Manipulating desaturase activities in transgenic crop plants.

The properties of edible vegetable oils are determined to a large extent by the relative content of the triacylglycerol fatty acids. The degree of saturation of these fatty acids can determine the functional, sensory and nutritional value of the oil. One method of altering the unsaturated fatty acid content of oilseeds is by manipulating the expression of desaturase genes of these plants. Manipulating the expression of desaturase genes in transgenic crops such as soybean, maize and canola (oilseed rape) has led to oils with improved functionality and nutrition. We have also been successful in manipulating the fatty acid content of domesticated oilseed plants by expressing heterologous desaturase and desaturase-related genes from exotic plants that produce unusual fatty acids. We have discovered that metabolic regulation, the number of genetic alleles that encode fatty acid biosynthetic enzymes, and the movement of fatty acids between complex lipids in the cell, all have a role in determining the effect of a transgene on the phenotype of the crop plant and the fatty acid composition of its seed oil.

Cloning of a higher-plant plastid omega-6 fatty acid desaturase cDNA and its expression in a cyanobacterium.

Oligomers based on amino acids conserved between known plant omega-3 and cyanobacterium omega-6 fatty acid desaturases were used to screen an Arabidopsis cDNA library for related sequences. An identified clone encoding a novel desaturase-like polypeptide was used to isolate its homologs from Glycine max and Brassica napus. The plant deduced amino acid sequences showed less than 27% similarity to known plant omega-6 and omega-3 desaturases but more than 48% similarity to cyanobacterial omega-6 desaturase, and they contained putative plastid transit sequences. Thus, we deduce that the plant cDNAs encode the plastid omega-6 desaturase. The identity was supported by expression of the B. napus cDNA in cyanobacterium. Synechococcus transformed with a chimeric gene that contains a prokaryotic promoter fused to the rapeseed cDNA encoding all but the first 73 amino acids partially converted its oleic acid fatty acid to linoleic acid, and the 16:1(9c) fatty acid was converted primarily to 16:2(9c, 12) in vivo. Thus, the plant omega-6 desaturase, which utilizes 16:1(7c) in plants, can utilize 16:1(9c) in the cyanobacterium. The plastid and cytosolic homologs of plant omega-6 desaturases are much more distantly related than those of omega-3 desaturases.

Cloning of higher plant omega-3 fatty acid desaturases.

Arabidopsis thaliana T-DNA transformants were screened for mutations affecting seed fatty acid composition. A mutant line was found with reduced levels of linolenic acid (18:3) due to a T-DNA insertion. Genomic DNA flanking the T-DNA insertion was used to obtain an Arabidopsis cDNA that encodes a polypeptide identified as a microsomal omega-3 fatty acid desaturase by its complementation of the mutation. Analysis of lipid content in transgenic tissues demonstrated that this enzyme is limiting for 18:3 production in Arabidopsis seeds and carrot hairy roots. This cDNA was used to isolate a related Arabidopsis cDNA, whose mRNA is accumulated to a much higher level in leaf tissue relative to root tissue. This related cDNA encodes a protein that is a homolog of the microsomal desaturase but has an N-terminal extension deduced to be a transit peptide, and its gene maps to a position consistent with that of the Arabidopsis fad D locus, which controls plastid omega-3 desaturation. These Arabidopsis cDNAs were used as hybridization probes to isolate cDNAs encoding homologous proteins from developing seeds of soybean and rapeseed. The high degree of sequence similarity between these sequences suggests that the omega-3 desaturases use a common enzyme mechanism.

A 62-kD sucrose binding protein is expressed and localized in tissues actively engaged in sucrose transport.

Sucrose transport from the apoplasm, across the plasma membrane, and into the symplast is critical for growth and development in most plant species. Phloem loading, the process of transporting sucrose against a concentration gradient into the phloem, is an essential first step in long-distance transport of sucrose and carbon partitioning. We report here that a soybean 62-kD sucrose binding protein is associated with the plasma membrane of several cell types engaged in sucrose transport, including the mesophyll cells of young sink leaves, the companion cells of mature phloem, and the cells of the developing cotyledons. Furthermore, the temporal expression of the gene and the accumulation pattern of the protein closely parallel the rate of sucrose uptake in the cotyledon. Molecular cloning and sequence analysis of a full-length cDNA for this 62-kD sucrose binding protein indicated that the protein is not an invertase, contains a 29-amino acid leader peptide that is absent from the mature protein, and is not an integral membrane protein. We conclude that the 62-kD sucrose binding protein is involved in sucrose transport, but is not performing this function independently.

Substrate Specificity of the H-Sucrose Symporter on the Plasma Membrane of Sugar Beets (Beta vulgaris L.) : Transport of Phenylglucopyranosides.

Previous results (TJ Buckhout, Planta [1989] 178: 393-399) indicated that the structural specificity of the H(+)-sucrose symporter on the plasma membrane from sugar beet leaves (Beta vulgaris L.) was specific for the sucrose molecule. To better understand the structural features of the sucrose molecule involved in its recognition by the symport carrier, the inhibitory activity of a variety of phenylhexopyranosides on sucrose uptake was tested. Three competitive inhibitors of sucrose uptake were found, phenyl-alpha-d-glucopyranoside, phenyl-alpha-d-thioglucopyranoside, and phenyl-alpha-d-4-deoxythioglucopyranoside (PDTGP; K(i) = 67, 180, and 327 micromolar, respectively). The K(m) for sucrose uptake was approximately 500 micromolar. Like sucrose, phenyl-alpha-d-thioglucopyranoside and to a lesser extent, PDTGP induced alkalization of the external medium, which indicated that these derivatives bound to and were transported by the sucrose symporter. Phenyl-alpha-d-3-deoxy-3-fluorothioglucopyranoside, phenyl-alpha-d-4-deoxy-4-fluorothioglucopyranoside, and phenyl-alpha-d-thioallopyranoside only weakly but competively inhibited sucrose uptake with K(i) values ranging from 600 to 800 micromolar, and phenyl-alpha-d-thiomannopyranoside, phenyl-beta-d-glucopyranoside, and phenylethyl-beta-d-thiogalactopyranoside did not inhibit sucrose uptake. Thus, the hydroxyl groups of the fructose portion of sucrose were not involved in a specific interaction with the carrier protein because phenyl and thiophenyl derivatives of glucose inhibited sucrose uptake and, in the case of phenyl-alpha-d-thioglucopyranoside and PDTGP, were transported.

Localization of a protein, immunologically similar to a sucrose-binding protein from developing soybean cotyledons, on the plasma membrane of sieve-tube members of spinach leaves.

Immunocytochemical studies using antibodies raised against a 62-kDa membrane protein isolated from developing soybean (Glycine max (L.) Merr.) cotyledons were performed on leaf tissue of spinach (Spinacia oleracea L.). This 62-kDa protein was labeled by 6'-deoxy-6'-(4-azido-2-hydroxy)-benzamidosucrose (HABS), a photoaffinity sucrose analogue (K. G. Ripp et al., 1988, Plant Physiol.88, 1435-1445). Western-blot analysis of spinach plasma-membrane proteins indicated a cross-reactive polypeptide identical in molecular mass to that found in soybean. Indirect immunogold labeling of resin-embedded sections of fully expanded leaf tissue resulted in specific localization of colloidal gold on the sieve-tube plasma membrane. The label was uniform and, except for a few non-specific gold particles over the cell wall, all other cellular organelles and membrane systems were free of label. With the exception of occasional gold particles associated with the companion-cell plasma membrane, all other cell types of the leaf contained little or no label. Control sections treated with non-immune rabbit immunoglobulin-G were also essentially free of label. Immunogold labeling of young leaves, in which the phloem contained no mature sieve-tube members, were free of label for the 62-kDa protein. However, young leaf tissue in which mature or nearly mature sieve tubes could be identified, contained immunolabel associated with the sieve-tube plasma membranes. Similar results were obtained with mature leaf tissue of sugar beet (Beta vulgaris L.). The results of the immunocytochemical studies are consistent with the suggestion that the concentrating step in the phloem-loading process in this species may occur across the sieve-tube plasma membrane.

Identification of membrane protein associated with sucrose transport into cells of developing soybean cotyledons.

The photolyzable sucrose derivative 6'-deoxy-6'-(4-azido-2-hydroxy)-benzamidosucrose (6'-HABS), competitively inhibited the influx of [(14)C] sucrose into protoplasts from developing soybean (Glycine max L. Merr cv Wye) cotyledons. Photolysis of (125)I-labeled 6'-HABS in the presence of 10 millimolar dithiothreitol and microsomal preparations from developing soybean cotyledons led to label incorporation into a moderately abundant membrane protein with an apparent molecular mass of about 62 kilodalton (kD) by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The 62 kD protein was partially protected from labeling by the inclusion of 100 millimolar sucrose in the photolysis medium and also by the inclusion of 10 millimolar phenyl alpha-d-thioglucopyranoside. Glucose, raffinose, or phenyl alpha-d-3-deoxy-3-fluoroglucopyranoside did not afford even partial protection from labeling. When the photolyzable moiety of 6'-HABS was attached to 6-deoxy-6-aminoglucose and (125)I labeled, the resulting photoprobe did not label the 62 kD protein above background. The labeled protein at 62 kD is therefore apparently a specific, sucrose binding protein. Sucrose influx into cotlyedons of less than 25 milligrams fresh weight (approximately 10 days after flowering) occurred by passive processes, but metabolically dependent uptake became dominant over the next 5 to 7 days of development. Both the Coomassie staining protein at 62 kD and label incorporation at that position in analysis of membrane proteins appeared concomitant with the onset of active sucrose influx. Polyclonal antibodies to the purified 62 kD protein bound specifically to a protein in the plasmalemma of thin sections prepared from cotyledons and density stained with colloidal gold-protein A. The results suggest that the 62 kD membrane protein is associated with sucrose transport and may be the plasmalemma sucrose transporter.

Transport and Metabolism of a Sucrose Analog (1'-Fluorosucrose) into Zea mays L. Endosperm without Invertase Hydrolysis.

1'-Fluorosucrose (FS), a sucrose analog resistant to hydrolysis by invertase, was transported from husk leaves into maize (Zea mays L., Pioneer Hybrid 3320) kernels with the same magnitude and kinetics as sucrose. (14)C-Label from [(14)C]FS and [(14)C]sucrose in separate experiments was distributed similarly between the pedicel, endosperm, and embryo with time. FS passed through maternal tissue and was absorbed intact into the endosperm where it was metabolized and used in synthesis of sucrose and methanol-chloroform-water insolubles. Accumulation of [(14)C] sucrose from supplied [(14)C]glucosyl-FS indicated that the glucose moiety from the breakdown of sucrose (here FS), which normally occurs in the process of starch synthesis in maize endosperm, was available to the pool of substrates for resynthesis of sucrose. Uptake of FS into maize endosperm without hydrolysis suggests that despite the presence of invertase in maternal tissues and the hydrolysis of a large percentage of sucrose unloaded from the phloem, hexoses are not specifically needed for uptake into maize endosperm.

Contributions of sucrose synthase and invertase to the metabolism of sucrose in developing leaves : estimation by alternate substrate utilization.

The relative contributions of invertase and sucrose synthase to initial cleavage of phloem-imported sucrose was calculated for sink leaves of soybean (Glycine max L. Merr cv Wye) and sugar beet (Beta vulgaris L. monohybrid). Invertase from yeast hydrolyzed sucrose 4200 times faster than 1'-deoxy-1'-fluorosucrose (FS) while sucrose cleavage by sucrose synthase from developing soybean leaves proceeded only 3.6 times faster than cleavage of FS. [(14)C]Sucrose and [(14)C]FS, used as tracers of sucrose, were transported at identical rates to developing leaves through the phloem. The rate of label incorporation into insoluble products varied with leaf age from 3.4 to 8.0 times faster when [(14)C]sucrose was supplied than when [(14)C]FS was supplied. The discrimination in metabolism was related to enzymatic discriminations against FS to calculate the relative contributions of invertase and sucrose synthase to sucrose cleavage. In the youngest soybean leaves measured, 4% of final laminar length (FLL), all cleavage was by sucrose synthase. Invertase contribution to sucrose metabolism was 47% by 7.6% FLL, increased to 54% by 11% FLL, then declined to 42% for the remainder of the import phase. In sugar beet sink leaves at 30% FLL invertase contribution to sucrose metabolism was 58%.

Substrate recognition by a sucrose transporting protein.

Protoplasts derived from developing soybean cotyledons were used to study substrate recognition by a sucrose transporting protein in plant membranes. When used as alternate substrate inhibitors of [14C] sucrose influx, five different fructosyl-substituted sucrose derivatives, phenyl-alpha-D-glucopyranoside, and phenyl-alpha-D-thioglucopyranoside proved to bind effectively to the sucrose carrier-active site. These results are interpreted to indicate that a large portion of substrate recognition by this carrier may arise from the interaction of a relatively hydrophobic portion of the sucrose molecule and a hydrophobic region of the carrier protein binding site. Binding of phenyl-alpha-D-thioglucopyranosides in which various substitutions were made for the glucosyl hydroxyls shows that the glucosyl hydroxyls at positions 3, 4, and 6 are involved in substrate recognition by the carrier protein.

Transport and metabolism of 1'-fluorosucrose, a sucrose analog not subject to invertase hydrolysis.

The novel sucrose derivative 1'-fluorosucrose (alpha-d-glucopyranosyl-beta- d-1-deoxy-1-fluorofructofuranoside) was synthesized in order to help define mechanisms of sucrose entry into plant cells. Replacement of the 1'-hydroxyl by fluorine very greatly reduces invertase hydrolysis of the derivative (hydrolysis at 10 millimolar 1'-fluorosucrose is less than 2% that of sucrose) but does not reduce recognition, binding, or transport of 1'-fluorosucrose by a sucrose carrier. Transport characteristics of 1'-fluorosucrose were studied in three different tissues. The derivative is transported by the sucrose carrier in the plasmalemma of developing soybean cotyledon protoplasts with a higher affinity than sucrose (K(m) 1'-fluorosucrose 0.9 millimolar, K(m) sucrose 2.0 millimolar). 1'-Fluorosucrose is a competitive inhibitor of sucrose uptake with an apparent K(i) also of 0.9 millimolar, while the K(i) of sucrose competition of 1'-fluorosucrose uptake was 2.0 millimolar. Thus, both sugars are recognized at the same binding site in the plasmalemma. Both sucrose and 1'-fluorosucrose show very similar patterns of phloem translocation from an abraded leaf surface through the petiole indicating that recognition of 1'-fluorosucrose by sucrose carriers involved in phloem loading is likely as well.1'-Fluorosucrose is a very poor substrate for invertase and as such is absorbed only slowly by corn root segments, a tissue in which sucrose hydrolysis by a cell wall invertase is required prior to active hexose uptake.The kinetics of 1'-fluorosucrose uptake by soybean cotyledon protoplasts indicate that membrane passage and substrate release to the protoplast interior are rate limiting to transport. Recognition of sucrose at the inner membrane surface of the carrier protein is apparently different than recognition and binding at the external surface.

Sugar transport in isolated corn root protoplasts.

Isolated corn (Zea mays L.) root protoplasts were used to study sucrose and hexose uptake. It is found that glucose was preferentially taken up by the protoplasts over sucrose and other hexoses. Glucose uptake showed a biphasic dependence on external glucose concentration with saturable (K(m) of 7 millimolar) and linear components. In contrast, sucrose uptake only showed a linear kinetic curve. Sucrose and glucose uptake were linear over a minimum of 1 hour at pH 6.0 and 1 millimolar exogenous sugar concentration. Glucose uptake showed a sharp 42 degrees C temperature optimum, while sucrose uptake showed a lower temperature sensitivity which did not reach a maximum below 50 degrees C. Uptake of both sugars was sensitive to several metabolic inhibitors and external pH. Differences between sucrose and glucose uptake in two different sink tissue (i.e. protoplasts from corn roots and soybean cotyledons) are discussed.

Sugar Transport into Protoplasts Isolated from Developing Soybean Cotyledons : II. Sucrose Transport Kinetics, Selectivity, and Modeling Studies.

The effects of metabolic inhibitors, pH, and temperature on the kinetics of sucrose uptake protoplasts isolated from developing soybean Glycine max L. cv Wye cotyledons were studied. Structural requirements for substrate recognition by the sucrose carrier were examined by observing the effects of potential alternate substrates for the saturable component on sucrose uptake.Uptake by the three components (saturable, sulfhydryl reagent-sensitive nonsaturable, and diffusive) was calculated over a range of sucrose concentrations. The saturable component dominated uptake at external sucrose concentrations below 12 millimolar and was approximately equal to the nonsaturable and diffusive components at 44 and 22 millimolar external sucrose, respectively. The three uptake components showed different temperature sensitivities.Increasing external pH decreased both the linear component and the V(max) calculated for the saturable component. Conversely, increasing pH increased the calculated K(m) (sucrose) for the saturable component.Sucrose uptake by the saturable component was insensitive to several mono- and divalent cations. Competition for uptake of 0.5 millimolar sucrose by several sugars suggested that the beta-d-fructofuranoside bond and molecular size of sucrose were particularly important in sugar recognition by the saturable component carrier.

Sugar transport into protoplasts isolated from developing soybean cotyledons : I. Protoplast isolation and general characteristics of sugar transport.

A procedure is described to isolated functional protoplasts from developing soybean (Glycine max L. Merr. cv Wye) cotyledons. Studies of sucrose and hexose uptake into these protoplasts show that the plasmalemma of cotyledons during the stage of rapid seed growth contains a sucrose-specific carrier which is energetically and kinetically distinct from the system(s) involved in hexose transport. For example, sucrose, but not hexose uptake: (a) is inhibited by alkaline pH and the nonpermeant SH modifier, p-chloromercuribenzene sulfonic acid; (b) is stimulated by fusicoccin; (c) shows both a saturable and a linear component of uptake in response to substrate concentration; and (d) displays a sharp temperature response (high Q(10) value and high activation energies).

Crop productivity and photoassimilate partitioning.

The photosynthetic basis for increasing the yield of major field crops is examined in terms of improving the interception of seasonal solar radiation by crop foliage, the efficiency of conversion of intercepted light to photosynthetic assimilates, and the partitioning of photoassimilates to organs of economic interest. It is concluded that, in practice, genetic and chemical manipulation of light interception over the season and of partitioning offer the most potential for achieving further increases in yield. During the history of improvement of genetic yield potential of crops, increase in the partitioning of photoassimilates to harvested organs has been of primary importance.

Experimental determination of the respiration associated with soybean/rhizobium nitrogenase function, nodule maintenance, and total nodule nitrogen fixation.

The total metabolic cost of soybean (Glycine max L. Mer Clark) nodule nitrogen fixation was empirically separated into respiration associated with electron flow through nitrogenase and respiration associated with maintenance of nodule function.Rates of CO(2) evolution and H(2) evolution from intact, nodulated root systems under Ar:O(2) atmospheres decreased in parallel when plants were maintained in an extended dark period. While H(2) evolution approached zero after 36 hours of darkness at 22 degrees C, CO(2) evolution rate remained at 38 degrees of the rate measured in light. Of the remaining CO(2) evolution, 62% was estimated to originate from the nodules and represents a measure of nodule maintenance respiration. The nodule maintenance requirement was temperature dependent and was estimated at 79 and 137 micromoles CO(2) (per gram dry weight nodule) per hour at 22 degrees C and 30 degrees C, respectively.The cost of N(2) fixation in terms of CO(2) evolved per electron pair utilized by nitrogenase was estimated from the slope of H(2) evolution rate versus CO(2) evolution rate. The cost was 2 moles CO(2) evolved per mole H(2) evolved and was independent of temperature.In this symbiosis, nodule maintenance consumed 22% of total respiratory energy while the functioning of nitrogenase consumed a further 52%. The remaining respiratory energy was calculated to be associated with ammonia assimilation, transport of reduced N, and H(2) evolution.

Radiotracer evidence implicating phosphoryl and phosphatidyl bases as intermediates in betaine synthesis by water-stressed barley leaves.

In barley, glycine betaine is a metabolic end product accumulated by wilted leaves; betaine accumulation involves acceleration of de novo synthesis from serine, via ethanolamine, N-methylethanolamines, choline, and betaine aldehyde (Hanson, Scott 1980 Plant Physiol 66: 342-348). Because in animals and microorganisms the N-methylation of ethanolamine involves phosphatide intermediates, and because in barley, wilting markedly increases the rate of methylation of ethanolamine to choline, the labeling of phosphatides was followed after supplying [(14)C]ethanolamine to attached leaf blades of turgid and wilted barley plants. The kinetics of labeling of phosphatidylcholine and betaine showed that phosphatidylcholine became labeled 2.5-fold faster in wilted than in turgid leaves, and that after short incubations, phosphatidylcholine was always more heavily labeled than betaine. In pulse-chase experiments with wilted leaves, label from [(14)C]ethanolamine continued to accumulate in betaine as it was being lost from phosphatidylcholine. When [(14)C]monomethylethanolamine was supplied to wilted leaves, phosphatidylcholine was initially more heavily labeled than betaine. These results are qualitatively consistent with a precursor-to-product relationship between phosphatidylcholine and betaine.The following experiments, in which tracer amounts of [(14)C]ethanolamine or [(14)C]formate were supplied to wilted barley leaves, implicated phosphoryl and phosphatidyl bases as intermediates in the methylation steps between ethanolamine and phosphatidylcholine. Label from both [(14)C]ethanolamine and [(14)C]formate entered phosphorylmonomethylethanolamine and phosphorylcholine very rapidly; these phosphoryl bases were the most heavily labeled products at 15 to 30 minutes after label addition and lost label rapidly as the fed (14)C-labeled precursor was depleted. Phosphatidylmonomethylethanolamine and phosphatidylcholine were also significantly labeled from [(14)C]ethanolamine and [(14)C]formate at early times; the corresponding free bases and nucleotide bases were not. Addition of a trapping pool of phosphorylcholine reduced [(14)C]ethanolamine conversion to both phosphatidylcholine and betaine, and resulted in accumulation of label in the trap.A computer model of the synthesis of betaine via phosphatidylcholine was developed from (14)C kinetic data. The model indicates that about 20% of the total leaf phosphatidylcholine behaves as an intermediate in betaine biosynthesis and that a marked decrease (>/=2-fold) in the half-life of this metabolically active phosphatidylcholine fraction accompanies wilting. Dual labeling experiments with [(14)C]choline and [(3)H]glycerol confirmed that the half-life of the choline portion of phosphatidylcholine falls by a factor of about 2 in wilted leaves.

Oxygen and Carbon Dioxide Effects on the Pool Size of Some Photosynthetic and Photorespiratory Intermediates in Soybean (Glycine max [L.] Merr.).

The levels of ribulose 1,5-bisphosphate (RuBP), 3-phosphoglyceric acid (PGA), glycolate, glycine, and serine were measured in soybean leaflets during photosynthesis in atmospheres ranging from 1 to 60% O(2) and from 0 to 500 microliters per liter CO(2.)The RuBP level remained constant as CO(2) concentration was decreased in atmospheres containing 20 or 60% O(2), but increased as CO(2) concentration was decreased in atmospheres containing 1% O(2.) PGA levels decreased at CO(2) concentrations near or below the CO(2) compensation point under all O(2) concentrations. The glycolate pool at 300 microliters per liter CO(2) increased slightly with increasing O(2) concentration, but remained nearly constant at very low CO(2). The serine pool showed no measurable change over the range of CO(2) or O(2) concentrations tested. The glycine pool did not change significantly with varying CO(2) concentration but increased linearly with increasing O(2) concentration.Measured RuBP levels indicate an RuBP concentration less than the estimated concentration of RuBP carboxylase/oxygenase active sites. The constant RuBP pool size in 20% O(2), however, indicates that RuBP level does not limit photosynthesis or photorespiration any more at 50 microliters per liter CO(2) than at 450 microliters per liter.

Translocation and metabolism of glycine betaine by barley plants in relation to water stress.

The glycine betaine which accumulated in shoots of young barley plants (Hordeum vulgare L.) during an episode of water stress did not undergo net destruction upon relief of stress, but its distribution among plant organs changed. During stress, betaine accumulated primarily in mature leaves, whereas it was found mainly in young leaves after rewatering. Well-watered, stressed, and stressed-rewatered plants were supplied with [methyl-(14)C]betaine (8.5 nmol) via an abraded spot on the second leaf blade, and incubated for 3 d. In all three treatments the added (14)C migrated more or less extensively from the second leaf blade, but was recovered quantitatively from various plant organs in the form of betaine; no labeled degradation products were found in any organ. When 0.5 µmol of [methyl-(14)C]betaine was applied via an abraded spot to the second leaf blades of well-watered, mildly-stressed, and stressed-rewatered plants, (14)C was translocated out of the blades at velocities of about 0.2-0.3 cm/min which were similar to velocities found for applied [(14)C]sucrose. Heat-girdling of the sheath prevented export of [(14)C]betaine from the blade. When 0.5 µmol [(3)H]sucrose and 0.5 µmol [(14)C]betaine were suppled simultaneously to second leaf blades, the (3)H/(14)C ratio in the sheath tissue was the same as that of the supplied mixture. After supplying tracer [(14)C]betaine aldehyde (the immediate precursor of betaine) to the second leaf blade, the (14)C which was translocated into the sheath was in the form of betaine. Thus, betaine synthesized by mature leaves during stress behaves as an inert end product and upon rewatering is translocated to the expanding leaves, most probably via the phloem. Accordingly, it is suggested that the level of betaine in a barley plant might serve as a useful cumulative index of the water stress experienced during growth.