<jats:title>Abstract</jats:title><jats:p>1,2,4-Butanetriol (BT) is a valuable chemical with extensive applications in many different fields. The traditional chemical routes to synthesize BT suffer from many drawbacks, e.g., harsh reaction conditions, multiple steps and poor selectivity, limiting its industrial production. In this study, an engineered <jats:italic>Escherichia coli</jats:italic> strain was constructed to produce BT from xylose, which is a major component of the lignocellulosic biomass. Through the coexpression of a xylose dehydrogenase (<jats:italic>CCxylB</jats:italic>) and a xylonolactonase (<jats:italic>xylC</jats:italic>) from <jats:italic>Caulobacter crescentus</jats:italic>, native <jats:italic>E. coli</jats:italic> xylonate dehydratase (<jats:italic>yjhG</jats:italic>), a 2-keto acid decarboxylase from <jats:italic>Pseudomonas putida</jats:italic> (<jats:italic>mdlC</jats:italic>) and native <jats:italic>E. coli</jats:italic> aldehyde reductase (<jats:italic>adhP</jats:italic>) in <jats:italic>E. coli</jats:italic> BL21 star(DE3), the recombinant strain could efficiently convert xylose to BT. Furthermore, the competitive pathway responsible for xylose metabolism in <jats:italic>E. coli</jats:italic> was blocked by disrupting two genes (<jats:italic>xylA</jats:italic> and <jats:italic>EcxylB</jats:italic>) encoding xylose isomerase and xyloluse kinase. Under fed-batch conditions, the engineered strain BL21ΔxylAB/pE-mdlCxylBC&pA-adhPyjhG produced up to 3.92 g/L of BT from 20 g/L of xylose, corresponding to a molar yield of 27.7%. These results suggest that the engineered <jats:italic>E. coli</jats:italic> has a promising prospect for the large-scale production of BT.