Vol. 173, No. 16

JOURNAL OF BACTERIOLOGY, Aug. 1991, p. 4914-4921

0021-9193/91/164914-08$02.00/0 Copyright © 1991, American Society for Microbiology

Cloning and DNA Sequence of amiC, a New Gene Regulating Expression of the Pseudomonas aeruginosa Aliphatic Amidase, and Purification of the amiC Product STUART WILSON AND ROBERT DREW*

Department of Biochemistry, University College London, Gower Street, London WCIE 6BT, United Kingdom Received 28 May 1991/Accepted 10 June 1991

Using in vitro-constructed deletions and subcloned DNA fragments, we have identified a new gene, amiC, which regulates expression of the inducible Pseudomonas aeruginosa aliphatic amidase activity. The DNA sequence of the gene has been determined, and an open reading frame encoding a polypeptide of 385 amino acids (molecular mass, 42,834 Da) has been identified. A search of sequence libraries has failed to find homologies with other published sequences. The amiC translation termination codon (A)TGA overlaps the initiation codon for the downstream amiR transcription antitermination factor gene, implying that the amiCR operon is coordinately regulated. Disruption of the amiC open reading frame by insertion and deletion leads to constitutive amidase synthesis, suggesting that AmiC is a negative regulator. This is confirmed by the finding that a broad-host-range expression vector carrying amiC (pSW41) represses amidase expression in a series of previously characterized P. aeruginosa amidase-constitutive mutants. The AmiC polypeptide has been purified from PAC452(pSW41), and N-terminal amino acid sequencing has confirmed the gene identification.

Pseudomonas aeruginosa PAC1 is able to grow on shortchain aliphatic amides by virtue of a chromosomally located amidase (EC 3.5.1.4) (8). Amidase activity is inducible and shows distinct inducer and substrate specificities (23). The P. aeruginosa amidase system has been used extensively by Patricia Clarke and coworkers to study experimental enzyme evolution, and this work has been recently reviewed (11). Initial cotransduction analysis showed that the amidase structural gene (amiE) and regulator gene (amiR) were closely linked (9), and investigations of the regulation of amidase expression have shown that at least two independent control systems operate. Genetic and biochemical studies have shown positive control of amidase expression by the product of amiR (16, 19), and genetic and physiological studies have shown the system to be subject to catabolite repression by succinate (39). To investigate the molecular mechanisms involved in amidase regulation, the genes from the magnoconstitutive mutant PAC433 (40) were cloned in bacteriophage lambda by selection for amidase activity in Escherichia coli (17). A 5.3-kb DNA fragment carrying the amidase genes was subcloned into pBR322, yielding plasmid pJB950, and the location of amiE was determined (12). The amiE gene was found to lie very close to one end of the DNA fragment, and subsequently both the DNA sequence and amino acid sequence of amidase were determined (1, 7). Codon usage of the amiE gene reflected the high G+C ratio of the organism and the fact that the gene can be highly expressed. The DNA sequence upstream of amiE has been determined for PAC433 and the wild-type strain PACL. The two sequences differ at only one position, in the ribosome binding site. amiE is expressed from an E. coli-like promoter 150 bp upstream of the start codon, and the positive control exerted by AmiR under inducing conditions is mediated by a transcription antitermination mechanism whereby transcription from the *

promoter reads through a rho-independent terminator and into the amiE gene (18). Studies using in vitro constructed deletions and subcloned DNA fragments in E. coli and P. aeruginosa have located the amiR gene in a 1.0-kb DNA region some 2 kb downstream of amiE and shown it to be transcribed in the same direction (13). These investigations additionally found preliminary evidence that the amiR gene product is a 23-kDa polypeptide by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and showed that the promoter for amiR lies in an undefined region some distance upstream of the gene itself. The DNA sequence of the PAC433 amiR gene has been determined, and the amiR open reading frame has been identified (28). Codon usage for amiR is very similar to that for amiE; however, on the basis of data base searches, the DNA sequence and amino acid sequence of the transcription antitermination factor appear to be unique. The early genetic studies of the amidase system used Ami- mutations to define the amiE gene and constitutive mutations to define amiR, and subsequent investigations, including the use of an amiR(Ts) mutant, showed that amiR functioned as a positive controlling element. We have now isolated the wild-type amidase genes and constructed a series of new plasmids to investigate amiR expression. These studies show that amiR expression or activity is regulated by a previously uncharacterized negative regulator gene, amiC, and that mutations in this gene lead to constitutive amidase synthesis. MATERIALS AND METHODS Bacterial strains and plasmids. The E. coli and P. aeruginosa strains and parental and newly constructed plasmids used in this study are listed in Table 1. Media. E. coli strains were grown at 37°C in Lennox broth, and P. aeruginosa strains were grown at 37°C in Oxoid no. 2 broth. Antibiotics were used for E. coli and P. aeruginosa as described previously (13). Growth of E. coli W3110 for

Corresponding author. 4914

amiC REGULATION OF P. AERUGINOSA AMIDASE EXPRESSION

VOL. 173, 1991

4915

TABLE 1. Bacterial strains and plasmids used in this study Strain or plasmid

Strains E. coli W3110 JA221 JM101 HB101 P. aeruginosa PAC1 PAC101 PAC111 PAC433 PAC452

Plasmids pRK2013 pBR322 derived pJB950 pDC5 pAS20 pAS21

pSWl pSW2 pSW3 pSW4 pSW5 pSW36 pSW37

Description or relevant features

Source or reference

Prototroph hsdR recA trp leu thi A(lac-proAB) F' [traD proAB+ lacIq lacZ AM15] hsdSB recA pro rpsL20

44 10 30 35

amiE+ amiR+, inducible Ami+ prototroph amiE+ amiRI, constitutive Ami+ amiE+ amiRII, constitutive Ami+ amiEJ14,120 amiRI, constitutive Ami+ CRPW ami-161, Ami deletion

23 9 9 40 M. Day

Kmr ColEl mob tra+ (RK2)

20

Apr amiE+ Apr amiE+ Apr amiE+ Apr amiE+ Apr amiE+ Apr amiE+ Apr amiE+ Apr amiE+ Apr amiE+ Apr amiE+ Apr amiE+

6 12 13 This This This This This This This This This

amiR+ 5.3-kb HindIII-SalI Ami+ fragment from PAC433 deletion derivative of pJB950 amiR+ 5.3-kb HindIII-SalI Ami+ fragment from PAC1 amiR+ amiR+ from pJB950 amiR+ from pJB950 amiR+ from pAS20 amiR+ from pAS20 amiR+ from pJB950 amiR+ from pAS20 amiR+ from pAS20

work work work work work work work work work

pKT231 derived pSW35 pSWlOl

Smr amiR+ 1.5-kb XhoI amiR+ fragment from pAS20 Smr amiE+ amiR+ 5.3-kb HindIII-SaIl fragment from pAS20

2 This work This work

pMMB66HE derived pSW40 pSW41

Apr amiR+ 1.5-kb XhoI amiR+ fragment from pAS20 Apr amiC+ 1.4-kb KpnI-PvuII amiC+ fragment from pAS20

21 This work This work

a CRF, resistance to catabolite repression.

preparation of phage vector L47 (27) and selection for X-ami recombinant phages on nitrogen-free glucose-acetamide minimal medium were as described previously (17). For amidase assays on cells grown under inducing, noninducing, and repressing conditions, the media and conditions were as described previously (16). DNA manipulations. Preparation of phage L47 DNA and P. aeruginosa chromosomal DNA was as described previously (17). Plasmid DNA was isolated from E. coli and P. aeruginosa by the method of Birnboim and Doly (5) and, if necessary, was purified by ethidium bromide-cesium chloride centrifugation (3). Ligated DNA samples were used to transform calcium chloride-treated E. coli JA221 as described previously (16). Restriction enzymes, DNA ligase, Klenow fragment, and BamHI linkers were purchased from Anglian Biotechnology. Plasmid mobilizations. Recombinant broad-host-range plasmids were constructed and characterized in E. coli JA221 and mobilized into P. aeruginosa strains either by using pTH10 as described previously (13) or by using HB1O1(pRK2013) in triparental matings (14). Amidase assays. Amidase activity in intact cells was measured by the transferase assay (8) with acetamide as the substrate. Activity levels presented in this article are the

mean values of duplicate assays carried out on at least three separate occasions. One unit represents 1 ,umol of acetohydroxamate formed per min per mg of bacteria. DNA sequencing. Restriction enzyme fragments from pAS20 were isolated from agarose gels as described previously (28) and ligated either into appropriately cut M13mpl8 or M13mpl9 vectors or pUC18 or pUC19 vectors (31, 33, 43). Single-stranded DNA from M13 recombinants for use as sequencing templates was purified by standard methods (36). Plasmid DNA for sequencing was prepared by the standard minipreparation method (32). DNA sequences were determined by the chain-termination method (37) for M13 recombinants with universal primer and for plasmid recombinants with universal and reverse primers. Sequencing reactions utilized standard and 7-deaza-dGTP premixes (Pharmacia) and [35S]dATP (NEN). The sequencing strategy is shown in Fig. 1. DNA sequence analysis. The DNA sequence was analyzed with the GCG software package (version 6.1) from the University of Wisconsin (15). The codon preference statistic (22) and third-position GC bias (4) were each calculated over a window of 25 codons with a codon usage table obtained by using published P. aeruginosa sequences from the EMBL library, release 14.0, plus the amiE sequence (7).

WILSON AND DREW

4916

XhoI

XorII

I

I

AccI XhoI

I

I

J. BACTERIOL.

Clal

I

Clal PvuII

I I

a) H

p

XK

I

r

*-

I

100Obp ~-41FIG. 1. Sequencing strategy for the amiC gene. The map shows some of the restriction targets within the amiC region. The amiC coding region is shown by the thick line. The arrows indicate the direction and extent of the sequencing reactions. The sequence was obtained by cloning unique fragments after restriction enzyme digestion.

AmiC purification. AmiC was purified from PAC452 pSW41 grown in Oxoid no. 2 broth to an optical density at 670 nm of 0.5. After addition of 5 mM isopropyl-p-Dthiogalactopyranoside (IPTG), growth was continued for 4 h, and the cells were harvested by low-speed centrifugation. The cell pellet was resuspended in 1/50 of the original volume of cell lysis buffer (20 mM Tris HCl [pH 8.0], 1 mM dithiothreitol, 0.5 mM phenylmethylsulfonyl fluoride, 5 mM EDTA) and disrupted by sonication on ice (MSE Soniprep 150). Disrupted cells were clarified by microcentrifugation (5 min at 12,000 x g), and the cleared lysate was aliquoted and stored at -70°C after addition of 10% (vol/vol) glycerol. The crude cell extract was fractionated by fast protein liquid chromatography (FPLC) (Pharmacia) gel filtration and ionexchange chromatography. Two hundred microliters of cell extract (10 mg of protein) was applied to a Superose HR 10/30 gel filtration column prewashed with elution buffer (EB) (20 mM Tris HCl [pH 8.0], 1 mM dithiothreitol, 1 mM EDTA) and eluted with EB at a flow rate of 0.3 ml/min. The large peak eluting at 13 to 15 ml was collected in three 1-ml fractions. Each fraction was then loaded separately onto a MonoQHR 5/5 ion-exchange column prewashed with EB. The column was eluted with a linear gradient of 0 to 0.6 M NaCl at a flow rate of 1.0 ml/min. The fraction eluting at 0.32 M NaCl was characterized as AmiC by SDS-PAGE and appeared to be substantially free from other proteins. The purified AmiC protein was stored at -70°C in EB with 15% (vol/vol) glycerol. N-terminal amino acid sequencing. The FPLC-purified AmiC was concentrated and desalted by high-pressure liquid chromatography (HPLC) using a Brownlee Aquapore RP300 (2.1 [inner diameter] by 60 mm; Anachem, Bedford, United Kingdom) with a solvent gradient of 5 to 65% acetonitrile in water at a flow rate of 0.13 ml/min. The acetonitrile contained 0.04% trifluoroacetic acid, and the water contained 0.06% trifluoroacetic acid, and the A214 was monitored. Apart from the salt flowthrough, a single large peak with a retention time of 21.8 min was seen and was followed by a broad tail. The main peak was collected and used for N-terminal amino acid sequence analysis. Approximately 80 pmol (5 ,ug) of HPLC-purified AmiC was subjected to N-terminal amino acid sequence analysis. The first 19 amino acids were sequenced by automated Edman degradation using an Applied Biosystems 470A gas phase sequencer with on-line detection of amino acid phenylthiohydantoin derivatives with a 120A HPLC (Applied Biosystems, Warrington, United Kingdom). The AmiC polypeptide gave a single amino acid sequence with an initial yield of 70% and repetitive yields of 91%. Protein electrophoresis. Protein gel electrophoresis (24)

I

I

aniR

5 5.3

KWb

b) H

Im

I

0 l Li

l

fmm pJB950 x

x

x

pSWl

1--I I

I

I

I

I

I

I _

p

SW2

I pSW5

fron pAS20

c)

I

S

X

XCCP

L

I

I

pSW3

Il

l

I pSW4

FIG. 2. Restriction maps of P. aeruginosa DNA inserts in pBR322-derived recombinant plasmids. (a) Map of pJB950 (13), which carries DNA from PAC433, and pAS20, which carries the same fragment from the wild-type PAC1. The positions of amiE and amiR are shown with thick lines. Restriction sites are Hindlll (H), PvuII (P), XhoI (X), KpnI (K), ClaI (C), and Sall (S). (b) Restriction maps of pJB950 derivatives pSW1, pSW2, and pSW5. (c) Restriction maps of pAS20 derivatives pSW3 and pSW4.

was used to investigate cell extracts of strains carrying recombinant plasmids and to monitor the AmiC protein purification. Samples were mixed with loading buffer containing SDS and electrophoresed on SDS-8 to 12.5% polyacrylamide gels as described by Laemmli (24). The gels were fixed and stained with Coomassie brilliant blue. Nucleotide sequence accession number. The DNA sequence presented in this article will appear in the EMBL/GenBank/ DDBJ nucleotide sequence data base under the accession number X13776 as an update of the existing (amiR) entry.

RESULTS Isolation and characterization of the wild-type (PAC1) amidase genes. A Sau3A partial digest of PAC1 chromosomal DNA was ligated to BamHI-digested arms of the A vector L47 (27), and the mixture was packaged in vitro. The phage preparation was propagated on E. coli W3110 and X-ami phages were identified as described previously (17) by growth haloes around plaques after infection of W3110 on nitrogen-free glucose-acetamide plates. DNA was prepared from the phage of five Ami+ plaques and restriction enzyme mapped with HindlIl and Sall. A 5.3-kb HindIII-SalI amiER fragment was isolated from one of the recombinants and subcloned into pBR322. Plasmid pAS20 thus carries the wild-type amidase genes (Fig. 2a) and is similar to pJB950 (12), which carries the same fragment from the magnoconstitutive strain PAC433. Restriction enzyme analysis of pAS20 showed no changes from the pJB950 map. Investigation of the amiR promoter region. Previous studies with plasmid pJB950 indicated that the amiR promoter lay between KpnI (2.4 kb) and ClaI (3.25 kb), more than 400 bp upstream of the amiR open reading frame (Fig. 2a) (13).

amiC REGULATION OF P. AERUGINOSA AMIDASE EXPRESSION

VOL. 173, 1991

4917

TABLE 2. Amidase assays of E. coli JA221 carrying plasmids pJB950, pAS20, and reconstructed derivatives Amidase activity under indicated growth conditionsa Plasmid

pJB950 and derivatives pJB950 pSW1 pSW2 pSW5

Mean amidase

Glucose

Glucose-lactamide

Succinate

Succinate-lactamide

Succinate-butyramide

3.8 0.4 0.5 7.0

4.0 0.3 0.3 7.8

3.2 0.3 0.2 9.9

3.8 0.3 0.3 7.8

3.5 0.3 0.3 10.9

3.7 0.3 0.3 8.7

0.0 2.1 0.1

0.1 1.6 0.1

0.0 1.6 0.0

0.0 1.6 0.0

0.0 1.6 0.0

0.0 1.7 0.0

activity

pAS20 and derivatives

pAS20 pSW3 pSW4

a Amidase activity was measured as micromoles of acetohydroxamate produced per minute per milligram of cells. Lactamide (0.2%) was used as a nonsubstrate inducer and butyramide (0.2%) was used as a repressor of amidase synthesis.

Plasmids pJB950 (PAC433 genes) and pAS20 (PAC1 genes) were each cut with XhoI and SaiI and religated. After transformation into E. coli JA221, plasmids pSW1 through pSW5 were identified and characterized (Fig. 2b and c). Amidase assays were performed with strain JA221 carrying these plasmids and the two parental plasmids grown under inducing, noninducing, and repressing conditions (Table 2). The parental plasmid pJB950 shows low constitutive amidase activity (13). pSW1, which has the small (658-bp) XhoI fragment inverted, shows a 10-fold decrease in activity compared with pJB950; loss of the 658-bp XhoI fragment and inversion of the large XhoI fragment (pSW2) cause a similar decrease in amidase activity; and pSW5, which has just the 658-bp XhoI fragment deleted, shows a high level of constitutive amidase activity, as seen previously with Bal 31generated deletions in this region of the DNA (13). The level of amidase activity from the wild-type genes in pAS20 is too low to be measured accurately under normal growth conditions (Table 2); however, loss of the 658-bp XhoI fragment causes amidase to be expressed constitutively (pSW3), and loss of this small fragment and inversion of the amiR gene XhoI fragment lead to residual activity (pSW4). As suggested previously (13) and now shown here, the amiR promoter does not lie within the 1.5-kb XhoI fragment, since inversion of this fragment causes loss of amidase activity (pSW2). These results suggest that the amiR promoter lies within the 658-bp XhoI fragment. In addition, the constitutive amidase expression from pSW3 indicates that regulation of amiR activity or expression determines the amidase phenotype. Studies of amidase inducibility. To investigate amidase inducibility, several new plasmid constructs were made (Fig. 3a). Plasmid pAS21 carries the PAC433 amiE gene HindIIIXhoI fragment joined to the PAC1 1.5-kb amiR gene fragment in the normal orientation. This plasmid yields a high level of constitutive amidase activity of 11.0 U under all growth conditions. This result, a confirmation of the pSW3 result, shows that either the site of action for a regulator or a regulator itself is disrupted in these constructs, and this disruption leads to constitutive amiR expression and thus constitutive amidase expression. Plasmids pSW36 and pSW37 (Fig. 3a), derivatives of pAS20, were constructed to distinguish between these possibilities. pSW36 is a deletion derivative of pAS20 lacking the 250-bp Clal fragment which is 200 bp upstream of the amiR open reading frame and downstream of any proposed control elements; this plasmid shows a constitutive amidase expression level of 2.3 U under

all growth conditions. pSW37 has an 8-bp BamHI linker inserted into the unique EcoRV target in the 658-bp XhoI fragment, and this in vitro mutation causes a constitutive expression level of 2.0 U. To test the hypothesis that amiR expression or activity is normally regulated and that the amount of active AmiR produced determines the level of antitermination and thus the level of amidase expression, a transcomplementation system was constructed in E. coli (Fig. 3b). Plasmid pDC5 carries the HindIII-XhoI amiE fragment from PAC433 (13), and pSW35 is a pKT231 derivative carrying the wild-type 1.5-kb XhoI amiR fragment. Strain JA221 carrying pDC5 and pSW35 produces a high level of amidase activity (mean value = 38.6 U) under all growth conditions. Such expression is AmiR dependent (13), and by showing high amidase expression with amiR expressed from the vector neomycin phosphotransferase II promoter, the hypothesis is confirmied. We conclude from these studies that the level of amidase activity is determined by the level of expression or activity of amiR and that this regulation of amiR is at least partly determined by a new gene, amiC, which lies upstream of amiR. DNA sequencing studies. The DNA sequence of the 658-bp XhoI fragment and the left end of the 1.5-kb XhoI amiR gene H

P

amiE1 I

XE

I 2

X

XCC P

I3

'' amiR4 ---

S

I Kbp

a) H I

PAC433 PA43

X

PAC1 PAC

X

I

I

X pAS21 A2

pSW36

CC

T

I

pSW37

b) I

PAC433

PACI i

o

I

pDC5

_

pSW35

FIG. 3. Restriction maps of pBR322-derived plasmids designed to locate amiC. Restriction targets are HindIll (H), PvuII (P), XhoI (X), EcoRV (E), ClaI (C), and Sall (S). (a) Restriction maps of pAS21, pSW36, and pSW37. (b) Plasmids used in the amiR complementation system, pDC5 (13) and pSW35.

4918

J. BACTERIOL.

WILSON AND DREW 10

20

~

30

60a

40

GGTACGCGTGGCCGAGCATCTGCTCGATCACCACCAGCCGGGCGACGGGAACTGCACGATCTACCGT 80

120

130

GGGGAGCCTGGAG~CACGAGCGGGTTCGCTTCGGT'ACGGCGCTGAGCGACAGTCACAGG-AGA-GOAAACGGG 140

00GGG MET

160

TCG

GIyS

CAC

rHi

200

GA

0CC

A

210

ATC GAG

TCG

r

e

OGT

GGC

Iy V

CGCG

GAG

As

ysTOG

GAG COG

A COG CGCAG

ACGO

GAO

GAG

s 400

GAG

CTG TAT

TTC ATC

GGC

COGC

GAG

680 690 CCC TCC GAG

GAG

GAG

TTG

TTC GAO

ATC Ie T

TAG

As

TAT

yr

700 GC

CGA

GCGC

750,

740 GTC TTC TCC ACC

VaIPhe

SerTh rVa I

800

GGC

TAG

Ly

s

AGO

GAG AGT

Me

G

Va

IAIa

0CC AGC COG A

ACC Th

830

880 GAG

COGC

AAC

r9 G Iu S

P

CTG

amniE

GAG

nA

840 ACC ACC AGC GAG

900 TAG

0CC

eA

GCG

hwhmlg

hc

from

an

are

16S rRN

some

CGT

TAT

bindinig

eedn

otecnessro

prsmbyacts

eugns Vra I. Mr

GACGOTG COGC

which

site (42). amiC starts with

an

asthe

ribo'-

ATG codon and

Ia ArgAr

start

codon. AmiC is

a

polypeptide with

385-amino-acid

a

predicted size of 42,834 Da. Codon usage in amniC is similar to but less extreme than that of arniE and arniR (7, 28).

850 GAG

GTG OGOCG

nui'lt

910 TTC

idtp

ftecoe

ee,adfv

mds

Hindlll-SalI fragment from wscntutdi .cl lsi S 11 di tiona-"y- tw-o bra-otrng lsis ""'W' and pSW41YA ArgV

1020

1010

expressed

nucleotides 8 to 24 and 65 to 82,

two blocks of sequence,

GTG ATGO

790 0CC COT

0CC ATC

950

990 0CC TOO 0CC GAG

is

(18); however, upstreamn of amiC

TCC AGC ATCG r Phe Ser Ty SerIl hec ative of pKT231 (2) carrying the 970 0CC TOGC CAT GOT TTC TTC CCG GAG AAC OGOG Ia Cy sH i sG IyPhe Phe Pr-o G uAsn A I a p

940

0CC TTC GTC

GGC

r9AIaApV

COGC

E. coli-like promoter

730

GCG COGC

780

890

Ia SerArg A IaPheVaiG I

GOGC

CAT

720 GAG

ATC TAG

TAG

550 GAG TAG

ATT CCG

GTG GTG GTC OGCG CCT G nVa Va Va IA Ia Pro GAG

uG I y

GI

CTG ATT

'OGI uL euTy rAr9A Ia

0CC AGC

GTC

ATC

CCG COG GAA AGC

TAG

I

aG GAGp AOG

540

0CC GAG CTG TAT

rA I

Previous studies have shown that GTO

TTG

uAO

TCG CCG AAC

TAT

uAgIIeT

820

930

980

GGC

CCG

430

TaIG

oT VaI

P

770

Iy ThrG Iy

GACGOTG GCA

Asp

uSe

OAT GTACGO CCC Asp Th rPro

~~~~710 COGC

GTC GAG

CCG CCG ATC

COO

870

ATG t

G

810

GGC

GAG

860 AAG

Va

ACC

GTA CGOG

CTC GAG GAA ATC TAG

760

GTG GTG GGC

r

370

420

GCG GCG

CCG CTG

erApApApLuGInAgAIaV

Pr

ATe

GGC GGC ACGO GTG

GAG

CGG

530

TCG GAG

er

GGC

91an116adlastani-rmdetonfprtfth C-terminal region of amiC. The 8-bp BamHl liniker insertion inpS 3 cassafaehf mutation atthe 5' end of amiC. sc ssp Pa msroi

GAG

A 360

AAG

rsde

is between

COG

eArgAsnArgGOI yVa IAr-gPhe LeuVa

VaIT

GCG

GTG GTG V V

I I

GGC GGC

A

410

510 520 GAG AAC AGT CCG AAC

GCG

CGOG

ATT

h

AAG

CCG ACC CCC TAG

TAG

TTG

AGOGCOGy

rA

TAGAT e e

S 350

pSW36

in

250

CT0 AAC aIGIuGInLuA

GTC GAG CAA

GAG GAG

h

GGG

GCG

e

ofthe ClalIeeto

GGC OGT AGCeeincrsde ThrGl Iy Va ITh r-

GAA ACC

240

TTG CTC

~34'0

390

TCC

euPhe.er,GG IuT

e

COG

yrAgLuC

CGCTC TOGC 500 GOT CCG

TOGC

CGOG CGT

TAT

r 380

OG

GOT

GGC GCA

yrGIy

G Iy A

y

330

320

C

GCG TAT

GAG

position

190 CTG TTC

230

220

COGC

s GGC GGC

170

CCG CTG ATC GGC CTG nGlIauA rg P roL euGI IeGIy LeuL

GAG GAG COG

sG I

alyi,the motprobable opnreading faefor amClies betweenbae 136 an 25(data ntpresented).Th

1030

GCG 0CC TAG TOG CGA ACC TTG TTG CTC GGC COGC 0CC GCG Th rLIIIaTh r AyI, Abroad-host-range a VI-aAJI Lrp yr rp G n h r eu eu eu y 1040 1050 1070 1080 1090 1060 were made. based on the ex pression vector GAG 0CC OGCA GGC AAC TOG COG GTG GACGOTG GAG COO GAG CTG TAG GAG ATC GAG ATC G I Va IG I i Ia AIa Ar GInA Iy AsnTrp 9VaIG uAsp nAr-gH sLouTy r Asp II eAsp II amiR geefragment carries the 1.5-kb

ACC ATC ACC

pMMB66HE

1110

OGCG CCA

GAG

CCG

GAG

1160

COGC Arg

ATC

uI

I I

COGC

OAT

CCC GAG

CCT

TAT

eArg Proa Asp Proa Ty r 1280

COO

1130

GTG

GAG

COGC

1180

GAG

OrG TTC

Va

AAC

GAG

AAC

GTCGOTG

CAT AAC

Va

Va

H Is

Va

1200

TCT TCG

1210

COGC TOG GAG TCG CCC IAr-gTr-p G I nSerProG I

1250

GTC

AGC COGC CTG

CCG

GAG GTC

G InVa IPhe

1240

1050pS

1140

1190

GCG COGC GGC

eAspAI a ArgG Iy

I

1230

1220 ATT

GTC

1170

GCG IeAIa G

ATC I

1120

CTC

GAG

GAG

1260 TOG TCC

Asn Laeu Asp Asp Trp Se r

uPro

1270 0CC

AI

AGC ATG GGC Ser Met GoIy

nosa

~after-

G I I A I a G L eu Pro

FIG. 4.

Kpnl-Pvull

pS4 ares the wild-type amiC gene on a fragment downstream of tac (Fig. 2a). The constructs were characterized in E. coli JA221 and mobilized into P. aerugi-

1290

GGGyGtAIGGaTCCA.PTGAy y sequence

DNA runs

sequence

from the

of the

Kpnl

Nucleotides

are

arniC gene

region. The DNA

site to the arniC termination codon.

The DNA strand shown has the

same

orientation

numbered from the intact

acid translation of amiC is shown. The

promoter sequences, nucleotides 8

Kpnl

as

the mRNA.

strains. Amidase assays growth

under inducing,

conditions (Table 3). pSW1O1 host which carries

proposed rpoN-dependent

to 24 and 65 to

81,

are

under-

the strains

mobilized into PAC452,

a

A42pW0)sosnra

pAS2O

confirm that the

HindIII-Sall quences

fragment from pAS2O has been determined. sequence shown (Fig. 4) runs from the unique

1.3-kbp

The

Kpnl target to

the arniR initiation codon. On the basis of the codon

prefer-

CC bias, and rare-codon usage

expression

inducible amidase

elevated because of the copy

E. coli grown under nitrogen-limiting (unpublished observation). These findings wild-type amidase genes are present on the

in

growth conditions

Amidase phenotype8

was

on

and repressing

number (Table 3). A similar result has been obtained with plasmid

as are the proposed Shine-Dalgarno sequence (nucleotides 121 126) and the EcoRV target in the N-terminal region (nucleotides 198 to 203).

Strain

performed

noninducing,

chromosomal deletion which includes

h,aiaegns

alo

to

statistic, third-position

a

were

target. The amino

lined,

ence

fo

Xhol

pW

1100

are

Strain PACi when

fragment and suggest that no other DNA serequired for inducible amidase expression. shows normal inducible

plasmidless,

and

PAC1(pSW40)

amidas'e

expression

grown in the absence

of LPTG shows low noninduced and normal induced levels of amidase

expression (Table 3).

This

indicates incomplete

TABLE 3. Amidase assays of P. aeruginosa strains carrying recombinant plasmids Presence or Amidase activity under indicated growth conditionsb Plasmid

absence of

IPTG

Succinate

Succinate-lactamide

Lactate

Lactate-lactamide

Lactate-butyramide

-

8.1 4.6 6.6

0.1 0.7 9.0

2.5 4.6 8.2

0.0 0.3 4.2

PAC1 PAC1 PAC1

Ind Ami+ Ind Ami+ Ind Ami+

pSW40 pSW40

+

0.1 0.8 8.1

PAC452

AmiA&

pSW101

-

1.1

51.9

0.6

30.7

0.9

PAC101 PAC101 PAC101

Con Ami+ Con Ami+ Con Ami+

None pSW41 pSW41

-

17.8 3.5 2.8

11.3 2.0 2.1

8.8 1.6 2.9

3.5 1.8 1.5

7.3 3.4 1.2

PAC111 PAC111 PAC111

Con Ami+ Con Ami+ Con Ami+

None pSW41 pSW41

6.7 0.5 0.4

9.2 0.4 0.2

2.7 0.1 0.3

3.4 1.2 0.6

0.1 0.1 0.1

None

Ind, inducible; Con, constitutive. b Amidase activity was measured as

+

+

micromoles of acetohydroxamate produced per minute per milligram of cells.

amiC REGULATION OF P. AERUGINOSA AMIDASE EXPRESSION

VOL. 173, 1991 a

b 1 2

3

4

kDa 661-

kDa

4511 66m-

-

3611

.sezffi

_ amiC

":.

45,.

36_-

20_ 141_

FIG. 5. (a) SDS-polyacrylamide gel of cell extracts. Samples were run on an 8% polyacrylamide gel. Lanes: 1, PAC452; 2, PAC452(pMMB66HE) + 5 mM IPTG; 3, PAC452(pSW41); 4, PAC452(pSW41) + 5 mM IPTG. (b) SDS-polyacrylamide gel of FPLC-purified AmiC. Samples were run on a 12% polyacrylamide gel. Lanes: 1, PAC452(pSW41) + 5 mM IPTG; 2, FPLC-purified AmiC.

repression of the tac promoter by the lac repressor in pMMB66HE, since the region upstream of amiR has no promoter activity (13). PAC1(pSW40) grown with IPTG shows constitutive amidase activity under all growth conditions by the provision of AmiR in trans. This finding confirms the pDC5-pSW35 result and shows that AmiR functions in the absence of the normal amide inducer and that regulation of amiR expression or activity determines the amidase phenotype. To test the role of amiC as a negative regulator of amidase synthesis, pSW41 was mobilized into two amidase-constitutive mutants, PAC101 and PAC111 (9). PAC101 is the original constitutive parent of the magnoconstitutive strain PAC433 (40) and shows constitutive amidase activity under all growth conditions (Table 3). pSW41, the amiC expression vector, carries the potential rpoN-dependent promoter sequences described above, upstream of the amiC open reading frame, and no differences are seen between results obtained in the presence and in the absence of IPTG in the growth medium. The presence of pSW41 causes a 75% reduction in amidase activity in PAC101, which is not relieved by the presence of amide inducer. PACll is an amidase-constitutive mutant which remains sensitive to butyramide repression, and pSW41 causes a greater than 10-fold reduction (7% residual activity) in amidase activity, which is again not relieved by the addition of inducing amide. Overexpression, purification, and N-terminal sequencing of AmiC. For studies of the AmiC protein, plasmid pSW41 was mobilized into the amidase deletion strain PAC452. Preliminary studies showed that cell extracts of PAC452(pSW41) grown in the presence or absence of IPTG contained a major new band with a molecular mass of 43 kDa. This band was not present in the plasmid-free strain or in PAC452 carrying the pMMB66HE vector (Fig. 5a). The AmiC polypeptide was purified to homogeneity by FPLC on the basis of its appearance on SDS-polyacrylamide gels. The initial fractionation using gel filtration showed that AmiC migrated as at least a dimer. A second-stage fractionation using ionexchange chromatography eluted AmiC as a single peak at 0.32M NaCl. The AmiC fraction from the ion-exchange stage was essentially free from other proteins (Fig. Sb).

4919

The purified AmiC protein was subjected to N-terminal amino acid sequence analysis for confirmation of the purification procedure and gene identification. The first 19 amino acids were sequenced and shown to be in complete agreement with the DNA sequence (data not presented). However, the amino acid analysis showed 5% N-terminal methionine and 95% N-terminal glycine. It is thus concluded that, under normal circumstances in vivo, the N-terminal methionine is removed. DISCUSSION Previous investigations of the regulation of amidase synthesis were carried out with cloned genes from a magnoconstitutive mutant, PAC433. The location and DNA sequence of the amiE gene and amiR gene were determined, and AmiR was shown to function as a transcription antitermination factor. These studies additionally showed that the amiR promoter lay some distance upstream from the gene itself. We have now isolated the wild-type genes in order to investigate the induction of amidase synthesis. Studies with new recombinant plasmids derived from both the magnoconstitutive and the wild-type parental strains showed that the amiR control region lay outside the large XhoI fragment carrying the amiR gene and that at least a part of it lay within the 658-bp XhoI fragment. One of these constructs, pSW3, which carries functionally intact wild-type amiE and amiR genes, expressed amidase constitutively. A similar result was shown by pAS21, and these results suggested to us that induction does not occur by the inducing amide binding to AmiR to produce a functional antiterminator, but rather that amiR expression and/or activity is regulated and production of functional AmiR leads to amidase synthesis. This was confirmed by transcomplementation studies using constructs with the wild-type amiR gene expressed from vector promoters. In E. coli JA221(pDC5 [amiE], pSW35 [amiR]), where amidase expression is strictly pSW35 dependent and amiR is expressed from a pKT231 promoter (13), a high amidase activity level is seen under all growth conditions. The only difference between the strain carrying pJB950 (amidase activity, 3.7 U) and the strain carrying pDC5 and pSW35 (amidase activity, 38.6 U) is that for the latter the amiR gene is present at a lower copy number but is expressed from a vector promoter. In the case of pSW40 (the controllable amiR expression vector) in the wild-type strain PAC1, in addition to the normal amide-inducible amidase activity shown by PAC1, production of AmiR in trans leads to amide-independent amidase activity. To investigate the role of the DNA sequences upstream of amiR in the regulation of amiR expression, two plasmids were constructed. Both pSW36, which carries a 250-bp deletion 200 bp upstream of amiR, and pSW37, which carries an 8-bp insertion 900 bp upstream of amiR, express amidase constitutively. This led us to conclude that a regulatory gene (amiC) lay in this region, which was confirmed by DNA sequence studies, protein purification, and N-terminal amino acid sequencing. In previous studies, constitutive mutations of the amidase system were used to define amiR. It now appears likely that at least some of these mutations lie in amiC, since the amiC mutant plasmids pSW36 and pSW37 express amidase constitutively. To test this prediction, the amiC expression vector pSW41 was mobilized into two constitutive mutants, PAC101 and PAC111. Both of these strains have been previously characterized as high constitutives, but they differ with respect to butyramide repression. PAC101 is resistant to butyramide repression, and PAC111 is sensitive.

4920

WILSON AND DREW

In both strains, pSW41 caused repression of amidase synthesis which, while incomplete, could not be relieved by addition of amide inducers. It thus appears that the constitutive mutations in strains PAC101 and PAC111 are amiC mutations which are complemented by pSW41. This experiment confirms the function of amiC as a negative control element. These studies indicate that the amidase system comprises two separately regulated units, the amiE gene regulated by transcription antitermination and the amiCR genes coordinately regulated from the rpoN-dependent promoter(s). The problem of induction and the function of inducing amides in the process remains to be solved and requires further investigation. At this preliminary stage, we have considered three different basic mechanisms for AmiC regulation of amidase expression. Firstly, AmiC is a conventional repressor which binds to an operator located either upstream of amiE, close to the amiCR promoter(s), or possibly both. This model seems unlikely for three reasons. AmiC shows none of the characteristic features of DNA-binding proteins (25). The studies of Farin and Clarke (19) would have been expected to generate conventional amiC(Ts) mutants, that is, amidase inducible at low temperature and amidase constitutive at high temperature, and despite extensive searches, no mutants of this type were found. Finally, using filter-binding and gel retardation assays with the whole 5.3-kb DNA fragment and smaller fragments, we have been unable to detect DNA binding with the FPLC-purified AmiC (unpublished observation). A second model for AmiC regulation could involve direct protein-protein binding between AmiC and AmiR to regulate the antitermination activity of AmiR. In this model, inducing amides would be expected to bind to AmiC and cause release of AmiR. The final model for AmiC regulation of amidase expression is one in which AmiC covalently modifies AmiR to regulate the antitermination activity. We have preliminary evidence that the purified AmiC exhibits protein kinase activity, and at this stage we believe that the covalent modification model represents the way in which AmiC regulates amidase expression. This type of two-component regulatory system with a repressor and an activator is not unique to the P. aeruginosa amidase system. Two other apparently similarly regulated catabolic systems in unrelated bacterial species have been described. The bgl operon of E. coli, which degrades aromatic P-glucosides, consists of three genes expressed from a single, normally cryptic promoter. bglG encodes a transcription antitermination factor; bglF encodes a membranebound ,-glucoside permease-phosphotransferase and the structural gene, bglB (29, 38). DNA sequence analysis has shown homology between bglG and the sac Y gene of Bacillus subtilis, another transcription antitermination factor (41). sacY is closely linked to and coordinately regulated with sacX, and the two genes regulate sucrose-inducible expression of an unlinked extracellular levansucrase gene, sacB (26). In overall terms, it appears that the inducible amidase activity of P. aeruginosa is at least functionally similar to these two systems with an activator gene and a repressor gene. However, the amiC and amiR genes show no homology to the equivalent bgl and sac operon genes at the DNA or amino acid levels. We have shown that in addition to the uncharacterized catabolite repression effects on amidase synthesis, there exists, at least, a two-component mechanism for regulating amidase expression. AmiC exerts a negative controlling

J. BACTERIOL.

effect upon the expression or, more likely, the activity of AmiR, and via transcription antitermination the amount of functionally active AmiR produced determines the level of amidase expression. At present the role of the rpoN-dependent promoters remains unclear. Previous studies by Potts and Clarke (34) have shown that amidase synthesis in the constitutive mutant PAC111 is not subject to control by the P. aeruginosa nitrogen-regulatory system. However, it is possible that the induction process requires expression from these promoter sequences. ACKNOWLEDGMENTS Thanks are due, as always, to Pat Clarke for continuing support and encouragement, to Alison Sparrow for technical assistance in the early stages of the project, to Peter Rice (EMBL) for the codon preference analysis, to Andy Shearer and Chris Taylorson for advice and assistance with the protein purification, and to Brian Coles of the CRC Molecular Toxicology Unit for the N-terminal amino acid sequencing. We gratefully acknowledge their contributions. S.W. was the recipient of an SERC Research Studentship.

REFERENCES 1. Ambler, R. P., A. D. Auffret, and P. H. Clarke. 1987. The amino acid sequence of the aliphatic amidase from Pseudomonas aeruginosa. FEBS Lett. 215:285-290. 2. Bagdasarian, M., R. Lurz, B. Ruckert, F. C. H. Franklin, M. M. Bagdasarian, J. Frey, and K. N. Timmis. 1981. Specific-purpose cloning vectors. II. Broad host range, high copy number, RSF1010 derived vectors and a host-vector system for gene cloning in Pseudomonas. Gene 16:237-247. 3. Bazaral, M., and D. R. Helinski. 1968. Circular DNA forms of colicinogenic factors El, E2 and E3 from Escherichia coli. J. Mol. Biol. 36:185-194. 4. Bibb, M. J., P. R. Findlay, and M. W. Johnson. 1984. The relationship between base composition and codon usage in bacterial genes and its use for the simple and reliable identification of protein coding sequences. Gene 30:157-166. 5. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7:1513-1523. 6. Bolivar, F., R. L. Rodriguez, P. J. Green, M. V. Betlach, H. L. Heynecker, H. W. Boyer, J. H. Crosa, and S. Falkow. 1977. Construction and characterisation of new cloning vehicles. II. A multipurpose cloning system. Gene 2:95-113. 7. Brammar, W. J., I. G. Charles, M. Matfield, L. Cheng-Pin, R. E. Drew, and P. H. Clarke. 1987. The nucleotide sequence of the amiE gene of Pseudomonas aeruginosa. FEBS Lett. 215:291294. 8. Brammar, W. J., and P. H. Clarke. 1964. Induction and repression of Pseudomonas aeruginosa amidase. J. Gen. Microbiol. 37:307-319. 9. Brammar, W. J., P. H. Clarke, and A. J. Skinner. 1967. Biochemical and genetic studies with regulator mutants of Pseudomonas aeruginosa amidase. J. Gen. Microbiol. 47:87102. 10. Clarke, L., and J. Carbon. 1978. Functional expression of cloned yeast DNA in Escherichia coli: specific complementation of argininosuccinate lyase (argH) mutations. J. Mol. Biol. 120:517-532. 11. Clarke, P. H., and R. E. Drew. 1988. An experiment in enzyme evolution. Studies with Pseudomonas aeruginosa amidase. Biosci. Rep. 8:103-120. 12. Clarke, P. H., R. E. Drew, C. Turbeville, W. J. Brammar, R. P. Ambler, and A. D. Auffret. 1981. Alignment of cloned amiE gene of Pseudomonas aeruginosa with the N-terminal sequence of amidase. Biosci. Rep. 1:299-307. 13. Cousens, D. J., P. H. Clarke, and R. E. Drew. 1987. The amidase regulatory gene (amiR) of Pseudomonas aeruginosa. J. Gen. Microbiol. 133:2041-2052. 14. Deretic, V., P. Tomasek, A. Darzins, and A. M. Chakrabarty.

VOL. 173, 1991

15. 16. 17.

18.

19.

20.

21.

22.

23. 24. 25.

26.

27.

28.

amiC REGULATION OF P. AERUGINOSA AMIDASE EXPRESSION

1986. Gene amplification induces mucoid phenotype in rec-2 Pseudomonas aeruginosa exposed to kanamycin. J. Bacteriol. 165:510-516. Devereux, J., P. Haeberli, and 0. Smithies. 1984. A comprehensive set of sequence analysis programs for the Vax. Nucleic Acids Res. 12:387-395. Drew, R. E. 1984. Complementation analysis of the aliphatic amidase genes of Pseudomonas aeruginosa. J. Gen. Microbiol. 130:3101-3111. Drew, R. E., P. H. Clarke, and W. J. Brammar. 1980. The construction in vitro of derivatives of bacteriophage lambda carrying the amidase genes of Pseudomonas aeruginosa. Mol. Gen. Genet. 177:311-320. Drew, R. E., and N. Lowe. 1989. Positive control of Pseudomonas aeruginosa amidase synthesis is mediated by a transcription anti-termination mechanism. J. Gen. Microbiol. 135:817-823. Farin, F., and P. H. Clarke. 1978. Positive regulation of amidase synthesis in Pseudomonas aeruginosa. J. Bacteriol. 135:379392. Figurski, D. H., and D. R. Helinski. 1979. Replication of an origin-containing derivative of plasmid RK2 dependent on a plasmid function provided in trans. Proc. Natl. Acad. Sci. USA 76:1648-1652. Furste, J. P., W. Pansegrau, R. Frank, H. Blocker, P. Scholz, M. Bagdasarian, and E. Lanka. 1986. Molecular cloning of the plasmid RP4 primase region in a multi-host-range tacP expression vector. Gene 48:119-131. Gribskov, M., J. Devereux, and R. R. Burgess. 1984. The codon preference plot: graphic analysis of protein coding sequences and prediction of gene expression. Nucleic Acids Res. 12:539549. Kelly, M., and P. H. Clarke. 1962. An inducible amidase produced by a strain of Pseudomonas aeruginosa. J. Gen. Microbiol. 27:305-316. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature (London) 227:680-685. Latchman, D. S. 1990. Eukaryotic transcription factors. Biochem. J. 270:281-289. Le Coq, D., A. M. Crutz, R. Richter, S. Aymerich, G. GonzyTreboul, M. Zagorec, M. C. Rain-Guion, and M. Steinmetz. 1989. Induction of levansucrase by sucrose in Bacillus subtilis: involvement of an antitermination mechanism negatively controlled by the PTS, p. 447-456. In L. 0. Butler, C. Harwood, and B. E. B. Moseley (ed.), Genetic transformation and expression. Intercept Ltd., Andover, United Kingdom. Loenen, W. A. M., and W. J. Brammar. 1980. A bacteriophage lambda vector for cloning large DNA fragments made with several restriction enzymes. Gene 10:249-259. Lowe, N., P. M. Rice, and R. E. Drew. 1989. Nucleotide sequence of the aliphatic amidase regulator gene (amiR) of Pseudomonas aeruginosa. FEBS Lett. 246:39-43.

4921

29. Mahadevan, S., and A. Wright. 1987. A bacterial gene involved in transcription antitermination: regulation at a rho-independent terminator in the bgl operon of E. coli. Cell 50:485-494. 30. Messing, J. 1979. A multipurpose cloning system based on single stranded DNA bacteriophage M13. Recomb. DNA Tech. Bull. 2:43. 31. Messing, J. 1983. New M13 vectors for cloning. Methods Enzymol. 101:20-78. 32. Mierendorf, R. C., and D. Pfeffer. 1987. Direct sequencing of denatured plasmid DNA. Methods Enzymol. 152:556-562. 33. Norrander, J., T. Kempe, and J. Messing. 1983. Construction of improved M13 vectors using oligonucleotide-directed mutagenesis. Gene 27:101-106. 34. Potts, J. R., and P. H. Clarke. 1976. The effect of nitrogen limitation on catabolite repression of amidase, histidase and urocanase in Pseudomonas aeruginosa. J. Gen. Microbiol. 93:377-387. 35. Rodriguez, R. L., R. Tait, J. Shine, F. Bolivar, H. Heynecker, M. Betlach, and H. Boyer. 1977. Characterisation of tetracycline and ampicillin resistant cloning vehicles, p. 73-84. In W. A. Scott and R. Werner (ed.), Molecular cloning of recombinant DNA. Academic Press, Inc., New York. 36. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: a laboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 37. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467. 38. Schnetz, K., and B. Rak. 1988. Regulation of the bgl operon of Escherichia coli by transcriptional antitermination. EMBO J. 7:3271-3277. 39. Smyth, P. F., and P. H. Clarke. 1975. Catabolite repression of Pseudomonas aeruginosa amidase: the effect of carbon source on amidase synthesis. J. Gen. Microbiol. 90:81-90. 40. Smyth, P. F., and P. H. Clarke. 1975. Catabolite repression of Pseudomonas aeruginosa amidase: isolation of promoter mutants. J. Gen. Microbiol. 90:91-99. 41. Steinmetz, M., S. Aymerich, G. Gonzy-Treboul, and D. Le Coq. 1988. Levansucrase induction by sucrose in Bacillus subtilis involves an antiterminator. Homology with the Escherichia coli bgl operon, p. 11-15. In A. T. Ganesan and J. A. Hoch (ed.), Genetics and biotechnology of bacilli, vol. 2. Academic Press, Inc., San Diego, Calif. 42. Toschka, H. Y., P. Hopfl, W. Ludwig, K. H. Schleifer, N. Ulbrich, and V. A. Erdmann. 1988. Complete nucleotide sequence of the 16S ribosomal RNA gene from Pseudomonas aeruginosa. Nucleic Acids Res. 16:2348. 43. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequence of the M13mpl8 and pUC19 vectors. Gene 33:103-119. 44. Yanofsky, C., and J. Ito. 1966. Nonsense codons and polarity in the tryptophan operon. J. Mol. Biol. 21:313-344.