Genetic Engineering of Monoterpenoid Indole Alkaloid Biosynthetic Pathways

Early attempts at engineering plants with cDNAs from monoterpenoid indole alkaloid pathways were limited to those plant species for which transformation and regeneration protocols were available. Later attempts have been made with C. roseus cell cultures. Although C. roseus can be transformed with Agrobacterium, there are not yet reports of regeneration of rooted plants. Selected examples of each are discussed in the following.

One of the first introductions of a monoterpenoid indole alkaloid cDNA was tdc from C. roseus into tobacco (Songstad et al., 1990, 1991). Transgene expression was driven by the CaMV 35S promoter. Young, fully expanded leaves of the transgenic tobacco plants had up to 45 times greater tryptophan decarboxylase activity than did control plants. Tryptamine accumulation in the plants was proportional to the tryptophan decarboxylase specific activity in protein extracts. Transgenic plants accumuated up to 1 mg tryptamine per gram fresh weight, but were unaffected in their indole-3-acetic acid levels.

Transgenic tobacco plants containing the C. roseus tdc gene and accumulating the protoalkaloid tryptamine have been analyzed for their effects on insect feeding (Thomas et al., 1995). An advantage of this type of system is that it is possible to test plants in parallel that differ only in their protoalkaloid content. In addition, tryptamine was found in phloem extracts. The sweet potato whitefly Bemisia tabaci was allowed to feed and reproduce on transgenic and control tobacco plants. The sweet potato whitefly, like the aphid, pierces phloem cells with a stylet, thereby obtaining nutrients from the vascular system of the host plant. Sweet potato whitefly pupae emergence was reduced by as much as 97% in transgenic plants as compared to control plants. The mechanism of this action of tryptamine on whitefly reproduction is not yet understood, but poses an interesting possible use for tryptamine in insect control.

Tobacco plants expressing tdc from C. roseus, tydc (tyrosine decarboxylase) from Papaver somniferum or both transgenes were analyzed for effects on metabolism and visible phenotype (Guillet et al., 2000). Expression of these transgenes tdc and tydc should create an artificial metabolic sink for the aromatic amino acids tryptophan and tyrosine, respectively. The results obtained are complex, but may be summarized as follows. Reduction of the tryptophan and tyrosine pools affected the phenylalanine pool in a light-dependent manner. Also perturbed were pool sizes of the nonaromatic amino acids methionine, valine, and leucine. This depletion of amino acids correlated with increases in activity of enzymes of the shikimate and phenylpropanoid pathways in older, light-treated seedlings. In addition, nicotine and chlorogenic acid levels were also increased.

Expression of the C. roseus tdc cDNA in Brassica napus (canola) created an artificial sink for tryptophan and resulted in reduced levels of indole glucosinolates (Chavadej et al., 1994). Transgene expression driven by the CaMV 35S promoter accumulated tryptamine while lower levels of tryptophan-derived indole glucosinolates accumulated in all plant parts compared to control plants. Significant to this particular study, seeds from transgenic plants contained as little as 3% indole glucosinate compared to control plant seeds. This potentially yielded a more palatable protein fodder that is produced after extraction of the oil from seed material.

The final nonnative plant example considered here is expression of the C. roseus tdc cDNA in potato tuber under transcriptional control of the CaMV 35S promoter (Yao et al., 1995). Again, an artificial metabolic sink was created for tryptophan. Transgene expression altered the balance of substrate and product pools in the shikimate and phenylpropanoid pathways. Transgenic tubers accumulated tryptamine and contained decreased levels of tryptophan, phenylalanine, and phenylalanine-derived phenolic compounds compared to control tubers. Wound-induced accumulation of chlorogenic acid was reduced as was the accumulation of soluble and cell wall-bound phenolics. The transgenic tubers were also more susceptible to infection by Phytophthora infestans possibly due to the modified cell wall.

Attempts at engineering monoterpenoid indole alkaloid profiles in C. roseus have also recently been made. str1 and/or tdc cDNAs were introduced into leaves from 6- to 8-week-old C. roseus seedlings by Agrobacterium tumefaciens-mediated transformation (Canel et al., 1998). Transgenic cell cultures showed that CaMV 35S-driven str1 demonstrated tenfold higher strictosidine synthase activity than untransformed cultured cells and higher levels of strictosidine, ajmalicine, catharanthine, serpentine, and tabersonine. Alkaloid production in these cell lines was, however, found to be unstable. Overexpression of tdc was apparently not necessary for alkaloid overproduction, but rather was detrimental to normal growth of the transgenic cultures. Feeding of a transgenic cell line of C. roseus that contained an str1 transgene with the monoterpenoid indole alkaloid biosynthetic precursors tryptamine and/or loganin indicated that utilization of tryptamine for alkaloid biosynthesis increases flux through the indole pathway (Whitmer et al., 1998). Overexpression of tdc is not necessary for high rates of tryptamine biosynthesis, and addition of tryptamine to the cultures did not increase overall levels of alkaloid. Addition of either loganin or loganin and tryptamine to the cell cultures resulted in increased accumulation of monoterpenoid indole alkaloids, suggesting that flux through the secoiridoid biosynthetic pathway is rate limiting. Rate limitation of the secologanin rather than the tryptamine pathway has also been demonstrated with C. roseus hairy root cultures (Morgan and Shanks, 2000). Methodologies for controlled expression in C. roseus using a glucocorticoid-inducible promoter system (Hughes et al., 2002) as well as models for monitoring metabolic flux in plant metabolism using HPLC and NMR spectroscopy have been developed (Morgan and Shanks, 2002; Rijhwani et al., 1999).

As a final point for consideration, transcription factors could also prove useful tools for engineering plant alkaloid accumulation where gene transcription is rate limiting. Analysis of the promoter regions of the C. roseus tdc and str1 genes led to the identification of cis elements and trans factors that affect gene expression (Menke et al., 1999; Van Der Fits and Memelink, 2000, 2001). Ectopic expression of one of these trans factor-encoding cDNAs, ORCA3, in cultured cells of C. roseus increased expression of tdc, str1, and d4h, but did not lead to an increase in accumulation of monoterpenoid indole alkaloids. Although ORCA3 acted pleiotropically on the transcription of several alkaloid biosynthetic genes, it did not activate all of the biosynthetic genes indicating that additional regulatory factors could be involved. Methods for identifying alkaloid gene regulatory factors and their potential use in metabolic engineering of alkaloid pathways have been reviewed and will not be discussed further here (Gantet and Memelink, 2002; Van Der Fits et al., 2001).