In the work presented here, the use of aqueous-organic, two-liquid-phase systems was tested for the production of secondary metabolites with plant cell cultures. The cells used were hairy root cultures obtained by the transformation of Tagetes patula with Agrobacterium rhizogenes. Tagetes produces thiophenes, heterocyclic sulfurouscompounds with biocidal activity. First, suitable organic solvents were selected with growth and secondary metabolite production as biological criteria. Physical criteria were the partition coefficient for the thiophenes and the density difference with medium. From the nine solvents tested, only those with a log P value higher than 5 proved to be suitable. FC40 and hexadecane were the solvents of choice for further experiments in two-liquid-phase bioreactors. Three fermenter types were tested, i.e. the liquid-impelled loop reactor, the stirred tank reactor, and the bubble column. The liquid-impelled loop reactor was blocked by fast-growing hairy root cells after 1 week. Immobilization of hairy roots in calcium alginate beads was not an adequate solution to this problem. In the two other reactor types, the stirred-tank reactor and the bubble column, several runs of about 1 month were carried out. In the stirred-tank reactor, the cells showed a higher volumetric growth rate than in the bubble column. The growth rate of the cells was higher for FC40 as the dispersed phase than for hexadecane. When thiophene production is also taken into account, the best results were obtained with the bubble column and hexadecane as the dispersed phase. The linear growth rate in this system was 0.18 g l-1 day-1, with a total thiophene production at the end of the run of 465-mu-mol in a total volume of 6 dm3. When hexadecane was used as the organic phase, the excretion of thiophenes was 30%-70% of the total thiophenes produced,in contrast to one-phase systems where the excretion is less than 1% with these cells. With FC40, the excretion was 10%-20%. The distribution of the various thiophenes inside and outside the cells was not the same, indicating different excretion patterns for the various thiophenes.
