Roles of the narJ and narI gene products in the expression of nitrate reductase in Escherichia coli.

Nitrate reductase, released and purified from membrane fractions of Escherichia coli, is composed of three subunits. Formation of the enzyme depends on induction of the nar operon, narGHJI, which is composed of four open reading frames (ORF). Previous studies established that the first two genes in the operon narG and narH encode the alpha and beta subunits, respectively, while formation of the gamma subunit, cytochrome bNR, depends on expression of the promoter distal genes. The narJ and narI genes were subcloned separately into plasmids where each was under the control of the nar promoter. Expression of these plasmids in a mutant which forms only alpha and beta subunits revealed that expression of the narI gene is sufficient to restore normal levels of cytochrome bNR, but expression of both genes is required for assembly of fully active, membrane-bound nitrate reductase. The amino acid composition, the N-terminal sequence, and the sequence of cyanogen bromide fragments derived from the isolated gamma subunit corresponds to that expected for a protein produced by the narI ORF. A protein corresponding to the narJ ORF did not appear to be associated with the purified nitrate reductase complex or with the complex immunoprecipitated from Triton X-100-solubilized membrane preparations. We conclude that narI encodes the gamma subunit (cytochrome bNR) and that while the product of the narJ gene is required for assembly of fully active membrane-bound enzyme it is not tightly associated with the active enzyme.

100-solubilized membrane preparations. We conclude that narl encodes the y subunit (cytochrome &R) and that while the product of the narJ gene is required for assembly of fully active membrane-bound enzyme it is not tightly associated with the active enzyme.
Nitrate reductase is a membrane-bound enzyme which is required for the utilization of nitrate as electron acceptor in anaerobic respiration in Escherichia coli. The purified enzyme is composed of three subunits a, p, y with approximate molecular masses of 143,000, 60,000, and 20,000 kDa, respectively (1-3).
The formation of nitrate reductase is under the control of the nar operon which is expressed only under anaerobic conditions and is induced to maximal expression by the presence of nitrate. Based on characteristics of mutants generated by transposon insertion, the nar operon was postulated to include the structural genes for the three subunits of nitrate reductase organized in the order promoter -narG (a) -nurH * This work was supported by Public Health Service Grant GM 19511 from the National Institute of General Medical Services. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
j To whom correspondence should be addressed Dept. of Biochemistry and Molecular Biology, P. 0. Box 20708, The University of Texas Medical School at Houston, Houston, TX 77225.
(p)narl (y) (4-6). Mutants with insertions in the narl region produced nitrate reductase which was active only with artificial electron donors such as reduced methyl viologen (MVH)' and was located primarily in the cytoplasm (4-7). Because this "soluble" nitrate reductase was composed only of the a and p subunits and the mutants which produced it lacked cytochrome b N R (7, 8), it was concluded that the y subunit was required for binding of nitrate reductase to the membrane as well as for the transfer of electrons from physiological electron donors involving membrane dehydrogenases such as formate or glyceraldehyde 3-phosphate.
We have recently shown by DNA sequencing that the narI region is composed of two open reading frames, designated narJ and narl, respectively (7). Both open reading frames were translated upon induction of the nar operon, but it was not possible to establish which gene encoded the y subunit of nitrate reductase or what the role of the additional gene product was in the expression of nitrate reductase activity.
In the studies presented here we establish that the fourth open reading frame of the operon (narI) encodes the y subunit and that while the product of narJ is also required for the formation of fully active, membrane-bound nitrate reductase, it does not appear to be associated with the active nitrate reductase complex.

MATERIALS AND METHODS
Strains and Plasmids-The bacterial strains used in this study and plasmids pSL962, pES203, and pMV4 have been described previously (7). Plasmids pES203.1 and pMV5 were constructed from plasmids pES203 and pSL962, respectively, as diagrammed in Fig Although plasmid pES203 produced only an intact narJ gene product, strains transformed with it tended to lyse when grown anaerobically, presumably due to the production of the narl::lacZ fusion protein. Plasmid pES203.1 was constructed by deleting much of the lac operon components as diagrammed in Fig. lA. After digestion with SstI and SmaI, blunt ending and ligation with T4 ligase, plasmid pES203.1 was isolated from ampicillin-resistant transformants of strain RK5274 and its structure confirmed by restriction mapping. This plasmid permitted normal anaerobic growth of transformed strains.
Plasmid pMV5 was constructed as diagrammed in Fig. lA by deleting all of nurH and most of narG and narJ, leaving nurl intact and still under the control of the nar promoter. Plasmid pSL962 was digested with BamHI and BglII and the mixture religated with T4 ligase. Plasmid pMV5 was isolated from an ampicillin-resistant transformant of RK5274 produced from the mixture, and its structure was confirmed by restriction endonuclease mapping.
Growth, Extract Preparation, and Assays-Cells were grown in L broth (9) with or without 1% nitrate and supplemented with the appropriate antibiotic in the case of plasmid-bearing strains. Ampicillin was used at 50 pg/pl and tetracycline at 20 pg/pl. Aerobic The abbreviations used are: MVH, reduced methylviologen; SDS, sodium dodecyl sulfate; HPLC, high pressure liquid chromatography. narJ and narI Gene Products 16157 cultures were grown on a rotary shaker and anaerobic cultures in stationary, filled bottles at 37 "C. Whole cell suspensions were prepared by centrifuging cultures for 10 min at 10,000 X g and resuspending at 10 g wet weight/100 ml in 50 mM potassium phosphate, pH 7.0. Crude extracts were prepared by passing the whole cell suspension through the French press two times at 15,000 psi and centrifuging 10 min at 10,000 X g to remove unbroken cells. The crude extract was centrifuged for 60 minutes at 100,000 X g to yield a supernatant fraction and a membrane pellet fraction which was suspended in 50 mM potassium phosphate, pH 7.0.
MVH-and formate-nitrate reductase activities were assayed as previously described (10, 11). The low temperature-reduced minusoxidized cytochrome spectra were determined on membrane fractions containing approximately 1 mg of protein/ml as previously described (12). The relative content of cytochrome ~N R was calculated as the peak height of the a band, i.e. at the emax, divided by the mg of protein present in the 1-ml sample. This calculation includes all cytochrome b components which contribute to the a band absorption and tends to minimize the changes specifically in cytochrome bNR.
Purification of Nitrate Reductase-MVH-nitrate reductase was released from the membrane pellet by deoxycholate treatment and concentrated by ammonium sulfate precipitation as previously described (13). This fraction was absorbed to a phenyl-Sepharose column and the column was eluted with 3 volumes of 1 mM Tris'HCl, pH 7.5, followed by a gradient in the same buffer of 0-0.8% Triton X-100. The active fractions which eluted in the gradient were pooled, concentrated by ultrafiltration with a PM-20 filter, and the buffer replaced with 5 mM potassium phosphate, pH 6.8, containing 0.1% Triton X-100. The concentrated fraction was observed to hydroxyapatite (Bio-Gel HPT), equilibrated with the same buffer. The enzyme was eluted from this column with a gradient from 5 to 250 mM potassium phosphate, pH 6.8, containing 0.1% Triton X-100. The active fractions were concentrated by ultrafiltration with a PM-10 filter, and the buffer was replaced with 20 mM Tris-acetate, pH 7.4. The concentrated fraction was finally purified by HPLC on a 0.9 X 25-cm column of 10 pm Vydac TP silica coated with polyethyleneimine (14) which was eluted with a 0-1 M sodium acetate gradient in 20 mM Tris-acetate, pH 7.4. The major activity peak was composed of undegraded a, p, and y subunits and had a specific activity of 30-50 units/mg of protein when assayed under standard assay conditions (11).
Isolation of Subunits-The a and p subunits were separated from y subunits as follows. The purified enzyme was heated at 60 "C for 20 min and then, after cooling, centrifuged at top speed in an Eppendorf centrifuge for 15 min. The y subunits aggregated under these conditions and was completely recovered in the pellet while the a and p subunits remained in the supernatant fraction in an active complex. To isolate the o( subunit, the supernatant fraction was lyophilized and the residue was dissolved in 6 M guanidinium.HC1, 100 mM dithiothreitol, and 100 mM sodium phosphate, pH 6.1. The 01 subunit was resolved from the other subunits by chromatography on a Sepharose CL4B column equilibrated and run with the same buffer. The fractions containing pure a subunits, as determined by SDS-polyacrylamide gel electrophoresis, were combined and dialyzed against 4 M urea in 200 mM Tris.HC1, pH 7.5, and then 0.2% SDS prior to N-terminal sequencing.
To isolate the j3 subunit, the lyophilized residue of the ap-containing supernatant fraction was dissolved in a solution containing 1% SDS and 10 mM dithiothreitol and heated 2 min at 100 "C. The cooled fraction was chromatographed on a Bio-Gel A-15m column equilibrated and run with 0.1% SDS. The fractions containing pure j3 subunit, as determined by SDS-polyacrylamide electrophoresis, were pooled and used directly for N-terminal sequencing.
To isolate the y subunit, the pellet fraction resulting from heating the purified enzyme at 60 "C was dissolved in a solution containing 1% SDS and 10 mM dithiothreitol. After heating 2 min at 100 "C, the fraction was cooled and centrifuged. The supernatant fraction was chromatographed on a Bio-Gel A-15m column equilibrated and run with 0.2% SDS. This procedure led to some loss of the y subunit, but a significant amount of pure y subunit, determined by SDS-polyacrylamide electrophoresis, was recovered in the included volume fractions of the column. Fractions containing the pure y subunit were pooled and used directly for amino acid analyses and N-terminal sequencing.
CNBr Cleavage of the y Subunit-To isolate y subunit for this procedure, purified nitrate reductase was suspended in 125 mM Tris. HC1, pH 6.8, and 10% glycerol supplemented with (per 100 pg of protein) 1 p1 of p-mercaptoethanol, 1 mg of dithiothreitol, and 0.1 mg of SDS. The suspension was incubated at 37 "C for 1 h. The treated fraction was subjected to SDS-polyacrylamide gel electrophoresis, and a band of gel containing the y subunit was excised and electroeluted in 10 mM Tris-acetate, pH 8.6, with 0.1% SDS. The eluted fraction was lyophilized, and the pellet was dissolved in 70% formic acid. The solution was carefully degassed, mixed with CNBr (100 pg/ pg protein) in 70% formic acid, and incubated for 8 h in the dark. The solution was mixed with 3 ml of water and lyophilized. Addition of water and lyophilization was repeated twice, and the residue was finally suspended in 0.5 ml of water and used directly for N-terminal sequence analysis.
Deblocking the y Subunits-Because direct N-terminal analysis revealed that the y subunit contained a blocked N-terminal, the y subunit preparation was partially deblocked by treatment with trifluoracetic acid (15). The y subunit prepared by gel elution was precipitated by mixing with an equal value of cold 50% trichloroacetic acid, incubation in ice for 30 min, and sedimentation in an Eppendorf centrifuge. The pellet was washed once with cold acetone, dried, mixed with 300 p1 of 25% trifluoracetic acid and heated for 2 h at 53 "C. The resulting suspension was used directly for N-terminal analysis.
Analytical Procedures-Hydrolyses of proteins and amino acid analyses were carried out as described by Walker and DeMoss (16). N-terminal sequence analyses were carried out as described by Sodergren and DeMoss (7).
SDS-polyacylamide gels prepared from a gradient of 7.5 to 15% acrylamide were prepared and run according to the procedures of Laemmli (17). The gels were stained with Coomassie Blue R-250.
obically in 30 ml of M63 minimal medium (17) to a density of 150 Immunoprecipitation of Nitrate Reductase-Cells were grown aer-Klett units. The culture was supplemented with 10 p~ sodium selenite, 10 p~ sodium molybdate, 1% potassium nitrate, and [35S]-methionine at 20 pC/pl. After sparging with 95% NI and 5% CO,, the culture was incubated for 60 min at 37 "C to induce nitrate reductase (6). Cells were harvested by centrifugation, washed two times with 50 p~ Tris. HCI, pH 7.4, and the cell pellet frozen overnight. The pellets were suspended in 5.0 ml of Tris.HC1, pH 8.0, passed through the French press two times at 1,500 psi and then centrifuged at 100,000 X g for 90 min. The supernatant fraction was used directly for immunoprecipitation, and the pellet was suspended in 5.0 p1 of 50 mM potassium phosphate, pH 7.3, with 2% Triton X-100, incubated at 37 "C for 30 min, and centrifuged at 100,000 X g for 90 min. The supernatant from the Triton X-100-treated pellet was the Triton X-100-released membrane fraction in the immunoprecipitation experiments.
The a subunit-specific antiserum used in the immunoprecipitation experiments was purified from a polyclonal antiserum prepared against purified nitrate reductase. Pure a subunit was separated from the other subunits by SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose. The antiserum was absorbed to the nitrocellulose strip containing the a subunit and antibodies absorbed to it were eluted at low pH (18). This a-specific antiserum reacted only with the a subunit in Western blots of either the crude or purified enzyme.
The supernatant and Triton X-100-released membrane fractions were incubated with varying amounts of antiserum at room temperature for 60 min. Protein A-Sepharose (200 mg/ml) was incubated with the mixture for 60 min on ice and then removed by centrifugation. The pellet was washed two times with 50 mM Tris. HCl, pH 7.5, containing 0.05% Nonidet P-40, 150 mM NaC1, 5 mM EDTA, and 0.02% sodium azide. The pellet was finally suspended in 40 pl of electrophoresis sample buffer and heated for 3 min at 100 'C. The sample was cooled, centrifuged, and the supernatant subjected to SDS-polyacrylamide gel electrophoresis on gradient gels containing 7.5-15% acrylamide. Autoradiography was carried out at -80 "C.

RESULTS
Complementation of a narJI Mutant with Subcloned narl and narJ-E. coli mutant RK5274 was isolated and characterized by Stewart and MacGregor (5) as a nar::TnlO mutant which produced only the a and ,6 subunits of nitrate reductase. In this mutant, nitrate reductase, assayed with reduced MVH as the electron donor, was located primarily in the cytoplasm, and the mutant was unable to reduce nitrate with physiological electron donors such as formate. This mutant was also shown by Hackett and Bragg (8) t o have an altered cyto-narJ and narl Gene Products chrome spectrum, indicating that cytochrome bNR was not produced under the conditions which induce nitrate reductase formation. Based on these observations it was concluded that RK5274 contained a TnlO insertion located 3' to the narG' and narH genes in the nurl gene, which was assumed to be the structural gene for cytochrome ~N R (5). This assumption was supported by the demonstration that plasmid pMV4 (Fig.  l ) , containing a subcloned fragment of the 3' end of the nar operon under the control of the nar promoter, complemented mutant RK5274, restoring normal levels of cytochrome ~N R and promoting the assembly of nitrate reductase on the membrane in a fully functional form (7). DNA sequencing of the 3' end of the operon, however, established that there are two open reading frames (7), designated narJ and narl respectively, in addition to the narG and narH genes which are located proximal to the promoter (6,19,21). Southern blots (22) revealed that the TnlO insertion in RK5274 was located downstream from narH, but it was not possible to discern whether the insertion was in the narJ or the narl open reading frame. We therefore provisionally designate mutant RK5274 and a narJI mutant.
In order to determine which of the two functions are required for complementation of RK5274, we subcloned the narJ and nurl open reading frames separately by constructing plasmids pES203.1 and pMV5 (Fig. 1). In both cases, as well as in plasmid pMV4, the subcloned genes remained under the control of the nar promoter region so that growth anaerobically in the presence of nitrate would induce cells carrying the plasmids to express the subcloned genes.
As previously shown (7), mutant RK5274 produced reduced levels of MVH-nitrate reductase and 80% of the activity was located in the supernatant fraction (Table I), in contrast to the parental wild type strain, RK4353, where the activity was associated mainly with the membrane fraction. The mutant strain was completely devoid of formate-nitrate reductase activity, reflecting the inability of the cytoplasmic nitrate reductase to accept electrons from membrane-associated dehydrogenases such as formate dehydrogenase.
Table I also shows the effects of transforming mutant RK5274 with plasmids which express the narJand narI genes. Transformation with plasmid pMV4, which expresses both the narJ and narI genes, restored both formate and MVHnitrate reductase activities to approximately 65% the wild type level and more than 80% of the MVH-nitrate reductase was associated with the membrane fraction. Transformation with plasmid pES203.1, which expresses nnrJ only, did not significantly change the levels of the nitrate reductase activities or the distribution of MVH-nitrate reductase between the supernatant and membrane fractions, indicating that the narJ product alone is not sufficient to restore the wild type phenotype. Transformation with plasmid pMV5, which expresses narl only, resulted in low but significant increases in  Relative contents of cytochrome bNR was calculated from the spectra in Fig. 2 and equal to the peak height of the a band of cytochrome b divided by the pg of proteins/ml in the sample  Table I. The frat-gel run as in lane 1; the Sample Was heated 3 min at 100" c in 2% tions were prepared as described under "Materials and Methods" and SDS before electroPhoreSiS. spectra determined at the protein concentration shown.

rnglrnl
Wavelength (nrn) the levels of both MVH-and formate-nitrate reductase activities. In addition the MVH-nitrate reductase activity was distributed equally between the supernatant and membrane fractions. These results indicate that both the narJ and narZ gene products are required for the formation of full levels of nitrate reductase and for the assembly of the enzyme on the membrane in a fully functional form; however, a significant amount of nitrate reductase can apparently be assembled on the membrane in the absence of the narJ gene product.
The reduced uersus oxidized spectra of the membrane fractions from the above strains are shown in Fig. 2 and the emar and relative content values for cytochrome b are summarized in Table I. As previously shown (11,12), under conditions of full nitrate reductase induction the emax for the a band of cytochrome b in the wild type strain was 556 nm. In mutant RK5274, the relative content was reduced about 50% and the tmax for the a band was shifted to approximately 558 nm. Transformation of the mutant with either plasmid pMV4 (narJ+narZ+) or pMV5 (narZ+) restored the emax for the a band and the relative content of cytochrome b to the wild type values. Transformation with plasmid pES203.1 (narJ+) appeared to affect neither the emax nor the relative content of cytochrome b found in RK5274. These results suggest that the expression of both narJ and narl are required for the assembly of fully active nitrate reductase but that a functional narJ and narl Gene Products narl gene is sufficient for the formation of the cytochrome b component of nitrate reductase.
Identification of the y Subunit Structural Gene-To determine directly which open reading frame encodes the y subunit, we resolved this subunit from purified nitrate reductase (Fig.  3, lane 1) by the procedure described under "Materials and Methods." As described by Chaudhry and MacGregor (3), this subunit is extremely hydrophobic and tends to aggregate irreversibly when heated. We therefore resolved the y subunit on a SDS-polyacrylamide gel after treatment at 37 "C with buffer containing 0.2% SDS (Fig. 3, lane 2) and isolated the subunit by elution from the gel (Fig. 3, lane 3 ) .
In Table I1 the amino acid composition of the purified y subunit is compared ot the amino acid compositions deduced for the theoretical products of the narJ and narl open reading frames. The composition of the y subunit appears to correspond closely to that deduced for the narl product and to be significantly different from that deduced for the narJproduct (see "Discussion"). Initial attempts to determine the N-terminal sequence for the isolated y subunit established that the subunit contained a blocked N terminus. Therefore, in order to derive sequence information the isolated subunit was subjected to cyanogen bromide cleavage as described under "Materials and Methods.'' The products of cyanogen bromide cleavage were extremely hydrophobic, as was the isolated subunit, and it was not possible to resolve individual peptides by either gel electrophoresis or HPLC. For that reason, the entire cyanogen bromide-cleaved mixture was subjected to sequential Edman degradation to derive sequence information.
In Table I11 the amino acid residues identified after each round of Edman degradation of the mixture resulting from cyanogen bromide treatment of the y subunit are compared to the residues expected for the cyanogen bromide fragments predicted from the nnrJ and nnrl sequences. There is excellent agreement between the theoretical and experimental data for the narl gene product and the y subunit. When fragments shorter than 5 residues were ignored, essentially all the residues expected for cyanogen bromide fragments derived from the NarI protein were present in each of the first 15 cycles of Edman degradation of the y subunit. The exceptions were residues which are poorly recovered (such as W, C, and T) and residues which are close to the carboxyl end of the fragments. In contrast, most of the residues expected for cyanogen bromide fragments from the NarJ protein were not found in the mixtures from each cycle of Edman degradation of the y subunit. Significantly, essentially every residue found in each cycle could be rationalized by the theoretical composition of the NarI protein while only a limited number of residues (12 of a total of 66) found corresponded to those expected from the theoretical composition of the NarJ protein.
Finally, it was possible to establish an N-terminal sequence for the y subunit after treatment of the isolated preparation with 30% trifluoroacetic acid, a procedure which has been shown to remove the formyl group from N-terminal blocked peptides (15). Although this hydrolytic procedure results in a high background at each Edman degradation cycle of the treated protein, the major amino acid residues released at each cycle could be clearly discerned. The sequence observed corresponded to that expected for the NarI protein as deduced from the nucleotide sequence of the narl open reading frame   (Table IV) and was distinctly different from that expected for the NarJ protein.
Attempts to Identify the NarJ Protein-We conclude from the above data that the narI gene encodes the y subunit of nitrate reductase. However, it was still unresolved where the NarJ protein is localized and what its role is in the expression of nitrate reductase activity.
T o examine the possibility that the NarJ protein is part of the nitrate reductase complex but that it had not been recognized previously because of its similarity in size (26.5 kDa) to that of the NarI protein, the y subunit (25.5 kDa) (7), we determined N-terminal sequences for isolated a and @ subunits and for the purified nitrate reductase complex (Table   IV). Both the isolated a and p subunits readily yielded Nterminal sequences with the a sequence corresponding to the first open reading frame of the nar operon (narC) minus the N-terminal Met (20, 21). The isolated y subunit contained a blocked N terminus and yielded no amino acid derivatives in multiple rounds of Edman degradation. The purified nitrate reductase complex (Fig. 3, lane I ) yielded a mixture of residues at each cycle which corresponded to an equal mixture of a and @ subunits and no other N-terminal sequences corresponding to either the narJ or narI sequence were detectable (Table IV). Therefore, any peptides present in purified nitrate reductase other than the a and @ subunits must contain blocked N termini. We previously demonstrated that a narJ-lac2 fusion produced a hybrid protein with an N-terminal sequence expected for the NarJ protein, indicating that this open reading frame produced a protein with an unblocked N terminus when the nar operon was induced (7). Since the narJ-lac2 fusion protein had an unblocked N terminus we assume that the NarJ protein would be similarly unblocked and therefore is not present in purified nitrate reductase.
It was also possible that the NarJ protein is part of the assembled nitrate reductase complex but is lost during the lengthly purification procedure. To examine this possibility, antibodies specific for the a subunit were purified by the procedure described under "Materials and Methods" and utilized to precipitate the nitrate reductase complex directly from Triton X-100-treated membrane fractions from 35S-labeled cells (Fig. 4). The antigen-antibody complexes were absorbed to protein A-Sepharose, and the labeled peptides were analyzed on a SDS-polyacrylamide gel. The labeled proteins were released by heating the protein A-sepharose pellet in 2% SDS for 0,3, and 10 min at 100 "C. In each case bands corresponding to the a, @, and y subunits in purified nitrate reductase were released. In the case of the unheated sample ( l a n e 2, Fig.  4), an additional band was present on the gels midway between the a and @ bands, which we have shown to be undenatured, active a@ complex (data not shown). As reported by Chaudhry and MacGregor (3), the apparent levels of y subunit were diminished by boiling in SDS, presumably due to its aggregation by heat and removal during centrifugation with the protein A-Sepharose pellet. In each case there is no indication of an additional band in the y subunit region; the radioactive bands in this region precipitated from the crude fraction corresponded precisely with the stained y subunit band in the purified complex.

DISCUSSION
It was previously established that the a and @ subunits of nitrate reductase are encoded by the first two genes in the nar operon, the narc and narH genes, respectively (4-6, 20, 21), and it was proposed from the phenotype of nar mutants lacking cytochrome b N R that a third gene, narl located 3' to the narc and narH genes, encoded the cytochrome b N R or y subunit of nitrate reductase (4, 5). Nucleotide sequencing, however, established that in addition to the narG and nurH genes there are two open reading frames at the 3' end of the operon which were designated narJ and narl, respectively (7). We conclude from the data presented here that the last open reading frame in the operon, narl, is the structural gene for the y subunit. This conclusion is based on the correspondence of the structure and properties of the isolated y subunit with those expected for a protein with the sequence and composition encoded by the narl reading frame. Based on theoretical amino acid compositions the NarJ protein would be a distinctly hydrophilic protein with an acid PI (a ratio of acidic to basic amino acid residues of approximately 2) while the NarI protein would be distinctly hydrophobic with a basic PI (a ratio of acidic to basic amino acid residues of approximately 0.5). The y subunit preparation isolated by Chaudhry and MacGregor (3) contained 45% hydrophobic amino acids and had an isoelectric point of greater than 9.5 which corresponds to the general composition of the y subunit which we isolated for this study (Table 111). It should be pointed out, however, that the amino acid composition reported for the y subunit by Chaudhry and MacGregor (3) differs significantly from the composition reported here. The final step in the isolation procedure employed in both studies was elution from an SDSpolyacrylamide gel. The two procedures differed in that we isolated a completely denatured protein band (Fig. 4) while Chaudhry and MacGregor isolated a partially denatured protein band which ran close to the ion front and may have been contaminated with other peptides. In any case, the amino acid composition of the y subunit preparation isolated in our studies (Table 11) corresponds closely to the composition expected for the narl open reading frame and is quite different from that expected for the narJ open reading frame (7). This correspondence between the narl gene and the y subunit is confirmed by the sequence information from the mixture of fragments generated by cyanogen bromide treatment and from the N terminus of the subunit which was partially deblocked by heating with triflouroacetic acid.
The cytochrome ~N R spectrum was restored in mutant RK5274 by the plasmid containing only the subcloned narl gene, and a low level of nitrate reductase was assembled on the membrane in a form capable of accepting electrons from formate.While the precise position of the transposon insertion in the nar operon of mutant RK5274 has not been established, the fact that both the narJ and narl genes are required to restore maximum levels of membrane-bound nitrate reductase argues that the insertion is in the narJ gene. This location was supported by Southern blot results with this strain (22) which indicated that the TnlO insertion was located 3' to the unique SstII site which is within the narJ gene (7). If the TnlO lies within narJ, the complementation results with narl show that the narJ protein is not absolutely required for expression of membrane-bound nitrate reductase activity. However, the complementation results with the cloned genes indicate that the NarJ protein is required for the assembly of full levels of membrane-bound enzyme suggesting that it may in some way facilitate the assembly process.
There is no evidence that the NarJ protein is directly associated with the nitrate reductase complex. As shown here it is not present in either the purified enzyme or in immune precipitates of the crude enzyme which had been solubilized by detergent treatment of membranes. Furthermore, it is not associated with immune precipitates of the soluble complex from extracts of RK5247 containing the plasmid expressing only the narJ gene.2 We cannot rule out the possibility, however, that it is loosely associated with nitrate reductase but lost during purification of during washing of the immune precipitates. It is also possible that the NarJ protein plays some other role in regulation or the uptake of nitrate, but further study will require that the protein be identified and its localization in the cell established.