Selection and experimental evaluation of universal primers to study the fungal microbiome of higher plants

The impact of primer choice on results of metabarcoding studies was experimentally evaluated by analyzing fungal communities associated with leaves of four plant species. Significant differences in target specificity of primers were highlighted by a percentage of plant reads ranging from almost nothing to 30 to 35% of the total detected sequences. Overall, primer sets targeting the internal transcribed spacer 1 (ITS1) region proved to be more specific than those targeting the ITS2 region. A comparable taxa coverage was revealed for all investigated primer sets. However, each primer set detected only around 50% of the overall detected taxa highlighting that a consistent part of the actual fungal diversity remains undetected in studies conducted using a single couple of primers. The coverage was increased to 70 to 80% by combining results from two different primer sets. Some fungal taxa were preferentially or exclusively detected by certain primer sets and this association between primers and taxa was generally recurrent on several plant hosts. Data highlighted that a perfect set of primers to investigate the whole fungal diversity does not exist and that whatever the choice, only a fraction of the actual microbial diversity will be investigated. However, provided information may be valuable to select the best primers according to the objective of the analysis.