Perfluoroalkyl substances (PFASs) are a complex group of man-made chemicals with high stability and mobility leading to ubiquitous environmental contamination and accumulation in the food chain. In human serum/plasma samples, perfluorooctanoic acid (PFOA) and perfluorooctane sulfonate (PFOS) are the lead compounds. They are immunotoxic in experimental animals, and epidemiological studies provided evidence of a diminished production of vaccine antibodies in young children. However, information on children of the first year of age is missing but relevant, as they have a relatively high exposure if breastfed, and may have a higher susceptibility as their immune system is developing. In a cross-sectional study with 101 healthy 1-year-old children, internal levels of persistent organic pollutants and a broad panel of biological parameters were investigated at the end of the 1990s. Additional analysis of PFASs resulted in plasma levels (mean ± SD) of PFOA and PFOS of 3.8 ± 1.1 and 6.8 ± 3.4 µg/L, respectively, in the 21 formula-fed children, and of 16.8 ± 6.6 and 15.2 ± 6.9 µg/L in the 80 children exclusively breastfed for at least 4 months. The study revealed significant associations between levels of PFOA, but not of PFOS, and adjusted levels of vaccine antibodies against Haemophilus influenza type b (Hib, r = 0.32), tetanus (r = 0.25) and diphtheria (r = 0.23), with no observed adverse effect concentrations (NOAECs) determined by fitting a ‘knee’ function of 12.2, 16.9 and 16.2 µg/L, respectively. The effect size (means for PFOA quintiles Q1 vs. Q5) was quantified to be − 86, − 54 and − 53%, respectively. Furthermore, levels of PFOA were inversely associated with the interferon gamma (IFNɣ) production of ex-vivo lymphocytes after stimulation with tetanus and diphtheria toxoid, with an effect size of − 64 and − 59% (means Q1 vs. Q5), respectively. The study revealed no influence of PFOA and PFOS on infections during the first year of life and on levels of cholesterol. Our results confirmed the negative associations of PFAS levels and parameters of immune response observed in other epidemiological studies, with high consistency as well as comparable NOAECs and effects sizes for the three vaccine antibodies investigated, but for PFOA only. Due to reduction of background levels of PFASs during the last 20 years, children in Germany nowadays breastfed for a long duration are for the most part not expected to reach PFOA levels at the end of the breastfeeding period above the NOAECs determined.

The acute toxicity of cyanide is determined by its peak levels reached in the body. Compared to the ingestion of free cyanide, lower peak levels may be expected after consumption of foods containing cyanogenic glycosides with the same equivalent dose of cyanide. This is due to possible delayed and/or incomplete release of cyanide from the cyanogenic glycosides depending on many factors. Data on bioavailability of cyanide after consumption of foods containing high levels of cyanogenic glycosides as presented herein were necessary to allow a meaningful risk assessment for these foods. A crossover study was carried out in 12 healthy adults who consumed persipan paste (equivalent total cyanide: 68 mg/kg), linseed (220 mg/kg), bitter apricot kernels (about 3250 mg/kg), and fresh cassava roots (76–150 mg/kg), with each “meal” containing equivalents of 6.8 mg cyanide. Cyanide levels were determined in whole blood using a GC–MS method with K¹³C¹⁵N as internal standard. Mean levels of cyanide at the different time points were highest after consumption of cassava (15.4 µM, after 37.5 min) and bitter apricot kernels (14.3 µM, after 20 min), followed by linseed (5.7 µM, after 40 min) and 100 g persipan (1.3 µM, after 105 min). The double dose of 13.6 mg cyanide eaten with 200 g persipan paste resulted in a mean peak level of 2.9 µM (after 150 min). An acute reference dose of 0.075 mg/kg body weight was derived being valid for a single application/meal of cyanides or hydrocyanic acid as well as of unprocessed foods with cyanogenic glycosides also containing the accompanying intact β-glucosidase. For some of these foods, this approach may be overly conservative due to delayed release of cyanide, as demonstrated for linseed. In case of missing or inactivated β-glucosidase, the hazard potential is much lower.

Bile acids play an active role in fat metabolism and, in high-fat diets, elevated concentrations of fecal bile acids may be related to an increased risk of colorectal cancer. This study investigated concentrations of fecal and serum bile acids in 36 vegans and 36 omnivores. The reduced rank regression was used to identify dietary patterns associated with fecal bile acids. Dietary patterns were derived with secondary and conjugated fecal bile acids as response variables and 53 food groups as predictors. Vegans had higher fiber (p < 0.01) and lower fat (p = 0.0024) intake than omnivores. In serum, primary and glycine-conjugated bile acids were higher in vegans than in omnivores (p ≤ 0.01). All fecal bile acids were significantly lower in vegans compared to omnivores (p < 0.01). Processed meat, fried potatoes, fish, margarine, and coffee contributed most positively, whereas muesli most negatively to a dietary pattern that was directly associated with all fecal bile acids. According to the pattern, fat intake was positively and fiber intake was inversely correlated with bile acids. The findings contribute to the evidence that, in particular, animal products and fat may play a part in higher levels of fecal bile acids.

A vegan diet could impact microbiota composition and bacterial metabolites like short-chain (SCFA) and branched-chain fatty acids (BCFA). The aim of this study was to compare the concentrations of SCFA, BCFA, ammonia, and fecal pH between vegans and omnivores. In this cross-sectional study (vegans n = 36; omnivores n = 36), microbiota composition, fecal SCFA, BCFA, and ammonia concentrations and pH were analyzed in complete stool samples. A random forest regression (RFR) was used to identify bacteria predicting SCFA/BCFA concentrations in vegans and omnivores. No significant differences in SCFA and BCFA concentrations were observed between vegans and omnivores. Fecal pH (p = 0.005) and ammonia concentration (p = 0.01) were significantly lower in vegans than in omnivores, while fiber intake was higher (p < 0.0001). Shannon diversity was higher in omnivores compared to vegans on species level (p = 0.04) only. In vegans, a cluster of Faecalibacterium prausnitzii, Prevotella copri, Dialister spp., and Eubacterium spp. was predictive for SCFA and BCFA concentrations. In omnivores, Bacteroides spp., Clostridium spp., Ruminococcus spp., and Prevotella copri were predictive. Though SCFA and BCFA did not differ between vegans and omnivores, the results of the RFR suggest that bacterial functionality may be adapted to varying nutrient availability in these diets.