Progress 01/26/01 to 07/24/04

Outputs
1. What major problem or issue is being resolved and how are you resolving it (summarize project aims and objectives)? How serious is the problem? What does it matter? Aflatoxins are highly toxic chemicals produced by the fungi Aspergillus flavus and A. parasiticus. As such, they are strictly regulated by both the FDA and foreign countries importing American food products. They contaminate a vast array of agricultural products and have impacted agricultural and economic policies worldwide. The objectives of this project are: (1) Identify characteristics of nuts which contain aflatoxin and identify those process streams in which such contaminated nuts occur; (2) Develop and demonstrate, in field conditions, real-time, economically feasible sorting methods to remove such nuts from process streams; and (3) Establish sample and single nut distributions in order to develop sampling theories and to discover the source of aflatoxin. This program falls within Component 4.1 (Toxic Chemicals: Mycotoxins) of NP108 (Food Safety) and focuses primarily on problem areas 1.1.1.2 (Develop methodology for toxin detection) and 1.1.1.3 (Develop sampling protocols). 2. List the milestones (indicators of progress) from your Project Plan. Year 1 (FY 2000) Completion of selection of training set for peanut sorting. Year 2 (FY 2001) Theory of subsampling. Comparison of European slurry method for pistachios. Distribution functions for insect infested almonds with sampling and risk calculations. Selection of features for peanut sorting. Year 3 (FY 2002) Distribution functions for insect infested walnuts with sampling and risk calculations. (Possible)Distribution function for Iranian pistachios with sampling and risk calculations. Comparison of European slurry method for almonds and walnuts. Year 4 (FY 2003) Construction of a bench-top peanut sorter (not real-time). Distribution functions for insect infested peanuts including the effects of storage, comparison with risks for negative binomial distribution, and calculation of sampling and risk. Year 5 (FY 2004) Construction of a demo peanut sorter. Study of possible chemical absorption of aflatoxin on nut matrices and ist affect on chemical analysis and mammal absorption. 3. Milestones: A. List the milestones that were scheduled to be addressed in FY2004. How many milestones did you fully or substantially meet in FY 2004 and indicate which ones were not fully met, briefly explain why not, and your plans to do so. The following are the milestones listed for FY 2004 in our project plan: Construction of a demo peanut sorter. Study of possible chemical absorption of aflatoxin on nut matrices and its affect on chemical analysis and mammal absorption. The milestones listed for FY2004 were not accomplished. Selection of NIR features as a basis for a real-time sorting device for aflatoxin in peanuts were inconclusive, so plans to build the sorter were abandoned. Chemical absorption studies planned were not conducted as the Principal Investigator for this project has retired. B. List the milestones that you expect to address over the next three years. As stated above, the lead scientist for this project has retired. Furthermore, the project has reached the end of its five year plan and has been redirected for FY 2005. 4. What were the most significant accomplishments this past year? A. Determining the single kernel aflatoxin distribution for peanuts is an important element in the understanding of proper sampling methodologies. Research conducted in the Plant Myotoxins Research Unit in Albany, CA. has led to the establishment of a single kernel aflatoxin distribution for peanuts, essentially confirming the widely reported negative binomial distribution. This is the first experimental determination of the single kernel distribution in peanuts and the first test of the negative binomial distribution as it relates to aflatoxin content in peanuts. This information will be useful in determining the single kernel aflatoxin distribution for peanuts. B. None. C. None. D. None 5. Describe the major accomplishments over the life of the project, including their predicted or actual impact. A sampling plan was developed for pistachios that reduces the amount of nuts and testing effort required to characterize a lot for aflatoxin. Use of this sampling method results in the same precision for aflatoxin determination while reducing the sample size from the current 110 lbs of nuts to 23 lbs. Adoption of this method will save the industry approximately $1,000,000 per year in sample and test costs. This sequential sampling test method for pistachios has been taught to the California Pistachio Commission, which is recommending it to the industry as a standard method. A new sampling plan is now being developed for the Calif. Pistachio Comm. in light of the newly approved FMO. (Mycotoxins 1. 1.1.3) A rapid, non-destructive method has been developed to determine the aflatoxin content in peanuts. This method has allowed the creation of a set of peanuts of known aflatoxin content which is in use for training and testing of a sorting device to remove peanuts contaminated with aflatoxin from the process stream in real-time. (Mycotoxins 1.1.1.2) The aflatoxin distribution of comestible peanuts in three representative, highly contaminated, dry farming lots of Florunner peanuts has been established and shown to be similar to that found in pistachio nuts but at a tenfold higher level. The form of the distribution was found to be of a more general form than the negative binomial, used until now. This distribution will help determine the parameters required for the peanut sorting device under development, will be used in future sampling studies, and should enable calculations of rejection risk over more extensive ranges. (Mycotoxins 1.1.1.3) Sampling of aflatoxin levels in walnuts (Serr cultivar) has demonstrated that, contrary to popular opinion, there is a contamination problem in that product, especially when the nuts are subject to insect infestation. The research indicates that aflatoxin levels are low in light colored meats, but dark colored meats are susceptible to extremely high levels. It is expected that the industry will implement color sorting of walnuts to obtain a safer product. (Mycotoxins 1.1.1.3) A relation between insect damage and aflatoxin content in almonds has been quantified. It has been determined that only feeding damage and gross damage result in contamination, and that the risk of lot rejection by the European Union can be kept below 5% if such damage is kept below 1- 3%, depending on cultivar. (Mycotoxins 1.1.1.3) A theory relating sub-sampling variance to particle size has been developed. Tests show agreement for dry grinding, as practiced in the U.S. In the case of wet slurry grinding, as practiced in the European Union, additional variance reduction appears to be due to extraction during grinding. This variance is now being included in data analysis in the industry. (Mycotoxins 1.1.1.3) 65,000 peanuts were tested using the non-destructive dipping method, yielding roughly 120 contaminated nuts of known aflatoxin content. These nuts are being used in testing spectroscopic methods to detect aflatoxin contamination in peanuts. An algorithm based on the results of these tests will allow detection of aflatoxin in peanuts in a real-time sorting device. (Mycotoxins 1.1.1.3) It has been determined that the wet slurry method of testing for aflatoxin contamination in pistachio nuts yields a 32% higher result than the dry grinding method used in the U.S. and is easier to apply. It is expected that U.S. labs will adopt the wet slurry method and as a consequence aflatoxin concentration limits will, in effect, be reduced by 32%. (Mycotoxins 1.1.1.3) The single kernel aflatoxin distribution for peanuts was established. It was possible to show that the widely reported negative binomial distribution matched part, but not all, of this distribution. This is the first experimental determination of the single kernel distribution in peanuts and the first test of the negative binomial distribution as it relates to aflatoxin content in peanuts. 6. What science and/or technologies have been transferred and to whom? When is the science and/or technology likely to become available to the end- user (industry, farmer, other scientists)? What are the constraints, if known, to the adoption and durability of the technology products? None for FY2004, although there has been significant transfer of technology to the industry over the life of the project as described in question 5 above.

Impacts
(N/A)

Publications

• Schatzki, T.F., Toyofuku, N. 2003. Sample preparation and presampling of pistachios. Journal of Agriculture and Food Chemistry. 51(20):6068-6072.
• Schatzki, T., Toyofuku, N. 2004. Sampling and sample preparation of pistachios. In: Barug, D., Van Egmond, H.P., Lopez-Garcia, R., Van Osenbruggen, W.A., Visconti, A. Meeting the Mycotoxin Menace. Netherlands: Wageningen Academic Publishers. p. 221-235.

Progress 10/01/02 to 09/30/03

Outputs
1. What major problem or issue is being resolved and how are you resolving it? Tree and ground nuts, like many food products, can become infected with molds, either before harvest or during storage. In nuts, some of these molds produce a mycotoxin, aflatoxin, which is among the most potent liver carcinogens known. When present, aflatoxin is contained in a very few nuts, but these infected nuts carry individually a very high level of toxin. Sampling lots for such contamination is a serious problem as it is statistically quite likely that good lots are rejected and bad ones accepted, simply due to sampling errors. We have developed methods to characterize these aflatoxin distributions. Using the distributions we can estimate the likelihood of sampling errors. However, very large samples (hundreds of pounds) may be required with standard methods. Another, more desirable, solution to the problem is to sort the nuts before final shipment, removing the contaminated ones. We are developing automatic sorting methods that remove contaminated nuts from the process stream. 2. How serious is the problem? Why does it matter? In the case of tree nuts, contamination is relatively rare, although not negligible, and is primarily limited to pistachios among U.S. crops. The average level among peanuts is considerably higher, particularly during drought years, although generally still not dangerous. There is a serious market resistance to aflatoxin containing foods, particularly in countries with strong consumer movements, such as Japan and Europe. As a result, the impact on overseas markets is important and is increasing rapidly. 3. How does it relate to the National Program(s) and National Program Component(s) to which it has been assigned? National Research Program 108 (100%) states: Mycotoxins. The presence and potential for the presence of mycotoxins in crops is not only a direct food safety problem, but it threatens the competitiveness of United States agriculture in the world market. Major goals are to control aflatoxin in peanuts, corn, cottonseed, nuts and figs; and reduce insect damage leading to fungal infection. 4. What were the most significant accomplishments this past year? A. Aflatoxin levels in pistachio nuts are a major concern for producers exporting to Europe, and a common method of testing between U.S. and E.U. labs is needed. Researchers at Western Regional Research Center (WRRC) compared two different methods (dry ground and water slurry) for measuring aflatoxin levels in samples of pistachio nuts, and a theory for the variance, based solely on non-adjustable parameters involving the distribution of kernel content, particle size, and sample size was developed. It was determined that wet slurry grinding (the E.U. method now in use) yields a 32% higher aflatoxin level than dry grinding (the classical U.S. method), and the slurry method is easier to apply. It is expected that U.S. labs will switch to slurry grinding and as a consequence aflatoxin concentration limits will, in effect, be reduced by 32%. 5. Describe the major accomplishments over the life of the project, including their predicted or actual impact. Over the nine years of the current and preceding projects, we have established methods to ascertain how much and where aflatoxin occurs in the process streams of almonds, pistachios and peanuts. We have investigated testing methods. We have developed sorting equipment which can reduce the aflatoxin level in pistachios below measurable levels. The pistachio industry has followed our suggestions for improved processing and is now installing these sorters. We have found methods of detecting insects or insect damage within nuts which, at least in the case of almonds, should reduce aflatoxin levels in that commodity as well. A rapid laboratory method of sorting peanuts for high aflatoxin content has been developed and been applied to develop a training set of infected peanuts as a step in developing a peanut sorter. 6. What do you expect to accomplish, year by year, over the next 3 years? FY 2004 Complete spectroscopic work on peanut sorting. Begin construction of a prototype peanut sorter. FY 2005 Complete a prototype sorter and demonstrate it at industrial shelling plants to transfer technology. FY 2006 Apply technology to other commodities where appropriate. 7. What science and/or technologies have been transferred and to whom? When is the science and/or technology likely to become available to the end- user (industry, farmer, other scientists)? What are the constraints, if known, to the adoption and durability of the technology products? Pistachio sorting for aflatoxin based on visible imaging, has been licensed and installations in pistachio processing plants are expected to increase. We are told that major pistachio processors are operating their plants guided by the distribution and sorting papers we have published. A highly improved sorter to remove non-split pistachios during processing has been licensed and is being constructed commercially. It is already in use at least one processor. In all these areas, market and cost constraints operate. We expect that U.S. producers of pistachio nuts will adopt the slurry method for aflatoxin measurements.

Impacts
(N/A)

Publications

• Kim, S.W., Schatzki, T.F. Detection of Pinholes in Almonds. Transactions of the American Society for Agricultural Engineers. 2001. v. 44(4). p. 997- 1003.
• Schatzki, T.F., Ong, M.S. Dependence of Aflatoxin in Almonds on the Type and Amount of Insect Damage. Journal Agricultural and Food Chemistry. 2001. v. 49(4) p. 4513-4519.
• Schatzki, T.F., Haddon, W.F. Rapid, Non-Destructive Selection of Peanuts for High Aflatoxin Content by Soaking and Tandem Mass Spectrometry. Journal Agricultural and Food Chemistry. 2002. v. 50(10). p. 3062-3069.

Progress 10/01/01 to 09/30/02

Outputs
1. What major problem or issue is being resolved and how are you resolving it? Tree and ground nuts, like many food products, can become infected with molds, either before harvest or during storage. In nuts, some of these molds produce a mycotoxin, aflatoxin, which is among the most potent liver carcinogens known. When present, aflatoxin is contained in a very few nuts, but these infected nuts carry individually a very high level of toxin. Sampling lots for such contamination is a serious problem as it is statistically quite likely that good lots are rejected and bad ones accepted, simply due to sampling errors. We have developed methods to characterize these aflatoxin distributions. Using the distributions we can estimate the likelihood of sampling errors. However, very large samples (hundreds of pounds) may be required with standard methods. Another, more desirable, solution to the problem is to sort the nuts before final shipment, removing the contaminated ones. We are developing automatic sorting methods that remove contaminated nuts from the process stream. 2. How serious is the problem? Why does it matter? In the case of tree nuts, contamination is relatively rare, although not negligible, and is primarily limited to pistachios among U.S. crops. The average level among peanuts is considerably higher, particularly during drought years, although generally still not dangerous. There is a serious market resistance to aflatoxin containing foods, particularly in countries with strong consumer movements, such as Japan and Europe. As a result, the impact on overseas markets is important and is increasing rapidly. 3. How does it relate to the national Program(s) and National Program Component(s) to which it has been assigned? National Research Program 108 states: Mycotoxins. The presence and potential for the presence of mycotoxins in crops is not only a direct food safety problem, but it threatens the competitiveness of United States agriculture in the world market. Major goals are to control aflatoxin in peanuts, corn, cottonseed, nuts and figs; and reduce insect damage leading to fungal infection. 4. What was your most significant accomplishment this past year? A. Single Most Significant Accomplishment During FY 2002: A reliable and tested aflatoxin distribution among single nuts in a lot is required to develop sampling distributions for any size, estimate a statistical risk of rejection and evaluate hazard, if any, to human populations. At WRRC (Albany, CA), with the cooperation of the NPRL (Dawson, GA) in acquiring lots, we measured the (200-nut) sample distribution of three sub-lots of dry-farmed Flo-runner peanuts. Using theory previously developed by us, we obtained the individual nuts distributions in each sub-lot and found them to be similar, but higher, than those in tree nuts. It is expected that more reliable sample distributions and risks can be calculated, following the pioneering work of T. Whitaker (Rayleigh, NC). B. Next Most Significant Accomplishment During FY 2002: Early work using a supply of intact peanuts with estimated aflatoxin content is needed to create a machine sorting algorithm. Spectroscopic work (NIR and fluorescence) indicates that a set of aflatoxin levels will be required to fit a continuous, rather than binary, sorting algorithm. Using the dipping method for non-destructive testing, described in the previous annual report, an additional 20000 peanuts were tested at WRRC (Albany, CA), yielding an additional 1000 contaminated peanuts of known aflatoxin content. These peanuts are now in used for testing spectroscopic methods of detecting aflatoxin contamination. C. None. D. None. 5. Describe your major accomplishments over the life of the project, including their predicted or actual impact? Over the nine years of the current and preceding projects, we have established methods to ascertain how much and where aflatoxin occurs in the process streams of almonds, pistachios and peanuts. We have investigated testing methods. We have developed sorting equipment which can reduce the aflatoxin level in pistachios below measurable levels. The pistachio industry has followed our suggestions for improved processing and is now installing these sorters. We have found methods of detecting insects or insect damage within nuts which, at least in the case of almonds, should reduce aflatoxin levels in that commodity as well. A rapid laboratory method of sorting peanuts for high aflatoxin content has been developed and been applied to develop a training set of infected peanuts as a step in developing a peanut sorter. 6. What do you expect to accomplish, year by year, over the next 3 years? FY 2002 Continue spectroscopic work on peanut sorting. FY 2003 Complete spectroscopic work on peanut sorting. Begin construction of a prototype peanut sorter. FY 2004 Complete a prototype sorter and demonstrate it at industrial shelling plants to transfer technology. 7. What technologies have been transferred and to whom? When is the technology likely to become available to the end user (industry, farmer other scientist)? What are the constraints, if known, to the adoption durability of the technology? Pistachio sorting for aflatoxin based on visible imaging, has been licensed and installations in pistachio processing plants are expected to increase. We are told that major pistachio processors are operating their plants guided by the distribution and sorting papers we have published. A highly improved sorter to remove non-split pistachios during processing has been licensed and is being constructed commercially. It is already in use at least one processor. In all these areas, market and cost constraints operate. 8. List your most important publications and presentations, and articles written about your work (NOTE: this does not replace your review publications which are listed below) Schatzki, T.F., Toyofuku, N. Sub-sample Preparation of Pistachios. Proc. of the 14th Aflatoxin Elimination Conference, Phoenix, AZ. 2001. USDA-ARS. Schatzki, T.F., Ong, M.S. Dependence of Aflatoxin in Almonds on the Type and Amount of Insect Damage. Proc. of the 14th Aflatoxin Elimination Conference, Phoenix, AZ. 2001. USDA-ARS. Schatzki, T.F., Toyofuku, N. Sub-sample Preparation of Pistachios", 6th International Symposium on Fruit, Nut and Vegetable Engineering, Potsdam, Germany. 2001. Institute fuer Agrartechnik Bornim e.V.

Impacts
(N/A)

Publications

• Kim, S.W., Schatzki, T.F. Detection of Pinholes in Almonds. Transactions of the American Society for Agricultural Engineers. 2001. v. 44(4). p. 997- 1003.
• Schatzki, T.F., Ong, M.S. Dependence of Aflatoxin in Almonds on the Type and Amount of Insect Damage. Journal Agricultural and Food Chemistry. 2001. v. 49(4) p. 4513-4519.
• Schatzki, T.F., Haddon, W.F. Rapid, Non-Destructive Selection of Peanuts for High Aflatoxin Content by Soaking and Tandem Mass Spectrometry. Journal Agricultural and Food Chemistry. 2002. v. 50(10). p. 3062-3069.