Foreman et al. [36] used oligonucleotide microarrays

(including 5,131 ESTs) to study the transcriptional regulation of biomass-degrading eFT-508 enzymes from T. reesei, a Trichoderma sp. of significance in the cellulose industry. In another study, the transcriptome of T. atroviride was analyzed using spotted microarrays (1,438 cDNA clones) but again not for the purpose of biocontrol [37]. The analysis reported here is based in a HDO microarray carrying probe sets representative of a total of 23,202 gene transcripts from thirteen Trichoderma strains, including 3,826 EST-based transcripts of the T. harzianum CECT 2413 biocontrol strain (Figure 1). Despite the redundant nature of EST libraries, a substantial representation of the T. harzianum CECT 2413 transcriptome

can be expected from the probe sets included on the HDO microarray for this strain, considering that already sequenced Trichoderma genomes have been estimated to contain 9,129-11,643 predicted genes [21, 22, 38]. Moreover, as shown in this work probe sets on the microarray designed from transcripts of Trichoderma BI 10773 strains other than T. harzianum CECT 2413 were also useful for obtaining information about gene expression in our strain. In particular, we found that nearly half of the probe sets revealing significant expression changes after hybridization with cDNA from T. harzianum CECT 2413 (strain T34) derived from other strains or species of Trichoderma. The fact that genes known to respond rapidly and sharply to chitin, including AG-881 in vitro those encoding the proteases PRA1, PRA2, PRB1 and PRB2 and the endochitinase these CHIT42 [26, 39], yielded the expected expression patterns, and that a homologue of the SM1 gene with demonstrated expression in the first stages of T. virens-root interactions [29] was also detected in our T. harzianum-root interaction system, provide

a high level of confidence that the microarrays identify differentially expressed genes. We are convinced that at present the Trichoderma HDO microarray proposed here offers the opportunity for extensive analyses of gene expression in Trichoderma strains whose whole genomes are not scheduled to be sequenced soon, such as those of T. harzianum, T. asperellum or T. viride. An improved microarray may now be possible for T. virens and T. atroviride, thanks to the release of their genome sequences and the availability of higher-density microarrays that ensure the coverage of complete genomes. For example, gene expression profiling based on entire genome tiling arrays will afford the possibility of monitoring the expression level of whole transcriptomes, avoiding the cloning biases of ESTs and allowing the data arising from different transcript variants that may not have been previously known or predicted to be distinguished. Furthermore, the introduction of new emerging technologies such as massive-scale RNA sequencing will in the near future enable us to overcome some of the limitations inherent to microarray-technology [40].