Vol. 169, No. 9 JOURNAL OF BACTERIOLOGY, Sept. 1987, p. 4024-4029 0021-9193/87/094024-06$02.00/0 Copyright © 1987, American Society for Microbiology Comparative Organization of Nitrogen Fixation-Specific Genes from Azotobacter vinelandii and Klebsiella pneumoniae: DNA Sequence of the nifUSV Genes JIM BEYNON,' ABDUL ALLY,' MAURA CANNON,' FRANK CANNON,' MARTY JACOBSON,2 VALERIE CASH,2 AND DENNIS DEAN2* BioTechnica International, Inc., Cambridge, Massachusetts 02140,1 and Department of Anaerobic Microbiology, Virginia Polytechnic Institute and State University, Blacksburg, Virginia 24060' Received 4 March 1987/Accepted 15 June 1987 In the facultative anaerobe Klebsiella pneamoniae 17 nitrogen fixation-specific genes (nif genes) have been identified. Homologs to 12 of these genes have now been isolated from the aerobic diazotroph Azotobacter vinelandii. Comparative studies have indicated that these diverse microorganisnms share strikiig similarities in the genetic organization of their nif genes and in the primary structure of their individual nif gene products. In this study the complete nucleotide sequences of the nifUSV gene clusters from both K. pneumoniae and A. vinelandii were determined. These genes are identically organized on their respective genomes, and the individual genes and their products exhibit a high degree of interspecies sequence homology. Nitrogen fixation is catalyzed by the enzyme nitrogenase, a complex two-component metalloenzyme. In the facultative anaerobe Klebsiella pneumoniae there are at least 17 genes whose products are likely to be required for the synthesis and assembly of a fully active nitrogen-fixing system. These genes include nifHDK (nitrogenase structural components); nipF and nifJ (electron transport components); nifQ, niJB, niJE, nijN, and nifV (FeMo-cofactor biosynthetic compo- nents); nifA and nifL (positive and negative regulatory elements); nifM (Fe protein maturation component); and nifY, nJX, nifU, and nifS (functions not known). For a recent review, see reference 7. The nif genes are clustered on the K. pneumoniae chromosome and are organized into eight transcriptional units (1). A gene-product relationship has been established,for many of the K. pneumoniae nif- specific components (9, 18), yet only those products encoded by the nitrogenase structural genes and those encoded by the electronl transport-specific genes (16, 24) have been purified in their native form. Since the function of a number of the uncharacterized nif gene products is likely to involve a catalytic action upon a nitrogenase structural component, for example, metallocenter assembly, it is probable that some of these gene products are present in nitrogen-fixing cells in only low amounts. This possibility and the prospect that many of the ancillary nif gene products will share with the structural components an extreme sensitivity to oxygen potentially present a formidable challenge for nif gene prod- uct isolation and biochemical characterization. One approach for initiating the characterization of nif gene products is the determination of their polypeptide sequences deduced from DNA sequence analysis. Such an approach can provide (i) insights into the structural features of the encoded polypeptides, (ii) DNA sequence information nec- essary for the construction of hybrid plasmids that can be used to enhance the production of the individual nif poly- peptides, and (iii) the information and materials needed for the directed mutagenesis of the individual nif genes. In this regard, comparative studies of the primary sequences of individual nif-specific components from diverse diazotrophic * Corresponding author. species are of particular relevance because conserved se- quences among homologous proteins are likely to indicate their important structural features. In the case of the nitrogenase structural components, a high degree of interspecies sequence homology has permit- ted the isolation of nitrogenase structural genes from numer- ous diazotrophic species (20). Similarly, the demonstration of interspecies homologies among other individual nif genes should also permit the identification and isolation of such genes from a variety of diazotrophs. Summarized here are results obtained by using this approach for the isolation and mapping of nif genes from the aerobic diazotroph Azotobac- ter vinelandii. A comparison of the physical organization of nif genes established for K. pneumoniae with that currently developed for A. vinelandii is presented. In addition, we have determined and compared the nucleotide sequences of the nifUSV gene clusters from both K. pneumoniae and A. vinelandii and compared their respective encoded polypep- tides. MATERIALS AND METHODS The results of physical and genetic mapping procedures used to define the K. pneulnoniae nifUSV gene cluster are summarized in reference 1. The identification of the A. vinelandii nifUSV gene cluster was based on its homology to the K. pneumoniae nifUSV gene cluster (see Results). The DNA sequences of the A. vinelandii and K. pneumoniae nifUSV gene clusters were determined by the dideoxy chain termination procedure (21) by using hybrids of filamentous phage vectors described by Messing (15). For A. vinelandii sequencing experiments, the hybrid plasmids used as DNA sources are shown in Fig. 1A. For these experiments purified DNA fragments were digested with the individual restriction enzymes Sau3A, EcoRI, SmaI, XhoI, PstI, Sall, HinclI, HinPI, and MspI or some combination of these restriction enzymes and ligated into the appropriately digested replica- tive form of the filamentous phage vector DNA. Approxi- mately 200 base pairs were determined in each sequencing experiment. All sequences were determined in an overlap- ping fashion and in both directions. For K. pneumoniae sequencing experiments, the hybrid plasmid pWK25 (17) 4024