The induction of cellulases by cellulose, an insoluble polymer, in the filamentous fungus Trichoderma reesei is puzzling. We previously proposed a mechanism that is based on the presence of low levels of cellulase in the uninduced fungus; this basal cellulase activity would digest cellulose-releasing oligosaccharides that could enter the cell and trigger expression of cellulases. We now present experiments that lend further support to this model. We show here that transcripts of two members of the cellulase system, cbh1 and egl1, are present in uninduced T. reesei cells. These transcripts are induced at least 1100-fold in the presence of cellulose. We also show that a construct containing the hygromycin B resistance-encoding gene driven by the cbh1 promoter confers hygromycin B resistance to T. reesei cells grown in the absence of cellulose. Moreover, cellulose-induced production of the cbh1 transcript was suppressed when antisense RNA against three members of the cellulase system was expressed in vivo. Experiments are presented indicating that extracellular cellulase activity is the rate-limiting event in induction of synthesis of the cellulase transcripts by cellulose. The results reveal a critical requirement for basal expression of the cellulase system for induction of synthesis of its own transcripts by cellulose.

Plant introns are typically AU-rich or U-rich, and this feature has been shown to be important for splicing. In maize, however, about 20% of the introns exceed 50% GC, and most of them are efficiently spliced. A series of constructs has been designed to analyze the cis requirements for splicing of the GC-rich Bz2 maize intron and two other GC-rich intron derivatives. By manipulating exon, intron and splice site sequences it is shown that exons can play an important role in intron definition: changes in exon sequences can increase splicing efficiency of a GC-rich intron from 17% to 86%. The relative difference, or base compositional contrast, in GC and U content between exon and intron sequences in the vicinity of splice sites, rather than the absolute base-content of the intron or exons, correlates with splicing efficiency. It is also shown that GC-rich intron constructs that are poorly spliced can be partially rescued by an improved 3' splice site.

No abstract

The mechanisms of intron recognition and processing have been well-studied in mammals and yeast, but in plants the biochemistry of splicing is not known and the rules for intron recognition are not clearly defined. To increase understanding of intron processing in plants, we have constructed new pairs of vectors, pSuccess and pFail, to assess the efficiency of splicing in maize cells. In the pFail series we use translation of pre-mRNA to monitor the amount of unspliced RNA. We inserted an ATG codon in the Bz2 (Bronze-2) intron in frame with luciferase: this construct will express luciferase activity only when splicing fails. In the pSuccess series the spliced message is monitored by inserting an ATG upstream of the Bz2 intron in frame with luciferase: this construct will express luciferase activity only when splicing succeeds. We show here, using both the wild-type Bz2 intron and the same intron with splice site mutations, that the efficiency of splicing can be estimated by the ratio between the luciferase activities of the vector pairs. We also show that mutations in the unique U-rich motif inside the intron can modulate splicing. In addition, a GC-rich insertion in the first exon increases the efficiency of splicing, suggesting that exons also play an important role in intron recognition and/or processing.

Heterologous introns are often inaccurately or inef®ciently processed in higher plants. The precise features that distinguish the process of premRNA splicing in plants from splicing in yeast and mammals are unclear. One contributing factor is the prominent base compositional contrast between U-rich plant introns and¯anking G C-rich exons. Inclusion of this contrast factor in recently developed statistical methods for splice site prediction from sequence inspection signi®cantly improved prediction accuracy. We applied the prediction tools to re-analyze experimental data on splice site selection and splicing ef®ciency for native and more than 170 mutated plant introns. In almost all cases, the experimentally determined preferred sites correspond to the highest scoring sites predicted by the model. In native genes, about 90% of splice sites are the locally highest scoring sites within the bounds of the¯anking exon and intron. We propose that, in most cases, local context (about 50 bases upstream and downstream from a potential intron end) is suf®cient to account for intrinsic splice site strength, and that competition for transacting factors determines splice site selection in vivo. We suggest that computer-aided splice site prediction can be a powerful tool for experimental design and interpretation.