Complete nucleotide sequence of the penicillin G acylase gene and the flanking regions, and its expression in Escherichia coli

87

Gene, 56 (1987) 87-97 Elsevier GEN 02070

Complete nucleotide sequence of the penicillin G acylase gene and the flanking regions, and its expression in Escherichia coli (Recombinant

DNA; gene regulation; terminator; overproduction;

precursor; phage Iz pL promoter; protein

processing)

Sang-Jin Oh, Young-Chang Kim*, Young-Woo Park, So-Young Min, In-Sook Kim and Hyen-Sam Kang Department of Microbiology, Seoul National University, Seoul 151 (‘Korea) Received I8 December 1986 Accepted 27 March 1987

SUMMARY

The pga gene coding for penicillin G acylase (PGA) in Escherichia coli ATCCll105 was cloned, and its complete nucleotide sequence including 5’- and 3 ‘-flankmg regions was determined. Two nonidentical subunits that constitute an active PGA enzyme complex are known to be formed by processing of a common precursor molecule [Bock et al., FEMS Microbial. Lett. 20 (1983) 141-1441. This novel type of protein processing was confirmed by a nucleotide sequencing study together with amino acid sequencing of two PGA subunits. In addition, it was found that the initiation codon, AUG, is preceded by an authentic ribosome-binding site, a consensus promoter sequence and putative CAMP receptor protein (CRP)-binding sites, and that the termination codon, UAA, is followed by a putative transcriptional terminator. The promoter function was confirmed by galactokinase assay using gallic fusion plasmids. A recombinant plasmid was constructed to overproduce the enzyme using phage 1 pL promoter. Unexpectedly, thermal induction led to accumulation of the 94-kDa polypeptide rather than active PGA in large amounts. Western immunoblot analysis showed that this large polypeptide is the real precursor of PGA. It is evident, therefore, that the synthesis of active PGA in E. coli is affected by growth temperature and that the precursor processing step(s) is temperature-sensitive.

INTRODUCTION

Penicillin G acylase (PGA), an industrially important enzyme, hydrolyzes penicillin G to phenyl-

Correspondencelo: Dr. H-S. Kang, Department of Microbiology, Seoul National University, Seoul 151 (Korea) Tel. (877)2130, ext. 2581. * Present address: Department of Microbiology, Chungbuk National University, Cheongju 3 10 (Korea) Tel. 043 l-4-6181, ext. 444.

acetic acid and 6aminopenicillanic acid, a key intermediate for the synthesis of many semisynthetic penicillin derivatives. This usefulness in the enzyme industry has led to the accumulation of data mainly Abbreviations: aa, amino acid(s); Ap, ampicillin; Cm, chloramphenicol; CRP, CAMP receptor protein; GL-“IACA, 7@-(4-carboxybutanamido) cephalosporanic acid; HPLC, high-performance liquid chromato~aphy; nt, nucIeotid~s); ORF, open reading frame; PAGE, polyac~i~ide gel electrophoresis; PGA, penicillin G acylase; pga, gene coding for PGA; pL, major leeward phage 1 promoter; R, resistant; SDS, sodium dodecylsulfate; Tc, tetracycline; [ 1, designates plasmid-carrier state.

0378-l 119~87~$03.50 0 1987 Efsevier Science Publishers B.V.(Biomedical Division)

88

on regarding the aspects of PGA application. Recently, attempts to unravel the structure and expression of the pga gene were made in several laboratories (Mayer et al., 1979; Kim et al., 1983; Bock et al., 1983; Bruns et al., 1985; Oliver et al., 1985). Primarily PGA is synthesized as a 94-kDa precursor poiypeptide (Bock et al., 1983) from which the signal peptide of 26 aa is cleaved off to produce a proenzyme (Oliver et al., 1985). As in many eukaryotic hormones and secretory enzymes, it is further cleaved into a small and a large subunit which form an active enzyme together (Bock et al., 1983). The nucleotide sequence of the pga gene had been determined partially by Bruns et al. (1985) and Oliver et al. (1985). The PGA is induced by phenylacetic acid and subjected to catabolite repression by glucose (Kaufman and Bauer, 1964; Gang and Shaikh, 1976). The glucose repression can be overcome by CAMP (Gang and Shaikh, 1976). Moreover, CAMP itself is thought to induce the synthesis of PGA (Gang and Shaikh, 1976). The maximum level of PGA activity was found in cells grown at 28°C or beiow (Kaufmann and Bauer, 1964), whereas only a low level of activity was detected at 37°C. This low acylase activity is perceived to result from a decrease in the amount of active enzyme, since the enzyme itself is active above 37°C (Daumy et al., 1985). The regulation by CAMP and the effect of temperature on the enzyme synthesis, however, have not been understood at the molecular level. During the preparation of this manuscript the nucleotide sequence of the structural gene of PGA was published (Schumacher

TABLE

I

Bacterial

plasmids Source

AND METHODS

(a) Bacterial strains and plasmids E~c~e~c~~ coli ATCC 11105 as a source of chroSewatia marcescens mosomal DNA and ATCC27117 for PGA assay were purchased from American Type Culture Collection. In most transformation experiments, E. coli HBlOl was used. E. coli strains M5219 SmR lacZam trpA~{~~~0252~1857~Hl) and NlOO dpalK, recA, pro, Iac + ) were used for transformation with plasmids containing the a pL promoter and the gal&C gene, respectively. Plasmids used are listed in Table I.

(b) DNA manipulation and sequence analysis Plasmid DNA was prepared by the SDS-NaOH lysis procedure of Bimboim and Doly (1979). Reaction conditions for various enzymes and linkers used were as recommended by the manufacturers. The nucleotide sequence of the pga gene was determined by the chemical cleavage method (Maxam and Gilbert, 1980). The sequence data were analyzed

or reference

Genetic

pKM300

ApR, TcR

Kim et al. (1983)

pPAKS 1

TcR,

Kim et al. (1983)

pPAKS63

pga+ TcR, pga f CmR, pga +

pKOl1

ApR, galK

ATCC (McKermey

~28 pPAKSP,l

ApR> PL +

O.J. Yoo (Reed,

CmR, pL + , ma +

this paper

a

a

MATERIALS

Plasmid

pPAKS2

characters

et al., 1986). However, the report did not include the 5’ and 3’ regulatory sequences of the gene. In this communication, we report the complete nucleotide sequence of pga gene including the 5’- and 3’-noncoding flanking regions and describe several features of its regulatory regions. We also demonstrate that the processing of the precursor to active PGA is temperature-sensitive.

Kim et al. (1983) Kim (1986) et al., 1981) 1981)

pga+ , presence of the pga gene; pL + , presence of the active pL promoter.

89

using computer (1984).

programs

developed

For the C-terminal amino acid sequence, subunit protein was digested with carboxypeptidase Y (Klemm, 1984), and the resulting aa were analyzed by the Pica-Tag HPLC system (Waters).

by Staden

(c) Amino acid analysis

N-terminal amino acid sequences of subunit proteins were determined by automated Edman degradation using a gas-phase protein sequencer (470A protein sequencer; Applied Biosystem). The resulting phenylthiohydantoin derivatives were analyzed on HPLC with a 120A PTH analyzer (Applied Biosystem).

(d) Other procedures

Peptide mapping using Staphylococcus aureus V8 protease was performed following the method of Cleveland et al. (1977). Galactokinase assays were performed as previously described by McKenney et al. (1981). Immunization and Western immuno-

DPAKS~~ S.Okb

pPAKS2 14 kb

pPAKS1 24.5kb

B

C I

I

I

I

II

II

‘co--,

II

I

I

. -

z

c----o represent

cloned fragments,

of transcription

and lines indicate

are shown by horizontal

by pPAKS 1. This is probably promoters

next to the represent

e---o--

. strategy.

(A) Physical

the region of deletion as compared

arrows.

due to the reduced

The PGA activities size of plasmids

directed

maps of pPAKS 1 and the derivatives.

to pPAKS 1. The position by pPAKS2

and pPAKS63

and the efficient read-through

transcription

of

Open boxes

pgu gene and the direction

were much higher than that from the one of the vector

pgu gene. (B) Schematic diagram of the pga gene and flanking regions. The ORF and the direction of transcription

are shown by a horizontal Numbers

--

.

pga region and the sequencing

I

--co--,

l

2

IllIll

b-w

--

of the cloned

III

I

l

-

--

Fig. 1. Structure

III

arrow. Solid boxes represent

the nucleotide

position

the small- and large-subunit

as in Fig. 2. Possible

CRP-binding

coding regions, with N and C termini being indicated. site, promoter

site, ribosome-binding

site (RBS) and

pga region. Arrows show the extent and directions of sequencing. Abbreviations: Bg, BglII; C, ClaI; Dde, D&I; E, EcoRI; H or Hd, HindUI; Hf, HinfI; HpI, HpaI; HpII, HpaII; P, &I; S, SmaI; Sau, Suu3AI; T or Taq, 7”qI.

terminator

are shown. (C) Restriction

cleavage

map and sequencing

strategy

for the

90

Hindlll 181 MCMT

PI

-35

lC7 CTC CAAAlACAT%ACC~C

SIGWAL SfOUEWCf UT cm ATC A7C 6% AAC TG7 CT7 AC7 cc7 TCC krt Al-ym 118 VSi &I# cys VSl Rp Ah SS? LSu M

Tyr Tyr Trp %r

nm1 291ICC7C7CCTT~7sCCA7CTCC'tCcU T CAC ACT W nf PW val LN Ala T*, bp WbI VSI AIS 9 Pm y CIn kr CIy Rw mulmMG7u~aTmsK6c7McKsc7c Ly~m CIu AmPh~Cly

Tyr

AryLy~IkrUu

Lal pra Ah

A77 CCT CCC UT II* AIS m, W

LU

Ah

cUACAW7 CAT YEW Cly Th* VSI W LyS II

TAT CM @TaMi CTG Z:U Tyr c1U rW EIA LW

MC cnS~7CTGuSsaCcrtm;WrcCut~ck~Wbn‘2~ LyS CIn w VSI au Ah as tyr Clu ySF CIA ciu VII Lal nir VSI

91

blotting

procedures

were as described

by Towbin

et al. (1979).

aa (Fig. 1B). The protein

specified by this 2538-bp

ORF has been calculated

to be 94.3 kDa, which is

close to the iI4, estimated RESULTS

(a) Structure of the pga gene

gene and flanking

control

(Fig. 2). We initially which the presence

and CAMP on the expression

Strain

Penicillin

b

M9-glucose

within

a given

b

CAMP

With CAMP

ATCCll105

5.1 (100%)

2.8 (54%)

3.7 (73%)

HBlOl[pPAKSl]

7.5 (100%)

4.3 (57%)

6.5 (87%)

HB 101 [pPAKS2]

46.5 (100%)

38.5 (83%)

38.2 (83%)

HBlOl[pPAKS63]

113.5 (100%)

105.5 (93%)

106.5 (93.5%)

a E. coli strains

were grown

at 28°C in 250-ml flasks on a rotary

source. The cells were washed glucose

resuspended

with 0.05 M phosphate

(0.2%) and CAMP (2.5 mM), and incubated

in 1 ml of 0.05 M phosphate

nm and the remainder

for checking

(100 mg/ml in 50 mM phosphate solution.

base (Bomstein independent

shaker

the intensity and Evans,

in the 50-ml minimal

buffer (pH 7.5), resuspended at 28°C with aeration

enzyme

activity.

of the color generated

at 45°C for 1 h. The reaction

by reacting

containing

of the supernatant x reaction

washed

the absorbance

cells and 0.5 ml of penicillin

was measured

as

with glycerol

were harvested,

was stopped

and

at 600

G solution

by 7 ~01s. of color-reagent

at 415 nm. This method

6-APA with p-dimethylaminobenzaldehyde

1965). The activity is given as A,,,/(&,,

0.2% succinate

supplemented

was used for measuring

For the enzyme assay, 0.5 ml of bacterial

buffer pH 7.5) were incubated

medium

in the medium

for 4 h. 5-ml samples

buffer (pH 7.5). 0.5 ml of the 1 ml suspension

After cell debris was spun down, the absorbance

on the measuring

(PDAB)

depends

to form a Schiffs

time) x 100. Each figure is the average

of at least three

determinations.

b In parentheses:

Fig. 2. Nucleotide ribosome-binding terminator

of a promoter

restriction fragment led to the expression of galK gene upon proper insertion into the vector. We have

activity”

Without

working

promoter

vector system, in

of the pga gene

acylase

M9-glycerol

(0.2x),

the pga

screening

II

Effect of glucose

a carbon

localized

region by the promoter

region

was determined by the method of Maxam and Gilbert (1980). Analysis of the nucleotide sequence revealed an uninterrupted coding sequence for 846 TABLE

(Bock

except for several nucleotides (see Fig. 2). The differences may have resulted from the genetic variation. The initiation codon AUG is preceded by a putative ribosome-binding site with 4 nt in between

We have previously reported the cloning of the pgu gene and the characterization of its derivatives (Kim et al., 1983). The complete nucleotide sequence of the pgu structural

by SDS-PAGE

et al., 1983). There is general agreement with the sequence reported by Schumacher et al. (1986)

AND DISCUSSION

y0 activity

relative

to the glycerol-grown

sequence

of pgu gene inluding

site (RBS) and restriction

(facing horizontal

shown. The remainder

arrows)

The sequence

and amino acid residues

of the amino acid sequence are shown by downward

from nt position

207-2799

its flanking

regions.

The presumed

sites are shown. The start and stop codons

at the right and let? of each line, start arbitrarily peptide and the subunits

culture.

is deduced

determined

arrows.

is identical

by N- and C-terminal

from the nucleotide

from the beginning Dashed

promoter

of the sequence.

sequence.

amino

The nucleotide

The posttranslational

lines above the 5’-flanking

to that determined

site (-35

are doubly underlined.

by Schumacher

region indicate

of Valle et al. (1986), are indicated

by V.

-10

regions),

The inverted

acid sequencing sequence processing

the

repeat of

(boxes)

numbers,

are

written

sites for the signal

possible CRP-binding

et al. (1986). except for nt at positions

(T to C), 1073 (T to C), 1140 (T to G), 2152 (G to C), 2153 (G to C) and 2154 (C to G). The positions A, A, in the report

and

of the additional

(v)

sites. 647

4 nt, A, T,

92

-26 Penicillin G Acylase: : GL-7ACA Acylase

Met

Ile

Val

Ala

Ala

Ser

Tyr Pro

Trp Ser l ia Val

Ala

-cl

Lys

Leu

Arg Ara cl

Vi1

I;e

p"'I Ser

-29

Asn Ala

Cys

Leu Ala

Pro [d,Gln A a Leu Phe Ala Leu

Leu

Asn -

Val

Thr Met

a

m Ala

Ser Thr

Ala

Glu +l

Pro

Ser Thr

Asn Val

Leu

His

Giym'T'

Ser Thr

L:L

Glu Pro

Gln

9 Ile

_

_

_

_

_

-

_

-

Pro

Ile

Ala

Ala

Tyr

Pro

Arg

Ser

Asn

Glu

Met

Pro His c]Gly

Ile

Tyr

Ala

E;y

yr Gly KGly]Vai

Asn Val

Asp m

Thr Ala

Trp Pro

His Ser

Phe

Tyr

Gly

Tyr

Gly

Tyr Trp

Val Ala

Val Gln

Asp Ser

Arg His

Leu Gly

Phe Asp

Met

Ala

Arg Tyr

Arg Gly

Ser Glu -m

Thr Ala

Gln Arg

sp a

Phe Tyr

Val Glu

Lys Gln

Phe Thr

Asp Thr

Lys Val

Asp Trp

Leu

Ile

Arg Arg Leu

Thr

Asp Glu

Ala Arg

Ile Ala --

Arq Gln

Ala Gtn

Gln Trp

Ile Ala Tyr cl Ala

Ala Gln

Leu Gln

Ser Ser cl

Pro Glu Asp Met Ser Pro Asp Phe Arg Ala

Asn

Leu cl

Asp

Val -

Asn -

Thr -

Glu Asp

Leu

Leu

Ile

Ser

m

Gln Asn

Met

Glu

Ile

Leu Arg

Val Tyr

Leu Tro

m

b:i

cl

Ile

20

Val

Leu Gln Gly Tyr Ala

Phe

102 Trp Tyr

Thr Ala

Asp Gln

Lys Gln

113

116 Thr Asp

124

Fig. 3. Comparison of the amino acid sequences of penicillin G acylase and GL-7ACA acylase. The first amino acid of the small subunit of PGA is designated as + 1. Dashes are used for the best alignment of amino acids except in the region from aa residues P-102. Boxes represent regions of identical amino acid sequences. The underlined sequences in the region between the aa residues P-102 indicate identical sequences occurring at shifted positions. The part extending from aa 117 to the C terminus of the small subunit shows no apparent homology. The downward arrow indicates the signal peptidase cleavage site. The sequence of GL-7ACA acylase 8ene is by Matsuda and Komatsu (1985).

cloned restriction fragments containing various regions of the pga gene into a pK0 gaIK-expression vector. E. coli NlOO was transformed by the plas-

mid and plated on ~acConkey-galactose plates containing 50 pg Ap/mi. Colonies bearing the 360-bp HindUI-HpaII (Fig. 2) fragment appeared red due to

93

the acid production cating quence.

from galactose

the presence

of an inserted

We have tentatively

assigned

utilization promoter

and

-10

(TATACA)

se-

the promoter

between nt 215 and nt 242 (Fig. 2). The putative (TTGCTA)

indi-

regions

-35 show

homology with the E. coli consensus sequences, -35 (TTGACA) and -10 (TATAAT). The spacing between

these sequences

sensus range (16-18 (Hawley and McClure,

The ‘highly’ expressed comprised

of optimal

genes are almost

codons

1982), whereas in ‘moderately’ genes the proportion

(Grosjean

entirely

and Fiers,

or ‘weakly’ expressed

of optimal

codons

decreases.

Based on these criteria, the pga gene is a ‘moderately’ or ‘weakly’ expressed

gene, although the physiologi-

cal role of PGA is not yet understood.

(16 bp) is within the conbp) of E. coli promoters 1983). Close examination of

the 360 bp downstream from the stop codon, UAA, revealed that there is only one inverted repeat sequence of 11 bp (nt 2813-2839). This sequence can form a hairpin structure, common to most of the terminators (Fig. 1B). Mayer et al. (1980) reported that the recombinant strains harboring pHM8 and pHM12 exhibit constitutive and non-catabolite-repressible acylase synthesis. The partial nucleotide sequence from pHM8 and pHM12 (Bruns et al., 1985) differs from our sequence data for the region upstream from Hind111 site (nt l-206). The regulatory region of these strains might have been rearranged, because the Hind111 site used for initial cloning of the pga gene is close to the beginning of the pga gene. To avoid this problem, we have cloned the pga gene using EcoRI and PstI instead of Hind111 (Kim et al., 1983). E. coli ATCCl1105andHBlOl[pPAKS1],whichhavethe contiguous chromosomal DNA region upstream from the Hind111 site (Fig. lA), are repressed by glucose, as measured by the pga gene expression (Table II). The catabolite repression is partially overcome after the addition of CAMP to glucose-grown culture of E. coli ATCC11105 and HBlOl[pPAKS l] (Table II). In the cells harboring pPAKS2 and pPAKS63 (Fig. lA), in which the region upstream from the Hind111 site was deleted, the glucose repression was not overridden by CAMP (Table II). These suggest that the nucleotide sequence of the deleted region contains a regulatory site, e.g., the CRP-binding site of the pga gene. This idea is supported by the fact that the sequence nt 152-173 is 5’-&-TGTGA--G---TAA-TAA and the sequence nt 110-131 is 5’-TA-TGTGA--I---ACG-AJ’C on the antisense strand, which is close to the consensus sequence characteristic of the CRP-binding sites: 5’-AA-TGTGA--T---TCA-ATT (W indicates either A or T) (Ebright et al., 1984).

(b) Characterization

and amino acid sequencing of

PGA The PGA separated

enzyme

of E. cofi ATCC11105

by SDS-PAGE

was

into two nonidentical

subunits. The calculated sizes for the small and large subunits, based on the nucleotide sequence, were 23.9 kDa and 61.5 kDa, respectively. These values are close to those obtained from SDS-PAGE (Fig. CA). The translational start codon of the pga structural gene was identified by determining the N-terminal amino acid sequence of the purified subunits. Sequences of 20 aa from the N termini of the two subunits correlate perfectly with the nucleotide sequence of the two regions at nt 337-396 and nt 1126-1185, respectively (Fig. 2). Thus the region coding for the small subunit precedes that coding for the large subunit on the structural gene. A single precursor polypeptide is predicted for two subunits, since no termination codon is found upstream from the large subunit gene (Fig. 2). Digestion of the small subunit with carboxypeptidase Y yielded a C-terminal sequence of Thr-Ala (nt 958-963) consistent with the result of Schumacher et al. (1986). The small and the large subunits are separated by 54 aa (Fig. 2). The peptide between the subunits may be involved in proper folding of the precursor and must be removed posttranslationally as is the C-peptide of proinsulin. Recently, this type of protein processing was also reported in 7/3-(4-carboxybutanamido)cephalosporanic acid acylase (GL-7ACA acylase) originated from a Pseudomonas strain (Matsuda and Komatsu, 1985) and the PGA from Kluyvera citrophila (Barber0 et al., 1986). The amino acid sequences of connecting peptides in E. coli PGA and K. citrophila PGA, are conserved (Barber0 et al., 1986), supporting the hypothesis that this amino acid structure might be essential for the protein processing. The homologies reach 80% at the nucleotide level and 87% at the

94

amino

E. coli and

acid level between

pga-coding quences

region

(Barber0

et al., 1986). The se-

of the 5’- and 3’-noncoding

ever, show no homology.

regions,

Comparison

minal amino acid sequences E. coli PGA

K. citrophila

how-

of the N-ter-

of the small subunit

and Pseudomonas

GL-7ACA

of

acylase

shows partial homology (Fig. 3). It is speculated

that

these proteins share a common origin and divergent evolution accounts for the difference in their substrate specificity. and accumulation

of

the precursor To construct the plasmid bearing a strong promoter upstream from the pga gene, a restriction fragment of ~18 containing the I pL promoter was cloned into pPAKS63 using BAL 3 1 nuclease digestion and a Hind111 linker in the proper orientation Fig. 4). The newly constructed plasmid, pPAKSP,l, expressed the enzyme constitutively using the pga promoter at 28 ‘C. Upon temperature shift to 42 oC, not increase enzyme activity did the

-

c fragment pPAKS63 DNA

m

the phage 3, pL promoter

are indicated.

Abbreviations:

gene segments,

Cm’

pACYCl84

is

the

(Maniatis

Phoge h

Bg, BglII;

of the plasmid C, &I;

H,

the large and the

The shaded

containing

chloramphenicol-resistance et al., 1982).

from the plasmid

sources

respectively. fragment

The 800-bp

into the Hind111 site of

Hp, HpaI; T, TaqI. L and S denote

small subunit

was immunoreactive

directed

against

PGA

of

(Fig. 6B). From the

accumulated

with the antibody

holoenzyme

double-diffusion

(Fig. 6C).

analysis

revealed

a

precipitin line with the precursor (not shown). Also, limited proteolysis of the precursor molecule isolated by SDS-PAGE

with staphylococcal

overproduced

94-kDa polypeptide

sor of the PGA. The maximum level of PGA in M5219[pPAKSP,l] cells below, whereas only a low detected in the cells grown at ATCC 11105 showed a similar

V8 protease

segment

the pga progene

from

activity was observed grown at 28°C or level of activity was 42’ C (Fig. 5). E. cofi response to tempera-

ATCClll05

28’C

.

42%

ATCCIIIOS

0

28°C

M5219

[PPAKSP~I~

l

42’C

M5219

[ppAKsp,_ll

expression

E. coli ATCCll105 (M5219) carrying

after

temp.

from pPAKSP,l (triangles)

pPAKSP,I

and

symbols)

was divided

or transferred

quent growth, absorbance in the assays pared

was determined

cells at

arrow)

at 28°C (open During

subse-

at l-h time interby measuring

the

in Table II. The discrepancy

activity of ATCCI 1105, when com-

with Table II, is due to the different

and because growth.

by a downward

were withdrawn

at 415 nm as described of the enzyme

and 42°C.

lysogenic

and either maintained

activity

(h)

at 28°C LI857

to 42°C (solid symbols).

l-ml samples

vals. The enzyme

shift

(circles) were grown in L-broth

28°C to A 600 -- 0 3 At time 0 (marked each culture

is the true precur-

A

Time

pACYCl84

(5.8 kb) plasmid.

with BAL 31 was placed

is the 360-bp HindIII-HpaII moter.

HP

using Hind111 linker. Various

HindIII;

analysis,

precursor Ouchterlony

in the synthesis

was observed

immunoblot

Fig. 5. PGA

of pPAKSP,l

containing

~18 treated

a 94-kDa polypeptide Western

increase

yielded a digestion pattern similar to that of PGA (not shown). These observations confirm that the

(c) Construction of pPAKSP,l

Fig. 4. Structure

(Fig. 5) but a dramatic

the cells were harvested

growth

media used

at the different

stages of

95

A

c

B

c

b

a

b

c

d

e

94

67

43

20 Fig. 6. SDS-PAGE was prepared

of PGA and its precursor. (A) SDS-PAGE

and immunoblotting

as described

by Kutzbach

and Rauenbach

1% SDS and 1 mM 2-mercaptoethanol. (120 x 1.8 cm) equilibrated lane a, purified

The large and small subunits

(B) Overexpression

polyacrylamide induction;

immunoblot

were prepared

nitrocellulose. the position

analysis

2 h induction of crude extracts

as in (B). The cell lysates

by chromatography

from uninduced

M5219[pPAKSP,l]

from the degradation

E. coli M5219[pPAKSP,l]

was grown at 28°C

Then cells were induced

at 42°C for 2 h and harvested

using antiserum

directed

precursor

against PGA holoenzyme

the migration

of the PGA (lane c) or precursor

with respect to the PGA synthesis. This phenomenon may be explained by the fact that posttranslational processing of the precursor is temperature-sensitive. It is possible that the higher temperature affects either the endopeptidase(s) which processes the precursor into subunits or the autoproteolytic processing (Barber0 et al., 1986). Alternatively, it may affect the proper conformation of precursor for processing. ture

(d) Conclusions The complete nucleotide sequence together with the amino acid sequence of two subunit proteins

molecules

and large subunit,

without respectively.

(42°C) E. coli M5219[pPAKSP,l].

polyacrylamide

gel and proteins

of the precursor

The crude

were transferred

to

and ‘251-labeled protein A. Lane c shows

enzyme. Lane d: total lysates ofuninduced

and P denotes

in SDS-lo%

brilliant blue. Lanes: a, M5219[pPAKSP,l]

(28°C) and induced on SDS-lo%

size markers;

specify kDa values

P and L on the right indicate

were separated

PGA

on a Biogel A5m column

Lane m, protein

in sample buffer and boiled for 5 min before electrophoresis

at 42°C.

ofthe large and the small subunit ofpurified of induced

as control.

subunits.

of PGA (23 kDa). The numbers

of I pL promoter.

analysis

The pellets were dissolved

The filter was immunoblotted

total cell lysates been derived

for SDS-PAGE

to Laemmli (1970). Gels were stained with Coomassie

b, M5219[pPAKSP,l].

(C) Western extracts

centrifuge.

gel, according

of PGA (62 kDa); lane c, small subunit

of pga gene under the control

to A 600 -- 0 3 A l-ml sample was removed using an eppendorf

were separated

PGA and individual

against 20 mM Tris . HCl (pH 7.8) containing

with 20 mM Tris . HCI (pH 7.8), 0.1% SDS and 1 mM 2-mercaptoethanol.

PGA; lane b, large subunit

for the markers.

of purified

(1974). The enzyme was dialyzed

M5219[pPAKSP,l]. molecule.

Lane e shows

The minor bands

may have

(lane e).

confirms a common precursor for subunits and the sites for processing as reported by Schumacher et al (1986). With respect to the transcriptional regulation of the pgu gene, we localized the putative CRP-binding sites, promoter and terminator. We also showed that the overproducing strain bearing pPAKSP,l accumulated the precursor at 42°C (Fig. 6B). This implies that processing of the precursor to PGA is temperature-sensitive and a possible limiting step of the enzyme synthesis. How the internal cleavages are produced and which proteolytic enzyme(s) is responsible for the maturation remain to be answered. The strain harboring pPAKSP,l, should be useful

96

for the further investigations of the protein processing in PGA.

Cleveland,

D.W., Fischer,

U.K.:

Peptide

dodecyl

S.G., Kirschner,

mapping

sulfate

by limited

and analysis

M.W. and Laemmli, proteolysis

in sodium

by gel electrophoresis.

J. Biol.

Chem. 252 (1977) 1102-l 106. Daumy,

G.O.,

Danley,

D. and McCall,

A.S.: Role of protein

in Proteus rettgeri penicillin

subunits

G acylase.

J. Bacterial.

163 (1985) 925-932.

ACKNOWLEDGEMENTS

Ebright,

We would like to thank K-S. Yoon (Applied Biosystem) and S-H. Kweon (GINSCO) for analyzing the amino acid sequence, D-H. Lee for computer analysis, and S-S. Kim for technical assistance. We also thank Dr. C-K. Park and Dr. J-H. Roe for helpful comments and M. Lee for careful proofreading.

R.H., Cossart,

Mutations

that

catabolite

P., Gicquel-Sanzey,

alter the DNA

gene activator

B. and Beckwith,

sequence

specificity

protein ofB. coli. Nature

J.:

of the

3 11 (1984)

232-235. Gang,

D.M. and Shaikh,

K.: Regulation

E. coli by cyclic AMP.

Biochim.

of penicillin

Biophys.

acylase

in

Acta 425 (1976)

110-114 Grosjean,

J. and Fiers, W.: Preferential

karyotic

gene.

The

optimal

codon

usuage

codon-anticodon

in pro-

interaction

energy and the selective codon usuage in efficiently expressed gene. Gene 18 (1982) 199-209. Hawley,

D.K. and McClure,

W.R.: Compilation

Escherichiu coli promoter

and analysis

DNA sequences.

of

Nucl. Acids Res.

11 (1983) 2237-2255. ADDENDUM

Kaufmann,

W. and

After our paper was accepted, we became aware of the report by Valle et al. (1986) whose sequence coincides with ours except for their additional 4 nt, one of which is located in the CRP-binding site (see Fig. 2). Although the promoter site as proposed by Valle et al. (1986) is possible, ours was supported by the biological data of galK fusion assay and the high expression of the PGA with Hind111 fragment alone.

Bauer,

K.: The production

by Escherichia coli ATCC9637.

amidase

of penicillin

J. Gen. Microbial.

35

( 1964) 4. Kim, Y-C.: Studies Cloned Doctoral

on the Structure

Penicillin

G Acylase

thesis.

Seoul National

and

Gene

Expression

of the

Escherichiu co&

from

University,

Seoul,

Korea,

1986. Kim, Y-C., Koo, Y-B., Oh, S-J. and Kang, H-S.: Studies on the structure

and expression

Cloning

of

the

ATCClllO5. Klemm,

of the penicillin

penicillin

G

Kor. J. Microbial.

P.: Carboxy-terminal

teins and peptides (Ed.), Methods

gene

gene; I. E. cofi

of

21 (1983) 95-102.

sequence

determination

with carboxypeptidase

in Molecular

G acylase

acylase

of pro-

Y. In Walker,

J.M.

Biology. Humana

Press, Clifton,

E.: Preparation

and general

NJ, 1984, pp. 225-259. Kutzbach,

C. and

properties

REFERENCES

of

Rauenbusch, crystalline

acylase

ATCCI 1105. Z. Physiol. Barbero,

J.L., Buesa, J.M., Buitrago,

Aranda,

A. and Garcia,

of the penicillin

G.G., Mendez,

J.L.: Complete

Laemmli,

E., Perez-

nucleotide

gene from Kluyveru citrophilu. Gene

acylase

H.C. and Doly, J.: A rapid alkaline

dure for screening

recombinant

plasmid

extraction

Matsuda,

Buckel,

P.: The

two

G., Schumacher, subunits

processed

from a common

20(1983)

141-144.

Bomstein, nation

G., Lang, G. and

of penicillin

precursor.

acylase

FEMS Microbial.

are Lett.

J. and Evans, W.G.: Automated of 6-aminopenicillinanic

W., Hoppe,

J., Tsai,

Collins, J. and Mayer,

calorimetric

acid in fermentation

H., Bruning,

H.: Structure

of the penicillin

( 1985) 36-44.

proteolytic

processing.

Harbor,

the

227 (1970)

K.I.: Molecular

for 7/?-cephalosporanic

Pseudomonas strain. J. Bacterial. Maxam,

J.: Molecular

Cloning.

Laboratory,

Cold

NY, 1982, pp. 9.

A. and Komatsu,

of the gene

Mayer,

media. F.,

A.M. and Gilbert,

cloning and structure acid

acylase

from

a

163 (1985) 1222-1228.

W.: Sequencing cleavage.

H., Collins, J. and Wagner,

In Timmis,

end-labeled

Methods

DNA

Enzymol.

65

F.: Cloning of the penicillin

Amsterdam, 3

Mayer,

and

Commercial

Importance.

plasmids. of Medical, Elsevier,

1979, pp. 459-470.

H., Jaakkola,

and Wagner,

on multicopy

K.N. and Ptihler, A. (Eds.), Plasmids

Environmental

acylase

enzyme that under-

J. Mol. Appl. Genet.

during

(1980) 499-560.

determi-

H.J., Maywald,

proteins T4. Nature

Cold Spring Harbor

G acylase gene of,?. co/i ATCClllO5

gene from Escherichia cofi: a periplasmic goes multiple

E.F. and Sambrook,

Manual.

with base specific chemical

Anal. Chem. 37 (1965) 576-578. Bruns,

T., Fritsch,

Spring

DNA. Nucl. Acids

of structural

ofthe head of bacteriophage

A Laboratory

proce-

Res. 7 (1979) 1513-1523. Bock, A., Wirth, G., Schmid,

Cleavage

co/i

Chem. 355 (1974) 45-53.

680-685. Maniatis,

49 (1986) 69-80. Birnboim,

U.K.:

assembly

sequence

Escherichiu

from

P., Schemer,

F.: Enzymatic

U., Senger,

hydrolysis

ofsome

A., Kemmer, penicillins

T. and

97

cephalosporins in

Escherichia cob acylase.

by cloned

Young, M., Robinson, Biotechnology,

In Moo-

C.W. and Vezina, C. (Eds.), Advances Vol.

I,

Pergamon,

London,

1980,

Brady,

K., Shimatake,

polymerase. Nucleic

signals

Santamaria,

Valle,

F.,

G.,

Sizmann,

Amsterdam,

P., Gosset,

R.R.: Transposon-mediated

Cell 25 (1981) 713-719.

of

Gomez-Pedrozo,

Bolivar,

M.,

F.: A common

of the penicillin

acylase

from

methods

acid sequences.

to determine

T. and Gordon,

recombination.

P. and

Bock,

A.:

relation.

the function ofnucleic

J.: Electrophoretic

from polyacrylamide

sheets: procedures

trans-

gels to nitrocellulose

and some applications.

Proc. Natl. Acad.

Sci. USA 76 (1979) 4350-4354. Characterization coli penicillin

G., Tenorio, acylase

structural

119-122. Communicated

B., Oliver,

of the regulatory

Gene 40 (1985) 9-14. site-specific

Buckel,

Nucl. Acids Res. 12 (1984) 521-538.

H., Staehelin,

fer of proteins

Valle, F., Gosset,

1981, pp. 383-415.

F.,

G. and

for the two subunits

Analysis

D.,

acylase from E. cob: unique gene-protein

Nucl. Acids Res. 14 (1986) 5713-5727.

Towbin,

T.S. (Eds.), Gene

Vol. II, Structural

Rossetti,

Escherichia coli ATCCll105. Reed,

U.,

by Escherichia coli RNA

J.C. and Papas,

and Analysis,

Acids. Elsevier,

G.,

D., Schmeissner,

M.: A system to study promoter

recognized

In Chirikjian,

Amplification

precursor

H., Court,

C. and Rosenberg,

and terminator

Oliver,

Penicillin

Staden, R.: Graphic

pp. 83-86. McKenney,

Schumacher,

by A. Nakazawa.

G. and Bolivar,

F.:

region of the Escherichia gene.

Gene

50 (1986)