek et al., 2003). This suggests that the local ATP distribution is developmentally regulated and coupled to the photosynthetic capacity, but is not influenced by the actual photosynthetic activity.

The bioluminescence-based imaging method presented here is, at the moment, not sensitive enough to allow resolution at the subcellular level. Such compartment-specific distributions and concentrations of adenylates were calculated for potato tuber cells based on a non-aqueous fractionation method (Farréet al., 2001). Most of the ATP and ADP were associated with plastids. The plastidial ADP:ATP ratio was higher than the cytosolic ratio. This situation is similar in spinach leaves (Heineke et al., 1991). Assuming an analogous distribution in Vicia cotyledons, the general increase of ATP is probably associated to processes on the level of plastids.

During greening, the embryo acquires a higher energy state

Seeds of Vicia and other legumes possess morphological and biochemical characteristics that potentially promote oxygen limitation (Rolletschek et al., 2002). Hypoxic conditions predominantly occur during stages IV and V before embryos become green, as evidenced by low levels of oxygen, AEC and ATP, as well as high fermentation activity shown by ethanol emission rates and upregulation of alcohol dehydrogenase (Rolletschek et al., 2003). During the process of greening (stage V), the embryos overcome hypoxia, which is shown by increasing AEC and ATP, whereas fermentative activity disappears. From approximately 200 mg seed FW onwards, Vicia embryos become fully green and can be regarded as being no longer hypoxic.

Overcoming hypoxia represents an adaptation process and is coupled to developmental and metabolic changes. The gain of photosynthetic capacity enables the embryo to acquire a higher energy state. We assume that one important event is the differentiation of plastids. Within green embryos, a specific type of plastid, which has been termed as photoheterotrophic, is present. This differs from chloroplasts in leaves by both morphological and physiological aspects (Asokanthan et al., 1997). Photoheterotrophic plastids import carbon, mainly Glc-6-P, phosphoenol pyruvate (PEP) and pyruvate, via specific translocators (Eastmond and Rawsthorne, 1998; Rawsthorne, 2002). In addition, photoheterotrophic plastids are also photosynthetically active. Their capacity for photosynthetic CO2 fixation is low as has been reported for embryos of rape (Asokanthan et al., 1997; Eastmond et al., 1996), pea (Flinn, 1985) and soybean (Saito et al., 1989). However, photoheterotrophic plastids have significant electron transport activity, a high chlorophyll a:b ratio and abundant proteins associated to photosystem II (Asokanthan et al., 1997). Rubisco in pea embryos is near the detection limit and several magnitudes lower than in the pod or leaves (Hedley et al., 1975). The main CO2 re-fixing enzyme here is most probably PEP-carboxylase (Golombek et al., 1999). On the other hand, canola seeds obviously follow another strategy with much higher activities of Rubisco in the growing embryo (King et al., 1998).

Here, we present a model for the photoheterotrophic metabolism in embryos (Figure 9). It implies that the main role of embryonic photosynthesis is to produce both energy-rich compounds and oxygen. Energy-rich molecules from photosynthesis can be directly used for biosynthesis. Oxygen production is especially important for embryogenesis when oxygen is limited because the heterotrophic embryo is mainly dependent on mitochondrial respiration. Upon illumination, oxygen produced in green embryonic tissue supports mitochondrial respiration and thereby enriches the cytoplasmic ATP pool (Figure 5). Thus, the greening of embryos increases their overall energy state, and light supply may therefore lead to increased biosynthetic fluxes. The fact that ATP patterns are not affected by photosynthetic activity seems to contradict our model, and we are not yet able to estimate the contribution of photosynthetic oxygen and ATP production. However, we have to differentiate between the steady state pool sizes of ATP and the flux through this pool. Recently, we showed that metabolic fluxes are increased upon light conditions without affecting ATP levels (Rolletschek et al., 2003). Changing metabolic fluxes at stable ATP pools requires a tight coupling between ATP supply and demand. This can be achieved by co-localization of high respiration and photosynthesis rates within the same regions of greening embryos.

image

Figure 9. A model of the photoheterotrophic metabolism in legume embryos.

ATP is mainly supplied via mitochondrial respiration. Under hypoxia, the oxygen limits respiration and thereby ATP supply (dotted arrows). Photosynthetically active plastids in green tissues produce oxygen (and ATP) supporting respiration and overall metabolic activity (green arrows).

Download figure to PowerPoint

Embryonic energy state and its implications for storage product synthesis

The increase of the energy state in cotyledons is coupled to the greening process. Both events start during stage V and proceed from the abaxial to the adaxial face. Storage activity in Vicia embryos also starts at this stage, but highest rates were achieved only during stage VI when the embryos became fully green and reached their highest ATP content. This suggests that the photoheterotrophic state of plastids is important to realize high fluxes into storage products. Starch and protein accumulation in Vicia embryos is very similar on a temporal scale (own unpublished results); however, the spatial patterns are markedly different and opposed to each other. Legumin-mRNA started to accumulate in the abaxial and adaxial region face (Figure 8b), whereas legumin protein, which is photosynthetically active and is characterized by high-energy state, is accumulated only in the outer region. On the other hand, starch grains, together with AGP-mRNA, first appeared within the adaxial regions with a lower energy state. This may implicate that the energy state plays a role for the partitioning of assimilates into starch or proteins. The energy demand is highest for lipids, followed by proteins, and is lowest for starch. Starch synthesis in isolated cauliflower plastids is saturated at much lower ATP concentrations than that of lipids (Möhlmann et al., 1994). Soybean embryos accumulate starch from 10 days after pollination, closely followed by protein synthesis. Lipid biosynthesis, however, starts only 10 days later when embryos have become fully green (Yazdi-Samadi et al., 1976). From a gene expression study in Arabidopsis seeds, it was concluded that the maximal photosynthetic activity coincides more closely with oil than with starch synthesis (Ruuska et al., 2002). In summary, our results suggest that the relative fluxes of substrates into different classes of storage products may be controlled not only by the developmental stage and the genetic programme but also by the energy state of the tissue.

It is unknown in which way the differentiation of photoheterotrophic plastids during embryogenesis is triggered. There is some evidence for the role of sugars. Glucose feeding to tobacco leaves induces Glc-6-P uptake into chloroplasts, which is a typical feature for heterotrophic plastids (Quick et al., 1995). Rape embryos cultured on 0.35 m sucrose develop photoheterotrophic plastids, which function in storage. Without sucrose, they germinate and become photoautotrophic (Johnson et al., 1997). Stage IV Vicia embryos become green when cultured on sucrose but not on glucose (own unpublished results). In accordance with the greening process, the principal sugars in Vicia embryos changed at stages V and VI from hexoses to sucrose (Borisjuk et al., 1998, 2002; Weber et al., 1995). In this context, it is interesting that in stage V embryos, highest sucrose concentrations are present within the abaxial half of the cotyledons with much lower levels towards the interior (Borisjuk et al., 2002). This pattern spatially and temporally coincides with that of ATP and chlorophyll distribution. The role of sucrose for the induction of the photoheterotrophic plastid metabolism will be analysed in more detail in the future.

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Plant material

Faba bean (V. faba L. var. minor cv. Fribo) was grown in growth chambers under a 16-h light/8-h dark regime at 20°C/18°C. For the isolation of embryos, pods of different ages were collected and processed further.

Imaging of local ATP concentrations

The method of imaging bioluminescence was used to measure ATP concentrations in 20-µm cryosections of the specimens. The technique allows for the histographical mapping of metabolite concentrations within selected histological regions of tissue sections (Mueller-Klieser and Walenta, 1993; Walenta et al., 2002).

Embryos were dissected from seeds, shortly washed in ice-cold assay buffer without enzymes and subsequently shock-frozen in liquid N2. Embedding medium (O.C.T.; Miles Inc., USA) was given drop-wise onto the frozen cotyledons and became immediately frozen after tissue contact. This minimized solute exchange between tissue and medium, as well as diffusion. The embedding medium did not penetrate the tissue, but only held it in an appropriate position. Subsequently, 20-µm thick cryosections were prepared from the frozen embryos. For measurement, cryostat sections adhered to a glass slide were immersed into an enzyme solution using a mould, which was positioned on a thermostatically controlled (20°C) microscope stage (Axiophot, Zeiss, Oberkochen, Germany). The enzyme solution linked ATP to the reaction of firefly luciferase, i.e. an enzyme that emits photons with an intensity proportional to the content of ATP. The solution consisted of 0.05 g ml−1 desiccated firefly tails (Sigma, Taufkirchen, Germany) homogenized in buffer solution (0.2 m Hepes, 0.1 m sodium arsenate (pH 7.6); all chemicals: Sigma). The light emission was registered with a photon-counting video system (Hamamatsu, Herrsching, Germany) connected to the microscope, allowing for the detection of single photons including the exact location of its emission within the viewing field. The method was calibrated using standards (nanomoles per gram tissue). Calibration was performed at the beginning of each experiment using at least three sections with different ATP concentrations in ranges of 0–400 nm. Digitalized images of the ATP distributions were processed using image analysis software (Optimas Co., Bothell, Washington State, WA, USA). Representative maps for each stage were selected after analysing a minimum of three different embryos per stage. Average mean concentrations were calculated using all pixel values within selected tissue areas. The results are well in accordance with values determined biochemically (Rolletschek et al., 2003). The spatial resolution of the method depends on the lateral diffusion within sections during measurements, as well as on the microscopic magnification. Therefore, conditions chosen here for the experiments allowed a resolution of around 100 µm. Further details of quantitative bioluminescence and imaging photon counting have been published elsewhere by Mueller-Klieser and Walenta (1993) and Walenta et al. (2002). HPLC analysis of nucleotides was performed as described by Rolletschek et al. (2003).

Fixation, sectioning and histochemical staining

Seeds or parts of seeds were fixed in 2.5% glutaraldehyde, 50 mm sodium cacodylate (pH 7) or in 4% (w/v) paraformaldehyde, 50 mm potassium phosphate buffer (pH 7.0), under slight vacuum for 4 h at room temperature, rinsed in cacodylate buffer, dehydrated and embedded in Paraplast Plus (Sherwood Medical, St Louis, MO, USA). Sections were cut at 1–10 µm on a microtome, transferred on poly l-lysine-treated slides (Sigma Diagnostics, München, Germany) and were dried overnight at 45°C. As a general stain, toluidine blue was used according to Gerlach (1977) and Carson (1990). Immunostaining was performed using monospecific antibodies against V. faba legumin as described by Panitz et al. (1995).

In situ hybridization

In situ hybridization was performed according to Weber et al. (1995). The following cDNA fragments were used as probes after labelling with [33P]-dCTP: the complete cDNA of V. faba AGPC (Weber et al., 1995), a 520-bp PstI fragment from pVfc70 containing a V. faba legumin B4 sequence (Wobus et al., 1986).

Determination of mitotic index and cell expansion index

Mitotic indices were determined as described by Borisjuk et al. (1998). To quantify cell expansion, longitudinal sections of cotyledons were used to count the number of cells in a given unit area under the microscope. For each stage, at least four cotyledons were counted. Up to five areas from both the adaxial and abaxial half of the cotyledons were counted and averaged. The relative value of the degree of cell expansion (cell expansion index) was expressed as a quotient of the cell number per selected area from the cotyledon under observation and the cotyledons in the pre-expansion phase (stage IV).

Pigment analysis and starch determination

Frozen material (20–100 mg FW) was extracted once with 600 µl acetone and 4 µl 10 µm KOH and centrifuged for 5 min at 4°C and 10 000 g. The pellet was re-suspended with 300 µl acetone and 2 µl KOH. After centrifugation, the supernatants were combined and immediately measured spectrophotometrically according to Porra et al. (1989). Starch was determined as described by Borisjuk et al. (2002).

Photosynthesis and respiration measurement

O2 concentration inside seed tissues and gross photosynthetic rates were measured using micro sensors (Presens, Neuburg, Germany) as described in detail earlier (Rolletschek et al., 2002). Briefly, the microsensor was inserted using a micromanipulator at steps of 100–200 µm, and O2 concentration was measured. Gross photosynthetic rate was estimated as the rate of decrease in O2 concentration during the first few seconds following extinction of light (light intensity about 400–450 µmol m−2 sec−1). This so-called dark/light method is based on the following assumptions: (i) a steady state oxygen profile before darkening, (ii) a constant rate of respiration before and during the first few seconds of darkness and (iii) identical diffusive fluxes during this time (for theoretical discussions see Glud et al., 1992; Revsbech et al., 1981). To measure respiration, seeds were harvested, FW was determined and intact Vicia embryos were isolated and incubated in liquid medium. Respiration was measured as O2 uptake in darkness (for details, see Rolletschek et al., 2003).

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

This work was supported by Deutsche Forschungsgemeinschaft (DFG). We are grateful to Katrin Blaschek and Angela Schwarz for excellent technical assistance. We thank Ursula Tiemann, Karin Lipfert and Heike Ernst for the graphical artwork. We thank Bernhard Claus and Michael Melzer for confocal microscopy.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Asokanthan, P., Johnson, R.W., Griffith, M. and Krol, M. (1997) The photosynthetic potential of canola embryos. Physiol. Plant. 101, 353360.
  • Atkins, C.A. and Flinn, A.M. (1978) Carbon dioxide fixation in the carbon economy of developing seeds of Lupinus albus (L.). Plant Physiol. 62, 486490.
  • Borisjuk, L., Weber. H., Panitz, R., Manteuffel, R. and Wobus, U. (1995) Embryogenesis of Vicia faba L.: histodifferentiation in relation to starch and storage protein synthesis. J. Plant Physiol. 147, 203218.
  • Borisjuk, L., Walenta, S., Weber, H., Mueller-Klieser, W. and Wobus, U. (1998) High resolution histographical mapping of glucose concentrations in developing cotyledons of V. faba in relation to mitotic activity and storage processes: glucose as a developmental trigger. Plant J. 15, 583591.
  • Borisjuk, L., Walenta, S., Rolletschek, H., Mueller-Klieser, W., Wobus, U. and Weber, H. (2002) Spatial analysis of plant metabolism: sucrose imaging within Vicia faba cotyledons reveals specific developmental patterns. Plant J. 29, 521530.
  • Browse, J. and Slack, C.R. (1985) Fatty acid synthesis in plastids from maturing sunflower and linseed cotyledons. Planta, 166, 7480.
  • Carson, F.L. (1990) Histotechnology: a Self-Instructional Text. Chicago: ASCP Press.
  • Ching, T.M., Crane, J.M. and Stamp, D.L. (1974) Adenylate energy pool and energy charge in maturing rape seeds. Plant Physiol. 54, 748751.
  • Eastmond, P. and Rawsthorne, S. (1998) Comparison of the metabolic properties of plastids isolated from developing leaves or embryos of Brassica napus. J. Exp. Bot. 49, 11051111.
  • Eastmond, P., Kolacna, L. and Rawsthorne, S. (1996) Photosynthesis by developing embryos of oilseed rape. J. Exp. Bot. 47, 17631769.
  • Farré, E.M., Tiessen, A., Roessner, U., Geigenberger, P., Trethewey, R.N. and Willmitzer, L. (2001) Analysis of the compartmentation of glycolytic intermediates, nucleotides, sugars, organic acids, amino acids, and sugar alcohols in potato tubers using a nonaqueous fractionation method. Plant Physiol. 127, 685700.
  • Flinn, A.M. (1985) Carbon dioxide fixation in developing seeds. In The Pea Crop: a Basis for Improvement (Hebblethwaite, P.D., Heath, M.C. and Dawkins, T.C.K., eds). London: Butterworths, pp. 349358.
  • Flinn, A.M. and Pate, J.S. (1970) A quantitative study of carbon transfer from pod and subtending leaf to the ripening seeds of the field pea (Pisum arvense L.). J. Exp. Bot. 21, 7182.
  • Gerlach, D. (1977) Botanische Mikrotechnik. 2. Stuttgart: Auflage, Thieme Verlag.
  • Glud, R.N., Ramsing, N.B. and Revsbech, N.P. (1992) Photosynthesis and photosynthesis-coupled respiration in natural biofilms quantified with oxygen microsensors. J. Physiol. 28, 5160.
  • Golombek, S., Heim, U., Horstmann, C., Wobus, U. and Weber, H. (1999) PEP-carboxylase in developing seeds of Vicia faba. Gene expression and metabolic regulation. Planta, 208, 6672.
  • Harvey, D.M., Hedley, C.L. and Keely, R. (1976) Photosynthetic and respiratory studies during pod and seed development in Pisum sativum L. Ann. Bot. 40, 9931001.
  • Hedley, C.L., Harvey, D.M. and Keely, R.J. (1975) The role of PEP-carboxylase during seed development in P. sativum. Nature, 258, 352354.
  • Heineke, D., Riens, B., Grosse, H., Hoferichter, P., Peter, U., Flügge, U.I. and Heldt, H.W. (1991) Redox transfer across the inner chloroplast envelope membrane. Plant Physiol. 95, 11311137.
  • Johnson, R.W., Asokanthan, P.S. and Griffith, M. (1997) Water and sucrose regulate canola embryo development. Physiol. Plant. 101, 361366.
  • King, S.P., Badger, M.R. and Furbank, R.T. (1998) CO2 characteristics of developing canola seeds and silique wall. Aust. J. Plant Physiol. 25, 377386.
  • Kollöffel, C. and Matthews, S. (1983) Respiratory activity in pea cotyledons during seed development. J. Exp. Bot. 34, 10261036.
  • Levin, M., Björnhelden, T., Evaldsson, M., Walenta, S. and Wiklund, O. (1999) A bioluminescence method for mapping of local ATP concentrations within the arterial wall, with potential to assess the in vivo situation. Arterioscler. Thromb. Vasc. Biol. 19, 950958.
  • Möhlmann, T., Scheibe, R. and Neuhaus, H.E. (1994) Interaction between starch synthesis and fatty-acid synthesis in isolated cauliflower-bud amyloplasts. Planta, 194, 492497.
  • Mueller-Klieser, W. and Walenta, S. (1993) Geographical mapping of metabolites in biological tissue with quantitative bioluminescence and single photon imaging. Histochem. J. 2, 407420.
  • Neuhaus, H.E. and Emes, M.J. (2000) Nonphotosynthetic metabolism in plastids. Annu. Rev. Plant Physiol. Plant Mol. Biol. 51, 111140.
  • Panitz, R., Borisjuk, L., Manteuffel, R. and Wobus, U. (1995) Transient expression of storage protein genes during early embryogenesis of V. faba: synthesis and metabolisation of vicilin and legumin in the embryo, suspensor and endosperm. Planta, 196, 765774.
  • Porra, R.J., Thompson, W.A. and Kriedemann, P.E. (1989) Determination of accurate extinction coefficients and simultaneous equations for assaying chlorophyll a and chlorophyll b extracted with different solvents verification of the concentration of chlorophyll standards by atomic-absorption spectroscopy. Biochim. Biophys. Acta, 975, 384394.
  • Pradet, A. and Raymond, A. (1983) Adenine nucleotide ratios and adenylate energy charge in energy metabolism. Annu. Rev. Plant Physiol. 34, 199224.
  • Quebedeaux, B. (1981) Adenylate and nicotinamide nucleotides in developing soybean seeds during seed-fill. Plant Physiol. 68, 2327.
  • Quick, P.W., Scheibe, R. and Neuhaus, H.E. (1995) Induction of a hexose-phosphate translocator activity in spinach chloroplasts. Planta, 194, 193199.
  • Rawsthorne, S. (2002) Carbon flux and fatty acid synthesis in plants. Prog. Lipid Res. 41, 182196.
  • Revsbech, N.P., Jorgensen, B.B. and Brix, O. (1981) Primary production of microalgae in sediments measured by oxygen microprofiles, H14CO2-fixation and oxygen exchange methods. Limnol. Oceanogr. 26, 717730.
  • Rolletschek, H., Borisjuk, L., Koschorreck, M., Wobus, U. and Weber, H. (2002) Legume embryos develop in a hypoxic environment. J. Exp. Bot. 53, 10991107.
  • Rolletschek, H., Weber, H. and Borisjuk, L. (2003) Energy status and its control on embryogenesis of legumes. Photosynthetic oxygen production is coupled to diurnally changing biosynthetic fluxes and reduces hypoxia. Plant Physiol. 132, 11961206.
  • Ruuska, S.A., Girke, T., Benning, C. and Ohlrogge, J.B. (2002) Contrapunctal networks of gene expression during Arabidopsis seed filling. Plant Cell, 14, 11911206.
  • Saito, G.Y., Chang, Y.C., Walling, L.L. and Thompson, W.W. (1989) A correlation in plastid development and cytoplasmic ultrastructure with nuclear gene expression during seed ripening in soybean. New Phytol. 113, 459469.
  • Smith, A.M., Quinton-Tulloch, J. and Denyer, K. (1990) Characteristics of plastids responsible for starch synthesis in developing pea embryos. Planta, 180, 517523.
  • Tjaden, J., Möhlmann, T., Kampfenkel, K., Henrichs, G. and Neuhaus, H.E. (1998) Altered plastidic ATP/ADP-transporter activity influences potato morphology, amount and composition of tuber starch and tuber morphology. Plant J. 16, 531540.
  • Walenta, S., Schroeder, T. and Mueller-Klieser, W. (2002) Invited review. Metabolic mapping with bioluminescence: basic and clinical relevance. Biomol. Eng. 18, 249262.
  • Weber, H., Borisjuk, L., Heim, U., Buchner, P. and Wobus, U. (1995) Seed coat-associated invertases of Fava bean control both unloading and storage functions: cloning of cDNAs and cell type-specific expression. Plant Cell, 7, 18351846.
  • Weber, H., Heim, U., Borisjuk, L. and Wobus, U. (1995) Cell-type specific, coordinate expression of two ADP-glucose pyrophosphorylase genes in relation to starch biosynthesis during seed development of Vicia faba L. Planta, 195, 352361.
  • Weber, H., Borisjuk, L., Heim, U., Sauer, N. and Wobus, U. (1997) A role for sugar transporters during seed development: molecular characterization of a hexose and a sucrose carrier in Fava bean seeds. Plant Cell, 9, 895908.
  • Willms, J.R., Salon, C. and Layzell, D.B. (1999) Evidence for light-stimulated fatty acid synthesis in soybean fruit. Plant Physiol. 120, 11171128.
  • Wobus, U. and Weber, H. (1999) Sugars as signal molecules in plant seed development. J. Biol. Chem. 380, 937944.
  • Wobus, U., Bäumlein, H., Bassüner, R., Heim, U., Jung, R., Müntz, K., Saalbach, G. and Weschke, W. (1986) Characteristics of two types of legumin genes in the field bean (Vicia faba L. var. minor) genome as revealed by cDNA analysis. FEBS Lett. 201, 7479.
  • Yazdi-Samadi, B., Rinne, R.W. and Seif, R.D. (1976) Components of developing soybean seeds: oil, protein, starch, organic acids and amino acids. Agron. J. 69, 481486.