A 28-day oral toxicity study of echimidine and lasiocarpine in Wistar rats.

Pyrrolizidine alkaloids (PAs) are a class of naturally-occurring plant toxins. Echimidine is one of the predominant PAs found in honeys produced in Australia and New Zealand. There is a lack of information on the oral toxicity of echimidine on which to base regulatory decisions concerning the risk to humans of these honeys. This GLP study was conducted to assess the subchronic dietary toxicity of echimidine to rats compared to that of lasiocarpine as a positive control. Wistar rats, 10/sex, were fed diets containing 0, 0.6, 1.2 or 2.5 mg/kg bw echimidine. Positive control groups, 10/sex, were fed diets containing 0.6, 1.2 or 2.5 mg/kg bw lasiocarpine. Neither PA had any effect on survival, food consumption, clinical signs, gross lesions, or histopathology. Consumption of lasiocarpine, but not echimidine, decreased bodyweight gain in males at ≥ 1.2 mg/kg bw, and in females at 2.5 mg/kg bw. Slight alterations in white cell counts and serum ALT concentrations at 2.5 mg/kg bw of both PAs were not clinically significant, had no histological correlates, and were considered to be of equivocal relevance. In conclusion, the subchronic No Observed Adverse Effect Level (NOAEL) for echimidine is 2.5 mg/kg bw/day, whereas, on the basis of a treatment-related decrease in bodyweight gain in males at 1.2 mg/kg bodyweight, the NOAEL for lasiocarpine is 0.6 mg/kg bw/day.

Gastric mucosal irritation following oral exposure to sodium metabisulphite: A reproducible effect?

Sulphiting agents, such as sodium metabisulphite (SM), are used in food as bleaching agents and to prevent browning reactions. A 1972 repeat dose study in rats found that dietary sulphites caused irritation of the stomach with inflammation, hyperplasia and bleeding. We conducted a 7-day dietary study in rats to confirm that stomach lesions were the most sensitive toxicological endpoint. Rat feed was prepared daily with 0%, 0.25%, 0.5%, 1% or 4% (w/w) SM. Parameters included clinical signs, feed and water intake, bodyweight gain, haematology, serum protein chemistry, necropsy findings and gastrointestinal histopathology. There were no treatment-related clinical signs or gastrointestinal lesions. Mean bodyweight gain was markedly decreased in the 4% (w/w) SM group although feed consumption was marginally depressed. Slightly lower mean values for RBC, Hb, Hct, total WBC and lymphocyte count were observed in the 4% SM group with no evidence of compensatory haematopoiesis. The gastric lesions in rats observed in a 1972 study of dietary SM for 10-56 days could not be replicated. These findings create uncertainty around the most relevant toxicological endpoint to establish a suitable health based guidance value, which can only be overcome if a robust long-term dietary study is undertaken.

Human pharmacokinetic study of tutin in honey; a plant-derived neurotoxin.

Over the last 150 years a number of people in New Zealand have been incapacitated, hospitalised, or died from eating honey contaminated with tutin, a plant-derived neurotoxin. A feature of the most recent poisoning incident in 2008 was the large variability in the onset time of clinical signs and symptoms of toxicity (0.5-17 h). To investigate the basis of this variability a pharmacokinetic study was undertaken in which 6 healthy males received a single oral dose of tutin-containing honey giving a tutin dose of 1.8 µg/kg body weight. The serum concentration-time curve for all volunteers exhibited two discrete peaks with the second and higher level occurring at approximately 15 h post-dose. Two subjects reported mild, transient headache at a time post-dose corresponding to maximum tutin concentrations. There were no other signs or symptoms typical of tutin intoxication such as nausea, vomiting, dizziness or seizures. Pharmacokinetic analysis using a two-site absorption model resulted in a good fit to the observed concentration data. A novel analytical method subsequently revealed the presence of glycoside conjugates of tutin in addition to unconjugated tutin in honey. These pharmacokinetic data will be important to better define a safe maximum tutin concentration in honey.

Endogenous allergens in the regulatory assessment of genetically engineered crops.

A scientific approach to the assessment of foods derived from genetically engineered (GE) crops is critical to maintaining objectivity and public confidence in regulatory decisions. Principles developed at the international level support regulators and enable robust and transparent safety assessments. A comparison of key constituents in the GE crop with a suitable comparator is an important element of an assessment. In Europe, endogenous allergens would be included in the comparative analysis, however this approach has been hindered by technical limitations on the ability to accurately measure identified allergenic proteins. Over recent years, improved proteomic methods have enabled researchers to focus on major allergenic proteins in conventional food crops, as information on natural variability is largely lacking. Emerging data for soybean indicate that variability in levels of major allergens already in the food supply is broad. This raises questions about the biological interpretation of differences between a GE plant and its conventional counterpart, in particular, whether any conclusions about altered allergenicity could be inferred. This paper discusses the scientific justification for requiring proteomic analysis of endogenous allergens as part of the evaluation. Ongoing scientific review and corresponding international discussion are integral to ensuring that data requirements address legitimate risk assessment questions.

Health risk assessment for cyanobacterial toxins in seafood.

Cyanobacteria (blue-green algae) are abundant in fresh, brackish and marine waters worldwide. When toxins produced by cyanobacteria are present in the aquatic environment, seafood harvested from these waters may present a health hazard to consumers. Toxicity hazards from seafood have been internationally recognised when the source is from marine algae (dinoflagellates and diatoms), but to date few risk assessments for cyanobacterial toxins in seafood have been presented. This paper estimates risk from seafood contaminated by cyanobacterial toxins, and provides guidelines for safe human consumption.

A refined approach to estimate exposure for use in calculating the Maximum Residue Limit of veterinary drugs.

Maximum Residue Limits (MRLs) are standards that represent the maximum residue concentration expected to be found if a veterinary drug is administered according to good practice in the use of veterinary drugs (GVP). MRLs are established only where the exposure to residues in food resulting from particular use patterns of the veterinary drug pass a public health risk assessment. The current model diet as used by major regulators overstates mean consumption of food for populations when compared to results from food surveys of actual consumption. Exposure to residues is overestimated when calculating long-term (chronic) exposure using the model diet leading to the risk to consumers being overstated. Additionally the model diet underestimates the size of large portions eaten by the group of consumers that eat large quantities of a particular food in a single meal potentially leading to understating of risks associated with exposure to residues of drugs that produce an adverse effect after a single exposure. A revision of dietary consumption figures is proposed that will better match the consumption figures used in point-estimates of dietary exposure to the timeframe for consumption that is relevant to the reference dose.