In plants, the shikimate pathway occurs in the plastid and leads to the biosynthesis of aromatic amino acids. The bifunctional 3-dehydroquinate dehydratase/shikimate dehydrogenase (DHD/SHD) catalyses the conversion of dehydroquinate into shikimate. Expression of NtDHD/SHD was suppressed by RNAi in transgenic tobacco plants. Transgenic lines with <40% of wild-type activity displayed severe growth retardation and reduced content of aromatic amino acids and downstream products such as cholorogenic acid and lignin. Dehydroquinate, the substrate of the enzyme, accumulated. However, unexpectedly, so did the product, shikimate. To exclude that this finding is due to developmental differences between wild-type and transgenic plants, the RNAi approach was additionally carried out using a chemically inducible promoter. This approach revealed that the accumulation of shikimate was a direct effect of the reduced activity of NtDHD/SHD with a gradual accumulation of both dehydroquinate and shikimate following induction of gene silencing. As an explanation for these findings the existence of a parallel extra-plastidic shikimate pathway into which dehydroquinate is diverted is proposed. Consistent with this notion was the identification of a second DHD/SHD gene in tobacco (NtDHD/SHD-2) that lacked a plastidic targeting sequence. Expression of an NtDHD/SHD-2-GFP fusion revealed that the NtDHD/SHD-2 protein is exclusively cytosolic and is capable of shikimate biosynthesis. However, given the fact that this cytosolic shikimate synthesis cannot complement loss of the plastidial pathway it appears likely that the role of the cytosolic DHD/SHD in vivo is different from that of the plastidial enzyme. These data are discussed in the context of current models of plant intermediary metabolism.