Mol Gen Genet (1991) 229:189-196 0026892591002944 © Springer-Verlag 1991 Catabolite repression of the operon for xylose utilization from Bacillus subtilis W23 is mediated at the level of transcription and depends on a cis site in the xylA reading frame Sabine Jacob, Rudolf Allmansberger, Dagmar Giirtner, and Wolfgang Hillen Lehrstuhl ffir Mikrobiologie, Institut ffir Mikrobiologieund Biochemieder Friedrich-Alexander Universitfit Erlangen-Ntirnberg Staudtstrasse 5, W-8520 Erlangen, Federal Republic of Germany ReceivedMarch 22, 1991 / May 8, 1991 Summary. The Bacillus subtilis xyl operon encoding en- zymes for xylose utilization is repressed in the absence of xylose and in the presence of glucose. Transcriptional fusions of spo VG-lacZ to this operon show regulation of fi-galactosidase expression by glucose, indicating that glucose repression operates at the level of transcription. A similar result is obtained when glucose is replaced by glycerol, thus defining a general catabolite repression mechanism. A deletion of xylR, which encodes the xy- lose-sensitive repressor of the operon, does not affect glucose repression. The cis element mediating glucose repression was identified by Ba131 deletion analysis. It is confined to a 34 bp segment located at position + 125 downstream of the xyl promoter in the coding sequence for xylose isomerase. Cloning of this segment in the op- posite orientation leads to reduced catabolite repression. The homology of this element to various proposed con- sensus sequences for catabolite repression in B. subtilis is discussed. Key words: Bacillus subtilis - spo VG-lacZ fusions - Reg- ulation of transcription - Glucose repression - Transla- tional coupling 1984) and amyE (Nicholson et al. 1987), are subject to catabolite repression (Fisher and Sonenshein 1991). Un- like catabolite repression in Escherichia coli, where a general, well understood mechanism of transcriptional activation by catabolite gene activator protein (CAP) and cyclic AMP mediates this effect (see de Crom- brugghe et al. 1984 for a review), it is not clear whether a common mechanism exists for this phenomenon in B. subtilis. A cyclic AMP-dependent regulatory mecha- nism seems very unlikely since cyclic AMP can only be detected in low concentrations under oxygen stress con- ditions (Mach et al. 1984). Recently three cis-acting ele- ments for glucose repression of amyE (Weickert and Chambliss 1990), gnt (Miwa and Fujita 1990), and citB (Fouet et al. 1990; Fouet and Sonenshein 1990) have been analysed and turned out to consist of different nu- cleotide sequences. It is therefore interesting to charac- terize the molecular details of glucose repression for other B. subtilis genes. In this report we demonstrate that glucose repression of the xyl operon in B. subtilis occurs at the level of transcription, is independent of a functional repressor gene for xylose induction and depends on a cis sequence in the translated reading frame of xyIA. Introduction Enzymes encoded by the xyn and xyl operons of Bacillus subtilis cooperate in the utilization of complex hemicellu- loses (Hastrup 1988; Wilhelm and Hollenberg 1984). They are negatively regulated by the Xyl repressor and are inducible by xylose (Kreuzer et al. 1989). Preliminary indications have suggested that the xyl operon is also glucose repressed (Gfirtner et al. 1988). In B. subtilis, several genes for degradative enzymes, among them sdh (Melin et al. 1989), grit (Fujita and Miwa 1989), citB (Rosenkrantz et al. 1985), hut (Fisher and Magasanik 1984), sacC (Martin et al. 1989), bglS (Murphy et al. Offprint requests to: W. Hillen Materials and methods Bacterial strains and culture conditions. Bacterial strains used in this study are listed in Table 1. E. coli HB101 (Boyer and Roulland-Dussoix 1969) was used as a host for cloning experiments. All B. subtilis strains are deriva- tives of B. subtilis 168. They were transformed to Cm r and amy- phenotypes with the pDH32 derivatives indi- cated in Table 2 and show xylose-dependent fl-galactosi- dase expression (G/irtner et al. 1988). B. subtilis was made competent according to a published protocol (Hardy 1985). B. subtilis and E. coli were generally grown in LB medium (i0 g tryptone, 10 g NaC1, 5 g yeast extract per litre deionized water, pH 7.3). Mopso medium was used