Genome-scale investigation of phenotypically distinct but nearly clonalTrichodermastrains

Abstract

Biological control agents (BCA) are beneficial organisms that are applied to protect plants from pests. Many fungi of the genusTrichodermaare successful BCAs but the underlying mechanisms are not yet fully understood.Trichoderma cf. atroviridestrain LU132 is a remarkably effective BCA compared toT. cf. atroviridestrain LU140 but these strains were found to be highly similar at the DNA sequence level. This unusual combination of phenotypic variability and high DNA sequence similarity between separately isolated strains prompted us to undertake a genome comparison study in order to identify DNA polymorphisms. We further investigated if the polymorphisms had functional effects on the phenotypes. The two strains were clearly identified as individuals, exhibiting different growth rates, conidiation and metabolism. Superior pathogen control demonstrated by LU132 depended on its faster growth, which is a prerequisite for successful distribution and competition. Genome sequencing identified only one non-synonymous single nucleotide polymorphism (SNP) between the strains. Based on this SNP, we successfully designed and validated an RFLP protocol that can be used to differentiate LU132 from LU140 and otherTrichodermastrains. This SNP changed the amino acid sequence of SERF, encoded by the previously undescribed single copy gene “small EDRK-rich factor” (serf). A deletion ofserfin the two strains did not lead to identical phenotypes, suggesting that, in addition to the single functional SNP between the nearly clonalTrichoderma cf. atroviridestrains, other non-genomic factors contribute to their phenotypic variation. This finding is significant as it shows that genomics is an extremely useful but not exhaustive tool for the study of biocontrol complexity and for strain typing.