Cloning of the fructan biosynthesis pathway of Jerusalem artichoke.

To study the regulation of fructan synthesis in plants, we isolated two full-size cDNA clones encoding the two enzymes responsible for fructan biosynthesis in Jerusalem artichoke (Helianthus tuberosus): 1-sucrose:sucrose fructosyl transferase (1-SST) and 1-fructan:fructan fructosyl transferase (1-FFT). Both enzymes have recently been purified to homogeneity from Jerusalem artichoke tubers (Koops and Jonker (1994) J.Exp.Bot.45, 1623-1631; Koops and Jonker (1996) Plant Physiol. 110, 1167-1175) and their amino acid sequences have been partially determined. Using RT-PCR and primers based on these sequences, specific fragments of the genes were amplified from tubers of Jerusalem artichoke. These fragments were used as probes to isolate the cDNAs encoding 1-SST and 1-FFT from a tuber-specific lambdal ZAP library. The deduced amino acid sequences of both cDNAs perfectly matched the sequences of the corresponding purified proteins. At the amino acid level, the cDNA sequences showed 61% homology to each other and 59% homology to tomato vacuolar invertase. Based on characteristics of the deduced amino acid sequence, the first 150 bp of both genes encode a putative vacuolar targeting signal. Southern blot hybridization revealed that both 1-SST and 1-FFT are likely to be encoded by single-copy genes. Expression studies based on RNA blot analysis showed organ-specific and developmental expression of both genes in growing tubers. Lower expression was detected in flowers and in stem. In other organs, including leaf, roots and dormant tubers, no expression could be detected. In tubers, the spatial and developmental expression correlates with the accumulation of fructans. Using the 1-sst and 1-fft cDNAs, chimeric genes were constructed driven by the CaMV 35S promoter. Analysis of transgenic petunia plants carrying these constructs showed that both cDNAs encode functional fructosyltransferase enzymes. Plants transformed with the 35S-1-sst construct accumulated the oligofructans 1-kestose (GF2), 1,1-nystose (GF3) and 1,1,1-fructosylnystose (GF4). Plants transformed with the 35S-1-fft construct did not accumulate fructans, probably because of the absence of suitable substrates for 1-FFT, i.e. fructans with a degree of polymerization > or = 3 (GF2, GF3, etc.). Nevertheless, protein extracts from these transgenic plants were able to convert GF3, when added as a substrate into fructans with a higher degree of polymerization. Progeny of crosses between a 35S-1-sst-containing plant and a 35S-1-fft-containing plant, showed accumulation of high-molecular-weight fructans in old, senescent leaves. Based on the comparison of the predicted amino acid sequences of 1-sst and 1-fft with those of other plant fructosyl transferase genes, we postulate that both plant fructan genes have evolved from plant invertase genes.

High level fructan accumulation in a transgenic sugar beet.

We have transformed sugar beet into a crop that produces fructans. The gene encoding 1-sucrose:sucrose fructosyl transferase (1-SST), which was isolated from Helianthus tuberosus, was introduced into sugar beet. In H. tuberosus, 1-SST mediates the first steps in fructan synthesis through the conversion of sucrose (GF) into low molecular weight fructans GF2, GF3, and GF4. In the taproot of sugar beet transformed with the 1-sst gene, the stored sucrose is almost totally converted into low molecular weight fructans. In contrast, 1-sst expression in the leaves resulted in only low levels of fructans. Despite the storage carbohydrate having been altered, the expression of the 1-sst gene did not have any visible effect on phenotype and did not affect the growth rate of the taproot as observed under greenhouse conditions.

Purification and Characterization of the Enzymes of Fructan Biosynthesis in Tubers of Helianthus tuberosus Colombia (II. Purification of Sucrose:Sucrose 1-Fructosyltransferase and Reconstitution of Fructan Synthesis in Vitro with Purified Sucrose:Sucrose 1-Fructosyltransferase and Fructan:Fructan 1-Fructosyltransferase).

Sucrose:sucrose 1-fructosyltransferase (1-SST), an enzyme involved in fructan biosynthesis, was purified to homogeneity from tubers of Helianthus tuberosus that were harvested in the accumulation phase. Gel filtration under native conditions predicted a molecular mass of about 67 kD. Electrophoresis or gel filtration under denaturing conditions yielded a 27- and a 55-kD fragment. 1-SST preferentially catalyzed the conversion of sucrose into the trisaccharide 1-kestose (GF2). Other reactions catalyzed by 1-SST at a lower rate were self-transfructosylations with GF2 and 1,1-nystose (GF3) as substrates yielding GF3 and 1,1,1-fructosylnystose, respectively, as products. 1-SST also catalyzed the removal of the terminal fructosyl unit from both GF2 and GF3, which resulted in the release of sucrose and GF2, respectively, and free Fru. The purified enzyme did not display [beta]-fructosidase activity. An enzyme mixture of purified 1-SST and fructan:fructan 1-fructosyltransferase, both isolated from tubers, was able to synthesize fructans up to a degree of polymerization of at least 13 with sucrose as a sole substrate.

Uptake of [(14)C]sucrose in isolated minor-vein networks of Commelina benghalensis L.

Maceration with pectinase (4.5h) of Commelina benghalensis L. leaves stripped at either side yielded isolated vein networks consisting of four to five secondary veins and tertiary cross veins (=minor veins). Examination with Evans Blue and injection of Fluorescein F showed that 80% of the veins were viable. Proof of normal functioning of isolated minor veins was that [(14)C]sucrose fed to an apical vein network attached to the remaining intact part of the leaf was absorbed and finally arrived in the petiole. Sucrose uptake by veins obeyed Michaelis-Menten kinetics (K m 5·10(-4) mol l(-1); V max (light) 3.2 µmol h(-1) g(-1) fresh weight, V max (dark) 1.5 µmol h(-1) g(-1) fresh weight). A linear component, not inhibited by carbonylcyanide m-chlorophenylhydrazone and p-chloromercuribenzenesulfonic acid, was present. Maximal uptake took place at 5 mmol l(-1) K(+); concentrations of K(+) higher than 10 mmol l(-1) decreased the rate of uptake. The uptake rates by isolated veins and veins in situ (in disks) were in the same order of magnitude. Altogether, isolated veins promise to be a useful system for the study of loading.