Vol. 173, No. 20

JOURNAL OF BACTERIOLOGY, OCt. 1991, p. 6626-6631

0021-9193/91/206626-06$02.00/0 Copyright X) 1991, American Society for Microbiology

NOTES

Klebsiella aerogenes Catabolite Gene Activator Protein and the Gene Encoding It (crp) ROBERT OSUNAt AND ROBERT A. BENDER* Department of Biology, The University of Michigan, Ann Arbor, Michigan 48109-1048 Received 25 February 1991/Accepted 4 August 1991

The catabolite gene activator protein from Klebsiella aerogenes (CAPK) and the corresponding protein from Escherichia coli (CAPE) were shown to be nearly identical. Both CAPK and CAPE activated transcription from the CAP-dependent promoters derived from E. coli and K. aerogenes. The crp gene from K. aerogenes (encoding CAP) is tightly linked to rpsL. The nucleotide sequence of cip predicts an amino acid sequence for CAPK that differs in only one position from that of CAPE.

The enteric bacterium Klebsiella aerogenes has several operons (including lac and hut) that are regulated by catabolite gene activator protein (CAP)-cyclic AMP (cAMP) (26). The regulation of one of the histidine utilization operons (hutUH) by CAP-cAMP has been a subject of study in this laboratory for some time (24, 25). In the absence of CAPcAMP, RNA polymerase binds preferentially to a leftward promoter of unknown function (Pc) rather than to hut Up, the promoter of the hutUH operon (23), and runoff transcription proceeds leftward rather than rightward into hutUH (24). The presence of CAP-cAMP causes the simultaneous repression of Pc and activation of hutUp (24, 25). A good match to the DNA consensus sequence for CAP binding sites was identified at positions -72 to -92 relative to the start of hutUp transcription (22) with its center of symmetry at position -81.5. This site resides rather far upstream from hutUp, so direct contact between CAP-cAMP and RNA polymerase bound at hutUp seems unlikely. However, this putative CAP-cAMP binding site overlaps the binding site of RNA polymerase at Pc. These observations suggested that the CAP-cAMP complex might act indirectly to cause the positive regulation of hutUp by repressing the Pc promoter directly (24). Implicit in this model is the untested assumption that CAP from K. aerogenes (CAPK) is capable of recognizing and binding to a DNA sequence similar to that recognized by CAP from Escherichia coli (CAPE). This implies considerable similarity between CAPE and CAPK and raises a broader question: how closely does CAPK resemble CAPE in its regulation of RNA polymerase activity? The observation that the K. aerogenes hut operons were regulated by catabolite repression when present in an E. coli host (17), though not as strongly as in K. aerogenes, suggested that CAPK and CAPE could be functionally interchanged in vivo, at'least to some extent. Therefore, we compared CAPK with CAPE. We show here that their deduced amino acid sequences are almost identical and that the proteins appear to be functionally equivalent. Isolation and characterization of crp anId cya mutations. K. aerogenes mutants with defects in cya and presumably crp

have been reported (26), but have since been lost. Three independent mutants presumed to lack either CAP (crp) or adenylate cyclase (cya) were isolated from K. aerogenes KC1297 by the fosfomycin resistance selection method of Alper and Ames (2). All three scored negative on MacConkey agar containing lactose, maltose, or sorbitol, and all three failed to grow with lactose or histidine as the sole carbon source. All three grew poorly on L broth unless it was supplemented with glucose. One of these mutants, strain KC1628, could grow with histidine or lactose as the sole carbon source when cAMP was present, but the other two were not affected by the presence of cAMP. Thus, we assumed that KC1628 (cya4011) was defective in cya and that the other two strains might be defective in crp. The cya4011 mutation was about 50% linked to ilvA and about 5% linked to metB by P1-mediated generalized transduction. The genes that lie between rbs and glnA in E. coli are translocated in K. aerogenes to a site between argH and rpoB. Thus, the positions of the E. coli cya (rbs-ilvA-cya-glnA-metB) and the K. aerogenes cya (rbs-glnA-metB-ilvA-cya) genes are truly homologous. The two cAMP-nonresponsive mutants, KC1669 (crp4021) and KC1670 (crp4031), had mutations tightly linked by transduction to rpsL and grew poorly on L broth and not at all on minimal medium with histidine as the sole carbon source. Thus, we concluded that these strains were crp mutants and not pts mutants, the other class of cAMPnonresponsive, pleiotropisally sugar-negative mutants (4). K. aerogenes pts mutants have mutations linked to cysK and grow well on L broth and on minimal medium with histidine as the sole carbon source (4). To confirm that the crp strains lacked CAP activity, cell extracts from the mutants were assayed for cAMP binding protein as described previously (25). Crude extracts from strains KC1669 (crp4021) and KC1670 (crp4031) had less than 10% as much cAMP binding protein as did the wild-type strain KC1297; strain KC1628 (cya4011) had levels of cAMP binding protein at least as high as those of the wild type (data not shown). Cloning and sequencing crp from K. aerogenes. A DNA fragment able to complement the crp mutation of strain KC1669 was cloned from strain KC1043 by using the in vivo cloning procedure of Groisman et al. (18). A 3.2-kb BamHI

* Corresponding author. t Present address: Department of Biological Chemistry, UCLA School of Medicine, Los Angeles, CA 90024. 6626

VOL. 173, 1991

NOTES

TAACTCCAGTATCCCGAATA¶rCCTCAGATTACGCGCACATAACAATTCTCGCAACGA

K.a. (-70) E.c -* ----- C-G---- -T------ A----G ----- T--TA-C ---A--G --------S. t -------------------- T-C ---- AAG ------T-

CAP SITE 2

-

35

-10

K. a. GACAr (-10) rAG E .c G ---CAAAS.t ----------G----------------------------- CA ------------------ -- -- - --

-- -- -- -- -- -- -- -- -- -- -- -- - --

-- -- -- -- -- --

+1 CAP SITE 1 A A K. a. A&CAGTCTGGATGCTACAGTAAT C E .c -- -- - -- A-- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- - -- A- - -- CG - -- -- C -- -- -- -- -- -- -- ------ -CA-- ----S.t -- --- -A-AA--- -- -- -- -- -- -- -- -- --C-

TACGGGCGAAAAATCATCAGCCAGCTTCCCAGGTATAGGGGAACAATACGATTGCA - -C-TGCAGTAC-GTT-G- ------------ GC

K. a. E.c.

-C

-

S.t.

---A ----

-

-

*--GC --- TTTCG ---

(+51)

(111)

ACAATC-CG-* ---- CC ----------- GC --- *---GT --- TTT-----

F(S.t.) S.D. K. a. GTAACCGCGC (171) E.c. AT- -A- -GA-A-C-GC- -T - - -T-T* -----A ------------------------ T -- - --- -- -- -- -- -- -- -- -- -- --S .t . AC - -- - -- GA-A---- CC -T . AC-

TGCCCGGACGCCGTGTAGTGTCGGCGCCCGGAGATAGCTTATAACA (E.C.)

T T (S.t.)

10

MetValLeuGlyLyslProGlnThrAspProThrLeuGluTrpPheLeuSerHisCysHis ATCGTGCTGAAACCGCAAACCA CG- -TGACCATACT CTCATTGCCAC --C----------------------

K. a. E.c .------------------------S.t------.---------

-

20 30 I leHi sLysTyrProSerLysSerThrLeuIleHi sGlnGlyGluLysAlaGluThrLeu

ATTCATAAGTACCCATCAAAGAGCACGCTGATCCACCAGG AAAGCAGAAACGCTG -C----------T- -T----------------- G---------

K.a. E.c .-

(291)

---------------

S.t. -

- -- -- -- --

-----

-G-- -- -- -- -- --

--

---

40 K. a.

(231)

G--T --------------

-T--- -- -- -- -- -- -- -- -- -- -- -----

50

leValLysGlySerVaLAUaValLeuI TACTACATCGTTAAAGGCTCGTC ,A leLysAspGluGluGlyLysGluMet GGTTG GAAGAAGAGATG (351) TyrTyrI

E.c. - -------------------T----- A--G--G ---- A--C ---- G-------- A---

S..t

--------------------------A--G--G ---- A----------- G----- A---

K. a.

60 70 StuI I leLeuSerTyrLeuAsnGlnGlyAspPhe IleGlyGluLesuGlyLeuPheGluGluGly

E.c.

S..t

ATCCTCTCCTACCTCAACCAGGGCGAT TATCGTATTCGG GAATTAAGAGGGT ---------- T- -G- -T ---- T---- T- -T- -C- -C-G -----------------C ----T- -T--T--G- -T ---- T-T-----T------ C-G ------------- A--C

(411)

-

90

80

GlnGluArgSerAlaTrpvalArgAlaLysThrAlaCysGluValAlaGluI

leSerTyr K.a. (471) -----A-------- A-------- T----------- C----------- T----- T--G--C E.c. CAGGAGCGTAGCGCCTGGGTACGGGCGAAAACCGCATGTGAAGTGGCCGAAATCTCTTAT

S..t

----A- -C-------------- T----------------G- -C--T- -- T- -C- -C

K. a. E.c.

LysLysPheArgGlnLeuI1eGlnValAsnLProAspIleLeuMetArgLeuSezgjjln AAGAAATTCCGTCAGCTGATCCAGGTGAACCCGGACATGCTGATGCGTCTCTCTrCGCAA -G-A- -G -A T- -C-

100

Fl-

110

(531)

---AT- - - -T----- A-------------------- T-G- S.t. --A ---- T--C--AT-A ------- C-------- T----------- C-------- C--G

120

K.a.a

E.c. S.t.

PstI

130

MetAlaArgArgLeuGlnValThrSerGluLysValGlyAsnLeuAlaPheLeuAspVal C

GCC CA

AGTGGGCAACCTC CGTG (591) ---- G--T ------- A----- T--A ---------------- G--G ----------A-T --------------T-C---C -------T- -CT-A- -A ---- C----- A----140

6627

fragment from one such clone was subcloned into vector KSM13+ and shown to complement crp4021. This plasmid, pJP13, also complemented a well-characterized E. coli deletion of crp (27) for lactose fermentation, confirming the identity of the cloned gene as crp. A large open reading frame encoding 209 amino acids was identified in pJP13. The deduced amino acid sequence from this open reading frame (Fig. 1) was almost identical to that of CAPE (1, 10). The only difference was a serine at position 118 of CAPK rather than the alanine present in the CAPE. The Salmonella typhimurium CAP (CAPS) also has serine at position 118 as its only difference from CAPE (11) and thus was identical to CAPK. At the nucleotide sequence level, the crp genes from K. aerogenes, S. typhimurium, and E. coli differed from each other by about 12% in each pairwise combination. The variations in nucleotide sequence among the three genes were mostly confined to the third positions of codon triplets. With three nucleotide sequences, it was possible to propose a consensus or ancestral sequence for positions at which two or three of the sequences have the same nucleotide. Although the three crp sequences differ from this ancestral sequence at similar frequencies, the majority of the changes that were unique to the K. aerogenes crp sequence were clustered within the first half of the gene, whereas the unique changes found within the E. coli and S. typhimurium crp sequences were noticeably more abundant within the second half of the gene. The sequence conservation between the crp genes from K. aerogenes and E. coli extends beyond the coding region and includes regions upstream and downstream of the gene, regions known to be important for regulating expression of the E. coli crp gene. The promoter region of E. coli crp, including the RNA polymerase binding site, the start of transcription (+1), and two CAP binding sites (+30 to +53 and -71 to -50), was very similar to the same region of K. aerogenes crp. In contrast, the region between the promoter and the start of the coding region (i.e., from +54 to +140 with the coding region starting at + 172) was quite dissimilar in the two organisms, suggesting a lack of selective pressure on this region (Fig. 1). The presumed signal for termination of transcription lies just beyond the end of the E. coli crp coding sequence.

150

ThrGlyArgIleAlaGlnThrLeuLeuAsnLeuAlaLysGlnProAspAlaMetThrHis

K. a. ACGGGCCGTATCGCCCAGACGCTGCTGAACCTGGCGAAGCAACCGGATGCCATGACCCAC (651) -E.c .-------- C--T--A ---- T-------- T----- A--A ---- A--C- -TS.t. --C--G ------- T----------------------- A--G--C ---------- G---

160

170

ProAspGlyMetGlnIleLysIleThrArgGlnGluIleGlyGlnIleValGlyCysSer

A T C AATTACACGCCA K.a. CCGGACGrATGCAAATrAA GCTGCs (711) E.c .----------------- C-------- C--T -T-------------T- -S.t.----- T--G ---- G--C ---- C--T--T -----------C-C--------------- C

180 K.a.

190

ArgGluThrValGlyArgIleLeuLysMetLeuGluAspGlnAsnLeuIleSerAlaHis C (771) CGAAACTCTGGAAGATCAGAACCTGAIC

E.C. - ----------G--A--C ---C----G ---------------------- C----- A--S.t. - -C-------------------------------------- A-------- C-T------T 210

200

GlyLysThrIleValValTyrGlyThrArgEND

K.a. GGTAAAACCATCGTCGTCTACGGCACCCGrTAAGTTGCTAAG E.c.----------------- T--------------- TCCCGTCGG S.t. --C--G ----------------- T --------- TTCCGTCAG

-srGi'TTTr E. c .AGTGGCGCGTTACCTGGrAGCGCGCCATTr -> >

K. a. CGGCGTATrGCTCCCGCAGTACGCC -

)0-

O

S. t .AATGGCGCGTATCATGCGCCATGTT

FIG. 1. Nucleotide sequence (5, 31) of the K. aerogenes (K.a.) crp gene. The nucleotide sequences of the crp genes from E. coli (E.c.) (1, 10) and S. typhimurium (S.t.) (11) are shown for comparison. A dash indicates a nucleotide identical to that in K. aerogenes, a letter indicates a nucleotide different from K. aerogenes, an * indicates a gap relative to K. aerogenes, and a bracket indicates an insertion relative to K. aerogenes. Nucleotides are numbered (in parentheses) relative to the start of E. coli crp transcription, 172 bp upstream of the start of translation. Amino acids deduced from the K. aerogenes crp DNA sequence are numbered relative to the N-terminal valine found in the mature E. coli protein. Other sites indicated by underlining are as follows: CAP sites 1 and 2, sites known to bind CAP-cAMP in E. coli; -35 and -10, the promoter of the E. coli crp gene; +1, the start of crp transcription in E. coli; S.D., a match to the Shine-Dalgarno consensus near the start of the open reading frame; Stul and PstI, two restriction sites known to lie within K. aerogenes crp; Ala (boxed), the site where E. coli crp encodes alanine in contrast to the serine encoded by K. aerogenes and S. typhimurium. Potential stem-loop structures (indicated by opposed arrows) which are followed by a T-rich run and thought to terminate transcription of E. coli crp are shown at the end of the sequence.

6628

J. BACTERIOL.

NOTES

I

Ii

FIG. 2. Activation of the E. coli lacZ promoter by CAP. (A) Transcripts produced in the presence of CAPK and CAPE. The 25-,u transcription reactions (25) contained [a-32P]UTP as label, 0.2 pmol of a 203-bp DNA fragment containing lacZp, 1 pmol of E. coli RNA polymerase holoenzyme, and 1.2 pmol of either CAPK or CAPE as indicated. Plus and minus indicate the presence and absence of 1 mM cAMP in the reaction mixture. The resulting transcripts were separated by electrophoresis and visualized by autoradiography. (B) Titration of lacZp with CAPK. Transcription reactions labeled as described above contained (in 25 jjl) 0.1 pmol of lacZp DNA (as a 203-bp fragment), 1.5 pmol of E. coli RNA polymerase holoenzyme, and 1 mM cAMP. CAPK was added to the reactions to give the indicated molar ratios of CAPK/promoter fragment. Transcripts were detected as described above and quantified by densitometric scanning of the autoradiogram. One hundred percent lacZp activity is the value assigned to the maximum transcription signal detected. The different symbols represent experiments done on different days. (The concentration of CAPK at the ratio of 4:1 was about 32 nM.)

Although the K. aerogenes crp sequence showed little sequence identity in the region of the stem-loop, the position of the stem-loop in K. aerogenes was virtually identical to that in E. coli. The crp terminators from K. aerogenes, E. coli, and S. typhimurium each ended with a stretch of 7 to 10 T residues split by a single G in the middle of the run of T residues. Thus, although we have no information on the regulation of crp expression in K. aerogenes, the conservation of sites known to regulate E. coli crp expression leads us to conclude that the regulation of crp expression in K. aerogenes is probably similar to that in E. coli. Functional similarity of CAPK and CAPE. Although the DNA sequence predicted that CAPK would be virtually identical to the well-characterized CAPE, it was still important to demonstrate that there were no significant differences between CAPK and CAPE caused by differences in folding, modification, or other cytoplasmic interactions. Therefore, we compared the ability of CAPK (purified as described previously [25]) and CAPE (generously provided by Stephanie H. Shanblatt and Arnold Revzin of Michigan State University) to activate transcription by E. coli RNA polymerase holoenzyme (purchased from New England BioLabs) at a CAPE site (a 203-bp fragment from plasmid pRZ4004 [kindly provided by William Reznikoff of the University of Wisconsin] containing the lacZ promoter) and at a CAPK site [a 255-bp fragment of plasmid pOS2 (isolated in this laboratory) containing the hut(P) region of the K. aerogenes hut operons]. Preliminary experiments showed that CAPK could acti-

vate transcription from lacZp by RNA polymerase from E. coli (RNAPE) (25). The data in Fig. 2 show that this activation by CAPK resembles that by CAPE. In the presence of cAMP, but not in its absence, both CAPK and CAPE promoted the synthesis of runoff transcripts of similar size and of comparable intensities. The size of the transcripts (approximately 68 bases based on DNA standards) was in agreement with the expected size of a transcript initiating at lacZp (63 bases). The amount of CAP added in each case was approximately the same, and the reaction conditions were identical. Thus, when complexed with cAMP, both CAPK and CAPE directed RNAPE to initiate transcription at

lacZp. The results of a CAPK titration with the lacZp template (Fig. 2B) showed a dose response similar to that reported by Gamer and Revzin (16) with CAPE. At low concentrations of CAPK, runoff transcription increased with increasing CAPK. At high concentrations of CAPK, runoff transcription decreased with increasing CAPK. Since maximum transcription was detected at a 4:1 molar ratio, and since only one active CAP dimer is required at lacZp to activate transcription (14, 16), our CAPK preparation was estimated to be 25% active. To determine whether CAPE could recognize a CAPK site, we examined the ability of CAPK and CAPE to regulate K. aerogenes hutUp. The hutUp activation process also involves the repression of an overlapping divergent promoter (Pc) of unknown function (24). Both CAPK-cAMP and CAPE-cAMP gave the same set of runoff transcripts from

VOL. 173, 1991

NOTES

CAP cAMP

K K + 1 2

E E

CAPK

3 4

cAMP

t,~4,*

-

5nM lOnM 15nM 2OnM 3OnM 4OnM

11

1

*

lacZp

Fr

+ - + 2 3 4 5

-

1419

.:

6629

*

+ - + - + - + 6 7 8 9 10 11 12

v.

L

-,.

l.

_b

m

o

Im -

33 1

'A

i,

I PC -*.

9

*#

214

,a

hutU:

-65

FIG. 3. Regulation of the K. aerogenes hutU and Pc promoters by CAP. Transcription conditions were as in the legend to Fig. 2, and reactions contained 0.3 pmol of a 255-bp DNA fragment carrying the hutUp region (and Pc), 1 pmol of E. coli RNA polymerase holoenzyme, and 3 pmol of either CAPK or CAPE. The runoff transcripts from hutUp and Pc (the oppositely oriented, overlapping promoter) are indicated.

hutUp with comparable sizes and intensities (Fig. 3, lanes 1 and 3). At the same time, both CAPK-cAMP and CAPEcAMP repressed transcription from Pc to about the same extent. When cAMP was omitted from the reaction mixture, transcription proceeded primarily from Pc, and hutUp was not activated (Fig. 3, lanes 2 and 4). A CAP consensus sequence that strongly resembles that from E. coli lacZp has been identified near hutUp (12, 22), and thus we tested whether CAPK would bind to lacZp and hutUp similarly. When CAPK was mixed with an equimolar mixture of three DNA fragments, one carrying lacZp, one carrying hutUp, and one carrying no known CAP binding sites, both the lacZp- and the hutUp-containing DNA fragments showed binding at CAPK concentrations as low as 15 to 20 nM, and complete binding was observed at 30 to 40 nM (Fig. 4). The mobility of the smaller (control) DNA fragment remained unaffected. The previously identified CAP consensus sequence lies centered at position -81.5 relative to the start of transcription of hutUp, or bp 110.5 on the 255-bp SalI fragment of pOS2 (Fig. 5). To demonstrate that CAPK and CAPE were actually binding in this region, we further digested this 255-bp Sall fragment with Sau3A, SphI, and HgaI and tested the resulting fragments for CAP binding by using a gel mobility shift assay (Fig. 5A). For simplicity, a diagrammatic representation of these results indicating the relevant restriction sites is also provided (Fig. SB). When CAPK was incubated (in the presence of cAMP) with the products of the Sau3A, SphI, or HgaI DNA digests of the Sall fragment, the mobility of the band containing the region around bp 110.5 was retarded in each case (Fig. SA, lanes 3, 5, and 7). When cAMP was omitted, no such retardation was observed (lanes 4, 6, and 8). The Sau3A and SphI fragments which did not contain the region around bp 110.5 were not retarded even in the

FIG. 4. Comparison of the CAP binding sites at hutUp and lacZp. Plasmid pOS2 was digested with restriction enzymes into three fragments: one carried hutUp from K. aerogenes, one carried lacZp from E. coli (indicated at the left), and one carried no known CAP binding site. The fragments were end labelled and mixed with the indicated concentrations of CAPK in the presence (+) or absence (-) of 1 mM cAMP. The fastest-migrating fragment lacks a CAP binding site.

of cAMP. However, two of the HgaI fragments showed evidence of cAMP-dependent CAPK binding: the smallest fragment (F-3), which contains the 44-bp region around bp 110.5 and little else; and the largest fragment (F-1), which contains hutUp (Fig. 5A, lanes 7 and 8). Since the DNA to the right of the SphI site was not retarded by CAPK (lanes 5 and 6), we inferred that the CAPK binding to HgaI fragment F-1 involved the region between the HgaI and SphI sites. Like CAPK, CAPE also showed cAMPdependent binding to both the F-3 and F-1 fragments, but not the F-2 fragment (Fig. 5A, lanes 9 and 10). Thus, both CAPK and CAPE recognized the same two DNA sites within the hutUp region. Given the general similarity within the enteric group of bacteria, it might be argued that the similarity between CAPK and CAPE (and that between RNAPK and RNAPE) was to be expected. However, there are numerous examples of major differences between E. coli and K. aerogenes proteins. For example, glutamine synthetase from K. aerogenes differs significantly in both molecular weight (7) and enzymatic properties (6) from that of E. coli. Similarly, P-galactosidase from K. aerogenes differs greatly in primary sequence from that of E. coli (9), and the K. aerogenes

presence

is far more thermolabile under standard conditions than the E. coli enzyme (4). Even the sigma subunits of RNA polymerase from E. coli K-12, E. coli C, and S. typhimurium migrate differently on sodium dodecyl sulfate-polyacrylamide gels (19, 21). We had initially proposed a double-negative model for activation of hutUp expression by CAP-cAMP (24). In this model, CAP-cAMP would directly repress binding of RNA polymerase to a oppositely directed promoter, Pc, thus allowing transcription from hutUp. This model allowed us to explain how a CAP binding site centered at position -81.5 (identified by inspection of the DNA sequence) could activate transcription at hutUp from such a distance. However, even though hutUp and Pc clearly are mutually competitive, recent work suggested that CAP-cAMP does not directly

P-galactosidase

assay

6630

J. BACTERIOL.

NOTES ..o 1.

.

S ~-wa

4

8XU

FIG. 5. CAP binding sites in the hutUp region. (A) Gel retardation of hutUp fragments by CAP. The test fragment was a 255-bp SalI-SalI fragment carrying hutUp (lanes 1 and 2) or the same fragment further cleaved with Sau3A (lanes 3 and 4), SphI (lanes 5 and 6), or HgaI (lanes 7 to 10). (The slowest-migrating band in lane 10 is probably a partial digestion product containing fragments F-1 and F-3.) About 100 ng of CAPK (lanes 1 to 8) or CAPE (lanes 9 and 10) was added to the end-labelled fragments in the presence (+) or absence (-) of 1 mM cAMP, as indicated. The DNA fragments were separated by electrophoresis and detected by autoradiography, using Hinfl fragments of pUC19 as size standards (lane S). The three fragments resulting from HgaI digestion are labelled F-1, F-2, and F-3. (B) Schematic representation of CAP binding data. The upper drawing shows the hutUp fragment. The RNA polymerase binding sites at hutUp and Pc (24) are shown as boxes, with their direction of transcription indicated by arrows. The lower lines illustrate fragments which showed a CAP-cAMP-induced mobility shift (solid lines) and those that did not (dashed lines).

repress Pc since Pc can fill with RNA polymerase even in the presence of bound CAP-cAMP (25). Moreover, the formation of open complexes at Pc is so slow that it may not contribute significantly to a repression of hutUp (25). The finding of a second CAP-cAMP binding site at hutUp (Fig. 5) may help explain the activation of hutUp by CAPcAMP. This second site, located somewhere between -20

and -70 relative to the start of transcription, lies close to hutUp. In fact, a weak match to the CAP consensus can be identified centered at position -42.5 and thus overlapping hutUp. This arrangement calls to mind a similarly positioned CAP site required for CAP-cAMP activation at gal (8, 28, 30) and may provide a means for direct activation of hutUp by CAP-cAMP.

VOL. 173, 1991

Nucleotide sequence accession number. The nucleotide sequence reported in Fig. 1 has been submitted to GenBank (accession number M68973). We are grateful for the technical assistance of J. Perry and K. Whitely in the isolation of the K. aerogenes crp mutants and clones. This work was supported by Public Health Service grant A115822 from the National Institutes of Health to R.A.B. and by a Michigan Minority Merit Fellowship Award to R.O. 1. 2.

3. 4.

5. 6.

7.

8.

9. 10. 11.

12.

13. 14. 15.

REFERENCES Aiba, H., S. Fujimoto, and N. Ozaki. 1982. Molecular cloning and nucleotide sequencing of the gene for E. coli cAMP receptor protein. Nucleic Acids Res. 10:1345-1361. Alper, M. D., and B. N. Ames. 1978. Transport of antibiotics and metabolite analogs by systems under cAMP control: positive selection of Salmonella typhimurium cya and crp mutants. J. Bacteriol. 133:149-157. Bachmann, B. J. 1983. Linkage map of Escherichia coli K-12, edition 7. Microbiol. Rev. 47:180-230. Baldauf, S. L., M. A. Cardani, and R. A. Bender. 1988. Regulation of the galactose-inducible lac operon and the histidine utilization operons in pts mutants of Klebsiella aerogenes. J. Bacteriol. 170:5588-5593. Bartlett, J. A., R. K. Gaillard, and W. K. Joklik. 1986. Sequencing of supercoiled plasmid DNA. BioTechniques 4:208-209. Bender, R. A., K. A. Janssen, A. D. Resnick, M. Blumenberg, F. Foor, and B. Magasanik. 1977. Biochemical parameters of glutamine synthetase from Klebsiella aerogenes. J. Bacteriol. 129:1001-1009. Bender, R. A., and S. L. Streicher. 1979. Glutamine synthetase regulation, adenylylation state, and strain specificity analyzed by polyacrylamide gel electrophoresis. J. Bacteriol. 137:10001007. Busby, S. J. W., H. Aiba, and B. de Combrugghe. 1982. Mutations in the Escherichia coli operon that define two promoters and the binding site for the cyclic AMP receptor protein. J. Mol. Biol. 154:211-227. Buvinger, W. E., and M. Riley. 1985. Nucleotide sequence of Klebsiella pneumoniae lac genes. J. Bacteriol. 163:850-857. Cossart, P., and B. Gicquel-Sanzey. 1982. Cloning and sequence of the crp gene of E. coli K-12. Nucleic Acids Res. 10:13631378. Cossart, P., E. A. Groisman, M. C. Serre, M. J. Casadaban, and B. Gicquel-Sanzey. 1986. crp genes of Shigella flexneri, Salmonella typhymurium, and Escherichia coli. J. Bacteriol. 167:639646. deCrombrugghe, B., S. Busby, and H. Buc. 1984. Activation of transcription by the cyclic AMP receptor protein, p. 129-167. In R. F. Goldberg and K. R. Yamamoto (ed.), Biological regulation and development. Plenum Press, New York. Fried, M., and D. M. Crothers. 1981. Equilibrium and kinetics of lac repressor-operator interactions by polyacrylamide gel electrophoresis. Nucleic Acids Res. 9:6505-6525. Fried, M. G., and D. M. Crothers. 1983. CAP and RNA polymerase interactions with the lac promoter: binding stoichiometry and long range effects. Nucleic Acids Res. 13:141-158. Garner, M. M., and A. Revzin. 1981. A gel electrophoresis

NOTES

16.

17. 18. 19. 20. 21. 22.

23.

24.

25.

26.

27.

28.

29. 30.

31.

6631

method for quantifying the binding of proteins to specific DNA regions: application to components of the E. coli lactose operon regulatory system. Nucleic Acids Res. 9:3047-3060. Garner, M. M., and A. Revzin. 1982. Stoichiometry of catabolite activator protein/adenosine cyclic 3',5'-monophosphate interactions at the lac promoter of Escherichia coli. Biochemistry 21:6032-6036. Goldberg, R. B., F. R. Bloom, and B. Magasanik. 1976. Regulation of histidase synthesis in intergeneric hybrids of enteric bacteria. J. Bacteriol. 127:114-119. Groisman, E. A., and M. J. Casadaban. 1986. Mini-Mu bacteriophage with plasmid replicons for in vivo cloning and lac gene fusing. J. Bacteriol. 168:357-364. Harris, J. D., I. I. Martinez, and R. Calendar. 1977. A gene from Escherichia coli affecting the sigma subunit of RNA polymerase. Proc. Natl. Acad. Sci. USA 74:1836-1840. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. Nakamura, Y., T. Osawa, and T. Yura. 1977. Chromosomal location of a structural gene for the RNA polymerase a factor in Escherichia coli. Proc. Natl. Acad. Sci. USA 74:1831-1835. Nieuwkoop, A. J., S. A. Baldauf, M. E. S. Hudspeth, and R. A. Bender. 1988. Bidirectional promoter in the hut(P) region of the histidine utilization (hut) operons from Klebsiella aerogenes. J. Bacteriol. 170:2240-2246. Nieuwkoop, A. J., and R. A. Bender. 1988. RNA polymerase as a repressor of transcription in the hut(P) region of mutant Klebsiella aerogenes histidine utilization operons. J. Bacteriol. 170:4986-4990. Nieuwkoop, A. J., S. A. Boylan, and R. A. Bender. 1984. Regulation of hutUH operon expression by the catabolite activator protein-cyclic AMP complex in Klebsiella aerogenes. J. Bacteriol. 159:934-939. Osuna, R., S. A. Boylan, and R. A. Bender. 1991. In vitro transcription of the histidine utilization (hutUH) operon from Klebsiella aerogenes. J. Bacteriol. 173:116-123. Prival, M. M., and B. Magasanik. 1971. Resistance to catabolite repression of histidase and proline oxidase during nitrogenlimited growth of Klebsiella aerogenes. J. Biol. Chem. 246: 6288-6296. Sabourin, D., and J. Beckwith. 1975. Deletion of the Escherichia coli crp gene. J. Bacteriol. 122:338-340. Shanblatt, S. H., and A. Revzin. 1983. Two catabolite activator protein molecules bind to the galactose promoter region of Escherichia coli in the presence of RNA polymerase. Proc. Natl. Acad. Sci. USA 80:1594-1598. Streicher, S. L., R. A. Bender, and B. Magasanik. 1975. Genetic control of glutamine synthetase in Klebsiella aerogenes. J. Bacteriol. 121:320-331. Taniguchi, T., M. O'Neill, and B. de Combrugghe. 1979. Interaction site of Escherichia coli cyclic AMP receptor protein on DNA of galactose operon promoters. Proc. Natl. Acad. Sci. USA 76:5090-5094. Zagursky, R. K., K. Baumeister, N. Lomax, and M. L. Berman. 1985. Rapid and easy sequencing of large linear double-stranded DNA and supercoiled plasmid DNA. Gene Anal. Technol. 2:89-94.