Get documents about "Cyclohexadienyl Dehydratase from Pseudomonas aeruginosa
Document Sample
T H EJ O l i R N A L OF BIOLOGICAL
CHEMISTRY Vol. 267, No. 4, Issue of February 5 , pp. 2487-2493, 1992
(c’ 1992 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.
Pseudomonas aeruginosa
Cyclohexadienyl Dehydratase from
MOLECULARCLONINGOF THE GENE AND CHARACTERIZATION OF THEGENEPRODUCT*
(Received forpublication, August 22, 1991)
Genshi Zhao, Tianhui Xia, Randy S . Fischer, and Roy A. JensenS
From the Departmentof Microbiology and Cell Science, Universityof Florida, Gainesville,Florida 3261 1
The gene encoding cyclohexadienyl dehydratase (de- tioning in nature, an alternative route to L-tyrosine biosyn-
notedphec) was cloned from Pseudomonas aeruginosa thesis was discovered in cyanobacteria in 1974, whereby pre-
by functional complementationof a pheA auxotroph of phenate was transaminated to L-arogenate via prephenate
Escherichia coli. The gene was highly expressed inE.
coli due to the use of the high-copy number vector aminotransferase (2). L-Arogenate, a substrate for arogenate
pUC 1 .The P . aeruginosa cyclohexadienyl dehydra-
8 dehydrogenase in cyanobacteria, was initially identified as a
tase expressed in coli was purified to electrophoretic
E. precursor of L-tyrosine, but was later recognized as a precursor
homogeneity. The latter enzyme exhibited identical of L-phenylalanine in many microorganisms and in higher
physical and biochemical properties as those obtained plants (3). Phenylalanine biosynthesis in microorganisms ex-
for cyclohexadienyl dehydratase purified from P . hibits multiple aspects of diversity.Exclusive use of the
aeruginosa. The activity ratios prephenate dehydra-
of
tase to arogenate dehydratase remained constant phenylpyruvate route (e.g. Bacillus)or exclusive use of the L-
(about 3.3-fold) throughout purification, thus demon- arogenate route(e.g. Pseudomonas diminuta) may be in place.
strating a single protein having broad substratespec- On the other hand, both pathway routesmay coexist due to
ificity. The cyclohexadienyl dehydratase exhibited K , (i) the substrateambiguity of cyclohexadienyl dehydratase or
M M
values of 0.42 m for prephenate and 0.22 m for L- (ii) the presence of one of the possible pairs made up of
arogenate, respectively. The pheC gene was 807 base prephenate dehydratase, arogenate dehydratase, cyclohex- or
pairs in length, encoding a protein with a calculated
molecular massof 30,480 daltons. This compares with adienyl dehydratase. Another aspect of diversity is the pres-
a molecular mass value 29.5 kDa determined for the ence orabsence of a physicalorganization of prephenate
of
of
purified enzyme by sodium dodecyl sulfate-polyacryl- dehydratase as one catalytic domain a bifunctional protein
amide gel electrophoresis. Since the native molecular (denotedtheP-protein), which alsopossesses a catalytic
mass determined by gel filtration was 72 kDa, the domain for chorismate mutase. Two of the threemajor rRNA
enzyme probably is a homodimer. Comparison of the Gram-negative superfamilies (Superfamily A Superfamily and
deduced amino acid sequenceof pheC from P . aerugi-
B) possess the bifunctional P-protein. Most enteric bacteria
nosa with those of the prephenate dehydratases of
Corynebacteriumglutamicum,Bacillussubtilis,E. (but not Escherichia coli) and most members of the entire
coli, and Pseudomonas stutzeri by standard pairwise Superfamily B assemblage possess cyclohexadienyl dehydra-
alignments did not establish obvious homology. How- tase in addition to the P-protein 4). (3,
ever, a more detailed analysis revealed a conserved Pseudomonas aeruginosa, a Superfamily-B organism, was
motif (containing a threonineresidueknowntobe the first example of a microorganism possessing dual path-
essential for catalysis) that was shared by all of the
dehydratase proteins. ways to L-phenylalanine ( 5 ) . This is illustrated in Fig. 1. In
addition to thebifunctional P-protein, P. aeruginosa possesses
chorismate mutase-F and cyclohexadienyl dehydratase. Be-
cause the latter two enzymes are unrestrained by allosteric
Prephenate, a cyclohexadienyl-ring molecule formed from control,they havebeenreferred to as components of an
chorismate by chorismate mutase, is a precursor which is overflow pathway (6). The cyclohexadienyl dehydratase of P.
uniquely used for the biosynthesis of L-phenylalanine and L- aeruginosa has a broad substrate specificity that accommo-
tyrosine inmicroorganisms (1). The conversion of prephenate dates both prephenate and L-arogenate as substrates for L-
to phenylpyruvate (via prephenate dehydratase) or to 4-hy- phenylalanine biosynthesis ( 5 ) . The overflow pathway to L-
droxyphenylpyruvate (via prephenate dehydrogenase) was phenylalanine biosynthesis is widely distributed among Su-
first established in entericbacteria. Although the latter were
perfamily-B microorganisms (7).
assumed for many years to be universal enzyme steps func-
The physiological role of the overflow pathway in nature is
* Florida Agricultural Experiment Station Journal Series No. R- essentially unknown. The evolutionary relationship of cycloh-
01936. The costs of publication of this article were defrayed in part exadienyl dehydratase to monofunctional prephenate dehy-
by the payment of page charges. This article must therefore be hereby
marked “advertisement” in accordance with 18 U.S.C. Section 1734 dratase, to monofunctional arogenate dehydratase, or to the
solely to indicate this fact. dehydratase domain of the bifunctional P-protein remains to
in
The nucleotide sequence(s) reported thispaperhas been submitted be elucidated. In this paper,we report the molecular cloning,
the
to GenBankTM/EMBL Bank Data with accession number(s) expression and nucleotide sequence of the P. ueruginosa cy-
M74132.
$ T o whom correspondence and reprint requests should be ad- clohexadienyl dehydratase gene, as well as the purification
dressed. and characterization of its gene product.
2487
2488 Prephenate DehydrataselArogenate Dehydratase Gene in P. aeruginosa
TABLE
I1
Expression of the P. uerugimsu pheC gene in E. coli
Specific activity is defined as nanomoles ofphenylpyruvate or
phenylalanine formed per min/mg of protein. All the clones or sub-
clones listed were first transformed into E. coli JP2255, and the crude
extracts of the transformed JP2255 were used for enzyme assay.
Specific activity
Plasmid in strain
JP2255 Prephenate Arogenate
dehydratase dehydratase
pJZl 621 191
211
pJZ2 700
twc b pJZla 0 0
FIG. 1. Dual biosynthetic routes to L-phenylalanine in P. pJZlb 0 0
ueruginosu. The bifunctional P-protein (denoted by shading) con- pJZlc 0 0
sists of chorismate mutase ([lu]) and prephenate dehydratase ( [ I b ] ) PJZld-S 618 186
domains. The broad specificity cyclohexadienyl dehydratase catalyzes pJZld-0 5.6 1.7
the two reactions indicated by shading: prephenate dehydratase or pJZle 0 0
arogenate dehydratase.Enzyme 121 is the monofunctional chorismate pJZlf 0 0
mutase-F. Reactions [3] and 141 refer to a multiplicity of aminotrans- 387
PJZk 1314
ferase enzymes with overlapping substrate specificities (26) which pUC18/pUC19 0 0
transaminate phenylpyruvate or L-arogenate, respectively. Abbrevi-
ations: CHA, chorismate; PPA, prephenate; PPY, phenylpyruvate;
PHE, L-phenylalanine;AGN, L-arogenate; PLP, pyridoxal 5”phOS- had been digested with SphI and SmaI (data shown). not
phate. Localization of the P. ueruginosa pheC Gene and Expression
of the Gene in E. coli-Cleavage of pJZ1 with EcoRI, KpnI,
and SmaI yielded subclones denoted pJZla, pJZlb, and pJZlc,
MATERIALS AND METHODS’ respectively (Fig. 2 A ) . These plasmids were unable to com-
plement E. coli JP2255 and E. coli KA197, and cyclohexad-
RESULTS ienyl dehydratase activity was not detected in crude extracts
Cloning of the. Gene Encoding Cyclohexudienyl Dehydra- of transformants carrying the subclones (Table 11), indicating
tme-Approximately 4000 recombinants were obtained after the probable location of the gene within the KpnI-KpnI frag-
transformation of E. coli JM83 (see “Materials and Meth- ment. The KpnI-KpnI fragment (Fig. 2 A ) was cloned into
ods”). Purified plasmids from these recombinants were used pUC18 in both possible orientations. When the KpnI-KpnI
to transform E. coli JP2255. The transformants obtained were fragment was cloned in the same orientation as the original
allowed to grow in LB medium overnight at 37 “C, washed same
clone (designated as pJZld-S), the high level of activities
twice with M9 medium, and then plated on M9 plates which conferred by pJZl was found (Table 11). However, when this
were supplemented with L-tyrosine, thiamine, and ampicillin. fragment was cloned in the opposite orientation (designated
After incubation for 4 days at 37 “C, two colonies were ob- as pJZld-0), the activities observed were two orders of mag-
served. Plasmids were purified from cultures derived from nitude lower (Table 11). These results showed that the over-
each of these transformants. Each the two plasmids isolated
of expression of the enzyme activity of these clones was largely
(designated as pJZl and pJZ2) was found to be capable of dependent onthe lucZ promoter of the plasmid. The dataalso
transforming E. coli JP2255 to phenylalanine independence. indicate that the promoter of the cloned gene was able to
The transformants of E. coli JP2255 carrying the clones of function in E. coli, although not efficiently. Further localiza-
pJZl and pJZ2 were slow growers on M9 medium. However, tion of the gene was carried out by cloning the two S m I
the transformation of pJZl or pJZ2 into E. coli KA197 pro- fragments (released upon digestion of pJZld-S with SmaI)
duced faster growing transformants, apparently due to the into pUC18 at the SmaI site. The two resulting subclones,
presence of chorismate mutase encoded by tyrA. designated as pJZle and pJZlf (Fig. 2 A ) , were unable to
The presence of prephenatedehydratase and arogenate complement the pheA defects of E. coli strains JP2255 and
dehydratase activities was examined in crude extracts of E. KA197, suggesting that the pheC gene was localized within
coli JP2255 carrying plasmids pJZl or pJZ2. A high level of the two SmaI fragments. In order to obtain a SphI-SmaI-
both activities was evident in crude extracts of E. coli JP2255 SmaI fragment, the original clone was completely digested
carrying the pJZl and pJZ2 plasmids, whereas no enzyme with SphI and thenpartially digested with SmaI. The SphI-
activity was detected in a crude extract of E. coli JP2255 SmaI-SmaI fragment was isolated and cloned into pUC19,
carrying pUC18 (Table 11). yielding a subclone denoted pJZlg. Subclone pJZlg comple-
Digestion of pJZl and pJZ2 with HindIII, PstI, and KpnI mented E. coli strains JP2255 andKA197 and exhibited a 2-
showed that they carried two identical DNA fragments esti- fold increase in enzyme activities when compared with pJZld-
mated to be 5.7 kilobase pairs in length. The plasmid desig- S (Table 11).The increased enzyme activity was probably due
nated as pJZl was used for further study. to the decrease of the distance between lucZ promoter and the
Southern blot hybridization showed that when the SphI- transcriptional start site of the gene.
EcoRI fragment of pJZ1 was biotinylated, it hybridized with DNA Sequence of the P. aeruginosa pheC Gene and Its
a 3.5-kilobase pair fragment of P. ueruginosa chromosomal Flunking Regions-The complete nucleotide sequence of the
DNA obtained by complete digestion with SphI and EcoRI, 1259-base pair SmaI fragment is presented in Fig.3. The
but itdid not hybridize with E. coli chromosomal DNA which structural gene encoding cyclohexadienyl dehydratase was
located within a single open reading frame (807 base pairs in
Portions of this paper (including “Materials and Methods,” length). The deduced amino acid sequence presented in Fig.
Tables I and 111, and Figs. 2, 4, and 5) are presented in miniprint at
the end of this paper. Miniprint is easily read with the aid of a 3 indicates that the P. ueruginosa cyclohexadienyl dehydra-
standard magnifying glass. Full size photocopies are included in the tase contained268 residues with a molecular weight of 30,480.
microfilm edition of the Journal that is available from WaverlyPress. This compares with a value of 29,500 (Fig. 4) determined for
Prephenate
DehydrataselArogenate in
Dehydratase
Gene P. aeruginosa 2489
10 30
GGATCAGCTTCCCGGCCTACCAGGAGCACGGCCTGGAGATGCTGCTGCGCTACCACCCGG
50 with that of arogenate dehydratase remained constant
G S AS R P T R S T A W R C C C A T T R throughout the process of purification. Only one major band
70 110
MTGGCTGCAGGGGGTACCGCTGTCGATGGCGGTGTGAGGTCGTCAGCCGTTTCGCGCAC was resolved by SDS-PAGE after the Sephadex G-200 column
N G C R G Y R C R W R C E V V S R F A H (Fig. 4). The subunit molecular weight of the cloned cycloh-
170 130 150
TTTTTTCCGCTTCTCCTGCCGCATGCTCGGCCCGCGCCCCGGCGTCATCGGGCGTTCCCC exadienyl dehydratase was 29,500 as determined by SDS-
F F P L L L P H A R P A P R R H R A F P
230 190 210 PAGE, and the molecular weight of the native enzyme was
TGCATTCCGGGATTTGGCCGCGGCTGCCGACTTGCGTAGTCTCTCTGCGGTCCGCCATCC
C I P G F G R G C R L A '
72,000 as determined by gel filtration on Sephadex G-200.
250 290 Purification of the cyclohexadienyl dehydratase from P.
CGAGGAGTCGCCATGCCGMGTCATTCCGCCATCTCGTCCAGGCCCTGGCCTGCCTTGCG
M P K S F R H L V Q A L A C L A aeruginosa was essentially carried out under identical condi-
310 tions as those used for its isolation from E. coli. Both of the
CTGCTGGCCAGCGCCAGCCTCCAGGCGCAGGAGAGCCGCCTCGACCGCATCCTCG~GC
L L A S A S L Q A Q E S R L D R I L L S cylohexadienyl dehydratasepreparations failed to bind to
410 370 390 DEAE-cellulose at pH values lower than 7.4, and they were
GGCGTGCTGCGCGTCACCACCACTGGCGACTACMGCCCTTCAGCTACCGCACGGMGAG
G V L R V T T T G D Y K P F S Y R T E E found to migrate into equivalent fractions throughout the
430 470
GGCGGTTACGCCGGTTTCGACGTGGACATGGCGCAGCGCCTGGCCGAGAGCCTGGGGGCC purification process. A specific activity ratio of prephenate
G G Y A G F D V D M A Q R L A E S L G A
490 dehydratase to that of arogenate dehydratase of 3:l was
MGCTGGTAGTGGTGCCGACCAGTTGGCCGMCCTGATGCGCGATTTCGCCGACGACCGC maintained throughout the purification process, as found for
K L V V V P T S W P N L M R D F A D D R
590 550 570 the P. aeruginosa enzyme isolated from E. coli (data not
TTCGACATCGCCATGAGCGGCATCTCGATCMCCTGGAGCGCCAGCGCCAGGCGCATTTC
F D I A M S G I S I N L E R Q R Q A H F shown). The native molecular weight of this enzyme was
610 650 found to be 72,000, a value identical to that obtained for the
TCGATTCCCTACCTGCGCMCAGCMGACGCCGATCACCCTCTGTAGCGMGMGCGCGT
S I P Y L R N S K T P I T L C S E E A R cloned gene product. After the final step of purification, the
710 670 690 enzyme preparation obtained was not electrophoretically ho-
TTCCAGACCCTGGAGCAGATCGACCAGCCGGGCGTGACGGCCATCGTCMCCCCGGCGGC
F Q T L E Q I D Q P G V T A I V N P G G mogeneous (data notshown).
730 S770
I. I 750
ACCMCGAGMGTTCG~CGMCCTGMGMGGCCCGGATCCTGGTGCATCCGGAC K , values of 0.42 m for prephenate and of 0.22 m for L-
M M
T N E K F A R A N L K K A R I L V H P D arogenate were obtained for the enzyme produced from the
790 810 830
MCGTGACGATCTTCCAGCAGATCGTCGACGGCMGGCCGACCTGATGATGACCGACGCC cloned gene (Fig. 5). The corresponding values obtained for
N V T I F Q Q I V D G K A D L M M T D A
850 870 890 cyclohexadienyl dehydratase isolated directly from P. aeru-
ATCGAGGCCCGCCTGCAGTCGCGTCTGCACCCGGMCTCTGCGCCGTGCATCCGCAGCM
I E A R L Q S R L H P E L C A V H P Q Q
ginosa were 0.40 and 0.19 mM, respectively. vmax values of
950 910 930 307.7 pmol/min/mg for prephenate and 102.8 pmol/min/mg
CCCTTCGACTTCGCCGAGMGGCCTACCTGCTGCCGCGCGACGAGGCCTTCMGCGCTAC
for L-arogenate were obtained for the enzyme produced from
P F D F A E K A Y L L P R D E A F K R Y
ain aan __" 1nln
*"*"
GTCGACCAGTGGCTGCACATCGCCGAGCAGAGCGGCTTGTTGCGCCAGCGCATGGAG~A~
the cloned gene. Since the preparation of the cyclohexadienyl
V D Q W L H I A E Q S G L L R Q R M E H dehydratase isolated from P. aeruginosa was not homogene-
1030
TGGCTCGMTACCGCTGGCCCACCGCGCACGGCAAGTMTACAGGGGCGGCGAGGGTGGC ous, the V,, values were not determined. Prephenate dehy-
W L E Y R W P T A H G K *
1090
dratase activity of both preparationswas competitively inhib-
CGCGGGCCCGCGCGGCCTTCCTTGGCGGCGGC~CGTTATGGTCGGCGCCCCATCCT ited by L-arogenate with a K, value of 0.2 mM, whereas
1150
GGTGCCTGGTCCATGCGTTATCTACTGTTCGTCACCGTCCTCTGGGCGTTCTCCTTCMC arogenate dehydratase activity was competitively inhibited by
1210
CTGATCGGCGAGTACCTCGCCGGCCAGGTCGGCAGCTACTTCGCCGTGCTTACCCGGGG
prephenate with a K, value of 0.40 mM. The P. aeruginosa
cyclohexadienyl dehydratase was not subject to allosteric con-
FIG. 3. Nucleotide sequence of the P.aeruginosapheC gene
and of its lanking regions. The deduced amino acid sequence of trol by phenylalanine, tyrosine, and tryptophan when present
the gene along with its upstream flanking region is shown beneath singly or in combination.
the corresponding codons. The SmaI site of cistronic inactivation
midway through thepheC gene is also shown. DISCUSSION
The Identity of the Cloned Gene and Its Product-Cyclo-
the purified enzyme by SDS-PAGEZ(19). The sequence GAG- hexadienyl dehydratase was first described in P. aeruginosa
GAG, located 6 base pairs upstream of the start codon is (5), and theanalysis has now been extended to themolecular-
presumbly the ribosome binding site (22). The open reading genetic level. The successful cloning of P. aeruginosa pheC in
frame was terminated by a TAA codon. E. coli, in concert with the purification of a gene product
+
The G C content of the cyclohexadienyl dehydratase gene having all of the properties of partially purified cyclohexad-
was 65.6% which falls within the 60.6-66.3% range for P. ienyl dehydratase isolated directly from P. aeruginosa, proves
aeruginosa genomic genes (23). The codon usage of the gene that a single protein possesses both prephenate dehydratase
was typical of P. aeruginosa (23), exhibiting a striking pref- and arogenate dehydratase activities as a consequence of
erence for G or C in the third base position in 91.1% of the ambiguity for substrate recognition. This is consistent with
codons. As is thecase for most P. aeruginosa genes, C (52.4%)
kinetic results showing that prephenate competitively inhib-
was utilized more frequently than G (38.7%) in the third ited arogenate dehydratase activity, whereas L-arogenate com-
position. petitively inhibitedprephenate dehydratase activity. Re-
A portion of an unidentified open reading frame encoding
cently, similar biochemical results have been obtained for the
a truncated peptide of 74 residues (Fig. 3) was found upstream
cyclohexadienyl dehydratase purified from Erwinia herbicola
of pheC. Asearch of GenBank sequences did not reveal
obvious homology with any known gene sequences. (24).
The in Vivo Function of Cyclohexadienyl Dehydratase-The
Characterization of the P. aeruginosa pheC Gene Product
Purified from E. coli and Comparison with the Cyclohexadienyl reluctant auxotrophy of P. aeruginosa for phenylalanine fol-
Dehydratase Isolated Directly from P. aeruginosa-Purifica- lowing otherwise successful mutagenesis protocols was ex-
tion of the cyclohexadienyl dehydratase isolated from E. coli plained as the consequence of independent dual pathways to
JP2255(pJZlg) is summarized in Table 111. The ratio of 3:l phenylalanine (25). A mutant lacking the bifunctional P-
obtained for the activity of prephenate dehydratase compared protein has been identified (26).The mutant exhibited a leaky
requirement for phenylalanine, thus indicating that exclusive
The abbreviation used is: SDS-PAGE, sodium dodecyl sulfate- dependence upon cyclohexadienyl dehydratase for biosyn-
polyacrylamide gel electrophoresis. thesis of phenylalanine is rate-limiting togrowth. In contrast,
2490 DehydrataselArogenate
Prephenate Gene
Dehydratase in P. aeruginosa
a regulatory mutant of P. aeruginosa possessing a tyrosine- mutasecomponent of theT-proteinhasthepotentialto
insensitive DAHP synthase was found to excrete phenylala- participate in phenylalanine biosynthesis, even though the
nine, presumably through the unregulatedoverflow pathway activities of the two domains are tightlycoupled (31-33).
(6). Thus, the capacity for generation of phenylalanine via EvolutionaryImplications-The deduced amino acid se-
the overflow pathway is dramatically influenced by precursor quence of the P. aeruginosa pheC gene product was pairwise
levels i n vivo. aligned with those of the Corynebacterium glutamicum and
L-Arogenate is generatedby transamination of prephenate. Bacillus subtilis prephenate dehydratases (34, 35), and with
Five aromatic aminotransferases capable of transamination E. coli and P. stutzeri P-proteins (36, 37). The P. aeruginosa
of prephenate have been isolated fromP. aeruginosa (27) and enzyme was found tobe only marginally similar to thesefour
shown to have a relatively poor affinity for prephenate com- proteins, ranging in identity from 16.5 to 18.9%. In contrast,
pared with the prephenate dehydratase component of the the monofunctional prephenate dehydratases of C. glutami-
bifunctional P-protein. The most likely source of prephenate cum and B. subtilis have shown significantidentity to the two
molecules for transamination is via the catalytic action of a bifunctional P-proteins of E. coli and P. stutzeri (34, 35, 37),
monofunctional chorismate mutase denoted chorismate mu- indicating that the prephenate dehydratases probably share a
tase-F. However, it has a poor affinity for chorismate com- common evolutionary origin. The marginal similarity of the
pared with the chorismate mutase componentof the bifunc- P. aeruginosa cyclohexadienyl dehydratase to the prephenate
tional P-protein." Since the cyclohexadienyl dehydratase has dehydratasesas well as the bifunctional P-proteins might
a relatively poor affinity for both of its substrates compared suggest that the cyclohexadienyl dehydratase and the prephe-
withthe competing P-proteinprephenatedehydratase,it nate dehydratases evolved independently. However, a more
would seem that most phenylalanine ordinarily synthesized detailed analysis, focusing on shorthighly conserved sequence
is
via the bifunctional P-protein and that most phenylalanine segments (previously established in Ref. 37), rather than the
molecules are normally derived from phenylpyruvate rather total peptide, revealed a conserved motif which includes the
than from L-arogenate. Thus, under ordinary growth condi- essentialthreonine residue demonstrated by Hudsonand
tions, the cyclohexadienyl dehydratase along with the mono- Davidson (36) and a number of flanking residues. This motif,
functional chorismate mutase probably does not contribute shown in Fig. 6, suggests residues (within boxes) that may
significantly to phenylalaninebiosynthesis. prove to be commonto all of the dehydratases. is interesting It
Perhaps the overflow pathway (the monofunctional chor- that all of the prephenate dehydratases share the TRF se-
ismate mutase and the cyclohexadienyl dehydratase) exists as quence, whereas the cyclohexadienyl dehydratase sequence is
a backup system for phenylalanine biosynthesis, possibly in TIF. It remains to be seen whether other cyclohexadienyl
relationship to differential carbon input into aromatic biosyn- dehydratases will also possess TIF sequences. Note that in
thesis during growth on different carbon sources. It is inter- this alignment therewas considerable conservation of amino
esting that P. stutzeri, a very close relative of P. aeruginosa, acid sequence betweenthe peptidesof the P. aeruginosa CDT
lacks the overflow pathway altogether (28). The loss of the and the P. stutzeri P-protein.Theseorganismsare more
prephenate dehydratase activity the bifunctional P-protein closely related to one another than to any the other orga-
of of
of P. stutzeri has yielded a tightly blocked phenylalanine nisms examined.
auxotroph (29), in contrast to the bradytrophy of the corre- Two cyclohexadienyl dehydratases, one from E. herbicola
spondingmutant of P. aeruginosa (26). To have a better (24) and the other from P. aeruginosa, have now been char-
understanding of the role of the cyclohexadienyl dehydratase acterized in detail. Thetwo enzymes are similarwith respect
in vivo, mutants lacking this enzyme activity would be desir- to a lack of allosteric control, the broad substrate utilization,
able, and such mutants now can be obtained by using the andthe relativeaffinityforL-arogenate andprephenate.
cloned cyclohexadienyl dehydratase gene to target thecorre- However, the E. herbicola enzyme is a homotetramer, whereas
sponding region of chromosome in P. aeruginosa through the P. aeruginosa enzyme appears to be a homodimer. Fur-
gene-scrambling mutagenesis (30). thermore, the subunitmolecular weights of the two enzymes
Basis for the Ability of P. aeruginosa pheC to Complement also differed considerably,one being 18,000 ( E . herbicola) and
pheA Defects in E. coli-E. coli JP2255 was initially employed one being 30,480 (P. aeruginosa). Since the two organisms
to select for the clones carrying the bifunctional P-protein studied are relatively close to each other phylogenetically,
gene. Successful complementation of the pheA defect by P. such results were unexpected. One explanation might be that
aeruginosa pheC was not anticipated, because strain JP2255 a small ancestral gene (retained in E. herbicola) underwent a
is deficient in the two chorismate mutase species encoded by tandem duplication and fusion following divergence of the
tyrA and pheA. Even though we did not detect chorismate lineage leading to P. aeruginosa. This would explain the larger
mutase activity under standard assay conditions, a low and subunit size and the dimer instead of tetramer in P. aerugi-
perhaps unstable level of enzyme must exist in strain JP2255 nosa. However, a comparison of amino-terminal sequence
as was reported by Baldwin and Davidson (12) in order to with carboxyl-terminal sequence did not reveal any striking
explain the slow-growing pheC transformants recovered.
When the original clones and the resulting subclones were
transformed into E. coli KA197, fast-growing strains were
obtained, an indication that the limitationof phenylalanine
in vivo was relieved due to the elevation of chorismate mutase
activity. Since KA197, a pheA mutant lacking a bifunctional
P-protein,still possesses anintactbifunctionalT-protein
:
00 V
B s P D T 1 6 6 R D I Q D Y R D N H T R I L S P D E N
CgPDT: 1 7 3 D D V A D V R G A R T R F V A V Q A Q A A
E c P D T : 2 6 8 R I E A N Q R Q N F T R Y V V L A R K A I
P s P D T : 2 5 5 E K I E D R P D N S T R P L I I G S Q E V
PaCDT: 1 6 9 A R I L V H P D N V T I P Q Q I V D G K A
FIG. 6. Multiple alignment of dehydratase sequences ori-
ented to the sequence motif (37) containing the threonine
(chorismate mutase/cyclohexadienyl dehydrogenase), the pre-residue (0)shown to be essential for catalysis (12,36). Amino
phenate molecules generated in vivo were probably derived acid residues, beginning as numbered on the left, are compared for B.
subtilis PDT (Bs P D T ) , Corynebacterium
from the catalytic activity of the chorismate mutase compo- P D T ) , E. coli P-protein ( E c P D T ) ,P. stutzeri P-protein (PDT D T ) ,
glutamicum
Ps P
(Cg
nent of the T-protein. Our results indicate that the chorismate P. aeruginosa CDT ( P a C D T ) Identities between the residues of
and .
one or more of the P. aeruginosa CDT with those of the other four
' G . Zhao, T. Xia, R. S. Fischer, and R. A. Jensen, unpublished dehydratases are shown by shading. Residues invariant in all se-
results. quences are represented by boxing.
Prephenate
DehydrataselArogenate
Dehydratase Gene in P. aeruginosa 2491
identities, and the motif illustrated in Fig.6 was not present 16. Cotton, R. G. H., and Gibson, F. (1965) Biochim. Biophys. Acta
in two places. If the pheC gene in P. aeruginosa enlarged by 100, 76-88
gene duplication and gene fusion, the subsequent divergence 17. Zamir, L. O., Tiberio, R., Fiske, M., Berry, A., and Jensen, R. A.
(1985) Biochemistry 24, 1607-1612
has been extensive. It should be instructive to obtain thepheC 18. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254
gene sequence from E. herbicola. 19. Laemmli, U. K. (1970) Nature 227, 680-685
20. Dayan, J., and Sprinson, D. B. (1970) Methods Enzymol. 17A,
REFERENCES 559-561
21. Zamir, L. O., Jensen, R. A,, Arison, B. H., Douglas, A. W., Albers-
1. Bentley, R. (1990) Crit. Reu. Biochem. Mol. Bid. 25, 307-384 Schoenberg, G., and Bowen, J. R. (1980) J . Am. Chem. SOC.
2. Stenmark, S. L., Pierson, D. L., Jensen, R. A., and Glover, G. 102,4499-4504
(1974) Nature 247, 290-292 22. Shine, J., and Dalgarno, S. (1974) Proc. Natl. Acad. Sci. U. S. A .
3. Byng, G. S., Kane, J. F., and Jensen, R. A. (1982) CRC Crit. Reu. 71,1342-1346
Microbiol. 9, 227-252 23. West, S. E. H., and Iglewski, B. H. (1988) Nucleic Acids Res. 16,
4. Ahmad, S., Weisburg, W. G., and Jensen,R. A. (1990) J. Bacteriol. 9323-9335
172,1051-1061 24. Xia, T. H., Ahmad, S., Zhao, G. S., and Jensen, R. A. (1991)
5. Patel, N., Pierson, D. L., and Jensen, R. A. (1977) J. Biol. Chem. Arch. Biochem. Biophys. 286,461-465
252,5839-5846 25. Patel, N., Stenmark-Cox, S. L., and Jensen, R. A. (1978) J. Bid.
6. Fiske, M., Whitaker, R. J., and Jensen, R. A. (1983) J. Bacteriol. Chem. 253,2972-2978
154,623-631 26. Berry, A., Bhatnagar, R.K., and Jensen, R. A. (1987) J. Gen.
7. Ahmad, S., and Jensen, R.A. (1988) Curr. Microbiol. 16, 295- Microbiol. 133, 3257-3263
302 27. Whitaker, R. J., Gaines, C. G., and Jensen, R. A. (1982) J. Bid.
8. Maniatis, T., Fritsch, E. F., and Sambrook, J . (1982) Molecular Chem. 257, 13550-13556
C1oning:A Laboratory Manual, Cold Spring HarborLaboratory, 28. Byng, G. S., Whitaker, R. J., and Jensen, R. A. (1983) Arch.
Cold Spring Harbor, NY Microbiol. 136, 163-168
9. Saito, H., and Miura, K. (1963) Biochim. Biophys. Acta 72, 619- 29. Carlson, C. A., Pierson, S. L., Rosen, J. J., and Ingraham, J . L.
629 (1984) Arch. Microbiol. 140, 124-138
10. Yanisch-Perron, C., Vieira, J., and Messing, J. (1985) Gene 30. Mohr, C. D., and Deretic, V. (1990) J. Bacteriol. 172, 6252-6260
( A m s t . )33, 103-119 31. Hyde, E. (1979) Biochemistry 18,2766-2775
11. Davis, R. W., Botstein, D., andRoth, J . R.(1980) Methods 32. Hyde, E., and Morrison, J . F. (1978) Biochemistry 17,1573-1580
Enzymol. 65, 404-411 33. Hoch, G. L. E., Shaw, D. C., and Gibson, F. (1972) Biochim.
12. Baldwin, G. S., and Davidson, B. E. (1981) Arch.Biochem. Biophys. Acta 258, 719-730
Biophys. 2 11,66-75 34. Follettie, M. T., and Sinskey, A. J. (1986) J. Bacteriol. 167, 695-
13. Humphreys, G. O., Willshaw, G. A., and Anderson, E. S. (1975) 702
Biochim. Biophys. Acta 383, 457-463 35. Trach, K., and Hoch, J. A. (1989) J . Bacteriol. 171, 1362-1371
14. Prober, J. M., Trainor, G. L., Dam, R. J., Hobbs, F. W., Robert- 36. Hudson, G. S., and Davidson, B. E. (1984) J. Mol. Biol. 180,
son, C. W., Zagursky, R. J., Cocuzza, A. J., Jensen, M. A,, and 1023-1051
Baumeister, K. (1987) Science 238, 336-341 37. Fischer, R. S., Zhao, G., and Jensen, R. A. (1991) J. Gen. Micro-
15. Devereux, J., Haeberli, P., and Smithies, 0. (1984) Nucleic Acids bid. 137, 1293-1301
Res. 12,387-395 38. Holloway, B. W. (1969) Bacteriol. Reu. 83, 419-443
Continued on nextpage.
2492 Prephenate DehydrataselArogenate DehydrataseGene in P . aeruginosa
Thls study
This study
Thls study
Th15 study
Thls study
515-bp SmaI fragrncnt of pJZld-S subcloned
Into PUC18 Thli s t u d y
"EAE-
rellulose R3 5 10.616 1.518 3.02 8.1
. ,
passaqe through a PD-lo Sephadex column to remove small molecules, 15 referred
PUrlflcaLion of the Cyclahexadienyl Dehydratase Encoded by the Cloned pheC
Gene-E. coli JP2255 carrying the subclone pJZlq was qrown In 2 liters of LB
broth supplemented wlth 50 rq/ml Of ampicillin at 17OC. and harvested by
centrlfuqatxon during late exponentla1 growth phase. The cells were washed once
wlth 2 0 mn potasslum phosphate, 1 DM DTT. pH 1 . 5 (Buffer A ) , resuspended In
the same buffer. and disrupted by sonication. The resultlnq Suspension Wd5
centrifuged a t 150,000 g for 60 mi". The supernatant was applied to a DEAE-
Prephenate DehydrataselArogenate Dehydratase Gene in P. aeruginosa 2493
1 2 3 4 5 n
i
*- -
Fiq. 4 . SDS-polyacrylamide gel electrophoresis of the pheC gene prcduct
purified from E. coli JP2255(pJZlq). The protein samples were run on a 1st qel
and stained with Coomassie blue. The first four lanes show results obtained when
samples collected after each of the fractionation steps shown Table Ill were
1"
applied: lane 1, crude extract; l a n e 2, DUE-cellulose; lane 3 ,
hydroxylapatite; l a n e 4 . qel filtratmn. The molecular-weight standards used
are Shown in lane 5 .
Fig. 5 . Double reciprocal plots or P. aeruqinoss syclohexadienyl dehydratase
purified either from E. coli JP2255(pJZlql (panels A and 8 ) or directly from
P. aeruqinosn (panels C and D l . When assayed as prephenate (PPI) dehydratase
l m n e l s A and CI. v is expressed as moles of phenylpyruvate formed per mxn In
Related docs
