To the editor:

Your editorial in the February issue (Nat. Biotechnol. 22, 13, 2004) expressed concerns about the use of conventional crops, mainly maize, for the production of biopharmaceuticals. With the current capacity crunch for production of biopharmaceuticals by current-day technologies, and the projected demand for therapeutic proteins (such as antibodies) in the next few years1,2, it is imperative that the acceptance of plants for molecular farming proceed smoothly.

Throughout the debate on the use of plants as production platforms for pharmaceutical proteins, we have been puzzled by the persistent and exclusive focus on the use of existing crop varieties. This focus has, predictably, triggered and fueled concerns about this exciting new agricultural opportunity from stake-holding industry groups3,4. However, those participating in the debate might take note of the fact that existing commercial crop varieties are unlikely to be optimal for plant-manufactured pharmaceutical applications anyway, having been bred and selected entirely for their present-day, traditional uses. This is true of both food (e.g., corn) and non-food (e.g., tobacco) crops that might be used in such applications.

Fortunately, however, it may be quite unnecessary to turn to alternatives, such as undomesticated plants or rarely cultivated crops, which might require new development of appropriate gene expression technology and so on. We contend that modern plant breeding approaches can be used to develop novel and distinctive cultivars of the desired (conventional) crops, which will be optimally suited to biopharmaceutical applications in every respect. Together with appropriate crop-management protocols, the resulting crop system will completely obviate any compromise or contamination of the equivalent traditional crop.

To illustrate this approach with tobacco5,6,7, which is the subject of several plant manufactured pharmaceutical platforms, we note that the US tobacco industry has voiced a zero-tolerance standard for genetically modified (GM) contamination of the conventional crop (Dean Wallace, Council for Burley Tobacco, personal communication). Transgene escape can be ameliorated through the use of male sterile varieties8, although the proximity of fertile, non-GM tobacco nearby could result in the production of some seed and the potential for GM volunteer plants. However, we believe that through an F1 hybrid strategy, we can take advantage of the many benefits of tobacco as a platform for producing pharmaceuticals while at the same time eliminating its drawbacks. We have found that interspecific crosses between tobacco and certain other species of Nicotiana produce vigorous, completely sterile hybrid plants that are morphologically distinct from conventional tobacco (thus achieving identity preservation) and can be planted and grown in a manner similar to conventional tobacco (though at a higher plant density) using existing infrastructure (float beds, machinery, fertilizers, herbicides, etc.), yet can be mechanically harvested three or four times during the growing season. For production of pharmaceutical or industrial proteins, the maternal (tobacco) cultivar is transformed with a gene of interest, and homozygous transformants with high-level protein expression are selected and crossed with the other species. It is the resulting F1 hybrid offspring that are grown in the field, as with hybrid maize. Our present work involves breeding a tobacco cultivar that is optimized for plant-manufactured pharmaceutical applications, and we are in the process of evaluating the performance of our hybrids with the appropriate gene expression systems.

We feel that all parties participating in the debate over biopharmaceutical crops will benefit from consideration of developing plant varieties specifically tailored to plant manufacture of pharmaceuticals, which will permit the use of existing transformation and gene-expression technologies without imposing undue risk to conventional crops or the environment. The hybrid Nicotiana approach effectively addresses the main concerns about 'third-generation' GM crops9 and could very well lead to simplified regulatory requirements in the near future.