Within the shikimate pathway the bifunctional enzyme 3-dehydroquinate dehydratase / shikimate dehydrogenase (DHD/SHD) catalyzes the reversible conversion of dehydroquinate into shikimate. Despite its central function within the shikimate pathway, little is known about its biochemical characteristics or its in planta importance. To elucidate the in planta function of DHD/SHD enzyme, enzyme activity was strongly decreased by RNAi (RNA interference) in transgenic tobacco. Plants with suppressed DHD/SHD activity below 40 % of the wild type level displayed severe growth retardation. Secondary metabolites (chlorogenic acid and lignin) and aromatic amino acids decreased in transgenic plants. Surprisingly, silencing of DHD/SHD enzyme resulted in an accumulation of dehydroquinate (substrate) and shikimate (product) in transgenic plants. To investigate kinetics of shikimate and dehydroquinate during the silencing of DHD/SHD, an ethanol inducible silencing construct alc-DHD/SHD-RNAi was created. The induced silencing of DHD/SHD led to an early accumulation of dehydroquinate and a late accumulation of shikimate in the leaves of transgenic plants. This result strongly suggested that the accumulation of dehydroquinate is a direct consequence of the silencing of DHD/SHD, whereas the buildup of shikimate is derived from secondary effects. This leads us to put forward a model to interpret the buildup of shikimate in transgenic plants. We assume that a certain amount of dehydroquinate would be transported from chloroplasts to the cytosol, and be converted to shikimate by a cytosolic isoform of DHD/SHD. The cytosolic shikimate most likely is not efficiently re-imported into chloroplasts, thus further metabolism is impaired. The isolation of a full-size cDNA encoding a putative DHD/SHD isozyme (Nt-DHD/SHD-2) supports this hypothesis. Lacking a typical chloroplast transit peptides, Nt-DHD/SHD-2 was judged to be a cytosolic enzyme. Transient expression using Agrobacterium infiltration confirmed the enzyme activity and cytosolic localization of Nt-DHD/SHD-2. PEP is one of the substrates of the shikimate pathway. Because most plastids have little or no PGM (phosphoglycerate mutase) and enolase activity, 3-phosphoglycerate (3-PGA) cannot be metabolized to PEP in plastids. Plastids rely on the supply of cytosolic PEP via a shuttle mechanism. By introducing PGM and enolase into chloroplasts, we created a plastic PEP biosynthetic pathway. Under the subsequent catalyses of PGM and enolase, plastidic 3-PGA was converted to PEP in situ. As a result, PEP and pyruvate content increased substantially, whereas 3-PGA levels decreased dramatically in leaves of transgenic tobacco plants. This genetic manipulation led to many detrimental consequences in transgenic plants, including retarded growth, repressed photosynthesis and reduced production of carbohydrates. Unexpectedly, the elevated supply of PEP to plastids did not significantly enhance carbon flux into the shikimate pathway and secondary metabolism.