The article by Lin et al. ( 1 ) has been retracted at the request of the authors. The strain of bacteria reported in the article was Lactobacillus reuteri GMN-32. However, the authors have contacted the British Journal of Nutrition to state that in fact the strain of bacteria used was Lactobacillus paracasei-32. Because of the fundamental nature of this error, the findings cannot be interpreted in the manner stated in the original article, and therefore this paper has been retracted. The authors apologise for this error.