JOURNAL OF BACTERIOLOGY, Nov. 1992, p. 7104-7111

Vol. 174, No. 22

0021-9193/92/227104-08$02.00/0 Copyright © 1992, American Society for Microbiology

Regulation of narK Gene Expression in Escherichia coli in Response to Anaerobiosis, Nitrate, Iron, and Molybdenum TASSIA KOLESNIKOW, IMKE SCHRODER, AND ROBERT P. GUNSALUS* Department of Microbiology and Molecular Genetics and the Molecular Biology Institute, 5304LS, University of California, Los Angeles, California 90024 Received 21 August 1992/Accepted 11 September 1992

The regulation of the narK gene in Escherichia coli was studied by constructing narK-lacZ gene and operon fusions and analyzing their expression in various mutant strains in response to changes in cell growth conditions. Expression of narK-lacZ was induced 110-fold by a shift to anaerobic growth and a further 8-fold by the presence of nitrate. The fnr gene product mediates this anaerobic response, while nitrate control is mediated by the narL, narX, and narQ gene products. The narX and narQ gene products were shown to sense nitrate independently of one another and could each activate narK expression in a NarL-dependent manner. We provide the first evidence that NarL and FNR interact to ensure optimal expression of narK. IHF and Fis proteins are also required for full activation of narK expression, and their roles in DNA bending are discussed. Finally, the availability of molybdate and iron ions is necessary for optimal narK expression, whereas the availability of nitrite is not. Although the role of the narK gene product in cell metabolism remains uncertain, the pattern of narK gene expression is consistent with a proposed role of NarK in nitrate uptake by the cell for nitrate-linked electron transport.

the N-terminal coding regions of narK and narX from pLK63 (21), into M13mpl9. BamHI and EcoRI recognition sites were then introduced into the M13IS4 phage DNA at the 27th codon of narK and the 11th codon of the divergently transcribed narX gene, respectively, by using site-directed mutagenesis procedures (24). This phage was designated M13TS1. The narK-lacZ gene and operon fusions were constructed by cloning the 440-bp EcoRI-BamHI DNA fragment (see Fig. 4) from phage M13TS1 into the lacZ gene and operon fusion vectors pRS414 and pRS415 to give pTS4 and pTS8, respectively (36). The narK-lacZ fusion contained on pTS4 occurs between the 27th codon of narK and the 9th codon of lacZ. The pTS16 and pTS17 narK-lacZ fusion plasmids were constructed in a manner similar to that described for pTS4 and pTS8, except that a 235-bp BamHI fragment from M13IS14 was used. The BamHI fragment was created by introducing a second BamHI site into M13IS13 that contained a 440-bp EcoRI-BamHI fragment at position 144 relative to the start of narK translation (see Fig. 4). The junction of each fusion was confirmed by double-stranded DNA sequence analysis (33). The gene and operon fusions were then transferred onto the specialized lambda vector, XRZ5, and introduced into the chromosome of MC4100 as previously described (19). Lysogens that contained a single phage were isolated and stocked for subsequent study. Similarly, the indicated narK-lacZ phages were introduced into A(narXL), narX, narQ, narL, chlD, chlA, chlE, and fnr strains (Table 1). The wild-type copy of narK located at 27 min in each strain was preserved intact since the narK-lacZ gene fusion was integrated at the lambda attachment site at 17 min on the chromosome. Construction of a narXi deletion strain. By site-directed mutagenesis, a NsiI site was introduced into plasmid pIS19 at a location 55 bp 5' of the translational start of the narX gene to give pIS21. Plasmid pIS19 contains the narXL genes within a 5-kb HindIII fragment. The 2.5-kb NsiI fragment containing the narXL region was then deleted from pIS21 to give pIS24, which contains the regions flanking narXL. The resulting 2.5-kb HindIII fragment was moved into plasmid

The narK gene in Escherichia coli, located at 27 min on the chromosome, is near the structural genes for nitrate reductase (narGHJI) plus two regulatory genes that are involved in nitrate-dependent gene expression, narX and narL (1, 30, 40). The narK gene was recently sequenced, and a role in nitrate transport was suggested on the basis of its similarity in protein sequence to a family of cytoplasmic membrane transport proteins (31). This role is supported by observations that nitrate uptake and metabolism are altered in narK mutants (9, 17, 31). In a previous study, narK expression was examined under anaerobic conditions and found to be impaired in strains defective in either narL or fnr (40). However, the effect of oxygen and nitrate on narK expression was not examined nor was the effect of varying other nutritional conditions. To address these questions, narK-lacZ fusions were constructed and inserted into the chromosome in single copy, and their expression was analyzed under a variety of conditions. The role of the narX, narQ, chlD, chlE, chU, fiur, fis, and himA gene products on narK-lacZ expression was documented as was the effect of limiting iron and molybdenum availability. The conditions optimal for narK expression are consistent with a role for NarK in nitrate transport and/or metabolism.

MATERIALS AND METHODS Bacterial strains, bacteriophages, and plasmids. The genotypes and origins of the E. coli K-12 strains, bacteriophage, and plasmids used are given in Table 1. Strains TK100 (hiNMA) and TK101 (fis) were constructed by P1 transduction of the respective mutations from SD1 and SD6 into MC4100/ XTS8 (7, 35). By using the same approach, the TK102 (narQ), TK103 [A(narXL) narQ], and TK104 (fur) strains were constructed (Table 1). Construction of narK-acZ gene fusions. Phage M13IS4 was constructed by inserting a 1.9-kb PstI fragment, containing *

Corresponding author. 7104

narX, narL, narQ, fnr, AND him CONTROL OF narK

VOL. 174, 1992

7105

TABLE 1. E. coli K-12 strains, bacteriophages, and plasmids Strain, phage, or

plasmid

Strains MC4100 PC100 IS2 IS4 LK23R35 LK4441 RCC3 RCC4 RK4353 RK5201 RK5200 TK100 TK101 TK102 TK103 TK104 SD1 SD6

W3110ftr::TnS Phages XTS4 XTS8 XTS16 XTS17 M13mpl9 M13IS4 M13TS1 M13IS13 M13IS14 Plasmids pACYC184 pIS19 pIS21 pIS24 pIS25 pLK63 pLK634 pRS414 pRS415 pTS4 pTS8 pTS16 pTS17

Genotype or phenotype

Origin

MC4100 MC4100 IS2 MC4100 MC4100 RCC1 RCC2 MC4100 RK4353 RK4353 MC4100 MC4100 MC4100 IS4 MC4100 MC4100 MC4100 W3110

F- araDl39 A(argF-lac) U169 rpsL150 reLA1 flbS301 deoCi ptsF25 rbsR fnr250 zcj-637::TnlO AnarXL::TnlO kan AnarXL::TnlO kan recA TnlO::tet narL TnJOA16A17::kan chiD TnJOA16A17::kan narQ TnlOA16A17::kan recAS6 A(narXL) narQ TnlOA16A17::kan recAS6 srl::TnlO AlacU169 araD139 rpsL gyrA chlE201::Mu cts chL4200::Mu cts himAA82 XTS8 fis-767 XTS8 narQ XTS4 A(narXL) narQ XTS4 fur::TnS ATS4 himAA82 fis-767

fur::TnS cj (narK-lacZ) lacY+ lacA+ (Hyb) (gene fusion)

pTS4 pTS8 pTS16 pTS17

4'(narK-lacZ+) lacY+ lacA+ (operon fusion) ?)(narK-lacZ) lacY+ lacA+ (Hyb) (gene fusion) 4(narK-lacZ+) lacY+ lacA+ (operon fusion)

M13mpl9 M13IS4 M13TS1 M13IS13

1.9-kb PstI narK fragment Same as that of M13IS4, but with BamHI and EcoRI sites 440-bp BamHI-EcoRI fragment Same as that of M13IS13, but with additional BamHI site

pTZ19U pIS19 pIS21 pMak7O5 pACYC184 pACYC184

narX+ narL+ Ampr Same as that of pIS19, but with an NsiI site Ampr A(narXnarL) Ampr 2.5-kb NsiI fragment Cmr narX+ narL+ narK Cmr narX narL+ Cmr lacZ lacY+ lacA+ Ampr lacZ+ lacrY lacA+ Ampr 4'(narK-lacZ) lacY+ IacA+ (Hyb) Ampr 4D(narK-lacZ+) lacY+ lacA+ Ampr 4(narK-lacZ) lacY+ lacA+ (Hyb) Ampr 4D(narK-lacZ+) lacY+ lacA+ Ampr

Cmr Tcr

pRS414 pRS415 pRS414 pRS415

pMak7O5 (16) to give pIS25, which was transformed into strain MC4100/XLK1 (20). The narXL region of the chromosome was deleted by using the allele replacement procedure described previously (16). A kan gene was placed near the narXL deletion by P1 transduction from a pool of cells with mini::kan elements located randomly around the chromosome to give IS2/VLK1. Southern blot analysis of total DNA isolated from IS2 showed that the 2.5-kb narXL region was absent from the chromosome (data not shown). The recA allele was introduced into this strain by P1 transduction to give strain IS4/XLK1 (7). Cell growth. For plasmid, phage, and strain manipulations, cells were grown in Luria broth or on solid media. When required, ampicillin, chloramphenicol, and 5-bromo-4chloro-3-indolyl-p-D-galactopyranoside (X-Gal) were added to the medium at concentrations of 80, 30, and 40 mg/liter, respectively. For ,B-galactosidase assays, cells were grown

Source or reference

35 6 This study This study 20 6 3 3 39 39 39 This study This study This study This study This study Silvia Daire Silvia Daire 8

This study This study This study This study 25 This study This study This study This study 2 This study This study This study This study 21 21 36 36 This study This study This study This study

in a minimal medium containing glucose (40 mM) and Casamino Acids (0.1%) at pH 7.0, unless otherwise indicated (6). Aerobic growth and anaerobic growth were performed as previously described (6). When desired, sodium nitrate and sodium molybdate were added at final concentrations of 40 mM and 100 FjM, respectively (22). High aeration of cultures during aerobic growth was accomplished by shaking 10-ml culture volumes in 150-ml flasks at 250 rpm (4). Flasks or tubes containing the indicated medium were inoculated from overnight cultures grown under the same conditions, and the cells were allowed to double four to five times in mid-exponential phase prior to being harvested for enzyme analysis (optical density at 400 nm, 0.40 to 0.45; Kontron Uvikon 810 spectrophotometer). Cell growth with carbon sources other than glucose was achieved by substituting glycerol, sorbitol, xylose, gluconic acid, succinate, or lactate

7106

KOLESNIKOW ET AL.

J. BACTERIOL.

TABLE 2. Effect of oxygen, nitrate, and molybdenum on narK-lacZ gene expression' [3-Galactosidase activity"

Addition

MC4100 (wt)C

02

N03

+ + -

+ +

PC100 (fnr)

LK23R35 (narL)

LK4441 (chlD)/XTS4

XTS4

XTS8

XTS4

XTS8

XTS4

XTS8

-Mo

+Mo

23 73 2,560 21,400

26 70 3,430 22,700

25 28 43 76

30 31 54 77

26 23 350 378

35 31 483 562

23 29 3,090 9,240

43 75 1,970 17,400

a Cells containing the indicated narK-lacZ fusion were grown in a glucose minimal medium either aerobically or anaerobically as described in the text. The relevant genotypes are indicated in parentheses. Where indicated, sodium nitrate was added at an initial concentration of 40 mM whereas sodium molybdate was added at 100 ,LM. b Activity is expressed in nanomoles of ONPG hydrolyzed per minute per milligram of protein. c wt, wild type.

(40 mM) for glucose in the minimal medium containing 40 mM sodium nitrate and 0.1% Casamino Acids. P-Galactosidase assay. ,B-Galactosidase levels were determined by hydrolysis of ortho-nitrophenyl-3-D-galactopyranoside (ONPG) as previously described (6). Protein concentration was estimated by assuming that a culture A6. of 1.4 corresponds to 150 ,ug of protein per ml (27). Units of ,-galactosidase are expressed as nanomoles of ONPG hydrolyzed per minute per milligram of protein unless otherwise indicated. An extinction coefficient for ONPG of 0.0045 mM-1 cm-1 was used (27). The P-galactosidase values presented are the average of at least three independent experiments. Values did not vary more than +5% from the mean except for the chlA and chlE experiments which showed +10% variation. mRNA manipulations. mRNA was isolated as previously described (28). Primer extension reactions were performed with a synthetic oligonucleotide complementary to positions +94 to +74 of the narK message relative to the predicted start of translation (31). DNA sequencing reactions were performed by using an appropriate template to provide a nucleotide sequence ladder for comparison. Materials. ONPG, 2,2'-dipyridyl, and ampicillin were purchased from Sigma Chemical Co. (St. Louis, Mo.). Sodium nitrite, sodium nitrate, and sodium molybdate were purchased from Fluka Chemical Co. (Ronkonkoma, N.Y.). All other reagents used were of reagent grade. RESULTS Effect of oxygen and nitrate on narK-lacZ gene expression. To determine how oxygen and nitrate affect narK gene expression, narK-lacZ gene (XTS4) and operon (XTS8) fusions were constructed and inserted in single copy into the chromosome of MC4100. After cell growth under aerobic or anaerobic conditions and in the presence or absence of nitrate, cells were harvested in the mid-logarithmic growth phase and ,B-galactosidase assays were performed. Expression of the narK-lacZ gene fusion, XTS4, was lowest when cells were grown aerobically (Table 2) and was elevated about threefold by the addition of nitrate. During anaerobic growth in the absence of nitrate, gene expression increased 110-fold relative to aerobic growth. When nitrate was also present during anaerobic conditions, an overall 930-fold increase in gene expression relative to that under aerobic growth without nitrate was seen. Similar results were observed for the narK-lacZ operon fusion strain, XTS8, under all of the conditions tested (Table 2). Effect offnr and narL mutations on narK-lacZ gene expres-

sion. The contribution of theffnr gene product on narK-lacZ expression was examined by introducing the gene (XTS4) and operon (XTS8) fusions into a fnr strain, PC100. During aerobic growth in the absence of nitrate, narK-lacZ expression in PC100 did not vary significantly from that seen in the wild-type strain (Table 2). However, the threefold increase in activity, observed in response to nitrate, exhibited by the wild-type strain was abolished in the fnr strain. Anaerobic induction of narK-lacZ expression was also nearly abolished in theftzr mutant. Only a modest twofold increase in activity was seen upon addition of nitrate. In comparison with the wild-type strain, a narL mutant defective in nitrate regulation of the frdABCD and narGHJI respiratory genes was severely impaired in the anaerobic activation of narK-lacZ expression (i.e., 14- versus 110-fold for narL mutant and wild type, respectively). When nitrate was present, narKlacZ expression was not further increased, whereas an eightfold induction was seen for the wild-type strain. Thus, the approximately 900-fold induction of narK-lacZ expression was greatly impaired by a defect in either the narL orfnr gene. Effect of chL4, chlE, and chiD mutations on narK-lacZ gene expression. Molybdate ions are necessary for normal nitrate regulation offrdABCD, narGHJI, and dmsABC gene expression by the narX and narL gene products (6, 18, 22). To establish whether molybdate ions are also involved in narK gene expression, the narK-lacZ gene fusion (ATS4) was introduced into chU4, chlE, and chlD strains and its expression was examined (Tables 2 and 3). A chlD strain (LK4441), which is defective in molybdate accumulation, showed a 40%-reduced level of narK-lacZ expression relative to the TABLE 3. Expression of a narK-lacZ gene fusion in and chlE strainsa

Il-Galactosidase activity"

Addition

02 + +

chUA

02 N03-

N03 +

RK4353

RK5200

RK5201

33 67

26 621

37 65 33,900 41,900

(wt)c

(chU)

(chlE)

6,920 29,100 + 26,800 39,500 a Cells containing XTS4 were grown in a glucose minimal medium, and, where indicated, nitrate was added at an initial concentration of 40 mM. The three strains are isogenic and differ from the chlD and wild-type strains listed in Table 2. b Activity is expressed in nanomoles of ONPG hydrolyzed per minute per milligram of protein. c wt, wild type.

VOL. 174, 1992

narX, narL, narQ, fnr, AND him CONTROL OF narK

7107

15 °-) 0

r-10

>10 ._

5

5

01

I 0

FeCI2

+ + + FIG. 1. Effect of iron limitation on narK-lacZ expression. Activity is expressed in micromoles of ONPG hydrolyzed per minute per milligram of protein. 2,2'-Dipyridyl was added at a final concentration of 200 FM, and FeCl2 was added at 80 FM.

di-pyridyl

J ' ' 1 5 40 Concentration (mM)

FIG. 2. Effect of nitrite and nitrate on narK-lacZ expression. Activity is expressed in nanomoles of ONPG hydrolyzed per minute per milligram of protein. Sodium nitrate (0) or sodium nitrite (0) was added at the concentration indicated. were grown anaerobically in a glucose minimal medium with 0, 0.25, 1, or 5 mM sodium nitrite. Expression varied only

parental strain when cells were grown anaerobically in the presence of nitrate. Expression was nearly restored to wild-type levels when molybdate was added (Table 2). Defects in either the chU or the chlE gene products, which are involved in formation of the molybdo-pterin cofactor and are thus necessary for nitrate reductase synthesis and catalytic activity, resulted in levels of narK-lacZ expression when cells were grown anaerobically without nitrate that were significantly higher than levels of expression of the isogenic parental strain RK4353 (Table 3). When nitrate was present under these conditions, the level of expression of the fusion in the chU and chlE mutants was also greater than that seen in the parent. Interestingly, narK-lacZ expression was also elevated in the chU mutant when grown aerobically in the presence of nitrate. Thus, narK expression appears to be dependent on molybdate availability as well as on the ability of the cell to synthesize a molybdo-pterin cofactor. Effect of iron limitation on narK-lacZ expression. To determine the effect of iron limitation on narK expression, the iron chelator 2,2'-dipyridyl was added to MC4100/XTS4 cells grown anaerobically with nitrate present. Expression of narK-lacZ was reduced by 87% compared with that of cells grown without the iron chelator (Fig. 1). When ferrous iron (80 ,uM FeCl2) was added to the culture medium in addition to 2,2'-dipyridyl, a near-complete restoration of narK-lacZ expression was seen. Affur mutant (TK104/XTK4) which is defective in iron-dependent regulation of many iron uptake systems in E. coli (8) was also examined with respect to narK-lacZ expression. Compared with that of the wild-type strain, narK-lacZ expression was unchanged in the fur background (data not shown). Effect of other environmental stimuli on narK-lacZ gene expression. Expression of narK was also examined after growth with various carbon substrates (sorbitol, xylose, glycerol, succinate, lactate, or gluconic acid) substituted for glucose as the source of carbon and energy. There was less than a 20% variation in 3-galactosidase levels in cells grown with any of these compounds in place of glucose (data not shown). When cyclic AMP (10 mM) was added, narK-lacZ expression was not significantly affected. To test the effect of nitrite on narK-lacZ expression, cells

three- to fourfold (Fig. 2). This indicates that nitrite availability does not significantly affect narK expression relative to nitrate availability. Effect of mutations in narX and narQ on narK-lacZ expression. The narX and narQ genes have each been shown to be involved in nitrate-dependent activation of narGHJI gene expression and repression of frdABCD and dmsABC gene expression (3). To determine how each gene product affects narK-lacZ expression, we introduced XTS4 into strains defective in narX, in narQ, or in narX narQ. Under anaerobic growth conditions, narK-lacZ expression was relatively unchanged in a narQ strain (TK102/ATK4) (Table 4). When a multicopy narL+ plasmid (pLK634) was introduced, the nitrate response was about the same as that of the wild-type strain (16-fold). However, a narXnarL strain (IS4/XTS4) was defective for nitrate activation. Introduction of a narL+ plasmid (pLK634) caused elevated narK-lacZ expression, but only a twofold response to nitrate addition was exhibited. A narXL narQ triple mutant (TK103/ATS4) was also severely impaired for nitrate regulation. Again, this regulaTABLE 4. Effect of narQ and narX alleles on nitrate-dependent activation of narK-lacZ expression p-Galactosidase Strain (plasmid)"

Relevant genotype

activityb -NO3- +NO3

MC4100/XTS4(pACYC184) TK102VXTS4(pACYC184) TK102/XTS4(pLK634) TK10V2XTS4(pLK63) IS4/XTS4(pACYC184) IS4/XTS4(pLK634) TK103/XTS4(pACYC184) TK103/XTS4(pLK634) TK103/XTS4(pLK63)

Wild type narQ narQ (pnarL+) narQ (pnarX+L+) narXL narXL (pnarL+)

667 444 839

1,410

12,100 15,200 13,700 7,620

298

180

2,380

5,130

210 166 narXL narQ narXL narQ (pnarL+) 2,320 3,390 narXL narQ (pnarX+L+) 1,200 5,200 a Cells containing a narK-lacZ reporter fusion, ATS4, and the indicated plasmid were grown anaerobically with 30 Ag of chloramphenicol per ml. Where indicated, nitrate was added at an initial concentration of 40 mM. b Activity is expressed in nanomoles of ONPG hydrolyzed per minute per milligram of protein.

_.

_-'._ ':

7108

KOLESNIKOW ET AL.

J. BACTERIOL.

C G T A l 2

results suggest that the narX and narQ gene products can each activate narK-lacZ expression in response to nitrate availability, although NarQ appears to be less effective than NarX. For reasons not understood, the presence of plasmid pACYC184 or its derivatives in the strains described above reduced narK-lacZ expression by about one-half under the conditions tested. Location of the in vivo narK mRNA 5' terminus. Primer extension experiments were performed to locate the initiation site of narK transcription. Cellular mRNA was prepared from cells grown anaerobically in media that either contained or lacked nitrate. A single 5' mRNA terminus which corresponds to a position 26 nucleotides before the translational start of narK was identified under both conditions (Fig. 3). During anaerobic cell growth conditions when nitrate was present, the amount of the narK mRNA was elevated 8- to 10-fold relative to when nitrate was absent, in good agreement with the narK-lacZ gene fusion data (Table

_

,.

t..FK

..

%r^+,,slw,

.._S

u5S+wt- ..

_ _

_ _

_

w

_ t

-__wl

_,,,@ut_ A".

.E,,ll__ _

'*

_,_.9 ;

v_

_

_-

_

.

._tes;..........

._ _

'.,. .:....':

_

*|0N'

''-*

_.

_.

GC TA

CG TA GC T AAT GC CG TA TA TA GC CG TA CG AT

._*

.._W

;-C_

b:. '_ ,.._ve

FIG. 3. Location of the in vivo mRNA 5' terminus of narK. Lanes 1 and 2 represent primer extension reactions with mRNA prepared from cells grown anaerobically with and without nitrate (40 mM). Lanes C, G, T, and A show the DNA sequencing reaction products from the corresponding region upstream of narK.

tion was NarL dependent as evidenced by the ability of a multicopy narL+ plasmid (pLK634) to partially restore narK-lacZ expression in the narXL narQ strain. The addition of a narX+ narL+ plasmid (pLK63), however, restored a modest fourfold increase in nitrate control. Together, these

L inarX

0.1

2). Effect of upstream DNA, IHF, and Fis on narK-kacZ expression. To determine whether DNA upstream of position -144 relative to the narK transcription initiation site was necessary for anaerobic nitrate expression, narK-lacZ gene (XTS16) and operon (XTS17) fusions that contained a 235-bp BamHI promoter fragment were constructed (Fig. 4) (see Materials and Methods) and analyzed as described above for XTS8 (Table 5). An approximate 100-fold anaerobic induction of narK expression in MC4100/XTS8 was also observed in MC4100/XTS17, although the level of I-galactosidase was reduced considerably under either condition. The fivefold nitrate induction seen in a wild-type strain under aerobic or anaerobic growth was completely abolished when DNA upstream of position -144 was removed (Table 5). Similar results were observed for narK-lacZ gene fusions XTS4 and XTS16 (data not shown).

0.2

0.3

0.4

1I

AI I

I

j

EJu

0.5

narK

U

-35 -10 VW 5AAAA \

-8.0

-60

-20

-40

+1

+20

TAAATATCAATGATAGATAAAG TTATCTTATCGTTTGATTTACATCAAATTGCCTTTAGCTACAGACACTMAGGTGGCAGACATCGAAACGAGTATCAGAGGTGTCTATGAGTCACTCA '''ITT6ACAI

ITTGATnnnnATCAAI

-3

ITATAAT -10

M S H S

FNR Box FIG. 4. Nucleotide sequence at the narK promoter region. The DNA sequence shown in the lower portion of the figure is numbered relative to the 5' terminus of the narK mRNA (see Fig. 3). The adenine residue located 26 bp before the narK translational start site is indicated as position + 1. The consensus sequence of the RNA polymerase recognition sequences in the -35 and -10 regions are shown below the DNA sequence. Vertical dashes shown above a consensus FNR binding sequence (14) indicate sequence identity to the proposed narK FNR-binding site (31). The stippled box indicates the location of a putative NarL box (31). The wavy line represents narK mRNA.

narX, narL, narQ, fnr, AND him CONTROL OF narK

VOL. 174, 1992 TABLE 5. Effect of upstream DNA sequence and the himA and fis gene products on narK-lacZ gene expression' P-Galactosidase activity"

Addition MC4100

02

N03

ATS8

(wtr ATS17

TrU00

(himA)/XTS8

TK101 (fis)/XTS8

+ 8 56 52 40 + + 133 142 207 5 723 4,710 1,570 3,210 + 500 21,000 9,420 7,010 a Cells containing the indicated narK-lacZ phage were grown in a glucose minimal medium either aerobically or anaerobically as described in the text. Sodium nitrate was added at an initial concentration of 40 mM. The relevant genotypes are indicated in parentheses. bActivity is expressed in nanomoles of ONPG hydrolyzed per minute per milligram of protein. I wt, wild type.

To test whether either the himA orfis gene products affect narK-lacZ expression, mutations in the respective genes were introduced into the operon reporter fusion strains (see Materials and Methods). In the himA strain, narK-lacZ expression was reduced by about threefold during anaerobic growth with or without nitrate relative to that in MC4100. Anaerobic expression of narK-lacZ in the fis strain was also reduced but by about twofold relative to that in the wild-type strain. Thus, for optimal narK-lacZ expression, the himA and fis gene products in addition to DNA sequences upstream of the narK promoter are necessary.

DISCUSSION In this study, expression of the narK gene in E. coli was shown to be regulated by nitrate, oxygen, iron, and molybdenum availability. Furthermore, narK expression was shown to be induced more than 100-fold by the fnr gene product in response to anaerobiosis, as demonstrated by the near-complete loss of narK-lacZ expression in the cells defective in fnr (Table 2). In a narL strain, narK-lacZ expression was also severely impaired during anaerobic growth, even when no nitrate was present in the growth medium (Table 2). Induction of the narK gene expression is also dependent on the narX and narQ gene products (Table 4). According to the models of Kalman and Gunsalus (22) and Chiang et al. (3), the narL gene product (NarL) functions as a response regulator to activate narK transcription, while the narX and the narQ gene products function as sensortransmitters to detect and report nitrate availability (3, 15, 30, 41). Interestingly, narX and narQ gene products appear to each activate NarL when nitrate is present but not to the same degrees. NarX appears to provide a wild type-like nitrate control, while NarQ is only partially responsive (Table 4). The reason for this differential control is unknown, and it may be indicative of the cells need for two sensortransmitters in the narX, narQ, and narL two-component system. Interestingly, mutations in narQ and/or narX cause lower basal narK-lacZ expression when nitrate is absent. Further studies are in progress to explore the basis of these observations. As anticipated (21), when a multicopy plasmid containing narL + was introduced into a AnarXL narQ strain, narK-lacZ expression was elevated to an intermediate level regardless of nitrate availability. Thus, production of elevated levels of the response regulator (NarL) apparently stimulates narK expression when the sensor-transmitter proteins are both defective. It is noted that residual nitrate

7109

control remains in a narX narQ mutant. The reason for this is not presently understood biochemically or genetically. FNR is essential for anaerobic induction of narK expression (Table 2). The 110-fold difference in aerobic versus anaerobic expression in the wild-type strain is reduced to 2-fold in the mutant. FNR is also required to mediate narK activation in response to nitrate as evidenced by the degree of narK-lacZ expression in aftnr strain (Table 2). A threefold increase in aerobic expression seen in response to nitrate in the wild-type strain is lost. Anaerobically, the nitrate effect is only 1.5-fold in the fnr mutant compared to 8-fold in the parent strain. Conversely, it appears that NarL is essential for FNR function. In a narL strain, nitrate activation of narK expression was abolished while the anaerobic induction was greatly impaired (14-fold versus 110-fold in the wild type). Thus, we provide the first evidence that NarL and FNR interact to ensure optimal expression of narK. Additional regulatory proteins are involved in this process (see below). Because the integration host factor protein (IHF) was demonstrated to be essential for nitrate-dependent anaerobic expression of the nitrate reductase (narGHJI) operon (34), we also tested its role in nitrate-dependent narK-lacZ expression (Table 5). In a himA mutant, narK-lacZ expression was reduced by approximately one-half to two-thirds under most conditions tested. Apparently, IHF binding to narK regulatory DNA is required for full activation of narK expression in response to anaerobiosis and to nitrate. This protein, which is essential for normal expression of narKlacZ in vivo, apparently binds to narK regulatory DNA site(s) and may bend it as has been demonstrated for IHF regulation of other genes in E. coli (32, 42). The Fis protein appears to also affect anaerobic and nitrate activation of narK expression in a way similar to that of IHF. How Fis acts in this process is unclear but may also involve a change in DNA topology on the basis of the known DNA-bending properties of this protein (23). Molybdenum transport appears to be essential for full expression of narK (Table 2), as evidenced by the results of the chiD mutant studies. The pattern of molybdenum-dependent regulation of narK gene expression is similar to that observed for frdABCD, dmsABC, and narGHJI expression (6, 18, 22). Limiting amounts of molybdate in the medium impairs the ability of narX, narQ, and narL to control in response to nitrate (6, 18, 22). Roles for the NarL receiverregulator protein (6, 18, 22) and the NarX sensor-transmitter (6, 22) in molybdenum detection have been demonstrated for the narXLQ gene regulatory system (6, 18, 22). Interestingly, chUA and chlE mutations, which affect synthesis of a molybdo-pterin cofactor, do not reduce narK expression but increase it when cells are grown anaerobically either with or without nitrate (Table 3). Although a similar pattern of gene expression has been reported for the dmsABC genes (6), the basis for this observation is unclear. Because an active nitrate reductase requires molybdo-pterin cofactor production for assembly of the mature complex from the apoenzyme precursor, it is interesting that the regulatory system for mediating nitrate control also extends to sensing an essential metal and/or cofactor. Since narK is regulated in response to many of the same stimuli that affect narGHJI expression, perhaps the function of NarK is to support anaerobic electron transport to nitrate. Iron availability also affects narK expression (Fig. 2). Whether the narK gene product requires iron directly for its function or, alternatively, whether other proteins with which it may interact do is unclear. It is conceivable that NarK functions as a nitrate transporter whose synthesis would not

7110

KOLESNIKOW ET AL.

be needed unless the cell is also able to synthesize a functional nitrate reductase. This respiratory enzyme contains 16 nonheme iron molecules per complex in addition to 4 molecules of cytochrome b type heme and molybdenumcontaining pterin (25). The reduced level of narK-lacZ expression seen during iron limitation is similar to that for the anaerobic respiratory genes narGHJI, dmsABC, and frdABCD (5). This control appears to be independent of the fiur gene product and supports the proposal that an additional level of iron control exists in E. coli which functions upon severe depletion of iron from the medium (5). Whether this effect is related to the presence of iron in the Fnr protein, as has been demonstrated by several laboratories (29, 37), is suggestive. A molecular basis for this control has yet to be established. Primer extension experiments indicate that the narK gene contains one major promoter (Fig. 3) and the resulting narK transcript contains an mRNA leader sequence of 25 nucleotides (Fig. 4). Comparison of the narK gene and operon fusion data permits an assessment of the translational efficiency of narK+ and lacZ+ (Table 2), where the translational efficiency of narK is similar to that of the lacZ+ gene. The abundance of the narK message in anaerobic cells grown in the presence and absence of nitrate also correlates well with the results of the gene fusion data. The nucleotide sequence preceding the transcriptional start of the narK gene contains putative binding regions for RNA polymerase at positions -10 and -35, suggesting that binding of a sigma 70-containing RNA polymerase is required for narK promoter expression. A putative FNR-binding site for anaerobic activation of narK transcription is located at positions -34 to -48 and shows considerable DNA sequence similarity to otherfnrcontrolled genes (Fig. 4) (14). This sequence was also noted by Noji and coworkers (31). A putative nitrate regulatory site within the narK regulatory region was recently proposed by Noji et al. (31). This nine-nucleotide sequence (TAC TCCTTA) located 199 bp before the start of transcription of narK (Fig. 4) is similar to a 6-bp NarL box (TACTCC) predicted for narGHJI activation by Dong et al. (10). When the putative NarL box, located 199 bp upstream of the translation start of narK, was deleted (Fig. 4 and Table 5), the nitrate-dependent activation of narK-lacZ expression was abolished. This finding establishes the requirement for upstream DNA sequences for the NarL-dependent transcriptional activation process in response to nitrate. A mechanism that accounts for this process is unclear, but it may involve an alteration of the DNA topology at the narK promoter region due to IHF bending of the DNA as is the case for other IHF-dependent promoters (12). This event would, conceivably, position the upstream NarL-binding site and bound NarL protein close to the FNR and RNA polymerase proteins to provide the necessary conditions for narK transcription. A similar mechanism has been proposed for the IHF-dependent expression of the narG promoter

(34). The narK gene product was originally proposed to be involved in nitrate regulation of nitrate reductase and fumarate reductase on the basis of the phenotype of a narK strain on anaerobic indicator media (39). However, narK mutants were later shown to have no effect on nitrate-dependent control of fumarate reductase (frdABCD) gene expression (19). Estimation of the intracellular NarK protein concentration (calculated by the method of Grove and Gunsalus [13]) indicates that it may be present in relatively high amounts during anaerobic cell growth when nitrate is present (i.e., excess of 25,000 molecules per cell). This is consistent with

J. BACTERIOL.

a metabolic role rather than a regulatory role for NarK. In any event, its function appears to be related to nitrate metabolism. Notably, nitrite did not substitute for nitrate and cause activation of narK gene expression. In addition, the carbon source did not affect narK expression. NarK synthesis does not occur optimally if either molybdenum or iron is limiting in the cell. Thus, narK transcription correlates well with the synthesis of an active nitrate reductase complex. An intermediate level of NarK synthesis in cells grown anaerobically without nitrate suggests that its physiological role is to function in the absence of oxygen. Cells are apparently poised anaerobically in the absence of nitrate but are further stimulated by nitrate availability. The role for NarK as a putative nitrate transporter would thus facilitate nitrate uptake and subsequent reduction via respiration. This process is energetically favorable to growth by anaerobic fermentation. The pattern of oxygen and nitrate regulation of narK expression is similar to that seen for the narGHJI genes of E. coli (3, 38). However, the magnitude of anaerobic induction of narK expression by FNR is greater (100-fold) than the induction of narGHJI (10-fold). In contrast, the induction of narK in response to nitrate is less than that of narGHJI (10-fold versus 80-fold, respectively). For nitrate control of the frdABCD and dmsABC operons, the narXLQ gene products mediate 30- and 12-fold repression, respectively (6, 11, 19). Thus, the narX, narQ, and narL gene products can mediate a wide range of positive and negative gene regulation to coordinate cellular energy generation in the cell. They regulate not only the choice of electron transport pathways to be used but also the acquisition of an essential metal, molybdate, which is required for synthesis of several enzymes in the respiratory pathways. On the basis of the present studies, they also appear to control nitrate uptake from the cell exterior for nitrate respiration. ACKNOWLEDGMENTS We thank Robin Chiang and Sylvia Daire for providing bacterial strains and Peggy Cotter for technical advice. This work was supported in part by Public Health Service grant AI21678 from the National Institutes of Health. I.S. was supported by a grant from the Deutsche Forschungsgemeinschaft. T.K. was a trainee of a Genetics Training Grant GM07104 from the Public Health Service. REFERENCES

1. Bachmann, B. J. 1990. Linkage map of Escherichia coli K-12, edition 8. Microbiol. Rev. 54:130-197. 2. Chang, A. C. Y., and S. N. Cohen. 1978. Construction and characterization of amplifiable multicopy DNA cloning vehicles derived from the P1SA cryptic miniplasmid. J. Bacteriol. 134: 1141-1156. 3. Chiang, R. C., R. Cavicchioli, and R. P. Gunsalus. 1992. Identification and characterization of narQ, a second nitrate sensor for nitrate-dependent gene regulation in Escherichia coli. Mol. Microbiol. 6:1913-1923. 4. Cotter, P. A., V. Chepuri, R B. Gennis, and R. P. Gunsalus. 1990. Cytochrome o (cyoABCDE) and d (cydAB) oxidase gene expression in Escherichia coli is regulated by oxygen, pH, and the fir gene product. J. Bacteriol. 172:6333-6338. 5. Cotter, P. A., S. Danie, and R P. Gunsalus. The effect of iron limitation on expression of the aerobic and anaerobic electron transport pathway genes in Escherichia coli. FEMS Microbiol. Lett., in press. 6. Cotter, P. A., and R P. Gunsalus. 1989. Oxygen, nitrate, and

molybdenum regulation of dmsABC gene expression in Escheichia coli. J. Bacteriol. 171:3817-3823.

7. Davis, R. W., D. Botstein, and J. R. Roth. 1980. A manual for

VOL. 174, 1992

genetic engineering. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 8. DeLorenzo, V., M. Herrero, F. Giovanni, and J. B. Neilands. 1988. Fur (ferric uptake regulation) protein and CAP (cataboliteactivator protein) modulate transcription of fur gene in Escherichia coli. J. Biochem. 173:537-546. 9. DeMoss, J. A., and P. Y. Hsu. 1991. NarK enhances nitrate uptake and nitrate excretion in Eschenichia coli. J. Bacteriol. 173:3303-3310. 10. Dong, X.-R., S. F. Lei, and J. A. DeMoss. 1992. Upstream sequence elements required for narL-mediated activation of transcription from the narGHJI promoter of Escherichia coli. J. Biol. Chem. 276:14122-14128. 11. Egan, S. M., and V. Stewart. 1990. Nitrate regulation of anaerobic respiratory gene expression in narX deletion mutants of Escherichia coli K-12. J. Bacteriol. 172:5020-5029. 12. Friedman, D. I. 1988. Integration host factor: a protein for all reasons. Cell 55:545-554. 13. Grove, C. L., and RI P. Gunsalus. 1987. Regulation of aroH operon of Escherichia coli by the Trp repressor. J. Bacteriol. 169:2158-2164. 14. Guest, J. R., and S. Sprio. 1990. FNR and its role in oxygenregulated gene expression in Escherichia coli. FEMS Microbiol. Rev. 75:399-428. 15. Gunsalus, R. P., L. V. Kalman, and R. R. Stewart. 1989. Nucleotide sequence of the narL gene that is involved in global regulation of nitrate controlled respiratory genes of Escherichia coli. Nucleic Acids Res. 17:1965-1975. 16. Hamilton, C. M., M. Aldea, B. K. Washburn, P. Babitzke, and S. R. Kushner. 1989. New method for generating deletions and gene replacements in Escherichia coli. J. Bacteriol. 171:46174622. 17. Hooper, D. A., F. M. Dias, K. S. Denis, R. M. Veitullo, and J. J. Rowe. 1990. Nitrate transport and nitrate reductase activity in nar mutants of Escherichia coli. Abstr. 90th Annu. Meet. Am. Soc. Microbiol. abstr. K80 p. 233. 18. Iuchi, S., and E. C. C. Un. 1987. A molybdenum effector of fumarate reductase expression and nitrate reductase induction in Escherichia coli. J. Bacteriol. 169:3720-3725. 19. Jones, H. M., and R. P. Gunsalus. 1987. Regulation of Escherichia coli fumarate reductase operon expression by respiratory electron acceptors and the fnr gene product. J. Bacteriol. 169:3340-3349. 20. Kalman, L. V., and R. P. Gunsalus. 1988. The frdR gene of Eschenichia coli globally regulates several operons involved in anaerobic growth in response to nitrate. J. Bacteriol. 170:623629. 21. Kalman, L. V., and R. P. Gunsalus. 1989. Identification of a second gene involved in global regulation of fumarate reductase and other nitrate-controlled genes for anaerobic respiration in Escherichia coli. J. Bacteriol. 171:3810-3816. 22. Kalman, L. V., and R. P. Gunsalus. 1990. Nitrate and molybdenum-independent signal transduction mutations in narX that alter regulation of anaerobic respiratory genes in Eschenichia coli. J. Bacteriol. 172:7049-7056. 23. Kostrews, D., J. Granzin, C. Koch, H. Choe, S. Raghunathan, W. Wolf, J. Labahn, R. Kahmann, and W. Saenger. 1991. Three dimensional structure of the E. coli DNA-binding protein Fis. Nature (London) 349:178-180. 24. Kunkel, T. A. 1985. Rapid and efficient site-specific mutagenesis without phenotypic selection. Proc. Natl. Acad. Sci. USA

narX, narL, narQ, fnr, AND him CONTROL OF narK

7111

82:488-492. 25. Lin, E. C. C., and D. R. Kuritzkes. 1987. Pathways for anaerobic electron transport, p. 207-209. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: cellular and molecular biology. American Society for Microbiology, Washington, D.C. 26. Messing, J., and J. VMeira. 1982. A new pair of M13 vectors for selecting either DNA strand of double digest restriction fragments. Gene 19:269-276. 27. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 28. Murakawa, G. J., C. Kwan, J. Yamashita, and D. P. Nierlich. 1991. Transcription and decay of the lac messenger: role of an intergenic terminator. J. Bacteriol. 173:28-36. 29. Niehaus, F., K. Hantke, and G. Unden. 1991. Iron content and FNR-dependent gene regulation in Escherichia coli. FEMS Microbiol. Lett. 84:319-324. 30. Nohno, T., S. Noji, S. Taniguchi, and T. Saito. 1989. The narX and narL genes encoding the nitrate-sensing regulators of Escherichia coli are homologous to a family of prokaryotic twocomponent regulatory genes. Nucleic Acids Res. 17:2947-2957. 31. Noji, S., T. Nohno, T. Saito, and S. Taniguchi. 1989. The narK gene product participates in nitrate transport induced in Escherichia coil nitrate-respiring cells. FEBS Lett. 252:139-143. 32. Robertson, C. A., and H. A. Nash. 1988. Bending of the bacteriophage lambda attachment site by Eschenichia coli integration host factor. J. Biol. Chem. 263:3554-3557. 33. Sanger, F., B. J. Nichien, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467. 34. Schroieder, I., S. Daire, and R. P. Gunsalus. Nitrate dependent activation of the nitrate reductase (narGHJI) operon of Escherichia coli requires Integration Host Factor protein. Submitted for publication. 35. Silhavy, T. J., M. L. Besman, and L. W. Enquist. 1984. Experiments with gene fusions, p. xi-xii. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 36. Simons, R. W., R. Houman, and N. Kleckner. 1987. Improved single and multicopy lac-based cloning vectors for protein and operon fusions. Gene 53:85-96. 37. Spiro, S., R. E. Roberts, and J. R. Guest. 1989. FNR-dependent repression of the ndh gene of Escherichia coli and metal ion requirement for FNR-regulated gene expression. Mol. Microbiol. 3:601-608. 38. Stewart, V. 1982. Requirement of Fnr and NarL functions for nitrate reductase expression in Escherichia coli K-12. J. Bacteriol. 151:1320-1325. 39. Stewart, V., and C. H. MacGregor. 1982. Nitrate reductase in Escherichia coli K-12, involvement of chlC, chlE, and chlG loci. J. Bacteriol. 151:788-799. 40. Stewart, V., and J. Parales. 1988. Identification and expression of genes narL and narX of the nar (nitrate reductase) locus in Escherichia coli K-12. J. Bacteriol. 170:1589-1597. 41. Stewart, V., J. Parnles, and S. M. Merkel. 1989. Structure of genes narL and narX of the nar (nitrate reductase) locus in Escherichia coli K-12. J. Bacteriol. 171:2229-2234. 42. Thompson, J. F., and A. Landy. 1988. Emperical estimation of protein induced DNA bending angles: applications to lambda site-specific recombination complexes. Nucleic Acids Res. 16: 9687-9705.