A minimal nitrogen fixation gene cluster from Paenibacillus sp. WLY78 enables expression of active nitrogenase in Escherichia coli

Abstract

Most biological nitrogen fixation is catalyzed by molybdenum-dependent nitrogenase, an enzyme complex comprising two component proteins that contains three different metalloclusters. Diazotrophs contain a common core of nitrogen fixation nif genes that encode the structural subunits of the enzyme and components required to synthesize the metalloclusters. However, the complement of nif genes required to enable diazotrophic growth varies significantly amongst nitrogen fixing bacteria and archaea. In this study, we identified a minimal nif gene cluster consisting of nine nif genes in the genome of Paenibacillus sp. WLY78, a gram-positive, facultative anaerobe isolated from the rhizosphere of bamboo. We demonstrate that the nif genes in this organism are organized as an operon comprising nifB, nifH, nifD, nifK, nifE, nifN, nifX, hesA and nifV and that the nif cluster is under the control of a σ(70) (σ(A))-dependent promoter located upstream of nifB. To investigate genetic requirements for diazotrophy, we transferred the Paenibacillus nif cluster to Escherichia coli. The minimal nif gene cluster enables synthesis of catalytically active nitrogenase in this host, when expressed either from the native nifB promoter or from the T7 promoter. Deletion analysis indicates that in addition to the core nif genes, hesA plays an important role in nitrogen fixation and is responsive to the availability of molybdenum. Whereas nif transcription in Paenibacillus is regulated in response to nitrogen availability and by the external oxygen concentration, transcription from the nifB promoter is constitutive in E. coli, indicating that negative regulation of nif transcription is bypassed in the heterologous host. This study demonstrates the potential for engineering nitrogen fixation in a non-nitrogen fixing organism with a minimum set of nine nif genes.

Figure 1. Comparison of the Paenibacillus sp. WLY78 nif gene cluster with representative clusters from diverse diazotrophic bacteria and archaea.

(A) Azotobacter vinelandii, (B) Heliobacterium chlorum, (C) Clostridium acetobutylicum W5, (D) Frankia sp. EAN1pec, (E) Methanococus maripaludis, (F) Anabaena variabilis ATCC 29413, (G) Klebsiella oxytoca M5al, (H) Paenibacillus sp. WLY78.

Figure 2. The nif genes of Paenibacillus sp. WLY78 are organized in an operon as determined by RT-PCR.

(A) Outline of the strategy. Primers used and amplified products (numbered) are given below the schematic representation of the genes. (B) Result of RT-PCR reactions with RNA from Paenibacillus sp. WLY78 grown under N₂-fixing conditions. The numbering on the top of the gels corresponds to the product numbers drawn schematically in the outline given above. RT, standard RT-PCR reaction; (−), negative control in which no reverse transcriptase was added to the RT reaction; (+), positive control in which genomic DNA was used as template in the RT-PCR.

Figure 3. Characterization of the nif promoter of Paenibacillus sp. WLY78.

(A) Schematic representation of the Paenibacillus sp. WLY78 nif operon. (B) Nucleotide sequence of the nifB promoter and the putative terminator sequence flanking the 3′ end of nifV. The asterisks below TAA indicate the nifV stop codon. (C) Overexpression and purification of σ⁷⁰ from Paenibacillus sp. WLY78. Lane 1: protein marker; lane 2: uninduced protein; lane 3: induced protein; lanes 4: purified σ⁷⁰ factor. (D) Electrophoretic mobility shift assays (EMSA) demonstrating binding of Paenibacillus σ⁷⁰ to the 50 bp nifB promoter DNA fragment (final concentration 0.03 pmol). The protein concentration is indicated in pmol above each lane (left hand panel). In the right hand panel, the protein concentration was maintained at 2.4 pmol and unlabeled nifB promoter fragment was added as competitor (concentration indicated above each lane). (E) EMSA experiments demonstrating binding of E. coli σ⁷⁰-RNAP to the 50 bp nifB promoter DNA fragment (final concentration 0.03 pmol). The protein concentration is indicated in pmol above each lane (left hand panel). In the right hand panel, the protein concentration was maintained at 0.2 pmol and unlabeled nifB promoter fragment was added as competitor (concentration indicated above each lane).

Figure 4. E. coli σ⁷⁰-RNAP binds preferentially to the −35 region and −10 region of the nifB promoter of Paenibacillus sp. WLY78.

(A) Substitutions introduced in the nifB promoter sequence. The sequences of the −35 and −10 regions of the nifB promoter are underlined (Wt indicates the wild-type sequence). Base substitutions in the mutant promoter are indicated in red. (B) and (C) EMSA experiments comparing the binding of E. coli σ⁷⁰-RNAP to the wild-type nifB promoter fragment (panel B) with the mutant promoter fragment (panel C). The protein concentration is indicated above each lane. (D) DNase I footprinting of the interaction of E. coli σ⁷⁰-RNAP with the nifB promoter using an automated capillary sequencer. The top lane is an electropherogram obtained in the presence of σ⁷⁰-RNAP with the sequence protected from cleavage shown below. A control electropherogram obtained from a reaction containing BSA is shown in the bottom lane.

Figure 5. Expression of the Paenibacillus sp. WLY78 PnifB::lacZ promoter fusion is constitutive in E. coli.

Black bars indicate expression of ß-galactosidase driven by the nifB promoter; grey bars indicate the level of ß-galactosidase activity exhibited by the vector plasmid (pPR9TT) alone. Cultures were grown in nitrogen deficient medium, with 2 mM glutamate as nitrogen source, either anaerobically with the indicated concentrations of NH₄Cl (left panel) or with the indicated initial oxygen concentrations shown in the right-hand panel. Error bars indicate the standard deviation observed from at least two independent experiments.

Figure 6. Engineered E. coli strains and their nitrogen fixation abilities.

(A) Scheme showing the genetic organization of the engineered E. coli strains. (B) and (C) Nitrogenase activities of engineered strains and their deletion variants compared with Paenibacillus sp. WLY78 (bars marked as “WT”) and E. coli JM109 carrying the empty vector plasmid pHY300PLK (bars marked as “vector”). Strains were grown anaerobically in nitrogen-deficient conditions and the cultures were assayed either for acetylene reduction (panel B) or for ¹⁵N₂ incorporation (panel C). Error bars indicate the standard deviation observed from at least two independent experiments.

Figure 7. Effects of O₂ and NH₄ ⁺ on nitrogenase activity and nif gene transcription.

(A) and (B) Comparison of the acetylene reduction activities of Paenibacillus sp. WLY78 (panel A) and the engineered E. coli 78-7 strain (panel B), when cultures are grown in the presence of either oxygen or ammonium (at the initial concentrations shown on the y axis). Error bars indicate the standard deviation observed from at least two independent experiments. (C) and (D) Comparison of transcription of nifH and nifK as determined by RT-PCR in Paenibacillus sp. WLY78 (panel C) and E. coli 78-7 (panel D). Initial concentrations of ammonium and oxygen are indicated above relevant lanes. Lanes labeled “NH₄ ⁺,O₂” indicate that both 2 mM ammonium and 21% oxygen were present. Lanes labeled “+” indicate positive controls in which genomic DNA was used as template in the RT-PCR. Lanes labeled “−” indicate negative controls in which no reverse transcriptase was added to the RT-PCR reaction. In each case a parallel RT-PCR reaction was performed to detect the level of 16S rRNA, to provide a loading control (shown beneath relevant lanes).