Myxobacteria are very important due to their unique characteristics, such as multicellular social behavior and the production of diverse and novel bioactive secondary metabolites. However, the lack of autonomously replicating plasmids has hindered genetic manipulation of myxobacteria for decades. To determine whether indigenous plasmids are present, we screened about 150 myxobacterial strains, and a circular plasmid designated pMF1 was isolated from Myxococcus fulvus 124B02. Sequence analysis showed that this plasmid was 18,634 bp long and had a G؉C content of 68.7%. Twenty-three open reading frames were found in the plasmid, and 14 of them were not homologous to any known sequence. Plasmids containing the gene designated pMF1.14, which encodes a large unknown protein, were shown to transform Myxococcus xanthus DZ1 and DK1622 at high frequencies (ϳ10 5 CFU/g DNA), suggesting that the locus is responsible for the autonomous replication of pMF1. Shuttle vectors were constructed for both M. xanthus and Escherichia coli. The pilA gene, which is essential for pilus formation and social motility in M. xanthus, was cloned into the shuttle vectors and introduced into the pilA-deficient mutant DK10410. The transformants subsequently exhibited the ability to form pili and social motility. Autonomously replicating plasmid pMF1 provides a new tool for genetic manipulation in Myxococcus.Myxobacteria are gram-negative gliding bacteria that are phylogenetically located in the delta division of the Proteobacteria (29,34,41). The two most intriguing characteristics of myxobacteria are their complicated multicellular social behavior, which provides an excellent model for studies of cell-to-cell communication and evolution (6,18,39,47), and their excellent capacity for production of diverse and novel bioactive secondary metabolites. Their production of bioactive secondary metabolites makes myxobacteria an important source of potential new drugs, although this possibility has not been well explored (36). The study and utilization of myxobacteria have been limited by the formidable isolation and culture techniques required (35) and the difficulty of performing genetic manipulations. In the past few decades, genetic studies of myxobacteria were performed mainly with the model species Myxococcus xanthus using transduction (7, 23) and the more efficient electroporation protocols (19). Besides these studies, Sorangium strains were also studied using conjugation protocols (13,14,22,31,33). Sorangium is a special cellulose degrader among the 17 myxobacterial genera (34, 50) and produces almost one-half of the known secondary metabolites produced by myxobacteria (8). Because no naturally occurring self-replicating plasmid has been discovered previously and no broad-hostrange vectors can replicate in myxobacterial cells, all the genetic transfer systems used have been based on integration of introduced plasmids or phages into the recipient chromosomes. Consequently, some genetic manipulations are hard to perform in myxobacteria or are not v...