Identification of Burkholderia gladioli pv. cocovenenans in Black Fungus and Efficient Recognition of Bongkrekic Acid and Toxoflavin Producing Phenotype by Back Propagation Neural Network

Abstract

Burkholderia gladioli pv. cocovenenans is a serious safety issue in black fungus due to the deadly toxin, bongkrekic acid. This has triggered the demand for an efficient toxigenic phenotype recognition method. The objective of this study is to develop an efficient method for the recognition of toxin-producing B. gladioli strains. The potential of multilocus sequence typing and a back propagation neural network for the recognition of toxigenic B. cocovenenans was explored for the first time. The virulent strains were isolated from a black fungus cultivation environment in Qinba Mountain area, Shaanxi, China. A comprehensive evaluation of toxigenic capability of 26 isolates were conducted using Ultra Performance Liquid Chromatography for determination of bongkrekic acid and toxoflavin production in different culturing conditions and foods. The isolates produced bongkrekic acid in the range of 0.05-6.24 mg/L in black fungus and a highly toxin-producing strain generated 201.86 mg/L bongkrekic acid and 45.26 mg/L toxoflavin in co-cultivation with Rhizopus oryzae on PDA medium. Multilocus sequence typing phylogeny (MLST) analysis showed that housekeeping gene sequences have a certain relationship with a strain toxigenic phenotype. We developed a well-trained, back-propagation neutral network for prediction of toxigenic phenotype in B. gladioli based on MLST sequences with an accuracy of 100% in the training set and an accuracy of 86.7% in external test set strains. The BP neutral network offers a highly efficient approach to predict toxigenic phenotype of strains and contributes to hazard detection and safety surveillance.

Keywords: B. gladioli; MLST; black fungus; bongkrekic acid; neutral network; toxoflavin.

Figure 1

Correspondence of MLST phylogenetic tree based on concatenated sequences of 7 MLST housekeeping genes with a toxigenic phenotype heatmap. The tree was constructed with 26 isolates and 3 standard strains using the neighborhood joining method with a bootstrap value of 1000. The colors of strains were divided according to different STs. The heatmap on the right represents the production of bongkrekic acid (BA) and toxoflavin. CCs at the TLV threshold have been indicated, with CC1-3 being the main and CC4-5 being the minor.

Figure 2

Bongkrekic acid production in soaked black fungus and foods. (a) The measurement of four isolated and two standard strains for the production of bongkrekic acid in black fungus soaked for 72 h. (b) The measurement of NC18, NC22, Co14, and DSM 11318 for the production of bongkrekic acid in tremella, rice noodles, chow fun, and fresh black fungus foods. *** indicates diferences exist between strains and foods when p < 0.001.

Figure 3

The TLV level geoBURST analysis. The red dots represent strains from black fungus, the green dots represent environmental sources, and the yellow-brown dots represent sources of CF patients (Supplementary Files for details). The solid lines represent only one allele difference and the dotted lines represent two or three allele differences. The distance between STs has nothing to do with the length of the lines.

Figure 4

BP neural network prediction of toxigenic phenotype. (a) The BP neural network model with the two-layer structure of 2811−15−4; (b) the classification accuracy of the training set (I), validation set (II), test set (III) and overall data set (IV) of this model; (c) the ROC curve of the model on training set (I), validation set (II), test set (III) and overall data (IV), respectively; (d) external data validation results, (+) is toxin positive and (−) is toxin negative, the length of the color block serves as an indicator of the probabilistic assessment for each classification.