Comment submitted by J Leach, President

Comments of the American Phytopathological Society on Docket EPA-HQ-OPP- 2006-0643 - Exemption from the Requirement of a Tolerance under the Federal Food, Drug, and Cosmetic Act for Residues of Plant Virus Coat Proteins that are Part of a Plant-Incorporated Protectant (PVC-Proteins); Supplemental Proposal By: Jan Leach, President, The American Phytopathological Society Jan.Leach@ColoState.edu (970) 491-2924 I am pleased to submit these comments on behalf of the American Phytopathological Society (APS), the world?s premier plant pathology society of about 5000 scientists working toward the establishing and maintain the health of plants. The proposal to exempt PVCP-Proteins from FFDCA is of great interest to our members. Although we have had the opportunity to discuss this with only a few of our members because our annual meeting is still 10 days away, the Public Policy Board and the leadership of APS are supportive of this proposal. APS has long followed the efforts of government agencies, especially EPA and APHIS, to provide a regulatory framework so that these virus-resistant plants can be used to the benefit of food industries without risk to the human and animal consumers of plants. The basic premise for the exemption from the requirement for tolerance of residues of PVCP-Proteins is that plant virus coat proteins in naturally infected plants have been a part of the diet of plant-consuming plants and animals forever. No untoward effects of consuming virus-infected plants on human health have been observed - a conclusion that we believe to be valid and are pleased that EPA has used in this proposal. Plant Incorporated Protectants that are derived from entire coat protein sequences and function as residues of coat protein are believed to have structural limitations on the extent of modification of the molecule to be effective as a protectant for the specific virus. Thus any effective construct will likely be ?virtually unmodified?, because it has to have the property of self-assembly and interaction with the viral nucleic acid to be functional. Apparently in the SAP discussions, addition or deletions of 1-2 amino acids at the termini were permissible within the definition of ?virtually unmodified?, and we concur with this interpretation. We favor, however, the suggestion that EPA remove the limitation of which and how many amino acids can be added or truncated from either end and still be considered ?virtually unmodified?. Based on ultrastructural and protein modeling analyses, these added or deleted amino acids would be located on the outside of the virus particles and would not interfere with properties of aggregation and self-assembly, or blockage of disassembly. Thus any protein residues in PVCP-Protein plants would be expected to be in the same structural form as plant virus coat proteins in naturally infected plants, except for virus particles. From this, the conclusion can be drawn that there should be no change in safety to consumers. Other classes of potential PVC-protein modifications, such as change in amino acid sequences known to function as a vector binding site, should produce a coat protein as safe for consumption as an unmodified plant virus coat protein, and a coat protein that would have greater environmental safety in the case of a recombinational event. This type of change is with the natural variation limit of the PVC protein, because non-transmissible isolates are found in naturally infected plants. Such changes have been encouraged to decrease the risk of producing a new, variant virus by a recombinational event. It would be hard to predict categorically, however, how many and what types of changes would be relevant to product safety. There are simply not the data on which a prediction could be based, with the possible exception of predicting the structure of the modified protein. Thus we concur with the conclusions of SAP, that it is appropriate to compare to natural variation limits on a case-by-case basis. The question was raised whether exposure to PVC-proteins would be changed relative to natural virus infections if the source PVC-PIP were a virus that did not naturally infect that plant species. As long as the source virus infected a plant normally consumed, we do not believe there should be a difference. To be functional against a virus, the contributing PVC-protein is likely to be a virus of the same family, such as Potyviridae, in which there is a great conservation of both core coat protein sequence and structure. The level of expression of an allergenic protein is an important component of allergenicity, but we do not think it should be an important component of whether a PVC-protein has allergenic activity. In all cases known, the level of coat protein expressed in a natural infection is orders of magnitude greater than in a PVC- protein expressed from a plant genome, whether it is virtually unmodified or minimally modified. The proposal extensively discusses the options for viewing a PVC-protein that does not meet the virtually unmodified definition, and how thoroughly a protein should be searched for in plants if it were to be produced from a construct designed to express no protein. In this case, protection is by PTGS in which the nucleotide sequence is the pesticidal substance in FIFRA terms, not the protein. If PTGS is suppressed by a viral infection or by viral sequences known to be silencing suppressors, protection from the target virus infection fails. This would not, however, be expected to result in expression of protein from the PVC insert. The natural virus infection would be expected to produce a protein with amino acid sequences similar to the predicted amino acid sequences of the PVC-protein, so that it may be indistinguishable from the PVC-protein, further complicating the assays and the issue. Production of the putative PVC-protein in plants would be a difficult and costly task, and it is not clear what analysis by mass spectroscopy or glycosylation would reveal in terms of safety, in comparison to the proteins of the natural infection. In the absence of any evidence of allergenicity or other risk of these viral proteins, in silico comparisons based on deduced amino acid sequence and predicted structure are likely to give adequate results. As additional plants with resistance to virus are reviewed by EPA, and the safety record accumulates, this information will be valuable to future determinations of the extent of minimal modifications that can be made. Molecular evidence to date demonstrates that nucleic acids mediating resistance do not produce proteins, and that proteins are not the pesticidal substance. We encourage exemption from tolerance for these proteins, as they are unlikely to be present in these virus- resistant plants.