The amplifiable unit of DNA no. 1 (AUD1) of Streptomyces lividans consists of three 1 kb repeats (left direct repeat, LDR; middle direct repeat, MDR; and the slightly different right direct repeat, RDR) and two 4.7 kb repeats alternately arranged in identical orientation to each other. Both 4.7 kb repeats have been sequenced. They are identical and contain one open reading frame (orf4.7). The deduced amino acid sequence has a low similarity to chitinases, and two amino acid repeats present high similarities to fibronectin type III modules. Sequencing had previously shown that the ORF corresponding to each 1 kb repeat encodes a putative DNA-binding protein. Crude extracts of Escherichia coli overexpressing the orfRDR-encoded protein and of S. lividans Jni1, having a high amplification of AUD1 and therefore orfMDR, were used in gel retardation assays. The orfRDR- and probably the orfMDR-encoded proteins can bind to an imperfect palindromic sequence upstream from MDR and RDR and to another sequence downstream from RDR. An extrachromosomal DNA amplification system was constructed containing different combinations of the sequences composing AUD1. In mutants having a deletion of the chromosomal AUD1, the 4.7 kb repeats could be reduced in size, mutated or replaced by E. coli DNA without altering the ability to amplify when RDR was present. Therefore, the only function of the 4.7 kb repeats in amplification is to provide directly repeated DNA sequences. When RDR was lacking or mutated, no amplification was observed. This strongly suggests that the DNA-binding protein encoded by orfRDR is required for AUD1 amplification.

The amplifiable AUD1 element of Streptomyces lividans 66 consists of two copies of a 4.7 kb sequence flanked by three copies of a 1 kb sequence. The DNA sequences of the three 1 kb repeats were determined. Two copies (left and middle repeats) were identical: (1009 bp in length) and the right repeat was 1012 bp long and differed at 63 positions. The repeats code for open reading frames (ORFs) with typical Streptomyces codon usage, which would encode proteins of about 36 kD molecular weight. The sequences of these ORFs suggest that they specify DNA-binding proteins and potential palindromic binding sites are found adjacent to the genes. The putative amplification protein encoded by the right repeat was expressed in Escherichia coli.

In a spontaneous, chloramphenicol-sensitive (Cms), arginine-auxotrophic (Arg-) [DRs]) of 800 to 2,200 bp. The amplified DNA sequences (ADSs) consist of tandemly repeated units of one copy of the DR and one copy of the sequence flanked by the DRs. Class I amplifications arise within the same chromosomal region but differ in size and endpoints. Only a few, if any, directly or inverted repeated base pairs are found at the ends of the sequence used for amplification. Both types of amplifications are accompanied by deletions on one side of the amplified DNA. The deletions always occur on the same side of the amplified DNA. So far, it is unknown whether elements of the two classes amplify by the same mechanism, which enzymes and DNA structures are involved in the amplification process, and how it is correlated to the deletions.Streptomyces lividans is especially suitable for studying amplification processes. This species segregates chloramphenicolsensitive (Cms) mutants at an average frequency of 0.5% and in a second step arginine-auxotrophic mutants (Arg-) at a frequency of about 25% that of germinating spores (1). The mutants have extensive chromosomal deletions, and more than

No abstract