Gene, 78 (1989) 101-109
Elsevier GEN 02977
Identification and characterization of the nifl pEA3 of Enterobacter agglomerans 333
and nifJ promoter regions located on the nif-plasmid
(Recombinant DNA; nitrogen fixation; nzj”promoter, NifA-binding site; sequence similarity)
Roland Kreutzer “, Mahavir Singh b and Walter Klingmiiller” “Lehrstuhl ftir Genetik. UniversitiitBayreuth, 8580 Bayreuth (F.R. G.) and bGesellschaftftir Biotechnologische Forschung, Mascheroder Weg 1, 3300 Braunschweig (F.R.G.) Tel. (0531)6181431 Received
by G.N. Gussin:
29 June 1988
Small restriction fragments of the plasmid-borne Enterobacter agglomerans 333 nif region were cloned into a promoter probe plasmid as transcriptional fusions with the 1acZ gene. Identification of NifA-dependent promoters was accomplished by using a compatible plasmid which constitutively expresses the Kfebsiella pneumoniae nifA gene. /I-Galactosidase assays showed strong activation of the cloned E. agglomerans promoters in Escherichia coli by the heterologous K. pneumoniae nzyA gene product. The positions of the promoter fragments on the corresponding restriction map were determined by Southern hybridization. As contiied by sequencing data, the nzfl and nifJ promoters are situated at opposite end-points of the nif gene group and their -24 to -12 nucleotide sequences are similar to the consensus sequence of NtrA-dependent promoters. Also, typical NifA-binding motifs are present in both promoters. The agreement of the promoter proximal regions of nzm and nifJ with the corresponding K. pneumoniae sequences is about 80%. Also the upstream regions of these genes are in agreement to some extent.
Free-living nitrogen-fixing bacteria play an important role in the rhizosphere of grasses grown in temperate climates (Pedersen et al., 1978; Haahtela et al., 1981; Jagnow, 1988). Correspondenceto: Dr. R. Kreutzer, Universitat (F.R.G.)
The taxonomic analysis of bacteria in the rhizosphere of wheat grown around Bayreuth, F.R.G., revealed a large proportion of E. agglomerans (Kleeberger et al., 1983). Five nitrogen-futing strains out of this group harbored the nitrogenase structural genes (nzpDK) on large plasmids (Singh et al., stitutively
sidase; bp, base pair(s); guanosine; defined
kb, 1000 bp; LB, Luria broth; NFDM,
Iz termini; nitrogen
I; R, resistant; Tc, tetracycline;
reading P, pro-
of E. coli DNA
TE, 10 mM Tris/l
et al., 1974); nifA”, nifA gene con-
0 1989 Elsevier
mM EDTA pH XGal,
::, novel joint.
1983). One of these plasmids, the 11 I-kb plasmid pEA3, was cloned in a cosmid gene bank (Singh and Klingmtiller, 1986). Southern hybridization experiments and heteroduplex analysis showed strong similarities between the nif gene organization on pEA3 and that of the chromosome nif genes of 1y.pneumoniae M5al (Singh et al., 1988). The main difference relates to the nifJ gene which is located on the opposite end of the nif gene cluster when compared to its location in the n$f cluster of K. pneumoniae. In addition to the similarities in ng gene arrangement, we recently found that on a functional level there is also NifA-mediated positive regulation of n$ genes in E. a~Iomerans (Kreutzer et al., 1988). In the genome of K. pneumoniae 20 or 2 1 nif genes are organized into eight operons (MacNeil et al., 1978; Riedel et al., 1979; Merrick et al., 1980; Puhler and Klipp, 1981; Arnold et al., 1988). The genes nzjF and niff are transcribed in the direction opposite to that of all other nif operons (Beynon et al., 1983; Cannon et al., 1984). The genes n$L and nifA code for regulatory proteins (reviewed in Gussin et al., 1986); the nifA gene product is essential for transcription of the other nif operons (Dixon et al., 1980; Buchanan-Wollaston et al., 1981) and the nzfi gene product functions as a repressor for nif transcription under repressing culture con~tions (Hill et al., 198 1; Merrick et al., 1982). Ng promoters, originally characterized by their -24 to -12 consensus sequence 5’-CTGGN,-‘ITGCA (Beynon et al., 1983), belong to a unique class of promoters which are recognized by the alternative B factor NtrA (Hirschman et al., 1985). In addition, upstream activator sequences have been found in NifA-activated promoters (Buck et al., 1986). They are characterized by a TGT-N,,-ACA motif and have been demonstrated to bind the nfA protein at its C terminus (Morett and Buck, 1988; Morett et al., 1988). For transcriptional activation, the upstream bound NifA might be brought close to the RNA polymerase-NtrA complex bound downstream via the formation of a DNA loop (Buck et al., 1987). In this paper we report on functions elements involved in the nif regulation of E. agglomerans. We show that E. agglomerans nif -24 to -12 promoter sequences as well as the UAS agree with the consensus sequences of NifA-dependent genes from enteric bacteria and ~h~zobi~rn species. We also
report sequence similarities between the E. agglomeran~ niflC and ni$J promoter regions and the corresponding K. pneumoniae sequences.
MATERIALS AND METHODS
(a) Bacterial strains, plasmids, media and growth ~onditioas
Bacterial strains and plasmids are listed in Table I. Plasmid pCIS3 (Kennedy and Drummond, 1985) expresses nifA from the kanamycin promoter of pACYC177. Strains of E. cd were grown in LB medium (Miller, 1972). The antibiotics Tc and Ap were used at 15 and 400 pg/ml, respectively. To screen colonies for fiGal activity, 50 ~1of a 2 % XGal solution in dimethylfo~~ide was added to the plates. (b) Isolation and puriBcation of DNA
Small-scale purification of plasmid DNA was carried out by the alkaline lysis method (Maniatis et al., 1982) and large quantities of plasmid DNA were obtained according to the method of Humphreys et al. (1975) as modified by Singh et al. (1983). (c) Enzymes
The DNA modifying enzymes were used as specified by the manufacturers, either Boehringer M~nheim, F.R.G., or Bethesda Research Laboratory, Gaithersburg, U.S.A. BAL 31 digestion of linear cosmid DNA was carried out analogously to the description of Maniatis et al. (1982) for mapping restriction sites in DNA. Separation of DNA fragments was achieved by horizontal agarose gel electrophoresis and Southern hybridization was carried out as described earlier (Singh et al., 1988). (d) Cloning procedures
A new technique for DNA elution was used. Bands containing restriction fragments to be cloned were cut out of agarose gels and divided into small
103 TABLE I Bacterial strains and plasmids Strain or plasmid
Genotype or phenotype b
Alac-pro thi rpsL supE end4 sbcB15 hsdR [F’traD36 proAB+ Ia&” IacZAMlS]
Schneider and Beck (1986) Messing et al. (1981) Rosenberg et al. (1985)
galK thi rpsL recA 56
C SW (Los2 AB xi.sl red3 gamam210 cIts857 nin5 Sam’l)/l,
peaMS2-2 pCK3 pCB182 pUCl8 pSA30 pMK182KpH pMK182Pl pMK182P7 pMK182P8 pMK182P9
ApR, E. agglomerans nif cluster Tcu, K. pneumoniae nifA’ ApR ApR TcR, K. pneumoniae nifKDH ApR, K. pneumoniae PnifH:: IacZ ApR, E. agglomerans PnifJ : :IacZ ApR, E. agglomerans PnifH : :IacZ ApR, E. agglomerans Pnif-IacZ-fusion ApR, E. agglomerans Pnif-IacZ-fusion
Singh et al. (1988) Kennedy and Drummond (1985) Schneider and Beck (1986) Yanisch-Perron et al. (1985) Cannon et al. (1979) this work this work this work this work this work
a All the pMK plasmids contain promoter-lacZ fusions, which were achieved as indicated in Fig. 1. By nucleotide sequence analysis, the cloned promoters of pMK182Pl and pMK182P7 were identified to be PnifJ and PnifH (see Fig. 5). The cloned promoters of pMKl82P8 and pMK182P9 are located within the E. agglomerans nfgene cluster (see Fig. 3). However, they were not assigned to certain nif genes. b A, deletion; AB, EcoRI-B fragment of I, DNA was deleted; C, E. coli C; R, resistant; [ 1, designates plasmid (F’)-carrier state; ( ), indicates prophage; /a, resistant to phage 1.
pieces. After adding 200 ~1 of TE and 500 ~1 of Trissaturated phenol, the vortexed mixture was briefly frozen in liquid nitrogen. Centrifugation yielded a supernatant containing the DNA fragment. This was treated with one volume of chloroform and precipitated with ethanol and sodium acetate as described by Maniatis et al. (1982). Digested vector DNA was treated with calf intestinal phosphatase (Boehringer Mannheim, F.R.G.) to prevent self-ligation. For ligation, appropriate insert-to-vector ratios were mixed and treated with T4 DNA ligase (Boehringer Mannheim, F.R.G.) first for 4 h at 20’ C and then at 4°C overnight. Transformation was carried out as described by Cohen et al. (1972).
RESULTS AND DISCUSSION
(a) Cloning of NifA-dependent
To detect n$ promoters of E. agglomerans we used the promoter probe plasmid pCB182 which ‘has a
multiple cloning site located between promoterless IacZ and galK genes arranged in opposite orientation to each other (Schneider and Beck, 1986). In this work we only screened for IacZ transcriptional fusions. The 23-kb nif region of the E. agglomerans plasmid pEA3 had earlier been cloned into the 43-kb recombinant plasmid peaMS2-2 (Singh et al., 1988). For subsequent cloning of nif promoters most of the peaMS2-2 that retained no nif DNA was deleted. However, no suitable endonuclease restriction sites were present that would enable the complete nif cluster, by single or multiple enzyme digestion, to be obtained. We therefore used exonuclease BAL 31 treatment of linearized peaMS2-2 DNA to eliminate irrelevant sequences. The cloning procedure is shown in Fig. 1. The ligation mixture was used to transform the Lac - E. coli strain CB454 containing the plasmid pCK3, which constitutively expresses the K. pneumoniae nifA gene (Kennedy and Drummond, 1985); the transformed cells were plated on selective medium containing XGal as a color indicator for /?Gal activity. Fifty-four blue colonies among 500
Fig. I. Procedures
for the cloning
of the helper lysogen
of nif promoter
model MD-500) until the 0D5,snm followed by incubation from phages peaMS2-2
The thin lines indicate
E. agglomerans plasmid DNA. The open boxes indicate nif DNA and the tilled arrowed The dashed
Pol I k
of BAL 31 digestion. the plasmid
to the large-scale
(Left): For the isolation
were collected method
of in vivo linearized
by centrifugation by Maniatis
DNA, eight liters
and linear peaMS2-2
of the linear
were ligated into the BarnHI-linearized
(Right): For cloning of the K. pneumoniue nigh
region, pSA30 was cut with EcoRI. The resulting cohesive ends were filled in with PolIk. Digestion which was ligated into the BglII + SmaI-digested
cosmid DNA was isolated
et al. (1982). BAL 31 treatment
galK and 1uc.Z genes.
was raised to 44°C for 15 min to induce phage gene expression,
mostly nif DNA which was cut with Sau3A.
probe plasmid pCB 182. The resulting plasmids
DNA while the bold lines indicate
were grown at 37°C in a small fermenter
0.6. Then the temperature
at 38°C for 3 h. The bacteria
boxes indicate promoterless
with BglII provided
white colonies were isolated and shown to contain plasmids, designated with pMK182P numbers, having Sau3A insertions of various sizes in the range of 0.3 to 1.0 kb (Fig. 2). To identify the inserts with NifA-dependent promoters, plasmid DNAs were used to transform CB454 without the @it-expressing plasmid pCK3. About 70% of the corresponding clones which had formed blue colonies under the influence of pCK3 now remained white, indicating that they carried an insertion with a NifA-activated promoter. The other 30% formed blue colonies both in strains with and without pCK3 and thus should contain NifA-independent promoters.
The K. pneumoniae nigh promoter was also cloned into pCB182 in order to obtain a well-defined comparable plasmid for measuring nif promoter activities. The construction was achieved as indicated in Fig. 1. (b) Localization
of the cloned promoters
Plasmid DNA containing NifA-dependent promoters was digested with SmaI + Hind111 and run on a 1.5% agarose gel to separate the insertions. These fragments were eluted, radioactively labeled by nick-translation, and hybridized to several restric-
Fig. 2. 1.5%
of SmaI + HindIII-
DNA of clones with NifA-dependent
by the shotgun
in Fig. 1. Each
promoter as explained
of the promoter
plasmid pCB 182 and the insert. The sizes of the insertions from 0.3 to 1.0 kb. The faint bands partial
additional promoters present that were not identified by the shotgun cloning technique. Considering the strong similarity between K. pneumoniae and E. agglomerans nSfgenes, we would have expected to find at least eight promoters in the E. agglomeruns nif gene cluster. We localized two NifA-dependent promoter regions bordering the nif gene cluster. As reported earlier, the genes nzj2 and nzj7f are located at the ends of the nif cluster (Singh et al., 1988). Based on these results, it is expected that the two flanking promoters are PnifI and PnifII and the corresponding genes are transcribed in opposite directions. This was confirmed by ZacZ fusion studies and nucleotide sequencing data, which indicated that nifJ is transcribed rightward and nzm is transcribed leftward in the orientation shown in Fig. 3. Genes nifJ and n$H are also transcribed divergently in K. pneumoniae, although, in contrast to the situation in E. agglomerans, the r&J gene in K. pneumoniae is located at the same end of the nif gene cluster as nzm. The -24 to - 12 promoter sequences and the UAS of the two genes in E. agglomerans were identified by comparison with the nif promoter consensus sequence and the consensus sequence for the UAS (Beynon et al., 1983; Buck et al., 1986; see section d below). The transcription start points were not determined.
are due to
tion digests of the recombinant cosmid peaMS2-2 blotted on nitrocellulose. On the basis of these results we identified the positions of most of the nif promoter fragments in the corresponding restriction map. Four representative pMK182P-plasmids whose inserts were used as probes are indicated in Fig. 3. Unambiguous mapping was not possible with all clones since more than one Sau3A fragment was probably inserted in the vector, resulting in hybridization signals at different places in the nif cluster. The hybridizing sites of these clones are not shown in Fig. 3. The four probes shown in Fig. 3 identify four different promoter-containing regions of the E. ugglomeruns nif gene cluster. It is likely that there are
(c) Influence of N&4 protein on nif promoter activity
The activity of each representative cloned NifAdependent promoter fused to the IacZ gene was
E:EgRI H. -Hind
pMK182Pl Fig. 3. Restriction Regions
map of the E. agglomerans nif gene cluster and the corresponding
with nif promoter
(see Table I) are indicated
at the bottom
PI gene map (Singh et al., 1988, revised).
of the figure.
of nif promoter
: : 1ucZ fusions /?Gal activity
(in E. coli + pCK3
@I$4 ‘) < 10
K. pneumoniae PnifH
E. agglomerans PnifJ
a See Table I. b Average
of four experiments.
minimal medium (NFDM) supplemented
with 100 pg aspartic
tions were achieved
were grown for 20 h in by Cannon
et al. (1974),
tally using ONPG
(Miller, 1972) with a modified cell lysis proce-
dure as described
(d) Sequencing regions
12-ml serum bottles containing
6 ml of liquid medium. j?Gal assays were performed
and Koch (1975). The units were
as in Miller (1972).
determined quantitatively by measuring the /IGal activity with and without the nzj_X-expressingplasmid pCK3 in E. coli (Table II). The K. pneumoniue nzj7l promoter was also tested under the same conditions. The four promoters shown in Fig. 3 were all activated significantly in the presence of pCK3. It can be concluded that the K. pneumoniae NifA protein activates E. agglomerans nif promoters. Also, the NifA-dependent promoters of these two bacterial species might have very similar specificities. This is confirmed by nucleotide sequencing data (see section d below). It is interesting that, in the system
we used, the E. agglomerans PnifII and PnifI, as well as the other cloned promoters, are activated at least three times as strongly as the K. pneumoniae PnifII by the K. pneumoniae nifA protein. This may reflect a greater promoter activity. The promoter probe plasznid pCB182 is a multicopy plasmid, whereas pCK3 is based on an IncP replicon and hence has a lower copy number. It is therefore possible that not all NifA-dependent promoters in each cell clone become fully activated in our test. This might be the reason why the /?Gal activities are lower than those reported for a comparable system by Tuli and Merrick (1988) who used multicopy plasmids based on pBR322 for NifA overproduction.
the positions to the method
for the nifJ (a) and nigh (b) p romoter regions. The arrows show the extent of sequencing reactions in genes, which are indicated by boxes. The vertical bars in front of the genes indicate
map and the respective
of the promoters,
cloned in pUCl8 reaction
to the restriction
Four different DNA fragments were subcloned in the sequencing vector pUC18. Two inserts were subsequently deleted by the use of exonuclease III. The resulting clones were used to determine the nucleotide sequence of both DNA strands of the nifJ and nzfl promoter regions. The sequencing strategy is shown in Fig. 4. The sequences of both promoter regions and the promoter proximal regions of the corresponding genes are presented in Fig. 5. By computer analysis we identified regions similar to data base sequences of the K. pneumoniae nif gene cluster, which are also presented in Fig. 5 for comparison with respective sequences of E. agglomerans. These data confirm the high degree of sequence similarity between the plasmid-borne E. agglomerans
Fig. 4. Sequencing
of the nifJ
the circles indicate
of the UAS. For plasmid sequencing,
et al., 1985). To delete some of the inserts, the exonuclease
(1984). DNA was prepared
out with the Boehringer
to the method
kit as specified
was used according
(1981), and the sequencing
by the manufacturer.
/II/ IIIII II II CCTGCTGATCGCCAAAAATATTcAGcGAATTGGTCCCCAGCGcccGG
nif gene cluster and the chromosomal nif cluster of K.pneumoniae,
which had been predicted on the basis of Southern hybridization experiments and 97 48 CCACTtACRTGRARAACTGCGGGCAGGAGTTCACCttCRA heteroduplex analysis (Singh et al., 1988). The seII IIIII IIIII II I IIIII IIlIIllIIII IIIIIIIlIII 3487 3438 GC‘CTGACGTGAAAGACGCCCGGCAGCAG~~CACcGGCGA~~~~G~AcA~ quence similarity in DNA primary structure is 80% 147 9S GTTGGGTGTCRTTAACAACAACCCTTGCG;\RGCGGTATA~G~GG~CGCC~ in the promoter proximal region of nigh and 76% in II/III IIII/IIIII/IIIIIIIlllllIIIIlI I 3488 GTTGGGGATCATCAGCAGCAGCCCCTCtGACtCCGTATAGGCCG~A~AGG~GG=GG~‘A3537 the promoter proximal region of nzj3. The agreement between the ammo acid sequences deduced from the 197 148 GCGCACCAGCCTGCAAGGCACCGTGTATC;;CCCCGGCGG~GCCAGA~~C~ II/I II IIIIIII II Illll I III II II IIIII I II nucleotide sequences (data not shown) is even 3587 3538 GCGCCCCGGCCTGCAGCGCGCCGTGGACCGCGCCTGCCGCGCCGGCCTCC higher, as most differences in sequence are in the 247 198 GACTGCATCTCCATGATTTTCACCGTCTG;CCGAACAGG~~==~~~~~C~ IIIIII/IIIIIII I IIIII III II II I/III IIIII II third position. A preference for the use of G + C-rich 3737 3588 GACTGCATCTCCATTAAGCGCACCGGCTGGCCARIAAGCT~C==~~~CCC codons is evident in the sequenced nif gene regions SO/l 297 248 CTGTGCGGCCCATTCATCAACACTTTCTGCCATCGG-GGTGTA; of E. agglomerans (data not shown), when the nucleIIlIllIllllIIIII IIIlIIIIIlIIIIIlllII/ 3787 3738 CTGCGCCGCCCACTCGTCGACGTTTTCCCCCATCGGCGTGGAGGGGG~=A otide sequences are compared to the corresponding 298 TCGGATAAATCGCGGTRAC;TCCGTGAAGGCATAAGACA~ATAGGC~GC~ 347 regions of K. pneumoniae. I II IIIIII/I I II II II IIIIlIIIIII I III1 III 3788 3837 TGGGGTAAATCGCCGCGACCTCtGTAAAGGCATAAAGITCCAGGCCGCC There is lower agreement between the sequences nrfJ SD upstream of nigh and n$_J in the two bacterial 348 GCCGTATTCCCGTCCATGGTTTTCATTTTTCCCGGCATGA 397 II I II II IIIII IIIIIlIIIlIIII I III IIII II I species, which is typical for non-coding regions lackGCGGCGTTGCCATCCATTGTTTTCATTTTTCCG~,ACTTGTKAA~~~~C 3839 3887 SD ing a selective pressure. However, similarities in : . 398 CGCTCAATGTTGCTCI\ALTbCAACCTTTCAGAAPICG............,.. 432 I II I I I I III these regions are obvious when the known promoter GAAGGTGAGAGGCATCTTCGCCGCCTCAAAA~AAGCGGCAAACCCAG~~G~ 3888 3837 features are aligned by incorporating gaps. This 489 CAGAAGGGTTGTATTAATATATTG; 433 observation implies that the ends of the E. agI II I I I 3887 TCGCGGATGACAGAAGAGTTAGCGC 3838 glomerans nifgene cluster might have the same evolupromoter 529 490 ATTAAAAGGAATAGTTAGGGTGGGCTCTC;ATAGGGGGTT tionary origin. Since the nigh and nifJ genes are I I III I III I II 3937 GAArPCAACGCGTTATGRAGIGRGTCGCCGCCGCGCAGCGCGCCAAGAGA~~G 3888 located at opposite ends of the E. agglomerans nif “AS cluster but at the same end of the K. pneumonia nif 579 530 AATGTTTCAGCACACATTTGTGRGCAACTTGTCtTTACAC~GACA==ACC IIII I II III III I I IIIII I 3987 cluster, the 18-bp gap in alignment at nt position 432 CGTGGAATAkGACACAGGGG:CGACAAGCTGTTGARCAGGCGACAAAGCG 3938 “AS of the E. agglomerans nifJ promoter region may be 591 580 CGCAGGCGTiCTGCC related to rearrangement of nigh and nifJ in E. agI II I Ill + UAS 1037 CCCATGGCCCCGGCAGGCCCAATTGTTCTGTTTCCCACATTTGGTCGCCT 3988 III II I Ill !IIIl IIll II II I glomerans, compared to K. pneumoniae. CTCCCGAAACCCGCCATCG.~TTTCTTTGTTTTATAACAAAC=G=ACGCA 49 UAS We identified both typical -24 to - 12 nif promoter 1097 4038 TA~~G~G~CGTT~~G~~~~ACGTCCTGCGCGG~GA~A*A~AAC~AA~*~~ sequences and UAS elements in the E. agglomerans IIII III I I I II II I I I lllll 98 50 G*TTGTTTTCTCTGCTTCTAA~A=~~CCAAC~CATTA~~~~AAA*C.GC nz#H and nz$J promoter regions (Fig. 5). A G occurs 4137 4088 at nt position 131 in the E. agglomerans -24 to -12 IIII III I I III 147 99 ATATTTTTCAATGGGTTTACAGCAT nzm promoter sequence, deviating from the corpromoter_ SD nlfH responding C in the K. pneumoniae promoter. How4138 CTTCTCTGCTGGCAAACACTCAACAAC~~~~GAGAIGTCACCATGACCA~G~ 4187 II III I III1 II IIlIIIlI I III I III 148 ATAGCCTGTATGTCAACTGGACAC=AC~~~~GAGAAAACAA=GGCAA~GC 197 ever, examination of all available NtrA-dependent & SD promoter sequences reveals that the corresponding 4188 G~~AA~GC~CTA~~~*~GGTAAAGG~GG~*~~GG~A**~C~A~~A~C*CG 4237 3437
Fig. 5. Sequences of the E. qgglomerans nigh and nifJ promoter regions and the promoter pmximal regions of the nigh and nifJ genes aligned with the corresponding sequence ofK. pneumoniae as published by Arnold et al. (1988). The extent of the sequenced E. agglomerans regions is marked by open arrows for the nifJ promoter region (numbered from nt 1 to 594) and by filled arrows for the nigh promoter region (numbered from nt 1 to 295).
Numbering of the K. pneumoniae sequence is taken from Arnold et al. (1988). Relevant restriction sites areindicated by horizontal brackets. The -24 to -12 consensus sequences are boxed, the putative UAS are marked by lines, the gene starts are indicated by thin arrows and the putative SD-sequences are marked by dashed lines. The sequences were aligned with the aid of the University of Wisconsin Genetics Computer Group program GAP. The dots within the sequences are gaps introduced there to align the known promoter features.
nt is not conserved
and may not be of significance
promoter function. The results reported the K.pneumoniae beginning
in the literature
which is similar
of the E. agglomerans nifJ gene, are con-
data that an ORF
et al. (1983)
niJH and t&J, pointing (1988), however,
deduced is located
from their between
We would Steinlein
like to thank
for helpful discussions,
M. Buck for reading
Dr. W. Schumann
Dr. J. Dancer
away from n$H. Arnold et al.
found no stop codon from termi-
nating this putative ORF and took this region to be the beginning of the nifJ gene.
Since in E. agglomerans there is no stop codon present in the sequenced region downstream of PnifJ, we propose that this sequence codes for the N-terminal part of the E. agglomerans nifJ protein. Also, BAL 3 1 truncation of the pMK182Pl insertion demonstrated that there is no other promoter activity on the remaining downstream fragment after cutting off PnifJ (data not shown). We therefore ascribe expression of nz$J to the promoter identified from nt position 437 to 453 (Fig. 5).
A., Klipp, W., Priefer,
the entire nitrogen Beynon,
(2) The E. agglomerans nifH and nifJ promoter regions contain sequences that are very similar both to the -24 to -12 nif promoter consensus sequence and the characteristic NifA-binding site. Nz$ regulation in E. agglomerans may therefore involve very similar promoter elements as reported for other enteric bacteria. (3) Several observations indicate the close relationship between the nif gene clusters of E. agglomerans and K. pneumoniae. Cloned promoters of the E. agglomerans nif gene cluster are activated by the K. pneumoniae NifA protein. The promoter proximal parts of the coding regions of the E. agglomerans nigh and n$J genes have sequence agreements of 76-80% with the corresponding K.pneumoniae sequences. Since alignment of the nigh and r&J promoter regions in the two bacterial species reveals to some extent agreement in the sequences upstream of the n$H and n$J genes, the regions might be homologous and have the same origin in evolution.
of Klebsiella pneumoniae.
V. and Cannon,
ofKlebsiellapneumoniae have a charac-
F.: The nifpromoters
Cell 34 (1983) 665-671.
F.: Role of the n$4 gene product
in the regulation
in Klebsiella pneumoniae. Nature
of nif expression
294 (198 1)
(1) We suggest that the region corresponding to the earlier reported ORF in the region upstream of the K. pneumoniae nl$r-f gene is the promoter proximal part of the nifJ gene in E. agglomerans.
J. Mol. Biol. 203 (1988) 715-738.
Buck, M., Miller, S., Drummond,
U. and Pfihler,
of a 24,206 bp DNA fragment
M. and Dixon, R.: Upstream
in the promoters
320 (1986) 374-378.
of the Klebsiella pneumoniae nitrogenase
may involve DNA loop formation.
F., Dixon, R., Postgate,
J. and Primrose,
of Klebsiella nitrogen
Escherichia coli. J. Gen. Microbial. Cannon,
F., Riedel, G. and Ausubel,
80 (1974) 227-239. F.: Overlapping
of Klebsiella pneumoniae nif DNA cloned and characterised. Mol. Gen. Genet. Cannon,
174 (1979) 59-66.
J., Hill, S. and Cannon,
pneumoniae [email protected]
and n$J. In Veeger, C. and Newton, (Eds.), Advances Nijhoff/Junk, Cohen,
Shu, L.: Nonchromosomal
E. coli by R-factor
1984, p. 34.
(1972) 2110-2114. Dixon,
R., Eady, R.R., Guadalupe,
Kahn, D. and Merrick, pneumoniae nitrogen fusions. Gussin,
(nif) gene cluster
ofKlebsiella with gene
286 (1980) 128-132.
E., Hill, S., Iaccarino,
M.: Analysis of regulation
C.W. and Ausubel, genes.
F.M.: Regulation Genet.
V. and Skujin, J.: Root reduction)
bacteriaceae and Azospitillum strains in cold-climate Appl. Environ. Henikoff,
41 (1981) 203-206. digestion
for DNA sequencing.
(1984) 35 l-359. Hill, S., Kennedy, ([email protected]
in oxygen regulation
in K. pneumoniae. Nature Hirschman,
genes ntrA and ntrC of
in vitro: evidence
is a a factor. Proc. Natl. Acad. Sci. USA
D.W. and Quigley, M.: A rapid boiling method
193-197. G.O., Willshaw,
G.R. and Anderson,
method for the preparation DNA. Biochim. Jagnow,
in the rhizosphere
plasmid of wheat, in field
W.E. (Eds.), Nitrogen
Years After. Fischer-Verlag,
H. and Klingmtiller,
W.: The rhizo-
of wheat and barley with special reference
Bothe, H., de Bruijn, F.J. and Newton, Fixation:
in Enterobucter. In
W.E. (Eds.), Nitrogen
NifA resides activator
and can be separated of NifA. Nucleic K., Klucas,
H. and Trebst,
R.V. and Vivader,
35 (1978) 129-135.
A. and Khpp,
W.: Fine structure
of the gene
(nif) ofKlebsiellupneumoniae. In Bothe,
region for N,-fixation
A. (Eds.), Biology of Inorganic
198 1, pp. 276-286. Putnam,
S.L. and Koch,
in the simplest
cellular enzyme assay: lysis ofEscherichiu coli for the assay of Anal. Bioch. 63 (1975) 350-360. F.M. and Cannon,
Sci. USA 76 (1979) 2866-2870.
S.M., Stahl, M.M., Kobayashi,
in vitro packaging system
fixation (nif) genes of Klebsiellu pneu-
moniae. Proc. Natl. Acad. Rosenberg,
I. and Stahl, F.W.:
(1985) 165-175. Schneider,
K. and Beck, CF.:
Cold Spring Harbor
Gene 42 (1986)
Q.-T. and Wu, Q.-L.: An open
from the [email protected]
gene of Klebsiellu
Acids Res. 11 (1983) 4241-4250. W.: Cloning ofpEA3,
map of the Klebsiellu pneumoniue
a tine-structure fixation
C., Sibold, L. and
(nif) gene cluster.
J. Gen. Microbial.
A. and Klingmtiller, (nif) genes
in Klebsiellupneumoniue. Mol. Gen. Genet. J., Crea, R. and Seeburg,
DNA sequencing. J.: Experiments Laboratory,
E. and Buck,
M., Dixon, R. and
of the nifL gene product 185 (1982) 75-81.
P.H.: A system for shotgun
Acids Res. 9 (1981) 309-321.
Cold Spring Harbor, M.: NifA-dependent
that the upstream
NY, 1972. in vivo protection sequence
on indigenous Gen.
R., Acker, G. and Khngmiiller,
373-378. Singh, M., Kreutzer,
23 kb nif-
of Klebsiella pneumoniue
19 (1988) I-12.
Tuli, R. and Merrick, scriptional
of a plasmid-borne
from Enterobucter ugglomerans showing
ogy to the entire M5al.
of the nifA gene product
M., Hill, S., Hennecke,
a large plasmid
genes. Plant Soil 90 (1986) 235-242.
J.: The use of translocatable
Enterobacter ugglomeruns. Mol.
(1980) 509-520. Kennedy,
Shen, S.-C., Xue, Z.-T., Kong,
136 (1978) 253-266.
M., Filser, M., Dixon, R., Elmerich,
E.F. and Sambrook,
pneumoniue. J. Bacterial.
of Enterobucfer ugglomeruns containing
Singh, M. and Khngmiiller,
1988, p. 311. MacNeil,
site. Proc. Natl. Acad.
Riedel, G.E., Ausubel,
M.: The use of cloned nif regula-
lation in Azotobacrer vinelundii. J. Gen. Microbial.
chromosomal C. and Drummond,
tory elements from Klebsiellupneumoniue to examine nif regu-
p. 795. Kennedy,
of nif promoters
doses of N-fertilizers.
H., de Bruijn, F.J. and Newton, tion: Hundred
ES.: A simple
Acta 383 (1975) 457-463.
of the transcriptional
82 (1985) 7525-7529. Holmes,
in its carboxy
J. and Kustu,
activate g/n,4 transcription
that the ntrA product
290 (1981) 424-426.
of the nitrogen
J., Wong, P.-K., Sei, K., Keener,
is a protein
USA 85 (1988) 9401-9405. C. and Kavanaugh,
of Klebsiellu pneumoniue - the tran-
J. Gen. Microbial.
134 (1988) 425-432. Yanisch-Perron,
C., Vieira, J. and Messing,
phage cloning vectors of the
and host strains:
J.: Improved nucleotide
sequence 33 (1985)