Intercellular Localization of Assimilatory Sulfate Reduction in Leaves of Zea mays and Triticum aestivum

The intercellular distribution of assimilatory sulfate reduction enzymes between mesophyll and bundle sheath cells was analyzed in maize (Zea mays L.) and wheat (Triticum aestivum L.) leaves. In maize, a C₄ plant, 96 to 100% of adenosine 5′-phosphosulfate sulfotransferase and 92 to 100% of ATP sulfurylase activity (EC 2.7.7.4) was detected in the bundle sheath cells. Sulfite reductase (EC 1.8.7.1) and O-acetyl-l-serine sulfhydrylase (EC 4.2.99.8) were found in both bundle sheath and mesophyll cell types. In wheat, a C₃ species, ATP sulfurylase and adenosine 5′-phosphosulfate sulfotransferase were found at equivalent activities in both mesophyll and bundle sheath cells. Leaves of etiolated maize plants contained appreciable ATP sulfurylase activity but only trace adenosine 5′-phosphosulfate sulfotransferase activity. Both enzyme activities increased in the bundle sheath cells during greening but remained at negligible levels in mesophyll cells. In leaves of maize grown without addition of a sulfur source for 12 d, the specific activity of adenosine 5′-phosphosulfate sulfotransferase and ATP sulfurylase in the bundle sheath cells was higher than in the controls. In the mesophyll cells, however, both enzyme activities remained undetectable. The intercellular distribution of enzymes would indicate that the first two steps of sulfur assimilation are restricted to the bundle sheath cells of C₄ plants, and this restriction is independent of ontogeny and the sulfur nutritional status of the plants.     

Light Modulation and Localization of Sucrose Phosphate Synthase Activity between Mesophyll Cells and Bundle Sheath Cells in C(4) Species

Experiments were conducted with several Panicum species (representing the different C₄ subtypes) to examine the light modulation of sucrose phosphate synthase (SPS) activity and the effect of illumination on the distribution of SPS activity between mesophyll cells (MC) and bundle sheath cells (BSC). Activity of SPS in the light decreased in the order: C₄ > C₃-C₄ intermediate > C₃. In illuminated leaves, SPS activities were similar among the three C₄ subtypes, but SPS activity was higher for NAD-malic enzyme (NAD-ME) species with centripetal chloroplasts in BSC (NAD-ME(P) species) than for NAD-ME species with centrifugal chloroplasts in BSC (NAD-ME(F) species). Transfer of plants into darkness for 30 minutes resulted in decreased SPS activity for all species tested except Panicum bisulcatum (C₃ species) and Panicum virgatum (NAD-ME(P) species) which showed little or no change. All C₄ subtypes had some SPS activity both in MC and BSC. In the light, SPS activity was mainly in the MC for NADP-ME, NAD-ME(F) and phosphoenolpyruvate carboxykinase species, while it was mainly in the BSC for NAD-ME(P) species. In the dark, for all C₄ subtypes, SPS activity in the MC was decreased to a greater extent than that in the BSC. It is intriguing that NAD-ME(F) and NAD-ME(P) species differed in the activity and distribution of SPS activity between MC and BSC, although they are otherwise identical in the photosynthetic carbon assimilation pathway. Diurnal changes in SPS activity in the MC and BSC were also examined in maize leaves. SPS activity in the MC in maize leaves was high and relatively constant throughout the middle of the light period, dropped rapidly after sunset and increased again prior to the light period. On the other hand, SPS activity in the BSC was lower and changed more coincidently with light intensity than that in the MC. The results suggested that light activation of SPS activity located in the BSC may require higher irradiance for saturation than the SPS in the MC. We conclude that SPS may function in both MC and BSC for sucrose synthesis in the light, particularly at high light intensity, while in the dark, the major function may be in the BSC during starch degradation.     

Adenosine 5′-triphosphate-mediated activation of sucrose-phosphate synthase in bundle sheath cells of C4 plants

We report the ATP-mediated activation of sucrose-phosphate synthase in bundle sheath cells prepared from C₄ species. Sucrose synthesis was followed by measuring the incorporation of [¹⁴C]fructose 6-phosphate into sucrose in bundle sheath cells also provided with uridine 5′-diphosphoglucose (UDPGlc). Studies with Panicum miliaceum L. cells showed that activation was largely due to an increase in the affinity for UDPGlc and was therefore only evident at limiting UDPGlc concentrations. The apparent K _m UDPGlc for sucrose synthesis by cells pretreated and assayed with ATP was about 0.7 mM compared with 7–8 mM for control cells without ATP. The γ-thio derivative of ATP had a similar effect to ATP. The effect was also evident when ATP was rapidly removed from cells prior to assay. Sucrose-phosphate synthase activity in extracts from cells pretreated with or without ATP showed similar differences in K _m UDPGlc. These observations support the view that ATP is inducing a covalent modification of the enzyme. However, several protein kinase inhibitors did not prevent activation. Changes of more than threefold were observed for the K _m UDPGlc with sucrose-phosphate synthase extracted from bundle sheath cells rapidly isolated from attached leaves that were subjected to dark/light treatments. The possible relationship between these changes and those induced by ATP with isolated cells is discussed. Received: 22 October 1996 / Accepted: 7 January 1997     

Ultrastructure of leaves in C4 Cyperus iria and C3 Carex siderosticta

The ultrastructural aspects ofCyperus iria leaves showing the C₄ syndrome and the typical C₃ species,Carex siderosticta, in the Cyperaceae family were examined.C. iria exhibited the chlorocyperoid type, showing an unusual Kranz structure with vascular bundles completely surrounded by two bundle sheaths. The cellular components of the inner Kranz bundle sheath cells were similar to those found in the NADP-ME C₄ subtype, having centrifugally arranged chloroplasts with greatly reduced grana and numerous starch grains. Their chloroplasts contained convoluted thyla-koids and a weakly-developed peripheral reticulum, although it was extensive mostly in mesophyll cell chloroplasts. The outer mestome bundle sheath layer was sclerenchymatous and generally devoid of organelles, but had unevenly thickened walls. Suberized lamellae were present on its cell walls, and they became polylamellate when traversed by plasmodesmata. Mesophyll cell chloroplasts showed well-stacked grana with small starch grains. InC. siderosticta, vascular bundles were surrounded by the inner mestome sheath and the outer parenchymatous bundle sheath with intercellular spaces. The mestome sheath cells degraded in their early development and remained in a collapsed state, although the suberized lamellae retained polylamellate features. Plastids with a crystalline structure, sometimes membrane-bounded, were found in the epidermal cells. The close interveinal distance was 35–50 μm inC. iria, whereas it was 157–218 μm inC. siderosticta. These ultrastructural characteristics were discussed in relation to their photosynthetic functions.     

Aberrant chloroplasts in transgenic rice plants expressing a high level of maize NADP-dependent malic enzyme

 NADP-dependent malic enzyme (NADP-ME) is a major decarboxylating enzyme in NADP-ME-type C₄ species such as maize and Flaveria. In this study, chloroplastic NADP-ME was transferred to rice (Oryza sativa L.) using a chimeric gene composed of maize NADP-ME cDNA under the control of rice light-harvesting chlorophyll-a/b-binding protein (Cab) promoter. There was a 20- to 70-fold increase in the NADP-ME activity in leaves of transgenic rice compared to that in wild-type rice plants. Immunocytochemical studies by electron microscopy showed that maize NADP-ME was mostly localized in chloroplasts in transgenic rice plants, and that the chloroplasts were agranal without thylakoid stacking. Chlorophyll content and photosystem II activity were inversely correlated with the level of NADP-ME activity. These results suggest that aberrant chloroplasts in transgenic plants may be caused by excessive NADP-ME activity. Based on these results and the known fact that only bundle sheath cells of NADP-ME species, among all three C₄ subgroups, have agranal chloroplasts, we postulate that a high level of chloroplastic NADP-ME activity could strongly affect the development of chloroplasts. Received: 27 January 1999 / Accepted: 20 January 2000     

Photochemical Characteristics of Mesophyll and Bundle Sheath Chloroplasts from C4 Plants

Mesophyll and bundle sheath chloroplasts were isolated by differential grinding from the leaves of two NADP-ME C₄ plants, Setaria italica Beauv. cv. H-1, Pennisetum typhoides S & H. cv. AKP-2, and a NAD-ME C₄ species Amaranthus paniculatus L. The mesophyll chloroplasts of C₄ plants possessed slightly lower K_m for ADP and Pi than those of bundle sheath chloroplasts. The Hill reaction activities and noncyclic photophosphorylation rates of the bundle sheath chloropiasts from S. italica and P. typhoides were less than one-fifth of those by the mesophyll chloroplasts. But the bundle sheath chloroplasts of A. paniculatus exhibited high rates of Hill reaction, cyclic as well as noncyclic photophosphorylation. The pigment- and eyiochrome composition suggested a relative enrichment of PS 1 in bundle sheath chloroplasts of S. italica and P. typhoides. The chain exists in both mesophyll and bundle sheath chloroplasts. As much as 35–52% of leaf chlorophyll was located in the bundle sheath chloroplasts. The photochemical activities of bundle sheath chloroplasts are significant though a major part of leaf photochemical potential is associated with the mesophyll chloroplasts.     

Research into the characteristics of C3 and C4 plants inCeratophyllum demersum L.

Summary The leaf anatomy was investigated with respect to the arrangement of cells involved in photosynthesis. The full-grown leaf has one vascular bundle consisting mainly of phioem cells. In similarity to terrestrial C₄ plants the vascular bundle is surrounded by mesophyll bundle sheath cells. However, in contrast to C₄ plants, these cells do not contain chlorophyll or starch inCeratophyllum. The early products in photosynthesis (10 seconds¹⁴C labelling) were analyzed. Although no complete separation of all radioactivity in the plant extracts was reached, it was clear that malate was the major labelled component, indicating C₄ activity in the plants. No light saturation could be proven inCeratophyllum in several stages of post-dormancy in a statistically significant way, although a tendency to light saturation was observed at intensities higher than 36 Wm^–2. The photosynthetic activity was only slightly depressed by enhancement of the O₂ concentration in the medium.     

Sulfate assimilation and glutathione synthesis in C<Subscript>4</Subscript> plants

Sulfate assimilation and glutathione synthesis were traditionally believed to be differentially compartmentalised in C₄ plants with the synthesis of cysteine and glutathione restricted to bundle sheath and mesophyll cells, respectively. Recent studies, however, showed that although ATP sulfurylase and adenosine 5′ phosphosulfate reductase, the key enzymes of sulfate assimilation, are localised exclusively in bundle sheath in maize and other C₄ monocot species, this is not true for the dicot C₄ species of Flaveria. On the other hand, enzymes of glutathione biosynthesis were demonstrated to be active in both types of maize cells. Therefore, in this review the recent findings on compartmentation of sulfate assimilation and glutathione metabolism in C₄ plants will be summarised and the consequences for our understanding of sulfate metabolism and C₄ photosynthesis will be discussed.     

Two-Dimensional Electrophoretic Mapping of Proteins of Bundle Sheath and Mesophyll Cells of the C(4) Grass Digitaria sanguinalis (L.) Scop. (Crabgrass)

Two-dimensional electrophoresis was performed on proteins of bundle sheath and mesophyll cells isolated from the C₄ grass Digitaria sanguinalis (L.) Scop. Two-dimensional maps of these proteins were constructed and ribulose-1,5-biphosphate carboxylase and phosphoenolpyruvate carboxylase were identified. Of the total number of proteins found in both cell types, 36% were found only in bundle sheath cells, 17% only in mesophyll cells, and 47% in both cell types. By comparison, the distributions of 48 enzymes assayed in these cell types were 35%, 21%, and 44%, respectively.

Protein patterns were also compared with C₄ plants exhibiting different decarboxylation pathways and, in both bundle sheath and mesophyll cells, proteins were found which were unique to each species. Bundle sheath proteins of one C₄ species were found to be more like bundle sheath proteins of another C₄ species than like mesophyll proteins of the same species.

     

Purification and Properties of Mesophyll and Bundle Sheath Cell alpha-Glucan Phosphorylases from Zea mays L. : Equivalence of the Enzymes with the Cytosol and Plastid Phosphorylases from Spinach

Two major α-glucan phosphorylases (I and II) from leaves of the C₄ plant corn (Zea mays L.) were previously shown to be compartmented in mesophyll and bundle sheath cells, respectively (C Mateyka, C Schnarrenberger 1984 Plant Sci Lett 36: 119-123). The two enzymes were separated by chromatography on DEAE-cellulose and purified to homogeneity by affinity chromatography on immobilized starch, according to published procedures, as developed for the cytosol and chloroplast phosphorylase from the C₃ plant spinach. The two α-glucan phosphorylases have their pH optimum at pH 7. The specificity for polyglucans was similar for soluble starch and amylopectin, however, differed for glycogen (K_m = 16 micrograms per milliliter for the mesophyll cell and 250 micrograms per milliliter for the bundle sheath cell phosphorylase). Maltose, maltotriose, and maltotetraose were not cleaved by either phosphorylase. If maltopentaose was used as substrate, the rate was about twice as high with the bundle sheath cell phosphorylase, than with the mesophyll cell phosphorylase. The phosphorylase I showed a molecular mass of 174 kilodaltons and the phosphorylase II of 195 kilodaltons for the native enzyme and of 87 and of 53 kilodaltons for the SDS-treated proteins, respectively. Specific antisera raised against mesophyll cell phosphorylase from corn leaves and against chloroplast phosphorylase from spinach leaves implied high similarity for the cytosol phosphorylase of the C₃ plant spinach with mesophyll cell phosphorylase of the C₄ plant corn and of chloroplast phosphorylase of spinach with the bundle sheath cell phosphorylase of corn.     

Mechanism of c(4) photosynthesis: the size and composition of the inorganic carbon pool in bundle sheath cells

We sought to characterize the inorganic carbon pool (CO₂ plus HCO₃⁻) formed in the leaves of C₄ plants when C₄ acids derived from CO₂ assimilation in mesophyll cells are decarboxylated in bundle sheath cells. The size and kinetics of labeling of this pool was determined in six species representative of the three metabolic subgroups of C₄ plants. The kinetics of labeling of the inorganic carbon pool of leaves photosynthesizing under steady state conditions in ¹⁴CO₂ closely paralleled those for the C-4 carboxyl of C₄ acids for all species tested. The inorganic carbon pool size, determined from its ¹⁴C content at radioactivity saturation, ranged between 15 and 97 nanomoles per milligram of leaf chlorophyll, giving estimated concentrations in bundle sheath cells of between 160 and 990 micromolar. The size of the pool decreased, together with photosynthesis, as light was reduced from 900 to 95 microeinsteins per square meter per second or as external CO₂ was reduced from 400 to 98 microliters per liter. A model is developed which suggests that the inorganic carbon pool existing in the bundle sheath cells of C₄ plants during steady state photosynthesis will comprise largely of CO₂; that is, CO₂ will only partially equlibrate with bicarbonate. This predominance of CO₂ is believed to be vital for the proper functioning of the C₄ pathway.     

Activity, location, and role of asparate aminotransferase and alanine aminotransferase isoenzymes in leaves with C4 pathway photosynthesis

Further evidence has been provided that C₄-pathway species characterized by having low malic enzyme activity contain exceptionally high activities of aspartate and alanine aminotransferases. The total activity of both enzymes is distributed about equally between mesophyll and bundle sheath cells. However, the activity in the two cell types is due to different isoenzymes. In addition to the one quantitatively major isoenzyme associated with each cell type there were at least two additional isozymes of each aminotransferase detectable in the different species examined. Increases in activity of both aminotransferases of ten-fold or more were observed during greening of leaves of dark-grown plants. This increased activity was due specifically to the two quantitatively major isoenzymes associated, respectively, with the mesophyll and bundle sheath cells of green leaves, providing further evidence for their specific role in photosynthesis. Apparently, neither the aspartate nor alanine aminotransferases of mesophyll cells was associated with chloroplasts or other subcellular organelles. However, the major aspartate aminotransferase isoenzyme of bundle sheath cells was associated with mitochondria. These findings are discussed in relation to the probable role of aspartate and alanine aminotransferases in C₄-pathway photosynthesis.     

Pyruvate, pi dikinase in bundle sheath strands as well as in mesophyll cells in maize leaves

Mesophyll protoplasts and bundle sheath strands were isolated from maize leaves. Light microscopic observation showed the preparations were pure and without cross contamination. Protein blot analysis of mesophyll and bundle sheath cell soluble protein showed that the concentration of pyruvate orthophosphate dikinase (EC 2.7.9.1) is about one-tenth as much in the bundle sheath cells as in mesophyll cells, but about eight times greater than that found in wheat leaves, on the basis of soluble protein. Phosphoenolpyruvate carboxylase (EC 4.1.1.31) was barely detectable in the bundle sheath cells, while ribulose-1,5-bisphosphate carboxylase (EC 4.1.1.39) and NADP-dependent malic enzyme (EC 1.3.1.37) were exclusively present in the bundle sheath cells and were absent in the mesophyll cells. Whereas pyruvate, Pi dikinase was previously considered localized only in mesophyll cells of C₄ plants, these results clearly demonstrate the presence of appreciable quantities of the enzyme in the bundle sheath cells of the C₄ species maize.     

Low bundle sheath carbonic anhydrase is apparently essential for effective c(4) pathway operation

Bundle sheath cells from leaves of a variety of C₄ species contained little or no carbonic anhydrase activity. The proportion of total leaf carbonic anhydrase in extracts of bundle sheath cells closely reflected the apparent mesophyll cell contamination of bundle sheath cell extracts as measured by the proportion of the mesophyll cell marker enzymes phosphoenolpyruvate carboxylase and pyruvate,Pi dikinase. Values of about 1% or less of the total leaf activity were obtained for all three enzymes. The recorded bundle sheath carbonic anhydrase activity was compared with a calculated upper limit of carbonic anhydrase activity that would still permit efficient functioning of the C₄ pathway; that is, a carbonic anhydrase level allowing a sufficiently high steady state [CO₂] to suppress photorespiration. Even before correcting for mesophyll cell contamination the activity in bundle sheath cell extracts was substantially less than the calculated upper limit of carbonic anhydrase activity consistent with effective C₄ function. The results accord with the notion that a deficiency of carbonic anhydrase in bundle sheath cells is vital for the efficient operation of the C₄ pathway.     

Intercellular compartmentation of sucrose synthesis in leaves of Zea mays L.

The incorporation of ¹⁴C into sucrose and hexose phosphates during steady-state photosynthesis was examined in intact leaves of Zea mays L. plants. The compartmentation of sucrose synthesis between the bundle sheath and mesophyll cells was determined by the rapid fractionation of the mesophyll and comparison of the labelled sucrose in this compartment with that in a complete leaf after homogenisation. From these experiments it was concluded that the majority of sucrose synthesis occurred in the mesophyll cell type (almost 100% when the time-course of sucrose synthesis was extrapolated to the time of ¹⁴C-pulsing). The distribution of enzymes involved in sucrose synthesis between the two cell types indicated that sucrose-phosphate synthetase was predominantly located in the mesophyll, as was cytosolic (neutral) fructose-1,6-bisphosphatase activity. Stromal (alkaline) fructose-1,6-bisphosphatase activity was found almost exclusively in the bundle-sheath cells. No starch was found in the mesophyll tissue. These data indicate that in Zea mays starch and sucrose synthesis are spatially, separated with sucrose synthesis occurring in the mesophyll compartment and starch synthesis in the bundle sheath.     

A mathematical model of C4 photosynthesis: The mechanism of concentrating CO2 in NADP-malic enzyme type species

A computer model comprising light reactions in PS II and PS I, electron-proton transport reactions in mesophyll and bundle sheath chloroplasts, all enzymatic reactions and most of the known regulatory functions of NADP-ME type C₄ photosynthesis has been developed as a system of differential budget equations for intermediate compounds. Rate-equations were designed on principles of multisubstrate-multiproduct enzyme kinetics. Some of the 275 constants needed (ΔG₀′ and K _m values) were available from literature and others (V _m) were estimated from reported rates and pool sizes. The model provided good simulations for rates of photosynthesis and pool sizes of intermediates under varying light, CO₂ and O₂. A basic novelty of the model is coupling of NADPH production via NADP-ME with ATP production and regulation of the C₃ cycle in bundle sheath chloroplasts. The functional range of the ATP/NADPH ratio in bundle sheath chloroplasts extends from 1.5 to 2.1, being energetically most efficient around 2. In the presence of such stoichiometry, the CO₂ concentrating function can be explained on the basis of two processes: (a) extra ATP consumption for starch and protein synthesis in bundle sheath leads to a faster NADPH and CO₂ import compared with CO₂ fixation in bundle sheath, and (b) the residual photorespiratory activity consumes RuBP by oxygenation, NADPH and ATP and causes the imported CO₂ to accumulate in bundle sheath cells. As a wider application, the model may be used for predicting results of genetic engineering of plants. This revised version was published online in June 2006 with corrections to the Cover Date.     

Predominant localization of mitochondria enriched with glycine-decarboxylating enzymes in bundle sheath cells of Alternanthera tenella, a C3–C4 intermediate species

Mesophyll protoplasts and bundle sheath cells were prepared by enzymatic digestion of leaves of Alternanthera tenella, a C₃-C₄ intermediate species. The intercellular distribution of selected photosynthetic, photorespiratory and respiratory (mitochondrial) enzymes in these meso-phyll and bundle sheath cells was studied. The activity levels of photosynthetic enzymes such as PEP carboxylase (EC 4.1.1.31) or NAD-malic enzyme (EC 1.1.1.39) and photorespiratory enzymes such as glycolate oxidase (EC 1.1.3.1) or NADH-hydroxypyruvate reductase (EC 1.1.1.29) were similar in the two cell types. The activity levels of mitochondrial TCA cycle enzymes such as citrate synthase (EC 4.1.3.7) or fumarase (EC 4.2.1.2) were 2- to 3-fold higher in bundle sheath cells. On the other hand, the activity levels of mitochondrial photorespiratory enzymes, namely glycine decarboxylase (EC 2.1.2.10) and serine hydroxymethyltransferase (EC 2.1.2.1), were 6-9-fold higher in bundle sheath cells than in mesophyll protoplasts. Such preferential localization of mitochondria enriched with the glycine-decarboxylating system in the inner bundle sheath cells would result in efficient refixa-tion of CO₂ from not only photorespiration but also dark respiration before its exit from the leaf. We propose that predominant localization of mitochondria specialized in glycine decarboxylation in bundle sheath cells may form the basis of reduced photorespiration in this C₃-C₄ intermediate species.     

Distribution of Photosynthetic Enzymes between Mesophyll, Specialized Parenchyma and Bundle Sheath Cells of Arundinella hirta

Arundinella hirta L. is a C₄ plant having an unusual C₄ leaf anatomy. Besides mesophyll and bundle sheath cells, A. hirta leaves have specialized parenchyma cells which look morphologically like bundle sheath cells but which lack vascular connections and are located between veins, running parallel to them. Activities of phosphoenolpyruvate and ribulose-1,5-bisphosphate carboxylases and phosphoenolpyruvate carboxykinase, NADP-and NAD-malic enzymes were determined for whole leaf extracts and isolated mesophyll protoplasts, specialized parenchyma cells, and bundle sheath cells. The data indicate that A. hirta is a NADP-malic enzyme type C₄ species. In addition, specialized parenchyma cells and bundle sheath cells are enzymatically alike. Compartmentation of enzymes followed the C₄ pattern with phosphoenolpyruvate carboxylase being restricted to mesophyll cells while ribulose-1,5-bisphosphate carboxylase and decarboxylating enzymes were restricted to bundle sheath and specialized parenchyma cells.     

Intracellular localization of certain photosynthetic enzymes in bundle sheath cells of plants possessing the C4 pathway of photosynthesis.

Bundle sheath cells were enzymatically isolated from representatives of three groups of C₄ plants: Zea mays (NADP malic enzyme type), Panicum miliaceum (NAD malic enzyme type), and Panicum maximum (phosphoenolpyruvate (PEP) carboxykinase type). Cellular organelles from bundle sheath homogenates were partially resolved by differential centrifugation and on isopycnic sucrose density gradients in order to study compartmentation of photosynthetic enzymes. A 48-h-dark pretreatment of the leaves allowed the isolation of relatively intact chloroplasts. Enzymes that decarboxylate C₄ acids and furnish CO₂ to the Calvin cycle are localized as follows: NADP malic enzyme, chloroplastic in Z. mays; NAD malic enzyme, mitochondrial in all three species; PEP carboxykinase, chloroplastic in P. maximum. The activity of NAD malic enzyme in the three species was in the order of P. miliaceum > P. maximum > Z. mays. There were high levels of aspartate and alanine aminotransferases in bundle sheath extracts of P. miliaceum and P. maximum and substantial activity in Z. mays. In all three species, aspartate aminotransferase was mitochondrial whereas alanine aminotransferase was cytoplasmic. Based on the activity and localization of certain enzymes, the concept for aspartate and malate as transport metabolites from mesophyll to bundle sheath cells in C₄ species of the three C₄ groups is discussed.