Genetic Engineering of Tetrahydrobenzylisoquinoline Alkaloid Biosynthetic Pathways

Attempts at engineering plant cell cultures and differentiated plants that produce isoquinoline alkaloids using cDNAs from tetrahydrobenzylisoquinoline-derived alkaloid biosynthesis have been hampered by the difficulties associated with establishing transformation and regeneration protocols for each new species. Just the same, there are selected successes to be reported.

One of the first introductions of an isoquinoline alkaloid biosynthetic gene into an isoquinoline alkaloid-producing plant was the Agrobacterium-mediated transformation of the C. japonica 9-omt under transcriptional of the CaMV 35S promoter into C. japonica cell cultures (Sato et al., 2001). Ectopic expression of 9-omt in a high berberine-producing cell culture resulted in a 15% increase in berberine and columbamine. Likewise, the same construct was introduced into E. californica seedling segments from which transgenic cell cultures were derived. E. californica produces benzo[c]phenanthridine alkaloids rather than berberine alkaloids. Introduction of the 9-omt cDNA resulted in the accumulation of columbamine (not normally present in E. californica) and a reduction in the level of the E. californica native alkaloid sanguinarine. Ectopic expression of 9-omt successfully introduced a new branch point in the E. californica isoquinoline alkaloid pathway and redirected (S)-scoulerine away from benzo[c]phenanthridine alkaloid biosynthesis into berberine alkaloid formation.

Root cultures of E. californica have been engineered with bbe1 and cyp80b1 under transcriptional control of the CaMV 35S promoter by Agrobacterium rhizogenes-mediated transformation (Park et al., 2002, 2003). Transgenic root cultures containing either antisense bbe1 or antisense cyp80b1 accumulated lower levels of benzo[c]phenanthridine alkaloids compared to controls, whereas transgenic root cultures that overexpressed bbe1 showed increased accumulation of benzo[c]phenanthridine alkaloids compared to controls. These types of experiments demonstrate that manipulation of alkaloid biosynthetic pathways is possible, but illustrate that intact transgenic plants are difficult to obtain.

There are several reports in the literature of the transformation and

FIGURE 10.10 Benzylisoquinoline alkaloid quantitation in latex from (A) untransformed P. somniferum and (B) transgenic anti-bbe1 T2 plants (Frick et al., 2004). The data are presented in percentages (µg alkaloid/100 µg soluble protein in latex). The values given in (A) are the mean of individual analyses of 29 plants.

regeneration of P. somniferum. The earliest reports of regeneration of P. somniferum plants from cell culture preceeded the transformation work (Wakhlu and Bajwa, 1986). Transformation of P. somniferum with A. rhizogenes appeared shortly thereafter (Williams and Ellis, 1993; Yoshimatsu and Shimomura, 1992). The first report of Agrobacterium-mediated transformation of P. somniferum cell suspension cultures in which an Arabidopsis thaliana sam1 trangene was introduced occurred in 1997 (Belny et al., 1997). The first report of Agrobacterium-mediated transformation with subsequent regeneration of P. somniferum appeared in 1999 (Larkin et al., 1999). A second report appeared in 2000 (Park and Facchini, 2000). Attempts at metabolic engineering of alkaloid biosynthesis in P. somniferum are currently being made. The known genes of tetrahydrobenzylisoquinoline-derived alkaloid formation in P. somniferum have been reintroduced into a highly inbred, elite Tasmanian variety in the sense and antisense orientation. Several hundred transgenic plants have been produced by Agrobacterium-mediated transformation (S. Frick, P. J. Larkin, and T. M. Kutchan, unpublished data). The first set of transgenic plants for which the analyses are complete are those that contained the bbe1 gene in an antisense orientation under transcriptional control of the S4S4 promoter (Frick et al., 2004). These experiments were designed to reduce flow of the central intermediate (S)-reticuline into the sanguinarine pathway. A complex picture of ratios of tetrahydrobenzylisoquinoline intermediate alkaloids emerged from these experiments, but morphinan levels did not increase (Fig. 10.10). Preliminary analyses of plants that contain various other transgenes indicate that it is, however, possible to manipulate morphine levels. P. somniferum has tremendous potential for alkaloid engineering because the plant is the commercial source of pharmaceutically important morphinan alkaloids and the transformation and regeneration, although still a slow process, has been established.