-
THE JOURNAL 0 1993 by The American Society for Blochemistry
OF BIOLOGICAL CHEM!STRY and Molecular Biology, Inc.
Vol. 268, No. 1, Issue of January 5, pp. 633-639.1993 Printed in
U.S.A.
An Exopolyphosphatase of Escherichia coli THE ENZYME AND ITS ppx
GENE IN A POLYPHOSPHATE OPERON*
(Received for publication, July 6,1992)
Masahiro AkiyamaS, Elliott Crooket, and Arthur Kornberg From the
Department of Biochemistry, Stanford University School of Medicine,
Stanford, California 94305-5307
A gene, ppx, that encodes a novel exopolyphospha- tase of 513
amino acids (58,133 Da) was found down- stream of the gene for
polyphosphate kinase, ppk. Transcription of the ppx gene depends on
the ppk promoters, indicating a polyphosphate (polyp) operon of ppk
and ppx. Exopolyphosphatase, purified to ho- mogeneity from
overproducing cells, is judged to be a dimer of 68-kDa subunits.
Orthophosphate is released processively from the ends of polyp -500
residues long, but chains of -15 residues compete poorly with polyp
as substrate; ATP is not a substrate. Mg2+ (1 mM) and a high
concentration of K+ (175 mM) support op- timal activity.
Inorganic polyphosphate (polyP) includes linear polymers of
orthophosphate with chain lengths up to 1000 or more (1). Polyp is
ubiquitous, having been found in bacteria, yeast, amebas, and
mammals; yet relatively little is known about the enzymes that
metabolize polyp or the physiological func- tions of polyp (2).
Among the functions proposed for polyp are (i) phosphate and energy
reservoirs with obvious osmotic advantage over Pi and ATP, (ii) a
substitute for ATP for certain sugar kinases ( 3 4 , (iii) an
association with poly-@- hydroxybutyrate and Ca2+ in a membrane
domain of trans- formable cells (6), (iv) a pH-stat mechanism to
counterbal- ance alkaline stress (7), and (v) a regulator of
promoter selectivity by RNA polymerase in stationary phase
cells.2
An Escherichia coli enzyme responsible for polyp synthesis is
the homotetrameric polyphosphate kinase (8), which poly- merizes
the terminal phosphate of ATP into polyp in a freely reversible
reaction (nATP e nADP + polyP,) (9, 10). Poly-
* This work was supported in part by National Institutes of
Health Grant GM07581 and National Science Foundation Grant DMB87-
1007945. The costs of publication of this article were defrayed in
part by the payment of page charges. This article must therefore be
hereby marked advertisement in accordance with 18 U.S.C. Section
1734 solely to indicate this fact.
The nucleotide sequencefs) reported in this paper has been
submitted
LO6129. to the GenBankTM/EMBL Data Bank with accession
number(s)
3 Present address: c/o Dr. Bruce Stillman, Cold Spring Harbor
Laboratory, P. 0. Box 100, Bungtown Rd., Cold Spring Harbor, NY
11724.
Fellowship DRG-983. 8 Supported by Damon Runyon-Walter Winchell
Cancer Fund
The abbreviations used are: polyp, long-chain polyphosphate,
ADA, N-(2-acetamido)-2-iminodiacetic acid; CHES, 2-(cyclohexy-
1amino)ethanesulfonic acid; MES, 2-(N-morpholino)ethanesulfonic
acid; MOPS, 3-(N-morpholino)propanesulfonic acid; polyP3, tripoly-
phosphate; polyP,, tetrapolyphosphate; polyP16, polyphosphate mix-
ture with average chain length of 15; PPX, exopolyphosphatase;
PAGE, polyacrylamide gel electrophoresis; Tricine, N-tris-
(hydroxymethy1)methylglycine; kb, kilobase(s). A. Ishihama,
personal communication.
63
phosphate kinase is associated with the membrane fraction in
cell lysates and must be detached for purification; purified
polyphosphate kinase reassociates with the membrane frac- tion
(11). The ppk gene encoding polyphosphate kinase, lo- cated at 53.4
min on the E. coli linkage map (12), has been overexpressed
(11).
In the course of analyzing clones of the ppk gene, we discovered
an adjacent gene (ppx) encoding a novel exopoly- phosphatase (PPX);
the two genes constitute a polyp operon. Here we report the
identification of the ppx gene, its DNA sequence, the expression of
the gene by the ppk promoters, the purification of PPX, and
characteristics of its action.
EXPERIMENTAL PROCEDURES
Reagents and Proteins-Sources were as follows: ATP, S-Sepha-
rose fast flow, Mono Q HR5/5, and Superose-12 HR10/30, Pharmacia
LKB Biotechnology Inc.; [y3P]ATP, Amersham Corp.; DNase I, RNase A,
glucose 1-phosphate, and glucose 6-phosphate, Boehringer Mannheim;
DE52-cellulose, Whatman; bacterial alkaline phospha- tase, U. S.
Biochemical Corp.; ADP, tripolyphosphate (polyp3), tetra-
polyphosphate (polyP,), and a mixture (polyPI& Sigma;
polyethyl- eneimine-cellulose F and cellulose plates, Merck; and
dimethyl adi- pimidate dihydrochloride and dimethyl suberimidate
dihydrochloride, Aldrich.
Bacterial Strains-Bacterial strains used are derivatives of E.
coli K12: W3110 (X-,IN[rrnD-rrnE]I), DH5a (supE44,AlacUI69[~8OlacZ
aM15],hsdRI7,recAl,endAl.gyrA96,thi-l,relAl), and CA36 (A(lac-
proAB),supE,thi,srl-3OO::TnIO,recA56/F[traD36,proAB+,lacIq,
lacZaM151). W3110 served as a host strain for pBClO to overproduce
PPX for purification. DH5a or CA36 (11) was used to determine the
levels of polyphosphate kinase and PPX overproduction induced by
the plasmids. DH5a was also the host strain for preparation of
plasmid DNA.
Construction of Plasmids-Starting from X phage 10H6 of the
Kohara library of the E. coli genome (13), a 5.5-kb BglI fragment
was subcloned into the S m I site of pUC18 in either orientation to
produce pBC9 and pBClO (Fig. 1) (11). Similarly, a 3.8-kb EcoRI
fragment was subcloned into the EcoRI site of pUC18 in either
orientation to produce pBC5 and pBC6; both lack the ppk promoters
(Fig. 1). In pBC6 and pBC10, the reading frame of the ppx gene is
placed in the same orientation as that of the lac gene. Complete
digestion of pBC9 by KpnI endonuclease and self-ligation resulted
in pBC29; digestion
pBC30. of pBClO by KpnI endonuclease, followed by self-ligation,
produced
DNA Sequencing-DNA sequencing of theppx gene was performed by
Dr. Shauna Brummet (U. S. Biochemical Corp. Custom Sequenc- ing
Service) and Alan Smith (PAN facility, Stanford University) as
described (14). The strategy for sequencing the ppx gene is shown
in Fig. 2. For ppx, -97% of the final sequence of 1884 base pairs
was determined from both strands (see Fig. 3). The nucleotide and
de- duced protein sequences were analyzed with the FASTDB program
(IntelliGenetics) and with BLASTP and BLASTN (15).
Assay for Polyphosphate Kinuse-The assay measured the produc-
tion of acid-insoluble [32P]polyP from [T-~P]ATP (8). One unit of
enzyme is defined as the amount incorporating 1 pmol of phosphate
into acid-insoluble polyP/min at 37 C.
Preparation of P2P]PolyP-Synthesis was performed as in the assay
for polyphosphate kinase, except that the reaction volume was
increased 50-fold to 12.5 ml, and [T-~~P]ATP was used at 10
Ci/mol
;3
-
634 Exopolyphosphatase
FIG. 1. Polyphosphate operon and recombinant plasmids. A.
orzaniza-
- "
tion of ppk an> ppx on E.' coli DNA carried in X phage 10H6
(13). The wavy arrow indicates the start and the direc- purN PPk !
PPX ! tion of transcription of the polyp operon. The stippled box
is X phage vector DNA, and the other boxes are E. coli DNA. E,
EcoRI; G, Bg& Q, KpnI. The EcoRI site shown at the junction of
the E. coli DNA insert and the X vector DNA comes from the X vector
DNA. An additional KpnI site is present in the vector. B, DNA
fragments carried in the pUC18 vector. The plasmids were
constructed as de- scribed under "Experimental Proce- dures."
Arrows indicate the direction of transcription of both the lac and
bla pro- moters of pUCl8.
O 0 E
- pBCS pBC10 - pBC5 P B a
t- pBC29
pBC30
""e-+ c- c- "c- *"
+ c - "-+" - .-, +"4 -t H
100 bp FIG. 2. Strategy for sequencing ppx. Nucleotide sequences
in
ppx were determined as indicated by the directions and lengths
of the arrows. The open reading frames shown are ppk (hatched box)
and ppx (filled b o x ) . E, EcoRI; Q, KpnI. bp, base pairs.
TABLE I Polyphosphate kinase and PPX actiuity
of cells harboring various plasmids DH5a and CA36 cells
harboring the indicated plasmids were grown
in LB medium (23) containing ampicillin (50 pg/ml) at 30 "C. At
an absorbance (Amnm) of 0.04, the culture was shifted to 37 "C and
grown to an Amm of -1.0. Cell pellets were collected, resuspended,
and treated as described previously (8). The values include the ac-
tivities of both the soluble and particulate fractions.
0 1 2 3 4 5 6 I I kb
PPW PPX
Host 10' units/g -fold lo' units/g -fold cells ovemroduction
cells ovemroduction
~
DH5a p U C l 8 DH5a pBC6 DH5a pBClO DH5a pBC29 DH5a pBC3O
CA36 pUC18 CA36 pBC5 CA36 D B C ~
39 1.0 14 4.5 0.1 360
2500 64 830 2200 56 14
43 1.1 12
14 52
1100
26 59
1.0
1.0 0.9
1.0 3.7
79 PPK, polyphosphate kinase.
(8). With lo6 units of purified polyphosphate kinase, -50% of
the 32P was incorporated into acid-insoluble polyp after 3 h a t 37
"C. The reaction was stopped by adding EDTA (50 mM) and extracted
with phenol/CHC13 (1:l) once, followed by CHC13/isoamyl alcohol
(24:l) three times. The long-chain polyp was further purified with
slight modifications as described (16). The aqueous phase was mixed
with 2 volumes of 2-butanol and placed at -20 "C for 30 min.
[32P]PolyP was precipitated by centrifugation at 15,000 X g for 15
min (4 "C). The resulting pellet was washed twice with cold 70%
acetone and dried. The dried pellet was dissolved twice in 0.5 ml
of 50 mM EDTA and precipitated with 2 volumes of 2-butanol. The
recovery of acid- insoluble [32P]polyP was 80%, and the purity of
polyp was >99%. The average chain length of this ["P]polyP based
on polyacrylamide gel analysis (17) was -500 residues.
Assay for PPX-The assay for PPX measured the loss of acid-
insoluble [32P]polyP. The reaction mixture (100 pl) contained 50 mM
Tricine/KOH (pH 8.0),1 mM MgCl', 175 mM KCI, 50 nM [32P]polyP (25 p
M in phosphate residues), and PPX as indicated. MgC1, was added
last to avoid precipitation of the polyp. After incubation for 15
min at 37 "C, the reaction was stopped by adding 100 pl of 7% HClO,
and 20 pl of 2 mg/ml bovine serum albumin. The [32P]polyP remaining
was measured in a liquid scintillation counter after collec- tion
on a Whatman GF/C glass-fiber filter and washing with a solution of
1 M HCl and 0.1 M sodium pyrophosphate, followed by ethanol. One
unit of enzyme is defined as the amount that decreases the
acid-insoluble polyp by 1 pmol/min at 37 "C.
Assay for Bacterial Alkaline Phosphatase-Bacteria1 alkaline
phos- phatase was assayed by measuring hydrolysis of [y3'P]ATP. The
reaction mixture (10, 20, or 100 pl) was 1 M Tris-HC1 (pH 8.0), 25
p~ [y3'P]ATP, and bacterial alkaline phosphatase as indicated.
After incubation at 37 "C for 15 min, the reaction was stopped by
adding an equal volume of nonlabeled 10 mM ATP and chilling on ice.
A sample (2 pl) was spotted on a polyethyleneimine-cellulose F
plate, and chromatography was performed with 0.4 M LiCl, 1 M formic
acid. The ATP spot was located by 256 nm UV light and cut out with
scissors; the [T-~'P]ATP that remained was measured in a liquid
scintillation counter. Bacterial alkaline phosphatase specific
activity was measured using p-nitrophenyl phosphate (18). One
bacterial alkaline phosphatase unit is defined as the amount
hydrolyzing 1 pmol of p-nitrophenyl phosphate/min at 25 "C.
Analysis of PPX Hydrolysis Products-The PPX reaction (200 11)
was performed at 37 "C; samples (20 pl) were taken periodically.
The reaction was terminated by adding 5 pl of 450 mM Tris borate
(pH 8.3), 13.5 mM EDTA. The acid-insoluble polyp was detected by
HCIOI precipitation as in the PPX assay. The amount of total polyp
was determined as polyp remaining at the origin on a
polyethyleneimine plate after development with solvent system A
(0.25 M Licl, 1 M formic acid). Orthophosphate was separated by
chromatography on the polyethyleneimine plate with solvent system
A. To determine the amount of long-chain polyp, samples were loaded
on a polyacrylamide gel (6% acrylamide, 0.3% bisacrylamide, 8 M
urea, 89 mM Tris borate (pH 8.3), 2.7 mM EDTA), and electrophoresis
was performed at 750 V with electrophoresis buffer (90 mM Tris
borate (pH 8.3), 2.7 mM EDTA) (17). Short chains were determined on
a cellulose plate developed for 24 h with isobutyric
acid/NH,OH/water (25:3:12) con- taining 0.8 mM EDTA (7). As
references, sodium phosphate, polyP3, and polyp4 were used and,
after chromatography, visualized with an ammonium molybdate spray
(19).
Other Methods-Cross-linking of the protein with dimethyl adi-
pimidate dihydrochloride or dimethyl suberimidate dihydrochloride
(20), SDS-PAGE (21), protein concentrations (22) using bovine serum
albumin as a standard, and DNA manipulations (23) were performed as
described. The amino-terminal sequence of PPX was determined with
an Applied Biosystems 470 Gas-phase Protein Sequencer with on-line
high pressure liquid chromatography by Alan Smith.
RESULTS
Identification of Exopolyphosphutase and Its Gene, ppx-An
extract from cells bearing pBC6, a plasmid that carries a 3.8-
-
Exopolyphosphatase 635
10 20 30 40 50 60 70 80 90
CTTGGTTC~CAGTTTTCACATGGTCATAGCCCGTGTGGTAGATGGTGCCATGCAGATTATTGGCCGCCTG~CAGCGGGTGCATCTG
LeuGlySerAsnSerPheHisMetValIleAlaArgValValAspGlyAlaMetGlnIleIleGlyArgLeuLysGlnArgValHisLeu
GCGGACGGCCTGGGGCCAGATMTATGTTGAGTGMGAGGCMTGACGCGCGGTTT~CTGTCTGTCGCTGTTTGCCGMCGGCTACM
AlaAspGlyLeuGlyProAspAsnMetLeuSerGluGluAlaMecThrArgG1yLeuAsnCysLeuSerLeuPheAlaGluArgLeuG~n
GGGTTTTCTCCTGCCAGCGTCTGTATAGTTGGTACCCATACGCTGCGTCAGGCGCTGMCGCCACTGACTTTCTG~CGCGCGG-G
GlyPheSerProAlaSerValCysIleValGlyThrHisThrLeuArgGlnAlaLeuAsnAlaThrAspPheLeuLysArgA~aGluLys
GTCATTCCCTACCCGATTGATATTTCCGGTMTGMGMGCCCGTCTGATTTTTAT~GCGTGGMCATACCC~CCGG~GGT
ValIleProTyrProIleGluIleIleSerGlyAsnGluGluAlaArgLeuIlePheMetGlyValGluHisThrGlnProGluLysGly
CGCAAACTGGTTATTGATATTGGCGGCGGATCTACGGMCTGGTGATTGGTG~TTTCGMCCTATTCTCGTTG~GCCGCCGGATG
ArgLysLeuValIleAspI1eGlyGlyGlySerThrGluLeuValIleGlyGluAsnPheGluProIleLeuValGluSerArgArgMet
GGTTGTGTCAGCTTTGCCCAGCTTTATTTTCCTGGCGGGGTCATCMT~GAGMTTTTCAGCGCGCTCGCATGGCGGCAGCAC~
GlyCysValSerPheAlaGlnLeuTyrPheProGlyGlyValIleAsnLysGluAsnPheGlnArgAlaArgMetAlaAlaAlaGlnLys
CTGGAAACTTTMCCTGGCMTTCCGTATTCAGGGCTGGMCGTTGCMTGGGCGCTTCCGGTACCAT-GCCGCCCATGMGTGTTA
LeuGluThrLeuThrTrpGlnPheArgIleGlnGlyTrpAsnValAlaMetGlyAlaSerGlyThrIleLysAlaAlaHisGluValL@u
ATGGAAATGGGCGAGAAAGAC~GATMTTACCCCGGMCGTCT~~CTGGT-GMGTTTTACGTCACCGTMTTTCGCATCG
MetGluMetGlyGluLysAspGlyI1eIleThrProGluArgLeuGluLysLeuValLysGluValLeuArgHisArgAsnPheAlaSer
CTGAGTTTACCGGGTCTTTCCGMGAGCGGAAAACAGTCTTCGTTCCGGGACTGGCGATTTTATGCGGTGTGTTTGATGCTTTAGCCATC
LeuSerLeuProGlyLeuSerGluGluArgLysThrValPheValProGlyLeuAlaIleL@uCysGlyValPheAspAlaLeuAlaIle
CGTGMCTGCGCCTTTCTGACGGGGCGTTACGCGMGGCGTACTGTATG~TGG~GGACGTTTCCGTCATCAGGATGTGCGTAGTCGC
ArgGluLeuArgLeuSerAspGlyAlaLeuArgGluGlyValLeuTyrGluMetGluGlyArgPheArgHisGlnAspValArgSerArg
ACCGCCAGCAGCCTCGCCMCCAGTATCACATCGACAGCGMCAGGCCCGACGGGTGCTGGATACCACTATGC~TGTACG~CAGTGG
ThrAlaSerSerLeuAlaAsnGlnTyrHisIleAspSerGluGlnAlaArgArgValLeuAspThrThrMetGlnMetTyrGluGlnTrp
CGGGMCAGCMCCGMGCTGGCGCATCCGCMCTGGAGGCGCTACTGCGATGGGCCGCCATGCTGCATGAGGTCGGGTTGMTATCMC
ArgGluGlnGlnProLysLeuAlaHisProGlnLeuGluAlaLeuLeuArgTrpAlaAlaMecLeuHisGluValGlyLeuAsnIleAsn
CACAGCGGTTTGCATCGCCACTCCGCTTATATTCTGC-CAGTGACTTGCCGGGTTTTMTCAGGMCAGCAGCTGATGATGGCGACA
HisSerGlyLeuHisArgHisSerAlaTyrIleLeuGlnAsnS~rAspLeuProGlyPheA~nGlnGluGlnGlnLeuMetMetAlaThr
CTGGTGCGCTATCACCGTACGATTMGCTCGACGATCTACCGCGCTTTACCTTGTTT~GMG~CAGTTCCTGCCACTGATACAG
LeuValArgTyrHisArgLysAlaIleLysLeuAspAspLeuProArgPheThrLeuPh@LysLysLysGlnPheLeuProLeuIleGln
CTATTGCGCCTTGGCGTATTACTCMCMTCMCGTCAGGCMCCACCA~CCGCCMCATTGACGCTGATTACTGATGACAGTCACTGG
LeuLeuArgLeuGlyValLeuLeuAsnAsnGlnArgGlnAlaThrThrThrProProThrLeuThrLeuIleThrAspAspSerHisTrp
ACACTGCGTTTCCCGCATGACTGGTTTAGTCAGMTGCGCTGGTACTGCTTGATCTGG~GGAGCMGMTACTGGGMGGCGTGGCT
ThrLeuArgPheProHisAspTrpPheSerGlnAsnAlaLeuValLeuLeuAspLeuGluLysGluGlnGluTyrTrpGluGlyValAla
G G C T G G C G G T T G A A A A T T G M G M G ~ G T A C A C C A
G A A A T G T C G G C C C G C A T T A T T
GlyTrpArgLeuLysIleGluGluGluSerThrProGluIleAlaAla
CAGGCACTTTCGCGMTGGGTTCGATTTCATTCAGCGTATCMTTMCGGCTGCGGCTTACC~TMGATMCCCTGCATATMTCGATC
CCC~GAGAGCACCGCCTCGCGGATCTCTTCGTTTTCMCGTACTCTGCCACTACCAGCATTTTCTTCATTCGCGCCAGGTGGC~TC
GATGCCACTATCTGATMTCCAGACTATTTGACACMTATTGCGGATAAAACTGCCGTCMTTTT~GCAGATCCGGGG~TTC
90
180
270
360
4 5 0
5 4 0
630
720
810
900
990
1080
1170
1260
1350
1440
1530
1620
1710
1800
1 8 8 4
FIG. 3. Nucleotide sequence of ppx. The nucleotide sequence of
the noncoding strand of ppx is given from 5 to 3, starting at
nucleotide 40 and terminating at nucleotide 1581. The predicted
amino acid sequence is also shown. The amino-terminal amino acid
sequence determined from the pure PPX (underlined) agrees with the
deduced amino acid sequence. The 3-end of the ppk gene is indicated
by the box .
kb EcoRI fragment but lacks the intact gene for the poly-
phosphate kinase (Fig. l ) , not only showed a 10-fold decrease in
polyphosphate kinase activity compared with cells harbor- ing
vector pUC18 (Table I), but also hydrolyzed [32P]polyP to
orthophosphate more effectively than cells harboring the vec- tor
plasmid. These results suggested that a region downstream of ppk
carries a gene encoding a phosphatase for polyp.
The ppk gene had been located in a 5.5-kb DNA fragment from X
phage 10H6 (11) and cloned in pUC18 as shown in Fig. 1. High levels
of polyphosphatase activity were associated with both pBC6 and
pBC10, which carry the region down- stream of ppk (Fig. 1 and Table
I). Digestion of this region with KpnI endonuclease abolished the
phosphatase activity (as in pBC3O). DNA sequence analyses revealed
that this region contains an open reading frame of 1539 base pairs
(nucleotides 40-1578) that spans the two KpnI sites and encodes a
protein of 513 amino acids with a calculated mass of 58,133 Da
(Fig. 3). This open reading frame (ppx) , which encodes the PPX
(see below), is located 7 base pairs down- stream of ppk and
transcribed in the same direction as ppk.
Polyp Operon Made Up of ppk and ppx-PPX was not overproduced
from pBC6 in lacF cells (CA36), in which the lac repressor is
present at levels higher than in wild-type cells
(data not shown). In view of the overproduction of PPX from pBC6
in lad+ cells (DH5a), there would appear to be a loose repression
of the lac promoter. To study the expression ofppx without the
influence of strong promoters in the vector DNA, ppx was oriented
in the direction opposite to transcription of both the lac and blu
genes (Fig. 1). The resultant plasmids were pBC5 with a 3.8-kb
EcoRI fragment lacking a 0.9-kb BgZI-EcoRI fragment presumably
carrying the ppk promoters and pBC9 with a 5.5-kb BglI fragment
containing the ppk promoters (11). When CA36 cells harboring each
plasmid were grown ( l l ) , overproduction of PPX was 79-fold for
pBC9 and only 3.7-fold for pBC5 (Table I). Thus, althoughppx may
have its own very weak promoter, most of the expression of ppx
depends on the 0.9-kb fragment with the presumed ppk promoters.
Furthermore, insertion of the knn gene into ppk on the chromosome
disrupted not only polyphosphate kinase, but also the PPX
a~tivity.~ These data suggested that ppx constitutes a polyp operon
with the ppk gene. Whereas a Shine-Dalgarno sequence was identified
upstream of ppk, none is apparent for ppx (Fig. 3).
Purification of PPX-Inasmuch as ppx is cotranscribed
E. Crooke, M. Akiyama, and A. Kornberg, unpublished data.
-
636 Exopolyphosphatase TABLE I1
Purification of PPX from overproducing cells Strain W3110/pBC10
was grown in LB medium (23) containing 50 pg/ml ampicillin a t 30
"C. At an absorbance (Amnm) of 0.05, the
temperature was shifted to 37 "C, and growth was continued for 4
h to an Amnm of -1.4. Unless indicated, all manipulations were 0-4
"C. The harvested cells were resuspended in an equal volume of 50
mM Tris-HC1 (pH 7.5), 10% sucrose; frozen in liquid nitrogen; and
stored at -80 "C. Dithiothreitol (to 2 mM) and lysozyme (to 300
pg/ml) were added to the thawed cell suspension (300 g), and the
mixture was immediately transferred to Beckman Ti-45 centrifuge
tubes and incubated at 0 "C for 30 min. The cells were lysed by
exposure to 37 "C for 4 min, followed by immediate chilling in ice
water. The particulate fraction was collected by centrifugation in
a Ti-45 rotor at 20,000 rpm for 1 h (46,500 X gmex). The resulting
pellet (144 g) was dispersed by sonication in the presence of 50 mM
Tris-HC1 (pH 7.51, 10% sucrose, 5 mM MgC12, 10 pg/ml DNase I, and
10 pg/ml RNase A. The temperature during the sonication procedure
was maintained at or below 4 "C. To assure the dissociation of PPX
from the membrane fraction, solid KC1 was dissolved to 1.0 M,
followed by addition of 0.1 volume of 1 M Na2C03. The mixture was
stirred for 30 min at 0 "C. Debris was removed by centrifugation in
a Ti-45 rotor a t 44,000 rpm for 1 h (225,000 X gmaX); the
supernatant was immediately dialyzed (three times for 1 h each,
followed by an 8-h dialysis) against 3-liter volumes of 0.2 M
potassium phosphate (pH 7.0), 10% glycerol, 1 mM dithiothreitol,
and 1 mM EDTA. The dialyzed extract (1100 ml) was stored at -80 "C.
Five percent of the dialyzed extract (45 ml) used for further
purification was dialyzed further against two 500-ml portions (for
2 and 10 h) of Buffer A (50 mM HEPES/KOH (pH 7.5), 10% glycerol, 1
mM dithiothreitol and 1 mM EDTA) (fraction I, 54 ml). Fraction I
adjusted to 0.2 M NaCl was centrifuged in a Sorvall SS-34 rotor a t
6000 rpm for 20 min (43,000 X gmax). The supernatant (55 ml) was
applied to a DE52 column (165 ml, 2.5 X 34 cm) equilibrated in
Buffer A containing 0.2 M NaCI. Active fractions in the
flow-through fraction with a high ratio of were pooled (fraction
11, 83 ml). To Fraction 11, solid (NH4)*S04 was added (0.35 g/ml)
and stirred for 30 min. The suspension was centrifuged for 20 min
at 17,200 X g. The pellet was dissolved in 16 ml of Buffer A, and
the solution was dialyzed against two 500-ml portions (for 2 and 10
h) of Buffer A. The dialyzed solution was centrifuged, and the
supernatant (fraction 111, 17 ml) was applied to an S- Sepharose
fast flow column (30 ml, 2.5 X 7 cm) equilibrated in Buffer A. The
column was washed with 3 column volumes of Buffer A; bound proteins
were eluted with Buffer A containing 0.2 M NaCl. Active fractions
were pooled (fraction IV, 15 ml) and dialyzed against two 750-ml
portions of Buffer A containing 50 mM NaCl (for 2 and 10 h).
Dialyzed fraction IV was centrifuged in an SS-34 rotor a t 18,000
rpm for 20 min (38,700 X g m J , and the supernatant was applied to
a Mono Q HR5/5 column equilibrated in Buffer A containing 50 mM
NaCl. The column was washed with 3 column volumes of Buffer A
containing 50 mM NaCl and eluted with an 8-column volume linear
gradient (0.05- 0.7 M NaCl) in Buffer A. Active fractions eluted at
-0.2 M NaCl were pooled (fraction V, 1 ml). A subsequent
preparation yielded fraction V with a 5-fold higher specific
activity, presumably due to less inactivation of PPX.
Fraction Steu Protein PPX activitv Suecific activitv Yield
Purification A,,,, mg units X units X 10"lmg 75 -fold
I Sonicated lysate 194 95 0.5 100 1.0 0.6
I1 DE52 116 83 0.7 87 1.4 1.0 111 Ammonium sulfate 37 43 1.2 45
2.4 1.4 IV S-Sepharose 5.3 55 10 58 20 V Mono Q 0.9 20 22 21 44
pellet
FRACTION
I I1 111 IV , V , I 20,000-60,000 unlts
kDa " ""
200 -
97 - 68"
29-
18-
FIG. 4. Analysis of PPX by SDS-PAGE. Samples of PPX (2.0 X lo4
units) from fractions I-V (Table 11) were analyzed by SDS- PAGE
(15%); a larger sample of fraction V (6.0 X lo' units) was also
analyzed. Proteins were visualized by Coomassie Blue staining.
with ppk, PPX was overproduced in W3110 cells harboring pBClO
(1.3 X lo7 units/g of cells) as described for polyphos- phate
kinase (11). After heat lysis, -60% of PPX activity was found in
the particulate fraction of the lysate. An early passage through a
DE52 column in the presence of 0.2 M NaCl was needed to remove
nucleic acids and to facilitate subsequent steps. PPX purified
44-fold (Table 11) was a single
1 0 4 0 L ~ 0.1 I 0.15 I 0.2 I 0.25 I
'd
FIG. 5. Gel filtration of PPX. PPX (fraction V, 45 pg) was
applied to a Superose-12 HR10/30 column (Pharmacia) equilibrated in
Buffer A containing 0.2 M NaCl (see legend to Table 11). Elution
was performed with equilibration buffer at a flow rate of 0.4
ml/min. PPX and marker proteins were detected by monitoring UV
absorb- ance (Am nm). cat, catalase (232 kDa); aMo, aldolase (158
kDa); BSA, bovine serum albumin (67 kDa); oval, ovalbumin (43 kDa).
K d is the distribution coefficient.
polypeptide (molecular mass of 57 kDa) on SDS-PAGE (Fig. 4). A
sequence of 10 amino acids from the amino terminus of PPX agreed
with that predicted from the DNA sequence of ppx (nucleotides 43-72
in Fig. 3).
-
Exopolyphosphatase 637
50u 25 0 0 50 100 150 200 SALT (mM)
b: a 100 1 ,100 ,
TEMPERATURE PC) PH
FIG. 6. PPX response to MgC12, salts, temperature, and pH.
Purified PPX (fraction V, 25 fmol as dimer) was added to 100 pl of
the PPX assay (see Experimental Procedures). The conditions for the
assay were varied independently as follows. A, MgC12; B, the
indicated salts; C, temperature for incubation; D, pH. The buffers
(50 mM) used for each pH (D, 0) were MES (pH 5.5 and 6.0), ADA (pH
6.5), MOPS (pH 7.0), HEPES (pH 7.5 and 8.0), Tricine (pH 8.5), and
CHES (pH 9.0). With Tricine, various pH values were also assayed
(D, 0).
A 0
PPX dimer (fmol) BAP dimer (fmol)
FIG. 7. Substrate specificities of PPX and bacterial alkaline
phosphatase. Reaction mixtures were incubated at 37 C for 15 min.
Hydrolysis was detected as in the bacterial alkaline phosphatase
(BAP) assay for [-p3P]ATP and the PPX assay for [32P]polyP (see
Experimental Procedures). A , PPX specificity. Purified PPX (frac-
tion V) and either 25 p~ [y-*P]ATP or 50 nM [32P]polyP as polymer
were present in a reaction mixture of 100 pl. B, bacterial alkaline
phosphatase specificity. Bacterial alkaline phosphatase and 10 p~
[T-~~P]ATP or 10 pM [32P]polyP as polymer were present in a
reaction mlxture of 10 pl.
PPXAppears to Be a Dimer-Comparison of the Superose- 12 elution
volume of PPX with those of reference proteins indicated a
molecular mass of -100 kDa (Fig. 5), suggestive of a dimer. With
the bifunctional reagent dimethyl adipimi- date dihydrochloride or
dimethyl suberimidate dihydrochlo- ride, PPX was cross-linked,
appearing as a 120-kDa polypep- tide on SDS-PAGE; monomeric
ovalbumin remained as a 43- kDa band as expected (data not
shown).
0 0.5 1.0 a Glucose l-phosphab (mM) Glucose &phosphate
(mM)
OO 0.5 1.0 0 0.5 1.0 ATP (mM) ADP (mM)
FIG. 8. Inhibition of PPX and bacterial alkaline phospha- tase
reactions. PPX (fraction V, 160 units, 60 fmol as dimer) hydrolysis
of [32P]polyP and bacterial alkaline phosphatase (BAP) (100 units,
14 fmol as dimer) hydrolysis of [y3P]ATP were assayed as described
under Experimental Procedures in the presence of each bacterial
alkaline phosphatase substrate (unlabeled); the reac- tion volume
of the bacterial alkaline phosphatase assay was 20 pl.
Requirements for PPX-Under optimal pH (-8) and tem- perature (37
C), 1 mM M$ was required for PPX activity (Fig. 6). KC1 at 175 mM
stimulated PPX 21-fold (Fig. 6), but was inhibitory at higher
levels; NaCl was not as effective. Ammonium sulfate a t 50 mM
stimulated 7-fold. PPX was completely inhibited by 20 mM phosphate,
but was unaffected by 1 mM NaF (data not shown).
Substrate Specificity of PPX-The chain length of the [PI polyp
substrate synthesized by polyphosphate kinase is -500 as judged by
polyacrylamide gel analysis. Under optimal con- ditions, 0.2 pmol
(as dimer) of PPX hydrolyzed 5 pmol (as polymer) of polyp (Fig.
7A), yet showed no activity on 2.5 nmol of [y-PIATP (25 p ~ ) .
Assuming the length of [PI polyp to be 500, the K,,, of polyp for
PPX is -9 nM as polymer, a value well below that used in the
assays. Upon raising the polyp level in the bacterial alkaline
phosphatase assay to 10 p~ as polymer, comparable to the K,,, for
ATP (18), polyp hydrolysis by bacterial alkaline phosphatase was
still undetectable ( 4 % ) during an incubation in which 10 p~ ATP
was hydrolyzed completely (Fig. 7B). The hydrolysis of 12.5 nmol of
phosphate residues/pmol of PPX is 50 times that of ATP
molecules/pmol of bacterial alkaline phospha- tase. The capacity of
bacterial alkaline phosphatase substrates to compete with polyp for
PPX was also examined (Fig. 8). Glucose 1-phosphate, glucose
6-phosphate, ATP, and ADP were ineffective inhibitors for PPX (Fig.
8), but served as inhibitors for bacterial alkaline
phosphatase.
Inhibition by Substrate Analogs-In the short-chain polyp series,
showed little inhibition of PPX hydrolysis of
-
638 Exopolyphosphatase
polyp even a t a 2000-fold polymer molar excess over the long-
chain polyp substrate (Fig. 9A); polyp, inhibited 50% at a
2000-fold ratio, and polyP,, showed this level of inhibition when
present a t a 200-fold ratio. Inhibitory effects by these
millimolar levels of short-chain polyphosphate compounds were
observed even when the Mg2' concentration was in- creased to 10 mM
in the PPX assay. These short-chain polyphosphate compounds were
far more effective inhibitors of bacterial alkaline phosphatase
(Fig. 9B). A 50% inhibition
mM
1
I
2 2.5 I I I
1.5 2 2.5 mM
FIG. 9. Inhibition of PPX and bacterial alkaline phospha- tase
reactions by short-chain polyp. Assays were performed as described
under "Experimental Procedures" in the presence of each of the
indicated short-chain polyp forms. Whereas polyP3 and polyPl are
homogeneous compounds, polyPls is a mixture of polyp with an
average chain length of 15. A , PPX action on 50 nM [32P]polyP as
polymer. The PPX assay (100 pl) included 160 units of PPX (fraction
V). B, bacterial alkaline phosphatase action on 25 p~ [y3*P]ATP.
The bacterial alkaline phosphatase assay (20 pl) included 150 units
of bacterial alkaline phosphatase.
of ATP hydrolysis was achieved at an -50-fold polymer ratio for
polyPs, a 20-fold ratio for polyp,, and a 4-fold ratio for
POlYPlS.
PPX Releases Orthophosphate from Ends of Polyp-In the course of
hydrolysis of [R"P]polyP by PPX, Pi release matched the loss of
high molecular weight polyp (Fig. 1OA). Inorganic pyrophosphate,
polyP, and polyp, were not detected among the products (data not
shown). Inasmuch as no perceptible decrease in polymer size
occurred during the course of its extensive conversion to Pi (Fig.
10B), the hydrolysis from the chain ends appears to be highly
processive. A chain in the size range of 10-20 residues, as an end
product of hydrolysis of each polyPsoo chain, could have been
missed because of heterogeneity and relatively low abundance. The
level of phosphate residues in each member of a short-chain poly-
phosphate series may be
-
Exopolyphosphatase 639
is strongly stimulated by potassium salts and is not inhibited
by sodium fluoride. Direct comparisons of PPX with bacterial
alkaline phosphatase reveal their striking differences (Figs.
7-91,
Whether PPX of E. coli is closely related to the polyphos-
phatase identified in other microorganisms will require fur- ther
characterization of these other enzymes. Exopolyphos- phatases have
been obtained from Corynebacterium xerosis (25), Saccharomyces
cereuisiue (l), Aerobacter aerogenes (261, and Neurospora crassa
(1). The enzymes from A. aerogenes and N. crassa also require Mg2+
and a high concentration of K', characteristics also similar to
those of the E. coli PPX. Whether these factors influence the
structure of the polyp substrate or the enzyme is not known.
The action of PPX is directed to the ends of the polyp chain,
from which it removes orthophosphate processively (Fig. 10). During
the course of the reaction, the size of the remaining large-chain
polyp is undiminished, and only ortho- phosphate is released.
Inasmuch as polyp chains of -15 residues are weak competitors with
the long-chain polyp substrate and would presumably be poorly
attacked, it would seem that they should also accumulate as an end
product. However, with only one such small chain produced per long-
chain substrate and the likely heterogeneity of these small- chain
products, the failure to observe any discrete bands in the small
size range by electrophoretic gel analysis would be expected. The
mechanism that enables PPX to distinguish the ends of a long chain
from those of a short one is obscure and intriguing. As for the
cellular salvage and disposal of short chains, they may be used as
primers for polyphosphate kinase (8) or degraded by numerous other
phosphatases, such as the tripolyphosphatases of A. aerogenes (26)
and N. crmsa (1).
The membranous attachments observed for polyphosphate kinase
(11) and possibly for PPX suggest that their actions in making and
disposing polyp are membrane-oriented and should direct our
attention to where polyp is found in the cell and the clues that
such localization may offer to its physio- logical functions. Polyp
granules are prominent in many
microorganisms and, under some circumstances, fill the vac-
uoles of yeast cells (2). Although such polyp deposits are not
prominent in E. coli, they have nevertheless been seen as
electron-dense spheres, enveloped in some way and often attached to
a DNA fiber.6 These observations may be related to the proposal
that membrane domains of polyhydroxybu- tyrate, calcium, and polyp
may function as channels for DNA entry into transformable
microorganisms (6).
Acknowledgment-We are grateful to Tod Klingler for computer
analysis of the DNA sequence.
REFERENCES 1. Kulaev, I. S. (1979) The Biochemistry of Inorganic
Polyphosphates, John
Wile & Sons New York 2. Wood, h. G., a i d Clark, J. E.
(1988) Annu. Reu. Biochem. 67,235-260 3. Szvmona. M.. Kowalska. H..
and Pastuszak. I. (1977) Acta Bmchim. Pol.
, I , . , 24,133-142
4. Dirheimer, G., and Ebel, J. P. (1968) Bull. SOC. Chim. Biol.
60,1933-1947 5. Pepin C. A,, and Wood, H. G. (1987) J. Biol. Chem.
262,5223-5226 6. Reusc'h, R. N., and Sadoff, H. L. (1988) Proc.
Nutl. Aead. Sei. LI. S. A. 86,
AI W-AI Rn . - - - - - 7. Pick, U., Bental, M., Chitlaru, E.,
and Weiss, M. (1990) FEBS Lett. 274,
8. Ahn, K., and Kornberg, A. (1990) J. Biol. Chem.
265,11734-11739 9. Kornberg A., Kornberg S. R., and Simms E. S.
(1956) Biochim. Biophys.
15-18
Acta 20,215-227 10. Kornberg, S. R. (1957) Biochim. Biophys.
Acta 26,294-300 11. Akiyama, M., Crooke, E., and Kornberg, A.
(1992) J. Biol. Chem. 267,
12. Bachmann, 8. J. (1990) Microbiol. Reu. 64,130-197 13. Kohara
Y., Akiyama, K., and Isono, K. (1987) Cell 50,495-508 14. Henikoif,
S. (1984) Gene (Amst.) 28,351-359 15. Altschul S. F., Gish, W.,
Miller, W., Myers, E. W., and Lipman, D. J.
16. Robinson, N. A., Clark, J. E., and Wood, H. G. (1987) J.
Biol. Chem. 262,
17. Clark, J. E., and Wood, H. G. (1987) Anal. Biochem.
161,280-290 18. Garen A. and Levinthal C. (1960) Biochim. Biophys.
Acta 38,470-483 19. Stah1,'E. (ed) (1969) Thih-layer
Chromatography: A Laboratory Handbook,
20. Tinberg H. h. and Packer L. (1979) Methods Enz mol
56,622-629 21. Ito K., bate T'., and WickAer W. (1980) J. Biol.
C&m. '266,2123-2130 22. BrAdford, MT M. (1976) Anal. Biochem.
72,248-254 23. Sambrook J Fritsch, E. F. and Maniatis, T. (1989)
Molecular Clonin A
LaboratAj'Manuul, 2nd dd., Cold Spring Harbor Laboratory Press,
{old Spring Harbor, New York
24. Torriani-Gorini, A,, Rothman, F. G., Silver, S., Wright,
A,,, and YaFil, E.
gunisms, American Society for Microbiology, Washington, D. C.
(eds) (1987) Phosphate Metabolum and Cellular Regulutwn tn
Mccroor-
25. Muhammed, A., Rodgers, A., and Hughes D. E. (1959) J. Gen.
Microbiol.
26. Harold F. M., and Harold R. L. (1965) J. Bacteriol.
89,1262-1270 20,482-495
22556-22561
(199ojJ. MOL B ~ O L 215,403-410
5216-5222
2nd Ed., S ringer-Verlag, New York
J. Griffith, personal communication.